Perverse incentives and financial crisis
People can be relied on to respond rationally to incentives. If perverse incentives are created for them, they will produce bad outcomes. Worse, if a powerful minority can get to create its own incentives, bad outcomes for everyone else will proliferate systematically. As public policy makers consider the mistakes made that led to this financial crisis, they would do well to focus on failures to counter perverse incentives. But there are lessons for investors too. Understanding incentives is crucial to self-protection in a conflicted investment industry.
The banking crisis was an accident waiting to happen. Its root causes, though economic in nature, lie in the activities of the banks. But it certainly was not inevitable. It should have stopped in bank board rooms, well before it got out of hand. What should have stopped it was rational self-interest, as perceived by each of bank directors, their shareholders and regulators.
In this Insight, we explain in laymen’s terms how the way business is organised leads to certain predictable outcomes. Though the focus is on the business of banking and its regulation, the trails we explore echo to a recurring No Monkey Business theme: smart agents will always exploit weak principals. Monkeys do what monkeys do.
Follow the money
We see five trails in the banking crisis that lead back from bad outcomes to flawed incentives.
Since the recession of the early 90s, government policies in the areas of money, credit and employment have all tended to encourage ‘moral hazard’, interfering with the symmetry of punishment and reward that systems depend on if they are to function normally.
Changes in international rules for determining how much capital banks needed led directly to bankers shunting assets and liabilities off their balance sheet, without a care for their risk and return.
Rising volumes of increasingly complex transactions led bankers to rely for simplicity on rating agencies to assess credit quality, even though these agencies were paid by the issuers of the paper they were assessing.
Senior bankers wrote their own asymmetrical pay incentives so that they participated quickly and handsomely if they wrote high business volumes but did not participate in the losses if volumes came at the expense of lower quality and future defaults.
Bank shareholders should have been the ones to hold bank management’s incentives in check but shareholders are today represented by institutional investment managers who have collectively and individually created similar perverse incentives for themselves. Independent advisers are too weak to counter these incentives and have even been tainted by them.
Banking will always be tightly regulated, because banks have the power to create money, as deposits make for loans and loans create more deposits. This makes them an instrument of government (or central bank) monetary policy, not a ‘free market’. So in judging the incentives provided by regulation and supervision of banks, we have to leave behind many free market notions of competition.
Monetary policy in democratic, advanced economies has created a spoilt generation of voters, much as parents now seem to run scared of their children.
The first evidence of this emerged in the USA in the Greenspan-led Federal Reserve Bank as financial market actors started to behave as though any signs of either recession or weak asset prices would be met by monetary stimulus, to prevent a rerun of the recession of the early 90s.
In general, such thinking has been validated by subsequent events. Policy responses to the financial crisis in emerging markets and the collapse of Long Term Capital Management, though aimed at stability, in fact fed instability, in the form of the boom in technology and new media stocks. To ensure the subsequent bust did not feed back to real economic activity (a risk thought to have been increased by 9/11), the response was to hold interest rates at about 1% for several years. After that, not even rapidly accumulating financial imbalances between nation states, increasing leverage in personal and corporate balance sheets and an unprecedented bubble in real house prices were enough to put steel into central bankers’ spines. No wonder that in bank board rooms around the world decisions were made to hold (or trade) as many as possible of the assets being created by unchecked borrowing.
This narrative does not even need the added ingredient that agents behaved as if many banks were too big to be allowed to fail. More likely, this emerged only after the sub-prime lending bubble burst. Prior to that, the thought of failure probably did not even enter their heads.
The policy response of governments to the ensuing crisis has been more of the same. If the underlying financial imbalances are at some point to be corrected it will be in spite of governments, not because of them, and then only thanks to the rational self-interest of households (who are also voters) when guided by fear and mistrust.
Fighting over the rent
In banks, in addition to their role in monetary policy, the normal tension between shareholders and management has always been particularly acute. This is particularly so for the old ‘merchant banks’ and modern ‘investment banks’, where profit generation depends less on running a broadly diversified book of financial assets and liabilities and more on deal making and trading. These activities are less capable of being dehumanised as mere systems and processes.
Doing more of these activities shows up as both higher and more volatile operating profitability. It makes it more important to keep staff costs variable. Bonuses achieve this. But dependency on individuals rather than systems increases the risk that individuals will succeed in extracting too high a rent for their labour relative to the rent on the bank’s fixed and financial capital that they need to use.
Checking this balance between shareholders and managers was never explicitly the role of bank supervisors. However, the earlier form of personal interaction between the Bank of England and senior bankers was possibly more attuned to spotting causes of risky behaviour than is the more data-dependent and process-driven FSA. It is also possible (as the Conservatives believe) that the separation of powers between the Treasury (although this normally does delegate its powers), the Bank of England and the FSA weakened the effectiveness of supervision. We may safely anticipate, therefore, that the regulatory lessons learned will include a return to closer board-level observation and dialogue.
Lord Turner, arriving late as the head of the FSA, has argued in public that regulators placed too much reliance on the ‘efficiency’ of financial markets. Presumably what he had in mind is not just market efficiency in the narrow, technical sense of information flows, and rational responses to information, but also the more widely-understood meaning, that open competitive markets tend to be self-correcting because of inbuilt incentives, in the form of commercial rewards and punishments.
The first interpretation of efficiency serves as an excuse for regulators to assume that regulation of products, and particularly innovative products, is not a critical focus for them. They could not be more wrong. There is a depressingly long history of product failures in both banking and investment markets that should keep regulators constantly focused on the technical integrity of products.
The second interpretation of efficiency refers to the theoretical function of equity owners, in a public stock market: rewarding good and sustainable business models and punishing bad or excessively risky ones. If the regulators relied on this theoretical function working well in the bank sector, they were again deluded.
Where were the shareholders?
The owners of the primary bank capital, equity, are the people with the powers to ensure management incentives are checked and to influence business models and strategy. They used these powers weakly.
This is not a new story, a British one or one peculiar to banks. Shareholders in the US as well as the UK have been mindful of the distortions created by pay incentives but the effect has been to shift more of the variable remuneration onto share incentives. Executive options have exploded as a form of deferred and only partially contingent pay, more so in the US than here. Institutional shareholders were extremely slow to realise that they were being taken for a ride.
The same lack of engagement by shareholders is signalled by the rising share of reported bank profits going to labour, unless they glibly assumed that the rise in operating profits was itself the result of needing less capital (and by implication) less need for return to capital relative to labour. We have yet to see the full cost of their error. Regulators are now bound to increase the capital banks are required to hold. It represents a second wave of dilution of shareholders’ permanent claim on banking profits, following emergency funding by governments.
The fund management agency problem
Shareholders are today predominantly represented by ‘professional’ or institutional money managers. Though their clients may wish they were fiduciaries, like trustees, the fact is they are commercial agents. As such, they respond to their own incentives. Unfortunately these too have become perverse.
Consider the particular effects of the banking practices that, with hindsight, caused the collapse of confidence and liquidity and turned booked profits into restated losses. We can see that stockmarkets had rewarded, with strong absolute and relative share price gains, both exceptionally high volume growth on banks’ balance sheets and capital-efficient ‘wholesaling’ activities that moved ‘structured’ forms of lending off the balance sheet, for an ‘origination’ fee. As you should expect, what was happening within banks was a microcosm of the increasing dependence of GDP growth on leverage, which stockmarkets also failed to see as increasing risk rather than value.
Professional or institutional shareholders, supported by competing research on banks by brokers, should have known better. Maybe they did, but it then required them to balance long-term risk (deteriorating quality means no value is being created) against short-term advantage (rising volumes support rising share prices). As many people have observed in recent years, professional investment managers have become increasingly reluctant to take the longer view or to back discernment of quality and sustainability rather than short-term momentum.
In doing so, however, they are acting rationally as the structure of the competition in professional fund management means that short-term performance is better rewarded than long-term performance. There are two sets of perverse incentives at work here.
As collective forms of retail investment have grown, the nature of firms’ competition for assets has been dominated by relatively recent achieved investment returns. Academic evidence demonstrates that strong short-term relative performance adds more assets and revenues than are subsequently lost when performance reverses. This ‘game’ effect is much more important for business management than the overwhelming evidence of the ubiquity of randomness in performance data. No surprise, then, if firms decide to use the public’s misconception rather than disabuse it.
The second set of perverse incentives is the link between achieved investment returns and the remuneration of the managers supposedly responsible for those returns, either through explicit linkages or through discretionary bonuses or pay awards. Performance awards now mirror the three to five year horizons that matter for asset gathering, which is perfectly rational. But that has nothing to do with skill. It might not even be long enough to capture more than a single decisive asset allocation move, such as a change in market trend.
What about the independent advisers?
Theoretically, the agency problem associated with short-term performance would be corrected by the involvement of smart and independent advisers, through their influence over investors’ asset allocation decisions, manager selection and fund trading.
Not a bit of it. They have made themselves complicit in the fund management agency problem by making their revenue model dependent on product providers, who then use them as distribution agents in the asset gathering game. The way that game works, even random shifts in relative return between competing funds provide scope for churning fund positions, to generate new commissions paid by the investment houses to advisers. It is totally irrational for their customers, except (of course) they have been led to believe it is a worthwhile endeavour.
If these agency incentives in investment markets are to change radically, the ultimate investors, as customers of the industry, need to want to know everything about its dirty little secrets.