Some comments om Professor Charles Bean's paper - Aggregate-Demand Policies and Labour Market Reform
Introduction
I would like to begin by thanking the Economic Council for the invitation to take part in this conference. Unemployment in Europe is not only high, but it has persisted over as long a period as a couple of decades. A discussion of what can be done to overcome this problem is therefore very relevant. This discussion is also important viewed from a central bank perspective. It is important to clarify what central banks can and cannot do to contribute to increased employment and reduced unemployment.
I would also like to thank Professor Charles Bean for an interesting and thought-provoking essay. The starting point of his paper is the need for structural reforms to combat unemployment. In other words, he views unemployment in Europe as primarily structural and not cyclical. Professor Bean discusses in detail what should be done if such reforms are to lead quickly to the desired result. He also takes up how the establishment of the EMU may affect the prospects for success.
I am grateful for valuable comments from the following members of staff at the Riksbank: Mikael Apel, Eva Uddén-Jondal, Staffan Viotti and Anders Vredin. Special thanks are due to Yngve Lindh who gave useful suggests and together with Magnus Jonsson and Jonny Nilsson contributed the result of model simulations that have served as the basis for this work.
In my comments, I intend to take up three aspects of the problem area identified by Professor Bean. I would first like to underline Professor Bean’s starting point, namely the need for structural reforms to be able to overcome the high level of unemployment. I then comment on the role of monetary policy in connection with the implementation of structural reforms. Finally, I take up the EMU perspective and the possibility of carrying out successful, structural reforms.
How can increased employment and reduced unemployment be achieved?
If we assume a simple model where goods and services are produced with the input of the factors of production, labour and capital, the level of employment can in principle be affected in two different ways.(1)
1. A growing economy where capital and labour increase at the same rate
2. Changed capital intensity with a given total production
If we examine the actual development since the first oil crisis in 1973/74, the diagram (2)shows that the differences in the development of employment between EU and the USA can be explained by the different development of the capital intensity in the production process rather than by large differences in growth. Capital intensity in the EU has increased. The increase in production has entailed, in relative terms, a reduction in the use of labour compared with capital.(3)
As shown in the diagram, labour productivity has increased considerably more in Europe than in the USA during the period 1974 to 1995. The productivity of labour measured ex post reflects not only the changes in an exogenously determined growth in labour productivity but also is to a great extent the result of high wage costs which favour capital intensive technology and force less efficient operations to stop production or shrink.
In the USA the inputs of capital and labour have increased to about the same extent. Employment has therefore also increased more rapidly without growth overall having been markedly higher than in the EU. It is close at hand to explain the increased capital intensity in the EU with a rise in the total price of labour in relation to the price of capital. The real wage prices (including wage-related taxes) have increased more quickly in the EU than in the USA. A further indication of this development is that the return on capital in the USA has been higher than in the EU. On the whole, it seems that the relative price of labour has been higher in Europe than in the USA, and that this can be an important explanation for the weaker growth of employment.
In Sweden, this trend seems to have been stronger than in the rest of the EU. This can have contributed to our country’s weak economic growth during recent decades. A more stable macro-economic regime-- of the type which was established in Sweden during the 1990s—is itself a structural reform which can lead to a growth rate which will be higher than that of the 1970s and 1980s.
Structural reforms – a path to increased employment
There are many indicators that the possibilities for overcoming employment problems in Europe are connected with the possibilities of achieving both an increased production potential and a more balanced use in future of the production resources, labour and capital.
Structural measures are of strategic importance for the labour market, including wage formation, to be able to function better and for employment to increase. This is also Professor Bean’s starting point in his paper. He points out that the flexibility of real wages needs to increase and that mobility on the labour market needs to be stimulated.
It is not clear to me that Professor Bean’s analysis considers the effects on the employment level of a more balanced use of labour and capital. The focus of the analysis, if I have understood it correctly, is rather on how an increase can be achieved in overall growth, and in this way also in employment (4). However, a reform of the operation of the labour market should also affect the capital intensity of production. Even if this mechanism is not explicitly contained in Bean’s paper, it may very well follow from the structural reforms he advocates. It is in this case something that can soften the apparent conflict between the short-term and long-term effects that Bean analyses. If growth has a somewhat different content, with a greater input of labour for each unit produced, the time perspective for the increase in employment can be affected in a favourable direction.
The role of monetary policy in carrying out structural reforms
Basically, Bean’s view of the macroeconomic dynamics after the implementation of structural reforms seems to be reasonable. Both due to the individuals’ uncertainty about the reform, and its effects and the fact that prices and wages adapt slowly, it may, as Bean points out, lead to a short period of increasing unemployment. Subsequently, however, demand in the economy increases due to the capital stock being built up, and production and income will after a time be higher than before the structural reform was carried out. Employment increases and unemployment falls.
I would like to point out two factors that mitigate or even eliminate the problem that Bean highlights. My first point illustrates the advantage of a monetary policy directed at explicit inflation targets, which partially takes care of the problem indicated by Bean. The second point deals with the importance of taking into consideration the extent to which the actors in the economy anticipate the structural reforms and their effects.
A reform of the labour market that leads to an increase in the supply of labour can, according to Bean, lead to a reduced use of resources in the short-term and thus exert a downward pressure on prices. This naturally affects monetary policy, which is focused on an explicit inflation target. An inflation rate that is lower than the target normally leads to the central bank reducing the instrumental rate. Total demand and employment consequently increase. The negative effects that may arise are thus softened within the framework of the current monetary policy approach. If the instrumental horizon for monetary policy is one to two years, the central bank will accordingly be able to take into consideration the effects of the structural reform in good time. This will be the case if the political situation is such that it is highly probable that the reform will really be carried out.
However, I would like to warn against carrying out a monetary policy that entails risks in the area of inflation (5). An expansive monetary policy approach can lead to inflationary expectations being adjusted upwards. This would in turn entail that the costs in terms of reduced employment to get back to the inflation target would exceed the short-term gains in the level of employment of such a policy. How high the costs of an expansive policy would be would depend on the how inflation expectations are formed. For a country which, for a long run of years, has carried out a policy that led to high inflation and quite recently come down to lower levels, the risks of such a policy are probably even higher.
Neither, in my view, is the type of package solution that Bean seems to recommend especially practicable. His idea is that a supply reform should be announced at the same as the central bank provides assurances on increasing demand by reduced instrumental interest rates. The whole idea of separating monetary policy decisions from other political decisions by increasing the independence of the central banks is to counteract such confusion. Most countries – not least Sweden – have poor experiences of such arrangements. It can lead to periods of overheating, inflation and unstable growth. If the effects of a package of this type with stimulating monetary policy are built into wage contracts, nothing will be achieved. It would rather lead to other types of costs.
The EMU perspective
A central mechanism behind the initial problems which structural reforms are said to entail depends on the implicit assumption that these reforms come as a surprise. This explains the fact that there will arise a delay between the effects on potential and actual production respectively. This is exactly why the need for demand policy arises in Bean’s opinion. In an EMU perspective, this will, in his view, be an even greater problem since monetary and foreign exchange policy are carried out together and the probability for co-ordinated reforms in all eleven EMU countries is low. Moreover, fiscal policy is restricted by the Stability and Growth Pact.
A basic issue is how probable it is that structural reforms will come unexpectedly. It does in fact play a large role for the development of the economy if the reforms are instead foreseen.(6) Then it is on the contrary reasonable to assume that the positive effects will appear more quickly. Labour market reforms are discussed and investigated from all appearances for a rather long time before being implemented. The whole process is probably long and drawn-out and employers, workers and unions together as well as other economic agents should at least be able to foresee in part the way things are moving. It is then not unreasonable to think that the positive effects of the reform will come earlier than would be the case if they came as a surprise. There will then be less need for demand stimulation than Bean anticipates.
The fact that a country is a member of the EMU should therefore neither prevent nor make it more difficult to carry out structural reforms, provided that the reforms are to a reasonable extent expected. In addition, a EMU country has the option of using fiscal policy to support the work of reform if it nevertheless should prove necessary. Bean’s starting point is that fiscal policy is strictly limited by the Stability and Growth Pact. It is worth emphasising that this is only the case if the countries permanently balance at the utmost limits of the Stability and Growth Pact. It is not a law of nature that this should be so. Swedish economic policy, for instance, aims at building up a surplus in public finances. If other countries adopt the same approach, there is a scope for action also for fiscal policy.
Summary
It is important to discuss what can be done to reduce the high level of unemployment in Sweden and in Europe as a whole. Many factors indicate that measures are required that affect wage formation and the working of the labour market as a whole in order for employment to increase permanently to a significant extent. Professor Bean advocates such reforms and discusses in an interesting paper how they are to be implemented in the best possible way.
In my comments, I have tried to take up a couple of aspects that show that the implementation of such reforms need not be linked with the type of difficulties that Bean supposes. This is especially the case since they partially operate through other mechanisms than those that Bean explicitly discusses. Structural reforms of the labour market do not lead to increased employment only by increased economic growth but can also change the way in which this growth is achieved. A more balanced mix of capital and labour is also a way to increase employment. Furthermore if structural reforms are more or less expected, which is probably the case, the positive effects will likely be felt relatively soon. Otherwise such a situation with initial negative effects, like the one that Bean visualises, can be dealt with within the framework of a monetary policy focused on an explicit inflation target. Member countries of the Monetary Union can use fiscal policy since countries need not necessarily stay at the at the limits of the Stability and Growth Pact. Sweden is an example of a country with a goal of achieving a surplus in public finances.
(1) The relationship between the rate of increase of production, factor inputs and productivity can be expressed as
Q = PF + TFP = w*L + (1-w)* K +TFP, i.e. the increase of production (Q) is equal to the sum of the increase of the production factors (PF) and total factor productivity (TFP). The increase of the production factors is in turn a weighted sum of the increase of labour (L) and capital (C) where the weights consist of the wage portion of the value-added (w). The growth in production can also be calculated as the sum of the growth rate of employment and the growth rate of labour productivity. The growth rate of labour productivity can be written as: Q – L = TFP + (1-w) * (K-L), i.e. the increase in labour productivity [the difference between the increase of production (Q) and of employment (L)] is the equal to the sum of the increase in total factor productivity (TFP) and the increase in capital intensity [the difference between the increase in the capital input (K) and employment (L)] multiplied by the capital portion of value-added (1-w).
(2) The figures for USA and EU in the diagram have been taken from European Economy, Annual Economic Report for 1997, No. 63, 1997. The figures for Sweden are based on statistics from Statistics Sweden and calculations made at the Riksbank. It should be emphasised that the estimates for Sweden are approximate since they are based on measurements of the capital stock which only cover the business sector. The period is 1974 to 1995. Development is stated as an annual average in per cent. Rounding-off errors may occur.
For figuers, see the Word-file
(3) An increasing capital intensity is not negative per se. On the contrary, it can drive up labour productivity, thus providing scope for high growth and a rising standard of living. In a situation with high unemployment and a weak growth of employment, it can, however, be a source of problems on the labour market.
(4) In the first paragraph on page 3 of his essay Bean writes: ”…the increase in equilibrium supply of output that follows from the structural reform”. This should reasonably be interpreted as meaning that the way to increased employment and reduced unemployment is through increased potential and actual production viewed as a whole – and not through a reduced capital intensity.
(5) This is something that Bean takes up on page 4 in the third paragraph of his paper when he writes that ”With a fixed money wage, this requires a temporary increase in domestic inflation, and a fortiori in consumer price inflation”. Bean refers in this context also to an earlier paper where a temporary increase of inflation is directly advocated. His analysis of this is relevant for today’s discussion since Bean in the third paragraph on page 6 explicitly refers to Sweden and the United Kingdom being outside the EMU and states that it should be possible to carry out a monetary policy that stimulates demand in these countries.
(6) In the simulations made at the Riksbank in the macro-model RIXMOD 2.0, it clearly emerges that the development of unemployment and the economy as a whole is different depending on whether the structural reform is totally unexpected or whether it is fully anticipated. In reality, it is difficult to see that a reform of this kind could come as a total surprise Neither does it seem probable that the reform would be fully expected. The probable effect is rather somewhere between these two extremes. My point is, however, that the short-term negative effects will be smaller, the less unexpected the implementation of the reform is. This insight is therefore something that could be used by the political system if it is thinking of carrying out a reform.
