Persson: Monetary policy in a low-inflation economy
Introduction
Thank you for the invitation to take part in this morning seminar. I intend to begin by giving my views on the current prospects for the economy and inflation, including the risks associated with, among other things, the unwinding of the present global imbalances. I will do so against the background of the minutes from this year’s first monetary policy meeting, which were published last week and which gave an account of the Executive Board’s discussion on 27 January. Then I will discuss the Riksbank’s remit, different ways to conduct monetary policy and the possible implications of the low inflation rate for future monetary policy.
The current monetary policy situation
Since Christmas the news flow has been dominated by the huge tidal wave disaster that hit the countries in the Indian Ocean. Despite the scale and extent of the disaster the consequences for the world economy will probably be very small. That at least has been the conclusion from the different international meetings that have taken place after Christmas.
International economic activity is thus expected to continue to strengthen and resource utilisation in the world economy to become increasingly higher in the coming years. Inflation has risen slightly in most countries, owing mainly to the increasing oil price. But excluding oil, inflation has not risen at the same rate. There are plenty of spare resources in several of the biggest economies, and in recent years competition has toughened in many industries. As a result, international inflation is forecast on the whole to be relatively moderate in the coming years.
The economic situation in the US and the euro area
In the US the economic situation is continuing to improve. In the fourth quarter last year GDP in the US rose by 0.8 per cent compared with the previous quarter, partly due to a sharp increase in investment. This reflects a benign investment climate in the US with historically high profits and optimism among firms. Indicators of growth prospects for the fourth quarter provide a positive picture of the activity level in the US economy.
As resource utilisation has picked up the Federal Reserve has raised its key policy rate from 1 per cent to 2.5 per cent. Many factors suggest that we can expect the tightening to continue in the spring. Inflation is estimated to rise in the period ahead, partly because the weakening of the dollar in recent months has made US imports more expensive. In the main scenario of the Riksbank’s most recent Inflation Report the US economy is expected to continue to perform strongly in 2005 and 2006.
One source of concern for both the United States and the world as a whole is the large deficits in the US economy. In the third quarter last year the current account deficit totalled USD 165 billion, or around 6 per cent of GDP. Data for November show that the deficit has continued to grow, reaching a new record high of just over USD 60 billion. This may result in changes in interest rates and exchange rates that could have considerable consequences for world trade.
A large share, or about 40 per cent, of Sweden’s exports go to the euro area, and as a result economic developments there are especially important for us. The recovery in the euro area economy is going slowly. Growth in the third quarter was lower than expected. Available indicators for the fourth quarter provide a mixed picture. Germany, which is an important market for Swedish products, is showing some signs of an improvement. After several years of extensive cutbacks there should be a pent-up need among firms to invest. At the same time exports are being dampened by the euro’s rise against mainly the dollar.
The economic situation in Sweden
As regards Sweden the Riksbank has been positive in the past year in its outlook on economic developments. International demand has increased faster than anticipated, with above all exports of services proving unexpectedly robust.
Economic policy is also contributing to the relatively bright growth prospects. Interest rates are at an all-time low and the proposals in the Government’s Budget Bill imply a looser fiscal stance. Household consumption will continue to grow steadily in the period ahead, and business investment also appears to be picking up again following a number of years’ weakness. Taken together, this means that GDP growth in 2004 is forecast to be around 3.7 per cent, after which it will be dampened somewhat, standing at about 3 per cent in 2005 and 2006, which is a fairly high level in historical terms.
Despite the firm growth Swedish inflation is very low – in January it stood at
0.4 per cent, measured as the annual change in UND1X. The weak inflation is primarily due to surprisingly low import prices. Other causes are the strong productivity growth; the krona’s appreciation in the past half year, above all against the dollar but also against the euro; and low electricity prices. There are many indications that we won't be seeing any significant inflationary pressures in the coming year either. That made it natural for us to decide not to raise interest rates at the end of January. My personal view is that it will probably be a while before a rate hike comes into question.
The risk outlook
Surrounding this rather bright picture are a large number of risks that our growth and inflation forecasts will not be met in the coming two years. The key risk factor is how the US deficits will be corrected. Will it come about through a sharp correction in the dollar and, if so, what will the consequences be for the world economy?
The majority of forecasters agree that the large deficits in the US are not sustainable. It’s true that the weakened dollar is expected to be able to lead to a certain stabilisation of the current account, but since the US imports around 1.5 times more than it exports there would have to be a sharp reduction in imports or an even steeper rise in exports for the deficit to be able to narrow.
At the same time, demand in the US has to a great extent been the engine for the world economy in recent years. Expansionary economic policy has contributed to this and also resulted in a very large saving deficit. The inflow of capital to the US has stemmed, among other things, from Asian central banks’ purchases of US government securities. China has recently accounted for about one-third of the US external borrowing requirement. In that way, both China and Japan have built up substantial dollar reserves.
The logic behind this loss-making activity (if the dollar continues to fall in value) is that it enables Japan and China to maintain low values for their own currencies, thus facilitating continued exports to the US and other countries. For China the substantial exports translate into jobs and incomes in an economy where a very high percentage of the population still live in the conditions of a developing country. Fast growth means increasing prosperity, which in turn may result in a stable social and political climate. However, in order for China to switch from export-led to domestic demand at a quicker pace – which has to happen in the future – it will require institutional and other reforms that take time to implement. The financial infrastructure in particular is in great need of change and development.
Where will demand come from when the economy is tightened in the US? The weak growth in Europe and the uncertainty surrounding developments in the Japanese economy means that there is no obvious candidate to take the United States’ place as growth engine. While China has recently become the world’s third biggest trading nation it can hardly assume the mantle on its own. Domestic demand in Europe needs to pick up but the current upswing here is sluggish and is being restrained by efforts to tackle public finance deficits in countries such as Germany, Italy and France.
So how likely is it that China will abandon its fixed rate with the dollar in the period ahead? Most commentators believe that China will take a gradual, slow and cautious road towards a floating exchange rate. Should the dollar continue to weaken, therefore, the euro will continue to bear a large part of the adjustment in the future as well, thus having an adverse effect on European competitiveness and growth.
Another important growth factor is how interest rates develop in the period ahead. There is little to suggest that the US federal budget deficit will narrow to a sufficient extent. That is worrying, especially given that the country in the long term will see sharply increasing costs for medical care, health care and pensions. As a result there is a clear risk of rising market rates in the US that could trigger a fall in house prices, which in turn could dampen private consumption. A reduction in public expenditures would also have a tightening effect but in the long term would lead to a more balanced development and thereby prevent an overly quick rise in rates.
Another risk factor is the oil price. As the US is more oil-dependent than the euro area countries, its economy is more sensitive to changes in oil prices. Should the oil price remain at a high, fluctuating level it may curtail growth in the US.
All in all, this picture of large global imbalances implies a clear risk that growth will be more subdued over the coming two-year period. The question is how fast this slowdown may come about. If controlled, the moderation would be softer, while rapid changes could entail a considerable weakening of the world economy and thereby also check Sweden’s economic performance in the period ahead.
In addition to the competition problems that an ever stronger euro means for Europe there is the issue of how we are to meet the competition from low-wage countries, both the new member states of the EU and the emerging economies in Asia. The differences in labour costs are very large. The toughening competition places great demands on the ability of Swedish economic players to adjust. What can Sweden do to attract and keep new investment that can create jobs and incomes here? Swedish investment is among the lowest in Europe and must in all likelihood increase as a share of GDP to ensure continued growth.
The global exchange rate adjustments may result in more subdued growth, which makes them a big uncertainty factor in the period ahead. The rebalancing of demand between currency areas takes time, and the relative price changes via the currencies creates, in the same way as inflation, an uncertainty that may have an adverse impact on both investment and demand.
The Riksbank’s remit and inflation targeting policy
The Swedish parliament has decided that the Riksbank’s objective shall be to maintain price stability, and the Bank has since defined this to mean an annual change in inflation, measured in terms of the consumer price index (CPI), of two per cent, with a tolerance for deviations of plus/minus one percentage point. In practice we often choose to look at a measure (UND1X) that excludes the effect of changes in indirect taxes and subsidies as well as mortgage interest expenditure – i.e. what we can call political or administrative decisions.
Our interest rate decisions are based on forecasts of price developments one to two years ahead. If our forecasts are higher than the inflation target the Bank normally raises interest rates, and vice versa. In other words the policy rate is normally adjusted so that inflation 1-2 years ahead, taking into account the balance of risks, is in line with the target.
The Riksbank introduced the current regime of a floating exchange rate and inflation target in 1993, after a number of decades with a fixed exchange rate. The new regime has meant a new way of working whereby we attempt, with the aid of monetary policy, to adjust economic growth to what the inflation target allows. A common analogy is that the economy is like a car with certain characteristics that enable it to drive at certain speeds but not at others. In that case we in the Executive Board are the driver, using the policy rate to accelerate or slow down. If the car is going too fast, its tyres will start to overheat and we risk driving off the road, i.e. prices will rise and we have to raise the policy rate to slow down. Conversely, when the car is going too slowly we accelerate by lowering the policy rate.
In reality the analogy with a car is somewhat misleading. First, it is difficult to know exactly the speed at which the economy is actually growing. The statistics that we use when forecasting economic growth contain errors, do not provide a comprehensive picture of the economic situation and furthermore are produced with a lag. Another problem is the difficulty in knowing exactly how long it takes before a rate adjustment has the desired effect. Since it takes time – we allow for a time horizon of one to two years before the transmission mechanism has worked through the economic system – we have to form an opinion of the state of the economy in the long run.
Predicting the future is difficult for many reasons, not just due to statistical shortcomings but also because the economy is changing all the time. The economy can be hit by different kinds of shocks, e.g. the increased electricity prices when there was a shortage of water in Sweden’s reservoirs, and the price of meat and other food when cows in Europe were affected by a number of diseases. It is difficult to determine early on what changes in the economy are temporary and what are permanent. Another factor complicating the picture is that inflation expectations in society greatly influence price and wage formation, and thereby inflation.
In other words the car – to continue with the analogy – has a fogged windscreen, a speedometer that is very difficult to read and pedals that sometimes are sluggish and sometimes easy to push down.
How can we make the economy go at an appropriate speed?
In somewhat simplified terms there are two different ways to describe monetary policy based on inflation targeting: The first is to employ a simple monetary policy rule whereby the policy rate is set as a function of the current inflation rate and the output gap, which is a measure of the difference between actual and potential output and reflects resource utilisation in the economy. (1) When the output gap is negative, actual output is lower than potential output, which means that there are plenty of spare resources in the economy. In such a scenario price and wage increases are small and inflation is generally below target.
When the output gap is positive the situation is the reverse: resources are overstretched and the excessive demand generates inflationary pressures in the economy. So according to the rule the policy rate should be raised if growth is above its potential rate and inflation is higher than the inflation target, and vice versa in the reverse situation. When actual and potential output are the same the economy is in balance and inflation is normally in line with the target, provided that policy is credible.
A common argument in favour of conducting monetary policy in this way is that the central bank’s knowledge of how the economy functions is too limited for a forecast-based monetary policy to be formulated in a good enough way. Incorrect forecasts can also burden the central bank with credibility problems, which is one reason to prefer a simple rule.
The other method, applied by the majority of central banks including the Riksbank, is to forecast how the economy will evolve over the next couple of years and adjust monetary policy accordingly. One complication here is that something unexpected can always happen, causing the economy to move in a different direction than that expected. For this reason the central bank must take account of different risks when setting the policy rate. The forecasts are produced by experts through the use of various statistical methods and models. The results are often presented in the shape of a main scenario and one or more alternative scenarios.
A monetary policy that is based on forecasts requires more information about the economy than a rule-based policy, and according to its critics there is a risk that forecasters will overestimate their ability, perhaps leading to poorer results than would have been the case with a less complicated approach. Advocates of the method, on the other hand, say that we should make full use of the knowledge that we nevertheless have.
Since a forecast-based monetary policy requires the analysis of many more variables than is the case for a simple rule, it is more difficult for external players to predict how the central bank is going to act. For this reason a forecast-based policy places high demands on the bank’s openness and transparency in relation to external parties, as well as on its ability to communicate.
There are also other advantages associated with a forecast-based monetary policy. Since it is forward-looking the central bank can at least try to foresee future problems or structural changes and meet them through a rate adjustment. The ability to act quickly is essential, as monetary policy acts with some lag and the economic cost of increasing inflation becomes lower if it is checked at an early stage. Another benefit is that the central bank does not become as tied as with a rule, but instead can take account of alternative scenarios and risks. Moreover, these assessments can be regularly updated in the light of new information about economic developments. (2)
Flaws in the data underlying decisions and the difficulty of predicting what will actually happen in the future mean that the ability to evaluate and weigh up a large amount of complex information is what most distinguishes an accurate monetary policy from a less accurate one. Judgement is therefore highly significant for how good the final result is. This favours the use of a forecast-based monetary policy, which allows greater scope for judgements than a simple rule.
Openness and communication
The choice of monetary policy strategy is of key importance for a central bank’s external communication. Good communication with external players is extra important for central banks that conduct a forecast-based monetary policy. It is therefore important that the Riksbank is open about how we view economic developments and the inflation outlook and about how we have reasoned in our conduct of monetary policy. The aim is to create an understanding of what governs the Bank’s interest rate decisions and to generate support among the Swedish public for the inflation target.
This is also a reason why we in the Executive Board often give public speeches. Openness and transparency from our side can help stabilise expectations regarding inflation and interest rate developments and lower the risk of unease in the financial markets. So inflation expectations play a key role for monetary policy and for that reason we keep a close eye on how they are developing. (3) Perhaps the most important task of monetary policy is to influence inflation expectations since it is via market rates, and not the repo rate itself, that demand in the economy is affected. That is why communication and credibility are so significant. If we can get our message across we increase the chances of a successful monetary policy.
Monetary policy in a low-inflation economy
The Riksbank’s rule of action, which says that the policy rate should be adjusted when inflation is above or below target one to two years ahead, is not at all as mechanical as is sometimes perceived. The Executive Board’s clarification in 1999 made it clear that the practical formulation of monetary policy takes account of developments in the real economy.
According to the clarification there may be reason for the Riksbank to deviate from the target if inflation is being affected by a temporary shock that will not fully dissipate during the coming two-year period but that is nonetheless not deemed to have any lasting influence on inflation developments. Another reason is when a bigger shock makes a return to the target one to two years ahead associated with excessive undesirable effects on the real economy.
There are many examples in recent years of when we have taken such factors into account. At the turn of the year 2002-2003, for example, rising energy prices were pushing up inflation quickly. But since the high energy prices were expected to gradually fall back entirely we saw them as temporary and decided to leave the repo rate unchanged.
When the current regime of inflation targeting was established in the early 1990s it had been preceded by a long period of high, fluctuating inflation in both Sweden and other countries. Many countries around the world experienced a similar shift in regime around this time. Partly as a result of this, international inflation fell sharply during the 1990s, from about 30 per cent to below 4 per cent a year on average. (4)
The inflation targeting regime and its practical application has been underpinned by the conviction that high, variable inflation is bad for the economy. Now we are living in a world with very weak price increases, particularly for manufactured goods. Is there anything in the current circumstances that could motivate a change in the Riksbank’s thinking? The integration of China's and India's large, fast-growing economies into the world economy will most likely affect world growth and inflation for a long time. Globalisation has been given a powerful push forward. Even though setbacks are possible, it is reasonable to believe that the supply shock produced by Asia’s low-price exports will continue to depress prices in the world for many years ahead. China’s government has set a target of quadrupling its GDP during 2000-2020, as happened over the last twenty-year period. In rural parts of China there are hundreds of millions of people who dream of a better life in the cities and of working in a modern company.
In Sweden we can see how inflation in recent months has turned out lower than we expected, because price increases for imported goods have been weaker than anticipated. The productivity gains that we have witnessed, and that have also contributed to such a low inflation rate, may also be due to the increased global competition. Another dampening effect can be expected from the abolition of the import quotas for textiles.
At the same time there is as usual reason to believe that world market prices will increase as economic activity improves. The rapid growth in the new industrial countries is also pushing up the prices of commodities, not least oil. It is therefore uncertain how long import price pressures will remain very low. In any event this is a very important issue for monetary policy in the coming year, and at the Riksbank we are keeping a close eye on import price developments.
Allow me to round off with some thoughts regarding the stance of monetary policy. Given that inflation is anticipated to be below target for a large part of the coming two-year period, I think one can argue that there has been scope to lower the repo rate somewhat more than we have done recently. But I have, like my colleagues, decided against this. There are several reasons for that – interest rates do not always have to be changed when inflation prospects differ from the target.
The reasons are that growth is currently firm. We believe that the problems in the labour market are chiefly due to structural factors; it is therefore likely that they would only be marginally affected by lower interest rates. Moreover, the demand for labour is expected to begin to rise during 2005-2006, Swedish monetary policy is already loose, real interest rates are low, household borrowing is increasing quickly and house prices are rising - and have been for a long time. Even lower rates could further fuel the willingness to borrow, entailing risks in the long term in the shape of a growing property bubble.
In addition, if monetary policy is to try to fully offset the low import prices by driving up domestic prices, for example in the service sector, this could lead to a domestic inflation rate that would be difficult to stop if economic activity were to continue to develop strongly for a longer period and resource utilisation thereby to rise.
The clarification of monetary policy that was published in 1999 can thus be applied to today’s situation: Out of consideration for developments in the real economy, i.e. growth and employment, in the somewhat longer term, we can refrain from lowering the repo rate now, even though a mechanical application of our rule of action could justify such a move.
But should economic activity weaken, for example due to the risk scenario that the adjustment of the US deficits could trigger, the situation will be different. Then I can see reason to lower the policy rate further, especially if the weakening is coupled with a stronger krona. In such a situation I believe it would be an advantage that the repo rate is not even lower than it is today.
Literature
Apel, M. Nessén, M., Söderström, U. & A. Vredin (1999), “Different ways of conducting inflation targeting – theory and practice”, Sveriges Riksbank Quarterly Review 4, 13-42
Bernanke, B. “The Logic of Monetary Policy”, speech before the National Economists Club, 2 December 2004, The Federal Reserve Board
Palmqvist, S. & Strömberg, L. (2004), “Households’ inflation opinions – a tale of two surveys”, Sveriges Riksbank Economic Review 4, 24-43
Rogoff, K. “Disinflation: An Unsung Benefit of Globalization?” December 2003, Finance & Development
Notes:
4) This is known as the Taylor rule. The US economics professor John B. Taylor is Under Secretary for International Affairs at the US Department of the Treasury. He has formulated a simple rate-setting rule that has come to be known as the Taylor rule and that has become very popular. This is one of many examples of monetary policy rules.
2) See the speech by Federal Reserve Board Governor Ben S. Bernanke, "The Logic of Monetary Policy" (2004). In his speech Bernanke discusses advantages and disadvantages of following a simple monetary policy rule compared with a forecast-based policy. See also Apel, M., Nessén M., Söderström U. & A. Vredin (1999), "Different ways of conducting inflation targeting – theory and practice", Sveriges Riksbank Quarterly Review 4, 13-42, for a similar discussion that is linked to the monetary policy framework in Sweden.
3) Read more about this in, for example, Palmqvist S. & Strömberg L. (2004), "Households’ inflation opinions – a tale of two surveys" in Sveriges Riksbank Economic Review 4, 24-43.
4) See Rogoff, K. (2003), "Disinflation: An Unsung Benefit of Globalization?"
