Perspective on the Swedish economy
A central bank has the task of producing and distributing money and maintaining its value.
If the Riksbank's monetary stance is unduly expansionary, that is, if the repo rate is below the level that would be compatible with a balanced economic development, the economy is likely to become overheated. If production capacity is insufficient to meet the growing demand, the result will be price increases that constitute inflation.
If, on the other hand, the monetary stance is unduly tight—if the Riksbank sets the repo rate too high—the result may be decreased demand and deflation.
Inflation and deflation are equally undesirable.
This is the background to the decision by the Riksdag (Sweden's parliament) to make the Riksbank independently responsible for maintaining price stability. The Riksbank in turn has defined this objective as a 2 per cent rate of inflation, surrounded by a tolerance interval of ±1 percentage point.
Our monetary policy deliberations accordingly focus on economic tendencies and inflation prospects in Sweden, not on what other central banks are doing or refraining from doing with their instrumental rates. A reminder is called for in this respect.
When a low inflation regime has been established and people have come to expect low price increases in the longer run, the rate of inflation is determined by the relationship between total demand and production capacity. A monetary policy based on fulfilling an inflation target accordingly tends to smooth the economy's fluctuations around the long-term growth rate. In that these fluctuations alter the outlook for inflation and monetary policy targets inflation, the instrumental rate will be adjusted so that demand is brought back into line with the long-term path. This contributes to growth that is more sustained and stable. However, one should not expect too much from stabilisation policy's potential for economic fine-tuning.
In that high and fluctuating inflation adds to the element of uncertainty in economic decisions and thereby tends to impair long-term growth, it is reasonable to suppose that potential growth is favoured, if anything, by a rate of inflation that is low and stable. It seems that monetary policy is capable of influencing the growth trajectory indirectly.
But monetary policy does not have the means to affect long-term growth directly. Potential output is determined instead by changes in the population and the capital stock, together with developments in technology and the efficiency with which the productive resources are utilised. The Riksbank is not in a position to influence any of these factors directly.
Still, the Riksbank does need an approximate picture of how growth is actually developing in order to bring demand, which monetary policy is able to influence, into line with this. As I just mentioned, deviations from trend growth are an important cause of inflation and deflation. A major problem in this context, however, is the lack of a direct measure of potential output. Monetary policy therefore has to be conducted with stepwise adjustments of the repo rate in the light of various indicators. It can then be instructive to pause from time to time and take a closer look at growth trends in the longer run. That is what I shall be doing this evening and my talk is accordingly headed: Perspective on the Swedish economy.
Surprisingly low inflation and rising productivity in the 1990s
Since the end of the 1980s virtually the whole of Sweden's political system has been engaged in a deliberate effort to achieve a low rate of inflation. Annual inflation had averaged 8 per cent in the 1970s and '80s. In the 1990s the average annual rate has been around 3 per cent and the average since 1993 has been about 2 per cent.
At the same time, practically all forecasters, including the Riksbank, have found that models based on historical relationships have almost consistently overestimated inflation's actual outcome.
A major explanation for this—as well as for the successful combination of comparatively high growth with persistently low inflation in the Swedish economy in recent years—is of course the capacity surplus that existed, and still exists, after the recession in the early 1990s.
But the low rate of inflation is not exclusively a consequence of surplus capacity. It also has to do with economic changes of a more structural nature. Not the least of these is the presumed effect on price and wage formation from the announcement of the explicit inflation target in January 1993. Economic agents now expect a low rate of price increases and this influences their behaviour. Other structural measures of economic policy have also been taken since the mid 1980s.
There are also many signs of an increase in international competitive pressure, for example as a result of growing opportunities of locating even comparatively advanced production processes to countries where wages are considerably lower than in the Western world. Although this amounts, strictly speaking, to a series of one-off effects, the increased competition can still affect global inflationary pressure over a long period. Moreover, growing international trade, European economic and monetary union and various deregulations are all working in a similar direction.
The low level of inflation and the likelihood of it remaining low are evident from the Riksbank's latest Inflation Report. But let us take a closer look at the real economy to see whether developments in the 1990s represent a departure from the long-term trend. Labour productivity growth, for example, averaged approximately 1.4 per cent a year from 1974 to 1993; while the annual average since 1994 has been about 1.9 per cent. The economy has admittedly been in an upward phase since the summer of 1993, so it is hardly surprising that productivity growth has been above the trend. But the rate has also accelerated compared with the upward phase in the 1980s; productivity growth now seems to be approximately half of a percentage point higher than before. This is something that neither we at the Riksbank nor many other observers had anticipated.
An increase of half a percentage point may not seem much; over twenty years, however, it adds up to a difference of ten per cent in the level of production at the end of the period. It is actually equivalent to a considerable part of Sweden's decline in the international growth league since the beginning of the 1970s.
So do the lower inflation and stronger labour productivity in Sweden in the 1990s provide any clues to our economy's growth trend? Can it be said that in future the Swedish economy will be capable of sustaining higher growth? If so, it would contrast sharply with the 1970s and '80s and imply an upward shift in the potential growth rate from between 1.5 and 2 per cent to between 2 and 2.5 per cent, which would be more in line with other industrialised countries. The answer is no; it is still too early to draw any definite conclusions. But the situation does warrant a couple of reflections that point in a positive direction.
Growth in the post-war period
Growth in the 1950s and '60s was both high and stable. Historically, the '60s in particular was an exceptional decade for economic growth: the annual GDP growth rate averaged over 4.5 per cent; in the 1950s the rate had averaged 3.5 per cent. After that, from 1970 to the begining of the 1990s, growth became markedly slower.
Various explanations of this retardation have been offered, from demand shocks due to oil price hikes in the 1970s and the collapse of the inflated economy in the early 1990s to an emphasis on more fundamental factors such as systemic disorders or a lack of adaptation and adjustment to changes in competitive conditions. The explanations partly overlap. Each of the factors no doubt contributed to the marked retardation of growth from 1970 onwards. Over the years, the conceivable causes of the trend break have also been a subject for numerous analyses.
One explanation I want to consider today is that production in Sweden's corporate sector gradually acquired a composition that entailed a loss of competitiveness in world markets. From the leading position which Swedish business had established after the end of World War Two, ground was gradually lost.
As economic development in the industrialised countries picked up in the post-war period and competition became stronger, partly because trade barriers were lowered, the new situation called for changes in the direction of Swedish output. During much of the period, moreover, this was accompanied by rapidly rising wage costs. Business profitability in Sweden was accordingly under pressure from two fronts: growing competitive pressure from the rest of the world and rising wage costs.
In order to counter the pressure on profits, companies tried to cut production costs by means of rationalisation rather than investment in new products. Things still seemed to be going well on the whole at first. As I mentioned, the 1960s was a fantastic decade for growth. But the strategy proved to be defensive and the 1970s saw a period of stagnation. There are limits to rationalisation as a strategy for growth. Without renewal, growth normally slackens sooner or later.
The economic policy response in the 1970s was an attempt to underpin growth with a series of devaluations. Although the most recent devaluation was always described as the last, the cycle of rising wage costs followed by a devaluation recurred again and again. All this tended to conserve the corporate sector's existing structure, pull inflation expectations up and gradually create a climate for speculative investment. There were times when real estate and equity were more advantageous objects for investment than new products and production technology.
Growth and employment were also sustained by a rapid expansion of the public sector but this strategy also has its limits. In the longer run, an economic sector financed from taxes cannot grow faster than the sector on which the tax payments are based. At some point it becomes difficult to nurture the vitality that is essential for a market-based economy.
The endeavours to compensate for the corporate sector's lack of adaptability, which stemmed from the 1960s, simply did not work. The problems tended to be exacerbated, if anything, by the devaluations. Wage formation no longer had any clear rules to follow. Public sector growth called for levels of taxation that the tax system had difficulty in delivering and also contributed to higher inflation. The high level of public spending made government finances sensitive to cyclical fluctuations in economic activity.
The elimination of the most labour-intensive companies, combined with investment in capital-intensive technology and moderate growth of the private business sector, meant that private sector employment declined from the mid 1960s onwards. The absence of investment in new products and the increasing dependence on a gradually out-moded structure of production contributed to the imposition of a general adjustment in material standards. For many years, growth in Sweden was weaker than elsewhere. The standstill in business sector employment was not felt, however, until public sector employment also stopped rising in the early 1990s. Unemployment then rose to levels that Sweden had not experienced since the 1930s.
New tendencies in the 1990s
Something now seems to have happened in the past five years, though the period is still too short to justify any far-reaching conclusions. But there are some indications of sufficient interest to warrant their inclusion in the ongoing economic analysis and debate.
Since the upturn following the crisis in the early 1990s, annual GDP growth has averaged about 3 per cent. Compared with the recovery after the 1982 devaluation, this figure is not remarkably high. The point worth noting is that growth has been generated exclusively by an expansion of the corporate sector, above all in mining and manufacturing. The level of industrial output has risen around 45 per cent, which implies an annual rate of about 8 per cent. For Swedish manufacturing, this has been the most expansive five-year period in more than three decades.
It might be supposed that the industrial upswing has mainly been a recovery from the deep recession in the early 1990s and simply involved the activation of capacity reserves. In practice, however, it has been accompanied by a notable growth of investment. It is, in fact, difficult to find another five-year period since the 1950s with such strong investment growth in annual percentage terms.
The increased investment has mostly concerned machinery, though some growth has also occurred for investment in buildings and plant. By itself, this could indicate a main tendency to go on rationalising labour in order to reduce wage costs, a phenomenon we can call defensive investment. Other statistics, however, suggest that this period has not seen an increase in capital intensity, which might indicate that the investments have been more offensive. Unchanged capital intensity implies that the inputs of capital and labour are changing at much the same rate. If that is the case, it is the first time this has happened since the 1960s.
Another indication that the investments have been more offensive is the appreciable increase in the relative size of the sector exposed to international competition, a development that has been prescribed for the Swedish economy in the past twenty-five years of economic policy discussions. In the past five years the GDP share for the manufacturing sector has grown from under 20 per cent to over 23 per cent. Confirmation of this comes from the sizeable surpluses that are now being recorded on the current account, in contrast to what we have been accustomed to for several decades. Meanwhile, the public sector's GDP share has decreased from 29 to 27 per cent.
Turning now to the composition of Swedish exports of goods, trade in vehicles, iron, steel and forest products was notably stable for many years. It can be said that Sweden was dependent on basic industries. Since the late 1980s, however, a new pattern has been emerging; new industries are expanding and older activities are becoming relatively less important. The contribution to foreign trade from medical and tele-technical products, for instance, has doubled in the past ten years. This suggests that a renewal of Sweden's corporate sector may have been on the way already in the 1980s.
Moreover, the falling trend for corporate profits relative to the value of output since the early 1950s now seems to have stopped. Since the end of the 1970s there has been an upward trend, not dramatic but distinct. The improved profitability is probably an important explanation for the stronger growth of the capital stock, after a virtual standstill for more than ten years between the mid 1970s and the mid '80s.
Taken together, all this may be an indication that the historical growth trend could be exceeded. This might be the case in particular in that wage formation has improved in recent years and seems to be adjusting by degrees to the new low inflation regime, while inflation has been lower than predicted from historical relationships.
What is the explanation?
Some might argue that the development I have just outlined is simply a result of the krona's sizeable depreciation in connection with the move to a flexible exchange rate in the autumn of 1992. The familiar pattern from earlier devaluations is repeating itself. The economy is stimulated for a time and at first it may look as though a new path has been initiated. Setbacks occur before long, however, as a new period of price and wage increases takes over and leads to renewed problems. There are contributors to the debate who argue that this is going to happen again.
The fact remains, however, that recent developments cannot be explained solely in terms of the improved level of costs that followed the krona's fall in the autumn of 1982. The positive features of economic development appear to be more far-reaching than those associated with earlier write-downs of the krona. As I mentioned before, moreover, the underlying pattern on this occasion also looks different.
It took almost a decade, from the mid 1960s to the mid '70s, for the more subdued economic growth trend in Sweden to show up in earnest. By the same token, it can be many years before effects of the measures taken since the mid 1980s can be expected to materialise. The rising profit share and a more stable macroeconomic regime are elements of this picture. It takes a long time to build up a capital stock as the base for new products that can hold their own in the increasingly strong international competition.
Conclusion
Today I have briefly considered the Swedish economy and the discernible signs of a trend that is more favourable than before. Clearly, observations from just a few years are not a sufficient foundation for far-reaching conclusions. Most of all, confident statements that we are on the threshold of a new type of economy should be avoided. Caution is always necessary when attempting to interpret new long-term trends in the light of just a few observations.
At the same time, there is no denying that the Swedish economy now shows signs of a better long-term growth potential compared with the 1970s and '80s. A number of what I will call infrastructural issues have already been resolved. The allocation of responsibilities for Sweden's economy is now clearer, with fiscal policy focused on sound government finances and monetary policy on low inflation, which makes it the primary instrument for economic stabilisation. Other measures have been taken; important markets have been deregulated, accompanied by major changes in public sector activities in particular. The statistics indicate, moreover, that after weak growth for a couple of decades, conditions may now be in place for a trend break in Sweden.
Still, much remains to be done. The high level of unemployment calls for further measures to improve the workings of the labour market and wage formation. It is difficult to see how a cyclical upswing can suffice by itself to bring unemployment down to the levels that are being discussed in the political system, though no economist or central bank is in a position to pontificate about where unemployment's cyclical component ends and the structural component begins.
Everything points, moreover, to an increasingly tough competitive climate in the global economy. The growing possibility of locating even comparatively advanced production processes to countries where wages a considerably lower than in the Western world will accentuate the need to enhance human capital in the traditional industrialised countries. But the international mobility of human capital is extremely high. Old industrialised countries that aim to maintain and go on improving their inhabitants' material standard therefore presumably face a hard task. The issues here include the functioning of the educational system as well as the degree and construction of taxation.
