deindustrialisation uk case study

• 0 Shopping Cart

De-industrialisation in the UK

How have traditional industries declined in the UK?

De-industrialisation is the reduction of industrial activity or capacity in a region or economy, especially of heavy industry or manufacturing industry. De-industrialisation is one of the most significant economic processes to occur in the UK. De-industrialisation in the UK has involved the decline of heavy industries such as coal mining , shipbuilding and steel manufacturing.

During the twentieth century, the UK went from over 3000 coal mines to just 30. The last working deep coal mine in the UK closed in December 2015 . The graph below shows the rapid decline in employment in coal mining in the UK due to mechanisation, increasing costs of extraction and growing availability of cheap imports.

The graph below shows the rapid decline of coal production in the UK and the growth, then the rapid decline of imports as our reliance on coal has dropped.

Further reading: The death of UK coal in five charts

What impact has de-industrialisation had on North East England?

North East England was one of the first industrialised regions in the UK. Tens of thousands of people were employed in heavy industry including coal mining and shipbuilding. However, it was also one of the first regions to be affected by de-industrialisation with the closure of coal mines and shipyards.

De-industrialisation also led to a negative multiplier effect . Many smaller businesses that supplied and supported heavy industries closed, a knock-on effect affecting thousands of people.

North East England has suffered huge job losses and a rise in unemployment as factories and industrial sites closed. Many of those employed in heavy industries struggled to find new jobs with the skills they have.

North East England Region Profile 2013

The closure of the Easington Colliery has devastated the town of Easington when it closed in 1993. Over one thousand men were made unemployed by the closure of the pit. Unemployment in the area is still high and many people are on low incomes. In 2018 ITV reported the communities worst fears realised 25 years after the closure of the last deep pit on Durham coalfield .

How has the government responded to de-industrialisation?

Successive UK governments have tried a range of strategies to re-energise economic opportunities in North East England, including:

• investing in new infrastructure such as roads and industrial parks
• encouraging foreign investment e.g. Nissan opened a car plant near Sunderland in 1986 which now employs 7000 people
• setting up a regional development agency in 1999, which was replaced by a local enterprise partnership in 2012 which supports businesses, plans for economic growth and provides training

Premium Resources

Please support internet geography.

If you've found the resources on this page useful please consider making a secure donation via PayPal to support the development of the site. The site is self-funded and your support is really appreciated.

Related Topics

Use the images below to explore related GeoTopics.

Economic change in the UK

Topic home, the uk’s post-industrial economy, share this:.

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)
• Click to share on Pinterest (Opens in new window)
• Click to email a link to a friend (Opens in new window)
• Click to share on WhatsApp (Opens in new window)
• Click to print (Opens in new window)

If you've found the resources on this site useful please consider making a secure donation via PayPal to support the development of the site. The site is self-funded and your support is really appreciated.

Search Internet Geography

Top posts and pages.

Latest Blog Entries

Pin It on Pinterest

• Click to share
• Print Friendly

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Front Sociol

The Long Shadow of Job Loss: Britain's Older Industrial Towns in the 21st Century

Associated data.

The datasets generated for this study are available on request to the corresponding author.

This article takes a long view of economic change in Britain's older industrial towns, drawing on the authors' accumulated research into labor market trends in the places and communities most affected by deindustrialization. It begins by documenting the industrial job losses over the last 50 years and their impact on unemployment, economic inactivity and welfare benefit claims, highlighting the diversion onto incapacity benefits triggered by job loss that remains a major feature of the towns. It then looks at the evidence on the present-day labor market in the towns, identifying job growth at a slower pace than in the cities and continuing weaknesses in terms of earnings, qualifications and occupational mix. These are the on-going problems the authors describe as the ‘long shadow of job loss’. The evidence also shows that despite years of job loss, industry remains a key component of the towns' economy and that the towns are increasingly connected to surrounding areas, including nearby cities, by strong commuting flows.

Aim of the Paper

A widely-held view of the UK economy is that it has become ‘deindustrialized.’ The world's first industrial nation now employs far fewer workers in manufacturing and mining than was the case fifty or more years ago and the economy as a whole is now dominated by jobs and output in the service sector. That this deindustrialization has happened in the UK is indisputable but the massive consequences for the individuals and communities that once depended upon industry for their livelihood are only poorly understood.

This is the gap in knowledge that the present paper helps to fill by bringing to bear statistical evidence on the contemporary labor market in the places hardest hit by deindustrialization. In doing so, the paper is also intended to provide a context for the others in this volume.

Specifically, the paper summarizes the evidence from the authors' own research, now extending over three decades, into employment, unemployment and welfare benefits across the UK and adds a quantitative overview of the contemporary labor market in Britain's older industrial towns, drawing on official statistics and the authors' most recent research. Job losses, labor market adjustment and the contemporary labor market in older industrial towns have been covered in an earlier report (Beatty and Fothergill, 2018 ). Here the empirical evidence has been comprehensively up-dated, mostly from 2016 to 2019, and the conclusions revised accordingly.

Because so many of Britain's industrial job losses happened a generation or more ago and because so many of the redundant workers have themselves reached pension age or died it has become easy to assume that the problems arising from deindustrialization have passed into history. Indeed, as the second decade of the twenty first century drew to a close official figures pointed to unemployment levels in the UK that were lower than at any time since the mid-1970s, when large-scale industrial job losses were only just beginning to get underway. If official unemployment data is the preferred guide, then the problems arising from deindustrialization do indeed appear to have been overcome.

There are however more dimensions to labor market disadvantage than just unemployment, and national figures almost always hide important differences between places. Britain's older industrial towns—the smaller places beyond the big cities that generally have an industrial history extending far back into the nineteenth century—are where so much of British industry was once concentrated. These are the places where the loss of industrial jobs is likely to have been most keenly felt and where we might expect to observe lasting impacts. By contrast, although Britain's cities too nearly all have an industrial past they have always played a wider role in regional and local economies as service centers for their hinterlands, administrative headquarters, transport hubs and as home to major universities.

The recession provoked by the 2008 financial crisis was in many ways a wake-up call, prompting a rediscovery of the divergent labor market trends between North and South (Gardiner et al., 2013 ) and between city regions (Townsend and Champion, 2014 ; Centre for Cities, 2015 ; Swinney and Thomas, 2015 ; Martin et al., 2016 ; Pike et al., 2016 ). In particular, there has been a growing realization that Britain's older industrial towns may not after all be well on the way to recovery. In the 2016 referendum on EU membership, for example, older industrial towns in England and Wales generally voted ‘leave’ by a margin of two-to-one (Jennings, 2017 ). This has been widely interpreted as a reflection of rising disaffection and disenchantment with the economic impacts of globalization and deindustrialization. The big cities, by contrast, mostly voted ‘remain.’

As the realization of on-going problems has grown, the term ‘left behind places’ has gained widespread use in British political debate and it has been applied in particular to older industrial towns. A sign of shifting priorities has been the UK government's announcement of Town Deals, which make available additional funding for economic and social development (Ministry of Housing, Communities and Local Government, 2019 ). Of the first 100 towns across England invited to submit bids to the new fund, rather more than half could be described as ‘older industrial.’

Britain is not unique of course in having towns that have lost most or all of their original industrial base. Deindustrialization also characterizes parts of North East France, the former East Germany and the ‘rustbelt’ of the United States, to mention just three other examples. In Britain, however, the original industrialization happened earlier and the efforts to rebuild economies have mostly been in place longer. Reflecting the research on which it is based, the evidence in this paper focusses exclusively on Britain but there may well be pointers to the experience in other developed, post-industrial economies.

Structure of the Paper

The next section of the paper summarizes the industrial job losses that have taken place in recent decades in Britain, and their distinctive geography. The following section looks at how the labor market adjusted to these job losses, highlighting the extent to which the impacts have been on labor force participation rather than recorded unemployment.

The core of the paper then looks at the contemporary labor market in Britain's older industrial towns, presenting a range of new evidence covering employment, unemployment and economic inactivity, industry mix, job quality, skills, pay and welfare benefits. This is preceded by a short section on a working definition of the towns, necessary in order to deploy a range of official statistics.

The final part of the paper considers the labor market links between older industrial towns and neighboring cities, before concluding with an assessment of the position that Britain's older industrial towns occupy in the economy of the early twenty first century.

The Destruction of Industrial Britain

Employment in British manufacturing peaked in 1966 when 8.9 million people worked in this sector, accounting for 30% of all employment. By 2019, the UK government's Office for National Statistics put the numbers employed in manufacturing at just 2.7 million, or 7.7% of the employed workforce.

UK manufacturing employment fell especially steeply in the early 1980s during a recession triggered by a high exchange rate and high interest rates. The job losses during these years were documented in particular by Townsend ( 1983 ) and Fothergill and Guy ( 1991 ) and the process of deindustrialization more generally by Martin and Rowthorn ( 1986 ). The recession of the early 1990s added further major job losses (Gudgin, 1995 ). Thereafter, manufacturing output largely stagnated and manufacturing employment continued to slide even though the UK economy as a whole enjoyed 15 years of sustained economic growth. The recession triggered by the 2008 financial crisis reduced UK manufacturing employment still further and during the subsequent economic recovery manufacturing employment did no more than stabilize at a new lower level.

The coal industry's job losses go back much further. When UK coal production peaked in 1913, 1.1 million miners were employed in over 3,000 mines. For much of the rest of the twentieth century employment in the UK coal industry fell, though there were still 450,000 miners working in the mid-1960s. The more recent job losses began in earnest after the year-long miners' strike of 1984/5, which failed in its attempt to stop pit closures, and the final colliery closed in 2015.

The shift from industrial to service sector employment is not unique to the UK. In all advanced economies it has its roots in differential rates of growth in labor productivity—it is generally easier to replace people by machines in manufacturing than in most service activities—and the shift away from industry has been accentuated by globalization, which has resulted in much routine production moving to China and other emerging economies. In the UK, however, the process of deindustrialization has probably gone further and faster than elsewhere.

The UK's industrial job losses have been concentrated in specific parts of the country. Partly this reflects the distribution of manufacturing, which was always more important in some places than others, but partly it reflects the location of the industries such as coal, steel, shipbuilding, heavy engineering and textiles and clothing that experienced the biggest reduction in employment. Figure 1 illustrates the geography of this job loss. This map flags up the most significant job losses, in places where major industries have been reduced to a fraction of their former size or disappeared entirely. The closure of individual large plants accounts for some of these job losses but more often, in the case of the coal and textile industries for example, the job losses affected several sites across neighboring towns.

Major industrial job losses across Britain since the early 1980s. Source: Beatty and Fothergill ( 2017 ).

Manufacturing employment has fallen in just about all parts of the UK but the concentration of industrial job losses north of a line from the Severn estuary to the Wash is especially noticeable. It is the cities, towns and coalfield areas of the Midlands, North, Scotland and Wales that have been hit hardest—a pattern that will be familiar to anyone with a basic knowledge of Britain's economic geography.

Labor Market Adjustment

The first and most obvious consequence of the large-scale loss of industrial jobs was a rise in the number claiming unemployment benefits. This hovered around 3 million for several years in the 1980s but fell away after the recession of the early 1990s, eventually to less than 1 million for most of the 2000s.

By the end of the 1990s, however, it was clear that the job shortfalls across Britain were highly uneven and that they were no longer accurately reflected by unemployment data (MacKay, 1999 ; Webster, 2000 ; Erdem and Glyn, 2001 ). The present authors' research on the coalfields shed important new light. In this part of Britain, the mines had mostly closed but claimant unemployment was no higher than when the mines had been working. The evidence on the coalfields showed that the main consequence of job loss was in fact a diversion of working age men into ‘economic inactivity’ and in particular into what the Census called ‘permanent sickness’—in practice a withdrawal from the labor market onto incapacity benefits (Beatty and Fothergill, 1996 ).

A decade later, a follow-up study (Beatty et al., 2007 ) identified growing job creation in the former coalfields but confirmed the observation that the principal labor market adjustment in response to the loss of mining jobs was an increase in economic inactivity among working age men. By this stage many of the ex-miners had themselves reached state pension age so it was clear that the continuing high level of economic inactivity among men must be spread more widely across the local workforce. Indeed, it appeared that job loss for one generation was being passed on as higher economic inactivity among the next.

The labor market adjustments in the coalfields were not unique. Across the whole of older industrial Britain, from the mid-1980s through to the early 2000s, the numbers out of the labor market—‘economically inactive’—on incapacity benefits surged to unprecedented highs. This triggered the argument that much of the increase was a form of ‘hidden unemployment’—men and women who in a fully employed economy might have been expected to be in work but whose health problems or disabilities entitled them to incapacity benefits instead of unemployment benefits (Beatty and Fothergill, 2005 ).

The increase in the numbers claiming incapacity benefits occurred among women as well as men. At first this seemed hard to understand because many of the older industries shedding jobs—coal and steel for example—had a predominantly male workforce. The explanation turned out to be that in older industrial areas and elsewhere the male and female sides of the labor market interact, so the competition for jobs transmits a shortfall in opportunities for men into a difficult labor market for women in the same places. When women with health problems or disabilities are out-of-work they generally then claim incapacity benefits in the same way as their male counterparts (Beatty et al., 2009 ; Beatty, 2016 ).

Figure 2 shows the numbers claiming the three main out-of-work benefits across Britain as a whole between 1979 and 2019. By the end of this period, and before the recession triggered by the coronavirus crisis, the numbers claiming unemployment-related benefits were well down on the levels of the 1980s and early 1990s, though there was an upturn in the late 2010s resulting from the introduction of the UK's new all-encompassing benefit, Universal Credit, which counts partners and some on small hours or very low pay. The numbers claiming lone parent benefits peaked in the mid-1990s, when the evidence pointed to job loss among men as a driving factor (Rowthorn and Webster, 2008 ), but have fallen back as eligibility has been restricted just to those with the very youngest children and, more recently, as Universal Credit has been phased in.

Working age benefit claimants, GB, 1979–2019. Source: Department for Work and Pensions.

The striking feature of Figure 2 is the rise in the numbers out-of-work on incapacity-related benefits, these days Employment and Support Allowance or, as the changeover takes place, Universal Credit on the grounds of ill health or disability. The numbers on these benefits rose from around 750,000 to a plateau of more than 2.5 million in the early 2000s and have subsequently only fallen to around 2.25 million. It is impossible to explain the large increase in health terms alone at a time when general standards of health and physical well-being have slowly been improving.

The highest incapacity claimant rates are predominantly in older industrial Britain. Exposure to industrial injury and disease means that older industrial Britain has long had higher levels of incapacitating ill health but the surge in incapacity claimant numbers only occurred after the industrial jobs began to disappear. Ill health or disability is not always an absolute bar to employment, so what seems to have happened is that where there are plenty of jobs the men and women with health problems or disabilities have been able to hang on in employment or find new work if they are made redundant. However, where the labor market is more difficult, as in much of older industrial Britain, ill health or disability has ruined many people's chances of finding and keeping work.

In effect, the job loss in older industrial Britain has led to higher numbers out-of-work on benefits but not in ways that were perhaps expected. It is the numbers on incapacity benefits, not unemployment benefits, that in the long-run have been impacted most.

Although the rise in incapacity claimant numbers is the defining feature of labor market adjustment in much of older industrial Britain the phenomenon is now past its peak. The national reduction in incapacity numbers since the early 2000s has occurred particularly in older industrial areas, where the claimant rate has typically fallen from above 10% of all adults of working age to a new average of 7–8%, though still well above the 3–4% in the more prosperous parts of southern England. The most recent estimate of ‘hidden unemployment’ among incapacity-related claimants, for 2017, puts the figure at 760,000, well down on an estimated peak of 1.15 million in 2002 (Beatty et al., 2017 ).

That incapacity numbers in older industrial Britain have fallen back from peak levels owes something to job growth in these places. In the former coalfields, for example, the number of employee jobs increased by 138,000 between 2012 and 2017 (Beatty et al., 2019 ). The wider growth in UK employment since 2010 has also provided commuting opportunities, particularly in nearby cities. Falling incapacity numbers probably also owe something to revised medical assessments, more restrictive entitlements and new conditionality for some, though the increase in women's state pension age has pushed in the opposite direction, bringing an additional cohort of over-60s into the scope of incapacity-related benefits.

The Contemporary Labor Market in Britain's Older Industrial Towns

A working definition of the towns.

So what exactly does the labor market now look like in Britain's older industrial towns? The figures we are able to present here are mostly for 2019—at the end of a sustained period of recovery from the 2008 financial crisis but before the coronavirus crisis hit economies across the world. The recession triggered by coronavirus will have worsened labor market conditions just about everywhere but its impact specifically on Britain's older industrial towns will take some while to become clear. It seems unlikely, however, that deep-seated differences between places, and in particular between older industrial towns and more prosperous local economies, will be easily overturned.

Presenting quantitative evidence on the labor market in Britain's older industrial towns requires the use of local area statistics, which in turn requires a working definition of the places to be included. Although these towns are a substantial part of the country there is no official definition of them to assist in the deployment of statistics. In this article we therefore use the term ‘older industrial towns’ to include all Britain's older industrial areas beyond the main regional cities, and we use district and unitary local authorities as the building block because they are the smallest unit for which much of the contemporary labor market data is available. The list of authorities is shown in Table 1 . The list is taken from our earlier report on older industrial towns (Beatty and Fothergill, 2018 ) and has also been deployed in a recent study of labor market adjustment (Beatty and Fothergill, 2020 ), where a fuller description can be found.

Districts and unitary local authorities covering Britain's older industrial towns.

Source: Beatty and Fothergill ( 2020 ) .

The core of the list comprises the former coalfields of the Midlands, North, Scotland and Wales, where so much of the UK's early industrialization took place and where so many of the UK's older industries once flourished. The additions to this core cover the locations of job loss from the UK's main steelworks, shipyards and concentrations of heavy engineering and chemicals. The list also includes the former mill towns of Lancashire and West Yorkshire, where the textile industry has all but disappeared. All the local authorities are in parts of the country where job losses from older industries have long posed a problem, in contrast to southern England where older industries were generally a smaller component of the economy. In 2018 the local authorities on the list had a combined population of 16.8 million, or 26% of the GB total.

Four points are worth noting. First, some of the local authorities actually cover substantial cities—Sunderland, Hull, Bradford, Stoke and Swansea are examples—though none of these are the main city within their region. Second, a number of the local authorities cover numerous quite small towns, especially in former mining areas such as County Durham. Third, some of the local authorities include rural areas just as some of the obvious omissions (Northumberland is an example) include sub-areas that are older industrial in character. Fourth, the sheer extent of job loss over the years means that in some cases ‘industrial’ may no longer be a very good description of the present-day town.

Labor Market Status

It is useful to begin by looking in Table 2 at the labor market status of working-age (16–64 years old) residents in Britain's older industrial towns. The figures here show that of the more than 10 million working age residents in the towns, 7.5 million were in work in 2019—an employment rate of 73%.

Labor market status of 16–64 years olds in older industrial towns, 2019.

Source: Annual Population Survey .

By comparison, recorded unemployment was modest—just 380,000, equivalent to 3.7% of the working age population. The unemployment figure here comes from the government's Annual Population Survey and uses the International Labour Organisation (ILO) definition of unemployment, which counts anyone who is out-of-work, available to start work within 2 weeks and has looked for work within the last 4 weeks. The ILO unemployment figures do not depend on benefit status and in the years following the financial crisis have been notably higher than UK claimant count. The 3.7% of the workforce unemployed in older industrial towns on the ILO measure is therefore actually the higher of the UK's two official unemployment estimates, which underlines the extent to which recorded unemployment had receded prior to the coronavirus crisis.

The remaining adults of working age fall into a number of categories. In Britain's older industrial towns there were rather more than 500,000 students, many of whom will be still at school or college rather than in higher education. There were more than 500,000 who look after family or home on a full-time basis, and 300,000 who described themselves as ‘retired.’

At 730,000, and accounting for 7.1% of all adults of working age, the single largest group of among the non-employed in the towns were the temporary and long-term sick, of whom all but 50,000 were in the long-term category. The Annual Population Survey tends to under-record the size of this group: in older industrial towns in May 2019, Department for Work and Pensions benefit data put the number of 16–64 year olds out-of-work and claiming incapacity-related benefits at 776,000, or 7.5% of all adults of working age. These figures underline the point that a diversion out of the labor market onto incapacity benefits remains a key feature of the towns. Indeed, with roughly a quarter of the GB population, Britain's older industrial towns are home to more than a third of the country's incapacity-related claimants.

Table 3 draws comparisons between older industrial towns, the main regional cities, London and the GB average. There are similarities and important differences. On the ILO measure of unemployment there is little to differentiate older industrial towns: the rate in 2019 was a little higher than the national average but a little below the level in the main regional cities. In contrast, the incapacity-related claimant rate in older industrial towns (7.5%) was higher than the rate in the main regional cities (7.1%) and well above the GB average (5.7%) and the rate in London (4.4%).

Labor market status: comparisons, 2019.

All figures are percentages of 16–64 year old residents .

Sources: Annual Population Survey, Department for Work and Pensions .

The employment rate—the share of adults of working age in work—paints a complex picture. On the raw figures, the employment rate in older industrial towns in 2019 does not appear unduly low—a little lower than the GB average or than in London but distinctly higher than in the main regional cities. The raw figures are however misleading. The big distortion is the distribution of students across the country, who are concentrated in London and the main regional cities where so many universities are located. In older industrial towns students accounted for just over 5% of all 16–64 year olds; in the main regional cities they accounted for nearer 10%. This distortion to the figures has always existed but as student numbers have increased it has become more important.

A better measure is therefore the employment rate excluding students . This points to older industrial towns as a whole lagging three percentage points behind the national average. It also narrows the gap between older industrial towns and the main regional cities. On this measure the labor market in older industrial towns looks less convincingly healthy.

Employment Structure

Table 4 shows the sectoral breakdown of the jobs in older industrial towns. In total 6.4 million jobs were located in the towns in 2019. Manufacturing, energy and water—‘industry’—accounted for one-in-seven, or 950,000 jobs in total. These days employment in older industrial towns is dominated by the service sector, particularly by over 2 million jobs in education, health and public administration, which will mostly be in the public sector, and by retail, distribution, hotels and related activities (a further 1.2 million in 2019).

Industry breakdown of jobs in older industrial towns, 2019.

The number of jobs in public services, such as schools and hospitals, is in most places driven by the size of the local population and many jobs in other parts of the service sector, such as retailing, follow local spending power, which is population-related. By contrast, jobs in businesses that serve markets beyond the local area, including most manufacturing but also some parts of the service sector, play a key role in driving the whole local economy because they bring in income to an area which then recirculates and supports other local businesses and jobs. Manufacturing's on-going significance to the economy of older industrial towns is therefore substantially greater than its share of employment.

Table 5 compares employment in older industrial towns with the main regional cities, London and the national average. This bears out the point that jobs in several parts of the service sector—retailing and public services for example—are found in large numbers in all areas because they are tied to population. The differences between older industrial towns and other places are in other sectors.

Industry breakdown of employment: comparisons, 2019.

In particular, despite years of decline which has often led to the disappearance of whole industries, Britain's older industrial towns continue to have a higher proportion of jobs in industry than the economy as a whole, than the main regional cities or than London in particular. These may be ‘older industrial towns’ but they remain to a large extent the heartland of British industry. The converse is that older industrial towns have proportionally fewer jobs in banking, finance and business services than either the main regional cities or London. The drivers of the economy of older industrial towns remain very different from those in London or the main regional cities.

Job Quality and the Workforce

In older industrial towns the shift from mining and manufacturing to new forms of employment has inevitably resulted in major changes in the experience of work. The qualitative aspects of these changes—job satisfaction, a sense of identity, and control over tasks and workload for example—are very real no doubt, and deserving of study. Here we are limited to the insights that official statistics are able to offer.

Table 6 brings together a number of indicators, again for 2019. These cover the nature of the employment in older industrial towns and two measures of the local workforce—the proportion of working age residents with degree-level qualifications and the proportion born outside the UK.

Selected labor market indicators: comparisons, 2019.

One of the widespread assumptions about the UK labor market is that as the economy recovered from the recession caused by the 2008 financial crisis the growth in employment was skewed toward part-time and insecure working, including debased forms of self-employment. A common assumption, too, is that these forms of employment became particularly prevalent in weaker local economies, such as much of older industrial Britain, where welfare reforms made it increasingly difficult for many claimants to stay on benefits. The proliferation of ‘self-employed’ delivery workers and taxi drivers, for example, was in the popular view been a defining feature of the last decade.

The first line of Table 6 shows that in fact self-employment in older industrial towns accounts for a below national average share of the workforce and a considerably smaller proportion than in London. This snapshot for 2019 does not however tell the full story because ‘self-employment’ has gradually been increasing. Expressed as a proportion of all employed residents, the increase since 2010 in older industrial towns was only 1 percentage point, on top of a 1 percentage point increase between 2000 and 2010, but the absolute numbers are substantial—an increase of 140,000 between 2010 and 2019. Self-employment accounted for 30% of the increase in residents in employment in older industrial towns between 2010 and 2019.

The increase in self-employment is probably unwelcome. As the UK government itself has documented (Department for Business, Innovation and Skills, 2016 ), the self-employed as a group have seen falling incomes since the post-financial crisis recession, which mostly reflects their changing composition as a group. The modern self-employed worker is less likely to be a prosperous entrepreneur or freelance worker than a quasi-employee with diminished employment rights.

The second line of Table 6 deals with part-time employment. This points to a share of jobs in older industrial towns that is only a little higher than in the main regional cities and in line with the GB average, and in older industrial towns the share fell by 1 percentage point between 2010 and 2019. Once again, the raw figures do not tell the whole story. According to the government's Annual Population Survey, 2.5 million people across the UK as a whole were ‘underemployed’ in 2019 in that they wanted to work more hours, were able to start to do so within 2 weeks and were already working less than full-time. This was down on the peak level of around 3.1 million in the wake of recession but still higher than the pre-recession figure of just below 2 million. Across the UK, around one-in-ten part-time workers say they could not find a full-time job.

Additionally, there has been an increase in the number of employees on zero hours contracts (Office for National Statistics, 2018 ). A government survey of businesses puts the figure for 2017 at 1.8 million contracts that did not guarantee a minimum number of hours. The Annual Population Survey puts the national figure for 2019 at 900,000, or 2.7% of all people in employment. Since 2010 the numbers have risen sharply from around 200,000 but the Office for National Statistics takes the view that part of the observed increase appears to be due to increased recognition and awareness of this form of employment. No local figures are available. According to the Office for National Statistics, the people on zero hours contracts are more likely to be young, part-time, women or in full-time education when compared with other people in employment, and only around a quarter say they would like more hours, mostly in their current job.

Across the UK as a whole, the Annual Population Survey points to 3% of workers with second jobs, and 4% in temporary employment. Of those in temporary employment, a quarter say this is because they could not find a permanent job, a proportion that has fallen from around 40% in the immediate wake of the financial crisis. Again, no local figures are available but the small national percentages with second jobs or in temporary employment suggest that both are likely to be relatively marginal features of the labor market in older industrial towns, though that does not rule out the possibility of increases since the pre-financial crisis years.

But if self-employment and part-time working (and possibly zero hours contracts, second jobs and temporary working) do not sharply differentiate older industrial areas from other places the remaining indicators in Table 6 most certainly do. The share of white-collar jobs is far below the level in the cities, the proportion of the workforce educated to degree level is far lower and so is the proportion of the workforce born outside the UK.

There are issues here of cause-and-effect. One interpretation could be that it is the location of white-collar jobs that follows the location of highly educated workers. There will always be cases that fit this model. A more likely model is that the composition of the workforce in older industrial towns reflects the nature of the job opportunities and that there is a migration of highly-educated workers out of the towns to the cities where they are more likely to find appropriate employment. Some of this will occur as school-leavers from the towns move to university, mostly in the cities, and then never return. At the extreme, London's exceptionally high proportion of graduates clearly reflects the availability of higher-level jobs that attract graduates from elsewhere in Britain and from the rest of the world. The industrial and service jobs that form such a large component of the economy in older industrial towns do not have the same magnetic appeal.

Likewise, the low proportion of the workforce born outside the UK at least in part reflects the long-term weakness of the economy in older industrial towns. Migrants are attracted to the places where jobs are more readily available. It should be no surprise, therefore, that international migrants are fewer in number in towns where the economic base has been eroded. There are exceptions of course—Bradford and the Lancashire mill towns are examples, where there continues to be in-migration to established Asian communities. As a general rule, however, it is the strength of the local economy to which we should look for the prime explanation.

Pay and Welfare Benefits

Table 7 compares the median weekly earnings of employees in older industrial towns with the equivalent figures for the main regional cities, London and the GB average. The first part of the table shows the figures for the jobs located in the area and demonstrates that in older industrial towns these jobs pay less than the national average, less than in the main regional cities and substantially less than in London. The median earnings of jobs in London are almost £200 a week (or 45%) higher than in older industrial towns.

Median earnings: comparisons, 2019.

Source: Annual Survey of Hours and Earnings .

The second part of the table shows the earnings of residents. On this measure, older industrial towns are still below the national average and well below the level in London, though in both cases by a smaller margin, but residents' earnings are much the same as in the main regional cities. The differences between the two halves of the table reflect the influence of commuting: some of the residents of older industrial towns fill higher-paid jobs in the cities.

One of the consequences of low pay in older industrial towns is that there is a substantial financial burden on the Exchequer. This occurs because the UK tax and benefit system operates to prop up household incomes not just for those out-of-work but also for those in low-paid employment. To illustrate this point, Table 8 looks at in-work households in receipt of Tax Credits. These figures are for the 2015/16 financial year—the last before Tax Credits began to be subsumed into Universal Credit and as a result difficult to disentangle as a separate category—but it is unlikely that the geography and to some extent the magnitude of the payments will have changed radically.

In-work households in receipt of Tax Credits, 2015/16.

Source: Her Majesty's Revenue and Customs .

The absolute numbers for older industrial towns are large. In total, just over 900,000 in-work households in the towns received an annualized average of almost £6,500 in Tax Credits, at a cost to the Exchequer of £6bn a year. What is also clear is that the cost to the Exchequer of Tax Credits is greater in older industrial towns than in the cities. This is not because the size of the average claim is higher—in fact it is lower in older industrial towns than in the cities—but because low wages bring larger numbers of households into the scope of Tax Credits. Averaged across the whole of the working age population (the final line of Table 8 ), the expenditure on Tax Credits is higher in older industrial towns than in the main regional cities or London, and higher than the GB average.

Relationship to the Cities

In recent years the dominant view within economic geography (Jacobs, 1986 ; Krugman, 1991 ; Centre for Cities, 2015 ) and in policy making (HM Government, 2011 ; Core Cities, 2013 ) has been that cities are the motor of regional and local growth. By implication, surrounding industrial towns are increasingly expected to function as their satellites, providing a source of labor and an overspill location for businesses. This is very different from the way in which Britain's industrial towns first developed, when they were nearly all locations of business growth in their own right.

What is undeniable is that approaching three-quarters of the population of Britain's older industrial towns listed earlier in Table 1 live in the immediate hinterland of the main regional cities. Only the remaining quarter live in towns located further afield—places such as Barrow in Cumbria or Grimsby on the south bank of the Humber, or smaller cities such as Hull, Swansea, Dundee, and Stoke on Trent that stand at some distance from the main regional cities.

The evidence on job growth in the final years of the twentieth century and the early years of the 21st did not support the view that employment in the UK's main regional cities was growing consistently faster than elsewhere (Champion and Townsend, 2011 , 2013 ; Martin et al., 2014 ). More recent trends, however, show that job growth in the cities is now substantially faster than in Britain's older industrial towns. As Table 9 shows, this is especially marked when the growth is expressed in relation to the resident working age population. On this measure, between 2010 and 2019 job growth in the main regional cities was five times faster than in older industrial towns, and nearly six times faster in London.

Increase in employment in area, 2010–2019.

In a small country such as the UK, with high car ownership and a network of public transport, the labor markets of cities and surrounding towns are inevitably strongly intertwined, as Swinney et al. ( 2018 ) have documented. This can be an asset for some older industrial towns if they are close to a city with a strong economy but a problem for others if they are further away or if their neighboring city is less prosperous. Additionally, Pike et al. ( 2016 ) identified ‘overshadowed cities’ (most of which are actually large towns)—that is, cities with a larger neighbor that hosts higher-level functions and provides development opportunities—as one of the categories of UK places facing relative decline.

The labor market links between older industrial towns and neighboring areas are certainly strong. Annual Population Survey data for 2019 puts the total net commuting out of older industrial towns at 1.06 million, equivalent to 14% of all residents in employment. Not all the commuting will have been into the main regional cities but this figure is a net flow —the daily flow outwards will be significantly larger, offset in part by in-commuting. Looking at commuting flows from the other direction, in 2019 the 10 main regional cities had a net in-flow of 990,000 commuters, equivalent to 27% of all the jobs located there.

Net out-commuting from Britain's older industrial towns rose by 200,000 between 2010 and 2019, and net in-commuting to the main regional cities rose by 100,000 over the same period. Around Manchester, Edinburgh and Cardiff there is clear evidence of rising in-flows of commuters from older industrial towns in surrounding areas (Beatty and Fothergill, 2020 ) but in these cities the growth in employment has been especially strong and the corresponding growth in their older industrial hinterlands (e.g., in Cardiff's case the Welsh Valleys) has been poor. Elsewhere the gap in job growth has been more modest and the changes in commuting patterns more complex.

London's particularly rapid growth—an additional 900,000 jobs between 2010 and 2019—has however had little impact on older industrial towns. Because of the distances involved it was always unrealistic to expect that London's growth would attract daily commuters from beyond the south of England, though there are undoubtedly Monday-to-Friday flows from longer distance into the capital. London's rapid job growth might however have been expected to attract a net inflow of migrants from older industrial towns in the rest of the country. In practice this has not happened because the London economy has tapped other sources of additional labor: a large net inflow of international migrants, a surge in commuting from the rest of southern England, rising labor force participation among London residents and a natural increase in the size of the local workforce that reflects a population skewed toward younger groups. The balance of internal migration (i.e., the flow of UK residents) has actually been strongly out of London (Beatty and Fothergill, 2020 ).

London it seems is no longer acting as a powerful magnet for migrants from the older industrial towns of the Midlands, North, Scotland and Wales, in effect detaching its growth from the labor market north of a line from the Severn to the Wash. This observation from the post-financial crisis years confirms a trend first identified in the previous decade: the net flow of internal migrants into London and the South of England has essentially come to a halt (Rowthorn, 2010 ).

Let us now draw some overall conclusions. The first is that set against the backdrop of years of massive industrial job loss, the labor market in Britain's older industrial towns is not as distressed as might have been feared. In some towns, it is worth remembering, virtually the whole of the original economic base has disappeared and this might have been expected to have resulted in permanent mass unemployment or large-scale depopulation. In fact, taking Britain's older industrial towns as a whole, prior to the coronavirus crisis the unemployment rate was only a little higher than the national average.

Britain's older industrial towns are not, it seems, locked into a spiral of decline. It might have been expected that the loss of most of their former industrial base, which in historical terms happened quite recently, might have triggered a longer-term knock-on loss of jobs, in the local service sector for example as a result of reduced local spending-power. If these negative consequences are still happening in the towns they have so far been offset by other more positive developments, including no doubt the impact of substantial public sector efforts to rebuild their local economies. On balance, the number of jobs in Britain's older industrial towns was growing prior to the coronavirus crisis, though more slowly than elsewhere.

The second conclusion is that even if conditions in Britain's older industrial towns are not as bad as they might have been, the labor market in the towns still remains difficult. This is evident not only in below-par employment rates but also in the very large numbers on incapacity benefits, in low pay, the predominance of manual jobs and a high dependence on in-work welfare benefits. In many respects this is ‘the long shadow of job loss.’

Until very recently, perhaps, the nature of the problem has been different from what it was 20 or 30 years ago. Even allowing for distortions to the official figures, unemployment in the towns was down on peak levels and down on the immediate post-financial crisis years. For many of the workless the problem is therefore likely to have been not that they could not find any job at all, which was probably the case in the era of 3 million claimant unemployed, but rather that they had difficulty finding suitable work with acceptable pay and conditions. In older industrial towns there have simply not been enough of these ‘good’ jobs to satisfy everyone, not least because the destruction of so much industry over so many years has removed the layer of jobs that once filled this important gap in the labor market. The new recession, triggered by coronavirus, looks likely to bring back older difficulties, at least for some while.

The third conclusion is that to describe most of these towns as ‘post-industrial’ is inaccurate. Certainly, they have lost most if not all the industry that underpinned their original growth but the figures here show that manufacturing, energy and water together still account for 15% of local employment—nearly four times the proportion in London and twice the proportion in the main regional cities. Add in the now numerous jobs in warehousing—just another ‘industry’ to many residents, no doubt, and a major element of the economy in several former coalfields—and the proportion would be significantly higher.

The point here is that Britain's older industrial towns remain the heartland of British manufacturing. Unlike the big regional cities, and London in particular, which have shed most of their former industrial jobs and become centers of banking, business services and higher education, many of Britain's older industrial towns remain to an important extent ‘industrial’ with economies that still depend on the local and national performance of this sector.

The fourth conclusion is that Britain's older industrial towns need to be understood in their wider geographical context. They do not exist in isolation from the places around them, in particular from neighboring cities. They are increasingly becoming dormitories for men and women who work elsewhere, partly no doubt because the on-going slack in the local labor market encourages commuting to neighboring areas and further afield.

The scale of commuting should be kept in perspective: the net outflow from the towns is presently just over a million whereas nearly seven and a half million residents of the towns are in work. Moreover, it is not always the big regional cities that are the destination and there is no evidence at all that London's spectacular recent growth has been of any direct benefit to the labor market in Britain's older industrial towns. Nevertheless, commuting on this scale represents a significant redefinition of the role of the towns within wider urban networks.

Data Availability Statement

Author contributions.

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

• Beatty C. (2016). Two become one: the integration of the male and female labour markets in the English and Welsh coalfields . Reg. Stud. 50 , 823–834. 10.1080/00343404.2014.943713 [ CrossRef ] [ Google Scholar ]
• Beatty C., Fothergill S. (1996). Labour market adjustment in areas of chronic industrial decline: the case of the UK coalfields . Reg. Stud. 30 , 637–650. 10.1080/00343409612331349928 [ CrossRef ] [ Google Scholar ]
• Beatty C., Fothergill S. (2005). The diversion from “unemployment” to “sickness” across British regions and districts . Reg. Stud. 39 , 837–854. 10.1080/00343400500289804 [ CrossRef ] [ Google Scholar ]
• Beatty C., Fothergill S. (2017). The impact on welfare and public finances of job loss in industrial Britain . Reg. Stud. Reg. Sci. 4 , 161–180. 10.1080/21681376.2017.1346481 [ CrossRef ] [ Google Scholar ]
• Beatty C., Fothergill S. (2018). The Contemporary Labour Market in Britain's Older Industrial Towns . Sheffield: CRESR, Sheffield Hallam University. [ Google Scholar ]
• Beatty C., Fothergill S. (2020). Recovery or stagnation: Britain's older industrial towns since the recession . Reg. Stud. 25 , 1–12. 10.1080/00343404.2019.1699651 [ CrossRef ] [ Google Scholar ]
• Beatty C., Fothergill S., Gore T. (2017). The Real Level of Unemployment 2017 . Sheffield: CRESR, Sheffield Hallam University. [ Google Scholar ]
• Beatty C., Fothergill S., Gore T. (2019). The State of the Coalfields 2019: Economic and Social Conditions in the Former Coalfields of England, Scotland and Wales . Sheffield: CRESR, Sheffield Hallam University. [ Google Scholar ]
• Beatty C., Fothergill S., Houston D., Powell R., Sissons P. (2009). Women on Incapacity Benefits . Sheffield: CRESR, Sheffield Hallam University. [ Google Scholar ]
• Beatty C., Fothergill S., Powell R. (2007). Twenty years on: has the economy of the UK coalfields recovered? Environ. Plan. A 39 , 1654–1675. 10.1068/a38216 [ CrossRef ] [ Google Scholar ]
• Centre for Cities (2015). Cities Outlook 2015. London: Centre for Cities. [ Google Scholar ]
• Champion T., Townsend A. (2011). The fluctuating record of economic regeneration in England's second-order city regions 1984–2007 . Urban Stud. 48 , 1539–1562. 10.1177/0042098010375320 [ CrossRef ] [ Google Scholar ]
• Champion T., Townsend A. (2013). Great Britain's second-order city regions in recession . Environ. Plan. A , 45 , 362–382. 10.1068/a45100 [ CrossRef ] [ Google Scholar ]
• Core Cities (2013). Competitive Cities, Prosperous People: A Core Cities Prospectus for Growth . Manchester: Core Cities. [ Google Scholar ]
• Department for Business, Innovation and Skills. (2016). The Income of the Self-Employed . London: BIS. [ Google Scholar ]
• Erdem E., Glyn A. (2001). Jobs deficits in UK regions . Oxf. Bull. Econ.Stat. 63 , 737–752. 10.1111/1468-0084.63.spe1.7 [ CrossRef ] [ Google Scholar ]
• Fothergill S., Guy N. (1991). Retreat From the Regions: Corporate Change and the Closure of Factories. London: Jessica Kingsley/Regional Studies Association. [ Google Scholar ]
• Gardiner B., Martin R., Sunley P., Tyler P. (2013). Spatially imbalanced growth in the British economy . J. Econ. Geogr. 13 , 889–928. 10.1093/jeg/lbt003 [ CrossRef ] [ Google Scholar ]
• Gudgin G. (1995). Regional problems and policy in the UK . Oxf. Rev. Econ. Policy 11 , 18–93. [ Google Scholar ]
• HM Government (2011). Unlocking Growth in Cities. London: HM Government. [ Google Scholar ]
• Jacobs J. (1986). Cities and the Wealth of Nations . Harmondsworth: Penguin. [ Google Scholar ]
• Jennings W. (2017). Cities, Towns and the General Election of 2017. Part 1 of Cities and Towns: The 2017 General Election and The Social Divisions of Place. London: New Economics Foundation. [ Google Scholar ]
• Krugman P. (1991). Geography and Trade. Cambridge, MA: MIT Press. [ Google Scholar ]
• MacKay R. (1999). Work and nonwork: a more difficult labour market . Environ. Plan. A 31 , 1919–1934. 10.1068/a311919 [ CrossRef ] [ Google Scholar ]
• Martin R., Gardiner B., Tyler P. (2014). The Evolving Economic Performance of UK Cities: City Growth Patterns 1981–2011. London: Foresight, Government Office for Science. [ Google Scholar ]
• Martin R., Rowthorn R. (1986). The Geography of De-industrialisation . London: Macmillan. [ Google Scholar ]
• Martin R., Sunley P., Tyler P., Gardiner B. (2016). Divergent cities in post-industrial Britain . Camb. J. Reg. Econ. Soc. 9 , 269–299. 10.1093/cjres/rsw005 [ CrossRef ] [ Google Scholar ]
• Ministry of Housing, Communities and Local Government. (2019). Town Deals: Prospectus . London: MHCLG. [ Google Scholar ]
• Office for National Statistics (2018). Contracts That Do Not Guarantee a Minimum Number of Hours: April 2018 . London: ONS. [ Google Scholar ]
• Pike A., MacKinnon D., Coombes M., Champion T., Bradley D., Cumbers A., et al.. (2016). Uneven Growth: Tackling City Decline . York: Joseph Rowntree Foundation. [ Google Scholar ]
• Rowthorn R. (2010). Combined and uneven development: reflections on the North–South divide . Spat. Econ. Anal. 5 , 363–388. 10.1080/17421772.2010.516445 [ CrossRef ] [ Google Scholar ]
• Rowthorn R., Webster D. (2008). Male worklessness and the rise of lone parenthood in Great Britain . Camb. J. Reg. Econ. Soc. 1 , 69–88. 10.1093/cjres/rsm004 [ CrossRef ] [ Google Scholar ]
• Swinney P., McDonald R., Ramuni L. (2018). Talk of the Town: The Economic Links Between Cities and Towns . London: Centre for Cities. [ Google Scholar ]
• Swinney P., Thomas E. (2015). A Century of Cities: Urban Economic Change Since 1911 . London: Centre for Cities. [ Google Scholar ]
• Townsend A. (1983). The Impact of Recession on Industry, Employment and the Regions, 1976–1981 . Beckenham: Croom Helm. [ Google Scholar ]
• Townsend A., Champion T. (2014). The impact of recession on city regions: the British experience, 2008–2013 . Local Econ. 29 , 38–51. 10.1177/0269094213518885 [ CrossRef ] [ Google Scholar ]
• Webster D. (2000). The geographical concentration of labour market disadvantage . Oxf. Rev. Econ. Pol. 16 , 14–128. 10.1093/oxrep/16.1.114 [ CrossRef ] [ Google Scholar ]

De-industrialisation: a case study of Dundee, 1951-2001 – and its broad implications

by Jim Tomlinson (Economic and Social History, University of Glasgow)

This research will be presented during the EHS Annual Conference in Belfast, April 5th – 7th 2019. Conference registration can be found on the EHS website.

The huge loss of industrial employment – ‘de-industrialisation’ – has been one of the most important economic and social changes in Britain since the Second World War. But its timing, causes and effects are often misunderstood.

My study of Dundee, a typical post-industrial city, enables us to examine this process and to demonstrate important aspects of the process relevant to the whole country. The key messages, which I will present at the Economic History Society’s 2019 annual conference, are as follows:

• De-industrialisation in Britain began in the 1950s: since then, the proportion of industrial jobs has shrunk from over 50% to around 15%, with the fall in manufacturing jobs even more dramatic.
• De-industrialisation was greatly accelerated by the ‘Thatcherite’ policies of the 1980s, but the process began long before that date.
• In particular, the ‘old staple’ industries, such as textiles, coal and the railways, lost more workers in the 1950s and 1960s than in the 1980s.
• De-industrialisation was not mainly caused by the recent phase of ‘globalisation’.
• The most important causes were technological change and shifts in patterns of consumption.
• De-industrialisation doesn’t mean ‘we don’t make anything any more’; the trend in industrial output was upwards until the 1970s and roughly flat since then, but higher productivity means it takes far fewer workers to produces this output.
• Most job losses arose from either long, slow attrition of employment levels in existing firms, or the slow growth of new jobs, not from dramatic, large-scale closures.
• De-industrialisation matters especially because it has polarised the labour market much more into ‘lovely and lousy’ jobs; ‘lovely’ jobs are well-paid and relatively secure; ‘lousy’ jobs poorly paid and precarious.
• The number of ‘lovely’ jobs, such as professionals, administrators, managers and technicians, has increased across all sectors of the economy, including industry.
• The number of ‘lovely’ jobs has been particularly increased by the expansion of public sector employment, especially in health and education, and the numbers in these areas have barely been affected by recent austerity (unlike employment in local authorities).
• Public sector ‘outsourcing’ has increased the polarisation of the labour market, as many of the outsourced jobs have been the low-skilled ones where public employment previously provided some protection against the impact of weak bargaining power.
• ‘Lovely’ jobs commonly require significant educational qualifications, and average educational achievement has shot up in the period of de-industrialisation, especially in universities. Universities in turn have been a significant source of expansion of ‘lovely’ jobs.
• The disadvantages of low educational attainment have been magnified by de-industrialisation, which makes access to ‘lovely’ jobs almost entirely reliant on high levels of attainment.
• The transition from the dominance of industry has pushed many people out of the labour market, something that is evident not only in unemployment but also in much higher levels of long-term sickness and disability.

As a result of this transition, there has been a large increase in self-employment, much of which is poorly paid.

Latest news

Ape research grant, daphne jackson trust and fellowship scheme.

De-industrialization: a case study of Dundee, 1951–2001, and its broad implications

• Publisher DOI

Tomlinson J, Phillips J & Wright V (2022) De-industrialization: a case study of Dundee, 1951–2001, and its broad implications. Business History , 64 (1), pp. 28-54. https://doi.org/10.1080/00076791.2019.1676235

Abstract Using a case study of one Scottish city, Dundee, this article addresses some of the tensions involved in the use of the concept of ‘de-industrialization’. Widely used to try to understand economic and social change in the post-war years, this term is complex and controversial. This article unravels some of this complexity, arguing that the term is potentially very helpful, but needs careful definition, nuanced application and recognition of its limits. The focus here is on the impact of changing industrial structures on the labour market. After analysing the processes of firm births and deaths, the study looks at the decline of the ‘old staple’ industry, jute manufacturing in Dundee. The next sections assess the role of multinational enterprises in re-shaping the employment structure of the city, before looking at the contraction of some of the city’s other industries. Attention then turns to the impact of all these changes on the economic welfare of the city. The final section draws conclusions about our general understanding of de-industrialization from the Dundee case.

Keywords De-industrialization; employment; Dundee; Scotland; jute; Timex; decline; industrial closures; capital flight; multinational corporations

Journal Business History: Volume 64, Issue 1

Dr Valerie Wright

Research Fellow, Dementia and Ageing

Queen’s University Oral History, Technology & Ethics

Memory, conflict and class: deindustrialisation in belfast and county durham since 1970, introduction, political background.

The two case studies – both nationalised heavy industries – highlight the agency open to state actors when pursuing deindustrial agendas. Deindustrialisation, as Jim Phillips has written, ‘is a deliberate and willed human phenomenon’. Local political sensitivities influenced the time-scale, scope and intensity of shipyard deindustrialisation in Belfast. Loyalist/unionists equated British support for Harland & Wolff with British support for unionism, and portrayed reneging on shipyard aid as incipient economic evacuation from Northern Ireland. A group interview with retired welders reveals how some former shipyard workers feel exploited by unionist politicians. Rapid job losses against the backdrop of the Troubles, discussed by former storeroom clerk Maurice Davies, would have inflamed the security situation. Put simply, Harland & Wolff was protected because successive governments feared the grassroots backlash flowing from a closure decision. ‘Soft’ deindustrialisation was therefore applied – piecemeal dismantling, which was carefully stage-managed to shield the state from blame.

Policies applied to the National Coal Board/British Coal differed markedly, particularly after the 1984/5 miners’ strike. Steve Fergus discussed how Margaret Thatcher’s Conservative Government applied subtle (but significant) restrictions to British Coal’s commercial environment. Pit closures proceeded with minimal warning and consultation. The workforce was further demoralised by bullish management tactics, and lured with generous redundancy packages to accept closures. County Durham’s final coal mines were closed in 1993. In local authority areas such as Easington District, the economic lifeblood of many communities was eradicated at a stroke, with little planning – or hope – for adequate replacement. Geordie Maitland discussed the closure of Murton Colliery in 1991, arguing that pit closures were executed in an unnecessarily callous manner. Deindustrialisation is not inevitable and can be regulated if the political inclination exists.

• Kept it going
• Ian Paisley
• This woman is a threat
• Why did Murton close?

Job Loss – Contested Perceptions

Industrial redundancy elicited myriad personal responses. For some, the loss of shipyard/pit employment marked a dramatic, and undesirable, life rupture. Fred Hoskins’ testimony is a stark example of this. Others recalled deindustrialisation as a kind of liberation, freeing them to pursue new careers in more rewarding and less dangerous jobs, or enjoy a long retirement. Former H&W manager Billy Harrison and Horden miner Gordon McDowell narrated their positive personal experiences of redundancy. Gordon’s testimony, nonetheless, was qualified with a hint of what Andrew Perchard has termed ‘survivor guilt’: acknowledgement that others (especially younger generations) faced bleaker circumstances. Deindustrialisation also challenged traditional gender relations and certainties, and was recalled differently by men and women. Maureen Foster, a former Women Against Pit Closures activist, discussed how coal industry redundancy could alter the domestic balance of power. Louise Jenkins made an insightful analysis of the age profile of redundant coal miners. She argued that where families were located on the “age spectrum” affected their job loss strategy. The quality and longevity of post-pit/shipyard employment was of critical importance in how people recalled the deindustrialisation process. Drew Shaw, a printer at H&W between 1959 and 1987, discussed his struggles, post-shipyard redundancy, to secure alternative employment and reluctant decision to withdraw (prematurely, in his view) from the labour market aged 57. Those that struggled economically post-redundancy were among the most critical of deindustrialisation.

• Miss the shipyard
• Liberated women
• Age spectrum

Community Change

The demise of the industrial economy had significant social and cultural ripples, many of which can still be felt today. In mining communities in County Durham, the socio-economic scars of deindustrialisation are still visible. Unemployment and poverty rates rocketed, and remain above the national average. Industrial communities slowly disintegrated as many mining families moved out. In ex-pit villages private landlords, with negligible local accountability, purchased former NCB housing stock, supplying the local County Council with accommodation for “problem” tenants and people exported from other authority areas. Ex-miners Alan Johnson and Steve Fergus discussed the gradual and ongoing process of community change in their respective villages – Dawdon and Easington Colliery. Dave Douglass, a former National Union of Mineworkers official, talks to schools in the Doncaster area. Dave’s testimony gives a rather bleak impression of young people’s current economic opportunities, and sense of despair when confronted with stories of the (now absent) well-paid coal industry. He also commented that service sector employment tended to lack masculine-endowing prestige. The growing popularity of the annual Durham Miners’ Gala gives some grounds for optimism. At the “Big Meeting”, post-industrial communities are using the dead past of coal to reconstruct a sense of self and revive community cohesion.

In Belfast – particularly inner East Belfast, adjacent to the shipyard – the process of community change is murkier. The gradual nature of shipyard run-down protected these communities from catastrophic economic damage, argued ex shop steward Campbell Kell. The diverse nature of the city’s labour market also provided better job prospects for redundant shipyard workers, vis-à-vis redundant Durham miners. The removal of much of Belfast’s terraced housing system in the 1970s and 1980s also affected community integration as many families moved to new suburban estates. Gentrification has also played a part. Jackie Pollock reflected on the wider process of deindustrialisation in Belfast, and his assessment of the economic damage to East Belfast counter-balances some of the more optimistic social audits. Belfast’s Protestant working-class communities – formerly at the heart of Ulster’s industrial economy – are among the most deprived in Northern Ireland, and, for a variety of complex reasons, suffer from poor educational attainment levels and economic marginalisation.

• East Belfast impact
• Heart in boots
• John Lennon
• Teenagers-schools

Lived Legacies

The oral testimony used in this article paints a deeply fragmented picture of deindustrialisation. Despite the often negative socio-economic impact of closure on their lives and communities, few shipyard workers and coal miners wished for heavy industrial jobs to return. There was a degree of contentment that the industrial economy – in particular the damaging health effects associated with such work – had come to an end, albeit underpinned with criticism about how slowly economic regeneration has taken place. Reflecting on the heavy industrial economy, the shipyard welders group and ex-miner Stephen Jenkins foregrounded health hazards as a major factor behind their retrospective ambivalence about their former work. In Belfast and County Durham, the industrial economy has cast a long shadow. The cultural, economic and social aftershocks are still being felt, long after the initial moment of job loss and closure. Oral history is invaluable for unpicking the mixed meanings and complex social and cultural legacies of deindustrialisation.

• Cranes/Walking dead
• Blood snots shit

This case study is written by  Pete Hodson  and uses oral history testimony collected for his PhD research.

phodson01@qub.ac.uk 

• Search Menu
• Sign in through your institution
• Browse content in A - General Economics and Teaching
• Browse content in A1 - General Economics
• A10 - General
• A11 - Role of Economics; Role of Economists; Market for Economists
• A12 - Relation of Economics to Other Disciplines
• A13 - Relation of Economics to Social Values
• A14 - Sociology of Economics
• Browse content in A2 - Economic Education and Teaching of Economics
• A20 - General
• A29 - Other
• A3 - Collective Works
• Browse content in B - History of Economic Thought, Methodology, and Heterodox Approaches
• Browse content in B0 - General
• B00 - General
• Browse content in B1 - History of Economic Thought through 1925
• B10 - General
• B11 - Preclassical (Ancient, Medieval, Mercantilist, Physiocratic)
• B12 - Classical (includes Adam Smith)
• B13 - Neoclassical through 1925 (Austrian, Marshallian, Walrasian, Stockholm School)
• B14 - Socialist; Marxist
• B15 - Historical; Institutional; Evolutionary
• B16 - History of Economic Thought: Quantitative and Mathematical
• B17 - International Trade and Finance
• B19 - Other
• Browse content in B2 - History of Economic Thought since 1925
• B20 - General
• B21 - Microeconomics
• B22 - Macroeconomics
• B23 - Econometrics; Quantitative and Mathematical Studies
• B24 - Socialist; Marxist; Sraffian
• B25 - Historical; Institutional; Evolutionary; Austrian
• B26 - Financial Economics
• B27 - International Trade and Finance
• B29 - Other
• Browse content in B3 - History of Economic Thought: Individuals
• B30 - General
• B31 - Individuals
• Browse content in B4 - Economic Methodology
• B40 - General
• B41 - Economic Methodology
• B49 - Other
• Browse content in B5 - Current Heterodox Approaches
• B50 - General
• B51 - Socialist; Marxian; Sraffian
• B52 - Institutional; Evolutionary
• B53 - Austrian
• B54 - Feminist Economics
• B55 - Social Economics
• B59 - Other
• Browse content in C - Mathematical and Quantitative Methods
• Browse content in C0 - General
• C00 - General
• C02 - Mathematical Methods
• Browse content in C1 - Econometric and Statistical Methods and Methodology: General
• C10 - General
• C12 - Hypothesis Testing: General
• C13 - Estimation: General
• C14 - Semiparametric and Nonparametric Methods: General
• C18 - Methodological Issues: General
• C19 - Other
• Browse content in C2 - Single Equation Models; Single Variables
• C20 - General
• C21 - Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions
• C22 - Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes
• C23 - Panel Data Models; Spatio-temporal Models
• C25 - Discrete Regression and Qualitative Choice Models; Discrete Regressors; Proportions; Probabilities
• Browse content in C3 - Multiple or Simultaneous Equation Models; Multiple Variables
• C30 - General
• C32 - Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes; State Space Models
• C34 - Truncated and Censored Models; Switching Regression Models
• C38 - Classification Methods; Cluster Analysis; Principal Components; Factor Models
• Browse content in C4 - Econometric and Statistical Methods: Special Topics
• C43 - Index Numbers and Aggregation
• C44 - Operations Research; Statistical Decision Theory
• Browse content in C5 - Econometric Modeling
• C50 - General
• Browse content in C6 - Mathematical Methods; Programming Models; Mathematical and Simulation Modeling
• C60 - General
• C61 - Optimization Techniques; Programming Models; Dynamic Analysis
• C62 - Existence and Stability Conditions of Equilibrium
• C63 - Computational Techniques; Simulation Modeling
• C65 - Miscellaneous Mathematical Tools
• C67 - Input-Output Models
• Browse content in C8 - Data Collection and Data Estimation Methodology; Computer Programs
• C82 - Methodology for Collecting, Estimating, and Organizing Macroeconomic Data; Data Access
• C89 - Other
• Browse content in C9 - Design of Experiments
• C90 - General
• C91 - Laboratory, Individual Behavior
• C92 - Laboratory, Group Behavior
• C93 - Field Experiments
• Browse content in D - Microeconomics
• Browse content in D0 - General
• D01 - Microeconomic Behavior: Underlying Principles
• D02 - Institutions: Design, Formation, Operations, and Impact
• D03 - Behavioral Microeconomics: Underlying Principles
• Browse content in D1 - Household Behavior and Family Economics
• D10 - General
• D11 - Consumer Economics: Theory
• D12 - Consumer Economics: Empirical Analysis
• D13 - Household Production and Intrahousehold Allocation
• D14 - Household Saving; Personal Finance
• Browse content in D2 - Production and Organizations
• D20 - General
• D21 - Firm Behavior: Theory
• D22 - Firm Behavior: Empirical Analysis
• D23 - Organizational Behavior; Transaction Costs; Property Rights
• D24 - Production; Cost; Capital; Capital, Total Factor, and Multifactor Productivity; Capacity
• D25 - Intertemporal Firm Choice: Investment, Capacity, and Financing
• Browse content in D3 - Distribution
• D30 - General
• D31 - Personal Income, Wealth, and Their Distributions
• D33 - Factor Income Distribution
• D39 - Other
• Browse content in D4 - Market Structure, Pricing, and Design
• D40 - General
• D41 - Perfect Competition
• D42 - Monopoly
• D43 - Oligopoly and Other Forms of Market Imperfection
• D46 - Value Theory
• Browse content in D5 - General Equilibrium and Disequilibrium
• D50 - General
• D51 - Exchange and Production Economies
• D57 - Input-Output Tables and Analysis
• D58 - Computable and Other Applied General Equilibrium Models
• Browse content in D6 - Welfare Economics
• D60 - General
• D61 - Allocative Efficiency; Cost-Benefit Analysis
• D62 - Externalities
• D63 - Equity, Justice, Inequality, and Other Normative Criteria and Measurement
• D64 - Altruism; Philanthropy
• D69 - Other
• Browse content in D7 - Analysis of Collective Decision-Making
• D71 - Social Choice; Clubs; Committees; Associations
• D72 - Political Processes: Rent-seeking, Lobbying, Elections, Legislatures, and Voting Behavior
• D73 - Bureaucracy; Administrative Processes in Public Organizations; Corruption
• D74 - Conflict; Conflict Resolution; Alliances; Revolutions
• Browse content in D8 - Information, Knowledge, and Uncertainty
• D80 - General
• D81 - Criteria for Decision-Making under Risk and Uncertainty
• D82 - Asymmetric and Private Information; Mechanism Design
• D83 - Search; Learning; Information and Knowledge; Communication; Belief; Unawareness
• D84 - Expectations; Speculations
• D85 - Network Formation and Analysis: Theory
• D86 - Economics of Contract: Theory
• D87 - Neuroeconomics
• Browse content in D9 - Micro-Based Behavioral Economics
• D91 - Role and Effects of Psychological, Emotional, Social, and Cognitive Factors on Decision Making
• Browse content in E - Macroeconomics and Monetary Economics
• Browse content in E0 - General
• E00 - General
• E01 - Measurement and Data on National Income and Product Accounts and Wealth; Environmental Accounts
• E02 - Institutions and the Macroeconomy
• Browse content in E1 - General Aggregative Models
• E10 - General
• E11 - Marxian; Sraffian; Kaleckian
• E12 - Keynes; Keynesian; Post-Keynesian
• E13 - Neoclassical
• E16 - Social Accounting Matrix
• E17 - Forecasting and Simulation: Models and Applications
• Browse content in E2 - Consumption, Saving, Production, Investment, Labor Markets, and Informal Economy
• E20 - General
• E21 - Consumption; Saving; Wealth
• E22 - Investment; Capital; Intangible Capital; Capacity
• E23 - Production
• E24 - Employment; Unemployment; Wages; Intergenerational Income Distribution; Aggregate Human Capital; Aggregate Labor Productivity
• E25 - Aggregate Factor Income Distribution
• E26 - Informal Economy; Underground Economy
• E27 - Forecasting and Simulation: Models and Applications
• Browse content in E3 - Prices, Business Fluctuations, and Cycles
• E30 - General
• E31 - Price Level; Inflation; Deflation
• E32 - Business Fluctuations; Cycles
• E37 - Forecasting and Simulation: Models and Applications
• Browse content in E4 - Money and Interest Rates
• E40 - General
• E41 - Demand for Money
• E42 - Monetary Systems; Standards; Regimes; Government and the Monetary System; Payment Systems
• E43 - Interest Rates: Determination, Term Structure, and Effects
• E44 - Financial Markets and the Macroeconomy
• E49 - Other
• Browse content in E5 - Monetary Policy, Central Banking, and the Supply of Money and Credit
• E50 - General
• E51 - Money Supply; Credit; Money Multipliers
• E52 - Monetary Policy
• E58 - Central Banks and Their Policies
• Browse content in E6 - Macroeconomic Policy, Macroeconomic Aspects of Public Finance, and General Outlook
• E60 - General
• E61 - Policy Objectives; Policy Designs and Consistency; Policy Coordination
• E62 - Fiscal Policy
• E63 - Comparative or Joint Analysis of Fiscal and Monetary Policy; Stabilization; Treasury Policy
• E64 - Incomes Policy; Price Policy
• E65 - Studies of Particular Policy Episodes
• Browse content in F - International Economics
• Browse content in F0 - General
• F00 - General
• F01 - Global Outlook
• F02 - International Economic Order and Integration
• Browse content in F1 - Trade
• F10 - General
• F11 - Neoclassical Models of Trade
• F12 - Models of Trade with Imperfect Competition and Scale Economies; Fragmentation
• F13 - Trade Policy; International Trade Organizations
• F14 - Empirical Studies of Trade
• F15 - Economic Integration
• F16 - Trade and Labor Market Interactions
• F17 - Trade Forecasting and Simulation
• F18 - Trade and Environment
• Browse content in F2 - International Factor Movements and International Business
• F20 - General
• F21 - International Investment; Long-Term Capital Movements
• F22 - International Migration
• F23 - Multinational Firms; International Business
• Browse content in F3 - International Finance
• F30 - General
• F31 - Foreign Exchange
• F32 - Current Account Adjustment; Short-Term Capital Movements
• F33 - International Monetary Arrangements and Institutions
• F34 - International Lending and Debt Problems
• F35 - Foreign Aid
• F36 - Financial Aspects of Economic Integration
• F37 - International Finance Forecasting and Simulation: Models and Applications
• F39 - Other
• Browse content in F4 - Macroeconomic Aspects of International Trade and Finance
• F40 - General
• F41 - Open Economy Macroeconomics
• F42 - International Policy Coordination and Transmission
• F43 - Economic Growth of Open Economies
• F44 - International Business Cycles
• F45 - Macroeconomic Issues of Monetary Unions
• F47 - Forecasting and Simulation: Models and Applications
• Browse content in F5 - International Relations, National Security, and International Political Economy
• F50 - General
• F51 - International Conflicts; Negotiations; Sanctions
• F53 - International Agreements and Observance; International Organizations
• F54 - Colonialism; Imperialism; Postcolonialism
• F55 - International Institutional Arrangements
• F59 - Other
• Browse content in F6 - Economic Impacts of Globalization
• F60 - General
• F61 - Microeconomic Impacts
• F62 - Macroeconomic Impacts
• F63 - Economic Development
• F64 - Environment
• F65 - Finance
• Browse content in G - Financial Economics
• Browse content in G0 - General
• G00 - General
• G01 - Financial Crises
• Browse content in G1 - General Financial Markets
• G10 - General
• G11 - Portfolio Choice; Investment Decisions
• G12 - Asset Pricing; Trading volume; Bond Interest Rates
• G13 - Contingent Pricing; Futures Pricing
• G14 - Information and Market Efficiency; Event Studies; Insider Trading
• G15 - International Financial Markets
• G18 - Government Policy and Regulation
• G19 - Other
• Browse content in G2 - Financial Institutions and Services
• G20 - General
• G21 - Banks; Depository Institutions; Micro Finance Institutions; Mortgages
• G22 - Insurance; Insurance Companies; Actuarial Studies
• G23 - Non-bank Financial Institutions; Financial Instruments; Institutional Investors
• G24 - Investment Banking; Venture Capital; Brokerage; Ratings and Ratings Agencies
• G28 - Government Policy and Regulation
• Browse content in G3 - Corporate Finance and Governance
• G30 - General
• G32 - Financing Policy; Financial Risk and Risk Management; Capital and Ownership Structure; Value of Firms; Goodwill
• G33 - Bankruptcy; Liquidation
• G34 - Mergers; Acquisitions; Restructuring; Corporate Governance
• G35 - Payout Policy
• G38 - Government Policy and Regulation
• Browse content in G5 - Household Finance
• G51 - Household Saving, Borrowing, Debt, and Wealth
• Browse content in H - Public Economics
• Browse content in H1 - Structure and Scope of Government
• H10 - General
• H11 - Structure, Scope, and Performance of Government
• H12 - Crisis Management
• Browse content in H2 - Taxation, Subsidies, and Revenue
• H20 - General
• H22 - Incidence
• H23 - Externalities; Redistributive Effects; Environmental Taxes and Subsidies
• H25 - Business Taxes and Subsidies
• H26 - Tax Evasion and Avoidance
• Browse content in H3 - Fiscal Policies and Behavior of Economic Agents
• Browse content in H4 - Publicly Provided Goods
• H40 - General
• H41 - Public Goods
• Browse content in H5 - National Government Expenditures and Related Policies
• H50 - General
• H53 - Government Expenditures and Welfare Programs
• H55 - Social Security and Public Pensions
• H56 - National Security and War
• Browse content in H6 - National Budget, Deficit, and Debt
• H60 - General
• H62 - Deficit; Surplus
• H63 - Debt; Debt Management; Sovereign Debt
• H68 - Forecasts of Budgets, Deficits, and Debt
• Browse content in H7 - State and Local Government; Intergovernmental Relations
• H70 - General
• H74 - State and Local Borrowing
• H77 - Intergovernmental Relations; Federalism; Secession
• Browse content in I - Health, Education, and Welfare
• Browse content in I0 - General
• I00 - General
• Browse content in I1 - Health
• I10 - General
• I12 - Health Behavior
• I14 - Health and Inequality
• I15 - Health and Economic Development
• Browse content in I2 - Education and Research Institutions
• I20 - General
• I21 - Analysis of Education
• I23 - Higher Education; Research Institutions
• I24 - Education and Inequality
• I26 - Returns to Education
• Browse content in I3 - Welfare, Well-Being, and Poverty
• I30 - General
• I31 - General Welfare
• I32 - Measurement and Analysis of Poverty
• I38 - Government Policy; Provision and Effects of Welfare Programs
• Browse content in J - Labor and Demographic Economics
• Browse content in J0 - General
• J00 - General
• J01 - Labor Economics: General
• J08 - Labor Economics Policies
• Browse content in J1 - Demographic Economics
• J10 - General
• J13 - Fertility; Family Planning; Child Care; Children; Youth
• J15 - Economics of Minorities, Races, Indigenous Peoples, and Immigrants; Non-labor Discrimination
• J16 - Economics of Gender; Non-labor Discrimination
• J18 - Public Policy
• Browse content in J2 - Demand and Supply of Labor
• J20 - General
• J21 - Labor Force and Employment, Size, and Structure
• J22 - Time Allocation and Labor Supply
• J23 - Labor Demand
• J24 - Human Capital; Skills; Occupational Choice; Labor Productivity
• J26 - Retirement; Retirement Policies
• J28 - Safety; Job Satisfaction; Related Public Policy
• J29 - Other
• Browse content in J3 - Wages, Compensation, and Labor Costs
• J30 - General
• J31 - Wage Level and Structure; Wage Differentials
• J32 - Nonwage Labor Costs and Benefits; Retirement Plans; Private Pensions
• J33 - Compensation Packages; Payment Methods
• J38 - Public Policy
• Browse content in J4 - Particular Labor Markets
• J40 - General
• J41 - Labor Contracts
• J42 - Monopsony; Segmented Labor Markets
• J44 - Professional Labor Markets; Occupational Licensing
• J45 - Public Sector Labor Markets
• J46 - Informal Labor Markets
• J48 - Public Policy
• J49 - Other
• Browse content in J5 - Labor-Management Relations, Trade Unions, and Collective Bargaining
• J50 - General
• J51 - Trade Unions: Objectives, Structure, and Effects
• J52 - Dispute Resolution: Strikes, Arbitration, and Mediation; Collective Bargaining
• J53 - Labor-Management Relations; Industrial Jurisprudence
• J54 - Producer Cooperatives; Labor Managed Firms; Employee Ownership
• J58 - Public Policy
• Browse content in J6 - Mobility, Unemployment, Vacancies, and Immigrant Workers
• J60 - General
• J61 - Geographic Labor Mobility; Immigrant Workers
• J62 - Job, Occupational, and Intergenerational Mobility
• J63 - Turnover; Vacancies; Layoffs
• J64 - Unemployment: Models, Duration, Incidence, and Job Search
• J65 - Unemployment Insurance; Severance Pay; Plant Closings
• J68 - Public Policy
• J69 - Other
• Browse content in J7 - Labor Discrimination
• J71 - Discrimination
• J78 - Public Policy
• Browse content in J8 - Labor Standards: National and International
• J80 - General
• J81 - Working Conditions
• J83 - Workers' Rights
• J88 - Public Policy
• Browse content in K - Law and Economics
• Browse content in K0 - General
• K00 - General
• Browse content in K1 - Basic Areas of Law
• K11 - Property Law
• K12 - Contract Law
• K13 - Tort Law and Product Liability; Forensic Economics
• Browse content in K2 - Regulation and Business Law
• K20 - General
• K21 - Antitrust Law
• K22 - Business and Securities Law
• K23 - Regulated Industries and Administrative Law
• K25 - Real Estate Law
• Browse content in K3 - Other Substantive Areas of Law
• K31 - Labor Law
• K39 - Other
• Browse content in K4 - Legal Procedure, the Legal System, and Illegal Behavior
• K40 - General
• K41 - Litigation Process
• K42 - Illegal Behavior and the Enforcement of Law
• Browse content in L - Industrial Organization
• Browse content in L0 - General
• L00 - General
• Browse content in L1 - Market Structure, Firm Strategy, and Market Performance
• L10 - General
• L11 - Production, Pricing, and Market Structure; Size Distribution of Firms
• L12 - Monopoly; Monopolization Strategies
• L13 - Oligopoly and Other Imperfect Markets
• L14 - Transactional Relationships; Contracts and Reputation; Networks
• L16 - Industrial Organization and Macroeconomics: Industrial Structure and Structural Change; Industrial Price Indices
• Browse content in L2 - Firm Objectives, Organization, and Behavior
• L20 - General
• L21 - Business Objectives of the Firm
• L22 - Firm Organization and Market Structure
• L23 - Organization of Production
• L24 - Contracting Out; Joint Ventures; Technology Licensing
• L25 - Firm Performance: Size, Diversification, and Scope
• L26 - Entrepreneurship
• L29 - Other
• Browse content in L3 - Nonprofit Organizations and Public Enterprise
• L30 - General
• L31 - Nonprofit Institutions; NGOs; Social Entrepreneurship
• L32 - Public Enterprises; Public-Private Enterprises
• L33 - Comparison of Public and Private Enterprises and Nonprofit Institutions; Privatization; Contracting Out
• L39 - Other
• Browse content in L4 - Antitrust Issues and Policies
• L40 - General
• L41 - Monopolization; Horizontal Anticompetitive Practices
• L44 - Antitrust Policy and Public Enterprises, Nonprofit Institutions, and Professional Organizations
• Browse content in L5 - Regulation and Industrial Policy
• L50 - General
• L52 - Industrial Policy; Sectoral Planning Methods
• Browse content in L6 - Industry Studies: Manufacturing
• L60 - General
• L61 - Metals and Metal Products; Cement; Glass; Ceramics
• L66 - Food; Beverages; Cosmetics; Tobacco; Wine and Spirits
• L67 - Other Consumer Nondurables: Clothing, Textiles, Shoes, and Leather Goods; Household Goods; Sports Equipment
• Browse content in L7 - Industry Studies: Primary Products and Construction
• L78 - Government Policy
• Browse content in L8 - Industry Studies: Services
• L80 - General
• L82 - Entertainment; Media
• Browse content in L9 - Industry Studies: Transportation and Utilities
• L97 - Utilities: General
• L98 - Government Policy
• Browse content in M - Business Administration and Business Economics; Marketing; Accounting; Personnel Economics
• Browse content in M0 - General
• M00 - General
• Browse content in M1 - Business Administration
• M10 - General
• M12 - Personnel Management; Executives; Executive Compensation
• M13 - New Firms; Startups
• M16 - International Business Administration
• Browse content in M2 - Business Economics
• M21 - Business Economics
• Browse content in M3 - Marketing and Advertising
• M37 - Advertising
• Browse content in M4 - Accounting and Auditing
• M41 - Accounting
• M49 - Other
• Browse content in M5 - Personnel Economics
• M51 - Firm Employment Decisions; Promotions
• M52 - Compensation and Compensation Methods and Their Effects
• M54 - Labor Management
• M55 - Labor Contracting Devices
• Browse content in N - Economic History
• Browse content in N0 - General
• N00 - General
• N01 - Development of the Discipline: Historiographical; Sources and Methods
• Browse content in N1 - Macroeconomics and Monetary Economics; Industrial Structure; Growth; Fluctuations
• N10 - General, International, or Comparative
• N11 - U.S.; Canada: Pre-1913
• N12 - U.S.; Canada: 1913-
• N13 - Europe: Pre-1913
• N14 - Europe: 1913-
• N15 - Asia including Middle East
• N17 - Africa; Oceania
• Browse content in N2 - Financial Markets and Institutions
• N20 - General, International, or Comparative
• N23 - Europe: Pre-1913
• N24 - Europe: 1913-
• N25 - Asia including Middle East
• N26 - Latin America; Caribbean
• Browse content in N3 - Labor and Consumers, Demography, Education, Health, Welfare, Income, Wealth, Religion, and Philanthropy
• N30 - General, International, or Comparative
• N32 - U.S.; Canada: 1913-
• N34 - Europe: 1913-
• Browse content in N4 - Government, War, Law, International Relations, and Regulation
• N43 - Europe: Pre-1913
• Browse content in N5 - Agriculture, Natural Resources, Environment, and Extractive Industries
• N50 - General, International, or Comparative
• N51 - U.S.; Canada: Pre-1913
• N52 - U.S.; Canada: 1913-
• N55 - Asia including Middle East
• N7 - Transport, Trade, Energy, Technology, and Other Services
• Browse content in N8 - Micro-Business History
• N80 - General, International, or Comparative
• Browse content in O - Economic Development, Innovation, Technological Change, and Growth
• Browse content in O1 - Economic Development
• O10 - General
• O11 - Macroeconomic Analyses of Economic Development
• O12 - Microeconomic Analyses of Economic Development
• O13 - Agriculture; Natural Resources; Energy; Environment; Other Primary Products
• O14 - Industrialization; Manufacturing and Service Industries; Choice of Technology
• O15 - Human Resources; Human Development; Income Distribution; Migration
• O16 - Financial Markets; Saving and Capital Investment; Corporate Finance and Governance
• O17 - Formal and Informal Sectors; Shadow Economy; Institutional Arrangements
• O18 - Urban, Rural, Regional, and Transportation Analysis; Housing; Infrastructure
• O19 - International Linkages to Development; Role of International Organizations
• Browse content in O2 - Development Planning and Policy
• O20 - General
• O23 - Fiscal and Monetary Policy in Development
• O24 - Trade Policy; Factor Movement Policy; Foreign Exchange Policy
• O25 - Industrial Policy
• Browse content in O3 - Innovation; Research and Development; Technological Change; Intellectual Property Rights
• O30 - General
• O31 - Innovation and Invention: Processes and Incentives
• O32 - Management of Technological Innovation and R&D
• O33 - Technological Change: Choices and Consequences; Diffusion Processes
• O34 - Intellectual Property and Intellectual Capital
• O35 - Social Innovation
• O38 - Government Policy
• O39 - Other
• Browse content in O4 - Economic Growth and Aggregate Productivity
• O40 - General
• O41 - One, Two, and Multisector Growth Models
• O43 - Institutions and Growth
• O44 - Environment and Growth
• O47 - Empirical Studies of Economic Growth; Aggregate Productivity; Cross-Country Output Convergence
• Browse content in O5 - Economywide Country Studies
• O50 - General
• O51 - U.S.; Canada
• O52 - Europe
• O53 - Asia including Middle East
• O54 - Latin America; Caribbean
• O55 - Africa
• Browse content in P - Economic Systems
• Browse content in P0 - General
• P00 - General
• Browse content in P1 - Capitalist Systems
• P10 - General
• P11 - Planning, Coordination, and Reform
• P12 - Capitalist Enterprises
• P13 - Cooperative Enterprises
• P14 - Property Rights
• P16 - Political Economy
• P17 - Performance and Prospects
• Browse content in P2 - Socialist Systems and Transitional Economies
• P20 - General
• P21 - Planning, Coordination, and Reform
• P25 - Urban, Rural, and Regional Economics
• Browse content in P3 - Socialist Institutions and Their Transitions
• P30 - General
• P31 - Socialist Enterprises and Their Transitions
• P32 - Collectives; Communes; Agriculture
• P35 - Public Economics
• P36 - Consumer Economics; Health; Education and Training; Welfare, Income, Wealth, and Poverty
• P37 - Legal Institutions; Illegal Behavior
• Browse content in P4 - Other Economic Systems
• P40 - General
• P41 - Planning, Coordination, and Reform
• P46 - Consumer Economics; Health; Education and Training; Welfare, Income, Wealth, and Poverty
• P48 - Political Economy; Legal Institutions; Property Rights; Natural Resources; Energy; Environment; Regional Studies
• Browse content in P5 - Comparative Economic Systems
• P50 - General
• P51 - Comparative Analysis of Economic Systems
• P52 - Comparative Studies of Particular Economies
• Browse content in Q - Agricultural and Natural Resource Economics; Environmental and Ecological Economics
• Browse content in Q0 - General
• Q00 - General
• Q01 - Sustainable Development
• Browse content in Q1 - Agriculture
• Q15 - Land Ownership and Tenure; Land Reform; Land Use; Irrigation; Agriculture and Environment
• Q18 - Agricultural Policy; Food Policy
• Browse content in Q3 - Nonrenewable Resources and Conservation
• Q30 - General
• Browse content in Q4 - Energy
• Q41 - Demand and Supply; Prices
• Q42 - Alternative Energy Sources
• Q48 - Government Policy
• Browse content in Q5 - Environmental Economics
• Q50 - General
• Q54 - Climate; Natural Disasters; Global Warming
• Q56 - Environment and Development; Environment and Trade; Sustainability; Environmental Accounts and Accounting; Environmental Equity; Population Growth
• Q57 - Ecological Economics: Ecosystem Services; Biodiversity Conservation; Bioeconomics; Industrial Ecology
• Browse content in R - Urban, Rural, Regional, Real Estate, and Transportation Economics
• Browse content in R0 - General
• R00 - General
• Browse content in R1 - General Regional Economics
• R10 - General
• R11 - Regional Economic Activity: Growth, Development, Environmental Issues, and Changes
• R12 - Size and Spatial Distributions of Regional Economic Activity
• R15 - Econometric and Input-Output Models; Other Models
• Browse content in R2 - Household Analysis
• R20 - General
• Browse content in R3 - Real Estate Markets, Spatial Production Analysis, and Firm Location
• R30 - General
• R31 - Housing Supply and Markets
• R4 - Transportation Economics
• Browse content in R5 - Regional Government Analysis
• R51 - Finance in Urban and Rural Economies
• R58 - Regional Development Planning and Policy
• Browse content in Y - Miscellaneous Categories
• Browse content in Y1 - Data: Tables and Charts
• Y10 - Data: Tables and Charts
• Browse content in Y3 - Book Reviews (unclassified)
• Y30 - Book Reviews (unclassified)
• Browse content in Y8 - Related Disciplines
• Y80 - Related Disciplines
• Browse content in Z - Other Special Topics
• Browse content in Z0 - General
• Z00 - General
• Browse content in Z1 - Cultural Economics; Economic Sociology; Economic Anthropology
• Z10 - General
• Z11 - Economics of the Arts and Literature
• Z12 - Religion
• Z13 - Economic Sociology; Economic Anthropology; Social and Economic Stratification
• Z18 - Public Policy
• Browse content in Z2 - Sports Economics
• Z29 - Other
• Advance articles
• Editor's Choice
• Author Guidelines
• Submission Site
• Open Access
• About Cambridge Journal of Economics
• About the Cambridge Political Economy Society
• Editorial Board
• Advertising and Corporate Services
• Self-Archiving Policy
• Dispatch Dates
• Terms and Conditions
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

1. introduction, 2. sectoral specificity and deindustrialisation, 3. empirical analysis, 4. discussion and conclusion, appendix 2: technical detail.

• < Previous

Characterising deindustrialisation: An analysis of changes in manufacturing employment and output internationally

• Article contents
• Figures & tables
• Supplementary Data

Fiona Tregenna, Characterising deindustrialisation: An analysis of changes in manufacturing employment and output internationally, Cambridge Journal of Economics , Volume 33, Issue 3, May 2009, Pages 433–466, https://doi.org/10.1093/cje/ben032

• Permissions Icon Permissions

Deindustrialisation is typically conceptualised as a decline in manufacturing as a share of total employment. From a Kaldorian perspective deindustrialisation could have negative implications for long-run growth, given the special growth-pulling properties of manufacturing. However, defining deindustrialisation purely in terms of employment share is conceptually limiting given that some of the Kaldorian processes operate primarily through output rather than employment, as well as blunting empirical analysis by not focussing enough on changes in manufacturing share of gross domestic product (GDP). This study develops a new method using decomposition techniques to analyse changes in manufacturing employment levels and shares in 48 countries over periods of ‘deindustrialisation’. The analysis separates out changes in the levels and shares of employment manufacturing into components associated with changes in the share of manufacturing in GDP, the growth of manufacturing value-added, the labour intensity of manufacturing production and economic growth. The results indicate that in most cases the decline in manufacturing employment is associated primarily with falling labour intensity of manufacturing rather than an overall decline in the size or share of the manufacturing sector. We suggest that deindustrialisation should appropriately be defined in terms of a sustained decline in both the share of manufacturing in total employment and the share of manufacturing in GDP.

Many upper-income countries have undergone deindustrialisation over recent decades, as have some middle-income countries more recently. Deindustrialisation is most commonly defined as a decline in the share of manufacturing in total employment. This article questions the adequacy of this definition, in terms of the properties that are thought to give manufacturing its special role as an engine of growth. The issue of whether deindustrialisation ought to be defined in terms of employment share or also in terms of output shares is important not only conceptually but for policy as well. It has implications for whether countries should be concerned primarily about a fall in the manufacturing employment share, or with a fall in manufacturing share of gross domestic product (GDP), or both, and as to which particular dimension(s) of a decline of manufacturing could dampen long-term growth.

In order to explore these issues empirically, we go deeper into the actual experiences of countries that have ‘deindustrialised’. Many countries have experienced a fall in the share of their employment that is in manufacturing. But this commonality obscures the very different dynamics at work, especially in terms of trends in their manufacturing sector as a whole. We therefore seek to explore these different dynamics, going beyond trends in manufacturing share of employment to also examine countries’ varying experiences in terms of manufacturing output level and share, the labour intensity of manufacturing, GDP growth, and labour productivity.

Section 2 begins by reviewing why sectoral structure matters for growth, and specifically in what ways manufacturing is considered to have special growth-pulling properties and hence why deindustrialisation could be problematic for growth. We discuss whether, and in what ways, it is manufacturing output or manufacturing employment that is relevant for this growth-pulling, an issue that has not been adequately dealt with in the existing literature. The basic relationship between manufacturing output and employment is discussed, with reference to how their trends might diverge.

Section 3 empirically investigates deindustrialisation processes in 48 countries. A new method is proposed using decomposition analysis in order to separate out various components of the changes in level and share of manufacturing employment. Deindustrialisation associated with falling labour intensity in manufacturing is very different from deindustrialisation associated with a poor performance in manufacturing value-added, and the methods developed and applied here allow for an analysis of the different dynamics at work in individual countries as well as a typology of country experiences of deindustrialisation. Section 4 discusses the implications of the results, particularly in terms of distinguishing between different types of ‘deindustrialisations’ and the implications of these, and comes back to the question of what should really be considered as deindustrialisation.

2.1 The sector-specificity of growth

There has traditionally been a strong argument in branches of the heterodox economics literature that there is a sector-specificity in the economic growth process. This implies that a unit of value-added is not necessarily equivalent across sectors, notably in terms of its growth-inducing or growth-enhancing effects. Such an approach can be distinguished from those parts of the growth literature that tend to see growth as sector-neutral (as well as activity-neutral in the traditional Solow-type growth models and some endogenous growth theories, or activity-specific such as in the new endogenous growth theories, which emphasise the importance of research & development and human capital). 1

The classical development economics literature posits a strong relationship between changes in the sectoral composition of an economy and its rate of growth. The intersectoral reallocation of labour from low- to high-productivity activities is seen as central to increases in overall productivity in developing countries. Specifically, industrialisation and the growth of manufacturing is the engine of technical progress and economic growth. This differs from developed countries where technological innovation, rather than changes in the sectoral composition of the economy, is most important for raising aggregate productivity. 2 Further, in the absence of sufficient rates of economic growth, neither technological progress nor productivity-enhancing structural changes in the economy are likely to reduce unemployment.

2.2 The ‘special properties’ of manufacturing

The Kaldorian tradition in the heterodox literature has regarded the manufacturing sector as being imbued with special growth-enhancing characteristics that are not shared by the other sectors (or at least not to the same extent). By Kaldorian we refer to the ‘laws’ that Kaldor advanced as explaining the differences in rates of growth internationally. 3 The first of these laws states that the faster the rate of growth in manufacturing, the faster the rate of growth of the economy as a whole (in a fundamental causal sense associated with rates of productivity). The second law, also known as Verdoorn's Law, is that the growth rate of labour productivity in manufacturing is endogenous to the growth rate of manufacturing output. According to the third law, aggregate productivity growth is positively related with the growth of manufacturing output and employment and negatively related with non-manufacturing employment. 4

In some sense Kaldor's contribution might be regarded as formalising and rationalising the empirical regularities and stylised facts discussed by Kuznets and developed and tested by Chenery and Syrquin. To this Kaldor added an analysis of why manufacturing has such special qualities relevant for growth. The growth-supporting externalities of these characteristics may not be fully reflected in relative prices, and hence a market-based ‘equilibrium’ sectoral structure can be sub-optimal for growth. This implies that an intersectoral shift of employment (or similarly of other resources) may potentially increase aggregate productivity.

Others associated with this type of approach include Hirschman, Verdoorn, Kalecki, Prebisch, Pasinetti and Thirlwall. A view of the manufacturing sector having special properties leads to it being accorded a special place in understanding the causal relationships of the growth process, as well as suggesting that from a policy perspective there needs to be a particular focus on the manufacturing sector. 5

Several key special characteristics typically attributed to the manufacturing sector can be identified. Crucial is the idea that manufacturing growth ‘pulls along’ aggregate economic growth in ways that growth in other sectors of the economy does not.

One dimension of this is Hirschmanian-type (direct and indirect) backward and forward linkages between manufacturing and other sectors of the domestic economy. If these are indeed stronger than for other sectors of the economy, then manufacturing growth can exert a particularly powerful pulling effect on the economy. 6

Another channel through which manufacturing can act as an engine of growth relates to dynamic economies of scale. The presence of dynamic economies of scale in manufacturing would mean that the growth of productivity in manufacturing is higher the higher the growth in manufacturing output. 7 This is related to the notion that ‘learning-by-doing’ is more important in industry than in agriculture or services. Learning-by-doing, innovation and intersectoral linkages thus render overall productivity growth endogenous to growth in dynamic manufacturing sectors. This of course means that expanding the manufacturing sector would raise manufacturing (and non-manufacturing) productivity.

It is also argued that most technological change occurs in the manufacturing sector. Further, much of the technological change that does occur in the rest of the economy is regarded as tending to be diffused out from the manufacturing sector (see cumulative causation), in part through the use of higher productivity manufacturing inputs in the ‘production’ processes of the rest of the economy. These kinds of technological-change externalities are one form of Hirschman-type intersectoral linkages.

Finally, due to issues of import income elasticities and relative tradability, manufacturing is considered critical to alleviating balance of payments constraints that can impose a ‘stop–go’ pattern on developing countries’ growth. This is important for supporting sustained high growth rates, particularly in the absence of a strong primary commodity export sector with stable and favourable terms of trade.

2.3 Deindustrialisation

Deindustrialisation is typically understood in the current literature (see, e.g., Palma, 2008 ; Rowthorn and Coutts, 2004 ; Rowthorn and Ramaswamy, 1997 ) as a decline in the share of manufacturing in a country's total employment. 8

From the perspective that manufacturing has a special role to play in the growth process, deindustrialisation and premature deindustrialisation in particular would be regarded as problematic in terms of the implications for the rates and sustainability of growth. Most advanced economies have experienced deindustrialisation in the past few decades, as have a number of middle-income countries more recently.

Examining international patterns in the share of manufacturing, in 2003 manufacturing accounted for, on average, just 14.9% of total employment or 16.3% when countries are weighted by their size of manufacturing employment. 9 It is difficult to accurately quantify overall global trends in manufacturing employment given the incompleteness of available data, especially for earlier periods. The share of manufacturing in total value added fell in absolute terms by an annualised average of 1.8% over the period 1980–2006 (1.9% when weighted by a country's manufacturing value added) and by 2.7% between 1990 and 2006 (3.8% when weighted). 10 As an annualised percentage change the share of manufacturing in total value added fell by 0.4% per annum between 1980 and 2006 (0.6% when weighted) and by 1.1% between 1990 and 2006 (2.1% when weighted). Of course, these figures do not reveal the considerable heterogeneity internationally—manufacturing is certainly not falling universally in share of employment and even less so in share of GDP—but they do point to an overall relative shrinkage of manufacturing. Services now account for the majority of both employment and GDP in most countries of the world. 11

From a ‘Kuznetsian’ perspective the relative growth of manufacturing might be expected to level off as a natural phase in economic development. Baumol (1967) argues that activities with relatively low scope for cumulative productivity improvements but which maintain their relative share of output—notably services—will tend (under certain assumptions) to absorb an ever-increasing share of employment, and that this will result in a corresponding slowdown in growth. Further, to the extent that labour productivity is indeed higher in manufacturing than in the rest of the economy (as per the Kaldorian argument), this might be expected to lead to a fall in the share of manufacturing employment (contingent of course on any changes in the share of manufacturing in GDP).

In addition to higher rates of productivity growth in manufacturing than in services leading to slower employment growth in manufacturing than in services (even if output were to increase at the same rate), Rowthorn and Coutts (2004) review four explanations of deindustrialisation that have been advanced in the literature. The first of these is specialisation, referring to the domestic outsourcing of activities previously performed in-house in manufacturing to specialised service providers. 12 This results in an apparent decline in manufacturing employment that is actually a ‘statistical artefact’ rather than real. Second, a fall in the relative prices of manufactures (for exogenous reasons) means that they account for a smaller share of consumer expenditure. Thirdly, international trade might negatively affect manufacturing employment in advanced economies by increasing productivity through higher competitive pressures. This cuts low value-added activities or inefficient firms, and tends to replace labour-intensive activities subject to import pressures with less labour-intensive activities producing relatively sophisticated exports. Finally, decreases in the rate of investment will tend to decrease the share of manufacturing (in both employment and GDP), since a disproportionately large share of investment expenditure is accounted for by manufactures. To these explanations Palma (2005) adds Dutch Disease, understood as a country shifting from a ‘manufacturing’ path to a ‘primary commodity’ path (in cases where a country discovered significant natural resources, developed export finance or tourism, or as a result of policy ‘liberalisation’ in middle-income countries).

The levels of manufacturing employment corresponding to particular levels of GDP per capita have fallen over time ( Palma, 2005 ), and there is increasing evidence of ‘premature’ deindustrialisation in middle-income developing countries. Trade liberalisation in particular appears to have accelerated deindustrialisation in a number of emerging economies. This has raised concerns that such economies may not be able to take advantage of the broader benefits of manufacturing growth as much as they could have.

The declines in the share of manufacturing employment in developed economies in the 1980s were much more pronounced than were the declines in manufacturing share of GDP. This stands to reason, as the sources of deindustrialisation reviewed above would tend to affect employment more strongly than output. The causes of deindustrialisation discussed earlier would affect manufacturing output and employment in differing ways.

To the extent that deindustrialisation is a statistical illusion arising from the contracting out of activities to specialised service providers, the fact that these activities are generally more labour-intensive than manufacturing overall means that manufacturing employment would be reduced proportionately more than would manufacturing output. Deindustrialisation associated with productivity growth in manufacturing exceeding that in services would, of course, only have a negative effect on manufacturing employment, not output. International trade as a source of deindustrialisation would reduce employment more than output, both because the activities affected will tend to be more labour-intensive than manufacturing as a whole and also because of trade-induced pressures to increase labour productivity. On the other hand, the ‘consumption’ source of deindustrialisation—that falling relative prices of manufactures reduce total expenditure on manufacturing—would affect manufacturing output rather than employment (in fact it may even increase employment to the extent that falling prices increase actual consumption in ‘volume’ terms). Deindustrialisation associated with a fall in the rate of investment would affect manufacturing output more than employment.

The fact that the declines in manufacturing employment share have generally exceeded those in manufacturing output might partly explain the emphasis on the fall in the share of manufacturing employment in the deindustrialisation literature. The clearer trends in manufacturing employment share may have been considered more conducive for quantitative analysis than the declines in manufacturing output shares, which were not only less pronounced but were found in fewer countries. Further, apparent changes in manufacturing output share are complicated by concurrent changes in relative prices. The fall in the relative price of manufactures might make it difficult to pin down the real decline in manufacturing output, given the limitations of sectoral deflators, and this could be part of the reason for the focus in the literature on changes in manufacturing employment share rather than output share.

In addition, the decline in manufacturing employment was arguably more acutely felt in a broader sense in advanced capitalist economies in the 1980s than was the decline in manufacturing output. The shedding of manufacturing jobs, and the apparent inability of the rest of the economy to absorb those who lost their jobs, made this an important political and social issue. The relative visibility and ‘political’ nature of manufacturing job losses may have contributed to the focus on this dimension of deindustrialisation. In middle-income countries experiencing premature deindustrialisation, on the other hand, the loss of relatively sophisticated, demand-dynamic manufacturing activities previously protected, may show up as much in a decline in manufacturing output as in employment.

2.4 What matters for growth: manufacturing output or employment?

As discussed earlier, the current literature generally conceptualises deindustrialisation as a decline in the share of manufacturing in total employment. Approaching the issue strictly in these terms would give no explicit place to changes in the share of manufacturing in GDP. Two countries that experienced an equivalent decline in the share of manufacturing employment, but where the share of manufacturing in GDP fell in one and rose in the other, could be regarded as having experienced a similar degree of deindustrialisation based on a definition framed exclusively in terms of employment share. However, there would be very different dynamics at work, arguably with different implications for growth.

A falling share of manufacturing employment may be of concern for various economic, social and political reasons, as discussed elsewhere in this article. But in terms of growth specifically, does it matter what happens to the share of manufacturing in GDP, or is it only the share of manufacturing in employment that has growth implications? The Kaldorian processes in which manufacturing is of particular importance for growth operate through both employment and output channels. We now theoretically evaluate the ‘growth-pulling’ properties of manufacturing in terms of whether it is the share/growth of manufacturing output or of employment that is most relevant.

Firstly, the growth-pulling effects of manufacturing through backward and forward linkages with the rest of the domestic economy are related more to the share of manufacturing in GDP and the growth of manufacturing output, than to its share of employment or growth in manufacturing employment. Even if manufacturing's share of employment is shrinking, if the sector as a whole is growing then this will ceteris paribus give rise to higher demand for inputs from backward-linked sectors as well as providing stimulus and potentially lower input costs to forward-linked sectors.

Second, manufacturing might also pull growth through Keynesian-type demand multiplier effects, through wages paid. In this respect it would clearly be manufacturing employment, rather than output per se , that would be relevant. However, this can be considered a ‘special property’ of manufacturing only insofar as manufacturing wages are higher than those in other sectors of the economy. A positive differential between manufacturing wages and average wages in the domestic economy—as is typically found empirically—can give it particular growth-pulling properties, especially in an (effective) demand-constrained economy. The share of manufacturing in total employment is thus central in this particular regard. 13

Third, dynamic economies of scale would operate through both output and employment. Output and employment are both relevant to learning-by-doing. In terms of employment, on-the-job learning by workers means that the scale of employment in manufacturing affects the strength of the contribution of manufacturing to productivity and growth through this channel. Manufacturing jobs generally tend to require and to develop higher levels of skill than jobs in other sectors. However, manufacturing output is also relevant, as learning-by-doing applies not only at the level of individual workers but also in terms of management and the planning of production and technology. Further, it is the ‘replicability’ of manufacturing production processes that is one of the distinguishing features of manufacturing from agriculture or most services. These (static) increasing returns to scale are effective through the output of manufacturing.

Both output and employment are germane to the broader endogeneity of manufacturing productivity growth to manufacturing output growth. Learning-by-doing is one channel of this productivity endogeneity. Nevertheless, the conceptualisation of productivity growth as a function of output growth (as in the specification of Verdoorn's Law) suggests that it is primarily the growth in manufacturing output (as opposed to employment) that is most important for this dimension of dynamic economies of scale.

Another of the ‘special properties’ of manufacturing for growth discussed above is in terms of technological change and innovation. This relates both to productivity-raising technological change in the manufacturing sector itself, and diffusion from manufacturing out to the rest of the economy. It would seem that it is the growth of manufacturing output that is more relevant to this than the growth of manufacturing employment.

A final quality of manufacturing regarded as being important for overall growth is in terms of alleviating balance of payments constraints and freeing economies (developing economies in particular) from a ‘stop–go’ pattern of growth. It is the output of manufacturing that is most relevant to its net balance of payments position. Even a decline in the share (or level) of manufacturing employment would not be directly relevant to this. 14

This assessment of the relevance of manufacturing output and employment to the channels through which manufacturing can raise overall growth suggests that both output and employment are important. 15 The relative importance of each for an individual country is ultimately an empirical issue, contingent on the binding constraints faced by a particular economy at a particular time. However, it does seem that in general the growth of manufacturing output is at least as important as manufacturing employment. This strongly suggests that it is inadequate to focus exclusively on changes in manufacturing's share of employment.

Defining deindustrialisation as a fall in the share of manufacturing in employment is narrow as it neglects trends in the level or share of manufacturing output. This could give rise to misleading policy interpretations. For instance, a case where the share of manufacturing employment falls despite healthy growth in manufacturing output and a rising share of manufacturing in GDP, would not necessarily give rise to the negative consequences for growth typically associated with ‘deindustrialisation’. Such a trend may well be of concern for other reasons—especially in terms of manufacturing employment in its own right—but would not necessarily undermine the growth-pulling capacity of manufacturing or depress long-term growth.

2.5 Changes in manufacturing output and employment

If changes in manufacturing output and employment were monotonically related as well as being of similar magnitudes, the above discussion as to the relative importance of each for the growth-pulling properties of manufacturing would not be of much practical import. But empirically, changes in the levels or share of manufacturing output and employment can be not only of very different magnitudes but even in different directions, as will be seen in the empirical analysis that follows.

Disparate trends in the shares of manufacturing in total employment and in GDP can be understood in terms of changes in the labour intensity of production. 16 An expanding manufacturing sector could show declining levels of employment if falling labour intensity outweighs the growth in the sector. Similarly, the share of manufacturing in GDP could rise concurrently with a fall in the share of manufacturing in total employment if changes in manufacturing labour intensity exceed those in the rest of the economy by a sufficient magnitude to outweigh the increase in manufacturing's share of GDP. 17

Broadly speaking, the labour intensity of manufacturing (whether in terms of absolute trend or trend relative to the rest of the economy) can change through compositional changes in the manufacturing sector and/or through technological changes within manufacturing. In terms of the first, if the composition of manufacturing changes in favour of the relatively less labour-intensive sub-sectors of manufacturing, this will prima facie result in a lower labour intensity of manufacturing, and manufacturing employment growth below manufacturing output growth (or even negative manufacturing employment growth in conjunction with positive manufacturing output growth). In terms of the second, technological change can result in less labour being employed per unit of output. Causal factors behind such a shift might include exogenous increases in labour productivity, changing relative factor costs, import penetration, changes in workplace organisation, class struggle and labour-displacing technological advances.

Whatever the underlying economic reasons, a fall in manufacturing labour intensity can thus account for falling numbers of people being employed in manufacturing even in the face of an expansion of manufacturing output, and a fall in manufacturing labour intensity greater than that in the aggregate economy can account for a declining share of manufacturing employment even if manufacturing is increasing its share of GDP.

It is these separate processes that the analysis in this article seeks to distinguish. That is, to understand the extent to which a fall in the share of manufacturing in total employment can be accounted for by a shrinking of the manufacturing sector as a whole on the one hand and, on the other hand, by changes in the labour intensity of manufacturing. This is important for understanding the precise character of ‘deindustrialisation’ processes in specific countries over specific periods, the implications of these changes for the rates and sustainability of growth, as well as what ‘deindustrialisation’ should really refer to. Section 3 thus develops a method for decomposing these changes, and applies it to the analysis of ‘deindustrialisation’ experiences in various countries.

2.6 Korea versus the UK

Before proceeding to the decomposition analysis, a simple comparison of two countries illustrates the importance of the conceptual distinctions discussed above. The share of manufacturing in total employment has fallen in Korea and the UK at almost exactly the same rate over the periods of study in this article. In the UK, manufacturing employment as a share of total employment fell from 27.7% in 1980 to 14.9% in 2003, a decline of 2.66% per annum. In Korea, manufacturing employment fell from 27.8% of the total in 1989 to 19% in 2003, an annualised decline of 2.68%. 18 The similarity of these rates of decline is startling.

Of course, several important caveats need to be noted in this regard. The decline in the share of manufacturing employment in the UK began well before 1980, and hence the period of study in this article excludes a significant part of its deindustrialisation. The decrease has thus been over a much longer period in the UK, even though at the same rate as in Korea over the period of this analysis. Further, Korea began from a higher share in 1989 than the UK in 1989, and ended at a higher share. The cases are not directly comparable in a broader sense, given that the countries are at different stages of economic development (UK GDP per capita being almost double that of Korea). Notwithstanding these essential caveats, the point remains that if deindustrialisation is defined exclusively in terms of manufacturing share of employment, it could be said that the two countries experienced an equivalent degree of ‘deindustrialisation’ over the relevant time periods.

This is highly counterintuitive, and calls into question the value of a unidimensional definition of deindustrialisation. This is borne out with reference to the divergent performances in manufacturing output of the two countries. Manufacturing GDP in Korea grew by 7.5% per annum (1989–2003), whereas that in the UK shrank by 1.3% per annum (1980–2003). Similarly with manufacturing as a share of GDP: this increased by 1.4% per annum in Korea, while falling by 1.2% per annum in the UK. It is these output figures which tell the story of the deindustrialisation in the UK and the continuing robust performance of manufacturing in Korea, rather than the common trends in manufacturing share of employment.

This is not to suggest that the decline in the share of manufacturing employment in Korea is unimportant or is not a matter of concern. As discussed earlier, some of the Kaldorian-type channels of manufacturing growth-pulling are realised through manufacturing employment. This suggests that a fall in the share of manufacturing in total Korean employment may have negative repercussions for the sustainability of growth in that economy, despite the continued growth of manufacturing output in both level and share of GDP.

The charts displayed in Figure 1 summarise trends in manufacturing in the UK and Korea. The top two panels show trends in manufacturing GDP and manufacturing employment levels, while the other three pairs of panels show manufacturing shares of GDP and total employment. The stark difference between the UK and Korea is in terms of manufacturing GDP/value-added rather than manufacturing employment, particularly when considered as shares of total GDP and total employment, respectively (see the lower three pairs of panels).

Comparison of employment and value added trends in UK and Korea

Notes: All charts in 3-year moving averages Y-axes do not begin at 0. Scales are identical for UK and Korea charts for ease of comparison. The top panels (manufacturing GDP and employment) are shown in a natural log scale.

This study analyses countries during periods of (relative) deindustrialisation as defined in the existing literature, that is, when they experienced a fall in the share of manufacturing in total employment. The stylised facts of economic development suggest that this will be mainly the relatively advanced countries. Further, lack of data excludes many of the less developed countries, even if they did experience a falling share of manufacturing employment. As the coverage and quality of available data improves, it will be possible to apply the methodology developed in this article more widely.

The sample comprises 48 countries, of which 25 are currently classified by the World Bank as high-income OECD (Organisation for Economic Cooperation and Development), nine as high-income non-OECD, nine as upper-middle income, three as lower-middle income and two as low-income.   Appendix 1 lists the countries, the time periods analysed for each country and the changes in the number of manufacturing jobs and in the share of manufacturing in total employment for that period.

The time periods are generally taken from the high point of manufacturing employment share prior to a sustained decline (although data is not available for the pre-1980 period). In isolated cases (such as Pakistan) a declining share of manufacturing employment over a significant period of time has since been followed by a rising share; nevertheless, the period in which the share declined is still included in this study.

Value-added and GDP data are taken from the UN national accounting data. The dollar-denominated measures of manufacturing value-added and GDP in 1990 constant prices are used. The employment data is from the International Labour Organisation (ILO) Key Indicators of the Labour Market (KILM) database, specifically from the KILM 4 series, which disaggregate employment by sector.

In constructing the employment dataset each country was examined individually, in order to assess whether it met the criteria for inclusion in the sample (a sustained decline in the share of manufacturing employment); to determine the appropriate period for inclusion in the sample; and to check the continuity of the data series. Some series in the KILM database include breaks, arising from a change in methodology, scope of coverage, type of source or repository. Data series used in this analysis generally do not traverse breaks in the series, unless scrutiny of the data suggests no significant shift in level or trend. In some cases no break is noted in the database, yet examination of the data suggests that there was in fact an unrecorded change in methodology (for example an unexplained large jump in the series), and such cases were treated as a break and excluded from this analysis. The tendency has thus been to err on the side of caution, even though this means excluding some countries that did indeed undergo deindustrialisation, and shortening the periods of analysis in other instances. 19 Even so, employment data generally has wide confidence intervals, particularly at the sectoral level, and the results derived using this data should thus be treated as indicative rather than precise. The focus is on the sign of changes and relative magnitudes over periods of time, rather than on exact values.

3.2 Summary of manufacturing output and employment trends

Table 1 summarises the manufacturing performance of the sample countries, in terms of changes in the levels and shares of manufacturing employment during the relevant periods (as listed in   Appendix 1 ). Since only countries whose share of manufacturing employment declined are included in the study, the dimension of manufacturing employment share is extraneous to the table.

Typology of changes in manufacturing

Of the 48 countries, the level of manufacturing employment rose in six despite the share falling. Of the other 42 countries, in which both the share and level of manufacturing employment fell (the rightmost column of the table), in most (31 countries) the manufacturing sector grew in real terms, and in 11 of these also increased as a share of GDP.

Eleven countries had an all-round decline of their manufacturing sectors: manufacturing shrank in real terms and as a share of GDP, and manufacturing employment also fell as well as declining as a share of total employment—no debate is needed as to whether these cases can be classified as deindustrialisation. The latter group are essentially developing and ‘transition’ economies. In fact, most of them are fairly clear cases of ‘premature’ deindustrialisation.

These dynamics are investigated further in the analysis set out below, in which we use decomposition techniques to separate out the various components of the changes in manufacturing employment. The first two decompositions analyse changes in the level of manufacturing employment, while the third one looks at changes in the share of manufacturing in total employment. 20

3.3 First decomposition

Firstly, we undertake a two-way decomposition on changes in the level of manufacturing employment. This separates out changes in the value-added of the sector from changes in the labour intensity of that sector. The object is to understand how much of each country's decline in manufacturing employment is associated with changes in the overall size of manufacturing, and how much with changes in the labour intensity of that production.

The separation of these two vectors is useful in distinguishing different types of ‘deindustrialisation’. For instance, a given fall in manufacturing employment could be associated with either a falling labour intensity of production, or with a shrinkage of manufacturing as a whole (or of course with a combination of these factors). These two processes would be very different, even if associated with the same change in manufacturing employment.

Deindustrialisation associated predominantly (at least in this ‘accounting’ sense) with one or other of the two effects could have different causes and implications. A loss of manufacturing jobs associated primarily with the sector growth effect might suggest—in a simplistic way of course—that the issue is primarily one of the manufacturing sector as a whole and its lack of dynamism. On the other hand, a loss of manufacturing jobs associated primarily with the labour intensity effect might suggest that the manufacturing sector as a whole is not necessarily in decline, but that the ‘problem’ pertains more to its labour-absorbing capacity.

Results from first decomposition analysis.

For example, the coordinates of the USA (–75.8, 58.5) indicate that the fall in manufacturing labour intensity accounted (hypothetically) for a 75.8% fall in the level of its manufacturing employment, while the growth of manufacturing value-added accounted (hypothetically) for a 58.5% increase in manufacturing employment. The sum of these two effects, at –17.3%, is the actual percentage change in manufacturing employment in the USA, as shown in   Appendix 1 .

The position of a point in the North-East quadrant would indicate that both the labour intensity and sector growth effects were positive, such that the change in the labour intensity of manufacturing and the growth in manufacturing value-added each accounted for manufacturing employment creation, with unambiguously positive manufacturing employment growth. Conversely, the location of a point in the South-West quadrant would indicate that both effects were negative for the relevant period; in other words, the change in labour intensity and the change in manufacturing sector size each accounted for negative manufacturing employment growth, with an unambiguously negative change in manufacturing employment. A point in the North-West quadrant would indicate a negative labour intensity effect and a positive sector growth effect, that is, the drop in labour intensity accounted for a negative change in manufacturing employment while the sector growth accounted for a positive change in manufacturing employment. A point in this quadrant lying above the dashed diagonal line (along which the labour intensity effect and the sector growth effect are of equal magnitude but opposite sign) would indicate that the positive sector growth effect outweighed the negative labour intensity effect, hence net manufacturing job creation; and below the line the reverse, and thus net manufacturing job loss. Finally, a point in the South-East quadrant would indicate a positive labour intensity effect and a negative sector growth effect, denoting a case where the rise in labour intensity accounted for a positive change in manufacturing employment while the change in sector size accounted for a negative change in manufacturing employment. A point in the quadrant above the dashed diagonal line would have had net manufacturing employment creation, with the positive labour intensity outweighing the negative sector growth effect, with a point below the dashed line having had net manufacturing employment loss. Overall, any point falling below/to the left of the dashed diagonal line shows manufacturing employment loss for that country, while any point above/to the right of the dashed line shows manufacturing employment growth.

Several insights can be noted from these results. Firstly, manufacturing employment increased in just six countries (Cyprus, Israel, Netherlands, San Marino, Spain and Venezuela), despite a fall in the share of manufacturing in total employment. These countries are located to the right of the dashed diagonal line. However, these increases are relatively small, as can be seen from their proximity to the line.

Second, there is a clustering of countries in the North-West quadrant, that is, with a negative labour intensity effect and a positive sector growth effect. For these countries, the manufacturing sector grew, yet became less labour-intensive. Third, almost all of the latter group of countries fall below the diagonal line. The potential increase in manufacturing employment associated with the positive sector growth effect was outweighed by the negative labour intensity effect. Countries in this quadrant thus generally experienced a fall in the level of manufacturing employment. In those few countries in the quadrant whose manufacturing employment increased—Cyprus, Israel, Netherlands and Spain—the increases are of a small magnitude (as can be seen by their proximity to the diagonal line).

Fourth, no countries experienced both growth in the size of their manufacturing sector and an increase in its labour intensity (note the absence of any points in the North-East quadrant). 24 In the handful of countries in which manufacturing became more labour-intensive (such as Russia and Venezuela), manufacturing value-added shrank more than did manufacturing employment.

Fifth, in 13 countries the manufacturing sector actually shrank in real terms: Argentina, Barbados, Hong Kong, Jamaica, Latvia, Macao, Romania, Russia, Saint Lucia, San Marino, Suriname, Uruguay and Venezuela. These countries are located below the x -axis. This group of countries includes most of the developing countries in the sample, as well as the ‘transition’ countries of Eastern Europe/ex-USSR. These are the countries whose manufacturing sector declined in absolute terms. (Conversely, the 35 other countries of the sample experienced real growth in manufacturing output.)

3.4 Second decomposition analysis

The first decomposition analysis separated out changes in the level of manufacturing employment into changes in manufacturing output and in the labour intensity of that output. In a second decomposition analysis, we continue to analyse changes in the level of manufacturing employment, but are now interested in changes in the share of manufacturing in GDP, rather than the level of manufacturing output as in the first decomposition, as well as in changes in labour intensity. A fast-growing country could show growth in the number of manufacturing jobs even while the share of manufacturing in GDP declines and manufacturing becomes less labour-intensive, if the economy is growing fast enough.

Labour intensity effect =  1 6 ( ϕ i j t − ϕ i j t − h ) { ( δ i j t − h Q j t − h + δ i j t Q j t ) + ( Q j t − h + Q j t ) ( δ i j t − h + δ i j t ) }

Sector share effect =  1 6 ( δ i j t − δ i j t − h ) { ( ϕ i j t − h Q j t − h + ϕ i j t Q j t ) + ( Q j t − h + Q j t ) ( ϕ i j t − h + ϕ i j t ) }

Economic growth effect =  1 6 ( Q i j t − Q i j t − h ) { ( ϕ i j t − h δ j t − h + ϕ i j t δ j t ) + ( δ j t − h + δ j t ) ( ϕ i j t − h + ϕ i j t ) }

The ‘labour intensity effect’ 26 is the change in the level of employment in a sector associated with the changing labour intensity of that sector. For example, there could be a decline in manufacturing employment associated with a falling labour intensity of manufacturing production. The ‘sector share effect’ measures the change in employment associated with a change in the sector's share in GDP. For example, there could be a decline in manufacturing employment associated with a decline in the share of manufacturing in the overall economy. Third, the ‘economic growth effect’ is the change in employment in a sector associated (in a simple mechanical way) with the overall growth of the economy. So long as there is any economic growth, this effect would be positive for all sectors. The sum of these three effects is the change in employment in a sector over that period.

Results from second decomposition analysis.

The coordinates of the UK (–65.7, –23.4), by way of example, indicate that the fall in manufacturing labour intensity in that country accounted for a (hypothetical) decrease of 65.7% in manufacturing employment, while the decline in the share of manufacturing in GDP accounted for a (hypothetical) decrease of 23.4% in manufacturing employment. Note that the overall change in countries’ manufacturing employment share cannot be directly read off this chart (in the way that the net change in manufacturing employment could be seen in Figure 2 from the first decomposition), as the third component of this decomposition (the economic growth effect) is not included in this chart. In the case of the UK, economic growth accounted for a (hypothetical) increase of 49.8%. The net change in manufacturing employment in the UK was –39.3% (which corresponds with the figure shown in the summary of country trends in   Appendix 1 ).

Most countries fall in the South-West quadrant: declining labour intensity of manufacturing as well as a decrease in manufacturing share of GDP, both associated with falls in the level of manufacturing employment. For countries in this quadrant, which, nevertheless, experienced manufacturing employment growth, this can be accounted for entirely by overall economic growth (i.e. manufacturing was a declining share of a growing economy).

Manufacturing in countries located in the North-West quadrant also became less labour intensive but grew as a share of GDP. Interestingly, these countries generally tend to be somewhat better performers in the sample in terms of overall GDP growth.

Figure 4 shows the results of this second decomposition in more detail, by also including the economic growth effect as well as giving a sense of the relative contributions of the labour intensity and sector share effects.

Full results of second decomposition.

Note: Countries are arranged in ascending order of 2005 per capita $GDP.

The vertical axis shows the percentage change in the level of manufacturing employment in each country. For each country, each of the three bars thus shows the percentage change in manufacturing employment that can be accounted for by the labour intensity effect, the sector share effect and the economic growth effect. The sum of the three bars gives the net percentage change in the level of manufacturing employment in each country (corresponding to the middle column of figures in   Appendix 1 ). For example, in Pakistan the normalised labour intensity effect, sector share effect and economic growth effect are –94.9%, 8.3% and 72.7%, respectively, summing to –13.9%, which is the net decline in manufacturing employment in that country. In this case the fall in the level of manufacturing employment is entirely accounted for by the fact that manufacturing became less labour-intensive.

The labour intensity effect is negative for almost all countries (the exceptions being Venezuela, Russia, Saint Lucia, Macao and San Marino; only the first two of these are of real empirical significance). The sector share effect is somewhat more mixed, but is negative in 37 of the 48 countries. The labour intensity effect exceeds (in absolute terms) the sector share effects for most countries. In other words, for most countries the falling labour intensity of manufacturing can primarily account for falls in manufacturing employment or, in cases where manufacturing employment increased, for these increases not being higher than they actually were. The economic growth effect is positive for all but three countries (Russia, Latvia and Romania). Of course, even where the economic growth effect is positive, had growth been higher than it was this might have led to better performance in manufacturing employment.

3.5 Third decomposition analysis

The two decomposition exercises undertaken above analyse changes in the level of manufacturing employment. The first decomposition disaggregated these changes into those associated with changes in manufacturing value added and manufacturing labour intensity, respectively; the second was a three-way decomposition into changes associated with manufacturing labour intensity, manufacturing GDP share and overall economic growth, respectively. Now, in the third and final decomposition we analyse changes in the share of manufacturing in total employment. This is particularly relevant as deindustrialisation is conventionally defined in terms of a decline in the share of manufacturing employment.

This allows for a separation of changes in the sectoral share of employment into components associated with changes in sectoral labour intensity, sectoral share of GDP and economy-wide labour productivity, respectively, as follows: 28 Figure 5 shows the labour intensity and sector share effects of the sample countries. We focus initially on the labour intensity and sector share effects, as these are of principal interest to this study.

Results from third decomposition analysis.

The only quadrant in which no countries fall is the North-East quadrant: none had both positive labour intensity and sector share effects. In other words, in none of the 48 sample countries did manufacturing increase as a share of GDP and become more labour intensive. 29 There are five countries in which manufacturing became more labour-intensive but declined as a share of GDP; 11 countries in which manufacturing became less labour-intensive but increased as a share of GDP; and in the remaining 32 countries both effects were negative: manufacturing fell as a share of GDP while also becoming less labour intensive. The most extreme of the latter cases is that of Hong Kong. Indeed, Hong Kong experienced the largest absolute fall in manufacturing share of employment as well as the largest proportional decline in the number of manufacturing jobs of all countries studied.

In Section 2.6 the performance of manufacturing in Korea and the UK were compared. This discussion highlighted that the primary difference between them lay in trends in manufacturing output (in terms of both level and share of GDP), whereas the falls in the share of manufacturing employment were of a similar magnitude. Figure 5 further elucidates the difference in their performances. Manufacturing became less labour-intensive in both cases, yet in Korea (located in the North-West quadrant) manufacturing rose as a share of GDP while in the UK (South-West quadrant) not only did manufacturing become less labour-intensive but it also fell as a share of GDP.

Figure 6 shows the full results from the third decomposition, displaying all three components of the change in manufacturing employment share (that is, the changes in manufacturing employment share associated with changes in manufacturing labour intensity, manufacturing GDP share and economy-wide labour productivity, respectively). The sum of these is, of course, the net change in manufacturing employment share for each country. To take the case of Korea, for example, the (hypothetical) fall in the manufacturing employment share associated with declining labour intensity outweighed the positive sector share effect and the (economy-wide) labour productivity effect; the share of manufacturing in GDP fell by 8.8 points overall.

Full results of third decomposition.

The labour intensity effect was negative in all but five cases, and these were all countries experiencing serious economic problems. The sector share effect was positive in 11 countries—generally relatively better performing countries (such as Korea and Ireland). The economy-wide labour productivity effect accounted for a positive component of the change in manufacturing employment share in 42 of the 48 countries.

3.6 Summary of results and a typology of ‘deindustrialisations’

Table 2 consolidates the basic results of all three decompositions, abstracting from the relative magnitudes of the various components and focussing on the signs of each component by country. The labour intensity effect, shown in the first column, was calculated in all three decompositions, and although the magnitude of this component varies across the decompositions the sign of the effect is consistent for each country. The sector growth effect was calculated in the first decomposition. The sector share effect was derived from the second and third decompositions (with the sign of this component being consistent, for each country, across both decompositions). The economic growth effect was calculated in the second decomposition, and the labour productivity effect in the third decomposition; both of these are economy-wide variables. We classify countries into four broad categories as discussed below, based on the results from all three decompositions. Within each category, countries are shown ordered in terms of their income grouping.

Summary of signs of components from decompositions 1–3

In the first set of countries, from Austria to Pakistan, the labour intensity effects, sector growth effect, sector share effects, economic growth effect and labour productivity effect are all positive. These are, in fact, the only countries for which the sector share effects are positive (i.e., in which manufacturing grew as a share of GDP). The declining shares of manufacturing in the total employment of these 11 countries are primarily accounted for by the fact that their manufacturing became less labour-intensive (that is, labour productivity rose in manufacturing).

The second set of countries, from Australia to Mongolia, differs from the first set in that the sector share effect is negative. The decrease in the share of manufacturing in total employment in these countries is accounted for both by the negative labour intensity effect and by the negative sector share effect. Half of the countries in the sample fall within this category.

In the third set of countries, from Barbados to Suriname, not only is the sector share effect negative (as in the second category), but the sector growth effect is also negative (i.e. manufacturing value added shrank in real terms). The economic growth and labour productivity effects are positive for most, but not all, of these countries, unlike the first and second sets of countries for which these two effects are uniformly positive. Note that six of these countries are developing countries and the other two are East European.

Finally, the fourth set of countries are the only ones with positive labour intensity effects. The fact that manufacturing share of employment fell in these countries nonetheless is accounted for by the negative sector share effect as well as by a negative economic growth effect or negative labour productivity effect in certain of the cases.

There is, of course, considerable heterogeneity amongst each of these sets of countries, as this typology is based only on the signs of the components of the decompositions. In the earlier discussions we considered the relative magnitude and importance of each component in accounting for changes in the level and share of manufacturing employment in each country. The implications of this broad typology are considered in the next section.

A fall in manufacturing employment (whether level or share) associated primarily with changes in the labour intensity of production is very different from a fall in manufacturing employment associated primarily with a declining level/share of manufacturing output. These two types of falling manufacturing employment are quite different phenomena, likely to have different causes, different implications for growth, and to require different policy interventions should they be deemed undesirable. These distinctions render the analysis of this article important, both conceptually and from a policy perspective.

Of course, a fall in manufacturing share of employment can, in its own right, have a negative effect on growth, as discussed in Section 2.4. Some of the Kaldorian channels of manufacturing growth-pulling operate through employment, and hence a fall in manufacturing employment would be of concern irrespective of the performance of manufacturing output.

If a decrease in manufacturing employment share is primarily accounted for by falling labour intensity of manufacturing (as in the cases of, for example, Finland, Sweden, Switzerland, Ireland or Korea), this calls into question the extent to which ‘deindustrialisation’ is an appropriate characterisation. This is especially relevant in cases (the countries mentioned above and all those in the first set of Table 2 ) where the manufacturing sector is growing in real terms as well as increasing its share of GDP. There could be various underlying economic causes behind falling labour intensity in manufacturing, which might relate to the sub-sectoral composition of manufacturing and/or to processes within sub-sectors (including as a defensive response to cheap manufacturing imports from lower-wage countries). In an ‘optimistic’ scenario, falling labour intensity could essentially amount to exogenous increases in labour productivity, driven by factors such as improved skills or technology. On the other hand, falling labour intensity could be caused by labour-displacing capital intensification. The actual causes of falling labour intensity would vary across countries and time periods, and are not the focus of this study. The point is that a fall in the share of manufacturing employment that is mostly accounted for by falling labour intensity of manufacturing (i.e. increasing labour productivity of manufacturing) would not necessarily have a negative impact on growth. The impact on growth would be contingent on various conjunctural factors, including what the causes of the fall in labour intensity might be.

This is very different from the case (such as in Norway, Colombia, Latvia or Hong Kong) where the fall in the share of manufacturing employment is associated primarily with a decline of the manufacturing sector as a share of GDP (and especially in cases such as the latter two where manufacturing shrank in real terms as well). In such a scenario, an economy would be particularly at risk of losing out on the growth-pulling effects of manufacturing. This could be associated with a diminution of long-term growth prospects (although of course this would be contingent on the country's stage of development, on the nature of the manufacturing sectors in decline, of the sectors whose share of GDP is growing, 30 and so on). This strongly suggests the need to go deeper into the black box of falling share of manufacturing employment—as this article has attempted to do—before the effects on growth can be assumed.

One insight that emerges from this analysis is the significant heterogeneity of experiences that would be characterised as ‘deindustrialisation’ when considered exclusively in terms of the share of manufacturing in total employment (as per the conventional denotation of deindustrialisation). The common denominator of the 48 countries of the sample, over their respective time periods included in this study, is that they experienced a reduction in the share of manufacturing in total employment. Yet amongst these countries the share of manufacturing in GDP grew in some and fell in others; manufacturing value-added increased in some and fell in others; and countries underwent diverse experiences in terms of the labour intensity of manufacturing, economic growth and overall labour productivity.

This heterogeneity points to the difficulty in formulating a generic definition of deindustrialisation. We would argue that a case in which the sector growth effect, sector share effect, economic growth effect and labour productivity effects are all positive and the decline in manufacturing employment level and/or share is accounted for entirely by a negative labour intensity effect should not be characterised as deindustrialisation in any real sense. If labour productivity rises more rapidly in manufacturing than in the rest of the economy—as might be expected if manufacturing does indeed have the Kaldorian properties attributed to it—and if manufacturing does not increase its share of GDP commensurately, then the share of manufacturing in total employment would of course fall.

Yet it does not seem meaningful to characterise such a process as deindustrialisation, particularly when associating deindustrialisation with negative implications for growth. Rather than defining deindustrialisation in terms of the single dimension of falling share of manufacturing in total employment, as in the current literature, we propose that deindustrialisation should be regarded as occurring when there is a sustained decline in both the share of manufacturing in total employment and the share of manufacturing in GDP.

This would mean that the 11 countries grouped in the top category in Table 2 , in which manufacturing actually grew as a share of GDP, would not be regarded as having deindustrialised (at least over the periods analysed in this paper). These countries are: Austria, Belgium, Finland, Ireland, Korea, Sweden, Switzerland, Estonia, Slovenia, Poland and Pakistan. Note that in these countries manufacturing value added also grew (in real terms), as well as economy-wide labour productivity increasing and overall economic growth being positive.

In the other 37 countries of the sample, manufacturing fell as a share of GDP (as well as decreasing in its share of total employment). These countries would meet the criteria of deindustrialisation suggested here. The empirical analysis presented here is helpful in distinguishing between different types of deindustrialisation amongst these countries. For instance, countries in which manufacturing grew in real terms despite falling as a share of GDP (the second set of countries in Table 2 , such as Japan, Denmark and Portugal) need to be distinguished from those in which manufacturing actually shrank in real terms (the third set of countries in Table 2 , which are either developing countries or East European countries).

The methodology developed in this article can also be helpful in distinguishing between ‘positive’ and ‘negative’ deindustrialisation, a distinction originally introduced by Rowthorn and Wells (1987) and which has been discussed further in the subsequent literature. Positive deindustrialisation is regarded as ‘the normal result of sustained economic growth in a fully employed, and already highly developed, economy [which] occurs because productivity growth in the manufacturing sector is so rapid that, despite increasing output, employment in this sector is reduced, either absolutely or as a share of total employment’ ( Rowthorn and Wells, 1987 , p. 5). In positive deindustrialisation any workers displaced from manufacturing find employment in new jobs in the services sector, such that unemployment does not increase. This is contrasted with negative deindustrialisation, which is ‘a product of economic failure and occurs when industry is in severe difficulties…labour shed from the manufacturing sector—because of falling output or rising productivity—will not be reabsorbed into the service sector’ and unemployment will therefore increase. Rowthorn and Wells also identify a third type of deindustrialisation, in which the matter of net exports shifts away from manufactures towards other goods and services, leading to a transfer of labour and resources from manufacturing to other sectors of the economy’ ( Rowthorn and Wells, 1987 , p. 6).

The empirical results obtained through the methodology proposed in this article, considered in conjunction with the overall employment performance of an economy, can shed light on the ‘positive’ or ‘negative’ character of specific deindustrialisation experiences. For instance, the decomposition analysis is suggestive as to the extent to which increases in labour productivity in manufacturing account for deindustrialisation in each country, and the relationship between changes in manufacturing output and employment. The empirical analysis also draws attention to the heterogeneity of experiences among rich countries, and hence the difficulties in using a generic definition of ‘positive deindustrialisation’.

A full analysis of deindustrialisation also needs to take other considerations into account, such as whether any workers who lose their jobs in manufacturing can readily find jobs elsewhere in the economy, and the nature of the activities that are ‘replacing’ manufacturing employment.

Falling manufacturing employment—whether in absolute terms or as a share of total employment—may be regarded as a problem in its own right, over and above any effects on aggregate growth and its sustainability. Loss of manufacturing jobs (in terms of levels) is of obvious concern, particularly if other sectors cannot absorb these workers. A fall in the share of manufacturing employment can also be problematic, even putting aside any implications for growth. There are various reasons as to why manufacturing jobs may be regarded as more desirable than jobs in other sectors of the economy, such that a fall in the share of manufacturing in total employment would be undesirable. Blue-collar manufacturing jobs generally tend to be better paid and to develop higher levels of skills than equivalent jobs in the rest of the economy. Employment security in manufacturing tends to be superior to that in agriculture or services, and there is lower scope for and actual trends towards casualisation, outsourcing and other forms of atypical employment (at least domestically). Manufacturing is also easier to unionise than is agriculture and many services sectors, making manufacturing an important mainstay of trade union organisations.

All of these characteristics make manufacturing employment important, arguably more important than employment in most other (private) sectors of the economy. A decline in the share of manufacturing employment—‘deindustrialisation’ as narrowly defined—is thus of concern in its own right. It may have various negative consequences, particularly in terms of distribution.

A study of the data on sectoral levels and shares obscures the human dimension of these shifts and the upheaval wrought in people's lives. Behind the glibness of economic models or policies, people whose manufacturing jobs are lost may struggle to find new jobs or may never work again, and if they do the jobs may not be at an equivalent level of remuneration or job security. The experience of acute deindustrialisation suggests that a displaced auto or textile worker who does manage to gain new employment is likely to have a relatively insecure, non-unionised, low-paid services job. The long-lasting social problems in regions that have undergone severe deindustrialisation are testimony to the devastation belied by the economic data.

The empirical analysis of this article does not address these distributional and social ramifications of deindustrialisation. The concerns typically associated with deindustrialisation in the economics literature relate principally to the growth-pulling properties of manufacturing. A decline in the share of manufacturing employment remains relevant to these properties, especially in terms of learning-by-doing, and thus may well have deleterious effects on sustainable economic growth. However, the share of manufacturing output in GDP and the growth of manufacturing output are also highly relevant to the growth-pulling properties of manufacturing.

A proper empirical analysis of deindustrialisation thus needs to take into account trends in manufacturing output as well as employment. This study has considered both, attempting to disentangle the components of falling manufacturing employment associated with changes in declines in manufacturing output and in the labour intensity of this output. In most country experiences of ‘deindustrialisation’ analysed here, falling manufacturing employment is accounted for primarily by decreases in the labour intensity of manufacturing, rather than by an overall decline in manufacturing GDP or manufacturing share of total GDP.

I would like to thank Gabriel Palma, Paulo Gala, and two anonymous referees for helpful comments

Country sample and summary data

Weighted by a country's manufacturing employment (mean of the beginning and end values).

This sets out the derivation of the three components of the second and third decompositions.

Second decomposition analysis:

Third decomposition analysis.

Taking the means of each of the three terms from the three alternative formulations, 32

These three components sum exactly to the change in the share of employment in sector in the total employment of country j over the period h .

Google Scholar

Google Preview

See Palma (2005) .

Note also that developing countries may gain some of the benefits of technological innovation in a form embodied in imported machinery.

See Thirlwall (1983, 2003) for an explication of these laws.

While Kaldor originally conceptualised this law as operating through the sucking in of labour from lower-productivity non-manufacturing activities to a growing manufacturing sector, it could also operate through other mechanisms even in the absence of the sucking in of labour to manufacturing. Particularly relevant in this regard is the diffusion of technological change from manufacturing to the rest of the economy, raising aggregate productivity growth.

This article does not seek to investigate the empirical validity of these arguments. There is however an existing corpus of work that has found some empirical support for the validity of Kaldor's laws, using various techniques and testing across a range of countries and time periods. Recent studies include Leon-Ledesma (2000) , Felipe (1998) , Wells and Thirlwall (2003) , Hansen and Zhang (1996) , Atesoglu (1993) , Bernat (1996), Pons-Novell and Viladecans-Marsal (1999) , Harris and Lau (1998) , Harris and Liu (1999) , Jeon (2006) , Bairam (1991) , Diaz Bautista (2003) , Libanio (2006) , Fingleton and McCombie (1998) , and Drakopoulos and Theodossiou (1991) .

See Tregenna (2008A) for an analysis of these linkages in the South African economy.

However, note also that in an open economy, economies of scale may be associated with falling prices, depending in part on demand conditions and also on the degree of concentration in factor and product markets, which affects how much of the gains of productivity growth is transmitted to lower prices and how much is retained in higher wage or profit rates.

A prominent exception to this has been the approach of Singh (1977 , 1987) , who considers deindustrialisation to be problematic insofar as it is a manifestation of structural disequilibrium in the economy, in the sense that manufacturing becomes unable to not only satisfy domestic demand at least cost but also to export enough to pay for full employment level of imports (at a ‘reasonable’ exchange rate).

This includes the 84 countries for which the relevant employment data is available (of which China is not included).

This includes the 206 countries for which the relevant value-added data is available.

All figures in this paragraph from author's calculations based on ILO KILM data for employment and UN national accounts data for value-added and GDP.

See Tregenna (2008B) for an analysis of the extent of outsourcing between manufacturing and services in the South African economy.

Note, however, that in a dynamic sense, the effects of a fall in the share of manufacturing employment on growth through this wage-based demand multiplier channel would be mediated by the endogeneity of the wage differential between manufacturing and the rest of the economy to the decline in manufacturing employment itself. This is both compositional and causal, in opposite directions and with an ambiguous net outcome. The ‘compositional’ effect is in terms of the static difference between wages in the manufacturing jobs being shed and those remaining, and as deindustrialisation tends to affect the more labour intensive, low-productivity, low-wage jobs, average manufacturing wages may increase. On the other hand, job losses and the threat of further losses may put downward pressure on wages in the remaining manufacturing employment, eroding the manufacturing ‘wage premium’.

In fact, a decline in manufacturing output in the absence of a commensurate decline in manufacturing employment could even improve a country's balance of payments position, through a channel of reduced demand for imports arising either from lower absolute levels of employment or the shifting of jobs to lower-paid employment in other sectors such as agriculture or low-wage private services.

Insofar as it is manufacturing output that is relevant for growth-pulling, one might enquire further as to whether it is the ‘quantity’ of manufacturing activities that is important or the total ‘price’ of those activities. Put differently, whether it is trends in manufacturing output in constant or in current prices that is relevant. This varies across the channels discussed above. The growth-pulling properties of manufacturing through intersectoral linkages and to some extent technological change and endogenous productivity growth would not necessarily diminish if the share of manufacturing in GDP fell purely because of falling relative prices while the relative ‘quantity’ of manufacturing activities remained unchanged. The importance of manufacturing in alleviating balance of payments constraints, on the other hand, would be sensitive to changes in the relative prices of manufactures; although this would depend not only on relative prices domestically but also of imported manufactures.

Throughout this article, labour intensity refers to the ratio of employment to value-added.

The conceptual and empirical differences between manufacturing output and employment levels, and between manufacturing GDP share and employment share, also draw attention to the problems in using one as a proxy for the other (e.g. using manufacturing GDP share as a proxy for manufacturing employment share), either in the analysis of deindustrialisation or in other empirical work such as growth econometrics.

Looking at the UK over the same period as Korea, 1989–2003, manufacturing employment fell from 20.4% to 14.9%, an annualised drop of 2.2%. This is even less of a decline than that of Korea over the same period. However, this comparison would not be appropriate as the UK began its decline in manufacturing share of employment much earlier, and it would miss the somewhat more precipitous fall in the share of manufacturing in the UK in the 1980s (associated, at least in part, with exogenous shocks such as with North Sea oil and economic reform).

For example, Brazil is widely regarded as having experienced deindustrialisation, but the lack of a continuous series on manufacturing employment prevents its inclusion in this analysis.

It should be noted that decomposition analysis is a mechanical technique, which does not shed light on the underlying economic causes of the changes being examined, nor on the endogeneity of each of the components to the other. As will be discussed below, it simply distinguishes between different aspects of the changes in manufacturing employment, with the aspects as specified in the initial identity on which each of the decompositions is based. The literature on deindustrialisation has examined the underlying causes of deindustrialisation—see, for instance, Blackaby (1978) , Rowthorn and Ramaswamy (1997) and Palma (2005) —of which a brief overview was given in Section 2.3 of this article.

Throughout this article sector i is manufacturing. However, the methods proposed here could also be used for the analysis of changes in other sectors.

Which can of course be summed across sectors to give the aggregate change in employment in an economy.

One caveat to be mentioned in relation to the interpretation of the labour intensity effect throughout this paper is that the ‘statistical illusion’ dimension of deindustrialisation may lead to an overstating of the magnitude of the labour intensity effect. As discussed in sections 2.3 and 2.4, one of the ‘sources’ of deindustrialisation is that some activities previously undertaken within manufacturing enterprises have been contracted out to specialised service providers, and the output and employment associated with these activities then counted as part of the services sector. These activities tend to be more labour-intensive than the rest of manufacturing, and hence this shift would affect manufacturing employment more strongly than manufacturing output. To the extent that part of the decline in share of manufacturing employment recorded as deindustrialisation is merely a statistical artefact, this might be considered to lead to an overstating of the (negative) labour intensity effect.

This is, in part, related to sample selection, which is limited to countries and periods in which the share of manufacturing in total employment declined.

See   Appendix 2 for the derivation of these terms.

Note that the ‘labour intensity effects’ calculated in the three decompositions do not measure exactly the same thing, as they are based on different initial specifications, yet all are measures of the changes associated with changes in manufacturing labour intensity and the sign of the effect is constant for each country across the three decompositions.

Note that this term is simply the inverse of the economy-wide labour intensity.

Again, it can be noted that this is in part related to the sample being limited to cases where manufacturing fell as a share of employment.

For instance, Hong Kong developed a dynamic financial services sector.

We tested the sensitivity of the analysis to each of the above three formulations by computing the standard deviation of each component for each country and normalising this by the country mean for that component. The mean and median of the absolute values of the normalised standard deviation of the labour intensity effect (across countries) were 0.015 and 0.006, respectively; of the sector share effect were 0.029 and 0.021, respectively; and of the economic growth effect were 0.012 and 0.007, respectively. This indicates the robustness of the analysis to the three alternative decomposition forms.

The sensitivity of the results between the three alternative formulations was tested in the same way as described above for the second decomposition analysis. The mean and median of the absolute values of the normalised standard deviation of the labour intensity effect were 0.010 and 0.003 respectively; of the sector share effect were 0.021 and 0.014, respectively; and of the labour productivity effect were 0.013 and 0.007, respectively. The results are thus robust between the three alternative decompositions.

Email alerts

Citing articles via.

• Contact Cambridge Political Economy Society
• Recommend to your Library

Affiliations

• Online ISSN 1464-3545
• Print ISSN 0309-166X
• Copyright © 2024 Cambridge Political Economy Society
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

UK Industry and the World Economy: A Case of de-Industrialization?

Cite this chapter.

• Ajit Singh  

Part of the book series: Nijenrode Studies in Economics ((NIEC,volume 2))

116 Accesses

17 Citations

Over the last couple of years in the UK, there has been growing discussion of a phenomenon known as the ‘de-industrialization’ of the UK economy.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Unable to display preview.  Download preview PDF.

Aldcroft, D. H. (ed.), 1968, The Development of British Industry and Foreign Competition , 1875–1914, London, Allen and Unwin.

Google Scholar  

Bacon, R. W. and Eltis, W. A., 1975, ‘Stop-go and de-industrialization’, National Westminster Bank Quarterly Review , November.

Bacon, R. W. and Eltis, W. A., 1976a, Britain’s economic problem: too few producers , London, Macmillan.

Bagchi, A. K., 1967, ‘De-industrialization in Gangetic Bihar 1809–1901 and its implications’ in Barun De et al., Essays in honour of S. C. Sarkar , New Delhi, People’s Publishing House.

Bagchi, A. K., 1976, ‘De-industrialization in India in the 19th century: some theoretical implications’, Journal of Development Studies , January.

Bardhan, P. K., 1965, ‘Equilibrium growth in the international economy’, Quarterly Journal of Economics , August.

Beckerman, W., 1962, ‘Projecting Europe’s growth’, Economic Journal , December.

Black, J., 1970, Trade and the long-run growth rate’, Oxford Economic Papers , March.

Brittan, S., 1977, ‘De-industrialization revisited’, Financial Times , 26 January.

Cairncross, A. (ed.), 1971, Britain’s Economic Prospects Reconsidered , London, Allen and Unwin.

Cambridge Economic Policy Group, 1976, Economic Policy Review , 2, Cambridge, Department of Applied Economics, March.

Caves, R. E., 1970, ‘Export-led growth’ in W. A. Eltis et al (eds.), Induction, growth and trade: essays in honour of Roy Harrod , Oxford, Clarendon.

Caves, R. E. and Jones, R. W., 1973, World trade and payments , Boston, Little Brown.

CSO, 1970,The Index of Industrial Production and other related Output Measures, London, HMSO.

CSO, 1974, United Kingdom Balance of Payments 19 63–73 , London, HMSO

CSO, 1975, United Kingdom Balance of Payments 1964–74 , London, HMSO

CSO, 1976a, United Kingdom Balance of Payments 1965–75 , London, HMSO

CSO, 1976b, Economic Trends , London, HMSO, November.

CSO, 1976c, National Income and Expenditure , 1965–75, London, HMSO.

Conrad, A. H. and Meyer, J. R., 1965, Studies in econometric history , London, Chapman and Hall.

Corden, W. M., 1971, The effects of trade on the rate of growth’, in J. N. Bhagwati et al . (eds.), Trade, Balance of Payments and Growth, Papers in International Economics in Honour of Charles P. Kindleberger , Amsterdam, North Holland.

Cripps, T. F. and Tarling, R. J., 1974, Growth in advanced capitalist economies , Cambridge University Press, Cambridge, UK.

Deane, Phyllis and Cole, W. A., 1962, British Economic Growth : 1688 – 1959 , Cambridge University Press, Cambridge.

Department of Employment, 1975, ‘A view of industrial employment in 1981’, Department of Employment Gazette , May.

Feinstein, C. H., 1972, National Income, Expenditure and Output of the United Kingdom , 1855–1965, Cambridge University Press, Cambridge, UK.

Frenkel, J. A. and Johnson, H. G., 1976, The monetary approach to the balance of payments , London, Allen and Unwin.

Fuchs, V. R., 1968, The Service Economy , New York, National Bureau of Economic Research.

George, K. D. and Ward, T. S., 1975, The Structure of Industry in the EEC: An International Comparison , Cambridge University Press, Cambridge, UK.

Godley, W. A. H., 1976, ‘What Britain needs is growth - but it must be the right kind’, The Times . 1 November.

Hicks, J. R., 1959, Essays in World Economics , Oxford, Clarendon Press.

Hicks, J. R., 1969, A theory of economic history , Oxford, Clarendon Press.

Holland, S. 1976, The socialist alternatives’, The Economist , 3 April.

Houthakker, H. S. and Magee, S. P., 1969, ‘Income and price elasticities in world trade’, Review of Economics and Statistics , May.

Indian Economic and Social History Review , 1968, Debate on the re-interpretation of 19th-century Indian economic history, Indian Economic and Social Review , March.

Jacquemin, A. and de Lichtbuer, M. C., 1973, ‘Size, structure, stability and performance of the largest British and EEC firms’, European Economic Review , December.

Johnson, H. G., 1962, Money, trade and economic growth , London, Allen and Unwin.

Kaldor, N., 1966, Causes of the slow rate of economic growth of the United Kingdom, Cambridge University Press, Cambridge, UK

Kaldor, N., 1970, The case for regional policies’, Scottish Journal of Political Economy , November.

Kaldor, N., 1971, ‘Conflicts in national economic objectives’, Economic Journal , March.

Kaldor, N., 1975, ‘Economic growth and the Verdoorn law: a comment on Mr Rowthorn’s article’, Economic Journal , December.

Kaldor, N., 1976, Letter to The Times , London, 9 November.

Kindleberger, C. P., 1962, Foreign trade and the national economy , New Haven, Yale University Press.

Krueger, A. O., 1969, ‘Balance-of-payments theory’, Journal of Economic Literature , March.

Kuznets, S., 1971, Economic Growth of Nations , Cambridge, Mass., Harvard University Press.

Lamfalussy, A., 1963,The United Kingdom and the Six: an Essay on Economic Growth in Western Europe , Homewood, III., Richard D. Irwin.

Levine, A. L., 1967, Industrial retardation in Britain 1880–1914 , London, Weidenfeld and Nicholson.

Lewis, W. A., 1967, The deceleration of British growth 1873–1913’, Princeton, mimeographed.

London and Cambridge Economic Service, 1971, The British Economy: key statistics 1900–1970 , London, Times Newspapers Ltd.

Magee, S. P., 1975, ‘Prices, Income and Foreign Trade’, in Peter B. Kenen (ed.), International Trade and Finance, Frontiers for Research , Cambridge University Press, Cambridge, UK.

Matthews, R. C. O., 1973, ‘Foreign trade and British economic growth’, Scottish Journal of Political Economy , November.

Moggridge, D. E., 1969, The Return to Gold , 1925 , Cambridge University Press, Cambridge, UK.

Moore, B. and Rhodes, J., 1976, ‘The relative decline of the UK manufacturing sector’, Economic Policy Review , 2 , Cambridge, Department of Applied Economics, March.

Morgan, A. D., 1970, ‘Income and price elasticities in world trade: a comment’, Manchester School , December.

Mundell, R. A., 1968, International Economics , New York, Macmillan.

Myrdal, G., 1957, Economic Theory and Under-developed Regions , London, Duckworth.

NIESR, 1976, National Institute Economic Review , London, November.

Nurkse, R., 1956, ‘The relation between home investment and external balance in the light of British experience’, Review of Economics and Statistics .

Oniki, H. and Uzawa, H., 1965, ‘Patterns of trade and investment in a dynamic model of international trade’, Review of Economic Studies , January.

Panic, M. (ed.), 1976, The UK and West German manufacturing industry , 1954–1972 , London, National Economic Development Office.

Panic, M. and Rajan, A. H., 1971, Product changes in industrial countries’ trade : 1955–68 , London, National Economic Development Office.

Phelps Brown, E. H. and Browne, M. H., 1968, A century of pay , London, Macmillan.

Purdy, D., 1976, ‘British capitalism since the war: part 2, decline and prospects’, Marxism Today , October.

Ray, G. F., 1976, ‘Labour costs in OECD countries, 1964–75’, National Institute Economic Review , November.

Reddaway, W. B., et al., 1968, Effects of UK Direct Investment Overseas , Cambridge, CUP.

Rowthorn, R. E., 1975a, ‘What remains of Kaldor’s Law’, Economic Journal , March.

Rowthorn, R. E., 1975b, ‘A reply to Lord Kaldor’s comment’, Economic Journal , December.

Sayers, R. S., 1965, The vicissitudes of an export economy : Britain since 1880, Sydney, University of Sydney.

Singh, A., 1975, ‘Take-overs, economic natural selection and the theory of the firm’, Economic Journal , December.

Stern, R. M., 1973, The balance of payments , London, Macmillan.

Stern, R. M., Francis, J. and Schumacher, B., 1976, Price-elasticities in international trade , London, Macmillan.

Taplin, G. R., 1973, ‘A model of world trade’, in R. J. Ball, ed., The International Linkage of National Economic Models , Amsterdam, North-Holland.

Thorner, D. and A., 1962, Land and labour in India , London, Asia Publishing House.

Whitman, M. V. N., 1967, International and interregional payments adjustment: a synthetic view , Princeton, Studies in International Finance, No. 19.

Whitman, M. V. N., 1970, Policies for internal and external balance , Special Papers in International Economics, 9, Princeton, Princeton University.

Woodward, V., 1976a, ‘No cause for optimism over imports’, The Times , 23 June.

Woodward, V., 1976b, Government Policy and the Structure of the Economy , Cambridge, DAE, mimeographed

Worswick, G. D. N., 1971, ‘Trade and payments’ in Cairncross (1971).

Download references

You can also search for this author in PubMed   Google Scholar

Editor information

Rights and permissions.

Reprints and permissions

Copyright information

© 1977 H. E. Stenfert Kroese B. V.

About this chapter

Singh, A. (1977). UK Industry and the World Economy: A Case of de-Industrialization?. In: Jacquemin, A.P., de Jong, H.W. (eds) Welfare aspects of industrial markets. Nijenrode Studies in Economics, vol 2. Springer, Boston, MA. https://doi.org/10.1007/978-1-4613-4231-1_10

Download citation

DOI : https://doi.org/10.1007/978-1-4613-4231-1_10

Publisher Name : Springer, Boston, MA

Print ISBN : 978-1-4613-4233-5

Online ISBN : 978-1-4613-4231-1

eBook Packages : Springer Book Archive

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Cookies on GOV.UK

We use some essential cookies to make this website work.

We’d like to set additional cookies to understand how you use GOV.UK, remember your settings and improve government services.

We also use cookies set by other sites to help us deliver content from their services.

You have accepted additional cookies. You can change your cookie settings at any time.

You have rejected additional cookies. You can change your cookie settings at any time.

• Health and social care
• Medicines, medical devices
• Clinical trials and investigations

VDEC is supporting a GBS vaccine to prevent newborn deaths

Antimicrobial resistance to Group B Streptococcus (GBS) antibiotics is growing. This puts newborns at risk. VDEC is supporting a new vaccine to cut antibiotic use

The vaccine development and evaluation centre ( VDEC ) facilitates the development and evaluation of new vaccines and therapeutics.

VDEC ’s work includes vaccine discovery and surveillance . The Vaccine Assay and Immune Response team within  VDEC  provides world class scientific expertise and regulated facilities to enable the classification of diseases, pathogens and their mechanisms of action. They also provide continued national surveillance after the release of vaccines.

They have recently been supporting the creation of a maternal vaccine for Group B Streptococcus ( GBS ).

Work with VDEC

We work with industry, academia and government. Contact UKHSA today to see how we can help you.

Get in touch

GBS is the leading cause of vaccine-preventable infections in newborns in the developed world, and a significant cause of newborn infections and stillbirths worldwide. As such, GBS is a leading driver of antibiotic use in neonatal settings, and antimicrobial resistance is increasing.

To reduce early-onset disease (usually within the first 48 hours of life) many countries have introduced screening for GBS in pregnancy. Culture positive mothers are then given intravenous antibiotics in childbirth to protect both the mother and the infant. This has dramatically increased the use of antibiotics in childbirth. In some countries over 50% of childbirths now involve intravenous antibiotic use.

A maternal vaccine for GBS which protects infants from both early and late onset disease (for which intrapartum antibiotics have no effect) will have a positive impact on infant mortality and morbidity. It will also lead to a sharp reduction in the use of antibiotics in neonatal units worldwide.

Reducing antibiotic use in newborns will allow the development of appropriate commensal gut bacteria. This enables them to digest milk and reduces the risk of antimicrobial resistance.

The UKHSA Pathogen Immunology Group at Porton Down had previously been part of the GASTON consortium. This is an international consortium to develop standardised correlates of protection to allow the evaluation of new GBS vaccines and enable licensure of the leading candidates.

UKHSA Porton successfully developed a ‘Gold Standard’ opsonophagocytic killing assay ( OPKA ) and led a global interlaboratory study of the OPKA in public health, academic and industry labs. The assay was shown to be highly sensitive and reproducible in different laboratories, and standard GBS test strains have been selected by UKHSA and made available worldwide.

What VDEC offers

In 2022 to 2023, follow-on funding was awarded by the Bill and Melinda Gates Foundation ( BMGF ) to develop a higher-throughput version of the OPKA suitable for Phase II/III clinical trials and to provide international reference serum samples with defined units of opsonophagocytic activity. The high throughput OPKA is required as the current assay method is very labour-intensive which limits the number of sera that can be analysed in large vaccine studies.

Expected outcome

Licensure of a GBS vaccine will:

• prevent the deaths of tens of thousands of newborns worldwide
• reduce the need for antibiotic treatment in disease cases
• reduce the need for intrapartum antibiotics during childbirth

Future work

As part of this project, the team has also developed an OPKA for GBS serotype VII, which will be used in a follow-up PATH -funded Phase I/II clinical study of a novel 6-valent GBS vaccine currently being developed.

Is this page useful?

• Yes this page is useful
• No this page is not useful

Help us improve GOV.UK

Don’t include personal or financial information like your National Insurance number or credit card details.

To help us improve GOV.UK, we’d like to know more about your visit today. Please fill in this survey (opens in a new tab) .

Science and Technology Studies

STS offers degrees at each university level: undergraduate, masters, and PhD

• Warning Research Centre

From Policy to Practice: Case Studies in Resilience and Public Warning

21 May 2024

UCL WRC welcomed Everbridge for the launch of their collaborative report.

The UCL Warning Research (UCL WRC) centre recently welcomed Everbridge to Bloomsbury for the launch of our collaborative report " From Policy to Practice: Case Studies in Resilience and Public Warning ". This groundbreaking report is the result of a dynamic partnership between the UCL WRC and Everbridge. 

The launch event welcomed Dave Wagner (CEO, Everbridge), Carina Fearnley (Director, UCL WRC), Lorenzo Marchetti (Head of Global Public Affairs, Everbridge) and Valerie Risk (BP Public Safety International, Everbridge) alongside authors of the report Maryam Rokhideh (PhD, UCL WRC) and Marley Taylor (Public Affairs Intern, Everbridge). 

Our speakers led an engaging session, uncovering the intersection of policy and practice in the realm of resilience and public warning through in-depth case studies focused on the United States, Japan, and the European Union. 

Bringing together academia and industry expertise to contribute to the advancement of societal resilience, the report navigates the complexities of translating policy formulations into actionable practices, fostering a deeper understanding of effective implementation.

Key Takeaway points from the report:

• Government resilience approaches have made great progress, but integration and harmonization from the local-to-national is needed to increase systemic resilience.
• Governments, institutions, and private companies can enhance Resilience and Disaster Preparedness through Knowledge-Sharing and Coordination on Warnings.
• Public-private partnerships and companies like Everbridge can fill government gaps to ensure preparedness and resilience in the face of disasters.
• Everyone should be covered by warning systems, yet many are still not reached by current initiatives.
• More resilience action is needed to fill government gaps - companies such as Everbridge are able to do so now, before it is too late.

Watch a recording of the launch:

YouTube Widget Placeholder https://youtu.be/uJ0fD6rsuUs?si=4EJ1QAlH9X-PYJt2

Download the Report Now

University of Birmingham

Inadequate access to simple elective surgery in developing countries is storing up future health problems for patients and may create a spiral of future health complications putting more people’s lives at risk, a new study reveals.

Analysing the experience of more than 18,000 patients in 640 hospitals across 83 countries, researchers, experts used hernia repair to represent elective health care, concluding that such treatments are essential to prevent over-reliance on emergency systems.

The study reveals that inguinal hernias are treatable with simple day-case surgery, but, if neglected, the need for more complex emergency surgery increases substantially, leading to delayed recovery and far higher total health-care costs.

Led by experts at the University of Birmingham, the NIHR Global Health Research Unit in Surgery study notes that increasing reliance on emergency care has resulted in crisis management becoming routine across a wide range of conditions that respond well to early elective treatment.

Boosting the use of elective surgery for conditions that can be fixed simple and early treatments will reduce the risk of complex, and potentially risky, emergency surgery. Dr Maria Picciochi - University of Birmingham

Study co-author Dr Maria Picciochi, from the University of Birmingham, commented: “Boosting the use of elective surgery for conditions that can be fixed simple and early treatments will reduce the risk of complex, and potentially risky, emergency surgery.”

Study co-author Prof Aneel Bhangu, from the University of Birmingham, also added: “Health policy makers can use our findings as a proxy for other elective conditions, creating a system strengthening approach to integrate surgery into the wider system of health care. This would relieve pressure on emergency pathways and reduce the health burden on society and healthcare services.”

The study shows that inguinal hernias are mostly a disease of working-age patients around the world, and, if neglected, may require bowel resection. This more complex surgical treatment leads to slow recovery and far higher total health-care costs.

Researchers also found a clear global imbalance in access to mesh repair - reflecting poor access to simple medical devices in lower-income countries. Mesh is well proven to reduce long-term hernia recurrence, is simple to place, low-cost and scaleable.

The researchers identified actionable targets for system strengthening, which include:

• Educating communities and community health workers around hernia symptoms;
• Improving referral systems and increasing mesh repair for hernias;
• Establishing a global quality improvement programme in mesh placement for hernias - strengthening supply chains, making mesh affordable and increasing training; and
• Improving capacity for simple, cost-effective surgery.

“Our study showed multiple weaknesses in access and quality in current health-care systems, with a particular disadvantage in lower-income settings,” commented Dr. Picciochi. “As a result, there was higher emergency demand, which further reduced elective capacity and might create downward spirals.

“If weak access and quality persist over several electively treatable conditions, both surgical and non-surgical, multimorbidity can also become established, which makes future elective care harder and emergency care even more complicated.”

Findings of the study and its wider implications will be discussed during the side event hosted by the NIHR Global Surgery Unit at the World Health Assembly, on 28 th May. 

Featured staff

Professor Aneel Bhangu

Professor of Global Surgery

Notes for editors

For more information, please contact our Press Office . For out-of-hours enquiries, please call +44 (0)121 414 2772.

The University of Birmingham is ranked amongst the world’s top 100 institutions, its work brings people from across the world to Birmingham, including researchers and teachers and more than 8,000 international students from over 150 countries. Discover Birmingham .

‘ Access to and quality of elective care: a prospective cohort study using hernia surgery as a tracer condition in 83 countries ’ – NIHR Global Health Research Unit on Global Surgery is published in Lancet Global Health.

Participating countries included: Albania, Algeria, Argentina, Austria, Australia, Aruba, Bosnia and Herzegovina, Bangladesh, Burkina Faso, Bulgaria, Burkina Faso, Burundi, Benin, Brazil, Cambodia, Cameroon, Canada, Chile, China, Colombia, Croatia, Cyprus, Czech Republique, Dominican Republic, Egypt, Ethiopia, France, Gabon, Germany, Georgia, Ghana, Greece, Guatemala, India, Ireland, Iran, Iraq, Italy, Japan, Jordan, Kazakhstan, Kenya, Lebanon, Libya, Lithuania, Madagascar, Malaysia, Mali, Malta, Mexico, Morocco, Namibia, New Zealand, Nigeria, Oman, Pakistan, Palestine, Paraguay, Peru, Poland, Portugal, Qatar, Republic of North Macedonia, Romania, Russian Federation, Rwanda, Saudi Arabia, Serbia, Slovenia, South Africa, Spain, Sri Lanka, Sudan, Sweden, Switzerland, Syria, Thailand, Togo, Tunisia, Turkey, Uganda, United Kingdom, United States, Yemen.

The National Institute for Health and Care Research (NIHR)

The mission of the National Institute for Health and Care Research (NIHR) is to improve the health and wealth of the nation through research.

We do this by:

• Funding high quality, timely research that benefits the NHS, public health and social care;
• Investing in world-class expertise, facilities and a skilled delivery workforce to translate discoveries into improved treatments and services;
• Partnering with patients, service users, carers and communities, improving the relevance, quality and impact of our research;
• Attracting, training and supporting the best researchers to tackle complex health and social care challenges;
• Collaborating with other public funders, charities and industry to help shape a cohesive and globally competitive research system;
• Funding applied global health research and training to meet the needs of the poorest people in low and middle income countries.
• NIHR is funded by the Department of Health and Social Care. Its work in low- and middle-income countries (LMICs) is principally funded through UK Aid from the UK government.

World-leading surgical research team gets fresh funding to save lives

A world-leading global surgical research team led by the University of Birmingham has received £7 million to continue life-saving work in developing countries.

16 February 2022