Capital: a very strange way to assess bank safety

Anyway your look at it, bank capital levels or ratios are a very strange way to assess bank safety.

Regular readers know that bank capital, in particular shareholder equity, is an accounting construct with two main components:  capital paid in to buy stock and retained earnings.

Capital paid in cannot be manipulated.  It is simply a reflection of the number of shares sold and the price at which they were sold.

Unlike capital paid in which is based on historic facts, retained earnings are a derived number based on a series of assumptions. Assumptions that can be easily manipulated.

Earnings can be manipulated by the banks.

For example, earnings can be manipulated by simply moving an investment (think opaque, toxic structured finance securities) from the trading account where it is subject to mark to market accounting to the bank's investment portfolio where it is held at acquisition cost to avoid taking a loss.  The result of this manipulation is to a) overstate earnings, b) overstate book capital and c) overstate the value of the assets.

Earnings can be manipulated by the regulators.

For example, by practicing regulatory forbearance, regulators allow the banks to engage in 'extend and pretend' and turn their bad loans into 'zombie' loans rather than recognize the losses on the loans.  The result of this manipulation is to a) overstate earnings, b) overstate book capital and c) overstate the value of the assets.

Because of the ease with which regulators can manipulate capital, the regulators use capital, both the balance sheet level and the ratio, to mislead the public and the financial markets.

For example, Tim Geithner said that the stress tests were rigged in an effort to restore market confidence.  What the stress tests were suppose to show was that the banks had sufficient capital.

Because of the ease with which earnings can be manipulated, even a simple ratio like shareholder equity to assets is meaningless as a measure of the bank's risk.  Piling on complexity by risk-adjusting the assets simply gives banks and regulators more ways to game the capital ratio and render it even more meaningless as a measure of the bank's risk.

Based on these facts, the OECD concluded that bank book capital and capital ratios are meaningless.

Regular readers know that your humble blogger thinks that the only way to assess the risk of a bank is by looking at its current global asset, liability and off-balance sheet exposure details.

I am not alone in this.

For example, Goldman's Lloyd Blankfein said that the only way Goldman knows to manage its own risk is to monitor every position every day.

For example, banks with deposits to lend stopped lending in 2008 when they realized they could not assess the risk of the banks looking to borrow.  The interbank lending market remains frozen.

Despite being meaningless and useless as a measure of risk, bank capital does have its supporters.

In his Bloomberg column, Professor Simon Johnson tries to save bank capital as a measure of bank risk.

Global regulators have a peculiar way of assessing the soundness of big banks: Ask bankers how risky their investments are, then figure out if they have enough capital to absorb the potential losses....

The strategy of trying to make sure that expected losses are less than a bank's book capital level is not unreasonable given that the goal of the financial regulators is to protect the deposit insurance funds.

This method ... has failed repeatedly -- most spectacularly during the 2008 financial crisis....

Is the problem with this method the question of do banks have enough capital to absorb their potential losses or the regulators' ability to answer the question?

The Bank of England's Andrew Haldane observed that the problem lays with the regulators' ability to answer the question.

I don't think that the stability of the financial system should depend on whether the regulators can or cannot answer the question.  By requiring banks to provide ultra transparency and disclose on an ongoing basis their current global asset, liability and off-balance sheet exposure details, we can use all the market participants to assess the soundness of the banks.

Like most firms, banks finance themselves with a combination of debt, which they get by taking deposits and issuing bonds, and equity, which they get from their shareholders. The latter, also known as capital, is crucial to the bank’s survival. 

If bad investments cause the value of a bank’s assets to fall, its equity decreases by an equivalent amount.

Not necessarily as shown by the examples above.

If equity is depleted, the bank is insolvent.

By definition a bank is insolvent if the market value of its assets is less than the book value of its liabilities.  Capital does not directly factor into this definition.

What is critically important to note is that a bank can be insolvent and still continue operating.

There will be either bankruptcy or some form of government bailout.

Actually, there is a frequent third choice made by regulators.

A modern banking system is designed so that banks can operate with low or even negative book capital levels.  The reason banks can do so is the combination of deposit guarantees and access to central bank funding.

With deposit guarantees, taxpayers effectively become the bank's silent equity partner when the bank has low or negative book capital levels.

While the taxpayer is the silent equity partner, if a bank can generate earnings, it can rebuild its book capital levels.

This choice has been made repeatedly.  Sometimes successfully.  Sometimes not.

An example of when the choice was unsuccessful was the US Savings and Loans.  To rebuild their book capital levels, they gambled on redemption.

One of the reasons for requiring the banks to provide ultra transparency is that it allows market discipline to restrain bank risk taking and prevents gambling on redemption.

Hence the need for capital requirements....

The facts simply do not support this conclusion.

One big problem is incentives. Bankers like to use as much borrowed money as possible, relative to their equity. This boosts the return to shareholders in good times, but also presents a threat to the financial system and the broader economy. 

If bankers are in charge of calculating their own risk weights, they will try to understate the risks. This happened with mortgage-backed securities and associated derivatives in the U.S., with real-estate-related loans and assets in countries such as Ireland and Spain, and with sovereign debt in much of the euro area.

This problem goes away if banks are required to provide ultra transparency.  Market participants can independently assess the risk of each bank and adjust both the amount and price of their exposure to reflect this risk.

As a result, simply increasing leverage or risk will result in a higher cost of funds and not necessarily a better ROE.

Second, regulators are no better than bankers at figuring out the right risk weights. For one, they are often heavily influenced by the bankers -- a reality I have experienced personally in my conversations with central bankers and other officials, who meet with banking staff continually and even now are convinced that top executives really know how to measure and handle risk. Beyond that, they are out of their depth. The system of risk weights has become too complex, unwieldy and far too easy to game.

This is what happens when complex rules and regulatory supervision are substituted for transparency.

Actually, no one can calculate proper risk weights. They are unknowable. Analysts at credit-rating companies, even if one sets aside all the conflicts of interest they face, are just as prone to group think, fads and misconceptions as the rest of us. Academics would do no better. And the “wisdom of crowds” -- as reflected in the market for credit-default swaps -- suggested that Citigroup Inc. was a low-risk investment until 2007.

The reason that Citigroup credit-defaul swaps were mis-priced was the combination of lack of transparency into Citigroup's actual exposure details and the Federal Reserve claiming that risk had been taken out of the banking system.

Professor Johnson actually makes the case for requiring ultra transparency.  The "wisdom of crowds" can only be realized if they have the exposure details necessary to assess risk.

The right approach, as articulated by Hoenig, is to choose capital rules that are “simple, understandable, and enforceable.”...

Of course, for any capital rule or risk reduction rule like the Volcker Rule, the enforcement mechanism is market discipline made possible by ultra transparency.

The bottom line:  talk about bank capital is simply a distraction so that we don't talk about what is really necessary for assessing bank safety ... ultra transparency.