Wickman-Parak: Some reflections on the current situation
Hawks and doves – opinions may differ on this way of trying to characterise me and my colleagues on the Executive Board of the Riksbank, but it cannot be de-nied that it has had a major impact. How often the terms are used in the media varies somewhat over time depending on how great the difference of opinion among the Executive Board members is assumed to be at the moment. Prior to the most recent repo rate decision, the terms were used quite frequently and it was claimed that there was a fight between the hawks and doves on the Board about the direction the interest rate should take.
The minutes of the latest monetary policy meeting, which were published the day before yesterday, reveal that I am one of the doves – if I myself should now lapse into this ornithological terminology. This time, I would like to add – it is not my intention to characterise myself as a permanent dove, no more than I intend to become a permanent hawk. As a monetary policy decision-maker you cannot sit in a predetermined rut, even though this would make life easier for those who like to express themselves in clichés.
The minutes also reveal that the Board was divided right down the middle, as was indeed speculated on beforehand. It was the Governor's casting vote that de-cided the issue and that led to the repo rate being increased rather than being left unchanged.
Before I explain what lay behind the decision and present my own view, I would like to say something more general about the even result of the vote. There are a number of observations and reflections that can be made and that I feel are worth highlighting. I intend to conclude by saying something about the new sta-tistics that have been presented since our meeting. A great deal has happened recently, not least on the financial markets.
An entirely divided Executive Board is unusual
A first observation is that if we look to the past it is very unusual for the Board to be split down the middle. During the almost ten years that the Board has existed and the approximately 85 monetary policy meetings that have been held, this has in fact only happened four times, counting the latest meeting. The most usual result, not surprisingly, is that the Board has been unanimous. Fairly frequently, however, at about a third of the meetings, at least one member has entered a reservation against the interest rate decision.
As an aside it may be worth pointing out that for some time now it has perhaps been more noticeable than before if we are in disagreement. These days, we also have to decide on a forecast for the repo rate. At the meeting in July, for exam-ple, all of the Board members agreed to increase the interest rate. But three of us felt that the forecast for the repo rate path in the period ahead should be some-what lower. This disunity attracted a lot of attention in the media and also pro-vided some clues that made it easier to forecast what the vote would be at the latest meeting.
Previously, before we began publishing interest rate forecasts, there were also different opinions among the Board members about how the interest rate should develop in the future. But as no forecast for the repo rate was published and there was nothing concrete to enter a reservation against, any differences of opinion were not visible to the outside world. It is probably also the case that the future development of the repo rate was not discussed as much internally then as it is today.
Disagreement a healthy sign
Another thing that I feel is worth pointing out is that disagreement among the members of the Board should not be seen as a sign of weakness. This would per-haps be a reasonable interpretation if you held the view that the Board members should be so wise that we can simply "calculate" which decision is the only pos-sible correct one and agree on it. Unfortunately, the world is far too complicated for this to be possible. And if it were possible there would be no real reason to let the monetary policy decisions be made by a Board with six members. It would be enough to have a single wise decision-maker.
In recent decades, however, the trend has been in the opposite direction. All over the world, it has become more and more common to let monetary policy deci-sions be made by a group of individuals – a committee or board – rather than by a central bank governor alone. Probably the most important explanation for this is that many central banks have become more independent in this period. This has made it seem more natural to delegate the right to make decisions to a group of people rather than to one individual.
This development has meant that research on monetary policy decision-making in groups has accelerated. 1) One result that has emerged is that groups tend to make better decisions than individuals. 2) This may be because the members of the group have different backgrounds, knowledge and experience that they can bring to the monetary policy discussion. These differences mark the decisions that the group makes in different ways so that these decisions tend to be better than the decisions that an individual would have made alone - the whole becomes greater than the parts.
This applies irrespective of how the decision is made – by voting, as at the Riks-bank, or in some other way. So, the fact that we disagree should be seen as a sign that everything is working the way it is supposed to. It would in fact be sur-prising if the entire Board supported every interest rate decision made.
Unusually difficult to assess the situation
The third thing I would like to point out is perhaps the most important. This is that the even vote is of course more than anything else a sign that it is unusually difficult to assess the current situation – there is a fine balance to be drawn. I think that it is important to say this as this is not really the picture that has emerged from the debate. Instead, an unchanged repo rate has most often been portrayed as the only reasonable and self-evident alternative. Even though I my-self eventually came to the conclusion that the interest rate should remain un-changed, I must say that I did not feel that this was an entirely self-evident deci-sion.
The Executive Board thus decided by the smallest possible margin to increase the interest rate by 0.25 percentage points. Three main reasons were given for this. First, inflation has continued to increase. In July it reached the highest level measured since the mid-1990s and was far above the inflation target. The much-publicised revision of the CPI by Statistics Sweden last week does not significantly alter this picture, by the way, even though the figure for June is now higher than that for July.
Secondly, inflation expectations are also still high. Different measures paint slightly different pictures, but all of the measures are above the inflation target. Thirdly, cost pressures have increased. This is largely related to the fact that growth in productivity has been surprisingly weak for some time.
The assessment made by the majority was that an increase was needed to pre-vent price rises for energy and food from spreading to other areas – to prevent the high level of inflation from becoming entrenched. The increase is expected to help bring inflation back to a level close to the target within a couple of years. Resource utilisation is also expected to fall but to be roughly normal at the end of the forecast period.
At the same time as the repo rate was increased, the repo rate path for the period ahead was adjusted downwards compared to the decision taken in July. This is partly because growth has slowed down more than expected in Sweden and abroad and partly because the price of oil and the prices of other commodities have fallen.
The increase may be a way of safeguarding the real economy in the long term
A common reaction to the decision to increase the interest rate was that the Riksbank pays too much attention to inflation and too little attention to the real economy. I would like to comment on this claim.
One reason for increasing the interest rate was that an increase was needed to prevent the high level of inflation from becoming entrenched. This could happen if the price increases for energy and food begin to spread to other areas. This could ultimately lead to a risk of the Riksbank's ability to hold inflation in check being called into question and of inflation expectations remaining at a consis-tently high level. This is what happened in the 1970s in many parts of the world, although inflation at that time became entrenched at much higher levels than would be the case today. But even though the levels are much lower today, it could still be relatively costly to get inflation back to the target. It cannot be ruled out that this could require quite a long period with high interest rates and re-stricted demand.
There is thus a risk that conducting too expansionary a policy today would sooner or later give rise to costs in the real economy that, furthermore, could be quite considerable. I do not feel, therefore, that it is correct to say that the increase in the interest rate is an indication that the Riksbank does not care about the real economy. It is rather the case that by focusing clearly on the inflation target we will be able to avoid considerable costs to the real economy in the slightly longer term.
I should perhaps add that the forecast in the Monetary Policy Update, as I just mentioned, is that resource utilisation will be roughly normal, rather than low, at the end of the forecast period. So from this perspective too we can question the claim that the Riksbank pays too little attention to the real economy.
In my view, all the arguments for an increase in the interest rate that were put forward by the majority are reasonable. So even though I voted to leave the repo rate unchanged I respect the majority's view. But this does not mean that I share it. It is possible to have different views of how great the risk of contagion effects is, and here I have a different view than the majority.
My own view
So, what was my thinking at the time? As I said in the minutes, I already saw clear signs of a slowdown in the Swedish economy at the time of the monetary policy meeting in July. The outcome figures and indicators that have been pre-sented since the July meeting also clearly indicate a weakening of economic activ-ity. Resource utilisation, as calculated by the Riksbank using various measures, is currently estimated to be higher than normal, but is falling rapidly. Capacity utili-sation according to Statistics Sweden and the National Institute of Economic Re-search, is still at a high level but is moving downwards. The downswing in the Swedish economy has come sooner and been more pronounced than was ex-pected only a few months ago. It covers the manufacturing industry as well as the retail and service industries.
Households have clearly become more pessimistic, not least with regard to the national economy and the labour market. For example, the survey conducted by the National Institute of Economic Research at the end of 2007 showed that an excess amount of 40 per cent of the households believed that unemployment would be lower in a year's time. Now, however, an excess amount of 50 per cent of the households believes that unemployment will be higher one year ahead. This deterioration in general sentiment has been accompanied by a fall-off in pri-vate consumption. This applies mainly to capital goods, which is common at the beginning of an economic downswing.
But it is not only private consumption that has declined; investments and exports have also weakened. GDP growth has been revised downwards for both 2008 and 2009, although the forecast predicts a recovery in 2009 in line with devel-opments abroad and it is believed that Sweden's exports will be an import driving force in this context. However, I am far from sure that such an upswing in eco-nomic activity will actually take place. Developments in the USA play a key role here and the recovery that is expected to begin next autumn presupposes that the housing market there has stabilised, that the repercussions of the financial crisis have faded and that employment has begun to increase again. As the inter-national forecast is highly uncertain, this also applies to the recovery that is pre-dicted to take place here in Sweden next year.
When it comes to inflation, it is of course impossible to deny that it is uncom-fortably high today. However, I do not believe that the outcome figures contain evidence that shows that the price increases for oil and other commodities have spread to other areas. The repo rate increase in July was also intended to prevent such contagion effects. The risk of these effects has also declined recently, partly due to the fall in the price of oil and other commodities and partly because eco-nomic activity is declining.
The development of productivity has been a disappointment and we do not have a really good explanation of why it has been so weak. Weak productivity also helps to keep cost pressures high. However, I was more worried about this earlier when economic activity was higher and the pressure on the labour market was stronger. In the current economic situation it is probably difficult for the compa-nies to pass cost increases on to the consumers.
There is every reason to continue to keep a close eye on inflation expectations as they are still some way above the inflation target. My assessment is, nevertheless, that they will decline as economic activity weakens and the actual rate of inflation falls.
To sum up it can be said that the Executive Board agreed on a great deal – that economic activity is weakening, that inflation is declining and that it will be possi-ble to reduce the interest rate in the period ahead. Our disagreements related primarily to our different assessments of the strength of the processes and of how much monetary policy would need to help to reduce inflation. My final assess-ment was that inflation in the period ahead will move towards the target, in roughly the same way as illustrated in the Monetary Policy Update, without the need to increase the interest rate. I thus voted to leave the repo rate unchanged.
Nor did I support the repo rate path presented in the Monetary Policy Update. The path had been lowered compared to the assessment made in July. But I felt that the force of the economic downswing may easily make it necessary to re-duce the interest rate more than is indicated in the new path.
New statistics and developments on the financial markets
Some important statistics have been published since we made our decision on the interest rate. These include the inflation figures for August which showed that the CPI increased by 4.3 per cent over the last 12 months. This was 0.2 percentage points lower than our forecast in the Monetary Policy Update. Everyone who produces inflation forecasts knows that it is difficult to exactly assess monthly changes and there is nothing remarkable about a forecast deviation of 0.2 per-centage points. A factor that contributed to the forecast error was that Statistics Sweden, when it published the new figures, also downwardly revised the inflation figures for the period January to the end of July this year. It is of course unfortu-nate that miscalculations are made in a variable that is of such central importance to the Riksbank and many others, but it should be pointed out that this is very rare. The revision attracted a lot of media attention and there was speculation that the Riksbank would have decided differently if this miscalculation had been known at the time the decision was made. The July figure for the CPI was then 4.4 per cent, while the revised figure is 4.1 per cent.
Allow me therefore to comment briefly on this. It need hardly be said that my decision, and that of the two colleagues who also entered a reservation, to rec-ommend an unchanged interest rate would not have been effected. Whether it would have affected the decision of the majority I cannot of course say. But it should be remembered that it is the processing and analysis of a great quantity of information that leads to the overall assessment of economic development and inflation prospects. And it is the forecast derived from this information that largely determines the interest rate decision. In our latest assessment this was based on the belief that inflation was somewhat higher than now proves to be the case. But monetary policy does not have such a degree of precision that it can control inflation to a tenth of a degree. And, moreover, inflation in July was high and far above the inflation target even after the downward revision of the figure.
In the discussion ahead of the interest rate decision, I and other members of the Executive Board highlighted the weak labour market indicators which uniformly pointed to a weaker development of employment, which is also in line with the Riksbank's forecast. After the decision, the Labour Force Survey for August has been presented. This shows that employment increased by a moderate 0.5 per cent on a 12-month basis, which is marginally lower than the Riksbank's forecast.
At our September meeting, we had access to the initial version of the National Accounts for the second quarter. These showed that GDP did not increase at all compared to the previous quarter and increased by only 0.7 per cent on a calen-dar-adjusted12-month basis. Now that the revised version of the National Ac-counts has been published, we can see that growth in the first two quarters has been marginally adjusted downwards in both cases. This means that growth in the first quarter was 2 per cent on a 12-month basis and with calendar effects taken into account. The corresponding figure for the second quarter was 0.6 per cent. The new figures thus confirm that there has been a sharp downturn in the growth rate and completely support the picture that could be discerned at our latest meeting.
Increasing cost pressures have been an important factor behind the series of in-terest rate increases introduced by the Riksbank. The new figures from the Na-tional Accounts show that the figures for hours worked have been revised up-wards compared to the version available to us in September. Together with the marginal downward revision of the figures for growth, this means that productiv-ity declined more than indicated by the preliminary figures. As the figures for la-bour costs per hour worked have also been adjusted upwards, the new figures show that unit labour costs increased by 5.3 per cent on a 12-month basis in the second quarter. This was almost a percentage point more than in the preliminary version of the National Accounts.
The fact that productivity is weak may be because companies have not had time to adjust their workforce as the downturn in economic activity has been so sud-den. We can now see that the number of vacancies is falling and that plans to recruit new personnel are being toned down. An adjustment is underway which will be followed by a recovery in productivity, in line with the Riksbank's forecast. Rising costs are of course a factor that the Riksbank has every reason to monitor and analyse carefully. But as I mentioned earlier, my view is that given the down-turn in economic activity that is now becoming apparent it is difficult to see how the companies would be able to pass on increased costs.
At our latest meeting, the price of oil had fallen from its peak level in July, and it has fallen further since then.
Let me say a few words about developments on the financial markets. We are now seeing a new wave of unrest sweep across the international financial mar-kets. This is nothing new but something we have had to live with for over a year now. From time to time it has appeared that things may have calmed down somewhat, only for the unrest to return in other forms. Over the last few days, major American investment banks and a number of other players have experi-enced acute problems.
It is difficult to say at the moment where all this will end. We are of course fol-lowing developments closely. So far, we can note that financial stability in Swe-den is satisfactory. Swedish banks are only marginally involved as their exposure to the US market is limited. Nevertheless, the Swedish market is significantly af-fected by events on the international stage. The costs of borrowing are increasing and some financial markets are working less efficiently than normal. Yesterday, the National Debt Office, in consultation with the Riksbank, took steps to facili-tate the functioning of the market. The National Debt Office will now issue short-term treasury bills in a number of extra auctions. This is very positive and will al-leviate the shortage on this market that has arisen in the wake of the global fi-nancial turmoil. The Riksbank's assessment is that the unrest on the financial markets is not affecting financial stability in Sweden. We can note that the Swed-ish banks have made very good profits in recent years and are financially robust. However, if the banks were to experience liquidity problems that could threaten financial stability we are ready to take measures to provide liquidity assistance.
It will take time for the uncertainty on the financial markets to abate and for the situation to stabilise in a lasting way. We have already seen how the financial un-rest has left its mark on the development of the real economy, and uncertainty about how it will affect growth prospects in the period ahead remains. At the end of October, the Executive Board will once more make an overall assessment of the growth and inflation prospects.
Exaggerated focus on the current interest rate decision
I would like to conclude with a final reflection. It is natural and important that monetary policy is debated. Sometimes, however, it feels as though a little too much focus is placed on the current interest rate decision. To say that a lot has been written about the latest interest rate increase would be somewhat of an un-derstatement. The discussion conducted in certain quarters almost gives the im-pression that this increase will throw Sweden into a bottomless pit of recession. Opinions may differ on the value of such prophecies, but I still think it is difficult to see how an interest rate increase of 0.25 per cent could have such a decisive impact.
In this context there is an important point that has fallen into the background a little. It is not individual interest rate decisions but a series of decisions – an inter-est rate path – that determines the effects that monetary policy will ultimately have on inflation and the real economy. If we look back in, say, three years time, the development we will see will not of course depend solely on the fact that the Riksbank raised the interest rate in September 2008 but also on what happened with the interest rate thereafter, and to a certain extent before this too. And in the forecast that the Riksbank presented a couple of weeks ago the repo rate path was lowered and the interest rate is lower than in the July forecast for the entire forecast period.
If, moreover, it turns out that economic activity weakens even more and that in-flation falls more rapidly than the Riksbank now predicts it will of course be pos-sible to adapt the policy in line with these developments. It is naturally not possi-ble to withdraw an interest rate increase "retroactively". One can, on the other hand, if it becomes apparent that economic activity is declining, reduce the inter-est rate more rapidly than was originally intended. There is, in other words, a de-gree of flexibility in monetary policy, both upwards and downwards of course, that should not be underestimated.
With this comforting thought in mind I would like to thank you all for your atten-tion.
Notes: 1) For a review of current research see, for example, Blinder, A., ”Making Monetary Policy by Committee”, an essay presented at the conference ”International Experience with the Conduct of Monetary Policy Un-der Inflation Targeting”, 22-23 July, 2008, Bank of Canada.
2) See, for example, Blinder, A. and J. Morgan, ”Are Two Heads Better than One? An Experimental Analysis of Group versus Individual Decision Making”, Journal of Money, Credit and Banking 37, 2005, 789-811.
