Reserve Releases Remind Us that Bad Habits Die Hard in Banking Industry

Originally posted by Yuval Bar-Or on Oct 12, 2012  

It’s earnings season, and this time the focus is on reserve releases at banks. Reserves, as the name implies, are funds set aside to absorb expected losses. Reserve releases occur when banks feel that storm clouds have cleared and that there is no longer a need to maintain such a large buffer of reserves. In “releasing” reserves bank managers are able to redirect freed-up dollars to earnings. According to a Wall Street Journal analysis, “The four biggest U.S. banks have released $18.2 billion of reserves over the past four quarters reported, or 23% of their pretax income.” Among these are Bank of America Corp. (BAC) and Citigroup Inc. (C).

So, are reserve releases a bad thing?

On the face of it, the releasing of reserves (as an economy improves) is a reasonable action. It is typical to observe reserves building up as banks’ expected losses mount, and to see a reversal when the dust settles. But some caution is required.

Finance professionals (institutional investors and analysts) are well aware of the boost to earnings from reserve releases, and they adjust their analyses accordingly. Smaller, less sophisticated investors may not be aware that earnings have been boosted by reserve releases. They may incorrectly interpret the dramatically rising earnings as definitive signs that the economy has turned the corner, signaling clear skies ahead.

In fact, with concerns about Europe and China ongoing, as well as potential conflict in the Middle East, there are still significant clouds on the horizon.

Perhaps most importantly, recent reserve releases are a painful reminder that banks may still not be “getting it” when it comes to risk management best practices. Professional risk managers know that the key to managing risk is to be proactive. Instead of waiting for damage or loss to manifest and then trying to clean up the mess, it is far better to anticipate and take preventive measures in advance. The ongoing reserve releases are a painful reminder that banks took too long to set aside reserves to address significant housing market losses. Instead of being ahead of the curve, banks reacted after the fact. This is especially disturbing considering that reserves are set aside to cover expected losses, i.e., those that bankers are supposed to be able to predict. (In contrast, unexpected losses are addressed through economic capital).

While bankers reactively raised reserves several years ago, they are now proactively (and perhaps prematurely) releasing them.

If the risk management logic is so obviously against this behavior pattern, why do banks persist in exhibiting it?

A simple answer is that bank management teams are following incentives and catering to financial markets’ short term thinking. As a downturn looms, banks are reluctant to increase reserves because such actions lower earnings, potentially depressing stock prices and affecting management bonuses and promotions. When it becomes clear that reserves are urgently needed, the banks must make enormous reserve contributions to catch up. Many of them refer to this as “taking a hit,” and they prefer to take one big hit rather than steady reserve increases because the single big hit is soon forgotten by markets focused on 90-day (quarterly) reporting windows. When markets stabilize, banks waste no time in releasing reserves, seeking to (as soon as possible) improve their earnings picture and benefit from improved bonuses.

No  One Wants to be First

Imagine one bank that decides to do the right thing. As storm clouds gather, the bank increases reserves significantly and proactively. Here’s where this sad story turns truly ironic. Instead of being given credit for responsible, forward-looking risk management, the market concludes that this bank has deeper problems than the rest, otherwise why would it be the only one increasing reserves? And so, the market hammers that bank’s stock price, while the less-responsible competitors are unaffected. Fear of such an outcome, in markets focused on quarterly earnings reports, constrains bank management teams from doing the right thing.

Whenever we see large reserves being released, which is inconsistent with proactive risk management and proper strategic planning, we are reminded that bad habits die hard in the banking industry.

Too Big to Fail: Too Big to Bail

Originally posted by Yuval Bar-Or on May 11, 2009  

Large banks have historically benefited from an implicit guarantee known as too big to fail. The idea is that even if not legally obligated to do so, governments may have to provide assistance to financial institutions whose bankruptcy could lead to a destructive domino effect throughout the entire financial system.

The logic invoked in defense of too big to fail is that it serves at least two important functions: it can calm skittish depositors who would otherwise withdraw their funds from the bank, causing a panic and a run on the bank that would surely force it into insolvency, and it can protect investors who would otherwise likely see the value of their entire investment wiped out by panic.

The problem is that with so many large banks struggling for survival, taxpayers are now facing upward spiraling bailout costs in the hundreds of billions (and possibly trillions) of dollars. Too big to fail has become too big to bail.


This blog entry first appeared on

The Law of One Number

Originally posted by Yuval Bar-Or on May 02, 2009  

Humans seem to have an ingrained need to simplify, to shorten, and to reduce. In economics we’ve embraced the law of one price, assets with similar return and risk characteristics must have the same price; in religion we’ve largely embraced monotheism, the law of one god; and in relationships many societies have adopted the law of one spouse.

These choices appear to have served us reasonably well over the centuries. Unfortunately, other simplifications have been our undoing. Specifically, our financial risk management efforts have taken oversimplification to an extreme, with an insistence that risk be summarized in one number. A bank CEO may be told that her firm’s portfolio of loans has an unexpected loss of X; the CEO of a brokerage firm may be told that his firm’s portfolio has a Value at Risk of Y; the arranger of an asset backed security may be told that the correlation among underlying assets is Z; an asset manager may summarize the standalone risk for a security by the standard deviation of its historical returns. Non-financial institutions are not exempt: a manufacturing firm’s CEO may be told that the Economic Value Added of a global division is Q.

Each of these examples represents a gross oversimplification – much in the same ways that a sports team or league insists on voting for one MVP, and the Oscars anoint one Best Actress. No single player can achieve greatness without teammates’ active contributions, no precise and objective way exists to differentiate between two or more brilliant acting performances. Similarly, in the risk management field, no single measure of financial risk tells the entire story. We know that a number of systematic (market wide) and idiosyncratic (investment specific) variables contribute in non-linear fashion to the risk of financial investments and strategic business decisions. We know that the outcomes will be manifested from an underlying distribution of events, and we know that no single number contains the rich complexity of all that information. Simply put, we know that oversimplification means the loss of critical information – yet somehow we continue to insist on summarizing a whole swath of critical knowledge into a single, convenient, yet dangerously misleading figure.

Surely we’ve learned the hard way that in some areas, simple is not good enough.

Should Rescued Banks be Paying Bonuses with Taxpayer Money?

originally posted by Yuval Bar-Or on Jan 09, 2009  

As a dismal 2008 ended, investment and commercial banks announced their bonus payment policies. Despite having collectively presided over a disastrous period in American (and global) economics, many institutions paid bonuses. In fairness, a number of high profile executives elected to forego bonuses entirely, and average bonuses paid were significantly lower than in prior years. Nevertheless, the perception that private companies used taxpayer money to pay bonuses to irresponsible, inept, and in some cases, corrupt players infuriated many. The consistent response to taxpayer and politician outrage was that many of the institutions had to pay bonuses in order to secure the loyalty of their employees, thereby ensuring continuity in their services.

This is an unsatisfactory argument for several reasons:

  1. Wall Street has always claimed it is about merit and performance. No one can deny that performance has been horrible in 2008. (And it now seems that the positive performances logged in prior years were also hollow). Paying a bonus, however small, creates an immense moral hazard: “you still get paid even if you do poorly (devastating the world along the way).”
  2. The idea that employees may leave is absurd. Where exactly are they going to go? Are there so many other healthy financial institutions out there who can afford to add everyone else’s employees to their payrolls?
  3. If there are institutions who’ve done well enough during this disaster to attract talent from underperforming firms, they’ve won the right to do so

The whole point in a capitalist system is that those who prove their commercial superiority should attract the best resources, including both financial and human capital.

It’s quite telling that these bastions of capitalism and cheerleaders of meritocracy have been so quick to embrace “socialist” handouts funded by taxpayer money.

Considering the destruction caused by Wall Street excesses in recent years, it’s impossible to conceive of an overwhelming argument in favor of any bonuses. Those firms whose leadership failed to make good decisions deserve to be subjected to Darwinian justice, and employees who knew the risks should be required to face the consequences of their risk-taking. This is one of those cases where survival of the fittest is consistent with social values.