The impact on welfare and public finances of job loss in industrial Britain

Abstract It is important to take a long view of many economic problems. This article explains how the large-scale loss of industrial jobs in parts of Britain during the 1980s and 1990s still inflates the contemporary budget deficit in the United Kingdom. Drawing on the findings of several empirical studies by the authors, the article shows that although there has been progress in regeneration the consequences of job loss in Britain’s older industrial areas have been near-permanently higher levels of worklessness, especially incapacity benefits, low pay, and a major claim on present-day public finances to pay for both in-work and out-of-work benefits. Furthermore, as the UK government implements reductions in welfare spending the poorest places are being hit hardest. In effect, communities in older industrial Britain now face punishment in the form of welfare cuts for the destruction previously wrought to their industrial base.


INTRODUCTION
Most discussion about the UK economy tends to be framed within short time-horizons, with elections or at most changes in government setting the outer edges of memory and debate. This may convey immediacy and relevance but the effect is all too often to obscure the longer-term issues and trends.
In the United Kingdom, the defining feature of the economy over the last 30 or 40 years has been the big shift away from industry as an employer and generator of wealth. That this 'deindustrialization' has happened is widely understood. Yet the massive consequences for the contemporary economy and for present-day policy-making are generally overlooked. This is unfortunate because major economic changes, such as deindustrialization, have impacts that spill over from decade to decade.
This article begins to build the links between the loss of Britain's industrial base and the UK's present-day public finances. To do so, the article draws on the authors' own empirical research over the last three decades. The individual components of this research have been published elsewhere so the full details are not repeated here. However, by combining all of the pieces an overall picture becomes clear. The article's novel contribution is to build the links between job loss, welfare spending and present-day austerity, and to do so by taking a long-term perspective.
In brief, the argument is that the destruction of industrial jobs, which was so marked in Britain in the 1980s and early 1990s but has continued on and off ever since, has fuelled spending on welfare benefits which in turn has compounded the budgetary problems of successive governments. With the UK government set on welfare reform, the places that bore the brunt of job destruction some years ago are now generally facing the biggest reductions in household incomes. There is a continuous thread linking what happened to British industry in the 1980s, via the Treasury's budgetary calculations, to what is today happening on the ground in so many of Britain's hardpressed communities.
In particular, these links are explored using local data. This has been the distinctive contribution of the authors' research and its value is that it provides not just a level of detail that would otherwise be missing but, more importantly, it sheds light on the underlying processes at work. The UK Treasury knows it has a problem balancing public finances and that the government spends an awful lot on working-age welfare benefits. But it never seems to ask exactly wherewhich towns and cities -draw so heavily on benefits, or why these communities have become so reliant on welfare spending.
The focus of the argument and the data we present is distinctively British. However, deindustrialization is not unique to the UK. Welfare systems obviously vary from country to country, but the rather striking evidence from the UK may offer important pointers to how, in the long run, labour markets and public spending may have adjusted to industrial job loss in other countries too.

BACKGROUND
'Jobs' and 'welfare spending' are significant topics in their own right but in the UK they have generally been separate lines of academic inquiry.
Industrial job loss has attracted substantial attention. The major job losses of the early 1980s were documented in particular by Townsend (1983) and the process of deindustrialization more generally by Martin and Rowthorn (1986). The distinctive geography of the early 1990s recession was recorded by Gudgin (1995). By the end of the 1990s it was clear that the job deficits across the UK were highly uneven and that they were no longer terribly well reflected by unemployment data (Erdem & Glyn, 2001;MacKay, 1999;Webster, 2000). The long up-turn in the UK economy from 1993 to 2008 then began to generate an element of complacency, and the government quite misleadingly began to speak in terms of a return to full employment (HM Treasury, 2003). The recession provoked by the 2008 financial crisis was a wake-up call, prompting a rediscovery of the divergent trends between North and South (Gardiner, Martin, Sunley, & Tyler, 2013) and between city regions (Centre for Cities, 2016;Martin, Sunley, Tyler, & Gardiner, 2016;Pike et al., 2016;Swinney & Thomas, 2015;Townsend & Champion, 2014).
Debates around welfare have traditionally been the preserve of social policy analysts rather than economic geographers so they tend to be concerned less with the impact on places than on specific social groups (see, e.g., Alcock, 2014). 'Welfare reform' is not a new concept, but in the UK it became a prominent part of the jigsaw following the election of the Conservative-led Coalition Government in 2010. Austerity has clear links to welfare reform in that the need to curb public spending has been advanced as a key reason for reducing social security benefits (HM Treasury, 2010). Welfare reform has generally been explored in terms of its impact across the income spectrum, on vulnerable groups or on work incentives (Adams & Browne, 2013;Hirsch & Beckhelling, 2011;Hood & Johnson, 2016;Scottish Government, 2013). Webster (2006) is an exception in arguing that the effectiveness of welfare reform can only be understood in the context of the geography of worklessness.
However, the failure to link industrial job loss, claims on the welfare budget and the impact of austerity is in our view a failure to understand the origins of present-day problems.
Regional StudieS, Regional Science

THE DESTRUCTION OF INDUSTRIAL BRITAIN
To begin to weave the links between jobs, welfare and austerity in the UK it is appropriate to start by outlining the scale of industrial job loss and its distinctive geography.
Britain was once a major industrial employer. In 1966, when manufacturing employment peaked, 8.9 million worked in manufacturing and a further 500,000 in the coal industry. This compares with just 2.9 million employed in manufacturing in 2016 1 and none at all in the coal industry except at a handful of opencast sites and tiny drift mines. The shift from manufacturing to service-sector employment is a phenomenon shared by other advance economies (see Rowthorn & Wells, 1987), rooted in differential rates of productivity growth and accentuated by globalization. But in Britain the process of deindustrialization has gone further and faster than just about anywhere else.
As Figure 1 shows, UK manufacturing employment fell steeply in the early 1980s in a recession triggered by a high exchange rate and high interest rates. The recession of the early 1990s added further pain. Thereafter, manufacturing employment failed to recover even though the UK economy enjoyed 15 years of sustained economic growth. As a share of total UK employment, manufacturing has now fallen to just below 10%, compared with 30% in the 1960s and around 25% at the start of the 1980s. For the coal industry (which is not included within 'manufacturing'), the biggest job losses started a little later, after the 1984/85 miners' strike, but by 1992 two-thirds of the pre-strike workforce had gone. The final colliery closed in 2015.
These industrial job losses were concentrated in specific parts of the country -mostly but not exclusively in the 'older industrial areas' of the North, Scotland and Wales. In many cases the economic base of whole communities was destroyed. By contrast, London escaped relatively lightly and so did most of its vast hinterland in the South and East of England.
Partly this pattern reflected the pre-1980 location of UK manufacturing, and coalmining of course only took place where there was coal to be mined. Partly the pattern of job loss reflected the location of the industries that shrank most -coal, steel, shipbuilding, heavy engineering, textiles and so forth. Also, partly it was attributable to the closure of branch factories in the 'assisted areas' that had opened in the post-war years of economic growth and strong regional policy (Fothergill & Guy, 1991).  Figure 2 illustrates the geography of this job loss. Manufacturing employment has fallen just about everywhere across the UK so this map does not try to show the location of every industrial closure or redundancy over the last three and a half decades. Rather, it flags up the biggest or most significant job losses, where major companies or industries have shrunk to a fraction of their former scale or disappeared entirely. Some of the job losses shown here are the result of individual, large plant closures, but more often than not (e.g., in the case of the coal and textile industries) they reflect the run down and closure of employers spread across several neighbouring towns. Nevertheless, the concentration in a number of specific areas is especially noticeable. It is not the whole of the North, nor indeed the whole of Scotland or Wales, that has been hard hit. Many rural areas, for example, were less affected. Anyone familiar with the UK will recognize that it is mostly the industrial cities, towns and coalfield areas that suffered the big job losses -central Scotland (particularly the Glasgow area), the North East of England, West Cumbria, a large swathe of the North West of England and Yorkshire (from Liverpool via Manchester, Leeds, Sheffield and their surrounding towns across to Hull), parts of the Midlands (including the Notts/Derby coalfield, the Birmingham area and Stoke-on-Trent) and South Wales including the Valleys.
This distinctive geography of industrial job loss is a pattern that recurs in the evidence on welfare spending presented later. Our argument is not that the link is confined to a handful of cities or towns but that it characterizes large parts of the North, the Midlands, Scotland and Wales affected by industrial job loss. In these places, recovery from job loss has been incomplete and welfare spending is far higher as a result.

LABOUR MARKET ADJUSTMENT
The 1980s are remembered in Britain as a period of high unemployment. The number of claimant unemployed -that is, the number out of work claiming unemployment benefits -hovered around three million for a number of years, which was perhaps to be expected given the scale of the job loss. But after the early 1990s recession, claimant unemployment fell away, declining to less than one million for most of the 2000s, and after the 2008 financial crisis returning to below one million once more. If claimant unemployment alone was to be the sole guide, it might be argued that the UK economy has got over deindustrialization. Unfortunately, this optimistic assessment is wide of the mark. A closer look at how labour markets have adjusted explains why.
The first clue to what was really happening in older industrial areas came from the coalfields. By the early 1990s most of the pits had shut but claimant unemployment in the coalfields was no higher than when the pits had been working. This was not what most observers had expected. We therefore picked apart the trends in the coalfield labour market (Beatty & Fothergill, 1996). How much of the low claimant unemployment could be explained by commuting to neighbouring areas, by out-migration or by new job creation? The conclusion, based on Census data, was in fact that the main response to coal job loss was a diversion of working-age men into 'economic inactivity'. Looking closer, this was primarily a withdrawal from the labour market into what the Census called 'permanent sickness' -in practice onto incapacity-related benefits. Therefore job loss had indeed resulted in a lasting increase in benefit claims but not in the way that had been expected.
In a follow-up study (Beatty, Fothergill, & Powell, 2007) we brought the figures forward by a decade or so. A lot more had happened, particularly on the job creation front, but the fundamental conclusion remained the same: the large labour market adjustment in response to coal job losses was an increase in economic inactivity among working-age men. Furthermore, because many of the ex-miners had by then reached state pension age it was clear that the increase in economic activity must be occurring much more widely across the local workforce. In effect, job loss for one generation was being passed on as higher economic inactivity among the next. By 2008, economic inactivity among working-age men in the English and Welsh coalfields was still 150,000 higher than at the beginning of the 1980s (Beatty, 2016; Foden, Fothergill, & Gore, 2014). Regional StudieS, Regional Science The coalfields pointed the way but it quickly became apparent that their experience was not unique. Across the whole of older industrial Britain, from the mid-1980s through to the early 2000s there was a huge surge in the numbers out of the labour market -'economically inactive'on incapacity-related benefits (Beatty & Fothergill, 2005).
We argued that much of the increase in incapacity numbers was a form of 'hidden unemployment'. These were 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-related benefits (these days Employment and Support Allowance (ESA)) instead of unemployment benefits. Our estimates of the scale of hidden unemployment adjust for underlying variations in the extent of ill health. The most recent figures, for 2012 (Beatty, Fothergill, & Gore, 2012), suggest that as many as 900,000 unemployed are hidden on incapacity-related benefits, only around 100,000 down on our first estimate for 1997 (Beatty, Fothergill, Gore, & Herrington, 1997).
The increase in incapacity numbers in older industrial Britain occurred among women as well as men. At first this seemed hard to understand because the heavy industries shedding jobs had previously mainly employed men. What became apparent, after much detailed research in the former coalfields and elsewhere, is that these days the male and female sides of the labour market interact so that a shortfall in opportunities for men is transmitted, through competition for jobs, to a difficult local labour market for women in the same places. Out-of-work women with health problems or disabilities generally end up on incapacity-related benefits just like their male counterparts (Beatty, 2016;Beatty, Fothergill, Houston, Powell, & Sissons, 2009b).
So via the study of local data it became possible to draw firm conclusions about what really happened in response to the large-scale loss of jobs in older industrial Britain: yes, claimant unemployment did fall back to low levels. But the near-permanent effect has been to raise incapacity claimant numbers, both among men and women.

THE IMPACT ON PRESENT-DAY WELFARE SPENDING
We can now begin to explore what this has meant for welfare spending and for the UK Treasury's struggle to balance public finances. It is appropriate to begin by looking at the numbers claiming the three main out-of-work benefits, shown in Figure 3 for 1979-2016 for Britain as a whole.
As we noted earlier, the numbers claiming unemployment benefits -Jobseeker's Allowance ( JSA) from 1996 onwards and Universal Credit more recently -reached three million in the mid-1980s, fell back, rose again in the early 1990s and then declined to well under a million. In the wake of the 2008 financial crisis the numbers peaked at around 1.5 million before falling back once more.
The number claiming lone-parent benefits -Income Support for most of this period -rose from around 300,000 at the start of the 1980s to a peak of around one million in the mid-1990s. The evidence on the geography of lone-parent claims (Rowthorn & Webster, 2008) pointed clearly to job loss among men as a key factor. In the places where men's jobs had disappeared, such as older industrial Britain, the ability of men to provide financial support for women and children had been eroded. More recently, the numbers for lone-parent benefits have fallen, not least because eligibility has gradually been restricted just to those with the very youngest children.
The striking feature in Figure 3, however, is the rise in the numbers out of work on incapacity-related benefits, from around 750,000 to a plateau of around 2.5 million. The numbers have declined a little from the all-time high in the early 2000s, but not by much.

Incapacity-related benefits
There are two remarkable aspects of the incapacity numbers. First, they are largely invisible. The figures surface in the media from time to time but probably few beyond those who follow these issues would be aware that the numbers currently out of work on incapacity-related benefits exceed the numbers on unemployment benefits by more than three to one and that, the immediate post-financial crisis years excepted, this has been the situation since the end of the 1990s.
The other remarkable aspect of the incapacity numbers is that they have stayed so high for so long despite multiple efforts to bring them down. Reform in 1995 introduced more formal medical tests. More restrictive eligibility rules were introduced in 1999. A Pathways to Work programme for claimants was introduced in 2003. From 2008 onwards ESA replaced the previous incapacity benefits, with a new medical test, new conditionality and (from 2012 onwards) an extension of means-testing. But all this effort has had remarkably little impact on the headline figures. Clearly, the factors that underpin incapacity claimant numbers are very powerful indeed.
The key insight again comes from the local numbers. Figure 4 shows the share of adults of working age claiming incapacity-related benefits in February 2016, by district across the whole of Britain. It is immediately apparent that there are huge variations across the country. Moreover the pattern is systematic. The highest incapacity claimant rates are mostly found in older industrial Britain -places such as the South Wales Valleys, North East England, Merseyside and Clydeside. In contrast, the incapacity claimant rate in much of the South and East of England, especially outside London, is modest.
In fact, 18 of the 20 districts with the highest incapacity claimant rates cover older industrial areas. 2 Typically, the incapacity claimant rate in older industrial Britain is just above or below 10%, meaning that one in 10 of all adults between the ages of 16 and 64 in these places are out of the labour market on ESA or (in a diminishing number of cases) one of its predecessor benefits. In Blaenau Gwent and Neath Port Talbot in South Wales the incapacity claimant rate is 11.9%. This rate is also 11.9% in Glasgow. In Liverpool it is 10.8%, in Middlesbrough 10.3% and in Stoke-on-Trent 10.1%.
High incapacity claimant rates are emphatically not an issue in the places where the local economy is strong. Only one London borough is among the top hundred (Islington at number 95) and only four other districts in South East England, all of which cover seaside towns.
Of course, older industrial Britain often has higher underlying levels of ill health so we might expect to find higher incapacity claimant rates here. But it is worth remembering that the surge in incapacity claimant numbers in these places only happened after the industrial jobs began to disappear. When the mines, steelworks and the like still employed vast numbers, far more people were actually exposed to damaging impacts on their health but far fewer made incapacity claims.  What appears to be happening is that where there are plenty of jobs, the men and women with health problems or disabilities are able to hang on in employment or find new work if they are made redundant. But where the labour market is difficult, as in older industrial Britain, ill health or disability ruins many people's chances of finding and keeping work. Employers are well able to recruit the fit and healthy instead. Poor qualifications, low-grade work experience and advancing years all too often compound the difficulties that the sick and disabled experience in finding work.

Additional welfare benefits
This is nevertheless still a long way from attributing more than a small share of the present-day fiscal constraints on the Treasury to the consequences for the benefits system of industrial job loss. To expose the full impact we have to first consider the other welfare benefits claimed alongside incapacity-related benefits.
ESA, onto which nearly all incapacity claimants have been moved, is not overly generous, particularly bearing in mind that most claimants spend long periods out of work on this benefit. The current (2016/17) basic rates are £109 a week for an ESA claimant aged 25 or over in the 'Support Group' and £102 a week for one in the 'Work-Related Activity Group', although the latter is now means-tested after 12 months and is soon to be reduced to £73 a week (the same as JSA) for new claimants. But in practice many ESA claimants are also entitled to additional benefits, depending on their personal and household circumstances. These benefits include the following: • Means-tested top-ups, including for disability. Several of these are widely claimed by ESA claimants. At the present time a number of these benefits are being merged into Universal Credit, but that does not change the basic point because the rules governing entitlement are essentially carried over from the old system. DLA, which is currently in the process of being replaced by PIP, is worth singling out. This is paid to men and women with health problems or disabilities to help offset the additional costs they face. In total, 3.65 million men and women claimed DLA or PIP in February 2016, of whom 2.25 million were of working age. 3 The geography of DLA/PIP claims, shown in Figure 5, is remarkably similar to the geography of incapacity (ESA) claims. As we noted in a report for the Department for Work and Pensions in 2009 (Beatty, Fothergill, & Platts-Fowler, 2009), at that time around half of all incapacity claimants received DLA and around four out of five working-age DLA claimants received incapacity benefits. The report also noted that even in-work DLA claimants were concentrated in the same places as incapacity claimants.
It is hardly surprising therefore that older industrial areas account for around three-quarters of the districts with the highest DLA/PIP claimant rate. Neath Port Talbot in South Wales heads the list, where 11.2% of all 16-64 year olds -one in nine of the working-age population -claim DLA or its replacement PIP. In Glasgow this figure is 9.7%, in Liverpool 9.3% and in Barnsley in South Yorkshire 8.6%. By contrast, there are relatively few DLA/PIP claimants, either in work or out of work, in the most prosperous local economies of southern England. Across much of Surrey, for example, the DLA/PIP claimant rate is below 3%. The highest ranked London borough (Islington again) comes in at only 5.6% -161st out of 379 Great Britain (GB) districts. Regional StudieS, Regional Science In essence, DLA/PIP appears to function as an addition to ESA for many longer-term incapacity claimants whilst some of those who do return to work then retain their DLA/PIP entitlement, which is not dependent on employment status. Therefore, just as ESA claims are disproportionately concentrated in older industrial areas, DLA/PIP claimants are found disproportionately in older industrial areas as well.
It is not easy to assess precisely how much this all costs. The Treasury does not publish overall figures but some components can be measured directly and others can be estimated. The DWP's own data tell us that £14.9 billion a year is spent on working-age incapacity-related benefits, these days nearly all ESA. To this needs to be added an estimated £7.2 billion a year paid to the same claimants in the form of DLA/PIP, £7 billion a year in Housing Benefit and £3.2 billion a year in Tax Credits. The grand total for the benefits presented in Table 1 comes to just under £34 billion a year. This is a staggering sum, and one that has previously received little attention. But let us be quite clear: we are not arguing that this is financial support which individuals are not entitled to or should not receive. We are simply drawing attention to the total cost to the Exchequer. Furthermore, of this immense cost a good proportion -perhaps £10-14 billion a year given the distribution of incapacity claimants across the country -could be described as the price of job destruction in older industrial Britain.
The cost of higher claimant unemployment in older industrial Britain needs to be added to this. The numbers on JSA are far less than those on ESA, as we noted earlier, and some add-on benefits such as DLA/PIP are less widely claimed by the claimant unemployed. JSA alone cost the Exchequer £2.3 billion in 2015/16, but if we follow broadly the same logic as for incapacity claimants (in Table 1) the full cost of claimant unemployment, adding in other benefits, is more than £6 billion a year.
Once more, only part of this cost can be attributed to the destruction of jobs in industrial Britain and it is worth bearing in mind that there is always some unemployment even in fully employed local economies. But if we assume that around half of the claimant unemployment in older industrial Britain is rooted in job loss, and bear in mind that these places have rather more than 40% of all GB unemployment, 4 the higher claimant unemployment arising from industrial job destruction probably costs the Exchequer another £1-1.5 billion a year.

In-work benefits
The full cost to the Exchequer is greater still, however. One of the defining features of the industrial jobs that have been lost on such a grand scale is that they were often relatively high value-added, high-wage jobs. The skilled manual jobs in manufacturing are now far fewer in number. In older industrial Britain there has been job growth in the wake of industrial decline but all too often it has been in low-productivity, low-wage activities. In the former coalfields, for example, two of the prime sources of new jobs have been call centres and warehouses. The Yorkshire, Derbyshire and Nottinghamshire coalfields, for instance, have a central location and ready access to the motorway network and have become prime destinations for national distribution depots. A well-publicized example, on the site of the former Shirebrook Colliery in Derbyshire, is the national warehouse of Sports Direct, where most of the workforce is employed on zero-hours contracts and low wages (see, e.g., The Guardian, 2015). Beyond the call centres and warehouses, growth in consumer spending has fuelled job growth in shops, hotels, pubs, restaurants and takeaways. Few of these new jobs are well paid, and many are part-time.
It is the weakness of labour demand in older industrial Britain, stripped of its once dominant employers, that has enabled the new employers to get away with paying low wages. The ex-miners and ex-steelworkers may have baulked at the prospect of work in a call centre or warehouse and opted out of the labour market instead, cushioned by redundancy pay, early entitlement to pensions and disability benefits, but their sons and daughters have never faced the same choices (Alcock, Beatty, Fothergill, Macmillan, & Yeandle, 2003;Beatty, Fothergill, Houston, Powell, & Sissons, 2009a). With little possibility of remaining on JSA for long periods they have had to accept whatever work they can find, particularly as some employers have been quick to turn to migrant workers from Eastern Europe as an alternative supply of low-wage labour (Dench, Hurstfield, Hill, & Akroyd, 2006).
Women's growing involvement in the labour market adds a further twist (Beatty et al., 2009b). In the places once dominated by heavy industry the tradition used to be that male wages supported whole families. Relatively few women with children held paid employment, especially on a fulltime basis. That more women in these places now look for paid employment should be welcome progress but they do so in some of the most problematic labour markets in the country. Local economies have to grow very fast indeed if they are to not only replace the jobs that have been lost but also keep up with new labour supply. In practice, the growth has been insufficient and the result has been worklessness, part-time employment and low wages.
All this has further knock-on consequences for the Exchequer. Low wages generate low tax returns. But at least as importantly, low wages generate a high bill for in-work benefits. These benefits include the following: • Housing Benefit. • Child Tax Credit. • Working Tax Credit.
Spending on in-work benefits is concentrated in the places where low wages are prevalent and (in the case of Housing Benefit) where housing costs are high. To illustrate this point, Figure 6 shows the estimated spending on Tax Credits, per adult of working age, in every local authority district across Britain. 5 This presents a complex picture which is by no means a case of older industrial Britain versus the rest. Wherever low wages are the norm, spending on Tax Credits is high. This applies in a number of rural areas, in several seaside towns and in the parts of London where less well-off residents are concentrated. But it also applies across most of older industrial Britain.
Regional StudieS, Regional Science The overlap with the maps shown earlier for job loss and disability benefits is considerable. In Middlesbrough, the estimated spending on Tax Credits works out at £1,050 a year per adult of working age. In Stoke-on-Trent it is £1,000. Across most of older industrial Britain, in fact, it exceeds £850. The equivalent figure in Guildford in Surrey is just £290 a year, and in Kensington and Chelsea is £310.
So job destruction in older industrial Britain has resulted not only in higher spending on outof-work benefits but also higher spending on in-work benefits and depressed tax revenue. It is impossible to put a reliable figure on the cost in terms of in-work benefits and lost tax revenue. However, Tax Credits paid to in-work households cost the Exchequer almost £20 billion a year 6 and Housing Benefit to in-work households rather more than £5 billion a year. If low pay also means that the Treasury receives £1,000 a year less in tax from (say) five million workers -one in six of the workforce -that would be a further £5 billion a year. 7 Add these together and the Exchequer cost of low pay is perhaps £30 billion a year. Bearing in mind the prevalence of low pay across older industrial Britain, perhaps £10 billion a year might be attributed to the destruction of well-paid industrial jobs.
If we add this admittedly speculative figure to the earlier, more robust estimates for the costs of ESA and JSA claims, the present-day cost to the Exchequer, in welfare payments and lost tax revenue, of the destruction of jobs in industrial Britain almost certainly exceeds £20 billion a year.

Lower output, lower incomes
There is a final way in which the destruction of industry has undermined public finances. This is through the erosion of the UK's export base, and this is arguably the most powerful effect of all. This is not something our research has addressed but it would be wrong to overlook.
The UK economy in 2017 seems remarkably prosperous. Gross Domestic Product (GDP) exceeds pre-financial crisis levels and overall employment has reached record highs. But the prosperity is deeply precarious, and not simply because of the uncertainties created by impending departure from the European Union (EU). The UK economy is characterized by an extraordinary level of household debt, an alarming trade deficit with the rest of the world and a public-sector budget deficit that remains large despite the most draconian austerity measures in modern times These features of the contemporary UK economy are deeply interrelated. In essence, Britain is living beyond its means. Consumption and living standards are being sustained not by incomes earned by trading with each other and the rest of the world, but by ever-rising debt and the sale of UK assets -companies, property, government bonds -to foreign investors.
That debt has become the driver of UK economic growth is first and foremost the result of the erosion of the UK's industrial base. The UK no longer sells enough to the rest of the world to pay for what it imports and the UK manufacturing sector has become so hollowed out that even a substantial devaluation of sterling, such as occurred in the wake of the 2008 financial crisis and more recently following the Brexit vote, no longer provides sufficient stimulus to bring foreign trade back into balance.
Of course, the UK does not rely just on manufacturing to pay for imports. The economy has proved exceptionally good at selling services to the rest of the world -finance, legal, design, media, education and the rest -but this success has never been enough to offset the industrial failures. The fact remains that around half the value of all UK exports still comes from manufacturing and that manufacturing, with just 10% of the UK workforce, sells as much to the rest of the world as the other 90% put together.
The point here is that contemporary public finances are undermined not only by the direct cost of welfare benefits in former industrial Britain but also by the inability of a weakened manufacturing sector to deliver the sustainable growth that the UK economy so clearly needs.

THE ORTHODOX SOLUTION: WELFARE CUTS
Faced with a budget deficit that is high and has proved slow to bring down, successive governments have chosen to tackle the symptoms rather than the underlying causes. A key symptom has been the high spending on working-age welfare benefits. As we have argued, this should really be understood as a result of economic failure rather than of financial generosity. This is not, however, how the Treasury has interpreted the problem.
The pre-2010 Labour Government was far from immune to Treasury orthodoxy and from 2008 onwards began to replace old-style incapacity benefits by ESA, with a new medical test and greater conditionality. The Coalition Government that then took office introduced a major round of cuts to welfare benefits and its wholly Conservative successor, elected in 2015, has carried on with a further major round.
A distinguishing feature of the welfare reforms is that they focus almost exclusively on working-age claimants. By contrast, spending on state pensions -by far the largest component of welfare spending -has been entirely unaffected. The quite explicit assumption has been that reductions in working-age benefits incentivize claimants to find work. By targeting in-work benefits as well, the reforms also assume that reduced entitlement will encourage claimants to find a better-paid job or work longer hours. All this has been backed up by increasing conditionality and the wider application of sanctions. To put this another way, the assumption framing welfare policy is that unemployment and low pay are the fault of individuals. Claimants have let themselves become 'dependent' on welfare benefits and they should choose to 'do the right thing' and instead find work or increase their earnings. 8 This is a quite different view to the one set out here, which is that high spending on welfare benefits is the result of economic failure. The Treasury's orthodoxy makes the mistake of taking welfare spending out of its economic context. If the Treasury had a better understanding of what has happened to the economy of older industrial Britain it might not be so keen to blame welfare spending on the workshy or feckless.
We first documented the uneven impact of the welfare reforms in a 2013 report (Beatty & Fothergill, 2013). An updated report calculated that by March 2016 the reforms had resulted in a loss to claimants of £14.5 billion a year (Beatty & Fothergill, 2016b). The Treasury's own figures show that the reforms now in the pipeline will result in a further loss of £12.3 billion a year by 2020/21. 9 That brings the cumulative total to almost £27 billion a year. Figure 7 shows where across Britain this financial loss is occurring. 10 The methods underpinning these estimates are set out in our 2016 report. In essence, they take the Treasury's own estimates of the financial loss and translate them down to the local level using a range of official statistics on claimant numbers and spending. The measure shown in this map is the average loss per adult of working age -in other words, the financial loss spread across all 16-64 year olds, including those not in receipt of welfare benefits. This is the best measure of the intensity of the 'hit' in each place. The map shows the annual loss that can now be expected by 2020/21 as a result of all the post-2010 welfare reforms.
If this map shows similarities with those presented earlier it is not accidental. Welfare cuts inevitably impact most in the places where claimants are concentrated. Therefore it is no surprise that Britain's older industrial areas figure so prominently among the worst-hit places. Once more, it is places such as South Wales, the industrial North from Merseyside across to the Humber, North East England and the West of Scotland that stand out, whilst large parts of southern England around London are much less affected.
In Middlesbrough by 2020/21, the financial loss from all of the post-2010 welfare reforms is estimated to exceed £1,000 a year per adult of working age. For those actually in receipt of welfare benefits the average financial loss will obviously be larger -sometimes much larger. In Bradford the average loss per working age adult is £970 a year, in Oldham £950 a year and in Merthyr Regional StudieS, Regional Science Tydfil in the heart of the Welsh Valleys £920 a year. Across most of older industrial Britain the loss exceeds £750 a year. In Cambridge the equivalent figure is just £340 a year.
In fairness, older industrial Britain is not the only place hit hard by welfare reform. A number of less prosperous seaside towns are also hit hard, and a number of low-wage rural areas. Some parts of London also lose large sums, particularly because of the reductions in support for housing costs. Nevertheless, it is not difficult to argue that communities in older industrial Britain now face punishment in the form of welfare cuts for the destruction wrought to their industrial base all those years ago.

IS THERE IS AN ALTERNATIVE?
But if welfare cuts represent the orthodox Treasury-driven response to the budget deficit, is there an alternative? Two principles should perhaps be central.
First, the rhetoric about rebalancing the UK economy needs to be turned into reality. In the wake of the 2008 financial crisis there was much talk about the need to move away from an over-reliance on financial services towards an economy based more on exports and investment. This did not happen. If anything, the UK economy became more imbalanced as growth returned because the old model based on debt and the housing market was rekindled one more time. At first sight, the proposed new industrial strategy (HM Government, 2017) seems insufficient to mark a turning point.
A genuine revival in industrial production would be central to any rebalancing of the UK economy. It is salutary to remember that in Germany, where labour costs are generally at least as high as in the UK, the share of GDP accounted for by manufacturing is twice the level in the UK. In no small part as a result, Germany has a large trade surplus and a far smaller budget deficit. A rebalancing of the UK economy in favour of industry would be of direct benefit to much of older industrial Britain because, even after years of job loss, that is where so much of what remains of UK manufacturing is still located.
The other principle central to an alternative to the Treasury's welfare cuts is a revival of regional economic policy. The places where welfare claimants are concentrated, out of work or on low wages, need to be grown fastest. At the present time, the UK probably has its weakest regional economic policies since the Second World War. In England, for example, the well-funded Regional Development Agencies were abolished in 2012 and the Regional Growth Fund, which co-financed business investment in less prosperous areas, closed to new business in 2015. Looking ahead, the UK's impending departure from the EU threatens to weaken regional policy still further if there is no replacement for the financial support from the EU Structural Funds.
Indeed, what masquerades as UK regional policy is more often the promotion of competition between places, which in practice often widens the differences in economic well-being, or the devolution of powers to local authorities, which is really about governance and has the most tenuous connection to prosperity. The dominating position of London, in particular, has gone unchallenged and London's economy has pulled still further away from the rest of the country even though the downsides of the capital's success -congestion and high property prices -are all too evident.
The starting point needs to be that the economies of older industrial Britain can be rebuilt. The prize is lower spending on welfare, higher tax revenue and a reduction in the budget deficit that is not based upon hitting the poorest place hardest.

LESSONS FOR ANALYSIS AND POLICY
In conclusion, let us return to the theme at the very start of this article: that the focus on the short term obscures longer-term issues and trends. We have endeavoured to explain here how the destruction of industrial Britain in the 1980s still has profound repercussions for present-day public finances. What does this tell us about the way policy-makers and researchers should go about understanding issues, in Britain in particular but also perhaps in other mature economies where there has been profound and unevenly concentrated deindustrialization?
First and most obviously, it underlines the importance of a long-term perspective. Where we are now, as a society, is the product of long and still evolving economic processes. The financial crisis of 2008 is not the defining event in Britain's recent economic history, or even the main cause of the present budget deficit. The source of many current problems lies much deeper in the destruction of Britain's industrial base and all that has flowed from it.
Second, it is hard to understand what is happening to the economy or society without looking at the differences between places. It is disturbing that the Treasury and most of the economics profession rarely, if ever, look beyond national data and national trends. They end up failing to grasp causality and misdiagnose problems.
Third, there is a pressing need to stop thinking in silos. Jobs, or the lack of them, and public finances are profoundly interconnected. Allowing Britain's industrial base to wither so dramatically has not been costless and it has certainly not been absorbed by the smooth operation of market forces. It has resulted in persistent worklessness, low wages and an inflated welfare bill.