Development in Productivity in Denmark

Per Flink Iversen and Johanne Dinesen Riishøj, Economics

INTRODUCTION AND SUMMARY

Danish productivity, measured as gross value added per working hour, has decreased during the past year according to the available quarterly national accounts data. This masks a decline in output growth and a continued rapid increase in labour. Over the last decade Danish productivity has grown modestly both in relation to previously and compared to other countries.

Productivity growth has decreased despite Danish business enterprises' considerable investment in information technology and other capital stock, and a higher level of education of the labour force. The development is probably due to the fact that Denmark's ability to capture efficiency gains from other areas, such as improved procedures or more effective production processes, has been less pronounced compared to previously, and to other countries. This may be the result of more people, including marginal groups, gaining a foothold in the labour market, due to such factors as the labour-market reforms since the mid-1990s and the economic upswing of recent years. The composition of the business sector in Denmark, with a relatively small IT production sector, may also be part of the explanation. Moreover, the degree of research and development and the level of education of the labour force have been slightly lower in Denmark than in some of our key trading-partner countries.

Higher productivity makes it possible to produce more with the same input, and in the long term productivity growth is important for the underlying economic growth, real incomes and the standard of living. Productivity increases are also an important parameter in international competition. Productivity gains will become ever more important to Denmark in the years to come, when the labour supply will fall as a result of the demographic development.

This article outlines the development in Danish productivity in recent years and compares it with the development in other countries by way of harmonised data.

MEASUREMENT OF PRODUCTIVITY

The size of a country's output depends on the input of labour, capital and other production resources, including how efficiently they are applied. The efficiency is also called productivity.

There are several ways of measuring productivity. This article focuses on labour productivity, i.e. output in relation to labour input. Usually, labour productivity is calculated as output per employee or output per working hour (hourly productivity). If the average number of working hours per employee varies over time or from country to country, the two measures of labour productivity will be different. Hourly productivity is the productivity measure used below.

A key factor to be considered is whether the productivity measure includes the whole economy, or only the market-related part of the economy. The reason is that the productivity increase in the public sector per construction is zero, irrespective of any actual efficiency gains, cf. Box 1. Productivity growth in the public sector is important to e.g. fiscal sustainability, but as it is difficult to measure, this article only takes into consideration the development in the market economy.

The overall productivity pattern is best assessed over a number of years because productivity growth varies with the cyclical development. During a recession, a lower than normal productivity increase will typically be observed. This may be attributed to the tendency for business enterprises to retain valuable employees who can be of use when cyclical changes occur. Thus employment rarely decreases to the same extent as production, and productivity may even fall.

At the beginning of an upswing, there will typically be a higher than normal productivity increase due to low capacity utilisation as a result of the employees having been retained during the recession. Besides, business enterprises will be reluctant to take on new employees at the beginning of the upswing, as they are uncertain about its duration and force. As the upswing becomes more stable, business enterprises will employ more people, and productivity growth will drop to just below its normal level.

The flexibility in the labour market is a determining factor for cyclical fluctuations in productivity. High flexibility enables business enterprises to faster adjust the number of employees to the development in output. This tends to reduce the effect of cyclical fluctuations on productivity, cf. the article on flexicurity, p. 45 ff.

DEVELOPMENT IN PRODUCTIVITY IN DENMARK AND ABROAD

In the past decade, productivity growth in Denmark has been modest compared to previously, cf. Chart 1. From 1996 to 2006, annual growth in hourly productivity in the market economy was 1.0 per cent on average against 3.9 per cent in the previous 30 years. There are indications that the downward trend in productivity will be aggravated in 2007. From the 2nd quarter of 2006 to the 2nd quarter of 2007, hourly productivity in the private sector dropped approximately 2 per cent according to the available national accounts data. This masks a decrease in economic growth and a simultaneous rapid increase in the number of working hours. The latest figures are, however, preliminary and can be revised.

Growth in hourly productivity has also been weaker in Denmark than in our key trading-partner countries. While Denmark saw the second-largest average annual improvement in hourly productivity among the old EU member states in the period 1971-95, we came last but three in the following years, cf. Chart 2. Since the mid-1990s hourly productivity growth has also been slower in Denmark than in e.g. the USA, where output per working hour has increased by approximately 3 per cent p.a. for the period.

The weaker development in Danish hourly productivity has been broad-based. All main sectors, with the exception of the transport, postal and telecommunications sectors, have experienced lower hourly-productivity growth during the last decade compared with previous years, cf. Chart 3. Productivity growth was more than halved in most main sectors. In the business-service sector, output per working hour has even decreased since the mid-1990s, whereas it has been largely unaltered in the construction sector and in trade, hotels and restaurants, according to the available data.

FACTORS BEHIND THE WEAKER PRODUCTIVITY GROWTH IN DENMARK

Improvement in employment and the cyclical development
Employment has increased strongly in Denmark since the mid-1990s, a result of a number of labour-market reforms, among other factors. The increase in employment has also included people with limited labour-market experience. If average productivity of those who have entered the labour market has been lower than that of those who were already employed, the increase in employment may help explain the low growth in productivity.

The last year has seen an outright decrease in productivity in Denmark, while capacity pressures have been high. The current situation deviates from the typical trend for productivity growth during a business cycle. Normally, growth will be relatively high during an upswing and the reverse during a recession, cf. the curves in Chart 4 are usually closely correlated.

The current situation may reflect that output has not been able to keep up with demand, because business enterprises have been short of qualified labour and excess production capacity. While output has been limited due to capacity shortages, business enterprises have continued to take on more employees to catch up with the backlog. This may temporarily have curbed productivity growth. The same phenomenon was apparently also observed in the mid-1980s, when capacity pressures also were high.

The normal positive relation between the cyclical position and productivity implies that the cyclical position at the beginning and end of the period under review also influences productivity growth. Part of the explanation why for example Sweden and Finland have seen greater productivity gains than Denmark since 1995 is that those two countries experienced a recession at the beginning and middle of the 1990s. This probably meant that capacity utilisation was low from the outset, and that especially business enterprises with low productivity were forced to close down. The opportunities to boost productivity have thus been better in Sweden and Finland than in Denmark.

Research and development
Research and development is key to productivity growth in Denmark and the other industrialised countries, cf. the Danish Economic Council (2003), among others. The explanation is that investments in research and development enhance innovation and the implementation of new technologies, and makes more efficient production processes possible. In the period 1995-2005 Denmark invested 2.2 per cent of GDP in research and development according to Eurostat data. That is more than the euro area, which invested 1.8 per cent of GDP in the same period, but less than in e.g. Sweden and Germany.

Capital intensity, labour quality and total factor productivity
Productivity increases may be distributed on contributions to growth from capital intensity, labour quality and total factor productivity, TFP, cf. Box 2.

Capital intensity designates the amount of capital a worker has at his disposal. The capital intensity may grow by increasing both the amount of capital (more computers) and the quality of the capital stock (faster computers) per worker. Quality improvements are difficult to measure and make it difficult to assess how rapidly the capital stock should be written down. Moreover, there is a risk that some quality improvements may not be registered, which leads to incorrect assessment of the development in capital intensity.

Labour quality is closely connected to the level of education of the labour force. New technology can be introduced faster with a more highly educated labour force. A higher level of education may therefore contribute to an increase in productivity.

Growth in TFP illustrates the remainder of the increase in productivity. The factors behind this part of the productivity increase cannot be identified with any certainty, but they could be e.g. smarter production processes, more efficient work procedures and better organisation.

With data from EU KLEMS^[1], productivity growth may be decomposed in the above way, cf. Table 1. EU KLEMS is a statistical and analytical research project encompassing contributions to economic growth financed by the European Commission. Data covers the EU, the USA and Japan, and some of the variations in calculation methods mentioned in Box 1 have been taken into consideration. This has facilitated cross-border comparison.

Weak development in TFP
The low growth in hourly productivity in Denmark in recent years is mainly due to the fact that growth in TFP has ceased, cf. Table 1. Growth in Denmark's TFP has in recent years also been considerably lower than in our key trading partner countries. Shifts of labour and capital stock between sectors, e.g. from sectors with high productivity growth to sectors with low productivity growth, have an impact on TFP. Nevertheless, Denmark's lower TFP cannot be explained merely by sector shifts, as the growth rate has slowed in all main sectors, cf. Table 2.

A Harberger chart^[2] can provide a more detailed picture of how large a part of the economy and which sectors have made a positive contribution to TFP growth. This is shown in Chart 5 for Denmark. Here the business sectors' accumulated contributions to TFP growth from 1996 to 2005 are plotted against the business sectors' accumulated share of output. The sectors are organised in descending order according to TFP growth, so that the fastest growing sectors are closest to (0;0) and those with lower or downright negative growth are further to the right in the grouping. If all sectors had shown TFP growth, the slope of the entire curve would have been positive.

In Denmark, chemical production showed the largest percentage growth in TFP from 1996 to 2005. The sector made up about 2 per cent of output in 1995 and contributed to elevating total TFP by approximately 10 percentage points for the period. Consequently, chemical production has been placed in (0.02;0.1) in the Harberger chart. The chart shows that TFP increased in about half of the private sector in the period 1996-2005, cf. the slope of the curve is positive for the first approximately 50 per cent of output. In the remaining part of the private sector, TFP decreased. The decrease in TFP was strongest in the hotel and restaurant sector and in the business-service sector. These sectors are characterised by relatively high employment growth. On the other hand, more export-oriented sectors, such as the pharmaceutical industry and agriculture, have achieved higher efficiency gains. The same applies to IT-intensive sectors, such as the post and telecommunications and the financial sector. This phenomenon is also observed in our key trading-partner countries.

Large contribution from increased capital intensity
Growth in hourly productivity since 1995 is to a large extent due to the fact that Danish business enterprises have invested heavily in capital stock, entailing more capital per worker, cf. Table 1. Especially investments in IT and telecommunications technology have boosted productivity, and the contribution from increased IT capital intensity has been slightly larger in Denmark than in our key trading-partner countries. The electronics sector, the post and telecommunications sector and the utilities sector account for the highest capital intensity. In our key trading-partner countries, the contribution from IT-capital deepening has also been most significant in the electronics industry and the post and telecommunications sector, and contrary to Denmark, also pronounced in trade and transport.

Labour quality and education
The level of education of the Danish labour force has increased continually, and the quality of the labour force has thus improved. On average, the higher labour quality has contributed approximately 0.3 percentage points annually to productivity growth since 1980, cf. Table 1. This is on a par with the contribution in the countries with which Denmark normally compares itself.

The level of education of the labour force also affects TFP growth. A high level of education often means a more flexible labour force. Besides, there are positive side effects, as people who work together with highly-educated people also boost productivity. Finally, studies suggest that highly-educated people are more mobile, which enhances knowledge sharing in society. Data from EU KLEMS shows that the input of highly educated labour makes up a smaller part of the total labour input in Denmark than in our key trading-partner countries. However, the quality of education and training at the same levels may vary across countries. Moreover, in Denmark there is a deep-rooted tradition for further training of labour, which is not included in the EU KLEMS calculation.

Sound Danish economy despite weak productivity development
Although productivity growth has been low during the recent decade, the Danish economy has performed well. The unemployment rate has dropped markedly, the current-account surplus has been sound for a number of years, and corporate earnings have increased, among other things as a result of improved terms of trade, cf. Beier and Pedersen (2005).

Higher productivity growth in the future would benefit the Danish economy. It will contribute to Danish business enterprises remaining competitive. Moreover, it may alleviate some of the problems related to the future demographic development, with fewer people actively employed and more elderly people.

LITERATURE

Ahmad, N., F. Lequiller, P. Marianna, D. Pilat, P. Schreyer and A. Wölfl (2003), Comparing growth in GDP and labour productivity: Measurement issues, OECD Statistics Brief, No. 7, December.

Beier, Niels C. and Erik Haller Pedersen (2005), Wages, Competitiveness and the Balance of Payments, Danmarks Nationalbank, Monetary Review, 1st Quarter.

The Danish Economic Council (2003), The Danish Economy, Spring.

Harberger, A.C. (1998), A Vision of the Growth Process, American Economic Review, March, Vol. 88, no.1.