In September 2007, I wrote an article in which I argued that banks’ profits were under pressure and banks were facing a crisis. Consequently, the Fed and ECB would have to lower interest rates, and banks would have to take losses and increase their capital. The latter has now happened with more than $500bn in writedowns or losses and more than $350bn in capital increases, according to Bloomberg.
While the banking sector is the proximate cause of the crisis and there were other factors, I believe the crisis has its origins in regulatory changes with large unintended impacts in the financial sector, specifically, the 1988 Basel I capital accord (updated 1998 version here) prepared and completed under the supervision of the Bank of International Settlements, endorsed by the Group of Ten central bank governors, and fully implemented by 1992; and the 1990s reduction of bank minimum reserve requirements by the US Federal Reserve and other central banks.
These regulatory changes have to be seen in the context of the 80s. The growing role of non-bank financial intermediaries and high interest rates put the business model of banks under growing competitive pressure, which led banks to take on increasing risk. A banking crisis ensued, notably, the US Savings and Loan Associations crisis. The regulatory changes sought to protect taxpayers from liabilities arising from moral hazard incentives of deposit insurance, and sought to create the conditions whereby banks could more effectively compete with the new non-bank financial intermediaries.
Specifically, Basel I's aims were to ensure that banks kept high capital ratios, to guarantee a competitive “level playing field” between banks from different countries, and to penalize risky behaviour by means of a target risk weighted capital requirement, such that the higher the risk of assets held by the bank the higher the required capital.
However, Basel I created a strong incentive for banks to increase their exposure to certain types of assets, deemed less risky by Basel I planners, such as interest rate derivatives, agency securities, and mortgages. For example, under the Basel framework, the risk of holding Fannie Mae (FNM) and Freddie Mac (FRE) debt was 20% of the risk of holding an industrial loan of identical amount. Little wonder then that Fannie and Freddie were able to grow so much.
Indeed, nearly all financial asset classes that Basel I endorsed with lower risk weightings have grown rapidly since 1992, and are right at the center of the current financial crisis. According to the Bank of International Settlements, the notional amount of outstanding over-the-counter [OTC] interest rate derivative contracts, which had a very low risk weight, the lowest risk weighting among derivative instruments under Basel I, have grown to over $319 trillion, or 75% of the total OTC derivatives exposure. In summary, despite the good intentions underlying them, the risk weightings of the Basel I accord are a bad idea, and have contributed instead to an increase in the overall risk assumed by banks.
Fortunately, all US banks must comply with US-specific minimum leverage ratio regulations implemented in 1990, which do not depend on risk weights. In addition, according to these regulations, only institutions with at least 5% leverage are “well capitalized”, where the leverage ratio is defined as the ratio of so-called “Tier 1” capital to adjusted average total non-weighted on-balance-sheet assets. Unfortunately, these regulations do not include off-balance sheet assets.
While off-balance sheet accounting predates the accord, Basel I legitimized and made it acceptable to hold assets off-balance sheet, by defining the credit factor and risk weighting of different types of assets, and was also permitted under Financial Accounting Standards Board [FASB] accounting rules. This loophole meant US banks had an incentive to instead expand their off-balance sheet assets. Likely as a consequence, according to the Financial Times, there are an estimated $5 trillion in off-balance sheet entities that would have to be brought to the bank balance sheets according to new FASB rules proposal, an amount that does not include all off-balance sheet assets held by banks, and that represents nearly half of the total assets ($11 trillion) of US commercial banks.
According to March 2008 Federal Deposit Insurance Corporation data, the leverage ratio for US banks was a quite healthy 7.5%, slightly higher, in fact, than in December 1992, the earliest date Tier 1 data is available. However, an analysis of the annual reports of large US and international banks indicates that large banks have much lower leverage ratios.
In December 2006, on the eve of the financial storm, Basel I instilled a false sense of security by showing Tier 1 risk-weighted capital ratios of large banks around a conservative 8%. For example, Citigroup (C) had an 8.6% Tier 1 risk-weighted capital ratio and a 5.2% leverage ratio. However, according to its 2006 annual report, Citigroup had $1.7 trillion off-balance sheet assets, which meant its overall leverage ratio was 2.5%. In other countries, banks had even lower leverage ratios. For example, Barclays Bank (BCS), Deutsche Bank (DB), and UBS (UBS) had leverage ratios of 2.3%, 2.2%, and 1.7% to on-balance sheet assets, respectively, while displaying Tier 1 risk-weighted capital ratios of 7.7%, 8.9%, and 11.9%, respectively. If my analysis is correct, Basel I conferred legitimacy to increasingly risky behaviour by banks, by painting an incorrect picture of the financial strength of banks.
Finally, and most importantly, Basel I risk weightings directed the allocation of financial resources towards certain economic activities deemed by the Basel I planners to have less risk, as shown by the empirical results of several academic papers.
The Basel I capital accord was defined in such a manner that it meant common sense and facts had no relevance in decision-making, similar, in fact, to what often happens in planned economies and large bureaucracies. A bank faced with two financial applications, one for a Fannie Mae security and another for a promising industrial loan, might prefer the former since, in terms of scarce bank capital, the Fannie Mae security cost it only one fifth of the industrial loan. No matter how well intended and knowledgeable, it is the essence of market economies that the choices of central planners will ultimately prove inferior to those made by some risk taking market participants, as has now become apparent.
The second main factor that allowed banks to take increasing risks was the reduction in the minimum reserve ratios, implemented in several steps in the 1990s by the Federal Reserve and other central banks. These reductions in the minimum reserve requirements aimed at reducing the bank costs associated with non-interest yielding reserves.
As a consequence, average US bank reserves (including surplus vault cash) fell from 2.9% of total deposits in January 1990 to 0.9% in March 2008. While the change may seem small, as the ratio approaches zero, credit creation by banks grows exponentially. Explained in a somewhat oversimplified manner, in 1990, a deposit of a newly Federal Reserve issued US dollar would be expected to originate, on average, total loans of $33, whereas in 2008 the same one US dollar deposit would be expected to originate, on average, total loans of $110.
Banks face two inter-related but separate constraints when they attempt to grow their loan book and assets: bank capital and minimum reserve requirements. Basel I and the off-balance sheet loophole allowed banks to skimp on bank capital. The changes implemented in the 1990s by the Federal Reserve and other central banks, relaxed the other remaining and more important constraint on the banks ability to grow. Thus, there was what must have seemed like unlimited demand for Basel-endorsed assets, and the unregulated part of the market (think subprime mortgage lenders) duly rose to the occasion and supplied the banks with the goods in the format required by Basel (e.g., AAA rated CDOs).
While most consider the Federal Reserve’s decision to sharply lower interest rates in 2001-2003 the cause of this crisis, I believe it was an important aggravating factor. It provided an incentive to increase leverage and speculate on the short and long term interest rate differential, namely through interest rate derivatives. In addition, it also rescued several banks from having to take some hard decisions. Had the interest rate not been lowered by so much, a few of these banks would have failed, but more importantly, the managers and the strategies that had been followed at the banks that survived would have changed. Consider, for example, how many top managers lost their jobs since August 2007. Instead, the low short term interest rates allowed these firms to keep on using the same strategy. The consequence is that the level of risk in the financial system is much larger than it was in 2002 and therefore the crisis is far more difficult to address.
Still, the above-mentioned regulatory measures do not excuse banks from their ultimate responsibility. The regulations did not force banks to buy only what was perceived as being low-risk assets, nor force them to expand their off-balance sheet liabilities. Large banks, in particular, were free to exercise good judgment, but instead chose to lead.
There are few palatable options available. Bank failures are costly and painful events to be avoided to the extent possible. Yet, a market system is based, in no small part, on failure. Among other things, failure importantly means the loss of equity and debt for those economic activities where there was previously an excessive allocation of economic resources.
Unfortunately, the situation is such that losses of, for example, 5% of total assets, which are not unconceivable given the size of large banks’ trading assets and current estimates of the total losses by the IMF and others, would mean insolvency for several large US and international banks. Therefore, it is no longer realistic to think that capital increases alone will suffice to solve the crisis.
The financial health of large banks can only be restored through a restructuring of their liabilities, so as to put the large banks’ Tier 1 leverage ratios well above 5%. The question then is which groups of liabilities owners have to bear the cost of the restructuring. I believe that, despite moral hazard and legal considerations, deposits and inter-bank loans, which represent the bulk of bank liabilities, should be fully protected, to avoid bank panics and ensure that the credit creation process is not unduly affected by the crisis. This means that the remaining liability owners, foremost equity and debt holders, which represent around 25% of US bank liabilities, would have to bear the brunt of the balance sheet retrenchment. Indeed, the reduction of the large banks’ liabilities, necessary to put the banking system in a more sure footing, has to happen before the economies can recover. The sooner it is done, the better.