.jpg)
Ladies and Gentlemen, I am delighted to have been invited to speak at the Annual Conference of the Chartered Financial Analyst Institute on the important and at times controversial relationship between efficient markets and market regulation.
We are going through testing times, having over the last year and a half witnessed the biggest turmoil in financial markets in almost a century. These testing times invite us to reflect on and question some fundamental theoretical issues and intellectual assumptions which we had until now taken as conventional wisdom.
The theory of efficient markets is one of these assumptions and I will aim here today to set out some of my personal reflections on the Efficient Markets Hypothesis from a regulator’s perspective. This issue has been very well set out by Lord Turner in the Turner Review. As you may recall, Lord Turner distinguished five propositions in the Efficient Market Hypothesis which had implications for the regulatory approach followed by regulators. Summarising these in turn, they were that:
- Market prices are good indicators of rationally evaluated economic value
- Securitised credit has improved allocative efficiency and financial stability
- Mathematical analysis can deliver robust quantitative measures of trading risk
- Market discipline can be used as an effective tool in constraining harmful risk taking
- Financial innovation can be assumed to be beneficial
Recent events have shown that these assumptions could be challenged and that such challenges could have major implications for the future design of financial regulation. The predominant assumption had been that financial markets are capable of being both efficient and rational, with independently acting market participants making rational assessments so that the overall level of prices would have a strong tendency towards a rational equilibrium. In such a market, regulation would therefore principally limit itself to removing the impediments and barriers which might exist and which might lead to inefficient and illiquid markets. This would justify a strong presumption in favour of market deregulation.
But is this still the world we live in? Does market efficiency necessarily imply market rationality? Does individual rationality ensure collective rationality? Or indeed is individual behaviour always entirely rational as we have seen with large scale herd effects? What are the implications for regulators? These are questions which will occupy most of the people in this room, academics and regulators, for the next decade and it would be presumptuous of me to give a conclusive answer to them in the thirty or so minutes of this speech.
What I would like to do however is to first reflect on the particular issue and implications of one of the regulator’s main goals which is improving market transparency and touch upon behavioural critiques of the Efficient Markets Hypothesis with a focus on individual investors. Before the markets crisis, financial regulators had traditionally been interested in these two issues.
However, since the markets crisis the macro-perspective of efficient markets has come very much into the spotlight. I will therefore also discuss this issue before turning to some ideas on how regulators can deal with market inefficiencies in light of the current crisis.
Transparency and market efficiency
There are many indications that market efficiency decreased during the crisis. In many markets we have observed lower trading volumes, higher bid-ask spreads, higher volatility, etc. Let me illustrate this with two examples:
- Volatility in the stock markets increased sharply during the crisis; during the autumn 2008 we observed volatility levels unseen in the last decade;
- We have also seen that in many markets, trading activity decreased substantially. To take just one example, volumes traded in corporate bond markets decreased substantially.
What does that mean for the Efficient Markets Hypothesis and regulatory activity?
The Efficient Markets Hypothesis implies that asset prices incorporate all available information. From this perspective, market inefficiencies redistribute wealth between well informed investors and less well informed investors. A financial regulator will have an interest in imposing transparency rules on markets to ensure fairness and because of a potential lack of incentives for market participants to provide all the relevant information to the market. The most obvious example of fostering market transparency is our listing regime, where we are ensuring the provision of adequate information to investors, e.g. on major shareholders in listed companies. This is clearly intended to ensure that all investors have a comparable level of information and markets are fair.
Why is fairness important? If some market participants are consistently better informed about market developments than less informed investors, less informed investors may face a systematic disadvantage when trading. This can undermine confidence in the fairness of the market and reduce the willingness of the less-informed to participate. This in turn would lead to an increase in the overall cost of credit intermediation and thus reduce the investment opportunities that could provide a positive return, ultimately reducing growth in the economy.
On the other hand, increasing transparency comes at a cost. If a market participant has to disclose trading activity to the market, he might lose profitable investment opportunities. There would be no return to costly research activity or the development of an innovative trading strategy. Institutional market participants may therefore provide less transparency than would be ideal from society’s perspective. This is because they do not take into account the benefits additional transparency would create for other market participants.
These two perspectives, in a nutshell, describe the dilemma facing the regulator when thinking about transparency requirements. Regulators, through transparency regulation, will aim to increase efficiency in the market and achieve fairness in the market.
This explains the open academic and regulatory debate as to the degree to which some transparency in markets is actually beneficial. And many will question whether transparency regulation can help to significantly improve market outcomes or whether a certain degree of inefficiency is a fact of life.
A related question is whether we can actually rely on the valuation of securities? There is a push for greater transparency in many of the securities markets- rightly so. But, the crisis has shown that there is not only an issue with transparency as such, but also that a lack of confidence in the valuation of securities has been one of the main problems. Pre crisis the conventional response would have been to increase transparency. However, the crisis has shown that there is one precondition for increased transparency to deliver benefits: increased transparency needs to have a positive impact on the confidence of market participants. In order to have this positive impact on confidence, investors have to be able to absorb and trust the additional information they get. Just providing more information is not going to work if people don’t believe in the value of that additional information.
This is why the issue of whether more transparency is needed is likely to vary significantly across different securities markets, especially in times of severe market stress, and will depend on the characteristics of that market and how much of the information investors can absorb, understand, trust and use.
Irrationality, the Efficient Markets Hypothesis and the regulator
Let me now turn to irrationality and provide a regulator’s view. There is a long list of deviations from rational behaviour which has been used to question the Efficient Markets Hypothesis, such as exuberance and herding behaviours.
Looking first at irrationality from an individual retail investor’s perspective – something we have interest in given consumer protection is one of the 4 statutory objectives of the FSA. Irrational behaviour as well as a lack of financial knowledge can lead to consumers taking decisions which, in contradiction to the Efficient Markets Hypothesis, can lead to consumer detriment. For the regulator this raises the issue of whether a given product is suitable for investors. Many disclosure and suitability regulations in retail investment markets stem from this rationale. With respect to wholesale market customers, most market and bank regulators have tended to assume that wholesale market customers are by definition sophisticated and do not need protection.
However, recent experience – especially in the markets for securitised and structured products - has led us to review this assumption. Not long ago Andrew Haldane from the Bank of England provided a very vivid example of the complexity of structured products. An investor in an ABS CDO needs to read about 30,000 pages to completely understand what he is actually investing in. For an investor in a CDO of ABS CDO (a so-called CDO squared) the number of pages increases to approximately one billion pages.1
Perhaps one can then understand the wisdom of the FSA recently emphasising again the importance for all investors – including wholesale investors – only to buy products which they understand. It seems that the popular wisdom of caveat emptor still offers the best protection whether you are buying a CDO squared or a new i-phone.
In conclusion, it seems that many structured products are too complex even for sophisticated investors to understand. Further, even sophisticated investors were on an individual level overconfident about the risks involved in these structured products. They have also relied too much on credit rating agencies instead of themselves making judgments based on understanding the products they invest in.
The Efficient Markets Hypothesis - A market and macro- economic perspective
Up to now I have discussed the limits of transparency and the issue of individual irrational behaviour of retail and wholesale investors. Acknowledging these problems, we have to ask ourselves whether the market, despite its many, obvious and severe failings on an individual level, is nevertheless "efficient". In other words, is the market able to ‘iron out’ these failings on an aggregate basis?2
Opinions on this are divided. Interestingly, one of the founders of the Efficient Markets Hypothesis, Paul Samuelsson, has questioned whether the Efficient Markets Hypothesis still applies at a macro-level. He now questions whether the Efficient Markets Hypothesis still holds because we observe that prices for securities can be above or below their fundamental levels for a long time, i.e. we observe persistent overshooting or undershooting of market prices.3
On a market or macro perspective the Efficient Markets Hypothesis implies that the building up of asset price bubbles and the following bust is not possible. From this perspective the observations of the last decade – the technology bubble at the turn of the millennium or the housing market bubble – are a puzzle. The Efficient Markets Hypothesis suggests that it is always optimal for investors to speculate against a bubble. Deviations from a ‘rational’ price are an opportunity for profitable trades which should be exploited. In other words, market discipline should help to avoid persistent over- or undershooting of asset prices.
However, there is evidence (even before this market crisis) from the technology bubble that this is not always the case. Take hedge funds as an example: hedge funds should in principle be the first to trade and profit from overshooting prices. However, hedge funds rode the technology bubble and heavily invested in technology stocks. There is evidence that they did not help to correct stock prices during the bubble. US Hedge Funds increased their investment in NASDAQ technology stocks from slightly more than 10% of their assets in March 1998 to 31% at the peak of NASDAQ valuations in March 2000.4 This may be because they were – rightly - predicting the irrational behaviour of other market participants or because hedge funds were acting irrationally themselves.
As another example, we have seen a huge increase in derivative products like Credit Default Swaps (CDS), Collateralised Debt Obligations (CDOs), etc. in the last couple of years.
There is a big question mark whether the pricing of these products has been efficient in the sense of the Efficient Markets Hypothesis. I think it has now been widely accepted that the inherent risks of these products have not been correctly priced by the market and therefore these markets were actually not efficient.
Ultimately we have observed have that the complexity of these products has led to uncertainty about their quality and therefore uncertainty whether their valuation was accurate. This is why we have seen that some derivatives markets basically stopped working as the crisis started. With hindsight it is clear that pre-crisis the price for the risks in these products was too low in view of their complexity and underlying riskiness.
Why is this wrong pricing of risks a problem?
The issue is that it leads to macro-inefficiency of markets which can have serious drawbacks for the real economy and can also lead to consumer detriment. If markets work well, they provide the right goods for the right people at the right price. This means that well working markets allocate resources in an efficient manner and this is the reason why well working markets are beneficial for society. This is clearly why regulators want to make markets work as efficiently as possible.
How is the Efficient Markets Hypothesis linked to allocative efficiency of markets?
Evidence suggests that there is a feedback loop between security prices on secondary markets and real investment.5 Security prices convey useful information about investment opportunities. In this sense informational efficiency of markets is linked to allocative efficiency of markets. In case the Efficient Markets Hypothesis holds, the information in the market price of a security can improve investment choices. In case the Efficient Markets hypothesis does not hold, prices are sending the wrong signals. Wrong signals due to informationally inefficient markets are a problem, because the market might then provide the wrong goods for the wrong people at a wrong price.
As an example of how prices on secondary markets can have real effects on investment decisions, let me briefly talk about securitised products. How have securitised products contributed to the housing price bubble?
Securitised and structured products were one of the key contributors for the easy availability of credit in the years before 2007. With hindsight there is also little doubt that the risks in these products have been consistently underpriced. It is therefore clear that inefficiencies in the market for securitised products have contributed to the build-up of the housing price bubble, the subsequent markets crisis and the current recession.
In other words informationally inefficient markets contribute to the building up of asset price bubbles and the following bust. This is not only problematic because it leads to consumer detriment and effects on the real economy, but also because there is evidence that large swings in asset prices are followed by strains in the financial sector, such as the crisis we have witnessed.6 After the housing price bubble collapsed, a number of financial institutions had to be rescued by the taxpayer at a huge cost to society. Bank lending decreased significantly both to firms and consumers, hence contributing to the recession we are currently in.
Regulators and inefficient markets
So far I have discussed a number of issues related to the Efficient Markets Hypothesis from a regulator’s perspective. I have discussed the limits of transparency regulation and the view of a regulator on irrational behaviour of retail and wholesale investors. I have also discussed the macroeconomic impacts of market inefficiencies.
Where do these issues leave regulators? Can they successfully tackle these issues? Can they also help to prevent or at least mitigate macroeconomic impacts of market inefficiencies, i.e. can they prevent or mitigate the build-up of asset price bubbles?
In analysing these issues, regulators should look at a number of criteria:
- Is there a risk to the objectives of the regulator?
- Is there a case of market failure?
- Can regulatory intervention actually improve market outcomes?
The answer to the first question is yes, there clearly have been risks to the FSA objective of market confidence and consumer protection.
The answer to the second question is also yes. As we have seen in the last year, asset price booms and the following busts come at a huge cost for taxpayer, as governments had to bail out systemically important banks. These costs are not incorporated in market prices. This is a classic case of what an economist calls “an externality”. We have also seen that a lack of understanding of complex financial products even for wholesale market participants has been an issue.
The third question obviously is more difficult. This is one of the reasons why Lord Turner in the Turner Review set out as an open question whether financial regulators should actually play a rule in product regulation in both wholesale and retail markets. For wholesale markets he asked whether there should be product regulation for structured products and credit derivatives markets, i.e. whether the use of these products should be restricted. There are clearly arguments for restricting the use of CDS contracts as well as against. On the one hand, the scale and complexity of these markets as well as issues of pricing of the securities have created financial stability risks. On the other hand, they have a legitimate purpose in risk management for firms. So, as I set out in the first part of this speech, the classic pre-crisis response would have been to increase transparency in these markets. However, the crisis has shown that transparency cannot by itself be the whole answer in achieving the right regulatory regime for such complex, largely wholesale markets.
For retail markets Lord Turner asked the question whether mortgage products should be regulated by introducing limits for loan-to-income ratios or loan-to-value ratios. This could help both to avoid consumer detriment and to limit housing price bubbles. Over the last 40 years, UK house buyers have taken out significantly higher mortgages to finance the purchase of a home. Loan-to-income ratios have increased from under two times income at the end of the 1960 to well above three times income nowadays.
Could limiting mortgage borrowing help to avoid consumer detriment and help to prevent housing price bubbles? Again, there are arguments for and against limiting borrowing for mortgages.
The introduction of borrowing limits could protect customers against the consequences of imprudent borrowing; protect bank solvency against the consequences of imprudent lending; and constrain over rapid credit growth and excessive property price increases. However, it is also important to note important arguments against such restrictions: restrictions on borrowing for mortgages will tend to disadvantage new entrants to the housing market. Further restrictions on the amount of mortgage borrowing might also not be effective, as consumers can build up debt through other means like credit card debt or unsecured loans.
The question of whether and how regulators should intervene and
try to prevent the build-up of bubbles is therefore extremely difficult to
answer against the backdrop of the biggest market turmoil we have seen in
decades. The answer depends on the ability of regulators to judge the future
better or worse than markets do. In order to try to avoid the build-up of asset
price bubbles, regulators also need to take the punch-bowl away exactly at a
time when markets are apparently doing well. Therefore, interventions by
regulators involve not only difficult judgments (something we are used to), but
also significant reputational risks. But as the crisis has shown, for us
regulators to sit back and rely purely on the market to avoid asset price
bubbles does not work either.
Conclusion
To conclude I will revisit my introductory remarks. We are in testing times which confront all market participants with a high degree of uncertainty. This is also true for financial regulators. Past experience in regulation is no longer enough when deciding on appropriate policy measures.
In the last decades, we have taken the dictum of efficient markets for granted. However, in these testing times we must reflect on and question these assumptions. This is the duty of all players in the financial markets: practitioners, academia and financial regulators and we look forward to engaging in the debate.
1 See Haldane, A.; Rethinking the Financial Network; Speech delivered at the Financial Student Association Amsterdam; April 2009, pp. 17 and 37; http://www.bankofengland.co.uk/publications/speeches/2009/speech386.pdf
2 See e.g. Malkiel, B.; The Efficient Markets Hypothesis and its Critiques; The Journal of Economic Perspectives; Vol 17; No. 1; 2003; 59-82.
3 See Samuelsson, P. ; Summing up on business cycles: Opening Address; Speech at the Federal Reserve Bank of Boston; 1998; http://www.bos.frb.org/economic/conf/conf42/con42_02.pdf and Jung, J. and Shiller R., Samuelsson’s dictum and the stock market, Cowley Foundation Paper No. 1183, 2006, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=906229.
4 See Brunnermeier, M. and Nagel, S.; Hedge Funds and the Technology Bubble; The Journal of Finance; Vol. 59; No.5; 2004; 2013-2040.
5 See e.g. Dow, J. and Gorton, G.; Stock Market Efficiency and Economic Efficiency: Is There a Connection?; The Journal of Finance; Vol. 52; No. 3; 1997; 1087-1129 or Polk, C. and Sapienza, P.; The Stock Market and Corporate Investment: A Test of Catering Theory; The Review of Financial Studies; Vol. 22; No. 1; 2009; 187-217.
6 See e.g. Borio, C. and Lowe, P.; Asset Prices, Financial and Monetary Stability: Exploring the Nexus; BIS Working Papers; No. 114; 2002 or M. Higgins and C. Osler; Asset Market Hangovers and Economic Growth; Federal Reserve Bank of New York Research Paper No. 9704; 1997.
