Banking Soundness and Monetary Policy
Chapter

4 Capital Mobility and Its Impact on the Operations of a Central Bank

Editor(s):
Charles Enoch, and J. Green
Published Date:
September 1997
Share
  • ShareShare
Show Summary Details
Author(s)
JACOB A. FRANKEL

There are many topics that one might cover in a paper presented at a seminar on banking soundness and monetary policy in a world of global capital markets. One could speak about banking; or about soundness; or monetary policy. And then there is the central issue, namely global capital markets, with the emphasis on markets because that is where the constraints and the opportunities come into play. In principle, one can address these broad topics from either a domestic perspective, or an international perspective. However, “globalization” implies that the distinction between the domestic and international arenas becomes blurred. As a matter of fact, in a global environment it is hard to think of anything remaining purely domestic.

In this paper, I will touch on each of these three main topics and how they are interrelated. First, I will make some general observations about banking sector problems in a global setting. A key point is that banking system soundness is interlinked with macroeconomic policies and performance in a number of ways. In particular, the monetary authority cannot pursue medium-term price stability without a sound banking system and without appropriate fiscal policies. In the second part of the paper, I turn to Israel’s experience in dealing with inflation and the exchange rate when capital markets were opened as a means of illustrating the impact of capital mobility on central bank operations geared toward price stability. This practical experience with globalization highlights the difficulties posed by free capital movement and again the overall importance of sound policies. In this context, the central bank cannot achieve price stability on its own: it must work in partnership with the government to control spending or cut deficits when necessary.

Some Thoughts About Banking Problems

To begin, it is clear when one is speaking broadly on the issue of banking soundness or more narrowly about a specific country’s experiences, banking problems are no longer the esoteric subject they used to be. Just a few decades ago, banking dealt with the right and left sides of the balance sheet, and the relevant multipliers. The concept of crisis was not part of the picture. Yet, today, banking problems are prevalent both in industrial countries and in developing ones, and their implications and resolutions are of great concern.

Banking crises inflict high costs and lead to macroeconomic disruptions. So this issue is of interest to bankers and nonbankers alike. The causality also works in the other direction: macroeconomic disruptions, in an environment of global capital markets, can lead to banking crises. This observation has important implications for the design of macroeconomic policies. In previous seminars organized by the IMF, this two-way relationship has been a major theme, with the conclusion that one needs to follow stable macroeconomic policies so as to minimize the likelihood of macroeconomic disruptions and systemic problems in the banking sector. But, by the same token, one needs to follow sound banking principles so as to minimize the likelihood of banking crises and their implications for the macroeconomy.

The question is this: how can one assure the development of sound macroeconomic policies as well as that of a stable banking system? When the chancellor of the exchequer, Kenneth Clark, gave a speech saying he was against a boom, many people in the audience wondered how a finance minister could ever be against a boom. And his answer was that a boom must be followed by a bust; otherwise, it would not have been called a boom in the first place. A boom is not sustainable growth. A similar concept applies also to financial markets and banks, because as Michael Mussa (Director of the IMF’s Research Department) once indicated, the difficulties that are experienced by the banking sector relate to some extent to the macroeconomic cycle of boom and bust. During booms, banks often extend risky loans, and those risky loans turn sour during the bust. This risk is greater when the boom is induced by capital inflows that lead to excessive bank lending, especially, for consumption or for real estate. These flows create the ingredients of a potential bubble, which once it bursts may quickly deflate the whole economy. This then creates pressure on the banking system, and these pressures can be aggravated by the authorities if they delay the necessary remedy, thereby causing a (potential) loss of confidence.

Now, what is needed to promote sound banking? Sound banking requires effective regulation and effective supervision leading to the strict maintenance of capital adequacy ratios. These regulations should also facilitate governance and market discipline through improved information and better transparency. But more generally, sound banking cannot be a static concept because globalization itself is dynamic. And if globalization is a dynamic concept, so are the innovations in the capital markets. And if financial instruments and practices are evolving, so must the regulatory and the supervisory capacity evolve. Therefore, the concept of sound banking itself should always be dynamic. That means that many of the points made in this volume will be obsolete in ten years due to dynamic developments in this domain.

Intervention, Market Discipline, and Fundamentals

Israel, over the past few years, has transformed its exchange rate system. The authorities have intervened at several points to lend credibility to the exchange rate system and to reduce the capital inflows resulting from an appreciation of the currency. But as interest rates were raised to deal with the inflation problem, markets did not expect the authorities to allow an appreciation of the currency because they wanted to protect export profitability. But this implied that uncovered interest arbitrage became a more or less riskless operation or a one-way bet. At some stage, the central bank therefore decided to stop intervening. As we stopped intervening, a major debate ensued because the central bank stopped providing certainty to the markets, and suddenly the business sector was faced with additional risk.

In the end, such worries were unfounded. A short time after the central bank stopped intervening, the volume of transactions in Israel’s forward markets, in other related financial instruments, and in the so-called derivatives markets tripled and then quadrupled. Basically, the business sector found its own way to reduce risk related to the exchange rate mechanism.

It is argued that covering risk in the financial market is not free of charge, but this is a fallacy since the intervention in the foreign exchange market itself is hardly free. It is costly to the economy for a variety of reasons. There are the budgetary implications to long-term intervention, since the central bank must have the capacity to sterilize the intervention. But economically speaking, intervention in and of itself does not necessarily reduce risk. It only shifts it around in the economic system, and possibly makes it even more costly. But in any event, no intervention can right a situation in which fundamentals are out of equilibrium.

Therefore, since sooner or later the fundamentals must be brought into balance, it is better to do so before the problem gets too big. This is true for intervention, and it is true for banking. Regulators should never let a problem in the banking sector become too big, because then the syndrome called “too big to fail” arises. Powerful political interest groups will attempt to protect a large bank that is in trouble. Then all additional interest groups will start to put pressure on policymakers, and before long, there is a budget crisis (as in Israel), a savings and loan crisis (as in the United States), a social security crisis, or a pension system crisis—all crises that could have been avoided had the underlying problem been addressed earlier. An important lesson is this: deal with a banking problem promptly, because it will only worsen with time and by then the central bank will have lost credibility and stronger measures will be needed to achieve any given results.

If that is the case, and the problem becomes too big, then a lot of excuses will be offered as a rationale for delaying the necessary, painful measures and instead temporary steps will be introduced. Yet, there is nothing more permanent than a temporary measure. The temptation to adopt the wrong policies and, therefore, to lose credibility is one of the major reasons to deal with problems at an early stage. The policymaker’s main asset is credibility, and one should always remember that credibility is never owned, it is only rented.

Now, the less credibility accorded to a central bank, the smaller is the authorities’ capacity to deal with the problem at hand. This relates mainly to the short term. As far as the medium term is concerned, structural issues, such as a country’s level of financial market development, are the dominant influence. Intervention, by its nature, contravenes the market, and the more a central bank uses intervention, the more likely it is the development of market mechanisms may be slowed. With intervention, there is the illusory feeling of providing certainty to market participants, and with this a consequent complacency about the need for more sophisticated financial instruments that would strengthen the capacity of the system to withstand storms and stress.

A final remark in this regard—again, drawn from Israel’s experience—has to do with countries that have not yet completed financial liberalization. Without full financial liberalization, whether in the banking sector or other areas, some of the authorities may be excessively reluctant to open up the economy, to dismantle controls, and to agree to the deregulation and integration of capital markets. Each time, one hears apologetic slogans in favor of keeping the underdeveloped system in place. Frequently, this paternalistic approach retards the development of the capital market, thus preventing its integration into world markets and providing a false sense of security. In reality, the only way to ensure thriving capital markets is by adding depth and breadth, which is done by letting more and more swimmers into this pond. Then growth and maturation will occur.

But when a country liberalizes its capital market, should it be done slowly or rapidly? For a long time, I believed the answer was similar to the sequencing approach developed at the IMF—first do this, second do that, and so forth. Indeed, these rules make a lot of theoretical sense. However, I now think, largely because of the political realities faced by central banks, that whenever an opportunity to liberalize arises, one should seize it and not wait until all the preconditions are in place. In France, Pierre Mendès-France once tried to convince the French to stop drinking alcohol and to drink milk instead. A big sign was hung up in the metro of Paris that read: “Don’t drink alcohol. It will kill you slowly.” And somebody added a note underneath, saying, “That’s okay. I’m not in a hurry.” When it comes to reforms, then—and to those touching on banking soundness, supervision, regulation, and harmonization, in particular—one probably should be in a hurry.

With regard to lenders of last resort, the rationale is to provide relief to illiquid banks that otherwise are solvent. This supply of liquidity is intended to reduce the likelihood that a specific bank, or specific banks, in trouble will behave irrationally in the short term, such as trying to raise liquid funds at all costs. In addition, there are systemic considerations, as one bank failure undermines confidence in the banking system as a whole.

There is also a countervailing systemic issue, which comes from the danger of too frequent use and indiscriminate application of the lender-of-last-resort principle. This is moral hazard: if the lender-of-last-resort principle is invoked too frequently, it destroys an important regulatory mechanism, not that of a bureaucrat but that of the market. So one must allow for some failures to take place: some shareholders must lose, some managers must be replaced, where depositors are protected. The very knowledge that such a possibility exists may provide very effective disciplinary market-based regulation, which might contribute to the soundness of banking. I cannot overemphasize this point.

All this said, it is important to remember the primary role of central banking: to maintain price stability. If a country has high inflation, then it needs to be lowered to a stable rate. However, in many countries, the law also places bank supervision under the central bank’s jurisdiction, thus raising a possible conflict between the two objectives. Occasionally it may be necessary to inject liquidity to save the bank, but at the same time it may also be necessary to withdraw liquidity to preserve price stability. What does a central banker do?

To begin with, he or she should make sure that the lender-of-last-resort injection is done as rarely as possible. This is the biggest exception. Manuel Guitián in his writings (for example, see Chapter 3) emphasizes this point time and again: engage in preventive medicine rather than deal with physicians once you are ill. What is true in medicine is also true in banking, as well as in economics. In this way, the alleged conflict between the short term and the long term can usually be resolved.

Specifics of the Israeli Experience

Now I would like to discuss Israel’s experience in dealing with inflation, the exchange rate system, and capital market development, in other words, how does capital mobility impact on central bank operations. 1 begin with some of the background.

In the first half of the 1980s, Israel experienced hyperinflation, up to 450 percent a year (Figure 1). It then successfully implemented a stabilization program, which enabled a very significant reduction in inflation. Following this, inflation got stuck for several years—from 1986 to 1991—at an average rate of 18 percent. Not satisfied with this result, Israel adopted a new system that enabled it to cut inflation by about half. With respect to the future, it has adopted inflation targets aimed at achieving a continuous reduction in inflation.

Figure 1.Rate of Inflation

(In percent)

The Israeli stabilization program used a multiplicity of nominal anchors, and one of the first was the exchange rate of the dollar, which was held stable for a while. Subsequently, the country moved to a basket of currencies as an anchor and held that exchange rate basket steady until the late 1980s (see Figure 2).

Figure 2.The Israeli Shekel Exchange Rate Vis-à-Vis the Basket and the U.S. Dollar

(July 1985-May 1989)

At this point, the authorities realized something was not working. Inflation was about 18 percent and the exchange rate was fixed. Obviously there was a real appreciation of the currency and from time to time there was a need to adjust. But once an adjustment was made, the markets began to guess when the next adjustment would take place. At this point in time the capital markets were still relatively closed, and difficulties from world markets were not an issue. In 1989, Israel decided to adopt an exchange rate band (see Figure 3).

Figure 3.The Horizontal Band System: The Israeli Shekel Exchange Rate Vis-à-Vis the Basket

(January 1989-March 1992)

In the first stage of the band, the exchange rate was formally allowed to vary by 3 percent on each side from the central rate. Since inflation persisted, an adjustment of the band was needed. The band was shifted upward, but again inflation persisted, and the band was readjusted. In the course of this manipulation, the band was widened to 5 percent on each side, which did not help much and, with inflation continuing to be a problem, subsequent adjustments occurred: each time the exchange rate reached the top of the band, the band was adjusted. Somehow it seemed as though we were chasing our own tail.

In December 1991, the Bank of Israel decided to adopt the diagonal exchange rate band (see Figure 4). With its adoption it was very important to decide on three parameters. First, at what position should the central parity be? Second, what slope should it have? And, third, what should be the width of the band? With respect to the slope, which is the key parameter, the Bank of Israel made a strategic decision: it did not opt for a horizontal band because it would have been illusory. A horizontal band could not be sustained as long as Israeli domestic inflation exceeded that of its major trading partners. The central bank decided to use its central rate as a macroeconomic variable, so that if the inflation target is say, 12 percent, and inflation abroad is about 3 percent, then the slope should be about 9 percent. Thus, from the point of view of the exchange rate, the country does not have a built-in appreciation of the currency as an integral part of its system. Such an appreciation may still occur, but it depends on other things in the system. Figure 4 shows this slope started to decline gradually, as inflation started to come down. Initially, the slope was 9 percent, then it became 8 percent, and later 6 percent. Today, it is still 6 percent.

Figure 4.The Crawling Band System: The Israeli Shekel Exchange Rate Vis-à-Vis the Basket

(October 1991-June 1997)

Concerning the width of the band, the exchange rate has been moving within the band, up and down, and the interest rate has been actively used to fight inflation. But when the domestic interest rate was raised, capital flowed in—in response to the differential in expected returns. And as capital came in, the domestic currency started to appreciate, and then competitiveness started to erode. At that stage, the Bank of Israel had to decide what to do.

It could print more money, thereby creating a nominal depreciation, knowing full well that it was a short-term solution because, before long, prices and wages would go up and erase any real effect from the depreciation. If no real depreciation is achieved, there will be no result except for raising inflation. Or, alternatively, it could try to maintain monetary control. But to do so means being able to sterilize all the purchases of foreign exchange by the central bank to protect the exchange rate.

The first condition, then, to adopting such a system is that if a country wants to have an open capital account while trying to disinflate, it must be sure that its monetary policy still has bite in the economy. And how can it have bite? It must allow the currency to appreciate when the interest rate goes up. If at some stage it has to intervene, it should delay doing so as much as possible. When it does intervene, it should make sure it has enough instruments to sterilize the monetary injection.

The second condition is to have the monetary instruments available to sterilize the intervention. Obviously sterilized intervention cannot go on for very long. If authorities open up the capital account, they should be prepared to widen the exchange rate band sufficiently. Here, indeed, Israel did widen its band, but it did not help much because the interest rate remained high, and the currency appreciated. Once it has reached the floor (fully appreciated within the band), the degrees of freedom for policy become very limited: either lower the interest rate and give up the band, which is a “stop and go” situation, or alternatively follow the Chilean example and spread sand in the wheels, namely, tax capital inflows, which is very difficult to recommend. If a country does not have capital controls, it should not impose them; if it does have them, it should think twice before dismantling them. It is a one-way street, since it is very difficult to enforce capital controls after markets get used to free capital movements.

What, then, are a central bank’s options? Here, the government, the fiscal authorities, play a role. One of the reasons why raising interest rates causes appreciation of the currency is because government spending does not adjust. If at the same time that the monetary authority raises interest rates the policy mix remains sensible—namely, the government cuts its spending or cuts its deficit—then one creates room for offsetting the appreciation in real terms. A cut in government spending creates the potential for a real depreciation; and a rise in the interest rate facilitates the potential for a real appreciation, particularly in the short run. Thus, to prevent a real appreciation, policymakers should make sure that the budget is indeed, contractionary, or less expansionary, at the time that a contractionary monetary policy is planned.

Finally, in this regard, one should never allow the fight against inflation to be viewed as the sole concern of the central bank. It must be the decision of the government as a whole. Why? After all, is not the independence of the central bank desirable? Why is the government needed as a partner? Here an important distinction must be made between independence of objectives and independence of instruments and the capacity to implement policy. In democracies, the objectives must be set by the government. Therefore, the central bank should not work against the government’s objectives when fighting inflation. Fighting inflation should be the government’s objective. Independence simply allows the central bank to decide how to use the tools at its disposal to carry out the policy laid out by the government.

Conclusion

How does this partnership between monetary and fiscal authorities relate to the broad theme of banking soundness? It links through the fact that a fragile banking environment ties a monetary authority’s hands, because it is very difficult to implement a sensible monetary policy in such an environment. There are many reasons, and the government will find them all, why not to fight inflation when the banks are fragile.

Therefore, to carry out an anti-inflation policy in a medium-term perspective—the only perspective in which one can fight inflation—one must make sure that the two true preconditions are in place: that the fiscal policy in place does not overburden the budget, and that the soundness of banking be assured, because fighting inflation is going to be pretty tough on banks. When interest rates are raised, those banks whose lending portfolios are not fully robust may face difficulties. A stable overall banking environment thus gives the anti-inflation stance a chance to prevail, which is the primary role of monetary policy.

    Other Resources Citing This Publication