Legal provisions of SEC(2010)1056 - SUMMARY OF THE IMPACT ASSESSMENT Accompanying document to the Proposal for a Proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL ON SHORT SELLING AND CERTAIN ASPECTS OF CREDIT DEFAULT SWAPS - Main contents
Please note
This page contains a limited version of this dossier in the EU Monitor.
dossier | SEC(2010)1056 - SUMMARY OF THE IMPACT ASSESSMENT Accompanying document to the Proposal for a Proposal for a REGULATION OF THE EUROPEAN ... |
---|---|
document | SEC(2010)1056 ![]() |
date | September 15, 2010 |
EUROPEAN COMMISSION
Brussels, 15.9.2010
SEC(2010) 1055
COMMISSION STAFF WORKING DOCUMENT
IMPACT ASSESSMENT
Accompanying document to the
Proposal for a
REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL
on Short Selling and certain aspects of Credit Default Swaps
{COM(2010) 482 final}
{SEC(2010) 1056}
TABLE OF CONTENTS
1. Introduction 6
2. Procedural issues and consultation of interested parties 8
2.1. CESR and ESME reports 8
2.2. Public consultation 9
2.3. Steering Group 9
2.4. Impact Assessment Board 10
3. Policy context, Problem Definition, Baseline scenario and Subsidiarity 11
3.1. Background and context 11
3.1.1 Who short sells, what and why 11
3.1.2 Trades in Credit Default Swaps 15
3.1.3 Economic benefits of short selling and CDS 16
3.1.4 Stakeholders concerned by short selling and CDS transactions 18
3.1.5 Overview of current regulatory landscape and how it may evolve 18
3.2. Problem definition 21
3.2.1. Other related issues outside of the scope of this initiative 23
3.3. Problem 1: Risk of negative price spirals 25
3.4. Problem 2: Risk of settlement failure associated with naked short selling 27
3.5. Problem 3: Transparency deficiencies 29
3.6. Problem 4: Increased compliance costs and potential regulatory arbitrage arising from a fragmented regulatory framework 31
3.7. HOW WOULD THE PROBLEM EVOLVE WITHOUT EU ACTION? THE BASELINE SCENARIO 33
3.8. SUBSIDIARITY 35
4. Objectives 35
4.1. General, specific and operational objectives 35
4.2. Consistency of the objectives with other EU policies 38
5. Policy options 38
5.1. Policy options to ensure regulators have the power to restrict or ban short selling or CDS in distressed markets 38
5.2. Policy options to ensure certain requirements at the point of trading and strengthen settlement discipline 39
5.3. Policy options to ensure regulators and markets obtain data on short positions (including through CDS) 39
5.4. Policy options to ensure a coordinated response by EU member states to short selling and CDS 40
6. Analysis of impacts and choice of preferred options and instruments 40
6.1. Analysis of impacts of policy options 40
6.1.1. Policy options to ensure regulators have the power to restrict or ban short selling or CDS in distressed markets 40
6.1.2. Policy options to ensure certain requirements at the point of trading and strengthen settlement discipline 49
6.1.3. Policy options to ensure regulators and markets obtain data on short positions (including through CDS) 56
6.2. The preferred policy options and instrument 66
6.2.1. The preferred policy options 66
6.2.2. Choice of instrument to ensure a coordinated response by EU Member States to short selling and CDS 67
6.2.3. Impact on retail investors and SMEs 68
6.2.4. Impact on third countries 69
6.2.5. Social impact 70
6.2.6. Environmental impact 70
6.3. Estimate of impact in terms of compliance costs 70
7. Monitoring and Evaluation 73
8. Annex 1 – Glossary of terms 74
9. Annex 2 – Summary of public consultation on Short Selling 77
9.1. Summary of the responses 77
9.2. List of contributors to public consultation 81
10. Annex 3: Overview of measures in force in EU 27 on short selling/CDS 87
11. Annex 4 – Summary of discussions between DG MARKT and CESR-POL on Short Selling, Paris, 14 April 2010 89
12. Annex 5 – List of organisations represented in meeting of commission services with market participants and public authorities on credit default swaps 95
13. Annex 6 – regulatory regime in the united states on short selling 97
14. Annex 7 – bibliography of recent economic literature on short selling, committee on economic and market analysis (CEMA), CESR 99
15. Annex 7 – Data on shares outstanding on loan as a proxy for volume of short selling transactions in the European Union, provided by dataexplorers 104
16. Annex 8 – Estimate of compliance costs 107
17. Annex 9 – Table providing an overview of the scope of preferred options 115
1.Introduction
At the height of the financial crisis in autumn 2008, competent authorities in the United States and several EU Member States adopted exceptional measures to restrict or ban short selling in some or all shares1. They acted due to concerns that at a time of considerable financial instability, short selling was aggravating the downward spiral in the prices of shares, notably in financial institutions, in a way which could ultimately threaten their viability and create systemic risks2. The measures adopted by Member States were divergent as the European Union lacks a specific legislative framework for dealing with short selling issues.
Since that time, regulators have developed through the Committee of European Securities Regulators3 (CESR) a two-tier regime for the disclosure of short selling transactions to regulators and the public4, which they have called on the Commission to include in a proposal for a new European legislative framework. Finally, concerns have been raised5 about the possible implications in terms of settlement risk of so-called 'naked short selling'6.
Short selling is the sale of a security that the seller does not own, with the intention of buying back an identical security at a later point in time to be able to deliver the security7. It can be divided into two types: 'covered' short selling where the seller has made arrangements to borrow the securities before the sale and 'naked' short selling where the seller has not borrowed the securities when the short sale occurs. Naked short selling is often used for intra day trading.
The expression 'short selling' is sometimes used in a more general sense to cover a broad range of actions that allow an investor to profit from a price decline in an asset. In addition to short selling on cash markets, a net short position can also be achieved by the use of derivatives, whether they are traded on exchanges or over-the-counter (OTC).
A Credit Default Swap (CDS) is a derivative which acts as a form of insurance against the risk of credit default of a corporate or a government. In return for an annual premium, the buyer of a CDS is protected against the risk of default of the reference entity (stated in the contract) by the seller. If the reference entity defaults, the protection seller pays the buyer the par value of the instrument in exchange for physical delivery of the reference instrument, although settlement may also be by cash.
In March 2010, concerns were expressed by some governments also about the possible role played by derivative transactions, notably CDS, in relation to the prices for Greek sovereign bonds8. A number of Member States have adopted temporary or permanent restrictions at national level on short selling and CDS which are outlined in section 3.1.5.
In light of concerns about the regulation of short selling and CDS, the Commission decided to include in its Work Programme for 2010 a legislative initiative on short selling and Credit Default Swaps as a strategic initiative9. This was restated in the Commission Communication of 2 June 2010 on Regulating Financial Services for Sustainable Growth.10This impact assessment accompanies the proposal for a regulation on short selling and certain aspects of Credit Default Swaps.
It is important to note that this initiative is not the only one to address problems in the markets highlighted by the current crisis. As announced in its Communication of 2 June 2010, the Commission will complete its full financial reform programme in the coming months. Of the existing or pending proposals listed in the Communication, a number are related to this initiative.
The proposal for a Directive on Alternative Investment Fund Managers11 aims to create a comprehensive and effective regulatory and supervisory framework at the European level, providing robust and harmonised regulatory standards for all managers and enhancing transparency towards investors. Concerning effective supervision, the proposal for a Regulation establishing a European Securities and Markets Authority (ESMA)12 provides in its article 10 for ESMA to have the power to adopt individual decisions in emergency situations.
In terms of enhanced transparency, a forthcoming proposal which is closely related to the initiative on short selling and Credit Default Swaps is the proposal for legislation to improve the functioning of derivatives markets, which will strengthen the EU's financial market infrastructure, promote the standardisation of derivatives contracts and develop central clearing parties for derivative contracts to substantially reduce risk. The Commission will also propose improvements to the Markets in Financial Instruments Directive13 (MiFID) in order to strengthen pre- and post-trade market transparency and bring more derivatives onto organised trading venues.
Finally, in the context of strengthened responsibility and consumer protection, the Market Abuse Directive14 will be revised in order to extend its rules beyond regulated markets and to include derivatives in its scope of application.
2.Procedural issues and consultation of interested parties
The proposal for a regulation on Short Selling and certain aspects of Credit Default Swaps and its impact assessment have been prepared in accordance with the Commission's better regulation principles. The proposal takes into consideration the observations and analysis contained in the reports published by the Committee of European Securities Regulators (CESR). The proposal and the impact assessment also take into consideration the comments received from stakeholders participating in a public consultation from 14 June to 10 July 2010.
2.1.CESR and ESME reports
DG Internal Market services asked the European Securities Markets Expert Group (ESME15), an independent advisory group to the Commission composed of market participants, to prepare a report on short selling. The report containing a series of recommendations was adopted on 19 March 200916.
The Committee of European Securities Regulators (CESR) consulted in the second half of 2009 on a possible pan-European model for the reporting and disclosure of net short positions in EU shares17. It then recommends that the Commission introduces such a regime as soon as possible. CESR also published on 22 September 2008 (updated on 22 February 2010) a table setting out the measures adopted by national competent authorities on short selling.18
DG Internal Market services held a meeting with CESR experts in Paris on 14 April 2010 to consult national competent authorities on issues relating to short selling, net short positions and CDS. The summary of the discussion is included in annex 4. National competent authorities also responded to a questionnaire circulated by the Commission services in June 2010.
The Commission services held two meetings on 5 March 2010 with representatives of market participants and national competent authorities on issues relating to sovereign CDS. The list of participants is included in annex 5.
2.2.Public consultation
In April 2009, the European Commission asked some high level questions on a possible regulatory regime for short selling in the context of a call for evidence on the review of the Market Abuse Directive launched on 20 April 200919. Most contributors responded that the Market Abuse Directive was not the appropriate instrument for addressing short selling, as most short selling does not constitute market abuse in the view of market participants.
On 14 June 2010, the European Commission launched a public consultation on policy options for a possible legislative initiative on short selling. The Commission services received around 120 contributions. The non-confidential contributions can be consulted on the Commission's website20. The outcome of the consultation has been summarised in Annex 2.
2.3.Steering Group
The Steering Group for this Impact Assessment was formed by representatives of a number of services of the European Commission, namely the Directorate General Internal Market and Services, the Directorate General Competition, the Directorate General Economic and Financial Affairs, the Directorate General Enterprise, the Directorate General for Health and Consumers, the Legal Service and the Secretariat General. This Group met three times, on 18 May 2010, 7 July 2010 and on 23 July 2010. The contributions of the members of the Steering Group have been taken into account in the content and shape of this impact assessment21.
2.4.Impact Assessment Board
DG MARKT services met the Impact Assessment Board on 30 August 2010. The Board analysed this Impact Assessment and delivered its opinion on 31 August 2010. During this meeting the members of the Board provided DG MARKT services with comments to improve the content of the Impact Assessment that led to some modifications of this final draft. These are:
- Modifications to highlight the problem of fragmented regulatory approaches to short selling and CDS and the added value that the EU can bring in creating a common European framework to ensure the smooth functioning of the single market;
- A clearer distinction between risks which regulators have expressed concern about and which should be the subject of EU action as part of a precautionary approach, and problems for which empirical evidence exists;
- A more detailed explanation of the nature of EU level coordination and the respective roles of national competent authorities and ESMA, particularly with regard to the circuit breaker and measures in exceptional situations, and clarification of how implementing measures will be used to further specify exceptional situations;
- Clarification of which financial instruments are covered by each preferred option, notably through the addition of an overview table in the annexes, as well as of the explanation and presentation of some of the options;
- A fuller explanation of the rationale for exempting market making activities and how this option will be implemented so as to ensure the effectiveness and efficiency of the legislation;
- Clarification of how cooperation with non-EU authorities would ensure effective implementation and compliance;
- Clarification of the analysis of compliance costs;
- The inclusion of a more comprehensive glossary of key terms.
3.Policy context, Problem Definition, Baseline scenario and Subsidiarity
3.1.Background and context
3.1.1 Who short sells, what and why
Short selling is not a new phenomenon – it has been a feature of equity markets since at least the 17th century, when short sellers were accused of causing the collapse of shares in the Dutch East India company in 1609, and short sellers were accused of causing the 1929 stock market crash, which led to legislation in the United States to regulate short selling22.
In theory, a variety of different financial instruments are capable of being sold short: shares, credit instruments, interest rates, currencies or commodities. However, in practice short selling is more widespread and can be conducted more easily for some instruments than others, and the evidence of risks, or concerns about potential risks, is greater for some instruments (i.e. shares) than for others. With the exception of Credit Default Swaps, which are examined in more detail in section 3.1.2. below, derivatives will be considered in relation to their underlying instruments, as it is from this underlying instrument that they derive their value. Also, in most cases derivatives cannot really be sold short, but they can be used to create a position which is economically equivalent to short selling the underlying instrument.
Short selling in shares
Short selling is used by a variety of market participants including hedge funds, traditional fund managers such as pension funds and insurance companies, investment banks, market makers and individual investors.23 Short selling can be used for speculative purposes, to hedge a long position, for arbitrage and for market making. Examples of different reasons for short selling are provided in the box below.
Examples of reasons for short selling24
Speculation: for example, an investment bank trading for its own book ("proprietary trading") may sell a share short because it believes the price will decline, in order to profit from that decline by buying back the share at the lower price. However, short selling is a risky strategy, as if the share price rises instead of falling, then the short seller will have to close out the position at a loss or pledge more collateral to keep the position open.
Hedging: although pension funds and insurance companies tend to buy shares in order to profit from their (anticipated) rise over the long term, they often hedge their risk by selling short comparable shares to those in which they hold a long position. If the share price in their long position goes down, then they can limit their losses through the rise in the value of the short position. This is a common practice in most developed financial markets.
Arbitrage: for example when a hedge fund combines a short position and a long position in two different but inter-related shares, to make a profit from the price differential between the two shares.
Market making: in order to meet customer demand for securities which are not immediately available, market makers use short selling to fill client orders, i.e. to provide liquidity to the market.
Estimates of the volume of short selling of shares vary widely. Some studies estimate that short selling in the United States accounts for between 14 and 30% of equity trading volume25. Studies also show that short selling volume can be considerably higher in the US, particularly in shares of financial institutions at times of financial instability; for example, short sales of shares in some US financial institutions exceeded 40% of trading volume in those shares in summer 200826.
The availability and reliability of data on the volume of short selling in Europe is limited in the absence of marking of transactions or disclosure regimes, and of reporting of over the counter transactions27. In the consultation of regulators carried out by the Commission services, most regulators who responded did not have any data on the volume of short selling transactions on their markets or reported very little or no short selling on their markets. However, regulators in some Member States were able to provide some data, notably the UK, Spain and Greece, and data on securities lending, which can serve as a proxy for short selling, has been provided by Dataexplorers (see annex 7). For the UK, the FSA provided data on securities lending as a percentage of market capitalisation as a proxy for the volume of short selling stock lending; according to this data, securities lending averaged between 1 and 3% of market capitalisation between August 2006 and August 2009 (disregarding spikes which the FSA attributes to dividend payment periods). Since according to the FSA a large proportion of securities lending is for tax purposes, this data suggests lower levels of short selling. The UK FSA also provided data based on disclosure of significant short positions in financial sector shares only between January and May 2009; this showed that the largest proportion of short positions was in the range of 0.25-0.35% of the total share capital of the issuer, i.e. just above the UK disclosure threshold of 0.25%28.
For Spain, the CNMV provided data from the Spanish clearing house Iberclear using securities lending as a proxy for covered short selling. This data showing that covered short selling activity in shares amounted to 1.4% of the total number of trades in 2009, essentially stable from the figure in 2007, which was 1.36% . In terms of value, short selling represented 6.47% of trading in 2009, again broadly stable from the figure of 6.35% in 2007. This data appears consistent with that provided by the UK and also using securities lending as a proxy. Based on the disclosure of short positions between September 2008 and June 2010 in Spain, a total of individual 441 short positions were reported and published, from 56 entities (mostly hedge funds), with an average position disclosed of 0.46% of the share capital of the issuer. The Spanish regulator also provided data estimating the level of naked short selling in their jurisdiction based on failed transaction statistics. When considering these figures it has to be noted that Spain has a permanent ban on naked short selling, so these cases represent occurrences of naked short selling despite the ban, and at the height of the crisis the Spanish regulator reminded market participants of their obligations in this regard29. According to their estimates, naked short selling represented 0.04% of all trades in 2009, down from 0.11 in 2007 (or 0.08% of trading volume in 2009, down from 0.30% in 2007).30
Greece is the only EU Member State which has a requirement for short sale transactions on its market to be flagged, and the Greek regulator is therefore in possession of data showing the actual volume and value of short sale transactions on its market. According to data for January to June 2010, the volume of short selling transactions as a percentage of total transactions is in a range between 0 and 3.33%. In terms of value the range is between 0 and 3.36%.31
To conclude, it is difficult to obtain reliable data on the extent of short selling in Europe in the absence of marking of transactions, or of disclosure of short selling transactions. However, according to the data outlined above it could be estimated to represent between 1 and 3% of market capitalisation using securities lending as a proxy, or less than 1% of the total share capital of the issuer using data on disclosures of net short positions in the UK and Spain as a measure. It is difficult to estimate the volume of naked short selling in the absence of data from countries where this practice is permitted, but the data from Spain suggests that this is a fraction of the total volume of short selling. However, while the volume of short selling in Europe appears on this evidence to be limited, concerns remain about the risks short selling can pose, especially in distressed markets, and these are explored further in the problem definition below.
Short selling in credit instruments
Credit instruments include both corporate and sovereign bonds. In principle it is possible to sell corporate bonds short, but the liquidity of the corporate bonds market is limited as there is no active secondary market for a large part of the bonds outstanding. Several factors explain this: each issuer tends to have different issues outstanding, so secondary market activity is fragmented into these different issues; and the maturity of corporate bonds is limited compared to equities where there is no repayment schedule, and investors on corporate bond markets tend to buy and hold for the long term. Therefore, most of the short selling activities on corporate bonds take place through the CDS market, where the liquidity is much higher. Regarding sovereign bonds, this market presents higher liquidity then the corporate bond market as there tend to be fewer issues per issuer and each issue is of a larger size. In addition, the sovereign CDS market is less liquid then the corporate CDS market. Short selling a bond could therefore be considered easier and a short seller would face a more balanced choice between using the bonds or the CDS.
Short selling in interest rate instruments
This market consists mostly of derivatives. There is no real short selling of interest rate derivatives as any derivative contract does not pre exist the operation, it is only created when bought or sold; in other terms, there is no secondary market. Interest rate derivatives are very global instruments, and they are used by a wide range of participants for risk management purposes and through often complex strategies. There is no evidence of concerns related to short selling on this market.
Short selling currencies
Currency markets are open 24 hours a day, 7 days a week, and may be the most global market of all asset classes. Short selling does occur on this market and there is evidence of it being used in the past to drive down the price of a currency (most notoriously the speculation against the pound sterling by George Soros in the context of the instability in the Exchange Rate Mechanism in 1992; there have also been concerns expressed by some Member States in February-March 2010 about the possible role of short selling of the euro in the decline of the currency's value32). Nevertheless, short selling regulation has been very seldom used worldwide in the foreign exchange market, and when applied, it has only been done by a few emerging countries.
Short selling commodities
Strictly speaking, short selling of the underlying physical commodity is not an issue for financial regulation. Selling physical commodities short in anticipation of their price decreasing, which would then be bought back at the lower price, may occur but this is probably dampened by the costs of storing/producing the goods, and is an option mainly for commodity companies. Whether this practice is harmful, for example for consumers, is an issue for sectoral (e.g. energy) legislation rather than financial regulation.
Entering into equivalent short positions through commodity derivatives is possible, but attempting to disassociate this from legitimate hedging and price discovery is very difficult. This would also apply to short positions entered into either in the spot or derivative market as part of arbitrage strategies, and putting controls on this could have a negative impact on price convergence. In addition, any controls would require a highly differentiated approach according to each specific commodity market. Many commodity markets are global (e.g. crude oil, metals) and trading could easily shift to other jurisdictions, and continue to set prices for Europe. Furthermore, sector-specific legislative proposals to enhance transparency and market integrity in certain commodities markets is under preparation, notably for gas and electricity and for emissions allowances.
3.1.2 Trades in Credit Default Swaps
Credit Default Swaps (CDS) are financial derivatives. Derivatives are referenced on an underlying, which in the case of CDS is the credit risk of an issuer. The credit risk of an issuer is the risk of default of that issuer on its obligations towards its creditors. CDS transfer credit risk from one party to another.
It is easiest to understand a CDS contract by comparing it with a bond. In a typical bond, the issuer agrees to pay the investor a regular sum, the coupon, in exchange for the principal amount, the issued amount. The investor is exposed to credit risk, as the borrower may not return the principal: therefore buying a bond implies taking a position on the credit risk of the issuer. A similar economic effect can also be achieved by selling a credit default swap, but here no principal changes hands ex ante. In a CDS, the seller receives regular payments from the buyer, while his obligation materialises only in event of default, when he has to provide for the credit loss.
At the end of May 2010, the gross notional amount of the total CDS market was USD 14.5 trillion, with about 2.1 million contracts outstanding. The sovereign CDS market, which includes both sovereign indices and sovereign single names, reached USD 2.2 trillion, with about 0.2 million contracts outstanding. The outstanding gross notional amount of the Itraxx Sovereign Index Western Europe was USD 140 bn (and USD 10 bn in net terms)33.
There are four main groups of market participants in the CDS market: dealers, non-dealer banks, hedge funds and asset managers. The dealers are by far the largest players on the market.34
The aims of these market participants are diverse and they employ different strategies: CDS can be used for hedging, arbitrage or speculative purposes. Examples of the reasons for buying or selling CDS are provided in the box below.
Reasons for trading CDS
Hedging: CDS can be used to neutralise or reduce a risk to which the CDS buyer is exposed from another position. An example of such an 'insurable interest' would be a bondholder's exposure to the credit risk of the issuer of the bond; by buying a CDS he can reduce that risk by passing it on to the CDS seller. The bondholder has hedged his position with a CDS. Sometimes the buyer of a CDS is seeking to hedge a risk other than the credit risk of the bond issuer, for example the credit risk of a bank heavily exposed to the bond issuer.
Arbitrage: arbitrage is usually understood as the risk-free exploitation of price differences in connected markets. The typical arbitrage operation that involves CDS is the combination of buying a CDS and entering into an asset swap where the fixed coupon payments of a bond are swapped against a stream of variable payments. It is known in the market as a trade on the basis.
Speculation: CDS can also be used to take a position in order to exploit price changes by trading in and out. For example, a CDS seller has taken on risk (in exchange for the regular payments he receives from the CDS buyer); he will gain from the contract if the credit risk does not materialise during the contract's term or if the compensation received will exceed a potential payout. If a CDS buyer does not hold an underlying insurable interest (known as naked CDS), he is also taking a position: he would gain from the contract when the payout he receives from a possible default event exceeds the premiums he has paid, or when the market value of the CDS has increased above the acquisition price; such increase would reflect a perception by the market that the risk of default has raised.
3.1.3 Economic benefits of short selling and CDS
There are many studies on the economic effects of short selling of shares, and most conclude that it has a number of general benefits to the market (a bibliography of the economic literature on short selling is included in annex 5).
First, it significantly increases market liquidity in a security, as short sellers can sell a security without owning it35. Liquidity is important for a market as it makes the completion of trades more likely, and larger trading volumes reduce transaction costs and tend to increase efficiency36. According to a study by the consultancy Oliver Wyman, "as relatively substantial participants in equity markets, short sellers play an integral role in providing liquidity and maintaining market efficiency. When short sellers' level of participation decreases, market become less liquid, more expensive and more difficult to trade." 37 It is important to note that short selling typically involves not only the sale of the security concerned but also at a subsequent point the purchase of a similar quantity of the security to cover the short sale (so extra liquidity is provided from both sets of transactions).
Second, it can act as a balance to irrational overpricing of securities and can mitigate price bubbles by allowing investors to act where they believe that a security is overvalued38.
Third, it leads to more efficient price discovery as it prevents prices from reflecting only the views of the most optimistic investors in the market and leads to faster integration of information into the price of securities39. A review of the economic literature by the FSA concludes that 'in summary, the empirical literature tends to confirm, with some qualifications, the theoretical proposition that short selling allows negative expectations about share price developments to feed more directly into the actual share price and thus contributes to efficient pricing.'40
Also, covered short selling requires the lending of securities to the seller to cover the settlement of the short sale. Securities lending by institutions such as banks, insurance companies and pension funds generates additional significant revenue for longer term holders of securities.
Academic studies also acknowledge that introducing derivative securities increases the investment opportunities for investors, which in turn could make markets more efficient, lead to positive welfare effects, and make the derivatives market interact with the underlying securities market. For example, CDS allow market participants to hedge their positions or express specific views on pure credit risk. In general, CDS markets are considered more liquid than bond markets, as they are more standardised. For example all sovereign CDS on Greece have a unique reference which is the credit risk of Greece. On the contrary, the bonds issued by Greece will split into different issues with different specificities, that will co-exist on the market. These different bonds will not be completely interchangeable, at least not to the extent that CDS would be. The higher liquidity of the CDS market allows market participants to express their views more easily. This is especially the case with regard to taking a short position in an entity, which is easier to do on the CDS market than using bonds. On the bond market investors have to rely on a functioning repo market in order to borrow the bonds necessary for short selling. In several countries, the repo market is not very deep and therefore is less functional. In addition, the required outlay to enter into a short position on the CDS market is much less than its equivalent on the cash market. This is due to the built in leverage effect of derivatives.
While short selling and CDS transactions have benefits, there are also risks and these are analysed further below.
3.1.4 Stakeholders concerned by short selling and CDS transactions
In addition to the financial institutions or investors who engage in short selling or CDS transactions for the reasons explained above, many other stakeholders can be affected by such transactions, notably:
- issuers, who wish to maintain shareholder value and therefore may fear that their share price may be affected negatively41;
- financial institutions, who in an extreme case may face the risk of bank runs if short selling amplifies a fall in their share price in a disorderly way42;
- individual investors, who may suffer from information asymmetry if they do not know the extent to which the price of a security is being affected by short selling and whether short sellers will need to later buy back the shares they have sold short43;
- regulators, who have an interest in knowing whether short positions are building up which could pose a systemic risk or whether short selling is being used abusively44;
- governments, as issuers of bonds, who may fear a negative impact on sovereign bond spreads and the cost of raising funds as a result of short selling of government bonds in the cash market or through the use of bonds derivatives and CDS45.
3.1.5 Overview of current regulatory landscape and how it may evolve
As outlined in the introduction and detailed in annex 3, EU Member State regulators adopted divergent approaches to the regulation of short selling in autumn 2008, ranging from temporary bans on short selling of financial stocks to partial bans on naked short selling and to no action at all46. As explained in section 3.4, some Member States subsequently introduced short selling disclosure requirements with thresholds varying from 0.01% to 0.25% of the total share capital of the issuer.
Those Member States that introduced temporary measures did so due to concerns that short selling may have been aggravating the downward spiral in the prices of shares, notably in financial institutions, in a way which could pose risks to financial stability, orderly markets and market integrity47.
In addition to the measures adopted by a number of Member States in autumn 2008, the Greek regulator introduced the flagging of short sale orders and an uptick rule on 15 May 200948. The German regulator BaFin adopted on 18 May 2010 a temporary ban (until 31 March 2011) on CDS transactions where the reference liability is also a liability of a euro zone country and is not used to hedge default risks (i.e. a ban of naked CDS on sovereign debt); the ban also concerns uncovered short selling of debt securities of eurozone countries admitted to trading on a domestic exchange, and a ban on uncovered short selling of shares in 10 financial institutions 49. On 28 April 2010 the Greek Capital Markets Commission decided to temporarily ban short selling of shares listed in the Athens Exchange; this was lifted on 26 August 2010 but a ban on naked short selling was maintained50.
On 2 June 2010, the German Cabinet of Ministers endorsed a draft law that essentially bans naked short-selling of all shares and government bonds traded on a German regulated market including correlated derivatives on shares, and of certain credit derivatives and currency derivatives that are entered into in Germany without having a relevant hedging function51. On 10 June, the French National Assembly adopted in first reading a draft law on banking and financial regulation which reduces the settlement period for short selling from three days to one52. On 1 July 2010, new rules including the flagging of short sale orders became effective in Poland53.
At European level, national competent authorities discussed and agreed within CESR in March 2010 a two-tier regime for the disclosure of short selling transactions to regulators and the public54. CESR has recommended that those CESR members that already have the necessary powers at national level should begin to implement the model and that others should do so on a best endeavours basis. The report advocates a two-tier model for disclosure of significant individual net short positions in all shares that are admitted to trading on an EEA regulated market and/or an EEA multilateral trading facility (MTF). At the lower threshold of 0.2%, positions would be disclosed to the relevant regulator, at the higher threshold of 0.5% positions would be disclosed, in addition to the regulator, also to the market as a whole.
At EU level, when short selling is carried out in connection with abusive strategies, such as the spreading of false rumours in order to drive down share prices, it is prohibited by the Market Abuse Directive (MAD)55. However, there is currently no overarching EU legislative framework dealing with other risks that can arise from short selling.
With regard to the regulation of the CDS market, which is an over the counter (OTC) market operated by broker dealers, these brokers dealers need to be authorized under the requirements of article 5 of the Directive on Markets in Financial Instruments (MiFID)56. These firms are also subject to the operations conditions laid out in this same Directive (articles under Chapter II). Nevertheless, the participants in these markets are considered professional clients or eligible counterparties in the MiFID terminology (see article 24 and Annex II of the Directive). Therefore, they do not benefit from the same level of protection as of retail investors. Furthermore, the CDS market is not operated on regulated markets or MTFs and is not subject to the same transparency or reporting rules. Only a few countries have decided to extend the reporting obligations (UK, Spain), options which are defined in Recitals 45 and 46 of the MiFID Directive, to the CDS market.
Regulatory developments on short selling in the USA
Other regulatory developments which are relevant in this context have occurred in the United States in recent years (a more detailed overview and explanation of the US measures is provided in annex 6). The United States operated an uptick rule from 1937 until its abolition by the Securities and Exchange Commission (SEC) in 2007. In 2004, the SEC adopted Regulation SHO57 which introduced a number of additional requirements for short selling: a locate rule for short sellers, a flagging regime and a 'close-out' requirement for short positions. On 24 February 2010 the SEC adopted the 'revised uptick' or 'circuit breaker' rule. This rule restricts short sales of a share whose price has fallen by more than 10% compared to its closing price the previous day58.
During the financial crisis, the SEC introduced a number of temporary emergency measures. On 15 July 2008, the SEC introduced a temporary 'pre-borrow' requirement on short selling of shares in 19 systemically important financial institutions59. On 18 September the SEC imposed a temporary ban on short sales of 799 financial stocks60, which grew to 1000 issuers. In September 2008 the SEC also tightened the 'close-out' requirement so that it applied to all securities not settled the day after the normal settlement date, and imposed temporary reporting requirements on certain short sales and positions.61 Since 1 August 2009, the SEC has been working with self-regulatory organisations to make short selling volume and transaction data available to the public through the latter's web sites. The Wall Street Reform Act enacted into law by the US President on 21 July 2010 includes certain provisions on short selling, notably it requires the SEC to adopt rules for public disclosure, at least monthly; of the amount of short sales by institutional investment managers62.
Regarding CDS and especially sovereign CDS, no specific measures have been adopted by the US authorities for the time being. However, CDS fall within the scope of new US legislation on financial services just adopted by the US Congress, and the CFTC and SEC will be expected to produce joint rules to implement this.
As this overview shows, the regulatory landscape is not static, either in the EU or the United States. In light of the recommendations of CESR for a disclosure regime, Member States may adopt national legislation to implement these recommendations, although some may prefer to wait until a European legislative initiative is proposed. CESR is also considering a possible regime for disclosure to regulators of significant net short positions in sovereign bonds. These net short positions would take into consideration short positions obtained through bonds and credit derivatives, including CDS transactions on sovereign bonds. Finally, as explained above there has been a strong call for EU legislative action by some heads of state and government.
3.2.Problem definition
This section examines the main risks associated with short selling in the European Union and the problems they give rise to. These risks and problems can be summarised as follows: (i) risks of negative price spirals, which can result in disorderly markets and possible systemic risks; (ii) risks of settlement failure that may be associated with naked short selling; (iii) accompanying these risks is the problem of transparency deficiencies on short selling transactions which prevent markets and regulators from effectively preventing potential negative impacts of short selling, resulting in risks to financial stability, market integrity and information asymmetries between market participants. Finally several Member States have taken, or are likely to take in future, action to address the above risks and problems, but in an uncoordinated manner. This fragmented regulatory framework then leads to the most important problem, that of (iv) increased compliance costs for firms and potential regulatory arbitrage.
The following problem tree provides an overview of the various problems and risks that arise in the context of short selling. The issues are displayed in the problem tree in order of their importance, starting with the most important problem, regulatory fragmentation.
A detailed explanation will follow in the section below, which begins with the explanations of the risks of negative price spirals and settlement risks, followed by the problems of transparency deficiencies and regulatory fragmentation.
Drivers Risks/Problems Consequences
3.2.1. Other related issues outside of the scope of this initiative
In addition to the problems outlined above and described in the problem definition, there are a number of other risks and problems which are related to short selling and CDS which will be addressed in other pending initiatives:
- Mitigation of counterparty credit risk of CDS transactions: the options aimed at reducing counterparty credit risk (either through bilateral or central clearing) in the OTC derivatives market, including the CDS market segment, are examined in the impact assessment for the legislative proposal on derivatives and post-trade market infrastructures, due by the end of the summer, as this is an issue which is of concern for all derivatives and is therefore best addressed in a horizontal instrument;
- Lack of transparency on CDS positions: the options aimed at increasing the transparency of OTC derivatives positions, including CDS positions, are also considered in the impact assessment on derivatives and post-trade market infrastructures as this is a horizontal issue for all derivatives;
- Transparency deficiencies for regulators on all financial transactions: currently, only transactions on regulated markets are required to be reported63, whereas OTC transactions are not. The option to require reporting to regulators of all transactions, including OTC transactions, in all instruments, not just significant net short positions, will be examined in the impact assessment for the revision of the Directive on Markets in Financial Instruments (MiFID), due by the beginning of 2011, as it would provide regulators with an additional tool to detect possible systemic risks and risks to market integrity;
- Transparency deficiencies for regulators concerning all financial positions: the option to require the reporting of all significant financial positions, whether long or short, and the possibility of setting limits for such positions, will be assessed in the impact assessment for the MiFID review, as this option could provide regulators with a tool to prevent market participants from creating disorderly markets through excessively large positions and to detect possible systemic risks;
- Market manipulation through OTC derivatives: the option to extend the prohibition of market manipulation to all OTC instruments including derivatives which could impact the prices of financial instruments traded on a regulated market or MTF will be considered in the impact assessment for the review of the Market Abuse Directive, due by the end of the year. This option would complement this initiative by providing regulators with the tools to sanction possible market manipulation of underlying bond markets through CDS.
As the above problems and the possible options to tackle them are horizontal in nature, for reasons of efficiency and better regulation they are best tackled in other initiatives, which are all subject to impact assessment, and are therefore excluded from the scope of this impact assessment.
In addition, the Council and the European Parliament are discussing in the context of discussions on the ESMA Regulation64 the possibility of ESMA having the task to temporarily prohibit or restrict certain types of financial activities that threaten the orderly functioning and integrity of financial markets or the stability of the financial system under the conditions set out in specific legislative acts or in an emergency situation. The possibility of prohibiting some products could also be discussed within the revision of MiFID.
Finally, the European Parliament has proposed amendments to the proposal for a Directive on AIFM which would regulate short selling and naked short selling. However, the Commission considers that these issues are best addressed in a horizontal initiative aimed at all investors who engage in short selling, rather than an instrument aimed only at alternative investment fund managers, since all investors can potentially engage in short selling and legislation only addressed at one category of investors would leave regulatory gaps.
3.3.Problem 1: Risk of negative price spirals
Short selling can lead to more efficient price formation by preventing the prices of securities from reflecting only the views of the most optimistic investors. However, especially in distressed markets when financial confidence is lacking, there is a risk of short selling creating the impression that there is more supply on the market than there really is, and thereby inciting others to sell ('herding behaviour'), which can lead to excessive downward pressure on the price of securities. The risk of negative price spirals becoming self-fulfilling, which can lead to disorderly markets and even systemic risks, is the main concern expressed by regulators with regard to short selling65.
The risk of short selling of shares in financial institutions amplifying negative price spirals and leading to systemic risks (through a contagion effect) can be explained with reference to the financial crisis. While fundamentals and lack of investor confidence rather than short selling were arguably the major contributors to the underlying market volatility, a number of EU regulators and other regulators around the world introduced emergency bans on short selling of financial stocks in autumn 2008 due to concerns that excessive short selling could amplify the decline in prices, which could become a self-fulfilling downward price spiral and ultimately lead financial institutions into financial difficulties through bank runs and have a contagion effect66on other institutions.
It is important to note that short selling on cash markets is not the only way for an investor to secure a short position; this can also be achieved by the use of derivative transactions such as options or contracts for difference67. CDS can also be used to secure an economic short position. Buying a CDS without holding an underlying insurable interest ('naked CDS') is economically equivalent to short selling a bond, as the naked buyer benefits if the price of the CDS goes up. The CDS price will go up if the risk of default increases, or at least the perception of the risk of default by the market increases. Any perceived increase in the risk of default of a bond issuer will result in increased spreads and increased interest rates on the bonds, hence a decrease in the price of the bonds. So, the buyer of the naked CDS benefits from an evolution of price which is symmetrical to the one benefiting the person short selling the bond.
A specific problem which has been raised by several governments and regulators in Europe68 with regard to CDS concerns the interaction between CDS and bond markets, and the fear that this interaction could cause mispricing on bond markets and thus higher funding costs for governments. As CDS contracts tend to be more liquid than the underlying bonds, and are often used by investors who are more reactive in their investments than bondholders, CDS are sometimes considered as likely to integrate new information faster than the bonds.
Furthermore, as the investors using CDS are also more mobile than traditional bonds investors, CDS prices can be more volatile than bonds prices. Because of the arbitrage relationship mentioned before, CDS and bonds prices are interconnected. Nevertheless, this interconnection is complex. Furthermore, the structure of the CDS market tends to be concentrated.69 This concentration could contribute to higher short term volatility in CDS prices. For example, a sudden rise in the demand for CDS to hedge either pure sovereign risk or more general macroeconomic risks, or even to speculate on that market, may unleash disproportionate effects on the CDS prices if their supply cannot react swiftly enough in the short run. If market participants base their expectations concerning bond prices on CDS behaviour, this may contribute to a very fast decline of prices on the cash bond market, which may lead to self fulfilling expectations. Traditional bonds investors may take fright at the drop in value of their investments caused by the contagion of the CDS market and start disinvesting. The disposal of their bond positions by long term investors could accelerate the fall of the bond market.
Furthermore, because of its mechanisms, the CDS presents the danger of self-fulfilling effects that can open risks of moral hazard. If market participants buy CDS without holding a proportionate insurable interest in the underlying debt obligation or exposure to the underlying credit risk, it can be argued that such a possibility gives the holder of the CDS a perverse incentive to precipitate a default and therefore obtain the payout foreseen by the CDS. However, as it is difficult for a CDS buyer to achieve such an outcome directly, the mechanism is argued to be more indirect: creating strong demand for CDS drives up the credit risk premium, making it more expensive to issue bonds and roll over debt, which can create capital gains for investors with short positions on the underlying bonds; pushed to its extreme, this could ultimately result in a default of the issuer. Nevertheless, such an action is likely to be very difficult to achieve regarding sovereign CDS markets as they only represent a tiny fraction of the underlying bond markets70. A price evolution on CDS that would be 'too artificially' driven would be very short lived and counterbalanced by the price evolution on the bond market. The difference in size between the sovereign CDS and bonds market is also likely to play a role in limiting the effects of a negative price spiral.
There are other risks associated with CDS, for instance, the mechanism of the instrument itself: in case the entity covered by the CDS contract defaults, the seller of protection may have to disburse large amounts, and may not be in a position to meet these payments. This creates huge counterparty risks71. However, this risk is not addressed here as it is covered in the impact assessment for the separate Commission initiative on market infrastructures72.
In summary, the responses of regulators to the questionnaire by the Commission services revealed that many of them have concerns about the risk of short selling amplifying price falls in distressed markets, and that this could lead to bank runs through a contagion effect. While to date there is no clear empirical evidence showing a causal link between short selling and negative price spirals, the concern of regulators was sufficient to lead many of them to restrict or ban short selling during the financial crisis as a precautionary measure and most regulators have expressed support for clear powers in exceptional situations to guard against this risk in the future (see annex 4). Further, the fact that Member States have taken divergent regulatory responses to this risk has led to a fragmented regulatory framework which leads to increased compliance costs for market participants and potential for regulatory arbitrage (see problem 4).
3.4.Problem 2: Risk of settlement failure associated with naked short selling
The risk of the short seller failing to deliver the shares to the buyer by the settlement date, as well as a risk of increased price volatility, are the main risks associated by some regulators with naked short selling73.
Data on levels of settlement failure is very limited, but the data available from some Member States suggest they are low in Europe. According to the data provided by one regulator, settlement failures represented 3.89% of all equity transactions cleared by a central counterparty (CCP) in 2010, compared to 2.74% in 2009. In another Member State settlement failures represented 2.71% of all transactions in 200974. In Spain (where naked short selling is banned), settlement failures occurred in 0.11% of all securities trades in 2007 and 2008 and in 0.04% of trades in 2009; for sovereign bonds settlement failures represented 0.22% of transactions in 2009 and 0.003% of corporate bonds in the same year75.
When assessing the reasons for these settlement failures, the few regulators that were able to provide reasons cited administrative problems such as back-office mistakes, naked short selling or other factors such as the delayed transfer of funds for payment76. The Spanish regulator carried out a study of settlement failures between January and December 2008 which showed that 59% of settlement failures came from naked short selling activity, although only 0.39% of the total settled amount could be considered as settlement failure77.
In terms of qualitative responses to the consultation of regulators carried out by DG Internal Market78, one regulator cited several cases of 'massive naked short selling prior to capital increases which resulted in failure to settle on settlement day and market disturbances'. One regulator sanctioned an intermediary in September 2008 who was unable to deliver securities in due time as a result of uncovered short selling; they also identified a number of cases of suspected naked short selling end 2008 but had difficulties to determine the liability of the persons involved. On the other hand, 14 regulators had no evidence or experience of naked short selling or of settlement problems occurring as a result, or thought that the risks were limited or could be addressed by settlement discipline. While the UK FSA sees 'occasional cases of problems caused by naked short selling in the equity markets', they consider 'that the overall risks from naked short selling are minimal'. According to the FSA's discussions with exchanges and market participants, "naked short selling is by no means a principal source of settlement failures – more common are administrative errors or the knock-on effects of problems elsewhere in the settlement chain"79
In the US, concerns about high levels of settlement failure linked with naked short selling led the SEC to introduce an emergency order on 15 July 2008, imposing a requirement on short sellers of specific financial stocks to borrow those shares before entering into a short sale. An analysis by the US Office of Economic Analysis looked at settlement failures for 19 stocks listed in the emergency order before and after the order came into force, and found 'a large reduction in fails to deliver' of 64% in volume and 56% in value.80
It should be recalled that the Directive on markets in financial instruments (MiFID) imposes specific obligations on market operators to have in place effective arrangements to ensure transactions can be settled81, and most exchanges and Member States have rules on settlement discipline. The Commission services are also undertaking separately preparatory work for a horizontal initiative which will address settlement discipline, including the option of a possible harmonisation of settlement period.
At the same time, in their responses to the questionnaire by the Commission services or the public consultation, several regulators cited the risk of price volatility as being greater with naked short selling than covered short selling. Some regulators expressed concern that in extreme cases naked short selling can put enormous pressure on share prices, which can in turn endanger the stability of the financial system82. This is because naked short selling enables the seller to sell, in principle, an unlimited number of shares in a very short space of time as they do not have to ensure that their position is covered first by borrowing or locating shares to borrow.83 One regulator cited the Porsche-Volkswagen case as evidence of this, when the number of shares being sold exceeded by a significant margin the number available to borrow as Porsche had secured a significant holding in Volkswagen through the use of derivatives.84
In summary, while some regulators consider the risk of settlement failure to be limited and consider that to the extent that any action should be taken to address risks of naked short selling, this should be confined to enhanced settlement discipline, others do perceive a risk of settlement failure, as well as a risk of increased price volatility, arising from naked short selling which should be addressed as a matter of principle and as a precaution to guard against possible systemic and market integrity risks. In addition, the fact that Member States have adopted different measures in relation to the risk of settlement failure and naked short selling fragments the European regulatory framework, leading to higher compliance costs and potential for regulatory arbitrage (see problem 4 below).
3.5.Problem 3: Transparency deficiencies
In most EU Member States there are currently no disclosure requirements for short selling or CDS transactions, either to regulators or to the public, so these Member States have no direct access to data on the short positions held in their jurisdictions; although as a result of the financial crisis, a number of Member States have introduced different short selling disclosure requirements85. However, consensus was reached by all regulators within CESR to introduce a disclosure regime for net short positions relating to EU shares. For CDS, the Depositary Trust and Clearing Company (DTCC) in the US operates a trade information warehouse that claims to collect information on 95% of the trade done on CDS in the world. Therefore regulators can obtain this data by requesting it from the DTCC.
Regulators have expressed concern that the absence of information about short selling activity makes it difficult for them to detect the build-up of positions which could have implications for the stability of markets. Greater transparency through disclosure to the regulator could help regulators to identify when this is occurring, and serve as a deterrent to the taking of aggressive short positions which could contribute to disorderly markets86.
Another difficulty is that in the absence of transparency requirements for short selling which cover short positions obtained through OTC transactions or derivative transactions, regulators would not have a complete picture of which market participants held a short position and any transparency requirements could easily be circumvented by a migration of sales off market or to derivatives87.
The lack of information on short positions also means that regulators cannot properly monitor short selling transactions for possible market abuse, resulting in the possibility of abusive behaviour not being detected or sanctioned88. Although market participants argue that most short selling is not abusive, and abusive short selling is already prohibited by the Market Abuse Directive, short selling can constitute market abuse in several ways: when it creates misleading signals about the real supply or the correct valuation of a share; when it is used in conjunction with the spreading of false rumours to drive down the price of the share being sold short; or when it is carried out on the basis of inside information89. According to some regulators, there may be a greater risk of market abuse associated with naked short selling90. There is concern also that financial institutions may be particularly vulnerable to abusive short selling, particularly at a time of severe market instability.91
Concern has also been expressed by some regulators (in five Member States) that speculators may be driving down the prices of government bonds by using Credit Default Swaps (CDS)92. The concern of these regulators is that in the absence of information about derivative transactions it is more difficult for them to detect the build-up of positions which could cause financial instability, as well as possible market abuse. A similar comment can be made for sovereign bonds transactions.
For most regulators in Europe, the ease of obtaining access to data on derivatives varies according to the existence or not of a trade repository. In the case of CDS, the information can be obtained from the DTCC. Some regulators (notably in the UK and Spain) collect data on OTC derivatives through daily transaction reporting from market participants in their jurisdiction. They can also address ad hoc requests to market participants in their jurisdiction or via the CESR Memorandum of Understanding for other markets. However, it appears from the contacts that the Commission has had recently with the DTCC that most regulators do not seem until now to have routinely requested transaction reports on CDS. This could be explained by the fact that regulators have only started looking at CDS issues recently and not yet established systematic procedures to use DTCC data. In addition, DTCC has only recently increased the accessibility to its data. Finally, the information stored by DTCC deals with positions and not transactions. There is for instance no time stamp of transactions stored in the warehouse.
Due to the lack of disclosure requirements, there is also a risk of information asymmetries between informed short sellers and other less informed market participants. The disclosure of significant short positions to the market provides information to other market participants about the price movements which short sellers expect and this could improve the efficiency of price discovery if correctly interpreted. Transparency to the market would also ensure that more information about the opinions that investors hold on a particular security would be made available to all investors and issuers93. The disclosure of short selling information to the market also provides useful information as investors are then aware of the extent to which short selling is affecting the price of the security and also of the extent to which short sellers will later have to purchase a similar number of securities to cover their short sale (again bearing in mind that short selling typically involves two transactions – a short sale and a subsequent purchase of the equivalent number of securities).
In summary, there is clear evidence of a lack of transparency of data on significant short positions held by financial institutions, with fewer than half of Member States currently applying disclosure regimes and only one Member State operating a flagging regime. The current lack of transparency in relation to short selling data can also be seen from the difficulty in obtaining reliable data for the EU. This lack of transparency is also one of the main concerns of regulators with regard to short selling, as it prevents them from being able to detect at an early stage the development of positions which may cause risks to financial stability or market integrity. The concern of regulators is so great that they have already agreed on a model for a disclosure regime within CESR. Regulators also take the view that disclosure to the public would deter aggressive short selling, which could reduce the risks of negative price spirals. Issuers would welcome the disclosure of short positions in their share capital as it enables them to be informed about the market view of their performance and prospects. Finally, the fact that in response to concerns about lack of transparency, Member States have had different transparency rules in place and have not to date all implemented CESR's recommended approach to disclosure, results in a fragmented regulatory framework, and consequently higher compliance costs and potential for regulatory arbitrage (see problem 4 below).
3.6.Problem 4: Increased compliance costs and potential regulatory arbitrage arising from a fragmented regulatory framework
The divergent responses of Member States to issues relating to short selling pose the risk of regulatory arbitrage, as investors could seek to circumvent restrictions in one jurisdiction by carrying out transactions in another. This regulatory fragmentation will also lead to increased compliance costs for market participants, especially those operating on several markets, who will have to set up different systems to comply with different requirements in different Member States.
As it has been referred to above, EU Member State regulators adopted divergent approaches to the regulation of short selling in autumn 2008, ranging from temporary emergency bans on short selling of financial stocks to partial bans on naked short selling and to no action at all. Some Member States adopted temporary emergency measures restricting or banning short selling in financial institutions (e.g. Austria, Denmark, France, Germany, Ireland, Italy, Netherlands, UK), whereas others (e.g. Finland, Sweden) adopted no rules on short selling at all in response to the crisis. Of the Member States which restricted or banned short selling in financial institutions temporarily in 2008, some (e.g. Austria, Denmark and France) still have those measures in place today. Furthermore, some Member States banned only naked short selling in certain financial institutions (e.g. Belgium, France, Luxembourg, Portugal) or in all issuers (Greece), while another already had a ban on naked short selling in place (Spain). Some Member States subsequently introduced short selling disclosure requirements with thresholds varying from 0.01% to 0.25% of the total share capital of the issuer and Greece and Poland introduced flagging regimes.
Detailed evidence on the current fragmented regulatory landscape is displayed in Annex 3.
In its consultation of regulators, DG Internal Market asked whether they perceived a risk of regulatory arbitrage arising from the different regulatory responses to short selling in Europe. Of the respondents, six saw possible risks of regulatory arbitrage, 6 did not feel this was a realistic prospect and one thought it would depend on the circumstances. For one regulator, for the regulation of short selling to be effective it must have extra-territorial reach or it could easily be circumvented by those undertaking short selling outside the jurisdiction94.
As pointed out by one regulator in its response to a Commission questionnaire, if a company is listed in different Member States, it is potentially possible to take advantage of the different regulation concerning short selling, and the opportunities for regulatory arbitrage are greater for bonds, in particular sovereign bonds (although this may partly depend on the costs incurred in changing trading venues)95. Another example would be that if one Member State introduces a national ban on short selling in its financial institutions, investors might be tempted to short sell shares in financial institutions of a closely related neighbouring country as a proxy in order to continue to profit from the expected decline in financial shares96.
This regulatory fragmentation will lead to increased compliance costs for market participants, especially those operating on several markets, who would have to set up different systems to comply with different requirements in different Member States.
According to some market participants, in addition to increased costs, the different national responses to short selling since September 2008 led to a significant increase in legal uncertainty due to the lack of harmonisation, and to scope for regulatory arbitrage97.
The fragmented approach at national level also created difficulties and costs for market participants (many of whom operate businesses across Europe) in identifying and interpreting the different requirements, with the exemption of market makers in particular causing difficulties98. It also created significant costs for participants in setting up trading and reporting systems that could comply with the differing requirements99.
In addition to unanimous support of regulators within CESR for an EU-wide short selling disclosure regime, most national regulators have also expressed strong support for a coordinated European approach to the powers of regulators on short selling in exceptional situtations100.
To summarise, the problem of fragmented national rules on short selling is a significant problem at the European level, which cannot be addressed at the national level alone. Action to coordinate national responses is based on evidence of regulatory developments which are expected to affect market developments unfavourably.
3.7.How would the problem evolve without EU action? The Baseline Scenario
As outlined above, a number of Member States have already taken national action to restrict or ban short selling in distressed markets, and one Member State has introduced temporary restrictions on naked CDS. This suggests that even in the absence of EU action, at least some Member States consider that they have the power to act, and have done so. On the other hand, the actions taken by some Member States (e.g. the UK) with regard to short selling restrictions and disclosure requirements in an emergency were based on the Market Abuse Directive, which does not offer sufficient legal certainty and clarity for this purpose, since it does not contain any specific provisions on short selling. In addition, some Member States (e.g. Italy) have indicated that they would prefer a coordinated approach to the regulation of short selling at EU level. The existence of different and changing national rules on short selling would imply additional compliance costs for investors operating in several EU markets, such as legal costs in identifying the different national rules and setting up or updating compliance processes for these.
In addition to the possibility of action by national regulators, many stock exchanges already have the ability to suspend trading in a particular share whose price has fallen by a significant amount (circuit breakers)101, and so it could be argued that no further regulatory intervention is required. However, circuit breakers operated by stock exchanges only very briefly suspend trading in shares which breach the specified threshold. A degree of transparency on levels of short selling in particular instruments can be obtained through data on securities lending, which can be purchased from data providers. However, since short selling is not the only reason for borrowing securities, the use of stock lending as a proxy for levels of short selling presents uncertainties.
Overall, if no action is taken at EU level the problems defined in this section are likely to remain without a coordinated response and to occur again in the future. In particular, in the event of a renewed sudden and repetitive drop in the prices of sovereign bonds and the related CDS market, or of a future financial crisis, Member States are likely to continue to take different approaches, without prior coordination, to exceptional measures to reduce the negative price spirals which can be associated with short selling in these situations. There would be a risk of renewed stock market volatility as occurred following the decision of Germany to restrict naked short selling and naked CDS.
With regard to transparency, some Member States may impose national disclosure obligations as proposed by CESR in the absence of EU legislation; but it is likely to be only a minority of countries since national competent authorities have agreed unanimously in CESR to urge the Commission to include these obligations in an EU legislative framework. Therefore in the absence of EU or national action to require disclosure of short positions to regulators and the market, regulators will continue to lack information on short selling transactions and information asymmetries between short sellers and uninformed investors will persist. These transparency deficiencies imply potential risks for financial stability, as well for investor protection and market integrity.
Although believed to be small in Europe, settlement failures linked to naked short selling would also persist in the absence of measures to strengthen settlement discipline or require short sellers to locate or borrow shares prior to a short sale.
In addition to the above-mentioned problems associated with the lack of a regulatory response at national level, there are also risks of over-regulation at national level. Unduly stringent rules on short selling risk having a negative impact on liquidity and efficient price discovery.
Overall, uncoordinated national responses (or the lack of national actions) in the future are likely to lead to an unlevel playing field and opportunities for regulatory arbitrage, which could hinder the effectiveness of national actions. If one Member State bans short selling of some or all financial instruments on their market, investors could circumvent this ban by short selling the instruments if these are admitted to trading on markets in other Member States not subject to the ban.
Furthermore, the timing of a European measure is also relevant in this case. With the adoption of an early measure at EU level, national responses can be better coordinated, ensuring that market participants only incur compliance costs once. The longer it takes to propose a clear European initiative, the more Member States may be expected to introduce uncoordinated national rules, leading to a potential multiplication of compliance costs for the financial industry in the respective markets.
Finally, in the absence of EU action, the European Union will continue to lag behind other countries, notably the United States, in their regulatory response to the risks associated with short selling.
3.8.Subsidiarity
According to the principle of subsidiarity (Article 5.3 of the TFEU), action on EU level should be taken only when the aims envisaged cannot be achieved sufficiently by Member States alone and can therefore, by reason of the scale or effects of the proposed action, be better achieved by the EU. The preceding analysis has shown that although all the problems outlined above have important implications for each individual Member State, their overall impact can only be fully perceived in a cross-border context. This is because short selling a financial instrument can be carried out wherever that instrument is listed, or over the counter, so even in markets other than the primary market of the issuer concerned. Further, the CDS market is by its very nature a highly cross-border, even international, market. Therefore there is a real risk of national responses to short selling and CDS being circumvented or ineffective in the absence of EU level action.
Further, certain aspects of this issue are already partly covered by the acquis, notably: the Market Abuse Directive which prohibits short selling which is used to manipulate the market or in conjunction with insider information; the Transparency Directive102, which requires the disclosure of significant long positions; and the Markets in Financial Instruments Directive which imposes certain requirements on settlement discipline. Therefore a proposal on short selling and these existing legal instruments should complement each other, and this can best be achieved in a common effort. Against this background EU action appears appropriate in terms of the principle of subsidiarity.
4.Objectives
4.1.General, specific and operational objectives
In light of the analysis of the risks and problems above, the general objectives of the legislative proposal on short selling are to:
1. reduce the risks to financial stability;
2. reduce systemic risks; and
3. reduce risks to market integrity arising from short selling; and
4. prevent market fragmentation, thereby increasing the efficiency of the internal market.
The main general objective, following the weight of the above problems, is to address the problem of market fragmentation. The other three general objectives are based on the principle of risk prevention. However, it should be recalled that the prevention of market abuse is not specifically linked to short selling and this issue is therefore dealt with in the context of the review of the Market Abuse Directive rather than in this initiative.
Reaching these general objectives requires the realisation of the following more specific policy objectives:
1. Reduce the risks of negative price spirals arising from short positions (including those obtained through CDS);
2. Increase the transparency of short positions (including those obtained through CDS);
3. Reduce settlement risk linked with naked short selling; and
4. Reduce the scope for regulatory arbitrage and compliance costs.
The specific objectives listed above require the attainment of the following operational objectives:
1. Ensure regulators have clear power to restrict or ban short selling or CDS in distressed markets;
2. Ensure that regulators and markets obtain data on short positions (including through CDS);
3. Ensure certain requirements are introduced at the point of trading and strengthen settlement discipline;
4. Ensure a coordinated response by EU Member States to short selling and CDS.
An overview of the various objectives and their interrelationships is depicted in the figure below.
4.2.Consistency of the objectives with other EU policies
The identified objectives are coherent with the EU's fundamental goals of promoting a harmonious and sustainable development of economic activities, a high degree of competitiveness, and a high level of consumer protection, which includes safety and economic interests of citizens (Article 169 TFEU).
These objectives are also consistent with the reform programme proposed by the European Commission in its Communication Driving European Recovery.103 More recently in the Commission Communication of 2 June 2010 on 'Regulating Financial Services for Sustainable Growth' the Commission indicated that it would propose appropriate measures relating to short selling and credit default swaps104.
5.Policy options
In order to meet the objectives set out in the previous section, the Commission services have analysed different policy options. The first section reflects the most relevant policy options that have been considered in order to ensure regulators have the power to restrict or ban short selling or CDS in distressed markets. The second section examines the policy options that have been analysed in relation to ensuring regulators and markets obtain data on short positions (including through CDS). The third section studies the policy options considered with regard to ensuring certain requirements at the point of trading and strengthening settlement discipline. Finally, the fourth section considers the choice of instruments available to reduce regulatory arbitrage and compliance costs.
As explained in section 3.7, a number of options are not assessed below as they are being considered in the impact assessments for other related initiatives.
Where necessary, further explanation of the content of the options is included in the assessment of the options. In addition, an overview of the scope of the preferred options (with regard to the financial instruments covered) is included in annex 9.
5.1.Policy options to ensure regulators have the power to restrict or ban short selling or CDS in distressed markets
1. Option 1 – take no action at EU level.
2. Option 2 – introduce a power for national competent authorities to temporarily restrict short selling in a financial instrument admitted to trading on an organised market whose price has fallen by a specified quantitative threshold, e.g. 10% ('circuit breaker').
3. Option 3 – introduce a rule that prohibits short selling of a financial instrument admitted to trading on an organised market except at a price above the last traded price of the instrument, or at the last traded price if that price was higher than the price in the previous trade (an uptick rule).
4. Option 4 - introduce a ban on naked CDS (i.e. entering into a CDS contract without having an underlying insurable interest). This option is compatible with options 2 and 3.
5. Option 5 – grant national competent authorities the power to temporarily restrict or ban short selling of some or all financial instruments or CDS transactions in case of an adverse event or development that creates a threat to financial stability or market confidence ('exceptional situations'), with coordination by ESMA, in accordance with article 6a(5) of Regulation ??/EC establishing ESMA and without prejudice to ESMA's powers under article 10 of the same Regulation. This option is compatible with options 2, 3 and 4.
6. Option 6 – introduce a permanent ban on short selling of all financial instruments capable of being sold short. This option is compatible with option 4.
7. Option 7 - introduce permanent restrictions or ban on CDS. This option is compatible with options 2, 3 and 6.
5.2.Policy options to ensure certain requirements at the point of trading and strengthen settlement discipline
8. Option 1 - take no action at EU level.
9. Option 2 - introduce a requirement that before entering into a short sale, a person must have borrowed the share, entered into an agreement to borrow the share or have other arrangements which ensure that he will be able to borrow the share at the time of settlement (locate rule)
10. Option 3 – introduce EU rules on settlement discipline so that persons engaging in short sales which result in a failure to deliver face appropriate penalties, with buy-in procedures and fines in case of settlement failures
11. Option 4 - introduce a ban on naked short selling
12. Option 5 - exemption for market makers and certain primary market operations. This option is compatible with options 2, 3 and 4.
5.3.Policy options to ensure regulators and markets obtain data on short positions (including through CDS)
5. Option 1 - take no action at EU level.
6. Option 2 - introduce a system of flagging of short sale transactions so that regulators can identify which transactions are long and which are short.
7. Option 3 - notification of short positions to the regulator. This option is compatible with option 2.
8. Option 4 - disclosure of short positions to the public. This option is compatible with options 2 and 3.
9. Option 5 - aggregated disclosure of short positions (i.e. individual short positions of investors are not disclosed). This option is compatible with options 2, 3 and 4.
10. Option 6 - disclosure of individual significant net short positions. This option is compatible with options 2, 3 and 4.
11. Option 7 - exemption from disclosure requirements for market makers and certain primary market operations. This option is compatible with options 2-6.
5.4.Policy options to ensure a coordinated response by EU member states to short selling and CDS
This objective should be met by the above three categories of targeted options. In addition, the choice of legal instrument should also aim to ensure coordinated national responses.
6.Analysis of impacts and choice of preferred options and instruments
This section discusses the advantages and disadvantages of the different policy options against the criteria of their effectiveness in achieving the related objectives (to be specified for each basket of options), and their efficiency in terms of achieving these options for a given level of resources or at least cost.
The options are measured against the above-mentioned pre-defined criteria in the tables below. Each scenario is rated between '---' (very negative), 0 (neutral) and "+++" (very positive). The assessment highlights the policy option which is best placed to reach the related objectives outlined in section 5 and therefore the preferred one.
6.1.Analysis of impacts of policy options
6.1.1.Policy options to ensure regulators have the power to restrict or ban short selling or CDS in distressed markets
These options will be assessed primarily against their effectiveness in achieving the specific objectives of: reducing the risks of negative price spirals arising from short positions, including those obtained through CDS; and reducing the scope for regulatory arbitrage and compliance costs. These policy options will also be assessed on their efficiency in achieving these objectives for a given level of resources or at least cost while avoiding unduly negative effects on market efficiency. However, options will also be assessed against other objectives where appropriate.
6.1.1.1 Option 1 - no action at EU level
As explained in the problem definition, at least some Member States consider that they have the power to restrict or ban short selling in exceptional situations, and have done so. In addition, many stock exchanges already operate circuit breakers, and so it could be argued that no further regulatory intervention is required.
However, this option has shortcomings. It leaves scope for continued risks of negative price spirals if some Member States do not act, also because circuit breakers operated by stock exchanges only very briefly suspend trading in shares. Option 1 also leaves scope for regulatory arbitrage and increased compliance costs, since in the absence of EU action, Member States will continue to take divergent approaches, which could hinder the effectiveness of national actions. Finally, the existence of different and changing national rules on short selling would imply additional compliance costs for investors operating in several EU markets, such as legal costs in identifying the different national rules and setting up or updating compliance processes for these.
6.1.1.2 Option 2 – introduce a circuit breaker
Under this option, which could apply to all financial instruments, a national competent authority would be given the power to temporarily prohibit short selling (until the end of the next trading day) in a financial instrument admitted to trading on an organised market whose price has fallen by a specified quantitative threshold. For shares, this threshold could be 10% as it is in the United States. This threshold should be adapted for other financial instruments in implementing technical standards developed by ESMA and adopted by the Commission, where appropriate in consultation with other concerned regulators (e.g. for commodity derivatives). The regulator would be informed of the price fall by the trading venue, or by electronic market news sources. Prior to implementing the restriction the home competent authority would inform other competent authorities of countries where trading venues admitted to trading the instrument concerned by the restriction, so that they could also implement the circuit breaker. ESMA would also be informed but could not block a competent authority's decision to invoke the circuit breaker. However, if it were invoked repeatedly by several authorities this could be a signal to ESMA that longer lasting and more comprehensive exceptional measures might be required. This option is therefore an intermediate step which could be combined with a ban in exceptional situations (option 5).
This option has some advantages when compared with option 1. First, it would partially achieve the objective of ensuring regulators have the power to restrict short selling in distressed markets, by providing regulators with a short term possibility to suspend short selling on the basis of an objective price test, without needing to determine an exceptional situation. The second advantage is that this would prevent short sellers from carrying out further short transactions until the prohibition expired, which could (at least temporarily) support the price of the instrument and slow or stop a negative price spiral. The third advantage is that the discretion left to the regulator to impose (or not) the prohibition in the event of a significant price fall would allow the regulator to decide not to impose the prohibition if they considered that the price fall was due to factors unrelated to short selling.
However, there are a number of disadvantages to this option. First, there would be continued scope for negative price spirals, as the prohibition of short selling would only be for a very short period and would have to be re-imposed in the event of a further significant price fall. Second, if introduced in isolation this option still leaves room for regulatory arbitrage. By its nature a circuit breaker is designed to halt the decline in the price of an instrument105 admitted to trading on an organised market, and such a measure would not prevent investors from circumventing the prohibition of short selling on organised markets by securing a short position outside the market or through the use of derivatives.
The evidence from published studies on the effectiveness of circuit breakers is mixed. On the negative side, according to some studies, circuit breakers may cause volatility to be spread out over a long period of time instead of occurring in a one day jump, or may interfere with market liquidity106. On the positive side, some studies take the view that circuit breakers provide a cooling-off period for investors107.
The views of respondents to the Commission's public consultation on the circuit breaker option were mixed, and a large number of respondents did not provide comment on this area. The majority of those who did noted that it could be a useful tool to slow a price fall amplified by short selling, but that similar tools were already in place on exchanges and markets across the EU. No data was provided on the compliance costs linked to the circuit breaker option.
6.1.1.3 Option 3 – introduce an uptick rule
The advantages of option 3 compared to option 1 are as follows. It would address the first shortcoming of option 1, as a short sale transaction could only be entered at a higher price than the last trade price, which would limit the scope for negative price spirals to occur, without stopping short selling in a flat or advancing market. Also, unlike option 2, the prohibition of short selling is permanent, not temporary, so long as the short sale transaction is at a lower price than the last trade price. It would partially address the second shortcoming of option 1, as a common uptick rule for all EU Member States would give a greater degree of certainty on the regulatory regime for market participants.
However, this option also has several disadvantages. The tick rule has been criticised by some stakeholders for being ineffective as it only temporarily decelerates price declines108. In the United States, which operated a tick rule for nearly 70 years, this rule was abolished by the SEC in 2007 on the grounds that it modestly reduced liquidity and did not appear necessary to prevent manipulation109. Furthermore, like option 2, if introduced on its own a tick rule would leave scope for regulatory arbitrage through the use of derivatives to secure an economic short position, as a tick rule only applies to short transactions on cash markets.
Most empirical studies are critical of the effectiveness of uptick rules, arguing that they were ineffective in preventing price declines or that the price behaviour of securities subject to an uptick rule was not substantially different to those not subject to such a rule.110
6.1.1.4 Option 4 – ban naked CDS
This option would permanently prohibit investors from entering into a CDS contract unless they held an underlying insurable interest. This option only concerns CDS contracts without an underlying insurable interest, so it has a narrower scope than a short selling ban which would prohibit investors from selling short any financial instrument.
Option 4 has the advantage over option 1 that it may help to reduce the risk of negative price spirals in the bond markets linked to economic short positions obtained through naked CDS, thereby partially fulfilling the first specific policy objective. A common approach to banning naked CDS would also partially address the second shortcoming of option 1 (regulatory arbitrage and compliance costs) by partially harmonising the approaches of Member States on the CDS aspect.
However, option 4 has a number of potential risks and disadvantages. According to some academics, prohibiting naked positions in credit default swaps could dramatically impact the market, because if the CDS market is reduced to hedgers only, market liquidity is likely to drop substantially 111. There is also the risk that regulation of the CDS market could also have some side effects on the bond market. Under a naked CDS ban, CDS might become more like classical insurance devices, i.e. customised to closely-related exposure. This would reduce the market's ability to trade credit risk and make proxy-hedging112 impossible. As a result, the cost of bond market financing for the broader economy could increase.
Another potential disadvantage of a ban on naked CDS is that there is a question mark as to whether this would be effective in achieving the objective of reducing the risk of negative price spirals in the bond market. According to some market participants and regulators, there is no clear evidence that trading on the CDS markets can distort or manipulate the underlying bond market, although some regulators argue that there is a risk that trading on the CDS market can lead to instability on the underlying bond market.113
Moreover, concerns about the possible damaging effects of a ban on naked CDS have to be taken with care as there is no certainty about the positive impact of CDS on the accuracy of pricing on the bond markets. Empirically, this issue has been studied for corporate debt by Ashcraft and Santos (2009) with mixed findings.114 The authors do not find evidence that the average firm on which CDS are traded has benefited from a reduction in the credit spreads that it pays to issue in the bond market or to borrow from banks. However, there is evidence that the onset of CDS trading has increased the cost of debt financing for the riskier firms as well as those that are more informationally opaque. In contrast, safer and more transparent firms experience a small reduction in the spreads after their CDS started to trade.
For sovereign CDS, no similar study exists. As the assessment of sovereign creditworthiness by and large rests on public information, reduced screening incentives through CDS should play only a minor role. However, rather opaque public accounts have been important in the recent sovereign debt crisis.
Finally, this option has the disadvantage that if it were to be adopted in isolation there would be a risk of regulatory arbitrage, as there are other strategies than naked CDS to bet on a downturn on sovereign or corporate risk which would also need to be addressed in order for any ban to be effective: selling a future on the bond, buying a put option, selling a call option, or short selling the bond.
Overall, there are very varied views on how the bond market would be affected by a ban on naked CDS.
Most respondents to the public consultation who addressed this option in their contributions opposed a ban on naked CDS, arguing that there was insufficient evidence to suggest naked CDS positions had been used in an abuse manor. On the contrary, a number of financial and non financial institutions (e.g. manufacturers) cited the use of proxy-hedging as critical to the effective management of financial and non financial risks.
6.1.1.5 Option 5 – power for national regulators to temporarily restrict or ban short selling or CDS in exceptional situations (coordinated by ESMA)
Under this option, national competent authorities would, in case of an adverse event or development that creates a threat to financial stability or market confidence ('exceptional situations'), be granted a clear power to temporarily restrict (e.g. by banning naked short selling or naked CDS) or ban short selling or limit other similar transactions. A national competent authority could invoke these powers of intervention if it considered it necessary in the event of adverse developments which constituted a serious threat to financial stability or to market confidence in the Member State, another Member State or the Union. The scope of the power would be broad, covering any financial instrument, to provide flexibility to regulators to respond to an exceptional situation. The Commission would further specify the criteria for determining an exceptional situation by means of a delegated act, which gives the European Parliament and the Council the possibility to object and prevent it from entering into force.
To ensure coordination, ESMA would have to be notified not less than 24 hours before (unless there were exceptional circumstances) and ESMA would then issue an opinion. In this opinion, ESMA would state whether it considers that adverse developments have arisen which constitute a serious threat to financial stability or to market confidence in a Member State or the Union, whether the measures are appropriate and proportionate to address the threat and whether the proposed duration of the measures are justified. Where ESMA concludes that the action is justified, it would also state whether action by other competent authorities is necessary. The opinion would be published on ESMA's website. If a national regulator then wished to continue with action contrary to ESMA's opinion, it would have to publicly explain its decision. The short notification period and the possibility for Member States to give less notice in exceptional circumstances would ensure coordination while enabling competent authorities to act quickly in fast-moving situations.
Consistent with an amendment to the ESMA Regulation adopted in first reading by the European Parliament115, ESMA itself would be given the power of intervention on short selling where an exceptional situation arose in a Member State or the Union which had cross-border implications and a competent authority or authorities had not already taken action, or taken sufficient action. If the threat was particularly serious, ESMA could intervene without first issuing an opinion. The same notification requirements would apply to ESMA. Restrictions in an exceptional situation could be applied for up to 3 months and, if justified, could be renewed for further periods of the same duration, and ESMA would ensure that the renewal would not exceed the period which was justified and proportionate.
This option has a number of advantages. First, by providing regulators with the power to act in distressed markets it would also achieve the first specific objective of reducing the risk of negative price spirals. It would be more effective than options 2, 3 and 4 in fulfilling the first operational objective, since it would ensure that regulators would be granted the power to restrict or ban short selling or CDS in exceptional situations. Second, since the power to restrict or ban short selling or CDS would be confined to such exceptional situations, any possible negative impacts on market efficiency of such a ban would be limited to the duration of the threat to financial stability or market confidence, which would contribute to achieving general objective 4 (prevent market fragmentation/enhance efficiency of internal market).
Third, if ESMA were to be granted the power to ensure that the actions of regulators in exceptional situations were coordinated and consistent, this would meet operational objective 4 and reduce the scope for regulatory arbitrage (specific objective 4). This option would increase the effectiveness of national actions, since regulators would be required to cooperate with each other to enforce the measures. It would also reduce compliance costs as market participants would have to comply with coordinated measures across the EU in an exceptional situation, rather than a patchwork of different national responses.
The first disadvantage of this option is the potential for negative impacts on market efficiency. A study by Beber and Pagano which analysed short selling bans around the world between 2007 and 2009 concluded that bans and constraints on short selling were 'detrimental for market liquidity' and may not have the intended effect of supporting market prices116. A temporary ban could also have negative implications for efficient price discovery: an analysis of short selling data and restrictions in 46 markets across the world by Bris et al concludes that 'markets where short sales are allowed are more efficient because bad news appears to be more rapidly impounded into prices.'117 However, these impacts would only last for the duration of any temporary restrictions or bans on short selling or CDS, and have to be balanced against the benefits which such restrictions could bring in terms of reducing the risk of negative price spirals in distressed markets.
The second disadvantage is the cost for market participants of compliance with a ban, which would include, for example, legal advice and adjustments of IT systems to prevent short selling of the instruments concerned. The FSA conducted a cost benefit analysis of its temporary ban on short selling of shares in UK financial institutions in 2008, and estimated the one-off compliance costs to be GBP 40,000 per firm and the ongoing costs to be GBP 6,500 per firm per month.118 On the other hand, ongoing costs would only last for as long as the ban was in place, and these costs also have to be weighed against the reduced risk of negative price spirals which this option could bring.
The other main disadvantages of this option is, that if ESMA were not in practice to have the ability to ensure real and effective coordination and consistency in national measures taken in exceptional situations, there would be a risk of continued regulatory arbitrage and higher compliance costs for market participants, similar to the second shortcoming of option 1. However, since the coordination powers foreseen for ESMA under this option are extensive and would be without prejudice to its powers under article 10 (emergency powers) of the ESMA Regulation, this risk is limited as ESMA could always act itself or invoke article 10 to ensure that national regulators took the same action in an emergency.
Most responses to the public consultation favoured this option as they felt limiting restrictions to exceptional situations would be a strong signal to the market and act as a damper to mitigate speculation. There was also support from market participants for the clear definition of an exceptional situation, which would help maintain market clarity and confidence. On the other hand, public authorities mostly argued for flexibility in determining an exceptional situation.
6.1.1.6 Option 6 – permanent ban on short selling
This option would mean that investors would be banned permanently from short selling financial instruments, either through short selling on cash markets or through the use of derivatives to secure a net short position.
The main advantage of a permanent ban on short selling would be that it would eliminate the potential for the risk that short selling may amplify negative price spirals in distressed markets, as well as the risk of settlement failures linked to naked short selling. If short selling were to be banned, there would not be a need for a disclosure regime for short positions, and a total EU-wide ban would rule out the possibility of regulatory arbitrage in Europe and additional compliance costs caused by differing national rules.
However, there would clearly be major costs associated with this option. The economic costs (in terms of market liquidity and price efficiency) outlined for option 5 would be even greater under this option, as the ban would be permanent, not temporary. Similarly, the ongoing compliance costs would be greater than for option 5 as they would not be limited in time (although one-off costs would only have to be incurred once, whereas in a temporary regime these would recur with each new restriction). This option would therefore not meet the objective of avoiding unduly negative effects on market efficiency.
Further, since most other countries do not have a permanent ban on short selling in place, there would be a risk that such a measure might lead to challenges before the World Trade Organisation, or to economic retaliation by other countries opposed to this approach. There would also be a risk that the attractiveness of Europe as a base for financial institutions would be reduced, and that over time they might opt to locate themselves in jurisdictions which gave them greater freedom in their investment strategies119.
Only two of the 116 responses to the public consultation supported a full ban on short selling.120
6.1.1.7 Option 7 – permanent restrictions or ban on CDS
The advantage of this option is that it would eliminate the potential for CDS to be used to secure an economic short position in the underlying reference instruments, and would therefore further reduce the scope for negative price spirals in those instruments. Like option 6 it would eliminate the need for a transparency regime for economic short positions obtained through CDS.
However, this option has similar disadvantages to option 6. There would be considerable economic costs to this measure, as institutions would no longer be able to hedge credit risk through CDS or take positions for a profit. The ongoing compliance costs would be greater than for option 5 as they would not be limited in time. Since the CDS market is a highly global one, a ban could easily be circumvented by investors trading in CDS in non-EU countries not subject to the ban and there would be very little that EU regulators could do about this.
6.1.1.8 The preferred options
Impact on stakeholders | Effectiveness | Efficiency | |
Option 1 (baseline) | n.a. | n.a. | n.a. |
Option 2 (circuit breaker) | (++) regulators obtain possibility to temporarily ban short selling and have discretion not to ban (++) issuers gain a mechanism to temporarily support their share price when it faces a significant decline (-) financial institutions engaging in short selling are briefly restricted from doing so and face compliance costs | (++) partially achieves specific objective 1, but prohibition only for a short period (-) regulatory arbitrage: could be circumvented by derivatives | (-) compliance costs for investors; any effect on liquidity would be temporary |
Option 3 (uptick rule) | (+) regulators have an ongoing automatic mechanism (+) issuers gain an ongoing, automatic mechanism to support their share price (- - ) financial institutions face compliance costs and are permanently restricted form short selling | (+) partially achieves operational objective 1 as it restricts short selling in a falling marke (-) could be circumvented by derivatives | (- --) compliance costs; could reduce market liquidity |
Option 4 (ban on naked CDS) | (+) regulators gain a tool to eliminate risk of negative price spirals on bond markets linked to naked CDS (0) governments benefit from possible reduced volatility on sovereign bond markets, but risk of negative side effects on liquidity (- - -) financial institutions lose a way of taking negative positions on bond markets through CDS | (-) partially achieves operational objective 1 for bond markets, but effectiveness questionable. Could be circumvented by using other derivatives or trading CDS outside EU | (- - -) could substantially reduce liquidity of CDS market and could have negative side effects on bond markets |
Option 5 (power for national regulators to ban short selling/CDS in exceptional situations, subject to ESMA coordination) | (+++) ensures regulators gain powers to ban short selling/CDS in exceptional situations (++) issuers gain comfort of knowing that their share price can be supported by a temporary ban on short selling in distressed markets (0) governments benefit from possible reduced volatility on sovereign bond markets, but risk of negative side effects on liquidity (- -) financial institutions engaging in short selling may be prevented from doing so in exceptional situations for several months | (+++) achieves operational objective 1 fully and reduces risk of negative price spirals arising from short positions; ESMA coordination eliminates risk of regulatory arbitrage | (-) reduced compliance costs compared to other options due to coordinated EU approach. Temporary restrictions in exceptional situations avoid unduly negative impacts on market efficiency, although temporary restrictions on CDS could have negative side effects on bond markets. |
Option 6 (permanent ban on short selling) | (-) regulators lose discretion to only ban short selling in exceptional situations (++) share prices of issuers would be supported, but this could lead to them being artificially inflated (- - -) financial institutions prevented from short selling permanently | (+) partially achieves operational objective 1, since ban would not be discretionary and restricted to distressed markets. Rules out regulatory arbitrage. | (- - -)Permanent costs in terms of reduced market liquidity and efficient price discovery. Ongoing compliance costs. Risk of firms relocating. |
Option 7 (permanent ban on CDS) | (-) regulators lose discretion to only ban CDS in exceptional situations (0) governments benefit from possible reduced volatility on sovereign bond markets, but risk of negative side effects on liquidity | (+) partially achieves operational objective 1, since ban would not be discretionary and restricted to distressed markets. Could be circumvented by using other derivatives or trading CDS outside EU | (- - -) substantial costs as firms lose ability to hedge credit risk and could have negative side effects on sovereign bond markets |
Based on the analysis above, the highest scoring option is option 5, granting a power for national regulators to temporarily restrict or ban short selling or CDS in exceptional situations (coordinated by ESMA). However, option 2 also scores well and is compatible with option 5. In fact, if option 2 were to be combined with option 5, this would address the shortcomings of option 2: that short positions using derivatives are not captured, and that the halt to short selling in the event of is only a very short term one (until the end of the next trading day). A combination of the two options would give regulators an instrument to impose a short term ban on short selling on organised markets in the event of a significant price decline, as an intermediate step based on an objective price test, as well as the possibility to impose a temporary ban of a longer duration, capturing derivatives as well, in the event of an adverse event or development which posed a threat to financial stability or market confidence. So the preferred option is a combination of options 2 and 5.
With regard to the circuit breaker, the scope should cover all financial instruments, but the threshold of 10% would apply only for shares, whereas thresholds for other financial instruments would be adapted to each instrument in implementing technical standards adopted by the Commission on the advice of ESMA. With regard to the broader powers of regulators in exceptional situations the scope should extend to all financial instruments that may be sold short, as well as CDS, to provide the necessary flexibility for regulators in unforeseen situations.
6.1.2.Policy options to ensure certain requirements at the point of trading and strengthen settlement discipline
These options will be assessed against their effectiveness in achieving the specific objectives of reducing settlement risk linked with naked short selling, and reducing the scope for regulatory arbitrage and compliance costs. These policy options will also be assessed on their efficiency in achieving these objectives for a given level of resources or at least cost while avoiding unduly negative effects on market efficiency. However, options will also be assessed against other objectives where appropriate.
6.1.2.1 Option 1 – no action at EU level
As explained in the problem definition, the available data suggests that levels of settlement failures in Europe are relatively low, perhaps as a result of the existing obligations imposed on market operators by MiFID or national rules to strengthen settlement discipline. However, a number of Member States have imposed temporary or permanent restrictions on naked short selling because they perceive risks of settlement failure and price volatility, and in the absence of EU level action, these risks will persist. In addition, the current diversity of national responses to the issue would also continue, implying scope for regulatory arbitrage and compliance costs for financial institutions operating on several EU markets.
6.1.2.2 Option 2 - introduce a locate rule
This option would involve imposing a requirement that before entering into a short sale, an investor would have to either have borrowed the security, entered into an agreement to borrow it or have other arrangements which ensure that he will be able to borrow the security at the time of settlement. This is often known as a locate rule, as prior to carrying out a short sale transaction, the investor has to either borrow the securities or locate them with a view to borrowing them before settlement.
A locate rule has several advantages compared to the no EU level action option. First, it would meet the objective of reducing the risks of settlement failures which naked short selling may cause, leaving other causal factors (e.g. administrative errors) to be dealt with by other means (regulatory or non-regulatory). Second, a locate rule would slow down the speed with which a short seller could execute his strategy, as he would have to first borrow or locate shares to borrow entering into a short sale. More importantly, since there is always a limit to the quantity of securities available for lending, a locate rule would also place a limit on the volume of short selling of securities. Aggressive short term strategies which depend on placing a high volume of short sale orders very quickly would therefore become much more difficult to carry out. This would have the additional advantage of contributing to meeting the objective of reducing the risk of negative price spirals arising from short positions.
Third, introducing a harmonised locate rule at EU level would eliminate the scope for regulatory arbitrage as investors could no longer execute short sales in jurisdictions with little or no restrictions on naked short selling. It would also reduce compliance costs relative to the cost of the very different national requirements which currently exist for naked short selling, as institutions would only have to comply with one common obligation EU-wide.
A potential disadvantage with this option is that it could impose compliance costs, including legal costs and the cost of IT systems to ensure that sort sale transactions could only be executed once securities had been borrowed or located. However, these costs would be limited since most market participants said in their responses to a questionnaire by the Commission services that they already use arrangements to ensure securities are located prior to entering into a short sale, and this option foresees that such arrangements could continue to be used. The public consultation has not provided data on the compliance costs associated with existing restrictions on naked short selling.
A disadvantage of a locate rule would be that it could impose indirect economic costs on institutions which currently engage in naked short selling. These costs in terms of reduced market liquidity and efficient price discovery would likely be lower than those for a ban on short selling, but some impact could still be expected. However, the evidence gathered from the bilateral and public consultations carried out by the Commission suggests that the prevalence of naked short selling, without any attempt to locate securities for borrowing, is quite limited.
Another significant disadvantage would be if the rule did not contain appropriate exemption for firms engaged in market making activity which provide significant liquidity to the market and due to the nature of their business need to take numerous naked short positions throughout the trading day. Requiring such participants to enter into agreements for the significant number of trades they enter into would significantly impair their ability to provide liquidity to the market. This would have a significant detrimental effect on markets.
In summary, a locate rule could be considered to be a proportionate response to the risk of settlement failure, as the costs involved are likely to be limited since it would be building on an existing market practice already used by many market participants, and an exemption could be foreseen for market making activities so as to minimise any negative impact on the liquidity they provide. A harmonised EU requirement on a locate rule would also reduce the compliance costs for market participants relative to the different approaches currently followed by Member States in this respect.
6.1.2.3 Option 3 – introduce EU rules on settlement discipline
Under this option, trading venues or central counterparties would be required to have rules in place so that if a short seller was unable to deliver securities by the settlement date, procedures would be triggered to buy in those securities for settlement after a specified period (which could be T+4), at the expense of the short seller. Further under this option rules could require trading venues or settlement systems to have in place appropriate measures to fine participants who do not settle by the due date. Since the causes of settlement failures are broader than just naked short selling, under this option it is proposed that there would be only such level of harmonisation as is necessary to achieve the relevant objectives on short selling. More detailed harmonisation of issues such as settlement periods, buy in procedures and fining regimes would be addressed in the forthcoming European Commission initiative on securities law, which will also be subject to an impact assessment.
The advantage of option 3 is that it would reduce the risk of settlement failures linked with naked short selling by providing for buy-in procedures or fines in that situation, thereby meeting objective 3. However, unlike option 2, it would not in itself limit the speed with which investors could short sell, or the volume of short selling, as it would not introduce any requirements at the point of trading. The discipline on investors would rather arise from the knowledge that if they did not have the securities to settle a short sale transaction, they would face the costs of the buy-in procedures and fines that would be triggered. This would be a significant commercial deterrent against persons entering into short sales without having in place adequate arrangements to ensure they can settle the transaction.
As with option 2, option 3 would also reduce the scope for regulatory arbitrage and compliance costs arising from different national rules on settlement failures linked with naked short selling.
In terms of compliance costs, as many trading venues already have in place differing forms of buy in or fining arrangements for late settlement, some level of harmonisation as foreseen by option 3 should not impose significant additional costs. The public consultation provided no data on possible costs associated with this option. .
6.1.2.4 Option 4 – ban on naked short selling
This option would entail imposing a requirement that securities could not be sold short unless they had been borrowed and were held by the seller before he entered into the short sale (a pre-borrow requirement).
A ban on naked short selling has similar advantages to a locate rule: it would reduce the risk of settlement failures linked to naked short selling to an even greater extent (although the evidence from Spain, which has such a ban, is that it would not eliminate settlement failures entirely). It would also restrict to a greater extent the ability of short sellers to short sell quickly, as their ability to do so would be conditional on borrowing and actually obtaining the borrowed securities, without the flexibility offered by a locate rule to actually borrow the securities after the short sale but before settlement. A harmonised ban would also eliminate the scope for regulatory arbitrage.
However, the costs associated with a ban on naked short selling would be greater than those linked with a locate rule. Investors would face legal costs and would have to put in place systems to ensure that short sales were only entered into when the securities had been borrowed.
The indirect economic costs of this option are also likely to be greater than for the locate rule. According to the economic literature, a ban on naked short selling would have similar effects in terms of reducing liquidity and efficient price discovery as a ban on short selling.
As with a locate rule, if there were to be no exemption from a ban on naked short selling for market making activity, this would significantly impair their ability to provide liquidity to the market.
6.1.2.5. Option 5 – exemption for market making activity and certain primary market operations
There are a number of important and established market practices which are necessary for the efficient functioning of markets and where participants need to take covered or uncovered short positions to conduct such activities. These activities are market making activities and specific primary market operations. This option would exempt these activities from certain requirements for uncovered short selling (the locate rule), but not from the requirements for settlement discipline (although these would be adapted for market making activities and primary market operations).
Under this option, market making activity would be defined as the provision by financial institutions of liquidity on a regular and ongoing basis to the market by posting firm, simultaneous two-way quotations of financial instruments at competitive prices, or by fulfilling orders initiated by clients or in response to clients' requests to trade, and to hedge positions arising out of those dealings.
The use of this exemption would be subject to a number of conditions to ensure that it is not abused. Under this option, only the market making activities of financial institutions who meet the criteria of this definition and who have notified the regulator 30 days beforehand of their intention to make use of the market maker exemption, would be exempt from the requirement on naked short selling considered in this section. The competent authority that receives a notification from a person wishing to use the exemption will be able to refuse to allow the person to use it if they believe the person does not meet the criteria for using the exemption. The competent authority will also be able to request further information from persons using the exemption to ascertain if they are complying with all conditions. The proprietary trading activities of financial institutions which also engaged in market making would not be exempt.
Similarly, primary market dealers who have signed an agreement with an issuer of sovereign debt in order to provide assistance to the sovereign issuer for primary and secondary market operations relating to its sovereign debt would also benefit from an exemption from the requirements on naked short selling for these primary market operations. The exemption would also apply to primary market operations by a person for the purposes of stabilisation under the Market Abuse Directive. The same conditions would apply as outlined above.
The option is consistent with principle 4 of the International Organisation of Securities Commission's (IOSCO) principles on short selling, which recommends that short selling regulation should cater for market transactions that are desirable for efficient market functioning and development. According to IOSCO, activities that fall under this category include bona fide hedging, market making and arbitrage activities; in IOSCO's view these activities generally provide benefits to the market and are unlikely to pose risks that will destabilise the market121.
This option is also in line with the advice of CESR and most responses from market participants to the consultation, who consider that market makers and primary market dealers fulfil an important and legitimate function in providing liquidity, and this liquidity would be affected negatively by applying disclosure and other obligations to these activities.
The rationale for this exemption is that when acting as market makers or performing primary market operations, these participants will continually be required to take both long and short positions in a large number of financial instruments in order to execute market making transactions to meet the demand from market users or to perform their primary market operations. Their positions are likely to change continuously for each financial instrument and fluctuate between long positions to short positions. It is therefore not feasible to expect such activities to enter into specific agreements for the large number of transactions they undertake. To apply such a requirement would make it impossible for these activities to be provided and could have significant detrimental effect on markets including other investors and issuers.
It is difficult to provide detailed estimates of market making activity as increasingly traditional market makers are using new automated trading methods for their activities. Also it will vary from trading venue to trading venue. An estimate by one exchange puts the level of market making at 20-25% for the most liquid securities122. But this is likely to include some trading that will not qualify for the detailed definition of market making activities in the proposal. Therefore the actual amount that is exempted is likely to be significantly less. Further, this is an estimate for the most liquid shares, figures for less liquid shares may also be lower.
This option has the advantage that the objective of reducing the risk of settlement failure linked with naked short selling would be met, while avoiding the possible negative impact on liquidity that full application of the proposed requirements for naked short selling could have on activities that are essential to the efficient functioning of markets. In particular, issuers would continue to obtain the benefits of liquidity provision for their financial instruments provided by market makers and primary market dealers, and financial institutions providing this service would be able to continue to do so. If appropriate exemptions or adaptations for market making activities from the above-mentioned obligations relating to naked short selling were not provided for, this would have a significant adverse effect on their ability to provide liquidity to the markets. This would reduce liquidity significantly, resulting in wider spreads and affecting the prices and the ability of investors to purchase securities. This would in turn have a detrimental effect on issuers of the securities. Choosing this option would therefore greatly contribute to ensuring the proportionality of the measures concerning naked short selling.
Similarly for certain primary market operations participants need to take a short position (for example for stabilisation programmes and the issue of sovereign bonds). Failure to provide appropriate exemptions or adaptations would result in direct and indirect costs for issuers who wish to raise funds through the issue of securities.
Another advantage of this option is that objective 4 would also be achieved, as a harmonised exemption for market making, based on a common definition, would eliminate the scope for regulatory arbitrage as all Member States would have the same definition. This would also entail lower compliance costs than the current situation where different Member States apply different regimes for market makers.
This option has the disadvantage that theoretically, uncovered short sales carried out for market making activities may very occasionally result in failures to settle. However, this is unlikely in practice as most short positions for market making are held very briefly and the objective of market makers is for these positions to be flat at the end of the day. Further, this theoretical risk would be mitigated by excluding market making activities and primary market operations from the conditions on naked short selling, while subjecting them to tailored rules for settlement discipline. For example, in the United States market makers are subject to buy-in procedures if a trade has not been settled 6 days after the trade (T+6), a more lenient period than the T+4 foreseen for other market participants.
The other potential disadvantage of this option is that some financial institutions might seek to use this exemption to circumvent the requirements on uncovered short selling for activities which do not in reality constitute market making. However, this risk is very limited as only genuine market making activities will be exempted; the risk can be mitigated by using the harmonised definition of market making activities recommended by CESR and by providing for the controls on the use of this exemption outlined above. In particular, those wishing to benefit from the exemption would have to notify the competent authority in writing not less than 30 days in advance of their intention to use the exemption, and the competent authority would have the power to obtain information from any institution benefiting from the exemption.
6.1.2.6 The preferred options
Impact on stakeholders | Effectiveness | Efficiency | |
Option 1 (baseline) | n.a. | n.a. | n.a. |
Option 2 (locate rule) | (+++) regulators will be able to sanction naked short selling which does not comply with locate rule (+++) issuers gain comfort that number of shares sold short cannot exceed the number issued or available to borrow (+++) governments gain comfort that number of government bonds sold short cannot exceed the number issued or available to borrow (- ) some financial institutions may have to adapt their compliance systems, although most operate with a locate rule currently | (++) Objective of reducing settlement risk achieved in part by rules at the point of trading (+++) Objective 4: met in full (+) Contributes to reducing risk of negative price spirals | (-) limited impact in terms of ongoing compliance costs as many already operate such a rule; some impact on liquidity and efficient price discovery |
Option 3 (introduce rules on settlement discipline) | (+++) regulators gain comfort that settlement failures will be penalised (+++) issuers also obtain reassurance that settlement failures will be penalised (+++) governments – same benefit as for issuers, concerning bonds (-) financial institutions may face penalties or costs due to buy-in procedures but prefer this option | (++) Objective of reducing settlement risk achieved in part by strengthening settlement discipline (+++) Objective 4 : met in full (+) Contributes to reducing risk of negative price spirals | (0 ) limited impact in terms of compliance costs; |
Option 4 (ban on naked short selling) | (+++) regulators gain comfort that settlement failures due to naked short selling will be greatly reduced (+++) issuers also obtain reassurance that settlement failures due to naked short selling will be greatly reduced (+++) governments – same benefit as for issuers, concerning bonds (- - -) financial institutions' ability to short sell significantly impaired, imposing economic costs and ongoing compliance costs | (++) Objective of reducing settlement risk achieved in part but settlement failures caused by other factors will persists (+++) Objective 4: met in full (+) Contributes to reducing risk of negative price spirals | (- - - ) significant negative impact on liquidity and efficient price discovery and economic/compliance costs for financial institutions |
Option 5 (exemption for market making) | (-) regulators may have to ensure the exemption is not abused (+++) issuers and (+++) governments gain comfort that liquidity in their shares or sovereign bonds is not impaired (+++) financial institutions: their market making activities and primary market operations would be exempt | (0) Objective of reducing settlement risk unaffected as short positions held by market makers are briefly held, and market makers and primary market dealers would be subject to settlement discipline (+++) Objective 4 met in full as a harmonised definition of a market maker/primary market dealer exemption would avoid regulatory arbitrage and limit compliance costs | (+) compliance costs would be limited by harmonised exemption (+++) liquidity provided by market makers and primary market dealers would be unaffected |
Based on the analysis above, the highest scoring options in terms of effectiveness and efficiency are options 2 and 3, while option 5 scores highly in terms of efficiency. These three options are compatible with each other and so could be combined. If options 2 and 3 were combined, settlement discipline would be reinforced both by requirements at the point of trading and by buy-in procedures and fines, thereby meeting very effectively the related operational objective. By combining option 5 with options 2 and 3, the potential negative impact on liquidity would be mitigated by a harmonised exemption for market makers, and so would the potential for regulatory arbitrage and compliance costs associated with different exemptions across the EU. Option 5 would also contribute greatly to ensuring the proportionality of the overall approach.
So the preferred option is a combination of options 2, 3 and 5.
In terms of scope, these options should apply to shares and sovereign bonds only, consistent with the evidence of the problems raised and the positions expressed in the public consultation. However, market makers and primary market dealers should not be exempt from rules on settlement discipline; these could be tailored to their situation, for example by providing for buy-in procedures for market makers at T+6 as is foreseen in the United States.
6.1.3.Policy options to ensure regulators and markets obtain data on short positions (including through CDS)
These options will be assessed against their effectiveness in achieving the specific objectives of ensuring that regulators and markets obtain data on short positions, including through CDS, and reducing the scope for regulatory arbitrage and compliance costs. These policy options will also be assessed on their efficiency in achieving these objectives for a given level of resources or at least cost while avoiding unduly negative effects on market efficiency. However, options will also be assessed against other objectives where appropriate.
6.1.3.1. Option 1 – no action at EU level
Some Member States have already introduced disclosure obligations in the absence of action at EU level, and some others may impose them based on the CESR recommendations, but as explained in the baseline scenario this is likely to be a minority. Therefore this option has the disadvantage that regulators and markets in a number of Member States may continue to lack information on short positions. These transparency deficiencies imply potential risks for financial stability, as well as for investor protection and market integrity, because regulators will not be able to detect the build-up of short positions which may pose risks to orderly markets or be used for market abuse. This option will also mean the continuation of divergent national approaches to disclosure, which leaves scope for regulatory arbitrage and implies higher compliance costs for market participants operating in several markets subject to different rules.
6.1.3.2. Option 2 – flagging
Under this option, a share trading venue would be required to establish procedures to ensure that persons entering short sale orders on the trading venue would have to mark those orders as short orders. The trading venue would then publish a daily summary of the volume of short orders.
This option has several advantages over option 1. It would provide regulators with a tool to receive very precise information about the volume of short transactions relative to long transactions. This would be very effective in achieving specific objective 2, as it would be very transparent. This option also has advantages over a disclosure regime: importantly, according to CESR it 'has the potential to give regulatory authorities real-time data on short selling, including intraday positions, and is more suitable for micro-supervision of transactions'123. In addition, flagging could complement a disclosure regime by providing regulators with data on short positions which remain below any disclosure threshold, and would help regulators to enforce disclosure rules by providing them with a full picture of the short transactions of investors. By introducing a harmonised flagging regime, this would reduce the scope for regulatory arbitrage, although if implemented in isolation option 2 would still leave some scope for regulatory arbitrage, as short positions obtained through OTC derivatives would not be captured by a flagging regime. However, as CESR has recognised, it is possible to use flagging in combination with disclosure124. Another advantage of this option is that the EU regulatory framework would further converge with those of the United States and Hong Kong, which both operate flagging regimes.
The disadvantage of option 2 is that the costs associated with it are considered by market participants and some regulators to be significant. Compliance costs would include the costs of setting up the necessary IT systems, as well as training and compliance procedures. According to estimates received by the UK FSA, the cost of a flagging regime could be in a range of several hundreds of thousands of pounds to around 2 million pounds for a larger broker, to 20,000 pounds in system costs for a small broker.125 No detailed estimates were received as part of the public consultation, although some responses argued that the costs would be substantial. Several respondents, including one exchange, stated that introducing a flagging regime would require significant changes of the complete European trading infrastructure. At the same time, Greece and Poland operate a flagging system at present and the French regulator argued in its response to the public consultation that the benefits in terms of increased transparency would outweigh the costs in terms of compliance.
6.1.3.3 Option 3 – notification of significant net short positions to the regulator
Under this option, short positions would be notified only to the regulator, not to the market. To ensure that only significant short positions were notified, a threshold would be introduced – CESR has proposed a threshold of 0.2% of issued share capital for notification to regulators126. The related option of whether information on short positions should be notified in aggregate form is considered under option 5, whereas option 6 considers the alternative, which is the notification of individual significant net short positions.
Option 3 has several advantages compared to option 1. It is very effective as it meets fully the objective of increasing transparency for regulators in short positions, and is an improvement on option 2 as it also includes those short positions obtained through CDS, which is to the advantage of governments seeking transparency on sovereign CDS. It also achieves the objective of reducing the scope for regulatory arbitrage by setting a common EU-wide standard for notification of short positions to regulators. This option would therefore also entail lower compliance costs for financial institutions operating in several markets, when compared to the costs of complying with the different thresholds for reporting across the EU which currently exist.
One disadvantage of this option is that the market would not receive the data on short positions, only regulators. This would mean that the information asymmetries between short sellers and other investors described in problem 2 would persist. However, it could be argued that market participants are only interested in data on short positions of a certain magnitude, as to receive data on every short position, however small, would mean processing a lot of data which could be difficult.
A second disadvantage of this option is the cost of compliance and administrative burden on investors. According to the responses received from market participants to a survey by DG Internal Market, the ongoing costs of compliance with existing national reporting obligations are estimated to be between 50,000 and 60,000 euro per jurisdiction.
The impact on compliance costs could be mitigated if a threshold were to be introduced for the reporting of positions to the regulator. As indicated above, CESR has proposed a threshold of 0.2%. Such a threshold would mean that insignificant short positions would not have to be reported, which would exclude from the obligation those only engaging short selling it to a very limited extent. The burden on regulators would also be reduced as they would receive fewer notification and only those which reach a sufficient size to be useful for supervisory purposes.
A third disadvantage is the potential negative impact on market liquidity. While some studies have suggested that disclosure obligations have a negative impact on liquidity (see section 6.1.2.4. below), these concern primarily public disclosure. However, disclosure to the regulator may also inhibit firms from short selling if they wish to avoid regulatory scrutiny of their strategies.
Option 3 is favoured by the majority of market participants who responded to the public consultation. Although there were many different views on the levels and instruments to be covered under such an option, there was a significant feeling that this would be preferable than public disclosure. Indeed, respondents also stated the benefits this would bring to regulators, such as the detailed information needed to exercise regulatory duties and, an increased data set to aid the understanding of the effects of short selling.
However, some respondents (primarily regulators and non investment firms), as well as the association representing the interests of issuers in Europe, argued in favour of disclosure of short positions to the market as well as to the regulator. CESR has also recommended a two-tier model for disclosure which combines disclosure to the regulator at one threshold with disclosure to the market at a higher threshold, as this would "help deter and constrain aggressive short selling which may threaten orderly markets or pose the risk of market abuse, and provide early warning signs of a build up of large short positions, thereby alerting regulators to potentially abusive behaviour and enabling them to monitor and take action more effectively127" (see section 6.1.2.4 below).
6.1.3.4 Option 4 – disclosure of significant net short positions to the market
Under this option, net short positions would be disclosed to the market as well as to the regulator. In addition, it would then need to be decided whether this information should be disclosed in an aggregated form (see option 5) or as individual net short positions (option 6).
Option 4 has the same advantages as option 3 – it meets in full the objective of transparency of short positions to the market, including those obtained through CDS, and has the same advantages in terms of reducing the scope for regulatory arbitrage and compliance costs.
The advantage of option 4 over option 3 is that it provides markets as well as regulators with transparent access to data on short positions. According to CESR, this provides informational benefits to the market, improving insight into market dynamics and making available important information to assist price discovery, as well as providing a more effective potential constraint on aggressive large-scale short selling128.
However, this option also has disadvantages. In addition to the compliance costs which are of a similar magnitude to those for notification to the regulator, the main disadvantage of option 4 is, according to many responses by market participants to the public consultation, an adverse impact on market liquidity. This is attributed by market participants to concerns that if their positions are disclosed to the market, other investors may copy their strategies ('herding behaviour') and this could result in them making losses ('short squeezes'); in order to avoid this risk of herding behaviour, certain firms might opt not to short sell so as not to have to disclose their investment strategies, thereby damaging liquidity. However, CESR argues that national regulators who operate public disclosure regimes and have undertaken empirical analyses of the impact of these regimes have not seen these concerns materialise129. For example, the UK regulator, which instituted a temporary regime of public disclosure for UK financial securities, found no evidence of phenomena such as herding behaviour or short squeezes occurring while the regime was in operation.130
One study of the UK disclosure regime suggests that public short selling disclosure obligations reduced short sellers participation in equity markets by 20-25%, and that the securities subject to the disclosure regime experienced a 13% decrease in trading volumes, as well as a widening of bid-ask spreads of over 45%131. However, the results of this study may have been distorted by the comparison of data in a declining market in 2008 with data in a benign market from April 2009 onwards, and the wider bid-ask spreads on financial stocks may reflect their higher risk at that time. The UK Financial Services Authority (UK FSA) takes the view that "while [a public disclosure regime] may have some effect on levels of short selling" they 'do not expect it would do so to any degree that will reduce market quality.' Indeed, the UK FSA's analysis of stock lending data for the financial securities covered by the public disclosure regime levels concluded that levels of short selling were in line with what would be expected given underlying market trends132.
Option 4 was strongly opposed by a significant number of market participants who responded to the public consultation, notably banking and investment firms and their representative associations, and also to a lesser extent by a limited number of other respondents133. However, this option was favoured by the association representing the interests of European issuers in their response to the public consultation, on the grounds that investor relations departments need to know who is selling shares short in order to advise the company's management on the market's views of their shares.134 It is also the agreed position of Member State regulators within CESR that net short positions above a threshold of 0.5% should be disclosed to the market, as in addition to the substantial informational benefits this option would provide to the market, it would help to constrain particularly aggressive large-scale short selling which may involve unacceptable risks of abuse or disorderly markets135.
The concerns raised by some market participants could be mitigated by introducing thresholds for reporting to the regulator and to the market, so that only significant short positions of a certain magnitude, exceeding a certain threshold, would be disclosed. CESR recognises that the efficacy and compliance burden of such a disclosure regime would depend on the levels of such thresholds, and that it would not provide a complete picture of the overall levels of shorting as investors may limit their short transactions in order to remain below the disclosure threshold136. Nevertheless, they conclude that the benefits of having harmonised uniform thresholds outweigh the disadvantages137. As outlined above, CESR has proposed a threshold of 0.2% for notification to the regulator, and of 0.5% for disclosure to the market138. A disadvantage of having thresholds is that short positions held just below those thresholds would not be disclosed, but if this option were to be combined with option 2, regulators would still have access to data on these short positions.
Option 4 could also be combined with option 5 and data could be disclosed to the market only on an aggregated basis, an option supported by several associations of market participants139 (see section 6.1.2.5 below). However, combining these options would not provide the degree of transparency sought by issuers and would not have the same effect of discouraging aggressive short selling.
6.1.3.5 Option 5 – aggregated disclosure of net short positions
Under this option, short positions would not be disclosed individually, but would be compiled by the regulator and disclosed to the market on an anonymous basis: the total size of short positions held in a financial instrument would be made public, but not the identity of the individual institutions which held those short positions.
This option has the advantage that it would partially meet the objective of increasing transparency in short positions, including through CDS, and it would also meet the objective of reducing the scope for regulatory arbitrage and compliance costs through a common EU approach to disclosure. For short sellers themselves, it has the advantage that their individual short positions would not be disclosed, which might make them less likely to reduce their levels of short selling (for the reasons explained in the analysis of option 4 above), which would mean their contribution to liquidity in the instruments concerned would be less affected.
On the other hand, CESR considers that the fact that individual positions would remain anonymous would make this option less effective as a constraint on aggressive short selling140, and so this option would not contribute to achieving the objective of reducing the risk of negative price spirals.
This option has other disadvantages: the degree of transparency would be less than that provided by the disclosure of individual short positions (option 6), so issuers and other market participants would not know which individual institutions held significant short positions in specific financial instruments, which could be valuable information for them in formulating their business or investment strategies. Furthermore, regulators would have to process the data received in order to publish it in aggregated form, thereby imposing administrative costs on them. Financial institutions would also still face compliance costs to disclose.
Further, when compared to option 1, option 5 has the disadvantage that EU Member States which currently require disclosure of individual short positions would have to reduce the level of transparency by switching to aggregated disclosure. While this may be welcomed by institutions engaged in short selling themselves, the regulators and other market participants in the concerned Member States may not welcome this reduction in the level of transparency that they have become accustomed to.
Most market participants who responded to the consultation favoured this option. In contrast, national regulators have agreed in CESR on a disclosure model based on the disclosure of significant individual net short positions (see option 6).
6.1.3.6 Option 6 – disclosure of significant individual net short positions
Under this option, individual short positions would be notified on a net basis (i.e. short positions would be subtracted from long positions). A threshold would be introduced so that only significant net short positions would be disclosed. CESR has proposed a threshold of 0.5% of the issued share capital.
This option is a significant improvement on options 1 and 5. It fully meets the objective of transparency by providing regulators or the public with data not just on short positions, but also which institutions hold those short positions. Unlike the baseline scenario or option 5, under this option issuers gain full access to data on which institutions may hold short positions in their financial instruments, and other market participants can adjust their investment strategies based on a complete picture of who holds significant short positions at any given time. Option 6 also fully meets the objective of reducing the scope for regulatory arbitrage as short positions through the use of derivatives are captured and a common disclosure regime would apply to all Member States. This would also mean lower compliance costs for market participants compared to option 1.
The main disadvantage of this option, as explained above, is that financial institutions may reduce their levels of short selling in order to avoid disclosure of their investment strategies to the market, and this could have a negative impact on liquidity. However, this could be mitigated by providing for a threshold for disclosure to the market.
6.1.3.7 Option 7 – exemption from disclosure requirements for market making activity and primary market operations
Under this option, market making activities and primary market operations would be exempt from the disclosure requirements and the flagging requirement. The content and conditions for the exemption are as described in section 6.1.2.5.
As explained in section 6.1.2.5 above, the rationale for this exemption is that liquidity provided by market making activity is critical to the efficient functioning of markets, especially for less liquid securities. CESR has proposed an exemption for market making activities in their model for a pan-European short selling disclosure regime because, in order to be able to provide liquidity on demand throughout the day, market makers regularly have to take short positions, generally temporary ones, and may be at risk of unjustified commercial prejudice if those positions are known to the market. CESR recommends that the exemption should be in relation to both notification to the regulator and disclosure to the market141.
This option is also in line with most responses from market participants to the consultation (see section 6.1.2.5), who consider that market makers and primary market dealers fulfil an important and legitimate function in providing liquidity, and this liquidity would be affected negatively by applying disclosure and other obligations to these activities. Further, publication of information about positions at the end of the trading day would be of limited informational value to competent authorities or even misleading as it would only provide a snapshot of an arbitrary and fleeting position.
This option has the advantage that the objective of increasing the transparency of short selling would be met, while avoiding the possible negative impact on liquidity that full application of the proposed transparency requirements could have on activities that are essential to the efficient functioning of markets. In particular, issuers would continue to obtain the benefits of liquidity provision for their financial instruments provided by market makers and primary market dealers, and financial institutions providing this service would be able to continue to do so. At the same time, short selling carried out for speculative or hedging purposes would be captured by reporting or disclosure obligations, but not short positions carried out for market making or primary market operations.
Another advantage of this option is that objective 4 would also be achieved, as a harmonised exemption for market making, based on a common definition, would eliminate the scope for regulatory arbitrage as all Member States would have the same definition. This would also entail lower compliance costs than the current situation where different Member States apply different regimes for market makers.
This option has the disadvantage that regulators or the market may lose access to some data that they may occasionally consider to be useful. However, this risk is mitigated by the fact that competent authorities can require any data they consider necessary from any market participant using the exemption.
The other potential disadvantage of this option is that some financial institutions might seek to use this exemption to circumvent the transparency requirements for activities which do not in reality constitute market making. However, this risk is very limited and can be mitigated by using the harmonised definition of market making activities recommended by CESR and by providing for the controls on the use of this exemption outlined above. In particular, those wishing to benefit from the exemption would have to notify the competent authority in writing not less than 30 days in advance of their intention to use the exemption, and the competent authority would have the power to obtain information from any institution benefiting from the exemption.
6.1.3.8 The preferred options
Impact on stakeholders | Effectiveness | Efficiency | |
Option 1 (baseline) | n.a. | n.a. | n.a. |
Option 2 (flagging) | (+++) regulators obtain very precise information on short vs. long transactions, which captures intraday positions and aids enforcement (+++) governments comforted by regulators access to real time data (0) issuers – no obvious costs or benefits (0) individual investors – as above (- -) financial institutions bear compliance costs | (+++) Objective 2: very transparent, captures intraday positions (++ ) Objective 4: partially met, but this option in isolation would not capture short positions through OTC derivatives | (- - ) compliance costs for financial institutions |
Option 3 (notification of short positions to regulator) | (+++) regulators obtain data on short positions they currently lack (- - -) issuers do no have access to data on short positions in their shares (- - -) individual investors - information asymmetries persist (+++) governments gain comfort that regulators obtain data on short positions through CDS (- ) financial institutions face compliance costs to notify and may reduce levels of short selling | (+++) Objective 2:fully met for regulators (+++) Objective 4: fully met;, better than option 2 as short positions through derivatives are captured | (- ) ongoing compliance costs; could reduce market liquidity |
Option 4 (disclosure of short positions to the market) | (+++) regulators obtain data on short positions they currently lack (+++) issuers obtain access to data on short positions in their shares (+++) individual investors - information asymmetries eliminated (+++) governments gain comfort that regulators obtain transparency on short positions through CDS (- - -) financial institutions face compliance costs to disclose and likely to reduce levels of short selling | (+++) Objective 2:fully met for market (+++) Objective 4: fully met | (- -) ongoing compliance costs; more likely to reduce market liquidity than option 3 |
Option 5 (aggregated disclosure of net short positions) | (+) regulators have access to data on short positions but have to process it in order to publish it in aggregated form (+) issuers obtain less specific data on who holds short positions in their shares (+) individual investors see partial reduction in information asymmetries (++) governments gain transparency on short positions through CDS, but some may lose out if they currently require individual disclosure (- -) financial institutions still face compliance costs to disclose, but less likely to reduce levels of short selling | (+) Objective 2: partially met, less detailed transparency than option 6 (+) Objective 4: met, but at cost of reduced transparency for some Member States | (- )ongoing compliance costs, but lower impact on liquidity |
Option 6 (disclosure of individual significant net short positions) | (+++) ensures regulators obtain full transparency (+++) issuers obtain full transparency (+++) individual investors informed of short sellers' strategies (+++) governments obtain full transparency on short positions through CDS (- - -) financial institutions face compliance costs and may reduce levels of short selling to avoid disclosure of their positions | (+++) Objective 2: met in full (+++) Objective 4: met in full | (- -) ongoing compliance costs, more likely to reduce market liquidity than option 5 |
Option 7 (exemption for market making and primary market operations) | (-) regulators may lose some useful data or face risk of abusive use of exemption (+++) issuers obtain full benefits of liquidity provision by market makers and can use short selling for stabilisation (+++) individual investors also benefit from ease of buying and selling at efficient price (+++) governments gain comfort that liquidity in sovereign bond markets is not impaired (+++) financial institutions engaged in bona fide market making can continue their activities | (++) Objective 2 partially met as all short selling captured except when it is carried out by legitimate market makers (+++) Objective 4 fully achieved as exemption would be harmonised at EU level (+++) General objective of avoiding unduly negative effects on market efficiency achieved | (+++) liquidity provided by market makers and primary market dealers not affected |
Based on the analysis above, the highest scoring options are options 2, 3, 4, 6 and 7. All these options are compatible. In fact, a combination of these five options would address some of their individual shortcomings. In particular, option 3 only provides transparency to regulators, whereas option 4 provides transparency also to the market. By combining these two options, using one threshold for notification to the regulator and a second, higher threshold for notification to the market, the objective of transparency for both would be achieved. In addition, the higher threshold for notification to the market would mitigate any impact on liquidity that disclosure at a lower threshold might have, while ensuring that regulators obtain the data they require. Combining these two options in this way also has the advantage of being in line with the disclosure model drawn up by CESR. The option of requiring disclosure of individual significant net short positions (option 6) should be preferred to aggregated disclosure (option 5), as it meets the objectives more fully by providing the market with more detailed transparency. Combining option 2 (flagging) with the disclosure options would complement these very effectively by potentially providing regulators with real time data on all short positions, not just those above the threshold, thereby helping regulators with enforcement, and would also capture intraday positions which would not be captured by a disclosure regime. Finally, opting for an exemption for market making activities and primary market operations (option 7) would ensure that the important liquidity provision function of these activities would be able to continue, which would mitigate any potential impact on liquidity of a disclosure regime. Inclusion of this option will greatly contribute to ensuring the proportionality of the overall approach.
So the preferred option is a combination of options 2, 3, 4, 6 and 7.
With regard to the legal scope of the transparency regime, this should apply to shares and their derivatives, including CDS. However, with regard to bonds, the public consultation showed very limited support for the inclusion of corporate bonds and their derivatives in a disclosure regime, and therefore these should be excluded. With regard to sovereign bonds and their derivatives, including CDS, there was support for their inclusion by public authorities, but serious concerns were expressed about the potential negative impact on liquidity of public disclosure of sovereign bond and sovereign CDS short positions. Therefore the disclosure of sovereign bonds and sovereign CDS should be limited to regulators only. This scope is consistent with the evidence of problems and concerns raised by governments, regulators or issuers of transparency deficiencies; it is therefore also consistent with the principle of proportionality. The public consultation showed very limited support for other instruments to be captured in a disclosure regime.
An important issue of geographical scope also arises in the context of disclosure. Transactions in shares or sovereign debt admitted to trading on a trading venue in the EU can also take place outside the EU, and if such transactions are not captured by the disclosure regime, this could be circumvented. Therefore for this regime to be effective, it is important that notification and disclosure obligations apply no matter where a transaction takes place, including where it takes place outside the Union but in relation to a company or sovereign debt issuer that has shares or sovereign debt admitted to trading on a trading venue in the Union.
6.2.The preferred policy options and instrument
6.2.1.The preferred policy options
Based on the above analysis of the impacts, the preferred policy options to achieve the objectives set out in this impact assessment are the following:
Reduce the risks of negative price spirals arising from short positions (including those obtained through CDS)
- introduce a power for national competent authorities to temporarily restrict short selling in a financial instrument admitted to trading on an organised market whose price has fallen by a specified quantitative threshold, e.g. 10% ('circuit breaker').
- grant national competent authorities the power to temporarily restrict or ban short selling of some or all financial instruments or CDS transactions in exceptional situations, with coordination by ESMA and without prejudice to ESMA's powers under article 10 of Regulation ??/EC establishing ESMA).
Increase the transparency of short positions (including those obtained through CDS)
- introduce a flagging system.
- notification of short positions to the regulator at a threshold of 0.2%.
- disclosure of short positions to the public at a threshold of 0.5%.
- disclosure of individual significant net short positions.
- exemption from disclosure requirements for market makers.
Reduce settlement risk linked with naked short selling
- introduce a requirement that before entering into a short sale, a person must have borrowed the share, entered into an agreement to borrow the share or have other arrangements which ensure that he will be able to borrow the share at the time of settlement (locate rule)
- introduce EU rules on settlement discipline so that persons engaging in short sales which result in a failure to deliver face appropriate penalties, with buy-in procedures and fines in case of settlement failures
- exemption for market makers from the locate rule, but not from the buy-in procedures.
6.2.2.Choice of instrument to ensure a coordinated response by EU Member States to short selling and CDS
6.2.2.1 Non-legislative cooperation between Member States with guidelines by CESR/ESMA
One option to achieve the objectives set out in this report would be through cooperation between governments and regulators in the EU Member States, coordinated through CESR and in future through ESMA. This is to some extent already happening, as national regulators have discussed and agreed in CESR on a model for a two-tier short selling disclosure regime, and CESR has advised regulators to implement this model pending possible action at EU level. CESR is also working on a model for disclosure to regulators for sovereign bonds and sovereign CDS. If all Member States were to voluntarily implement these models based on the work of CESR, this would go a long way towards addressing the objective of ensuring that regulators and markets obtain data on short positions, including those obtained through CDS.
However, the disadvantage of this approach is that it is voluntary, and Member States may opt not to implement the CESR model at national level. In the absence of EU legislative action, there would be no obligation on Member States to implement this transparency regime, and the Commission would not be able to take infringement action against Member States that did not act, or which took a different approach than that proposed by CESR/ESMA.
The divergent responses taken by Member States to short selling in exceptional situations also illustrate the limits of non-legislative cooperation. Member States are not required to consult and cooperate with each other before introducing exceptional measures, which leaves wide scope for unilateral actions to be taken without prior consultation. In the absence of a European legislative framework, the effectiveness of national actions is also open to question, as there is no obligation on Member States to cooperate with each other, and the absence of such cooperation leaves scope for regulatory arbitrage.
Concerning the policy options to reduce potential settlement risks arising from naked short selling, cooperation between Member States also has limited effectiveness. It has not been possible for Member States to agree on a common approach to the regulation of naked short selling, with some Member States having a permanent ban (e.g. Spain) and others introducing temporary bans (e.g. France and Germany). Trading venues and Member States also take different approaches to the issue of settlement discipline.
Last but not least, the European Commission has opted in its Work Programme for 2010 to present a legislative initiative on short selling and Credit Default Swaps. In view of the disadvantages outlined above, the option of promoting non-legislative cooperation between Member States with guidelines by CESR/ESMA is therefore not the most effective and efficient instrument to achieve the objectives set out in this impact assessment.
6.2.2.2 Propose new stand-alone EU legislation on short selling in a Directive or a Regulation
Having discarded the option of a non-legislative instrument, this leaves the options of pursuing the objectives of this impact assessment through a harmonising legal instrument. A harmonising legal instrument would ensure that all Member States applied the same regulatory framework based on the same principles, thereby ending the current fragmentation of the regulatory response to short selling and ensuring that compliance costs would be lower. However, it is necessary to define whether the appropriate legal instrument should be a Directive or a Regulation.
Traditionally, the main legislative instrument chosen for EU financial services legislation has been a Directive. This was because the legislative proposals mainly sought to approximate national rules on the taking up of business and the provision of services in a gradual manner. The choice of a Directive enables Member States to integrate rules into their different legal systems, while allowing some margin for them to extend EU rules to areas uncovered by the EU legislation.
However, a Directive does not seem to be the right choice of instrument in view of the objectives of this initiative for a number of reasons. A Directive would leave some scope for Member States to maintain divergent rules, whereas a Regulation would ensure uniform rules throughout the EU. Nevertheless, a Regulation can leave some flexibility for national competent authorities in applying the rules. While a Directive requires national implementing provisions to be adopted, leaving scope for interpretation of the Directive, a Regulation is directly applicable without requiring national legislation, thereby ensuring greater legal certainty for those subject to the legislation.
Implementation of a Directive into national law can also be a time consuming process. In contrast, a Regulation is immediately applicable after adoption by the legislator and, while it is likely to require binding technical standards to be adopted through delegated acts in certain areas to ensure consistent application, these can be prepared in parallel to the process for adoption of the legislation. Further, a regulation does not require any monitoring of correct implementation by the Commission, and those concerned by the provisions of a Regulation would be able to depend on them immediately. Finally, a Regulation could be directly invoked by concerned parties before national administrations and courts, whereas this applies only in very limited circumstances for Directives.
For all these reasons, the Commission services consider that a Regulation rather than a Directive the most appropriate instrument for this initiative.
6.2.3.Impact on retail investors and SMEs
Given that it is a highly risky investment strategy, short selling is unlikely to be widely used by retail investors, so to the extent that this initiative may restrict levels of short selling, retail investors will not be significantly affected in a negative sense. However, retail investors would benefit from the enhanced transparency of short selling transactions, as they would have a clearer picture as to the reasons why some investors are selling certain securities, and could either emulate those strategies or decide to hold their securities in the knowledge that prices might rebound when short sellers closed their positions. Retail investors would also benefit from the enhanced stability of markets which coordinated national responses to short selling could bring at times when markets are distressed. Retail investors also benefit from a highly liquid market, so the option to exempt market making activities would help to preserve liquidity and make it easier for retail investors to buy and sell securities at efficient prices. The options which reduce the risk of settlement failure would also be to the advantage of retail investors, as they would face a lower risk that securities they had bought from a short seller would not be delivered in due time.
Shares in SMEs are unlikely to be subject to short selling as these are generally much less liquid markets, so small and medium sized issuers would not be substantially impacted by restrictions on short selling. However, SMEs as investors would obtain similar benefits to those outlined for retail investors above.
6.2.4.Impact on third countries
This initiative will impact on third countries in two main ways. As explained in section 6.1.2, it is proposed that in order for them to be effective, notification and disclosure obligations should apply no matter where a transaction takes place, including where it takes place outside the EU but in relation to a company or sovereign debt issuer that has shares or sovereign debt admitted to trading on a trading venue in the EU. If this were not to be the case, it would be possible for investors to circumvent the disclosure obligations by engaging in transactions in third countries, either over the counter or (where securities are listed in third countries) on trading venues in those countries. The public consultation showed strong support from most respondents and from public authorities for the application of the disclosure regime to extend to both persons within and outside the EU who have significant net short positions in EU shares or EU sovereign bonds.
This implies that investors or firms which are outside the EU will have to notify, to the regulator of the most liquid market in the EU for the share or sovereign bond, or disclose to the market, short positions which exceed the relevant thresholds. In order to enforce this requirement, EU regulators should be required by the short selling legislation to upgrade existing international agreements (e.g. the IOSCO MoU) or reach further cooperation agreements with regulators in the third countries where EU shares or sovereign bonds are mainly traded. Such agreements should, inter alia, provide for exchanges of information with regulators in third countries so as to ensure that that concerned investors in those countries are made aware of their obligations with regard to the EU legislation and comply with them. EU regulators could in specific cases exchange information with third country regulators to enforce the obligations in EU short selling legislation, as well to take reciprocal action in an exceptional situation. ESMA could be required to facilitate such cooperation by preparing a template agreement that could be used by national competent authorities. These authorities should also be required to inform ESMA when they propose to enter into an agreement with a third country regulator. This approach would be proportionate since by upgrading existing agreements and using an ESMA template Member States would be able to limit the costs of reaching cooperation agreements with third countries.
At the same time, it is foreseen that this initiative would not apply to shares for which, while they may be admitted to trading on a trading venue in the EU, their principal market is located outside the EU. This reflects that fact that shares in companies are increasingly traded on a number of different trading venues around the world. For example, many large overseas companies have shares traded not only on their home market but on a trading venue in the European Union. It is not appropriate or proportionate to apply short selling requirements where most trading of the share takes place outside the Union. This is potentially onerous for market participants and creates unnecessary complexity that may discourage issuers from having their shares traded on venues in the European Union.
In order to facilitate implementation of this provision by investors, ESMA should be notified at least every two years by national regulators of which shares fall into this category, and ESMA would then publish a list of those shares.
The fact that the legal framework on short selling can diverge across different countries leaves scope for international regulatory arbitrage, especially for instruments which are not traded on a regulated trading venues. In particular, the trading of sovereign bonds and credit default swaps is almost exclusively done over the counter, and can therefore move very easily from one jurisdiction to another. International regulatory cooperation through fora such as IOSCO is therefore very important in this area, and the European Union should take a lead in enhancing cooperation with its main international partners.
The proposed initiative on short selling aims at introducing greater coordination in Europe in the regulation of financial markets. As such, it could contribute to more integrated financial markets in Europe. Similarly to the effect of the single currency, this further integration could have the effect of enhancing the global nature of European financial markets, which could in turn increase their relative competitiveness compared to third countries, notably the United States.
The preferred options retained in this impact assessment would ensure that the European Union's regulatory regime would converge with that of the United States, which already operates a flagging regime, circuit breaker, a disclosure regime, a locate rule and buy-in procedures in case of settlement failures.
6.2.5.Social impact
To the extent that the options considered in this initiative will help to contain the effects of future financial crises on the real economy, they will also help reduce the social costs of those crises (e.g. unemployment).
6.2.6.Environmental impact
It does not appear that the options proposed in this will have any direct or indirect impacts on environmental issues.
6.3.Estimate of impact in terms of compliance costs
The preferred options include requirements for firms to ensure regulators and markets obtain data on short positions (including CDS) by notification of short positions to the regulator and by disclosure of short positions to the public. As part of this option, regulators would be expected to provide quarterly reports to ESMA on the net short positions notified or disclosed within their jurisdiction. This section provides an estimate of the compliance costs (including both one-off and ongoing costs) implied by these requirements. The full explanation of the methodology, assumptions and calculation of the estimate can be found in annex 8.
This section also provides an indication of the order of magnitude of the cost of disclosure of significant short positions in sovereign bonds which is also part of the preferred options. A more comprehensive assessment of the compliance cost of this approach has not proved possible due to the absence of data, since no Member State has such a disclosure regime at national level and no Member State has been able to provide estimates for the cost of such a reporting regime.
For some options assessed in this report it has not proved possible to carry out a detailed estimate of the compliance costs for the EU as a whole. This estimates of compliance costs in this section do not include the costs of flagging, because other than the estimates provided by the UK FSA (see section 6.1.2.2), no cost estimates have been submitted to the Commission services despite repeated requests for data from regulators and market participants. The analysis of the flagging option acknowledges that the costs of this option could be significant, but also that this option would offer benefits in terms of increased transparency.
Finally, the Commission services have not received from market participants estimates of compliance costs for the locate rule option. However, as explained in section 6.1.3.2, the costs of a locate rule would be limited since most market participants indicated to the Commission that they currently apply a locate rule as a matter of good business practice.
According to our estimation, the main part of the compliance costs related to notification and disclosure requirements would consist in a one-off investment in information technology and information systems (IT/IS) development, training and compliance procedures. We estimated the number of banks and investment firms with sufficiently large operations to be likely to have to disclose (based on data of existing disclosure regimes provided by 6 Member States). We then assumed that 10% of these firms would be large, complex cross-border institutions reporting across most EU jurisdictions, with the remaining 90% of firms mainly reporting in one jurisdiction.
Based on estimates provided by some market participants we estimated the average one-off costs in terms of IT/IS development, training and compliance procedures for complying with disclosure obligations for these two types of firms. We therefore estimate the EU-wide one-off costs in terms of compliance costs to be approximately 137 million euro. We estimate ongoing costs to maintain or upgrade IT/IS on a yearly basis to be 10% of this cost, or approximately 13.7 million euro.
In order to estimate the ongoing costs in terms of disclosure to the regulator when the relevant threshold is exceeded and to the public when the higher threshold is exceeded, we examined data provided by 6 Member States on the number of disclosures received in their current national disclosure regimes. Since these regimes are based on disclosure of significant short positions only in shares of financial institutions, we then extrapolated from this data to obtain an estimate of the yearly average number of disclosures for all shares. We also grouped Member States into three categories according to the likely frequency of disclosures per year: high, medium and low. Since these estimates do not take into account the multiple thresholds foreseen in the preferred disclosure option, these estimates were adjusted based on estimates for the frequency of disclosures at different thresholds provided by the UK.
We then estimated the yearly number of disclosures for each category based on the data available from 6 Member States. We estimated the average time it would take to produce one disclosure as 2 hours. Given that a large proportion of notifications are likely to be made by complex cross-border entities, often with a centralised compliance function in a high wage cost country, we thought it appropriate to apply an average hourly wage for managers from a large Member State (in this case the UK). We then estimated the costs per Member State of disclosure based on the estimated average number of disclosures.
As a result of the preceding analysis, we therefore estimate that the ongoing costs for disclosure of short positions in shares to be approximately 2.1 million euro per year. To this should be added the estimate of ongoing IT/IS costs of 13.7 million euro per year. Therefore the annual EU-wide compliance cost (not including one-off costs) could be estimated to be approximately 15.8 million euro.
For the administrative cost on regulators of quarterly reporting to ESMA of significant net short positions in their jurisdiction, we also assumed that compiling each quarterly report would take 2 hours. Based on the average hourly wages of managers in the 27 Member States, this gives an estimated total administrative cost for regulators in the EU 27 of 6.758 euro per year.
The costing of a disclosure regime on bonds should be largely based on the cost evaluation carried out for shares above: the number of reporting entities and the split between small and complex cross borders entities should be the same, as the same firms are active in bonds and in shares.
Along these lines, the articulation of the costs of the reporting of bonds positions should very similar. If the banks set up IT systems for reporting on shares, these systems are likely to allow reporting of positions on bonds to a large extent. They may need some adjustments at the margin in order to either increase the capacity of systems to also take into consideration reporting on bonds (or to connect to the bond positions management systems that may differ slightly from those for shares). In any case, the one off initial costs should be expected to be marginal compared to the initial costs for equities. Conservatively, we can estimate these costs at 25% of the initial costs for shares, with the same variation between complex and non complex organisations as we have estimated for shares. One-off compliance costs for the sovereign bond disclosure requirement could therefore be estimated to be of the order of 34.2 million euro.
The ongoing costs for bonds should be calculated applying the same ratio of 10% of initial costs, leading to an estimated ongoing cost of the order of 3.4 million euro.
However, it is much more difficult to assess the frequency of disclosures on sovereign bonds by financial institutions, particularly as the disclosure threshold is not yet known. Nevertheless, the frequency of disclosure is likely to be lower for two main reasons. The first is that there will be only one disclosure threshold for bonds, while there are two different thresholds for shares. The second is that bonds are far less frequently traded than shares142, so short positions on bonds are therefore very likely to vary far less than those for equities. The order of magnitude of the gap between the two could be further estimated by examining the difference between the average frequency of bond trading compared to the average frequency of share trading. Conservatively, it appears reasonable to take 75 % of the number of disclosures for shares, since the liquidity of sovereign bonds is higher than that of corporate bonds, but still less than that of shares. On this basis, we could estimate the ongoing compliance costs for sovereign bond disclosures to be of the order of 1.6 million euro per year. This gives an estimate of the ongoing EU-wide compliance cost for disclosure of sovereign bonds of the order of 5 million euro per year.
7.Monitoring and Evaluation
The Commission is the guardian of the Treaty and therefore will monitor how Member States are applying the changes proposed in the legislative initiative on short selling. When necessary, the Commission will pursue the procedure set out in Article 226 of the Treaty in case any Member State fails to respect its duties concerning the implementation and application of Community Law.
The evaluation of the consequences of the application of the legislative measure could take place two years after the entry into force of the legislative measure, in the context of a report to the Council and the Parliament on the appropriateness of the reporting and public disclosure thresholds, which has been recommended by CESR.
The main indicators and sources of information that could be used in the evaluation are as follows:
- Data on short positions gathered by national regulators and reported to ESMA on a quarterly basis;
- Data on liquidity and spreads in the markets affected by the short selling initiative; and
- A report (which could be undertaken by ESMA) on the experience gained by regulators in enforcing the short selling initiative and how cooperation has worked in practice.
8.Annex 1 – Glossary of terms
Short Sales143
What is a short sale? A short sale is generally the sale of a security you do not own (or that you will borrow for delivery). Short sellers believe the price of the security will fall, or are seeking to hedge against potential price volatility in securities that they own. If the price of the security drops, short sellers buy the security at the lower price and make a profit. If the price of the security rises, short sellers will incur a loss. Short selling is used for many purposes, including to profit from an expected downward price movement, to provide liquidity in response to unanticipated buyer demand, or to hedge the risk of a long position in the same security or a related security.
Example of a short sale: For example, an investor believes that there will be a decline in the stock price of Company A. Company A is trading at €60 a share, so the investor borrows shares of Company A stock at €60 a share and immediately sells them in a short sale. Later, Company A's stock price declines to €40 a share, and the investor buys shares back on the open market to replace the borrowed shares. Since the price is lower, the investor profits on the difference -- in this case €20 a share (minus transaction costs such as commissions and fees). However, if the price goes up from the original price, the investor loses money. Unlike a traditional long position — when risk is limited to the amount invested — shorting a stock leaves an investor open to the possibility of unlimited losses, since a stock can theoretically keep rising indefinitely.
How does short selling work? Typically, when a person sells short, their brokerage firm will loan them the securities. The securities they borrow come from either the firm's own inventory, the margin account of other brokerage firm clients, or another lender. The person is then able to sell these borrowed securities knowing that they will have to replace them at an arranged later date. If the securities which are borrowed pay a dividend, the person must generally pay the dividend to the owner from which the securities were borrowed.
Large institutional investors and hedge funds will typically have standing arrangements in place with specialist brokers and investment firms who offer an established and arranged facility to borrow securities. These are referred to as Prime Brokers and for a fee will manage the lending and administration of the securities as well as other services, e.g. cash management. Other market participants such as settlement systems also offer facilities that enable investors to ensure that the securities are covered and will be available when settlement is due.
Uncovered or 'Naked' Short Sales: In an uncovered short sale, the seller does not borrow or arrange to borrow the securities prior to executing the trade. If the seller is then unable to source the securities before the physical settlement date they will be unable to deliver to the buyer (known as a "failure to deliver" or 'fail').
Failures to deliver can result irrespective of if the seller physically has the securities or not and there may be some legitimate reasons for a fail. For example, human or mechanical errors or processing delays can result from transferring securities in physical certificate rather than book-entry form, which may happen even if the seller physically holds the security. Further, a fail may also result from an uncovered short sale when a market maker, in response to a customer order, sells a very thinly traded and illiquid security. If the market maker subsequently encounters difficulty in obtaining the security they may be unable to settle.
1. Credit Default Swaps
A Credit Default Swap (CDS) is a derivative which is sometimes regarded as a form of insurance against the risk of credit default of a corporate or government (or sovereign) bond. In return for an annual premium, the buyer of a CDS is protected against the risk of default of the reference entity (stated in the contract) by the seller. If the reference entity defaults, the protection seller compensates the buyer for the cost of default.
In addition to short selling on cash markets, a net short position can also be achieved by the use of derivatives, including Credit Default Swaps (CDS). For example, if an investor buys a CDS without being exposed to the credit risk of the underlying bond issuer (a so-called 'naked CDS'), he is expecting, and potentially gaining from, rising credit risk. This is equivalent to short selling the underlying bond.
2. Other Terms
A Central Counterparty (CCP) is an entity that acts as an intermediary between trading counterparties and absorbs some of the settlement risk. In practice, the seller will sell the security to the central counterparty, which will simultaneously sell it on to the buyer (and vice versa). If one of the trading parties defaults, the central counterparty absorbs the loss.
A Derivative is a security whose price is dependent upon, or derived from one or more underlying assets. Examples of derivatives include Options, Futures, CDS, and Contracts for Difference. The underlying assets can be diverse, however common examples include shares, bonds, commodities and interest rates.
Primary Market Operations are transactions performed by dealers to provide liquidity to issuers of new securities such as sovereign debt and for the purposes of stabilisation schemes (i.e. share issues intended to stabilise a share price). Stabilisation schemes are defined under the Market Abuse Directive.
A Market Maker is a firm that will buy and sell a particular security on a regular and continuous basis by posting or executing orders at a publicly quoted price. They ensure that an investor can always trade the particular security and in doing so enhance liquidity in that security.
Prime Brokers, or firms offering Prime Brokerage services, are firms which offer specialist bespoke services to, generally, large institutions and hedge funds. Prime brokerage covers a range of services, including securities borrowing and administration, and cash management. Many large investment banks provide Prime Brokerage services.
A Multilateral Trading Facility (MTF) is an electronic system which facilitates the exchange of securities between counterparties. The securities may include derivatives and instruments which do not have a main market, as well as traditional securities.
A Trading Venue is an official venue where securities are exchanged; it includes MTFs and regulated markets (e.g. typical stock exchanges).
A Long Position refers to the buying of a security with an expectation that the security will rise in value.
A Short Position refers to the selling of a security with an expectation that the security will fall in value.
A Net Position is where a short position has been subtracted from a long position. If the short position is greater than the long position the result will be a Net Short Position; if the long position is greater than the short position the result will be a Net Long Position.
9.Annex 2 – Summary of public consultation on Short Selling
DG MARKT services held a public consultation between 14 June and 10 July 2010. Responses to the public consultation were received from finance ministries and regulators, investors, intermediaries, exchanges and clearers, issuers and individual citizens (see list in section 9.2 below).
9.1.Summary of the responses
Some 123 contributions were received, of which 112 were authorised for publication, including 2 from issuers, 55 from investors (funds, banks, associations, investment funds, savings banks etc), 14 from exchanges, 19 from Member States and Financial Agencies (securities regulators, finance ministries and relevant bodies), 5 from non-financial institutions and 17 from others (e.g. chambers of commerce, citizens, academics).
Contributions received from stakeholders varied in detail; most elaborated comments were provided by investor groups (funds), exchanges and bodies representing issuers.
There was a fair degree of consistency in the responses.
The following general remarks were repeatedly raised with respect to the text:
- Most responses (especially from market participants but also from some public authorities) expressed strongly and consistently the view that the consultation paper refers to risks of short selling when there is little evidence of risks. Most responses asked for further evidence to be provided of those risks; however, some public authorities agreed with the consultation document's description of the potential risks of short selling. Many responses referred to the benefits of short selling to market efficiency and quality and some responses included further data supporting the positive benefits.
- At the same time, there was general support for the Commission's aims of introducing a harmonised regime dealing with short selling issues to avoid problems of unilateral approaches to these issues that have recently been seen.
- On scope the majority of responses argued that the rules should be limited to shares and (to a lesser extent) sovereign debt. There was opposition especially from bond issuer bodies to an extension to corporate bonds. There were also concerns by corporates about the lack of clarity and unintended consequences of extending measures designed for shares to other asset classes.
- On transparency many responses (primarily from market participants) argued for aggregated disclosure to the market of net short positions in shares (instead of individual disclosure). On a disclosure regime for sovereign bonds most respondents argued that this should only be disclosed to the regulator.
- Regarding naked short selling there was some support for introducing requirements related to settlement discipline but a majority were against any restrictions on naked short selling. Some public authorities favoured requirements at the point of trading, either on their own or in combination with rules on settlement discipline.
- There were mixed responses about emergency powers although most responses seem to accept the need for such powers in an emergency situation. But many argued that there needs to be more stringent definition of the powers and when they can be used. Also some responses argued that ESMA should have a stronger coordination role.
On exemptions for market makers there was widespread support for an exemption, although there was some limited opposition.
1. SCOPE OF THE PROPOSAL AND SUBSTANTIVE REQUIREMENTS
1.1. Scope
1) Which instruments give rise to risks of short selling and what is the evidence of those risks?
The vast majority stated that there is little evidence of risks. No new data to support any risks was submitted, although some public authorities agreed with the description of the potential risks in the consultation document. Many responses referred to evidence of benefits of short selling.
2) What should be the scope of the requirements?
The vast majority of responses argued that proposals should only apply to equities and associated derivatives. They felt that this is the only area where there is any evidence of risks.
A few supported limited measures for sovereign bonds/sovereign CDS.
There was opposition (especially from financial bond issuer bodies) to applying requirements to corporate bonds as any measures could not be justified and would be detrimental to corporate bond issuers.
There were also opposition from corporates who use derivatives about applying requirements that are designed for equities to other assets classes such as interest rates and foreign exchange. These respondents argued that such an approach was conceptually flawed and would create difficulties for many businesses that use these derivatives. They argued it was difficult to understand what the short selling requirements meant in the context of such instruments or how they would apply.
3) In what circumstances should measures apply outside the European Union?
There was broad support for applying some requirements to persons outside the EU, but many noted the difficulty of enforcing such a regime.
1.2. Transparency
1) What should be the scope of transparency requirements?
The policy option outlined in the consultation document is largely based on the two tier model for EU shares recommended by CESR in its report in March 2010. The CESR model provides that at a lower threshold notification of a short position should be made only to the regulator and at a higher threshold short positions should be disclosed to the market. Notification to regulators will enable them to monitor and if necessary investigate short selling that may pose systemic risks or be abusive. Publication of information to the market will provide useful information to other market users. The consultation document included also a specific regime for notification to regulators only of significant net short positions in EU sovereign bonds. Disclosure to regulators of significant net short positions relating to EU sovereign bonds could provide important information to assist regulators to monitor whether such positions are creating disorderly markets or systemic risks or are being used for abusive purposes. There was support for applying transparency requirements in relation to shares. There was some support for applying transparency requirements to sovereign bonds (but not for public disclosure of short positions in sovereign bonds).
2) Transparency for notification of short positions in EU shares
There was support for these transparency measures although many responses from market participants argued for aggregated anonymous disclosure to the market rather than disclosure of individual positions. These respondents argued that disclosure to the market exposed them to various risks. However, public authorities did not share this assessment of the risks of public disclosure of short positions in shares.
3) Notification to regulators of net short positions in EU sovereign debt (including through the use of CDS)
There was limited support for this measure although it was noted that public disclosure of individual positions would be extremely detrimental to sovereign debt markets.
The majority of responses argued that there should be an exemption for primary market operations and market makers, although a few argued against exemptions as this only involves private disclosure to regulators.
1.3. Uncovered short sales
1) What are the risks of naked short selling and the evidence of those risks?
Most responses thought there was limited or no evidence of such risks, but commented that the primary risks are settlement failure and market manipulation. Some public authorities argued strongly that there were risks of price volatility associated with naked short selling which could in extreme situations pose a risk to financial stability, and that these should be addressed.
2) Is there evidence of risks of uncovered short selling for instruments other than shares?
No submission provided evidence of risks, but some public authorities saw the same risk for short selling of bonds as for shares.
3) Should there be a ban on entering into naked CDS relating to sovereign issuers?
There was little response to this question. Most of those that did respond pointed out there was no evidence of a problem, although some public authorities saw risks of disorderly markets and market manipulation, especially in distressed markets. There were comments about the difficulty of defining this concept to take into account the various hedging purposes CDS are used for.
4) Restrictions on uncovered short selling.
Most responses argued against any restriction on naked short selling due to lack of evidence of problems, although the issuers association and some public authorities supported requirements for settlement discipline and/or requirements at the point of trading. Many respondents, especially market participants, argued that if a locate rule were to be introduced it should include the variety of arrangements that are currently used for covered short selling (i.e. not require an agreement to borrow in every case).
5) Should there be requirements for buy in procedures?
There was general support for including buy in procedures as this was considered the most appropriate means of dealing with settlement risk.
6) Should there be requirements for marking of orders?
Most responses argued that such a regime would be excessive and costly without adding much benefit, especially as a separate and more comprehensive transparency regime was being proposed by the Commission. One public authority responded that marking of orders was necessary as a complement to the transparency regime and to ensure the enforcement of the short selling regime.
1.4. Exemptions
1) Should there be an exemption for market makers?
On exemptions for market makers there was widespread support for an exemption, although there was some limited opposition to any form of exemption.
2) Should there be an exemption for shares where the principal market is outside the EU?
This question was widely misinterpreted, as most respondents assumed this was a reference to market makers outside the EU.
3) Other issues relating to market efficiency?
There was much comment about the potential adverse effect on market liquidity if adequate exemptions for market makers and primary dealers were not provided.
1.5. Emergency powers for competent authorities
1) Are the emergency powers appropriate?
There was general support for the emergency powers, but many responses (especially from market participants) argued that there needed to be more stringent definition of the powers and when they could be used. Some public authorities called for more flexibility for national authorities in the exercise of emergency powers.
2) Should emergency situations be further defined?
Consistent with the need for more stringent conditions around the use of banning powers there was support (especially from market participants) for more prescription of the circumstances in which powers can be used.
3) Should emergency bans be restricted in time?
The majority agreed that there should be a time restriction. Some thought it should be shorter than the max 3 month period suggested (i.e. with a 3 month extension in exceptional situations). A limited number thought an indefinite ban should be permitted.
4) Are there further measures that would ensure greater coordination of measures in an emergency?
There was support from respondents (although views were mixed among national authorities) for further coordination of emergency measures and for greater clarification of the role of ESMA.
5) Should there be a temporary circuit breaker requirement if prices fall by a certain amount?
Most responses did not see the benefit of such a measure, on the grounds that trading venues already had circuit breaker rules that applied if there was a significant price fall. However, some respondents argued that a circuit breaker should be introduced instead of, or in addition to, emergency powers.
1.6. Enforcement and definitions
1) Are there any special enforcement provisions necessary?
No views were expressed.
2) Are the proposed definitions adequate?
It was suggested that a definition of a 'person' be introduced as this was relevant to funds when calculating a net short position. It was suggested that there be further definition of 'net short position'. There was support for a definition of Market Maker.
9.2.List of contributors to public consultation
Exchanges and Clearers – Total 14
BATS Trading Limited
BME Spanish Exchanges
Chi-X Europe
Deutsche Börse AG
Euroclear
FESE
ICE Futures Europe/ICE Clear Europe
Irish Stock Exchange
KDPW (The Central Polish Securities Depository and Clearing House)
London Metal Exchange
London Stock Exchange
NASDAQ OMX
NYSE Euronext
Stuttgart Stock Exchange
Investors and Associated Bodies - Total 55
ACI
Af2i
AIMA
Allianz SE
Association française des marchés financiers (AMAFI)
Asociación Española de Banca
Association of British Insurers
Association of Private Client Investment Managers and Stockbrokers (APCIMS)
Assogestioni
ASSOSIM
Barclays Capital
British Bankers Association
Bloomberg Tradebook Europe Limited
Bundesverband Alternative Investments
BVI (Bundesverband Investment und Asset Management e.V.)
BWF (Bundesverband der Wertpapierfirmen an den deutschen Börsen e.V.)
Crédit Agricole Cheuvreux.
Danish Mortgage Banks' Federation and the Association of Danish Mortgage Banks, Joint Submission
Danish Shareholders Association
Deutsche Bank AG
EFAMA
Eumedion
European Association of Public Banks
European Banking Federation
European Savings Banks Group
Flow Traders B.V.
Fogain
French Banking Federation (FBF)
Futures and Options Association
German Insurance Association
Getco Europe
Hedge Fund Standards Board
HSBC
IMC Trading B.V
IntesaSanpaolo S.p.A.
Investment Advisors Association
Investment Management Association
Investment Quotient
ISDA, ISLA, AFME Joint Submission
Italian Banking Association
Legal & General Investment Management
Managed Funds Association (MFA)
MEDEF
Nordic Bank, Finance and Insurance Unions (NFU)
Optiver
REB (Raad van de Effectenbranche)
Rivoli Fund Management
Societe Generale
Spanish Association of Investment and Pension Funds (INVERCO)
State Street Bank and Trust Company
Association Française de la Gestion financière (AFG)
The Luxembourg Bankers Association
UniCredit Group
Verband der Auslandsbanken, Association of Foreign Banks in Germany
Zentraler Kreditausschuss
Issuers – Total 2
EuropeanIssuers
ICMA
Member States and Financial Agencies – Total 19
Autorité des Marches Financiers
Banca d'Italia
Banque de France
CONSOB
Consultative Committee of the Spanish CNMV
Czech National Bank
Danish Central Bank
Danish Ministry of Economic and Business Affairs
Det Kongelige Finansdepartement (Norwegian Finance Ministry)
ECB Eurosystem
Federal Ministry of Finance, Germany
Finland Ministry of Finance
IMF
Latvian Ministry of Finance
Ministry for National Economy Hungary
Ministry of Economy and Finance, Treasury Department, Italy
Ministère Français de l’Economie, de l’Industrie et de l’emploi
Swedish Financial Authorities Joint Response
UK Financial Authorities Joint Response
Non Financial Institutions – Total 5
Daimler AG
Oesterreichs Energie
Rolls Royce
Volkswagen Group
Man SE
Other – Total 17
Austrian Federal Economic Chamber
Bundesarbeitskammer Österreich
BUSINESSEUROPE
CFA Society of France
CFA Society of the UK
Confederation of Swedish Enterprise
Deutsches Aktieninstitut
Deutscher Industrie- und Handelskammertag
European Trade Union Confederation
Federation of German Business (BDI)
IE Business School
IG BAU
Interest Capturing Systems, LLC
International Centre For Financial Regulation
John Chapman
Stellungnahme des Deutschen Gewerkschaftsbundes(DGB)
World Economy, Ecology & Development
10.Annex 3: Overview of measures in force in eu 27 on short selling/cds
Disclosure of short positions | Ban on naked short selling | Temporary ban on short selling | Restrictions on CDS | Other – flagging, uptick rule | |
Austria | Yes – 0.25% net short position | Yes, temporary ban for protected financial institutions | No | No | No |
Belgium | Yes – 0.25% net short position | Yes – for protected financial institutions | No | No | No |
Bulgaria | Yes, recommendation to reg. mkts | No | No | No | No |
Cyprus | No | No | No | No | No |
Czech Republic | No | No | No | No | No |
Denmark | No | No | Yes for protected financial institutions | No | No |
Estonia | No | No | No | No | No |
Finland | No | No | No | No | No |
France | Yes – 0.25% net short position for protected financial inst. | Yes for protected financial institutions | No | No | No |
Germany | Yes – for selected financials, 0.2% net short positions to reg. and 0.5% to public | Yes for German fin. inst., EU sovereign bonds | No | Yes, ban on naked CDS | No |
Greece | No | Yes temporary ban for all shares on EL stock exchange | Yes for all shares but rescinded 01 September 2010 | No | Yes – flagging and uptick rule |
Hungary | Yes – 0.01% gross short position | No | No | No | No |
Ireland | Yes – 0.25% net short position | No | Yes for protected financial institutions | No | No |
Italy | No | No | Yes but expired 31.07.09 | No | No |
Latvia | No | No | No | No | No |
Lithuania | No | Yes – shares should be owned or borrowed before sale | No | No | No |
Luxembourg | No | Yes in financial institutions | No | No | No |
Malta | No | No | No | No | No |
The Netherlands | Yes – 0.25 net short position | No | Yes but expired 01.06.09 | No | No |
Poland | No | No – although updated requirements enhance settlement | Yes but only on illiquid securities | No | Yes - flagging |
Portugal | Yes – 0.2% net short position to regulator, 0.5% to public | Yes | No | No | No |
Romania | No | No | No | No | No |
Slovakia | No | No | No | No | No |
Slovenia | No | No | No | No | No |
Spain | Yes – 0.2% net short position to regulator, 0.5% to public | Yes | No | No | No |
Sweden | No | No | No | No | No |
UK | Yes – 0.25% net short position | No | Yes for protected fin. inst. but lifted | No | No |
Sources: Measures adopted by CESR Members on short selling, CESR 08/742, 22.09.08, updated: 01. 09.10 and
Clifford Chance, Short selling rules: the global picture, Client Briefing June 2009.
Note: The above table provides only a high level overview of the most recent available measures and specific details should be taken from each Member State's own regulation.
11.Annex 4 – Summary of discussions between DG MARKT and CESR-POL on Short Selling, Paris, 14 April 2010
On 14 April, officials from unit G3 of DG Internal Market and Services of the European Commission met experts on short selling from national regulators at the premises of CESR in Paris for an exchange of views on short selling. The discussion was structured according to a discussion paper circulated in advance by the Commission services. This note summarises the key issues discussed and positions taken by national regulators on each of the main topics discussed. The positions of individual regulators are anonymous at their request.
1. Scope of the legislation
The chairman of CESR-Pol summarised the views of regulators as follows: no Member State has measures in place for short selling of bonds; transparency and reporting measures are needed under MiFID for transactions in bonds and derivatives (including CDS); emergency powers for regulators on short selling should cover bonds and derivatives as well as shares; shares are not the same as bonds or derivatives – bond markets are more global and CDS markets more concentrated.
One regulator said there is some short selling of bonds on their market, but as they don't have flagging of transactions they cannot quantify it; there is no evidence of manipulation or wider risks arising from this which would justify a regulatory response, and they would need evidence of harm to bond market to justify introducing a net short position disclosure regime for bonds. They could envisage mandatory reporting for CDS transactions to trade repositories to enable proper monitoring of possible risks.
Another regulator believes there is evidence of market abuse on bond markets, they have had some investigations on this, but they lack the tools to know where, when and how it occurs. Should we wait for a crisis to occur or anticipate this?
A regulator said there was only anecdotal evidence of problems on bond markets due to the limited tools they have.
One regulator felt disclosure of short positions in bonds would have an adverse effect on liquidity if it was disclosed to the markets.
One regulator argued there was a risk of settlement failure for bonds due to short selling; another responded that such risks would only arise from naked short selling of bonds and this was not an argument for a disclosure regime; another regulator said that settlement failures were not necessarily linked to naked short selling of bonds.
No other regulators had evidence of short selling in bond markets.
2. Naked short selling
CESR-Pol chairman said that regulators positions were as follows: some regulators wanted a total ban; some wanted a strong locate rule; and others wanted a softer locate rule like the US.
He summarised the position as follows: naked short selling poses a risk of settlement failure, this can be dealt with by buy in rules and by sanctions for late settlement imposed by exchanges and clearing houses. If rules on this are not harmonised at EU level there is a risk of regulatory arbitrage.
One regulator argued that settlement risk does exist for short selling of shares, but in the their experience the evidence is that the vast majority of trades settle by the due date, and where they don't, this is less due to short selling but rather due to administrative problems. This regulator does not see a real threat; they have no empirical study on bonds but suspect the same is true as for shares. This regulator does not favour an EU regime for naked short selling, if pressed could envisage a soft locate rule. Requiring day traders to arrange to borrow shares would be totally disproportionate.
Another regulator said they have seen settlement problems due to naked short selling, although this is not the primary cause, and could provide evidence to the Commission. A locate rule should be stronger than the US rule which is too loosely worded. An EU provision should require action to borrow shares before the short sale.
Another regulator favours a locate rule and short sales should be flagged.
One regulator felt naked short selling might increase herding behaviour and destabilise the market, but there is not a big impact on settlement, so they were not sure a locate rule would be efficient – the power to ban in an emergency would be better.
Another regulator favours an obtain rule or failing that a strong locate rule; naked short selling causes stock inflation which could be contained by a strong locate rule.
One regulator supported a US style locate rule, felt it was not necessary to forbid intra day naked short selling.
Another regulator favours enhancing settlement discipline and could think about an uptick rule.
Two other regulators favour a US-style locate rule.
Regulators from ten countries all said they had no experience of naked short selling on their markets or saw no evidence of risks. One Member State had no strong feelings but felt a locate rule would be better than a ban.
One regulator said they have seen some problems before share issues, and they have a US-style locate rule.
One regulator has less than 1% of settlement failure as they have an obligation to settle on time (two other regulators said they have the same).
One regulator argued occasional problems with naked short selling do not justify action; they have imposed penalties in one case of short selling without the intention to settle and have exchange rule for settlement. Most short selling in this regulator's experience is intra-day with participants flat at the end of the day, so there is no settlement needed and no justification for a locate rule in these cases. If there are problems this is a matter for settlement rules; any locate rule should be a soft one and apply only to positions held at the end of the day.
One regulator has flagging of short sales and an uptick rule, and naked short selling is banned since May 2009.
One regulator felt that all were agreed on a locate rule, but argued this should comprise an obligation to borrow, with a contract.
When asked by the Commission services whether regulators supported an EU level rule obliging settlement by the due date as an alternative to a locate rule – one regulator agreed saying that a study of the economic literature by CEMA argued naked short selling was best dealt with by sanctions for settlement failure. Another disagreed, saying action should be at the trading point since naked short selling has a leverage effect out of proportion to the share capital of the company concerned. Another favoured stronger settlement discipline but as an addition, not as an alternative, to a locate rule.
3. Naked CDS
One regulator said their corporate bond market is small so the CDS market is also likely to be small; they have no evidence of risks from naked CDS but there is a lack of transparency; they see no technical grounds for action, only if it is abusive, but this is a political issue.
One regulator said they have a huge CDS market with many players, but lack data to say if there is an effect on the market so difficult to say if there are risks. If there was evidence of manipulation of sovereign debt markets, they would favour action, which could be a total ban on CDS, or certain holding periods, or banning naked CDS (although the latter would be very difficult to define).
One regulator said they have daily reporting of CDS transactions and have seen no evidence of wrongdoing which cannot be dealt with by MAD or transparency and reporting regimes under MiFID.
Another regulator felt CDS could be a vehicle for market abuse and is currently investigating a suspected case involving corporate CDS just before a market operation. CDS should therefore be included in the MAD; as there is little knowledge of CDS they favour a more transparent regime in the future derivatives legislation and they could envisage flagging of net short CDS positions. Banning naked CDS would not be useful.
Another regulator agreed there was no case to ban naked CDS, it would be difficult to define and there is no evidence of abuse or wider risks. CDS could be a vehicle for market abuse as could any financial instrument. This regulator opposes flagging as it would be very expensive for participants and difficult to operate, but they support the transparency and clearing initiatives foreseen by the Commission.
Thirteen regulators said they had no market for CDS or no evidence of risks but would support the power to ban CDS temporarily in an emergency.
One regulator said there have been allegations in the media of manipulation of certain sovereign bonds using CDS which they are investigating. Transparency and reporting under MiFID is needed and if those requirements are not met there should be a ban; the alternative would be to impose margin and capital requirements.
The Commission services summed up that all Member States supported increased transparency for CDS through transparency and reporting requirements under MiFID and more centralised clearing, and some would favour a flagging requirement.
4. Emergency powers
One regulator agreed emergency powers for national regulators to restrict or ban short selling were needed. Declaration of an emergency should be left to national discretion, not an EU-wide decision, and criteria should not be defined rigidly ex ante as this could prevent a regulator from acting quickly. There should be an obligation to notify other regulators, if necessary after the event, but consultation or coordination mechanisms would not work in an emergency; also not all Member States may need to act. The duration of an emergency should be restricted or it would risk becoming permanent – one regulator has a 3 month period which can be renewed once. This regulator was sceptical of the revised uptick rule of the SEC, as they felt there was no case for an automatic requirement as market conditions were too different.
Thirteen regulators agreed with this position, with some nuances. One wanted flexible criteria for defining an emergency and did not want to go beyond consultation of ESMA members. Another was concerned that the article 10 procedure in the ESMA regulation would prevent action, and that the duration of an emergency could not be defined ex ante. Another agreed, and said that the US circuit breaker approach should be studied.
One regulator said regulators needed similar powers and a long term legal base to act in an emergency, and favoured a role for ESMA.
One regulator favours national discretion on declaration of emergency, but with consultation or notification of ESMA members as a minimum. This regulator doubts the efficiency of a circuit breaker, this should be analysed as it may not be appropriate for our markets.
Another regulator agreed that it was doubtful if a circuit breaker would work and pointed out that many trading venues in Member States have mechanisms in place which have the same purpose as a circuit breaker of slowing the fall in share prices (e.g. volatility auctions when stock falls more than (5%)). When there is an emergency ESMA must play a role, should have a coordinating role before national measures. A distinction should be made between EU and national emergencies.
One regulator agreed that Europe has other mechanisms than a circuit breaker, a quantitative trigger is not appropriate. Member States should have broad powers in an emergency, there should be no inconsistency with ESMA regulation article 10, but we should allow for a differentiated approach by Member States.
Another regulator agreed stock exchanges already have quantitative triggers to stop trading, and a 10% drop happens more frequently in some markets than others.
One regulator argued a circuit breaker would need some coordination to be considered, so that trading in other markets would be considered.
Another regulator favoured a coordinating role for ESMA and suggested an MoU.
5. Powers of regulators
One regulator called for the power to ban; to introduce tighter disclosure requirements; to gather information from market participants; also full investigative powers and powers to compel people to comply and impose penalties for breaches; powers to cooperate with other regulators. Sanctions should be sufficient to punish and deter, and these should not be limited – in this Member State sanctions are unlimited. Another regulator disagreed saying sanctions should be limited for constitutional reasons.
One regulator called for a flagging regime to enable surveillance. Another proposed a harmonised reporting system including client IDs.
One regulator called for the power to suspend trading and the license of intermediaries to be defined in the regulation.
Another agreed saying they suspend licenses of intermediaries due to failures to deliver. This regulator also argued that saying sanctions should have a deterrent effect would not be enough to ensure convergence. We should not fix a maximum sanction, but a minimum. The publication of measures would also be important for enforcement.
One regulator argued that the EU regime should harmonise these powers.
Another called for the power to investigate, and to sanction also actions in other territories, and said sanctions should be large enough to deter.
One regulator said there should be a power to suspend trading in all or some instruments.
Another regulator said powers should be broadened so they cover CDS, or they could not take action in an emergency. This regulator said it could not ban naked short selling for a dual listed share if another Member State introduced such a ban.
One regulator said its emergency measures banned all short selling in financial instruments anywhere in the world, and this would need the help of other competent authorities to enforce.
The Commission services asked about concerns about extra-territoriality – one regulator said this was a concern, but the power was needed in a global market. Two other regulators agreed.
6. Coordination with third countries
One regulator said this was essential for any measures on CDS or these could be circumvented. Agreements could be bilateral or through IOSCO. If EU goes for a locate rule we should seek international agreement on this approach through IOSCO.
One regulator proposed a Memorandum of Understanding (MoU) with third countries. Another pointed out that MoUs are helpful but don't supersede national law, so all regulators should have the possibility to cooperate with other countries.
Two regulators proposed an ESMA-third country master MoU, but stressed that cooperation has to be between national authorities and third countries.
12.Annex 5 – List of organisations represented in meeting of commission services with market participants and public authorities on credit default swaps
5 March 2010
Morning Session
AMF
BaFin
Banque de France / ECB
Bundesbank
CESR
French Treasury
FSA
Hellenic FSA
NBB/BNB
NY FED
European Commission, officials from DG MARKT and DG ECFIN
Afternoon session
AIMA
Assenagon Fund Management
Axa Investment
Barclays Capital
BlackRock
Brevan Howard
Deutsche Bank
Fitch
Hypo Real Estate
ICMA
ING
Ithuba Capital
JP Morgan
Katholieke Universiteit Leuven
BNB/NBB
Banque de France / ECB
European Commission, officials from DG MARKT and DG ECFIN
13.Annex 6 – regulatory regime in the united states on short selling
A number of regulatory developments which are relevant in the context of this initiative have occurred in the United States in recent months. Before coming to those developments, it is necessary to recall the broad regulatory regime which the Securities and Exchange Commission (SEC) has applied to short selling over many years144. Following the 1929 stock market crash, the SEC was established by the Securities Exchange Act of 1934145 and, under the authority granted by this Act, issued in 1937 rule 10a-1, known as the 'uptick rule'. This limited short sales on national exchanges to those executed at an uptick (increase in price) over the last price at which the preceding sale was executed.
This uptick rule was adapted in several ways over the years but remained in force until 2007. In 2004, the SEC adopted Regulation SHO146 which introduced a number of additional requirements for short selling. In particular, Regulation SHO introduced a locate rule for short sellers, requiring them, before entering into a short sale, to borrow the share, have an agreement in place to borrow the share or have a reasonable belief that the share can be borrowed and delivered by the settlement date. Regulation SHO also required broker-dealers to mark sales in all equity securities 'long,' 'short,' or 'short exempt' and included a 'close-out' requirement, obliging brokers to purchase securities to close out a short position if this position had remained open for at least 13 days beyond the 3 day settlement period. This requirement originally only applied to certain securities that had large and persistent 'fails to deliver', i.e. which failed to be settled within the normal settlement period of three days.
In July 2007, the SEC abolished the uptick rule, after several years of analysis. Following a long consultation period in 2009, on 24 February 2010 the SEC adopted a new short sale rule, rule 201 of Regulation SHO, known as the 'revised uptick' or 'circuit breaker' rule. This rule restricts short sales of a share whose price has fallen by more than 10% compared to its closing price the previous day. The rule restricts the display or execution of a short sale order in such shares to a price above the national best bid for the rest of the trading day and the next trading day. The exceptions to the rule are limited, as there are no exemptions for bona fide hedging or market making147.
During the financial crisis, the SEC introduced a number of temporary emergency measures. On 15 July 2008, the SEC introduced a temporary 'pre-borrow' requirement on short selling of shares in 19 systemically important financial institutions148. On 18 September the SEC imposed a temporary ban on short sales of 799 financial stocks149, which grew to 1000 issuers. In September 2008 the SEC also tightened the 'close-out' requirement of Regulation SHO so that it applied to all securities not settled the day after the normal settlement date, and imposed temporary reporting requirements on certain short sales and positions on institutional money managers with assets under management of USD 100 million or more.150
On 1 August 2009 temporary rule 10a-3T, which required disclosure to the authorities of certain aggregated data on short selling transactions, expired. Since then, the SEC has been working with self-regulatory organisations to make short selling volume and transaction data available to the public through the latter's web sites.
Regarding CDS, these fall within the scope of the Wall Street Reform Act enacted by President Obama on 21 July 2010, and the CFTC and SEC will be expected to produce joint rules to implement this. In particular, the Act requires the CFTC and SEC to adopt rules on real time disclosure to the public of swap transactions, including price and volume; clearing of all swaps that the CFTC or the SEC determine should be cleared, except for commercial end-users; and all swaps subject to clearing should be traded on a securities exchange or similar facility151.
The Act also includes certain provisions on short selling, notably it requires the SEC to adopt rules for public disclosure, at least monthly, of the amount of short sales by institutional investment managers152. It also obliges broker-dealers to notify their clients that they can opt not to allow their securities to be lent out for short selling, explicitly makes manipulative short selling illegal and allows the SEC to adopt rules to enforce this provision. The Act also requires the SEC to carry out a cost-benefit study within a year on real time reporting of short sale positions and another study within two years on the state of play following the enactment of short selling rules.153
14.Annex 7 – bibliography of recent economic literature on short selling, committee on economic and market analysis (CEMA), CESR
Date: 10 June 2010
Ref: CESR/10-334
CEMA – Short selling WG
Recent economic literature on short selling and “naked CDS”
Academic economic literature on short selling
Working papers on short selling/ short selling measures
Naked short selling
Boulton, Thomas J., and Marcus V. Braga-Alves, 2009, Naked short selling and market returns, Working paper, Miami University.
Culp, Christopher, and J.B. Heaton, 2008, The economics of naked short selling, Regulation, Spring.
Fotak, Veljko, Vikas Raman, and Pradeep K. Yadav, 2009, Naked short selling: The emperor’s new clothes?, Working paper, University of Oklahoma.
Murphy, Austin, Hong Qian, and Yun Ellen Zhu, 2010, “An Empirical Examination of the Relationship between Naked Shorting and Share Prices Around the Announcement of a Firm’s Need for External Capital”, Oakland University - School of Business Administration
Stone, Elizabeth C. “Fails to Deliver: The Price Impact of Naked Short Sales”, Stanford University.
Putnins, Talis Janis, 2010, “Naked Short Sales and Fails to Deliver: An Overview of Clearing and Settlement Procedures for Stock Trades in the US”, Stockholm School of Economics.
Moffett, Clay M., Robert Brooks, and Jin Q Jeon , 2010, “The Efficacy of Regulation SHO in Resolving Naked Shorts”, University of North Carolina Wilmington, Cameron School of Business.
Short-sale bans 2008
Autore, Don M., Randall S. Billingsley, and Tunde Kovacs, 2009, Short sale constraints, dispersion of opinion, and market quality: Evidence from the short sale ban on U.S. financial stocks, Working paper, Florida State University.
Boehmer, Ekkehart, Charles M. Jones, and Xiaoyan Zhang, 2009, Shackling short sellers: The 2008 shorting ban, Working paper, Texas A&M University.
Boulton, Thomas J., and Marcus V. Braga-Alves, 2009, The skinny on the 2008 naked short sale restrictions, Working paper, Miami University.
Gagnon, Louis, and Jonathan Witmer, 2009, Short changed? The market’s reaction to the short sale ban of 2008, Working paper, Queen’s University.
Gruenewald, Seraina, Alexander F. Wagner, and Rolf H. Weber, 2009, Short selling regulation after the financial crisis – First principles revisited, Working paper, University of Zurich.
Grundy, Bruce D., Bryan Lim, and Patrick Verwijmeren, 2009, Do option markets undo restrictions on short sales? Evidence from the 2008 short-sale ban, Working paper, University of Melbourne.
Khanna, Naveen, and Richmond D. Mathews, 2009, Bear raids and short sale bans: Is government intervention justifiable?, Working paper, Duke University.
Kolasinksi, Adam, Adam V. Reed, and Jacob R. Thornock, 2009, Prohibitions versus constraints: The 2008 short sales regulations, Working paper, University of Washington.
Lobanova, O., S.S. Hamid, and A.J. Prakash, 2010, 'The Impact of Short-Sale Restrictions on Volatility, Liquidity, and Market Efficiency: The Evidence from the Short-Sale Ban in the U.S.' Department of Finance, College of Business, Florida International University.
Short selling since 2008
Avgouleas, Emilios, 2010, “A New Framework for the Global Regulation of Short Sales: Why Prohibition is Inefficient and Disclosure Insufficient” University of Manchester - School of Law
Autore, Don M., Thomas J. Boulton, and Marcus V. Braga-Alves, 2010, “Failures to Deliver, Short Sale Constraints, and Stock Overvaluation”, Florida State University - College of Business.
Battalio, Robert H., and Paul H. Schultz, 2010, “Regulatory Uncertainty and Market Liquidity: The 2008 Short Sale Ban's Impact on Equity Option Markets” University of Notre Dame - Department of Finance
Berber, Allesandro, and Marco Pagano, 2010, Short-Selling bans around the world: evidence from the 2007-09 crisis, CSEF working paper.
Blocher, Jesse, Joseph Engelberg, and Adam V. Reed, 2009, The long short wars: Evidence of incentive driven short selling by hedge funds, Working paper, UNC at Chapel Hill.
Boehmer, Ekkehart, and Julie Wu, 2009, Short selling and the informational efficiency of prices, Working paper, Texas A&M University.
Bardong, Florian, Sohnke M. Bartram, and Pradeep K. Yadav, 2008, Are short-sellers different from ‘regular’ trades?, Working paper, Lancaster University.
Boehmer, Ekkehart, Bilal Erturk, and Sorin Sorescu, 2008, Why do short interest levels predict stock returns?, working paper, Texas A&M University.
Boehmer, Ekkehart, Charles M. Jones, and Xiaoyan Zhang, 2008, Unshackling short sellers: The repeal of the uptick rule, Working paper, Texas A&M University.
Bris, Arturo, 2008, Short Selling Activity in Financial Stocks and the SEC July 15th Emergency Order, Working paper, IMD, European Corporate Governance Institute, and Yale International Center for Finance.
Comerton-Forde, Carole, and Tālis J. Putniņš, 2009: Are short sellers manipulating the market?, University of Sydney
Edwards, A. and K. Hanley, 2008, Short Selling in Initial Public Offerings, Working Paper.
Engelbert, Joseph E., Adam V. Reed, and Matthew C. Ringgenberg, 2010, How are shorts informed? Short sellers, news, and information processing, Working paper, University of North Carolina.
Fox, Merritt B., Lawrence Glosten, and Paul Tetlock, 2009, Short Selling and the News: A Preliminary Report on an Empirical Study, Working Paper.
Hansson, Fredrik and Erik Rüdow Fors, 2009, Get Shorty? Market Impact of the 2008-09 U.K. Short Selling Ban, University of Gothenburg.
Harris, Lawrence, Ethan Namvar, and Blake Phillips, 2009, Price Inflation and Wealth Transfer During the 2008 SEC Short-Sale Ban, Working paper, University of Southern California.
Henry, T.R., D.J. Kisgen, and J. Wu, 2010, 'Do Equity Short Sellers Anticipate Bond Rating Downgrades?', University of Georgia.
Khan, Mozaffar, and Hai Lu, 2009, Do short sellers front-run insider sales?, Working paper, MIT.
Lioui, Abraham, 2009, The undesirable effects of banning short sales, EDHEC working paper.
Lioui, Abraham, 2010, 'Spillover Effects of Counter-Cyclical Market Regulation: Evidence from the 2008 Ban on Short Sales'
Purnanandam, Amiyatosh, and Nejat Seyhun, 2009, Do short sellers trade on private information or false information?, Working paper, University of Michigan.
SEC Office of Economic Analysis Memorandum, January 14, 2009, Analysis of the July Emergency Order Requiring a Pre-Borrow on Short Sales,
Saffi, Pedro A., and Kari Sigurdsson, 2008, Price efficiency and short selling, Working paper, IESE Business School.
Wang, Jun, Steven X. Wei, and Bohui Zhang, 2008, Short sale constraints and price informativeness, Working paper, Hong Kong Polytechnic University.
Westerholm, P. Joakim, and MD Mostafa Ahmed, 2009, The Impact of Actual Market Wide Short Selling Restrictions and Liquidity Constraints During the Financial Crisis of 2008, University of Sidney
Published / accepted papers on short selling/ short selling measures
Goldstein I. and Guembel A. (2008). Manipulation and the allocation role of prices. Review of Economic Studies, vol 75, 133-164.
Henry, Tyler R., and Jennifer L. Koski, 2009, Short selling around seasoned equity offerings, Review of Financial Studies, 3rd round.
Karpoff, Jonathan M., and Xiaoxia Lou, 2010, Short sellers and financial misconduct, Journal of Finance, forthcoming.
Edwards, Amy K., and Kathleen Weiss Hanley, 2010, Short selling in initial public offering, Journal of Financial Economics, forthcoming.
Chen, Crystal Xiaobei, and S. Ghon Rhee, 2010, The impact of short sales on the speed of price adjustment: Evidence from the Hong Kong stock market, Journal of Banking & Finance 34, 471-483.
Christophe, Stephen E. , Ferri, Michael G., and Hsieh, Jim, 2010, Informed trading before analyst downgrades: evidence from short sellers, Journal of Financial Economics 95, 85-106.
Boehmer, Ekkehart, Brad Jordan, and Zsuzsa Huszar, 2010, The good news in short interest, Journal of Financial Economics, in press.
Regulatory literature on short selling
FSA, 2009, Short selling, Discussion paper.
IOSCO, 2009, Regulation of short selling, March.
SFC (Securities and Futures Commission), 2009, How short-selling activity affects liquidity of the Hong Kong stock market, April
Literature on short selling from market participants
Clifton, Matthew, and Mark Snape, 2008, The effect of short-selling restrictions on liquidity: Evidence from the London Stock Exchange, report commissioned by the London Stock Exchange.
Deutsche Bank, 2010, Short selling - Important business in need of globally consistent rules
Marsh, I., Niemer, N., 2008: The Impact of Short Sales Restrictions, International Securities Lending Association, (ISLA) the Alternative Investment Management Association (AIMA) and London Investment Banking Association (LIBA)
Oliver Wyman, 2010, Effects of short selling disclosures
Literature on “naked CDS”
Duffie, Darrell, 2010: Is There a Case for Banning Short Speculation in Sovereign Bond Markets?, Stanford University
Portes, Richard, 2010: Ban Naked CDS, London Business School and CEPR
15.Annex 7 – Data on shares outstanding on loan as a proxy for volume of short selling transactions in the European Union, provided by dataexplorers
16.Annex 8 – Estimate of compliance costs
The Commission services assessed the compliance costs of the preferred options. To assist the policy development process, DG MARKT issued a survey on the expected impacts of different policy options to firms with significant short selling activities and received 50 responses. We also received 20 Member States/ national competent authorities' responses with non-attributable background information on short selling.
Currently there are eleven Member States with short selling disclosure regimes (see Annex 3).
The preferred options include requirements for firms to ensure regulators and markets obtain data on short positions (including CDS):
- Disclosure of short positions to the regulator;
- Disclosure of short positions to the public.
As part of this option, regulators would be expected to provide quarterly reports to ESMA on the net short positions notified or disclosed within their jurisdiction. This section provides an estimate of the compliance costs (including both one-off and ongoing costs) implied by these requirements.
1. Compliance costs for firms subject to the disclosure requirements
It is assumed that the compliance costs are best described by the following elements:
- One-off costs for the development for training, setting up procedures and establishing Information Technology/Information Systems (IT/IS). These costs are assumed to depend on the complexity of the entity subject to the requirement.
- Ongoing costs for training, procedures and maintenance of IT/IS systems. These costs are assumed to be proportional with the one-off costs, and hence also depend on complexity of the entity.
- Ongoing costs depending of the number of notifications. This primarily concerns the time spent once the systems have flagged the need to perform a notification.
In order to estimate the one-off costs, firstly the number of reporting firms on short position is estimated, and secondly the average cost per firm (according to its complexity) is estimated. The firms are grouped into two categories: complex cross-border firms and other firms. The complex cross-border firms are characterised as global financial market players operating in all EU Member States. Other firms are all small and medium sized firms operational in one or few Member States.
The ongoing cost for maintenance of IT/IS systems, the due-diligence process, training and setting up procedures is assumed, on an annual basis, to be a fraction of the one-off-costs in that regard. The remaining ongoing costs are assumed to depend on the number of actual notifications. The requirement for publication when the 0.5% threshold is crossed is not accounted for separately, since this threshold is also requires a notification to the regulator.
1.1 One-off costs IT/IS development, training and compliance procedures
Based on the responses provided by the survey, an estimated percentage of investment firms, credit institutions authorised to provide investment services and their foreign branches subject to these disclosure requirements would consequently amount to around 12% (see table below). This assumption is based on the average of percentage of entities notifying short positions in the UK and Spain at the moment.
Table 1 – Ratio of firms subject to the disclosure requirements and being potentially exposed to compliance costs
The total number of investment firms, credit institutions authorised to provide investment services and their foreign branches is 11,999. We estimate that there should be approximately 1,440 firms (11,999 x 0.12) in the EU subject to the disclosure obligations. We also estimate that there may be 10%154 of complex cross-border firms and 90% of other firms. Furthermore, we extrapolate from the results of the responses provided from some market participants and from some regulators to make an assumption that the average costs for IT/IS development, training and compliance procedures could be estimated at approximately 500,000 Euro for complex cross-border firms and 50,000 Euro for other firms155.
Table 2 – Estimation of costs of IT/IS development, training and compliance procedures
1.2 Estimate of on-going costs based on national reporting requirements for financial institutions only
Based on the responses provided by the survey, we identify two types of on-going costs:
1. IT/IS upgrade and maintenance.
2. Reporting of short positions.
We estimate the IT/IS upgrade and maintenance costs to be 10% of the initial cost of IT/IS development (including training and compliance procedures) for all the EU reporting entities.
Table 3 – Estimates of IT/IS upgrade and maintenance costs (on a yearly basis)
Furthermore, an estimate of disclosure frequency is based on the relevant data provided by six national competent authorities. Five out of six national competent authorities have in recent months applied disclosure rules providing for a threshold of 0.25% of net short positions. Under this requirement, the disclosure of short positions to the relevant national competent authority was done after the crossing of the threshold. As the regulators have provided data for variable periods (see Table 4, disclosure period provided by regulators), we recalculated the disclosure numbers on a yearly basis.
Table 4 – Estimates of yearly disclosure numbers per Member State (MS) and yearly average disclosure number for six MS
Furthermore, we apply a grouping methodology to estimate the disclosure numbers of short positions for all the EU Member States. The countries are grouped into three categories according to the likely frequency of disclosures per year: high, medium and low. Based on available information (see table 4) and also based on our extrapolated number of reporting firms subject to disclosure regulation (see table 5, columns 1 and 2), we extrapolate the assumed weight of short positions to be disclosed and allocate the potential number of disclosures to each Member State (see table 5, column 5).
Table 5 – Assumed exposure to short positions and yearly average disclosure number (sorted in ascending order)
We estimate that reporting disclosure information will take 2 hours. The estimate is based on information collected from some market participants.
As regards the costs (hourly wage) we did not obtain relevant figures about the distribution from across the Member States. However, taking into consideration that a large proportion of notifications are likely to be made by complex cross-border entities, often with a centralised compliance function located in a high-cost country, it appears appropriate to apply an average hourly wage of managers from such a Member State. The hourly wage of managers156 in the UK is estimated to € 52.81, and is used in the following calculations.
Table 6 – Calculation of compliance cost of disclosure for Short position requirements for shares in financial institutions only
1.3 On-going costs for disclosure of short positions in all EU shares
Given that under the proposed option all EU shares will be subject to the disclosure regime of short positions so the number of disclosures will increase. The number of daily disclosures will vary according to the disclosure regime using the 0.2%, 0.3%, 0.4% and 0.5% threshold levels.
Table 7 – Statistical Findings in the UK157
A survey in the UK provides a basis for estimating the number of disclosures in the case where a broader set of shares are part of the regime and where multiple thresholds are used. As previously stated the actual number of notifications in the UK on an annual basis was 156 with a single threshold for 0.25%. The number of notifications at 0.2% and 0.3% is significantly higher most probably due to the broader scope of shares. The average of the notifications at these thresholds is 392 [(472 + 312)/2], indicating that the overall number of notifications would increase by a factor of 2.5 [(392/156)] due to the broader scope. This is reflected in the calculation (see tables 8 and 9). In addition the total number of notifications is likely to increase with the higher number of thresholds. The UK survey gives an indication of the relative frequency with which the various thresholds are likely to trigger a notification. On this basis, the factor with which the number of notifications will increase due to the increase to 4 thresholds instead of 1 can be estimated to 2.86 [(472 + 312 + 180 + 156)/392].
These results have been used for our extrapolation to get an estimate of the yearly disclosures all 27 EU regulators might receive (see table 8).
Table 8 – Estimates for reporting on-going costs per threshold (0.2; 0.3; 0.4; 0.5)
We estimate that the on-going compliance cost for disclosure of short positions will be approximately 2.1 million Euros (see calculation details in table 10).
1.4 On-going costs for disclosure of short positions by regulators to ESMA
The regulators/national competent authorities will be required to disclose quarterly short positions to ESMA. We estimate that reporting disclosure information will take 2 hours. In order to calculate the administrative cost, we use the hourly wages158 of manager in EU Member States (see table 9).
Table 9 – Estimates for reporting to ESMA on-going costs
17.Annex 9 – Table providing an overview of the scope of preferred options
The below table shows the instruments which are in scope for each of the preferred options (indicated with a Y).
Financial Instruments Preferred options | EU Shares | Derivatives relating to EU Shares | EU Sovereign Debt | Derivatives relating to EU Sovereign Issuers | CDS relating to EU Sovereign Issuers | Other Financial Instruments |
Transparency | ||||||
Notification to competent authorities of significant net short positions in shares | Y | Y | - | - | - | - |
Requirement to mark a short order on a trading venue as 'short' | Y | - | - | - | - | - |
Public disclosure of significant net short positions in shares | Y | Y | - | - | - | - |
Notification to competent authorities of significant net short positions in sovereign debt and credit default swaps | - | - | Y | Y | Y | - |
Information provided (quarterly) to ESMA | Y | Y | Y | Y | Y | - |
Uncovered Short Sales | ||||||
Restrictions on uncovered short sales | Y | - | Y | - | - | - |
Buy in procedures and fines for late settlement | Y | - | Y | - | - | - |
Exemptions | ||||||
Exemption when the principle trading venue is outside the Union | Y | Y | - | - | - | - |
Exemption for market making and primary market operations | Y | Y | Y | Y | Y | Y |
Powers in exceptional situations | ||||||
Disclosure in an exceptional situation | - | - | - | - | - | Y |
Restrictions on short selling in an exceptional situation | Y | Y | Y | Y | Y | - |
Restrictions on Credit Default Swap transactions in an exceptional situation | - | - | - | - | Y | - |
Power to temporarily restrict short selling of financial instruments in case of a significant price fall (10%) | Y | Y | Y | Y | Y | Y |
1Annex 3 includes a table giving an overview of the situation on key measures in the EU 27 taken from a note by CESR setting out the measures on short selling taken by Member States.
2For example, in Discussion Paper 09/1, Short Selling, February 2009, p. 3, the UK Financial Services Authority explained their reasons for introducing temporary short selling measures in September 2008 as follows: 'We did this at a time of extreme market turbulence, manifested in the forms of high and prolonged price volatility and downward pressure on the prices of financial stocks in particular. We were concerned by the heightened risks of market abuse and disorderly markets posed by short selling in these conditions.'
3CESR is an independent advisory group to the European Commission composed by the national supervisors of the EU securities markets. See the European Commission's Decision of 23 January 2009 establishing the Committee of European Securities Regulators 2009/77/CE. OJ L25, 23.10.2009, p. 18). The role of CESR is to improve co-ordination among securities regulators, act as an advisory group to assist the EU Commission and to ensure more consistent and timely day-to-day implementation of community legislation in the Member States.
4Committee of European Securities Regulators, Model for a Pan-European Short Selling Disclosure Regime, March 2010. The report advocates a two-tier model for disclosure of significant individual net short positions in all shares that are admitted to trading on an EEA regulated market and/or an EEA MTF. At the lower threshold of 0.2%, positions would be disclosed to the relevant regulator, at the higher threshold of 0.5% positions would be disclosed, in addition to the regulator, also to the market as a whole.
5On 8 June 2010, President Sarkozy and Chancellor Merkel wrote to President Barroso urging the Commission to bring forward a legislative initiative on short selling and sovereign CDS encompassing the possibility of an EU wide prohibition of naked short selling and naked sovereign CDS .
6Responses from some national regulators to a questionnaire by the Commission services. The respondents wished their responses to remain confidential. A summary by the Commission services of a meeting with national regulators on short selling is included in annex 5.
7A glossary of key terms on short selling and Credit Default Swaps is included in annex 1.
8In a joint letter dated 10 March 2010 to President Barroso and Prime Minister Zapatero, President Sarkozy, Chancellor Merkel, Prime Minister Juncker and Prime Minister Papandreou wrote: 'We therefore propose that the EU Commission initiates as quickly as possible at European level an inquiry into the role and impact of speculative practices in connection with CDS trading in the government bonds of European countries. Should the inquiry ascertain market abuses or that there is a well-founded suspicion that speculative practices are having a considerable impact on the development of yields, we should quickly examine measures to determine whether they are suitable and, if necessary, pass the appropriate legislation. These examinations should also consider introducing minimum holding periods for CDS trading, banning speculative CDS trading as well as banning the acquisition of CDS which are not being used for hedging purposes.'
9Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Commission Work Programme 2010 – Time to Act, COM(2010) 135 final, Vol. 1, p. 4 and Annex I, p. 4.
10Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the European Central Bank, Regulating financial services for sustainable growth, COM(2010) 301 final, 02.06.2010, p. 7.
11Proposal for a Directive of the European Parliament and of the Council on Alternative Investment Fund Managers and amending Directives 2004/39/EC and 2009/…/EC, COM(2009) 207 final, 30.4.2009.
12Proposal for a Regulation of the European Parliament and of the Council establishing a European Securities and Markets Authority, COM(2009) 503 final, 23.9.2009.
13Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments, OJ L145, 30.04.2004.
14Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (market abuse), L 96/20, 12.04.2003. Article 1.2 defines market manipulation and article 5 requires Member States to prohibit any person from engaging in market manipulation.
15The European Securities Markets Expert Group (ESME) was established by Commission Decision 2006/288/EC setting up a European Securities Markets Expert Group to provide legal and economic advice on the application of the EU securities Directives (30.03.2006). The objective of the Group was to advise the Commission on various issues in relation to the EU securities markets. The Group consisted of 20 members, chosen according to strict procedures to ensure the widest possible range of professional backgrounds (such as lawyers, economists and accountants) and geographical balance so that the different EU markets were covered. This non-binding advice assisted the Commission in fulfilling its tasks. ESME's mandate expired on 31.12.2009 and was not renewed.
16The full text of the report can be found at:
http://ec.europa.eu/internal_market/securities/docs/esme/report_20090319_en.pdf
17CESR/10-088, March 2010, c.f. footnote 4.
18Committee of European Securities Regulators, CESR/08-742, Measures adopted by CESR Members on short selling, 22 September 2008, updated 22 February 2010
19Call for evidence, Review of Directive 2003/6/EC on insider dealing and market manipulation (Market Abuse Directive). For the text of the call for evidence, see:
http://ec.europa.eu/internal_market/consultations/docs/2009/market_abuse/call_for_evidence.pdf
20See http://ec.europa.eu/internal_market/consultations/2010/short_selling_en.htm
21In accordance with the rules for the elaboration of impact assessments the minutes of the last meeting of the steering group have been submitted to the Impact Assessment Board together with this impact assessment.
22Short sellers have been the villains for 400 years, Reuters, 26.09.08
Factbox: milestones in short selling history, Reuters, 16.07.08
23Discussion Paper 09/1, Short Selling, February 2009, Financial Services Authority, pp. 7-9 for a more detailed analysis of who short sells and why.
24Ibid.
25"For the control stocks that are never subject to the shorting ban, NYSE short sales account for a cross-sectional average of 14.12% of NYSE trading volume during the pre-ban period from 1 August through 18 September ", Boehmer, Ekkehart; Jones, Charles; and Zhang, Xiaoyan, Shackling short sellers: the 2008 shorting ban, working paper, 18 November 2008, p. 7. 'Short sellers account for more than 20% of trading volume', Boehmer, Jones and Zang, 2008, quoted in Boehmer and Wu, Short selling and the informational efficiency of prices, 2009, p. 1. 'Some estimates have short sellers responsible for between 20-30% of equity trading volume', Oliver Wyman, The effects of short-selling public disclosure regimes on equity markets, 2010, p. 6.
26Analysis of the Pre-Borrow Emergency Order, Memorandum by the Office of Economic Analysis, 14 January 2009, p. 6. According to the US Office of Economic Analysis, short selling in the 17 NYSE listed financial stocks subject to a pre-borrowing requirement on short sales represented 41.81% of trading volume in those stocks from 12 June to 11 July 2008. This fell to 30.74% after the introduction of the pre-borrow requirement on 15 July 2008.
27In contrast, the United States has a marking regime and provides for reporting of OTC transactions. The latter is an option which will be considered in the context of the review of the Directive on Markets in Financial Instruments. One Member State, the UK, currently requires OTC transaction reporting at national level.
28Response by the UK FSA to Commission services questionnaire on short selling and Credit Default Swaps.
29"Agreement of the Executive Committee of the CNMV on naked short selling, adopted on September 22nd 2008". http://www.cnmv.es/Portal/verDoc.axd?t={0f1fd720-9592-45fc-b293-4fc99e82ce4f}
30Response by the Spanish CNMV to Commission services questionnaire on short selling and Credit Default Swaps.
31Response by the Greek HCMC to Commission services questionnaire on short selling and Credit Default Swaps.
32Financial Times, Lagarde wants look into murky market, 14 February 2010.
33See Depositary Trust and Clearing Company (DTCC) data
34Information provided by regulators and market participants to the Commission services at a meeting on CDS issues, 5 March 2010.
35Beber, Alessandro and Pagano, Marco, Short-Selling Bans around the World: Evidence from the 2007-09 Crisis, CSEF working paper, examines the effect of short selling bans on liquidity by using bid-ask spreads as a measure of liquidity. The paper states, p 21: "Italy emerges as the country where the ban on short sales was associated with the most dramatic deterioration of market liquidity, followed by Denmark, Australia and Norway. The U.S., U.K. and Ireland are in an intermediate group, while in the remaining countries short-selling bans have been associated with comparatively mild increases in bid-ask spreads – in the order of about 50 basis points or less."
36Gruenewald, Seraina; Wagner, Alexander; and Weber, Rolf, Short Selling Regulation after the Financial Crisis – First Principles Revisited, Working Paper, University of Zurich, 2009, p. 16.
37Oliver Wyman, The effects of short-selling public disclosure regimes on equity markets, 2010, p. 29.
38Battalio, Robert and Schultz, Paul, Regulatory uncertainty and market liquidity: the 2008 short sale ban's impact on equity option markets, University of Notre Dame, 2010, p. 9 cite Ofek and Richardson (2003) as suggesting 'that the inability to short led to high prices for internet stocks in 1999 and 2000, and the relaxation of constraints on borrowing shares for shorting led to the eventual collapse of prices for these stocks'.
39Bris, A., Goetzmann,W. N., Zhu, N., Efficiency and the Bear: Short Sales and Markets Around the World, 2007, Journal of Finance, Vol 62, No 3, p. 28.
40FSA, DP09/1 (February 2009), Annex 1 p. 3.
41“Short Selling Study: The Views of Corporate Issuers”, Opinion Research Corporation, October 17, 2008. In this survey of CEOs of NYSE and NASDAQ listed issuers, almost 60% of respondents consider short selling to be harmful to their company’s stock and to their shareholders, and 75% of respondents think that short selling should be prohibited in periods of heightened volatility.
42FSA DP09/1, February 2009, p. 12.
43Ibid, p. 13.
44Source: responses provided by regulators to a questionnaire by the Commission services.
45Joint letter dated 10 March 2010 to President Barroso and Prime Minister Zapatero, from President Sarkozy, Chancellor Merkel, Prime Minister Juncker and Prime Minister Papandreou.
46Some Member States adopted temporary emergency measures banning short selling in financial institutions (e.g. Austria, Denmark, France, Germany, Ireland, Italy, Netherlands, UK), whereas others (e.g. Finland, Sweden) adopted no rules on short selling at all in response to the crisis. Of the Member States which banned short selling in financial institutions temporarily in 2008, Austria, Denmark and France still have those bans in place today. Furthermore, some Member States banned only naked short selling in certain financial institutions (Belgium, Luxembourg, Portugal) or in all issuers (Greece), while another already had a ban on naked short selling in place (Spain).
47CESR report, Model for a Pan-European Short Selling Disclosure Regime, CESR/10-088, March 2010, p. 3.
48CMC rule 1/509/15.5.2009
49General Decree of the Federal Financial Supervisory Authority (BaFin) on the prohibition on contracting a credit derivative or entering into a transaction on the same if a no more than insignificant reduction of the protection buyer’s risk is achieved thereby, of 18 May 2010; General Decree of the Federal Financial Supervisory Authority (BaFin) on the prohibition of uncovered short-selling transactions in debt securities of Member States of the EU whose legal currency is the euro, of 18 May 2010; General Decree of the Federal Financial Supervisory Authority (BaFin) on the prohibition of uncovered short-selling transactions in certain shares of 18 May 2010
50See guidance on HCMC's decision regarding short selling ban:
http://www.hcmc.gr/photos/kefalaiagora/files/Guidance%20on%20Short%20Selling%20Ban_110510.pdf Lifting of ban – see press release: http://www.hcmc.gr/photos/nea/files/Short_Selling_260810.pdf
51Draft law on the Enhancement of Integrity of the Financial Markets, German Ministry of Finance, 25 May 2010
52Projet de loi adopté par l'Assemblée Nationale, de régulation bancaire et financière, TA no. 485, 10 June 2010, chapter 5 article 7bis.
53See http://gss.unicreditgroup.eu/gss/pdf/Poland/Newsflashes/PL20100526_016.pdf and http://www.gpw.pl/zrodla/papierywartosciowe/krotka/pdf/rules-ks.pdf
54CESR, March 2010.
55Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (market abuse), L 96/20, 12.04.2003, articles 1.2 and 5.
56Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments, OJ L145, 30.04.2004.
57Regulation SHO, Securities and Exchange Commission, 17 CFR PARTS 240, 241 and 242 [Release No. 34-50103; File No. S7-23-03] http://www.sec.gov/rules/final/34-50103.htm#P19_2741
58The SEC's New Short Sale Rule: Implications and Ambiguities, Davis Polk Client Memorandum, 08.03.2010, p. 4.
59SEC Release 34-58166, Emergency Order Pursuant to Section 12(k)(2) of The Securities Act of 1934 Taking Temporary Action to Respond to Market Developments, July 15, 2008. Available at: http://www.sec.gov/rules/other/2008/34-58166.pdf.
60SEC, Emergency Order Pursuant to Section 12(k)(2) of the Securities Exchange Act of 1934 Taking Temporary Action to Respond to Market Developments. Release 34-58572. September 17, 2008. Available at: http://www.sec.gov/rules/other/2008/34-58572.pdf.
61The D.E. Shaw Group Market Insights, May 2010, p. 6.
62Davis Polk, Summary of the Dodd-Frank Wall Street Reform and Consumer Protection Act, Enacted into Law on July 21 2010, 21 July 2010, p. 70.
63Article 25 of the Directive on Markets in Financial Instruments.
64COM(2009) 503 final, 23.9.2009.
65Discussion Paper 09/1, Short Selling, February 2009, p. 3, UK Financial Services Authority.
66Ibid, p. 12.
67FSA DP09/1, February 2009, p. 7.
68Cf Letter from France, Germany, Greece and Luxembourg to Mr Barroso dated 10 March 2010
69See European Central Bank report dated June 2009 on the sovereign CDS market
70At the end of May 2010, the outstanding of sovereign CDS stood at USD 2.2 trillion to be compared to an outstanding sovereign bond market of around USD 40 trillion (source: DTCC and BIS)
71See Commission Staff working paper dated 20 October 2009 accompanying Commission communication on a roadmap for ensuring, efficient, safe and sound derivatives markets
72See Public Consultation on Derivatives and Market Infrastructures, launched on 14.06.2010: http://ec.europa.eu/internal_market/consultations/2010/derivatives_en.htm
73Source: responses by national regulators to questionnaire by the Commission services. Unless the origin of the data is clearly indicated, sources are anonymous at the request of the regulators concerned who wished their identity to remain confidential.
74Source: responses by national regulators to questionnaire by the Commission services.
75Source: data provided by Spain in response to a questionnaire by the Commission services.
76Source: responses by national regulators to questionnaire by the Commission services.
77Source: data provided by Spain in response to a questionnaire by the Commission services.
78Source: responses by national regulators to questionnaire by the Commission services.
79Source: response by the UK FSA to a questionnaire by the Commission services.
80Office of Economic Analysis, Analysis of the July Emergency Order Requiring a Pre-Borrow on Short Sales, 14 January 2009, pp. 12-13. According to the OEA study, "prior to the Order, the 19 issuers in the order sample had 2.8 million in fails valued at $64.2 million per day. With the order in effect, these 19 issuers had 1 million in fails valued at $28.4 million per day, representing decreases of about 64% and 56% respectively. The number of securities with fails declined from 12 to 4.5 per day, representing a decline of about 63%. (…) In summary, we found that the Emergency Order led to a large reduction in fails to deliver and significantly reduced the frequency of new fails."
81Article 39 currently requires regulated market operators to have effective arrangements in place to facilitate the efficient and timely finalisation of transactions executed on their systems. A similar requirement for MTF operators can be found in MiFID article 14.5.
82Response by the German Federal Ministry of Finance to the public consultation, p. 5.
83Response by the AMF to the public consultation, p. 4.
84Confidential response by a regulator to the questionnaire by the Commission services.
85See table in annex 3. Concerning disclosure, one Member State required daily disclosure of all short selling transactions to the regulator (Greece, until 1.06.09), others require disclosure when a threshold of 0.25% of the total share capital of certain financial stocks is breached (Belgium, France, Ireland, Spain and UK) and others set a different threshold (Hungary – 0.01%, Greece – 0.10% from 1.06.09). The remaining Member States require no disclosure of short selling transactions at all. Greece and Poland have rules in place requiring the flagging of short sale orders.
86FSA DP09/1, February 2009, pp. 24-25/
87CESR/10-088, March 2010, p. 9 and Regulation of Short Selling – Consultation report, IOSCO Technical Committee, March 2009, p. 14.
88CESR report, Model for a Pan-European Short Selling Disclosure regime, CESR/10-088, March 2010, p. 5: 'CESR considers that improving the transparency of short selling would have distinct benefits which would outweigh the associated costs. It would help deter and constrain particularly aggressive large-scale short selling which may threaten the maintenance of orderly markets or pose the risk of market abuse and provide early warning signs of a build-up of large short positions, thereby alerting regulators to potentially abusive behaviour and enabling them to monitor and take action more effectively.'
89Khan, Moffazar and Lu, Hai, Do Short Sellers Front-Run Insider Sales?, MIT Working paper, find "significantly positive abnormal short sales in the days leading up to large insider sales, and peaking on the day of large insider sales" and they conclude that this 'is consistent with front-running facilitated by leaked information'.
90This risk of market abuse arises because the naked short seller is not constrained by the need to identify and borrow shares from the market before shorting them; this means that naked short sellers can sell as quickly as they can find buyers and can even sell more shares than the total issued share capital of a company. Naked short selling can therefore more easily give a false impression of the supply of shares for sale since the seller does not have to borrow them before the sale; although it should be noted that naked short selling without any intention or reasonable plan to settle the short position might be considered to be market abuse that is already prohibited under the Market Abuse Directive. For example, in one case in the UK, the competent authority took action against a party for market abuse for short selling 252% of the issued share capital of a company - See FSA DP09/1, p. 13.
91FSA DP09/1, February 2009, p. 11.
92Source: responses from national regulators to questionnaire by the Commission services.
93FSA DP09/1 (February 2009), p. 24.
94Source: responses by national regulators to questionnaire by the Commission services.
95Source: responses by national regulators to questionnaire by the Commission services.
96On 20 May 2010, the Dutch parliament adopted a motion calling for a similar ban on naked short selling as that introduced by Germany due to fears that it created "the danger that [speculators] are diverted to the Netherlands", Financial Times, Dutch Parliament calls for short selling ban, 21 May 2010.
97European Financial Management Association (EFAMA), Reply to CESR's Call for Evidence on Regulation of Short Selling by CESR Members, January 20 2009. EFAMA considers that the divergence created regulatory arbitrage possibilities 'due to the many differences in the short selling regimes and the cross-border nature of equity markets', which sometimes caused 'distortions that negatively impacted the trading in specific shares or markets'.
98Emilios Avgouleas, A new framework for the global regulation of short sales: Why prohibition is inefficient and disclosure insufficient, University of Manchester School of Law, p. 18: "The market maker exemption was a source of particularly serious problems. Most Member States either did not have any market maker exemption or uniform definition about the kind of market actors that qualified as market makers. Arguably, the whole episode showed in the most vivid manner how the lack of harmonisation of trading rules keeps the EU from achieving the objective of level playing field, arguably the holy grail of EU securities regulation."
99The Association of British Insurers (ABI) said in its response to CESR's consultation on short sales: 'The disparity of approaches taken not just in the EU but globally has represented a significant cost. For example, one large firm, which operates in several jurisdictions, reports that the annual legal and consultancy fees paid merely for identifying and tracking changes runs to tens of thousands of pounds.' Association of British Insurers (ABI), Regulation of Short Selling by CESR Members – the ABI's Response to CESR 08/1010.
100Source: discussion with national regulators (see annex 4) and individual responses by national regulators to questionnaire by the Commission services.
101For example, the London Stock Exchange operates automatic execution suspension periods which are designed to provide a pause in trading of approximately five minutes when a share price fall breaches a specified threshold, e.g. 5%.
102Directive 2004/109/EC of the European Parliament and of the Council of 15 December 2004 on the harmonisation of transparency requirements in relation to information about issuers whose securities are admitted to trading on a regulated market and amending Directive 2001/34/EC, OJ L390/38, 31.12.2004
103Communication for the spring European Council, Driving European recovery, COM(2009)114.
104Page 7 of the Communication of 2 June 2010 from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the European Central Bank.
105In fact in the United States, the revised uptick rule only applies to shares.
106For a review of the published literature on this issue see Gruenewald, S, Wagner A, Weber R, Short selling Regulation after the Financial Crisis – First Principles Revisited, University of Zurich, 2009, p. 37.
107Ibid, p. 37.
108Financial Services Authority, FS09/4, Short Selling: Feedback on DP09/1, October 2009, p. 11.
http://www.fsa.gov.uk/pubs/discussion/fs09_04.pdf
109See SEC press release: www.sec.gov/news/press/2007/2007-114.htm
110For an overview of empirical studies on uptick rules see Avgouleas, Emilios, A new framework for the global regulation of short sales: why prohibition is inefficient and disclosure insufficient, Stanford Journal of Law, Business and Finance, Spring 2010, p. 55.
111Rene Schultz Journal of Economic Perspective Winter 2010
112Proxy hedges are hedges which cover a non-deliverable obligation that is sufficiently correlated with the deliverable obligations. Examples would be using sovereign CDS to hedge exposure to banks or corporates in the same jurisdiction or to hedge exposure to other sovereigns, for which no sovereign CDS are traded. Since CDS are a synthetic market, it would of course be possible for market participants to write CDS contracts on any 'exotic' reference entity. Since market liquidity for that contract would be low, however, the CDS seller would require a significant premium beyond the credit risk. Therefore, proxy-hedging may be less precise but cheaper.
113Responses to DG MARKT Public Consultation on Short Selling. A number of respondents, including the Association of British Insurers, Danish Central Bank, UK Finance Ministries Joint Submission and the London Stock Exchange Group, noted that there was no clear evidence to support such a ban. The Banque De France also commented that they were not in favour of a ban, and noted the differences in correlation across the sovereign CDS/Bond market. The German Federal Ministry of Finance, AMF and CNMV noted that they believed there was such a correlation.
114Ashcraft, A. and Santos, J. (2009), Has the CDS market lowered the cost of corporate debt?, Journal of Monetary Economics, 56, pp. 514-23.
115Article 6a(5) of the ESMA Regulation as adopted by the European Parliament says that 'the Authority may temporarily prohibit or restrict certain financial activities that threaten the orderly functioning and integrity of financial markets or the stability of the whole or part of the financial system in the Union in the cases specified and under the conditions laid down in the legislative acts referred to in Article 1(2) or if so required in the case of an emergency situation in accordance with and under the conditions laid down in article 10. The Authority may also assess the need to prohibit or restrict certain types of financial activities and, where there is such a need, inform the Commission in order to facilitate the adoption of any prohibition or restriction. The authority shall review its decision at appropriate intervals and at least every three months. If the decision is not renewed after those three months, it shall automatically expire. A Member State can appeal against the decision of the Authority. In that case, the Authority shall decide in accordance with Article 29(1) subparagraph 2, whether it maintains its decision.'
116Beber, Alessandro and Pagano, Marco, Short-Selling Bans around the World: Evidence from the 2007-09 Crisis, CSEF working paper, 2010, p. 27.
117Bris, Arturo, William N. Goetzmann, W.N., Zhu, N., Efficiency and the bear: Short sales and markets around the world, 2007, Journal of Finance, Vol 62, No 3, p. 28.
118FSA DP09/1 (February 2009), annex 3, p. 2.
119In the response of the Alternative Investment Managers Association (AIMA) to a questionnaire by the Commission services, one hedge fund argued that a ban on short selling through the use of derivatives would force it to move outside the EU, and another argued that listings and trading would move to the US and Asian markets.
120Responses to DG MARKT Public Consultation on Short Selling. Stellungnahme des Deutschen Gewerkschaftsbundes and John Chapman.
121IOSCO Final Report on the Regulation of Short Selling, p. 19 http://www.iosco.org/library/pubdocs/pdf/IOSCOPD292.pdf]
122London Stock Exchange Group in its response dated 9 July 2010 to the Commission consultation has indicated that in Q4 2009 approximately 20-25% of trading in FTSE 100 shares was by technical traders whose strategies may include market making activities.
123CESR/10-88 (March 2010), p. 6.
124Ibid, p. 6.
125FSA, DP09/1 (February 2009), annex 3, p.7.
126CESR/10-88 (March 2010), p. 9.
127CESR/10-088 (March 2010), p. 5.
128Ibid, p. 6.
129Ibid, p. 8.
130FSA, FS09/3: Short Selling, October 2009, p. 16.
131Oliver Wyman Inc, The effects of short selling public disclosure regimes on equity markets, 2010, pp. 4-5.
132FSA, FS09/3: Short Selling, October 2009, p. 16.
133AIMA, BBA, ISDA, ISLA, AFME, Dutch Ministry of Finance, Danish Central Bank and VW group.
134EuropeanIssuers, Response to European Commission's Public Consultation on Short Selling, 9 July 2010, p. 3.
135CESR/10-088 (March 2010), p. 8.
136CESR/10-088 (March 2010), p. 8. Also, in their joint response to the public consultation, the Association of Financial Markets in Europe, (AFME) the International Securities Lending Association (ISLA) and the International Swaps and Derivatives Association (ISDA) stated that their prime broker member firms had 'confirmed that their clients avoid short selling beyond disclosure thresholds to avoid signalling their trading strategies to the broader market' (p. 8).
137CESR/10-088 (March 2010), p. 9.
138Ibid, p. 9. These thresholds were decided following a public consultation, an informal survey by one regulator and discussions with all regulators. CESR considers that the public disclosure threshold should be higher than that introduced by many regulators in 2008 as it would apply to all shares, not just financial sector shares, and the lower threshold for reporting to the regulator would mean that regulators would already have had advance notice of building short positions in the issued share capital of a particular issuer and would already have had the opportunity to consider their response.
139Alternative Investment Managers Association, European Commission's Public Consultation on Short Selling, 13 July 2010, p. 2. Also AFME, ISLA and ISDA, p. 7.
140CESR/10-088 (March 2010), p. 6.
141CESR/10-088, pp. 10-11.
142According to the analysis of corporate bond market activity in 2008 in a response by the Association for Financial Markets in Europe to a consultation by CESR on the Directive on Markets in Financial Instruments dated 4 June 2010: the average percentage turnover of the most liquid corporate bonds compared to issuance size was 121%, compared to an average of 257% for the total European equity market in 2008 (based on data by the Federation of European Stock Exchanges).
143Explanations are adapted inter alia from Key points about Regulation SHO, Division of Market Regulation, Securities and Exchange Commission, 11 April 2005.
http://www.sec.gov/spotlight/keyregshoissues.htm
144This overview of the current regulatory regime applicable to short selling in its historical context is taken from the following publication: Short Circuit: New Restrictions on Short Selling in U.S. Equity Markets, The D.E. Shaw Group Market Insights, May 2010, Vol. 2 No. 2.
145The Securities Exchange Act of 1934, 6 June 1934, ch. 404, title I, Sec. 1, 48 Stat. 881.
146Regulation SHO, Securities and Exchange Commission, 17 CFR PARTS 240, 241 and 242 [Release No. 34-50103; File No. S7-23-03] http://www.sec.gov/rules/final/34-50103.htm#P19_2741
147The SEC's New Short Sale Rule: Implications and Ambiguities, Davis Polk Client Memorandum, 08.03.2010, p. 4.
148SEC Release 34-58166, Emergency Order Pursuant to Section 12(k)(2) of The Securities Act of 1934 Taking Temporary Action to Respond to Market Developments, July 15, 2008. Available at: http://www.sec.gov/rules/other/2008/34-58166.pdf.
149SEC, Emergency Order Pursuant to Section 12(k)(2) of the Securities Exchange Act of 1934 Taking Temporary Action to Respond to Market Developments. Release 34-58572. September 17, 2008. Available at: http://www.sec.gov/rules/other/2008/34-58572.pdf.
150The D.E. Shaw Group Market Insights, May 2010, p. 6.
151Davis Polk, Summary of the Dodd-Frank Wall Street Reform and Consumer Protection Act, Enacted into Law on July 21 2010, 21 July 2010, pp. 52-63.
152Ibid,, p. 70.
153Ibid, p.68.
154The 10% estimate is reached by considering the total number of firms that are significantly more complex in their operations than the rest. This is typically large investment banks and other entities with a variety of activities across Member States. 10% amount to 144 entities across the 27 Member States, or approximately 5 per Member State. For some Member States this would be a low figure, but in average this is probably a slightly high estimate. The resulting estimates should therefore probably be considered prudent in terms of not underestimating the full compliance costs.
155An interview with one of the biggest investment banks indicated that their one-off implementation costs would probably be €0.5 million. On the other hand, the complexity of the business, and hence the number of positions to take into account for calculating the overall exposure, was considered very high compared to most other entities. The cost for less complex entities should on average therefore only be a fraction of that incurred by a large investment bank. It is likely that the difference in the degree of complexity could be said to be a factor of 50 or even 100. However, the cost does probably not decrease proportionally with less complex structures due to minimum costs incurred with system changes. As a result we have found it reasonable to assume that less complex entities will incur a cost of 10%, compared to the most complex companies. Due to the importance of this estimate effort has been made to understand the operational aspect of the requirement.
156The hourly wages are based on standardised ESTAT data (the four-yearly Labour cost survey and the annual updates of labour cost (ALC) statistics), reflecting 2006 figures. They already contain the standard 25% overhead costs, as required by the Standard Cost Model .
157Short selling disclosure thresholds: findings of FSA survey. Communication to CESR, 8/12/2009.
158The hourly wages are based on standardised ESTAT data (the four-yearly Labour cost survey and the annual updates of labour cost (ALC) statistics), reflecting 2006 figures. They already contain the standard 25% overhead costs, as required by the Standard Cost Model.
EN EN