Monetary policy during financial unrest
Thank you for the invitation to come and talk about monetary policy at your meeting today.
In Sweden, as in many other countries, monetary policy focuses on achieving price stability. There are two main reasons for this.
One reason concerns monetary policy's contribution to good long-term growth. Economic growth is determined in the long run by such factors as the creation and use of physical capital, the development and use of the labour force's know-how, and technical advances. None of these fundamental growth factors can be affected directly by monetary policy. But what monetary policy can do is create confidence in the future value of money and thereby contribute indirectly to a climate that favours growth.
The other reason has to do with economic stability. When economic activity slackens and resource utilisation decreases, inflationary pressure normally becomes lower and that often provides room for interest rate cuts. Similarly, a tightening of the monetary stance is usually called for when the growth of demand has been exceeding the increase in the economy's capacity, since that normally entails a higher risk of inflation. While the possibility of fine-tuning the economy should not be exaggerated, a focus on price stability should normally also enhance the real economy's stability.
Once price stability has been established as a credible benchmark, it can therefore be said that the path of inflation becomes a sort of thermometer that shows whether resource utilisation in the economy is high or low. It is, however, a thermometer that has to be read with care because it is susceptible to various kinds of temporary factors. But the relationship between capacity utilisation and inflation is a strong one.
According to the latest figures from Statistics Sweden, the price level in September 1998 was 0.5 per cent lower than a year earlier. That does not mean that the Swedish economy is in a phase of deflation. The average price level can fall for a variety of reasons, some of them good and others bad for economic development.
As true deflation is a sign of falling demand, the first thing to note is that the present decline in the Swedish price level is not primarily a consequence of low demand. Its main causes instead are altered indirect taxes, decreased interest rates and a low price rise for imports. It is a matter of shifts in relative prices, with effects on inflation that are mainly transitory. This in turn leads to temporarily higher real incomes and thereby tends to strengthen, not weaken, demand.
Transitory price effects of this type, whether the supply shock that generates them is positive or negative, are not something to which monetary policy should normally react. The underlying rate of domestic inflation, which is largely determined by resource utilisation in Sweden's economy, was not quite 2 per cent in September, which reflects a relatively favourable demand trend. GDP in the first half of 1998 rose 3 per cent from the first half of 1997, which means that the annual growth rate of about 3 per cent which the Swedish economy has achieved since the summer of 1993 has continued this year.
That does not mean, however, that we can sit back and relax. Dangers may be lurking round the corner that monetary policy has to be prepared for. We are in the throes of a global financial crisis that in a short time has changed the world economic outlook—from a situation with expectations of interest rate increases, for instance in the U.S. economy, to one where the talk instead is of a global recession and a need for more expansionary monetary policies. Representatives of the U.S. Federal Reserve have even referred to the financial crisis as the most serious since World War Two, with a greater risk of it worsening instead of easing.
Permit me now to take a quick look at what has happened and say something about what a financial crisis actually means. After that I shall review the various factors which the Riksbank has to consider in its conduct of Sweden's monetary policy.
Developments in the world economy
The countries in Southeast Asia were long regarded as economic miracles and generally dubbed the Asian Tigers. With apparently sound economic fundamentals in the form of good growth, high saving and investment ratios and orderly government finances, assets in these countries were seen as attractive for investment. There was strong confidence that the positive performance would continue. Capital accordingly flowed into these countries. In only half a decade the foreign debt to GDP ratio of the countries in Southeast Asia rose to almost 30 per cent. While that is not, in fact, a high figure, the problems in this case were that almost two-thirds of the debt was short term and that the capital had been accumulated over such a short period. A contributory factor was that, with fixed exchange rate regimes, all exchange risks had apparently been absorbed by the central banks.
The problems were, in fact, greater than investors believed. One of them was that bank systems were not fully prepared for the large inflows. The foreign capital and a rapidly growing supply of domestic credit had been used, for instance, for investment in real estate, shares and businesses that failed to deliver the high long-term return that had been expected. The bubble ultimately burst. Just what triggered the crisis has not been determined but when investors finally began to have their doubts during 1997, the awakening was quick. The capital flows went into reverse. An even stronger reaction occurred when the Thai currency was devalued on 2 July 1997. At the end of July the same thing therefore happened in Malaysia. Next came the Indonesian currency, which fell on 14 August, followed by the collapse of the South Korean currency on 16 December 1997.
My description points to clear similarities with the crisis in Sweden in the early 1990s. Under certain circumstances, a central bank that apparently absorbs exchange risks can create incentives for sizeable short-run currency flows. The term in Sweden was “basket loans”. Together with a notable expansion of domestic credit, a bank system that is not sufficiently mature to cope with a free capital market and a “bubble-generating” economy, this often sets the stage for a financial crisis. That was what happened in Sweden and it also applied in Southeast Asia.
It was generally foreseen that the direct economic effects in the OECD area would not be particularly great. So no one reckoned that the problems in Southeast Asia would constitute an appreciable threat to growth in the rest of the world. What nobody really appreciated, however, was that the Asian crisis would be exacerbated and spread through the world via the financial system.
In the first place the Asian crisis meant that the domestic problems in Japan and Russia, already considerable, became even worse. With the Russian Government's decision on August 17 1998 to suspend payments on rouble-denominated government debt, the crisis was further accentuated and spread on to other emerging markets as well as to the financial system in the West.
The altered risk assessments led initially to investors playing safe and preferring historically secure currencies like the German mark and the U.S. dollar. These currencies strengthened, accompanied by falling bond rates. In that other investments were wound up to reduce future risks or contain existing losses, smaller currencies like the Swedish krona were hit notwithstanding good economic fundamentals. Also hit were the currencies of other countries—Australia, Canada and New Zealand, for example—where the central bank likewise targets inflation with a flexible exchange rate regime.
As the unrest grew in the light of losses and problems in the financial sector, investors also displayed an increasing preference for even more liquid assets. Long bond rates accordingly rose. Efforts to attain the highest possible degree of liquidity are indicative of concern about difficulties in withdrawing from existing positions; quite simply, fears that the market will cease to function in some respect.
What is a financial crisis?
The present situation is troublesome in that it might develop—contrary to expectations—into a proper global financial crisis. What, then, are a proper financial crisis and the phenomenon known as a credit crunch?
To understand this we must look at how banks function in an economy. Commercial banks make a living by accepting short-term deposits from the non-bank sector and lending the funds for longer terms, a strategy that may seem somewhat risky. The business concept is based in large measure on confidence but also on the banks' ability to reduce the uncertainty about the borrowers and their chances of meeting commitments. Sound borrowers must first be distinguished from others, followed by continuous monitoring of well-founded credit assessments. Any negative effects of the uncertainty can also be limited by the banks requiring collateral for the loans they do provide.
If banks cease to be able to reduce this uncertainty and improve, or make up for, an initial lack of information, then they may refrain from providing credits. In such a situation banks may simply no longer be in a position to distinguish between good and bad borrowers and therefore stop lending at all. This cuts off the economy's supply of liquidity and hits lending to consumers and investors as well as to other banks. An involuntary contraction of the credit supply—a credit crunch—arises and may have contagious effects on the real economy in that consumers and firms can no longer obtain the credits their plans require. GDP falls and unemployment rises.
LTCM and the role of hedge funds
The concern that bank lending would be affected was accentuated when the hedge fund Long Term Capital Management (LTCM) ran into serious difficulties. The losses of this fund could have effects that spread to banks in the industrialised countries, both via their direct exposures to this fund and via increased general alarm about the solvency of other players. A sign of this alarm is that the shares of international banks have fallen more than the general level of stock markets.
LTCM has been operating since 1994. Its annual return in the early years was around 40 per cent. The strategy relied on complex mathematical models for identifying temporary interest rate differentials and the probabilities of these differentials returning to more normal levels. With turbulent financial markets, however, the interest rate differentials, instead of displaying the assumed narrowing predicted by the calculations, actually grew. With the large number of the fund's positions, which had previously served to spread risk and provide cover, the losses instead became greater and greater as interest rate differentials widened in virtually every market.
The lesson from this seems to be that no matter how sophisticated the models that are used for investment decisions, the only way of obtaining exceptionally high returns is by taking exceptionally high risks. Here I cannot help rephrasing Tage Danielsson's superb monologue on probabilities:
Before Long Term Capital Management (LTCM) collapsed it was, of course, extremely improbable that what happened would in fact happen. There are still those who think that what happened to LTCM was so incredibly improbable that it no doubt hasn't happened. One can appreciate their incredulity because, according to LTCM's own probability estimates, an event like this happens so seldom. And if what happened to LTCM really did happen, then the risk of it happening again is ridiculously minute. Really it was lucky that it did happen. Because now we know that it almost certainly won't happen again.
The statistics on the situation in the Swedish financial system indicate that compared with banks in other countries, Swedish banks do not have as extensive a degree of direct exposure to risks in the crisis-ridden countries. Their exposures to hedge funds are also small. Amid all the misfortunes of Sweden's bank crisis in the early 1990s there is still this ray of light: in recent years Swedish banks appear to have been fairly prudent about taking excessive risks and have appreciably improved their ability to assess risks.
However, the Swedish market has also been hit by a general aversion to invest in relatively risky and non-liquid assets. This is evident from interest rate differentials compared with the marketed assets, T-bonds, that are perceived to carry least risk. The return investors require to accept assets issued by the private sector, banks included, instead of sovereign assets (the differential between interest rate swaps and T-bonds, for example) has risen markedly. Today the ten-year swap spread in Sweden is roughly 0,8 percentage points. The differential between housing bond and T-bond rates conveys the same picture of aversion to credit and non-liquid investments; with the growing uncertainty since the summer, this differential has risen almost half of a percentage point. Here we have a manifestation of the general concern at present in international financial markets; it should not be over-interpreted as a lack of confidence in Sweden's financial sector and economic future.
Checking against the Inflation Report
In the September Inflation Report the Riksbank assumed that in the coming years annual growth in the OECD area would be between 2 and 2.5 per cent. This represented a further downward revision of our forecast for international economic activity. The main reason was the prospect of activity in Japan and the United States being weaker than envisaged earlier. In Japan this was mainly because of problems of confidence in the light of the situation in the bank sector, while in the United States the international crisis was assumed to act via lower asset prices that dampen consumption as well as via weaker net exports.
In the euro area, however, strong domestic demand was expected to lead to a favourable growth path in the years ahead.
What is the situation today? The outlook is still heavily clouded with unrest that makes it difficult to discern how the real economy will develop in the wake of the financial crisis. The global economy may suffer from the financial crisis through direct effects on production and asset prices. The extent of the consequences will also depend on the degree to which the confidence of households and firms is weakened by the crisis and how far banks are obliged to restrict their lending.
In the United States, households' expectations—previously a close indicator of private consumption—have worsened for the third consecutive month. The stock market, moreover, carries more weight there than in the European economies. Studies show that the earlier increase in share prices contributed around one-quarter of the growth of consumption in the period 1995–98. The appreciation of share prices has led to decreased saving. If households act to restore saving ratios from their very depressed levels, this could have effects on domestic demand.
In Japan, confidence and growth will hinge on the future handling of the new package for bank support. Expectations of the economic future are closely linked to how the problems in the bank sector will be dealt with. The Japanese economy is a clear example of the difficulties in exerting economic influence with political measures at a time when confidence in the long-term solution is low.
The European outlook is broadly positive. Apart from some countries, expectations in the euro area have not deteriorated appreciably. The impact of stock market developments is smaller here than in the United States. But confidence could be affected if European industrial activity were to slacken, with repercussions in the labour market.
A central factor for growth prospects in the world economy is the supply of credit. The central banks recognise this and are carefully following the solvency and lending potential of the banks. The interest rate cut by the U.S. Federal Reserve should be seen in the light of concern both for domestic demand and for the development of the bank sector.
Issues facing the Riksbank
At the end of September the Riksbank concluded that the repo rate should be left unchanged. This was indicated by a relatively positive picture of the Swedish economy, with good growth and a path for inflation in the forecast period that represented a successive return to the target from a low level.
At the same time, the Riksbank underscored the difficulties in assessing inflation prospects. If the turbulence in global financial markets were to have real economic effects, lower inflation might result. On the other hand, higher inflationary pressure might result from a lastingly weaker exchange rate that did not have a counterpart in a deterioration of economic fundamentals.
While it is too early to construct a new forecast, most of the evidence at present seems to suggest that growth in the rest of the world is more likely to be revised downwards than upwards. The key factors will probably be the development of confidence among households as well as among market players, and the extent to which the situation in the international bank sector has a negative impact on credit supply in different countries. The risks of a weaker future in the rest of the world seem, quite simply, to have increased. Under such circumstances it would naturally be unreasonable to suppose that the Swedish economy would be left untouched.
The next question is how domestic demand would be affected by the international course of events. The latest statistics suggest that households' expectations of the Swedish economy have become somewhat less positive but this does not yet appear to have weakened their personal economic expectations. There has also been some fall-off in the growth of corporate borrowing. Current tendencies look fairly strong for retail trade, for example, as well as for indicators of private consumption, but it must be borne in mind that the analysis has to be forward-looking. What matters, in other words, is how consumption develops under the new circumstances.
Meanwhile, the exchange rate tendency, at least to date, has been the opposite of that on which the Inflation Report was based. If it lasts, the marked weakening of the krona represents a strong economic stimulus. This is equally true in floating and fixed exchange rate regimes. The relevant aspect in the context of monetary policy is not so much the weaker exchange rate's effects on import prices as its impact on inflation via increased production and thereby higher resource utilisation. The reason for this is that the pass-through from earlier import price increases to consumer prices has been comparatively moderate. This is not remarkable now that medium-term inflation expectations seem to be fairly firmly anchored around the 2 per cent target.
What the Riksbank has to determine is which of the two effects will predominate. If the weaker international trend dominates the development of inflation, there is a risk of inflation being below the target in the time horizon of twelve to twenty-four months that is relevant for monetary policy. According to our line of thought, this implies lowering the repo rate. If the effect via a lastingly weaker exchange rare predominates, there is a risk instead of the inflation target being exceeded. In such a situation, a good development of demand with inflation in line with the target can only be achieved with a repo rate that is higher than at present. The cyclical outlook must thus be weighed against effects that the exchange rate may have on inflation prospects.
In this context it may be worth repeating once more that the Riksbank doe not target the level of the exchange rate. What matters is the combined assessment of all the factors that affect inflation prospects, including economic activity internationally as well as in Sweden and any stimulus from the exchange rate. The fact that we do not enthuse when the exchange rate fluctuates abruptly is something else again.
The monetary policy conclusion is that we must keep a close watch on the real economy, the financial system and the financial markets. At present we are waiting with a change in the repo rate but that decision may need to be altered in time as the picture becomes clearer.
