Limitations to Commitments Must be Recognized: A collapse of confidence triggers a crisis

KOBAYASHI Keiichiro
Senior Fellow, RIETI

The revelation of the fact that governments and banks do not have ability to keep their commitments and the resulting collapse of confidence are common factors underlying the major challenges observed in recent years, including the global financial crisis, sovereign debt crisis, and fiscal problems caused by an aging population. As the world has awakened to the natural fact that debtors—whether private companies or governments—may default on their obligations, the conventional wisdom whereby we can have confidence and trust in counterparties in transactions has collapsed.

A lack of commitment has long been subject to economic analysis as a factor for various economic problems. A typical example of this is the prisoner's dilemma in game theory, where two accomplices are interrogated by the police separately and must choose between "cooperating with the other and staying silent" and "defecting and confessing to the crime." In the case described in the Figure below, both prisoners choose the latter and gain a payoff of five each because such choice is believed by both to be more beneficial regardless of the choice of the other. However, if both prisoners are given a chance prior to the interrogation to make a commitment to cooperate with each other and if both are able to keep it, they would obtain a better result with each gaining a payoff of 10.

Figure: A case of the prisoner's dilemma
Two accomplices (prisoners A and B), being interrogated by the police in separate rooms, are to choose between "confessing" and "staying silent." The police tell them as follows: if both confess, they will be sentenced to two years in prison; if both remain silent, they will be sentenced to one year in prison; and if one confesses and the other remains silent, the former can go free on probation while the latter will be sentenced to three years in prison. The payoff for the two prisoners (or the degree of their satisfaction) is as follows:
Figure: A case of the prisoner's dilemma

♦  ♦  ♦

Recent years have seen an emphasis on the problem of commitment in financial and macroeconomic research. A series of analyses conducted by New York University Professor Mark Gertler, European Central Bank researcher Peter Karadi, Princeton University Professor Nobuhiro Kiyotaki, and others since 2009 show that impairment of a bank's equity capital due to the collapse of an asset bubble exposes the limit of its ability to keep commitments, and, unable to procure funds for lending, the bank is compelled to tighten credit further. Their conclusion is that a lack of banks' ability to keep commitments is an integral factor in amplifying the crisis.

The banking theory developed in the 2000s by University of Chicago professors Douglas Diamond and Raghuram Rajan explains the form of bank deposits known as demand deposit contracts (see the Keywords section) as a systemic device to make up for shortfalls in a bank's ability to keep commitments. Banks providing demand deposit contracts that offer "the freedom to make withdrawals at any time" cannot betray the trust of their depositors; any attempt to do so would see large numbers of depositors rush to withdraw their deposits, instantly reducing the bank to insolvency. In other words, a bank providing demand deposit contracts and creating bank run risk for itself is demonstrating to depositors its ability to keep its word, but the price it pays for this is the occasional bank crisis.

♦  ♦  ♦

A lack of ability to keep commitments on the part of policymakers also poses a significant problem. In the case of a commitment made between private-sector players, failure to fulfill it can be punished by the government (judiciary). However, governments are rarely punished if they fail to fulfill their policy commitments. This problem was first pointed out by Finn E. Kyland, currently professor at the University of California, Santa Barbara, and Edward C. Prescott, currently professor at Arizona State University, in their 1977 paper on "time inconsistency" (see the Keywords section). Subsequent development of research on the problem of time inconsistency in monetary policy has led to the current consensus that having central banks commit themselves to follow predefined rules is preferable to allowing them the discretion to set monetary policy in accordance with the situation.

Arguments by Prescott et al. are based on the assumption that policymakers are able to commit themselves to follow rules. In reality, however, it is impossible to make long-term commitments because political decisions in general can be reversed by an election. A 1998 paper written by Professor Timothy J. Besley of the London School of Economics and Political Science, and Stephen Coate, currently professor of Cornell University, points out that inefficient policy decisions stem from the problem that politicians in a representative democracy cannot commit to the future outcome of policy decisions which they make today. They call this problem "political failure" to draw a parallel with the notion of "market failure."

In a series of papers written in the mid-2000s, Professor Daron Acemoglu of the Massachusetts Institute of Technology (MIT) and Professor James Robinson of Harvard University analyzed the same phenomenon, which they called "political-economic failure," not only in democratic nations but also in non-democratic ones. For instance, those in power may intentionally make policy choices that would undermine the economic growth of their country and cause poverty to persist in order to maintain their power. The essential cause of such failure is a mutual lack of ability to keep commitments, for instance, between politicians and citizens and between different political parties.

♦  ♦  ♦

A lack of ability to keep commitments is closely related to redistribution mechanisms such as social security and public pension systems, because there is no way to force future generations—those not yet born—to maintain and conform to the mechanisms created by the past generations. Despite this, why do social security systems continue to gain political support and exist? Recently, many attempts have been made to analyze this question using a political-economic approach.

Regarding pay-as-you-go public pensions programs, in which benefits are paid out of contributions made by the current working population, standard economic theory (i.e., overlapping generations model) states that while such a program increases social welfare in a society where the population growth rate is higher than the real interest rate, it works to the opposite where the population growth rate is lower than the real interest rate or in a society with a declining population, resulting in a decrease in the social welfare of all generations. That is, in a society where the population continues to dwindle, today's young people will receive a smaller amount of pension benefits than today's elderly because the population of the next generation—those who are to support today's young people in their old age—will be smaller.

If we were to design institutional arrangements from scratch, a society with a declining population might be better off going without a universal pension scheme and instead implementing welfare policy specifically designed to rescue the elderly poor. It would be unrealistic for Japan to abolish the existing public pension system. Yet, considering the need for care, changes in typical family structure, and so forth, the country where the population is aging at an unprecedented pace may as well discuss the possibility of overhauling the system in a manner similar to the liquidation process.

Actually, in macroeconomic terms, a shift from the pay-as-you-go system to a funded system is almost equal to the complete privatization (or abolition) of the public pension system. In a 1999 paper, Thomas F. Cooley, currently professor at New York University, and Jorge Soares, currently associate professor at the University of Delaware, showed that it is politically possible, even in a society where the presence of a public pension system is detrimental to the social welfare of all generations, to choose to maintain the system. In another paper published in the same year, they described a possible approach to privatizing pension systems fully while maintaining or improving the social welfare of all generations.

Putting them into historical context, the ongoing sovereign debt and social security crises can be defined as a consequence of the grand experiment with the welfare state in the 20th century. The past century witnessed two major experiments on the roles of governments and markets, namely, communism in the Eastern bloc led by the Soviet Union and welfare states pursued by advanced countries in the Western bloc. The failure of communism taught us that governments are not more efficient than markets in distributing resources. The assumption that governments are wiser than the private sector proved to be wrong. The lesson we should be learning today from the experiment with the welfare state is that governments are just as incapable as the private sector of making commitments over a very long period of time, i.e., they cannot commit themselves to what the policy decisions made today will bring in the remote future.

Even if we draw a plan for a social security system that is claimed to be able to last for 100 years, no one in the current generation can guarantee it. Instead of making a promise that is impossible to keep, we should redesign the existing system into one to which we can commit ourselves and assume responsibility over during our lifetime, which I believe is our obligation to future generations. For instance, we can design the pension system in such a way that the level of benefits is adjustable depending on circumstances so as to ensure that the total amount of contributions paid in and that of benefits paid out over a 25-year period will be balanced. Such a system might be resilient to problems arising from a lack of commitment.

A lack of commitment over an extremely long period of time is a problem underlying not only social security but also many other policy issues including the question of whether or not to continue nuclear power generation. It is therefore necessary to redesign government policies in all areas premised on the fact that the government lacks the ability to make and keep commitments over an extremely long period of time.

>> Original text in Japanese

* Translated by RIETI.

Keywords

  • [Demand deposit contracts]
    Current deposits and ordinary deposits are demand contracts that allow depositors to withdraw a discretionary amount (up to their account balance) at any chosen time. Seen in the Diamond-Rajan model as a commitment device enabling banks to gain the trust of depositors, demand deposits are also generally regarded as insurance policies that provide depositors with insurance (liquidity insurance) should they face a sudden need for cash.
  • [Time inconsistency]
    Suppose that a government commits to a certain policy and the private sector has acted on the belief that the policy will be implemented. There are cases where the government feels compelled to scrap its promises and such a situation is referred to as "time inconsistency." A government promise to exempt taxes on capital assets as a way to promote investments is one example. The government is tempted to break its promise because, once constructed, capital assets are hard to be reduced even if they are taxed.

November 19, 2012 Nihon Keizai Shimbun

December 20, 2012

Article(s) by this author