Serious financial crises go through seven distinct phases. First is the precrisis phase in which the authorities should be, and sometimes are, practicing crisis prevention. Too often, the crisis may be brewing, but the authorities are either in ignorance, or in denial, of that fact.
Second is the outbreak of the crisis, which in retrospect is linked to a particular event, such as an action by a French financial institution to freeze access to funds it is administering. The action itself is irrelevant except for its use in dating the start of the crisis, which by that time was probably inevitable.
Third is the crisis management phase, in which authorities and institutions grapple with an ongoing cascade of events with little time to chart their next move or to ponder the implications of their moves.
Anna Gelpern has recently written about the fourth phase, which she calls "crisis containment."1 This is a phase in only the most serious crises like the present one, in which the rulebook is thrown away and the overriding objective is to stop the bleeding. Ultimately, the bleeding does stop and the fifth phase begins, the "mopping up" phase.
In the sixth phase of a crisis lessons are, or are not, learned. Seventh and finally, preparations are made to prevent or minimize the virulence of the next crisis. Generally, lessons are only partially learned and incompletely applied.
In any case, we know that financial crises are inevitable. Some say they are a healthy feature of market economies.
At present, we are somewhere between the containment and the mopping-up phases of the crisis of 2007-09. Consequently, it may well be premature to think that now is the time to learn and to apply the lessons of this crisis. However, it is not entirely unreasonable to start that process even if we lack full information and detachment.
Some lessons are common to all the serious financial crises that I have witnessed over more than 30 years.
First, serious crises are tamed, in the end, by a multipronged set of responses. Early searches for simple, one-dimensional solutions fail. Success comes only after a comprehensive, multipronged approach is adopted.
Second, serious crises require the application of overwhelming force. Half-measures are insufficient to arrest their evolution.
Third, the principal obstacle to applying the first and second lessons is the political and institutional context in which the crisis itself unfolds. The evolution of any crisis is highly path dependent.
I will illustrate some of these points in the context of the current global economic and financial crisis. I will first offer a few comments about what is different about this crisis. Next I will look at the policy responses that have been adopted. They can be usefully grouped under three headings: improving the macroeconomic environment, promoting market stability, and advancing structural repair. I will conclude with some cautious evaluation.
What Is Different About This Crisis?
Each financial crisis bears a resemblance to other crises and passes through similar phases. However, each crisis has its own unique characteristics. Three features distinguish this crisis.
First, the proximate origins of the crisis were in the United States. Regardless of the amount of blame for the crisis one places on US macroeconomic policies on the one hand, and on financial regulatory policies on the other, the fact is that the United States led the way into the crisis.
Second, if the largest economy in the world, whose currency and institutions are at the core of the global financial system, stops functioning, the fact that the resulting crisis is global should not be surprising. In this crisis, the citizens and authorities of a country can run, but they cannot hide.
Third, it is not unusual for a crisis to begin in the financial sector, spread to the real economy, cycle back to further weaken the financial sector, and thereby further weaken the real economy.
Diagnosis of the economic and financial situation, if anything, was more complicated in this crisis. During much of 2008, global growth appeared to be holding up in general, and inflation, particularly in commodity prices, was still rising. Policymakers were slow to learn that they were dealing with two severe crises on a global scale, in the financial system and in the real economy. The statistical evidence about dual crises in the traditional industrial countries on average points toward financial downturns leading downturns in the real economy. Upturns in the real economy lead the financial upturns, except for equity prices, which generally are contemporaneous.2
I think it is useful to group the policy actions taken in response to the crisis of 2007-09 under three headings: improving the macroeconomic environment, promoting market stability, and advancing structural repair.
Because of the connection between the real economy and the financial sector, it is useful to consider, first, the policies that have been employed to improve the macroeconomic environment during this crisis.
Starting in the summer of 2007, the monetary authorities generally acted quickly to adopt measures responding to the demand by financial institutions for increased access to central bank liquidity. Central banks relaxed the terms of access to discount windows and employed a variety of similar mechanisms. Those actions were reasonably well coordinated. Coordination between the major central banks was reinforced in December 2007 with the establishment of Federal Reserve swap facilities that, over the following 10 months, were expanded in size and in participation, though that expansion raised other issues.
For many countries, the use of conventional monetary policy and interest rate reductions came later. On occasion, monetary policy actions were coordinated, but this was common only during the second year of the crisis. As part of the containment phase of the crisis, a few central banks have embraced so-called unconventional monetary policies with respect to what they buy, on what terms, and whether they worry about the consequences for the liability sides of their balance sheets.
It was not until the second year of the crisis, starting in late 2008, that discretionary fiscal policy came to be widely used by those countries with scope to do so. It is difficult to conclude that fiscal policy actions have been coordinated, but they are all pointed in the same direction.
A second category of policy measures has been to promote financial market stability more directly. These measures often blend into monetary policy actions, and they contribute to structural repair as well as to market stability. However, this type of measures is distinct in that it is directed primarily at actual or perceived market failures rather than at institutions, individually or collectively. Generally, these measures employ the central bank's balance sheet, initially constrained by its interest rate target and more recently unconstrained. The broad objective is to provide liquidity to specific markets, for example the commercial paper market, and thereby to restore their functioning as suppliers of credit. One measure of this type that does not involve central banks is increasing the coverage of deposit insurance, which in effect is designed to stabilize the market for bank deposits as well as to stabilize deposit-taking institutions.
The United States has gone further in this direction than most other countries. I suspect that there are two reasons for this. First, nonbank financial institutions make up a larger part of the overall financial system in the United States than in other countries. Therefore, direct market-related measures are thought to be necessary to restore the flow of credit. Second, the Federal Reserve conveniently has the power to act under "unusual and exigent circumstances" to use its balance sheet to facilitate this type of action. Thus, it was easier within the US political system to go to the central bank, where the money is, than in some other countries.
A third category of policy measures has been aimed at advancing structural repair. They are the most controversial and diverse. At least five types of action with progressive levels of intrusion in the operations of financial institutions have been employed. First are guarantees of the liabilities (other than deposits) of certain types of financial institutions, generally banks but, for example in the United States, money market mutual funds as well. The use of this type of action has been widespread. Second are capital injections. They have been employed in many countries, but with different conditions attached depending on the jurisdiction and the circumstances of the institutions. Third are programs to deal with legacy, or toxic, assets by moving those assets off of the balance sheets of institutions or by isolating them on balance sheets. A fourth type of action is the takeover of a financial institution (or of group of institutions) either de jure or through a process that does not involve formal nationalization but amounts to substantial governmental control. Finally institutions have been forced into mergers or into bankruptcy. These two events differ in many respects but they have a common structural characteristic: The institution disappears.
I have identified a dozen policy actions in three broad categories. In principle, each action could be evaluated in terms of its effects and effectiveness. I think that is premature and not very illuminating. I will limit myself to six observations.
First, action in all three categories-improving the macroeconomic environment, promoting market stability, and advancing structural repair-has been essential in this crisis. I think it is potentially dangerous to examine any one action or even any one category in isolation. For example, does anyone really believe that for the global system as a whole, structural repair of a country's financial system could proceed successfully without improvement in the macroeconomic environment? To take another example, were the liquidity measures taken by central banks early in the crisis adequate and effective? Clearly they were not effective in halting the spread of the crisis. In that sense they were inadequate. However, that statement does not lead me to conclude that central banks should not have tried to stem the crisis using those means. Without such action the crisis almost certainly would have been worse.
Second, it was natural during the early phase of the crisis for the authorities to want to avoid extreme measures. Most, but perhaps not all, of us think that economies and financial systems are generally self-correcting. We would like to keep official interventions to a minimum because they shift the incidence of economic and financial consequences, for example from shareholders and creditors of specific institutions to the general public, and because they distort future incentives. At the same time, most of us support moderate countercyclical measures at least to cushion the effects of economic and/or financial crises.
Third, the very fact that the authorities wanted to avoid using extreme measures early in the crisis meant that they failed to be as forceful and decisive as they should have been in retrospect. For example, when the US Treasury asked the US Congress for what was, in effect, resolution authority over Fannie Mae and Freddie Mac, in retrospect it would have been wise to have obtained such authority covering all nonbank financial institutions. (Resolution authority already existed for banks but not for bank holding companies or other types of financial institutions.) The Congress might not have granted such authority, which probably is one reason why the Treasury did not ask for it. The Treasury also may not have asked for it because it did not want to signal a concern about a larger set of institutions.
Fourth, the example of the lack of resolution authority illustrates the constraints that the authorities operate under and the "path dependence" of their actions to deal with crises. To have a complete and appropriate "crisis management plan" means that the authorities must have perfect foresight about the evolution of the crisis, instant hindsight about its causes, and a comprehensive set of tools to deal with the crisis. Therefore, a certain amount of trial and error, learning by doing, and use of ad hoc measures is inevitable. One consequence may be that some courses of action are foreclosed, or at least their use is delayed. Does anyone really believe that the US Troubled Asset Relief Program could have received congressional approval in October 2007?
Fifth, it is useful to think about the cross-border effects of the dozen types of policy action that I outlined under the three categories. Almost all of them have at least indirect effects on other countries working through the political process, and most of them have direct economic or financial effects as well. Few of those effects are unambiguously positive. For example, fiscal stimulus in one country boosts demand in other countries, but the lack of parallel action in other countries may constrain the first country economically, financially, or politically from doing the right thing.
Sixth, this crisis has revealed a number of open issues in international financial crisis management. The most prominent issue is the international coordination of financial support for individual financial institutions or financial systems. On the whole, the level of cooperation has been admirable. However, it has not been perfect. The most egregious deficiencies, in my view, were in the areas of deposit insurance and other forms of guarantees of bank liabilities.
A second issue involves the process of cooperation and coordination of resolution authority with respect to financial institutions, including their bankruptcy. One wonders whether the global system would have been better off if the various authorities had put more financial institutions into bankruptcy or the equivalent, as some critics have advocated.
A third issue is transparency and the associated topic of accounting standards for financial institutions. For example, the US Supervisory Capital Assessment Program has been criticized here in the United States as well as in Europe. In the United States, most of the criticism is that the program did not go far enough, probe deep enough, or reveal sufficient information about the 19 financial institutions that were subjected to the assessment. In Europe, most of the criticism is that the program went too far in probing the balance sheets of individual institutions and in revealing the results of those probes. This is not an issue that is going away.
I have deliberately not included among my observations anything explicit about international regulatory reform. International regulatory reform is important. It will have indirect consequences for crisis management going forward, but it is part of the sixth and seventh phases of this crisis when some lessons will be learned and applied.
1.Gelpern dates the start of the containment phase of the current crisis to March 13, 2008 when the US authorities confronted the issue of what to do about Bear Stearns. Anna Gelpern, 2009, Financial Crisis Containment, Connecticut Law Review 41, no. 4 (May), 493-549.
2. Stijn Claessens, M. Ayhan Kose, and Marco E. Terrones, 2009, "A Recovery without credit: Possible, but . . ." VoxEU.org, available at www.voxeu.org (accessed on June 3, 2009). The authors' analysis is based on their IMF Working Paper 08/274, What Happens During Recessions, Crunches, and Busts?