Rethinking Risk Management Again
J. Rizzi
August, 2013
Draft #3 8-23-13
I. Introduction
The 2008/2009 financial crisis represents a teachable moment for risk management. In fact, it represents the second teachable moment. The first being Long Term Capital Management, which occurred a decade earlier, and served as an unnoticed dress rehearsal to the 21st century crisis. Large and small banks suffered severe damage despite substantial pre-crisis risk management investments. The 2011 MF Global Failure and the 2012 JPMorgan Chase London Whale $6B loss reminds us that something is still wrong with risk management. Furthermore, regulatory changes like Dodd Frank and BIS III are unlikely to improve matters. In fact, they may be counterproductive as they change risk management into a compliance function.
Risk management’s prediction focus is largely based on recent history. It developed in the 1990s based on an actuarial statistical approach to estimate future probabilities based on past events. It involves looking backwards to see into the future. Tranquil periods last long enough to seem to be the natural state. Crises seem sharp enough to be seen as aberrations instead of normal accidents. An unfamiliar crisis is seen as improbable, and not taken seriously. Substituting a probability distribution for uncertainty does not solve the problem. Out of a sample of events; Peso Risk, caused by unexpected regime changes, are difficult for individuals, regulators and organizations to understand due to behavioral biases. These errors underlie the failure of risk management and are reinforced by behavioral biases, such as overconfidence.
The biases are magnified in financial institutions by the “Killer B’s” of budgets and bonuses given that the budget bonus period is shorter than the risk horizon. The combination of these biases produced an unwarranted belief that risk could be controlled. This leads to the acceptance of what would otherwise be viewed as unacceptable risk. Institutions mistakenly assumed that risk created return. Therefore, risk appetite increases can create higher returns, while risk management can handle the increased risk. Unfortunately, risk is not static. It evolves in ways not fully understood.
Risk, the exposure to the consequences of uncertain events, is managed by people and not by models. We must accept randomness by emphasizing discipline and judgment over prediction. Markets are more complicated than many over-confident risk managers believe.
This article reexamines risk management primarily from a bank viewpoint. It identifies the errors in effective risk management that lead to massive losses at financial institutions during the financial crisis. Also, it helps place risk management in its proper context within the organization versus being viewed as a detached specialist function. The key is making risk count instead of just counting risk. Hopefully, it illustrates how modern risk management techniques can create unintended consequences in the management of financial institutions; how government policy and corporate governance can influence the management of risk in financial institutions; and how discipline and judgment are important in the management of risk.
II. Delusion About Risk
Many bankers reach for yield in low return environments. Unfortunately, it has potentially ruinous long-term side effects. The strategy is based on the mistaken belief that risk creates return. Thus, if you cannot get desired returns from safe investments, then increase your risk appetite and take more risk to increase your return.
Risk may be correlated with return. Risk is not, however, the raw material generating return. If risk caused return, then it would not be risky, and Jimmy Cayne and Dick Fuld, among others, would still be on Wall Street. The mistaken belief that risk creates return confuses correlation with causation. Investors take risk and receive a return. This does not mean they received a return because of risk. Rather, return is based on skill in creating value and managing risk. Taking risk to generate nominal return, however, is easier than creating a value added service that satisfies a client need. Thus, absent strong board of directors’ oversight, managers will focus on nominal returns as a key performance indicator instead of risk-adjusted return value drivers. This encourages increased risk taking even though the institution is undercompensated for the incremental risk.
Risk is a cost of return and not an opportunity. Seen in this light, it is something to be reduced – not increased. Additionally, it has capital implications. Higher risk without higher capital is like building in a flood plain without flood insurance. Capital is needed to absorb the volatility. The increased capital reduces return on equity, which illustrates risk alone does not increase value. In fact, many banks confuse investors with high returns based on risk not skill. Ultimately, the reality becomes clear resulting in “surprise” losses, shareholder value destruction and management changes.
We see only losses after risk is realized. Risk is usually and mistakenly measured by extrapolating historical loss data to calculate a capital change known as economic capital. The charge is deducted from an investment’s return to approximate a risk adjusted return. The nonstationary business cycle related macro component of risk is typically not reflected in the recent historical data and consequently is frequently ignored. Consequently, economic capital is understated as many banks discovered during the crisis. Losses tend to occur infrequently, but their effect is large. Since there are usually more good years than bad, high-risk strategies can appear successful for long periods of time. During a bull market, those who take on more risk appear to outperform their more conservative peers.
Reliance on a trailing weighted average of returns misses periodic shocks which can swamp shorter term averages. Not only does risk evolve over time, but so does return. As popular asset classes, like commercial real estate pre-crisis and commercial and industrial loans post crisis become crowded, returns fall requiring even higher risk to maintain return levels.
Avoiding permanent capital loss through skillful risk reduction represents a competitive advantage. Nonetheless, this fact received little attention pre-crisis during which time many banks were seduced into thinking they could not lose. Also, higher short-term nominal returns are more exciting than lower long-term risk adjusted returns, especially for bonus purposes.
There is nothing wrong with high risk-high return strategies provided the volatility implications are appreciated. The first cut is whether the institution has the skills needed to manage risk to an acceptable scenario based stress level. Next, sufficient capital and liquidity are required to withstand shocks.
The mechanical risk-return relationship inherent in the risk thermostat view of setting risk appetite and business strategy is wrong. Sustainable long-term returns are dependent on skill in managing risk, and not just in taking more risk. This is a subtle, but critical distinction.
III. Behavioral Finance Framework
Behavioral finance examines how managers gather, interpret, and process information. It recognizes that models can influence behavior and shape decisions. This influence can corrupt the decision process leading to suboptimal results.
Risk can be classified along two dimensions. The first concerns high-frequency events with relatively clear cause-effect relationships. Other risks occur infrequently. Consequently, the cause-effect relationship is unclear. The second dimension is impact severity. No matter how remote, high-impact events cannot be ignored because they can threaten an institution’s existence as was demonstrated in the financial crisis. The dimensions are reflected in the risk map in Figure 1.
Frequency
A B
C D
Impact
A: High frequency/low impact events: reflected in risk pricing.
B: Low frequency/low impact events: treated as a cost of business.
C: High frequency/high impact events: managed through control.
D: Low frequency/high impact events; frequently ignored.
Figure 1 Risk Map
Quadrant A events include retail credit products including credit cards. Many small defaults are expected. Screening helps identify groups with higher default probabilities. These groups are charged higher rates to offset the risk. Quadrant B represents many internal operational risks such as check processing errors. The costs are absorbed and the focus is on mitigation and prevention through improved processing and training.
Type C events include concentrated exposures to high risk borrowers. These well known risks are managed by constant management monitoring and control. Type D events are frequently ignored due to a low frequency. Examples include many of the structured finance products which represented short positions in an option. They offered a long period of steady income punctuated with occasional large losses.
Cyclical risks are low-frequency-high-impact events characterized by their negative skew and “fat-tailed” loss distributions. Investors incurring such risk can expect mainly small positive events but are subject to a few cases of extreme loss. These risks are difficult to understand. The difficulty stems from two factors. First, there is insufficient data to determine meaningful probability distributions. In this case, the statistics are descriptive, not predictive. Consequently, no amount of mathematics can tease out certainty from uncertainty.1 Second, and perhaps more important, infrequency clouds hazard perception. Risk estimates become anchored on recent events. Overemphasis on recent events produces disaster myopia during a bull market, as instruments are priced without regard to the possibility of a crash. These facts lead to risk mispricing and the procyclical nature of risk appetite.
Quantitative risk-management models are based on portfolio and option pricing theory and provide a framework on how risk managers should act. These models build on expected utility theory (EUT), which views individuals as expected utility maximizers.2 Empirical support of EUT is mixed with numerous reported anomalies.3 Examples of anomalies include holding losers, selling winners, excess trading, and herding.
An alternative, prospect theory,4 can explain these facts. Instead of being expected utility (E(U)) maximizers, investors are viewed as expected regret (E(r)) minimizers focusing more on loses than gains. This is reflected in Figure 2.
EUT focuses on wealth changes. The value function in prospect theory is based on gains or losses relative to a reference point, usually par or the original purchase price.
Value function value
+ Utility
. Convex slope indicates pain of
loss (regret exceeds value of gain)
. The conflict between E(u) maximizing
and E(r) minimizing underlies many
anomalies Losses Gain
. Investment decisions involve 3 Rs:
return, risk and regret
Reference point
-
Figure 2 Investors Minimize Expected Regret
Market signals are complex. They include both information and noise. Information concerns facts affecting fundamental values. Noise is a random blip erroneously interpreted as a signal.5 Risk managers have developed shortcuts, rules of thumb, or heuristics to process market signals. These belief-based heuristics incorporate biases or cognitive constraints. These biases are the hidden risk in risk management, and will now be investigated.
A. Regret
Risk is forward looking. Regret, however is backward looking. It focuses on responsibility for what we could have done but did not do. Regret underlies several biases. We try to minimize regret by seeking confirming data, suppressing disconfirming information, and taking comfort that others made the same decision. Consequently, regret can inhibit learning from past experiences.
Sunk costs are the first regret bias considered. Sunk-cost bias involves avoiding recognizing a loss despite evidence the loss has already occurred and a further loss is likely. Examples include the reluctance to sell impaired assets at reduced prices. Usually this is defended as the market prices being too low. Most institutions, however, reject the logical alternative of acquiring additional exposure at the market price to exploit the alleged under pricing; thus, illustrating in this instance, price is of secondary importance relative to regret.
Panic conditions are also based on a combination of regret and herding. In a crisis, the reference is pessimism, and we actively seek bad news to confirm our belief. Thus to minimize regret, we follow the herd not to be left behind and engage in panic selling. This further depresses prices leading to continued forced selling and the creation of a negative feedback loop as occurred in the fourth quarter of 2008.
Another regret-related bias is the house money effect. Risk managers will assume greater risks when they are up in a bull market and lower risk in a bear market. Regret is perceived to be less when risk of winnings is involved, than risk of initial capital. This procyclical phenomenon leads to “buy high and sell low” behavior.
It illustrates the George Soros reflexivity or feedback principle, whereby markets affect psychology and psychology affects markets. Positive feedback is self amplifying, while negative feedback is self corrective. For example, collateral values rise during a bull market. This increases their access to lower priced funding and liquidity, which fuels further gains.
Finally, regret leads to confusing risk with wealth. Larger, better-capitalized financial institutions can absorb more risk than smaller institutions. Their greater risk tolerance lessens their downside sensitivity, especially during a bull market when income levels are high. Thus, risk appetite increases with wealth. Risk and return are, however, scale invariant. Larger institutions confuse the ability to absorb risk provided by capital with the desirability of the risk position. Therefore, they acquire underpriced, higher-yielding, higher-risk assets in bull markets.6 The JPM Chase London Whale situation examined later is a recent example of this fact.
B. Overconfidence
Overconfidence occurs when we exaggerate our predictive skills and ignore the impact of chance or outside circumstances. It results in an underestimation of outcome variability.7 Overconfidence is reinforced by self-attribution and hindsight. Self-attribution involves internalizing success while externalizing failure. Structured finance bankers and quantitative risk managers took credit for results during the boom, failing to consider the impact of randomness and mean reversion creating an illusion of control.8 Hindsight involves selective recall of confirming information to overestimate their ability to predict the correct outcome, which inhibits learning. Disappointment and surprise are characteristics of processes subject to overconfidence.
Industry and product experts are especially prone to overconfidence based on knowledge and control illusions. Knowledge is frequently confused with familiarity. This is reflected in the number of industry experts including most famously the former Federal Chairman who missed the collapse of the housing and structured credit bottom.9 This is due, in part, to misguided overreliance on quantitative credit scoring models without understanding their limitations. Key model limitations include the following: