Why is the Financial System So Procyclical?

The global financial storm which hit the banking systems of major developed economies and also sent dangerous waves to the emerging world has shown clearly that the financial system exacerbates the cyclical movements of an economy.

The procyclicality of the financial system is not new. It is due to a number of different reasons, which need to be understood before assessing why some financial systems are more procyclical than others and how to best combat such procyclicality.

First of all, the financial system is prone to have a more lax assessment of risk in good times than in bad ones influenced by the economy's general environment. The idea of short-sightedness in economic or financial decision making was introduced by Kahneman and Tversky (1973) and then developed more by Kindleberger (1978) and Minsky (1982), whose contribution was to explain why it is an inherent component of our financial system, branded as the “financial-instability hypothesis.” The “excess,” or overlending which takes place during good times is then corrected during recessions.

Second, borrowers' net worth—as well as cash flow—is bound to be higher during upturns, facilitating their access to credit. This mechanism, introduced by Kiyotaki and Moore (1997), has been branded as “financial accelerator”. In the same vein, the value of collateral is bound to increase in the good times and fall in the bad times. Such asset price dynamics—and the related wealth effects—clearly increase borrowers' capacity to obtain collateralized lending during booms. However, during the subsequent slowdown, it will become clear that the collateral backing the loans did not have the expected value. This crisis shows clearly how important the asset channel can be as a sudden fall in the value of collateral in the mortgage market can have enormous effects on banks' balance sheets and the economy in general. In fact, asset price deflation has been exacerbated, as expected, by debtors' trying to liquidate their collateral to cover their financial obligations, as summarized in the Figure 1 [ PDF 17KB | 1 page ].

Third, banks may also be intermediating the procyclicality of other markets in so far as they depend on the funding they obtain and that is much scarcer in bad times. Such scarcity of funds (either equity or borrowing) will result in a diminished capacity to lend. This has actually been exacerbated by regulation as will be explained at the end of this section.

Fourth, investors—and thus financial institutions as intermediaries of savings—have a strong incentive to behave as their peers—the so-called herd behavior (Rajan 1994; Devenow and Welch 1996)—since their evaluation is done in relative and not absolute terms, which at an aggregate level fosters lending during booms and limits it during recessions. In fact, credit mistakes are generally judged more leniently if they are common to the whole industry. This crisis has definitely confirmed this idea.

Fifth, the classical principal-agency problem between bank shareholders and managers can also feed excessive volatility into loan growth rates. Managers, once they obtain a reasonable return on equity for their shareholders, may engage in other activities that depart from firm value maximization and focus more on managers' rewards. One of these strategies might be excessive credit growth in order to increase the social presence of the bank (and its managers) or the power of managers in a continuously enlarging organization (Williamson 1963).

Sixth, compensation policies are generally such that there is no need to have a classical principal-agency problem for managers of financial institutions to behave procyclically. Bonuses linked to business growth in good times and to business retrenchment in bad ones are a powerful reason for financial institutions to become very procyclical.

Seventh, human capital can not grow as fast as a financial institution does in good times. In fact, when the economy booms, loan officers need to grant loans faster and, probably, in a less rigorous way. Furthermore, the more time that has gone by since the last downturn, the less prepared are loan officers to realize that the economic environment can change very quickly. This is what Berger and Udell (2003) have called lack of institutional memory.

Eighth, the increasing sophistication, harmonization, and automaticization of risk management also add to procyclicality. If we take the example of Value at Risk (VaR) techniques, they basically transform large nominal amounts into much smaller values-at-risk. This reduces the perceived order of magnitude of risk exposures and gives a sense of comfort that may turn out to be wrong. In fact, the current crisis proved that nominal and notional amounts do matter when looking at risk exposures. Furthermore, network externalities also increase risk assumption in the good times and propagate financial distress in the downturn. These risk externalities will tend to be amplified when aggregated across the network as a whole. In order to address this problem, financial institutions should not look at unconditional VaR, but consider conditional VaRs (CoVaRs) (Haldane 2009).

Ninth, competition in the banking system, especially in the commercial banking system, is such that cross subsidization is more and more used to attract clients. An important aspect of cross-subsidization is offering credit access to clients so that they pay commissions for other products. In the good times, this can bring about a relaxation of credit standards for the sake of attracting new clients (Nys [2008] and Lepetit et al. [2008]).

Finally, and very importantly, financial regulation may be an additional source of procyclicality. In fact, traditional loan-loss provisions are tied to loan delinquency. That means that in the good times financial institutions hardly need to provision, while they need to step up provisioning as soon as delinquencies appear. This obviously reduces their available capital and, thus, their lending capacity when it is most needed. In addition, the traditional focus of risk sensitive capital adequacy requirements is not dependent on the macroeconomic situation but only on the type of asset in each bank's balance sheet. The way in which the Capital Adequacy Ratio (CAR) is defined, basically allowing for hybrid capital to be part of it, can also induce additional procyclicality. This is because hybrid capital, to the extent that it has a debt component, is subject to debt market swings. In this regard, a frequently discussed issue is whether Basel II fosters procyclicality even further due to the increased risk sensitivity of CAR. Making capital more risk sensitive is not controversial, the real issue being the right timing for acknowledging this risk: at the time the loan is granted (expected loss approach) or at the time delinquency appears (incurred loss approach). Moreover, as Caruana and Narain (2008) argue, how much Basel II exacerbates procyclicality very much depends on the mitigating measures the local regulator is willing to introduce, basically under Pillar II¹

The increasingly homogeneous assessment of risk and common trading techniques may also exacerbate herd behavior. The same can be said about the introduction of fair value in accounting standards, which may create an illusion of very good solvency based on high market prices during boom periods and a sudden change of such solvency situation as soon as market prices change (Jimenez and Saurina 2006; Taylor and Goodhardt 2006). . This is related to the rules versus discretion debate that will be dealt with in Section 2.1.

Finally, a vicious circle could even be created by procyclical regulation feeding the asset price bubble. This point has recently been formalized by Aiyagari and Gertler (1999) and Gruss and Sgherri (2009).

Download this Paper [ PDF 259.4KB| 32 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.