In the last thirty years, the share of manufacturing in total world GDP has decreased. Simultaneously, there has been a fall in relative and absolute manufacturing employment in the Western world. This structural change is defined as deindustrialization and started in the US at the beginning of the 1960’s. Since then, almost all rich countries have developed in the same direction, although at different speeds and to different extents.
Despite deindustrialization being a well-known phenomenon, knowledge about its causes is relatively limited. One explanation put forward is based on the composition of demand.
Daniel Bell has established the theory of the postindustrial society. This theory predicts as GDP per capita grows, there will be a decrease in household demand for manufactured goods in favour of a larger share of services. The stepping stone is the hierarchy of needs that the human nature is assumed to have. When our basic material needs have been satisfied, an increasing share of our consumption is directed towards services that are assumed to increase our quality of life. As a consequence, there will be a decrease in manufacturing employment – in absolute and relative terms – and in the sector’s share of nominal GDP.
William Baumol challenges this demand oriented explanation. He assumes that the demand for manufactured goods is independent of income. This means that the share of services in GDP is constant over time. From his point of view, deindustrialization is the result of differences in productivity growth. When productivity growth is higher in manufacturing than in the service sector, the share of service employment will increase as countries become richer. If wages in the service sector grow in line with the average wage growth, then the share of manufacturing in nominal GDP will decrease over time. Deindustrialization can thus occur despite relative demand for manufactured goods remaining unchanged over time.
Another supply oriented explanation is that firms increasingly buy the intermediate goods that were previously produced in-house. Due to technological change and globalization, firms have greater possibilities to increase their degree of specialization and separate parts of the production process. Not the least, for many manufacturing firms does this apply for services in general and business services in particular. Another aspect that points in the same direction is that many manufacturing firms today include an increasing number of services in their work at developing, producing, selling and marketing their products. A manufactured good is to a much smaller degree a “product” and increasingly a carrier of “services” that create added value. This means that deindustrialization in this sense is about statistical reallocations of businesses and the fact that more services are required to deliver a manufactured good.
In a forthcoming article in the journal International Productivity Monitor, the three possible explanations behind the deindustrialization are critically analyzed from a Swedish perspective. Since globalisation and technology developments are forces that truly affect the whole Western world, some general conclusions should be valid.
There has been a decrease in manufacturing employment in Sweden in both relative and absolute terms in the post-war period and there has been a gradual decrease in its share of nominal GDP. Yet, at the same time, manufacturing production has grown faster than GDP and there has been no decrease in domestic demand for manufactured goods over time.
However, the high productivity growth in manufacturing has lead to a fall in the relative prices of manufactured goods. This has not only resulted in a fall in relative employment, but there has also been an increase in the share of real GDP since 1980. In terms of number of units produced, we have, since at least 1980, been in a period of reindustrialization; the falling share of nominal GDP is entirely explained by the change in relative prices, not by a decrease in the demand for manufactured goods.
Despite this, since 1975 there has been a decrease in the number of employed to satisfy demand for manufactured goods. Though the loss appears less pronounced in official statistics. The explanation for this is the increased indirect manufacturing employment caused by a deeper interaction with the rest of the economy. This intensified interaction is particularly pronounced in relation to the service sector in general and its knowledge intensive part in particular.
This aspect has been of particular importance since the crisis years at the beginning of the 1990’s. The number of employed who were in some way involved in the production of manufactured goods did only decrease by slightly more than 9 000 between 1995 and 2005 and the share of total employment did only decrease from 26 to 24 percent. In the same period, production growth was twice as high as in the service sector, productivity growth was on average about seven percent per year and the share of real GDP increased by 70 percent.
The period since the 1990’s has been the most manufacturing intensive since the 1960’s. Considering that the number of employed who were in any way involved in the production of manufactured goods and that the share of nominal GDP did after all decrease somewhat, some people would call this positive deindustrialization, but in terms of production, productivity and the share of real GDP, others would claim that this is rather a case of reindustrialization.
Notwithstanding which interpretation seems most reasonable, the picture is considerably more complicated than what is too often being expressed in the public debate. Deindustrialization is both a myth and a reality.
A well functioning and a highly skilled service sector is an important condition for the competitiveness of the manufacturing sector and a competitive manufacturing sector is an important condition for future growth in the knowledge intensive service sector. This sheer insight is a decisive ingredient in a progressive growth policy for Europe in the years to come.