key: cord-0046271-ctvwm0kq authors: Mayhew, Ken; Anand, Paul title: Covid-19 And The Uk Labour Market date: 2020-06-02 journal: nan DOI: 10.1093/oxrep/graa017 sha: 1d35f70ff9ca6e1613fe95edfceff33815db777a doc_id: 46271 cord_uid: ctvwm0kq This article considers policy responses to the COVID-19 crisis as they affect the labour market, how these policies are evolving and some of the design issues they face. The concentration is on the UK, but other countries are also discussed for comparative purposes. The Job Retention Scheme is a successful innovation to keep temporarily stopped workers attached to their employers. However, since economic recovery will be slow, it is not sustainable in its current form. A sustained rise in unemployment is inevitable and alternative policies to mitigate this and the dangers of scarring are discussed. The structure of output will change, as therefore will the composition of jobs. A comprehensive active manpower policy will be needed to efficiently match job seekers to available jobs. The young are likely to suffer disproportionately from the recession and this makes it essential to introduce radical policies to boost work-based training and to enhance the contribution made by further and higher education institutions. This article considers policy responses to the COVID-19 crisis as they affect the labour market, how these policies are likely to evolve, and how the crisis might shape the labour market in the longer term. The concentration is on the UK, but other countries are also discussed for comparative purposes. The COVID-19 crisis has been described as a twin pandemica grave and global challenge to the health and wealth of nations. On the last day of 2019, the World Health Organisation received reports of a cluster of unusual pneumonia-like cases in the Wuhan province in China. The virus spread across the world and had reached the UK by the end of January at the latest. Concern about the economic implications started to emerge, though they lagged by several weeks. However, by mid-February 2020, the OECD was warning of the possibility of a serious global recession and by 11 March the UK's Chancellor and the Bank of England had embarked on the creation of fiscal and monetary policies that increasingly recognised that the shock was less of a conventional business downturn and more of a potentially existential threat to the national economy. As the virus continued to spread, on 25 March the Prime Minister announced stay-at-home orders covering all but essential front-line workers which had the effect of closing down significant segments of the economy literally overnight. One feature of the crisis is that events are moving more quickly than the data available to interpret them. On 19 March, the OECD (2020)estimated that lockdowns would initially reduce the level of output by 20% to 25% 'in many economies'. The longer run impact on GDP growth, it argued, would depend on the duration of lockdowns and the manner and extent to which they were gradually eased. Every month of 'strict containment' was calculated to reduce annual GDP growth by 2 percentage points. If anything, the Resolution Foundation (2020) was even more pessimistic about the UK's prospects, suggesting an impact effect of a fall of one-third in economic activity. The Foundation suggested that if the lockdown lasted for three months in more or less its present form, then 2020 would see a 10% fall in GDP. If it lasted for 6 months the fall would be 20% and 24% if it lasted for 12 months. The corresponding unemployment rates would be 5.4%, 14.1% and 20.8%. At the end of the first week of May the Bank of England was warning that output might fall by 14% during the year. Indeed, unemployment rates have already risen significantly across the OECD. In the UK the initial rise was more modest than in some other countriesthough in the last two weeks of March the number of Universal Credit claimants was ten times higher than the usual number in any given fortnight. The German unemployment rate has also risen relatively slowly, though commentators expect it to rise to about 6% later this year before falling back again. This contrasts sharply with the US where unemployment grew to over 30 million within the space of just over a month (Lambert, 2020) . In France the number of unemployed increased at the fastest rate since records began. Such early differences across countries can be explained by different policy responses. A centrepiece of UK policy has been its Job Retention Scheme (JRS) or 'furlough' scheme, whereby employers receive 80% (up to a limit of £2,500 per month) of the wages of employees who are temporarily stopped. It became active in late April and was initially set to last until the end of June. Early signs were that there would be a very high take-up and the logic of the policy is clear. Keep workers attached to their employers so that they can easily slot back into their jobs once lockdown restrictions start to be lifted. It is meant to avoid large-scale redundancies and the subsequent problems of large numbers of unemployed people trying to find their way back into work. Other countries have introduced policies with the same general aim, whether through furlough schemes, subsidised short-time working or other forms of wage subsidies. These countries include France, Germany, the Netherlands, Portugal, Sweden, Spain, Sweden and Switzerland (Fujita et al., 2020) . The precise content and strength of such policies differs from country to country. Even the US offers loans to firms who furlough their workers, but the reach of the policy seems quite weak, which in large part accounts for the huge increase in unemployment there. The UK's JRS is one of the more effective responses to the current crisis but inevitably it leaves large gaps. Some 5 million workers in the UK are self-employed while another 1 million are in temporary jobs, in the gig-economy or on zero hours contracts. Because the scheme failed to include these groups, some 20% of the UK's workforce were not covered and it took a couple of weeks for a separate policy directed at the self-employed to be developed and even that failed to effectively include many workers. The JRS policy raises the question as to whether 80% is enough or possibly too much for an optimal sharing of risk between businesses and taxpayers. Again, the absence of sufficient data makes it hard for government to know how generous it has to be (Anand, 2020), but reports of website use when the scheme opened for applications suggest that this aspect of the government's response will be fairly effective in the short term. By the end of its first week, no less than 4 million workers had been furloughed under its provisions, by early May the figure was 6.3 million and it was expected it would rise further to 9 million (about a quarter of the UK workforce). Certainly, when it comes to ease of take-up, it is far superior to the business loans schemes, not least because there is no need for commercial bank involvement or solvency tests to be applied to the employers who apply. The scheme for the self-employed appears to be working far less well, with sustained complaints being voiced about eligibility rules amongst other difficulties. In addition to the measures discussed above, social security benefits have been enhanced for those who have lost their jobs or who were already unemployed. Inevitably, however, a very large proportion of the UK population have experienced a fall in disposable incomethe Resolution Foundation estimates 68% of households. Younger people appear to be taking the hardest hit. A disproportionate proportion of them work in low-paid jobs which cannot be done from home (Hensvik et al., 2020) , and therefore they are more liable to be laid off or put on the Job Retention Scheme (without their employers necessarily making up the rest of their pay); and they are disproportionately represented in two of the hardest hit sectorsretail and hospitality. Even if they retain their jobs or get new ones, Bell et al. (2020) predict that young workers in small firms could see earnings losses of 8 to 9 %. According to the same authors, older women in large firms are also exposed to significant earnings losses. So, the Job Retention Scheme is undoubtedly a bold policy initiative but there are problems. It is open to employers to make up the rest of an individual's pay so that there is no loss of income at all, but it is too early to tell how many employers will do this. Even more important is the longer-term viability of the scheme. Initially governments were hoping for a V-shaped recovery, a quick bounce back from the present downturn. This depended on what now appear to be overly optimistic views as to how quickly the lockdown could be lifted and the economy could return to 'normal'. The crucial policy instruments depend on the ability to control transmission by social distancing and test-and-trace at scale. Can places of work, shops, restaurants, buses, trains and air-transport ensure enough distancing to limit transmission, and can sufficiently accurate and comprehensive testing and tracing be conducted to identify individuals who should self-isolate? Can workplaces and schools be perceived as sufficiently safe that they will attract back consumers, workers and pupils? The hope for a rapid V-shaped recovery was based on a short period of social distancing which might have been warranted if the outbreak had become self-limiting by the summer, as did SARS in 2002-4. There is still considerable uncertainty about duration. That said, it seems inevitable that some forms of social distancing will remain in the UK for many months to come, in which case it will take a long time for overall employment rates to recover. In truth, it is likely that return to normality will be limited until effective treatment drugs become widely available and/or mass vaccination is possible. Although the government is helping the private sector with a panoply of subsidised loans and tax waivers, the longer restrictions on normal economic life remain, the more companies will cease trading. Hopefully, the majority will survive but even they are likely to reduce their workforce or put significant proportions of workers on limited hours in the belief that demand for their products or services will not recover in the foreseeable future. The retail, hospitality tourism, leisure and parts of the transport sectors are especially vulnerable. Indeed, we already have early indications. The high street retail sector was already in trouble before the virus hit and Debenhams, for example, has gone into liquidation and has announced plans for store closures. Employers in other sectors, HSBC for instance, have simply postponed major redundancies. British Airways, calculating that it will take years for air travel to recover to pre-crisis levels, has made initial proposals for a large programme of redundancies with a switch to zero-hours contracts being offered for those that remain. Virgin Atlantic has also announced plans for 3,000 redundancies. Already some employers have taken the redundancy route rather than apply for the JRS. On 12 May the Government announced that it would extend JRS in its present form for a further monthuntil the end of July. However, there are two problems. The first is that the longer the period of time the scheme is extended, the greater the burden on public financesit has been calculated that it is currently costing £14bn per month. Put crudely, it is much cheaper to pay people out of work benefits than it is to maintain them on schemes like the JRS. Already by early May the Treasury was suggesting that it was unlikely that the scheme could continue in its initial form and in his 12 May announcement, the Chancellor said that, although the government was committed to continuing the scheme until the end of October, from the beginning of August employers would have to start sharing the costas of yet the details of this cost sharing are unspecified. The second problem is that the greater the number of employers that start to foresee medium-term downturns in demand, the more employers will withdraw from any extended JRS and instead lay off proportions of their workforces. Many universities, for instance, foreseeing their income streams compromised at least through the 2020-21 academic year, are drawing up redundancy plans. A time-extended JRS scheme may moderate the rise in unemployment but some significant rise is inevitableit is simply a matter of how great that rise will be. Now that a quick bounce-back of the economy to where we were seems highly unlikely, we have to think about the longer-term consequences for the labour market. Some insights can be gleaned from Germany's policy initiatives during the 2008 financial crisis and in its aftermath. Prominent amongst these was the 'short-time-work' (STW) scheme which paid firms 67 per cent of an individual's loss of pay resulting from shorter working hours for up to two years (Bosch, 2011). German government policy was possibly unusual at that time, but it seems to have influenced the UK and other countries in their recent adoption of furlough and similar schemes to keep employees attached to their organisations. A whole series of other features also helped it to limit the rise in unemployment. Burda and Hunt (2011) , for example, suggest that German employers tended to take a long-term view of employment decisions and had taken on fewer workers in the runup to the financial crash and as a result had to lay fewer off subsequently. Wage moderation was also important but so too was the widespread adoption of working time accounts. These essentially work on an averaging principle. The account lasts typically for anywhere between 6 months and 2 years. It allows for a worker to work longer or shorter hours in any given week, as long as the average over the relevant time period corresponds to the agreed weekly working hours, without having to experience variations in pay. The additional flexibility provided by working time accounts reduced the incentive for employers to lay workers off in the downturn. However, all the advocates of these instruments which provide flexibility agree that they only work well if there is an external demand shock which is short-lived (Arpaia and Curci, 2010) . This should be taken as a cautionary noteif social distancing persists into the medium term and has both supply-side and demand-side consequences, the furlough-type instruments countries are adopting are unlikely to be sufficient to prevent significant shrinkages in the demand for labour, even if they were deemed affordable. It is almost certain that by the end of the year unemployment will have reached highs we have not seen in the UK since the early 1980s or possibly even since the Great Depression of the early 1930s. Historical experience tells us that this creates problems far beyond the immediate because of 'scarring'. To understand how labour markets are evolving, and are likely to go on doing so, it makes sense to draw on what we know from previous recessions, while adjusting for severity, connections to health dynamics and a very uneven distribution of impacts across sectors of the economy. It is known, for example, that some groups of workers are disproportionately affected by downturns in economic activity and, in particular, that those entering the labour market for the first time can suffer from long-term negative impacts. Recovery from the financial crisis of 2008 also followed very different trajectories among higher-income countries, as Table 1 indicates. It compares 2010, by which time the impact effects of the crisis had worked through to unemployment, with the situation in 2018. Coulter (2016) argues that economic recovery, at least in labour market terms, in the UK was impressive on the grounds that significant reductions in public-sector employment were made up for by the creation of jobs in the private sector. However, he also notes that this came at the expense of falls in real wages and stagnant productivity with some evidence of increasing labour market polarisation. Many of the jobs created in the employment boom were low quality and, partly as a consequence of this, in-work poverty increased. It was not until at least 2014 that substantial numbers of good, full-time jobs started to emerge, a finding that mirrors experience in the US. It took nearly a decade for the real wages of UK workers to recover to their pre-2008 levels, shortly before the COVID-19 crisis hit. Allegretto (2018) observes that it took 6 years for the US economy to recover its jobs, that disparities in unemployment by region and race/ethnicity persisted despite a decade of sustained if weak recovery, that reductions in poverty were positive but modest, and that average household incomes also took 10 years to reach their pre-crisis levels. Germany managed to achieve a labour market recovery with less cost to real incomes. Declines in unemployment came more slowly in Spain where unemployment had peaked at a much higher level. Growth there also emerged around 2014 and continued until the onset of the present crisis, but there is some evidence that wage incomes were gradually falling to align better with productivity. These different experiences all suggest that recovery from a sharp spike in unemployment could take a decade and that it may come at the expense of real wages. Different this time around is the fact that rather than being focused on markets for housing and lending, there are risks to several sectorsaccommodation and food, real estate, business and administration, manufacturing, wholesale and retail and vehicle repair, transport, storage and communication, arts entertainment and recreation (ILO, 2020). It is also worth noting that the growth spurt reflected particularly in stock markets from around 2015 appears to have coincided with a dramatic fall in oil prices due to the emergence of shale energy production in the US. If the re-emergence of good-quality jobs requires a large positive shock to economies in years to come, its possible sources are far from obvious, though another article (Hepburn et al., 2020) in this issue argues that initiatives to promote green growth could be one. The potential consequences for many individuals are frightening. The research literature on scarring tells us that the experience of unemployment is associated with higher subsequent probabilities of worklessness and lower pay. In addition, there are also negative consequences in terms of poor mental health and increased perceptions of job insecurity with consequences that can be long lived (Daly and Delaney, 2013; Nilsen and Holm Reiso, 2014) . If prevention is the best policy, as Arulampalam (2001) concluded some time ago, and if, as we expect, the Job Retention Scheme loses its bite, then many workers face lifechanging threats. Scarring will be a particular problem for young people, and especially for those who have just entered the labour market or who are about to enter it. Many university students about to finish their courses are still to find jobs. Some who thought that they had secured employment are finding those job offers being withdrawn. School leavers who might have hoped to enter jobs which offered apprenticeships or other substantive training will find their hopes dashed. Some will become unemployed or inactive. Others will get jobs but lowerlevel ones than they expected. They will be mis-matchedin jobs which do not fully use their capabilities and are correspondingly low paid. The evidence tells us that historically it has been hard for individuals to switch to more suitable employment and the longer they remain mis-matched the harder it becomes. Research from the Resolution Foundation dubbed those young people who entered the labour force during the 2008 financial crisis the 'crisis cohort'. The research concluded that they 'continued to face higher unemployment, lower pay and worse job prospects up to a decade later as compared to other young people entering work before or after the downturn'. It seems likely that, at least in the medium term, we will emerge from the crisis with a rather different structure of production and therefore of available jobs. Firms in hospitality, retail, leisure and entertainment, some personal services and parts of transport will be slow to recover to pre-crisis levels, not least because of continuing social distancing rules. In the process some of them will go to the wall. Clearly many, but not all, of the lost jobs will be relatively low-paying ones. What sort of jobs will be available for the displaced workers? One possible development could be the corollary of long-standing concerns, exacerbated by the COVID-19 crisis, about the economy's exposure to the globalisation of value chains. Rincon-Aznar (2020) has estimated that global spillovers through trade linkages will substantially exacerbate the economic impacts of the COVID-19 shock. In the face of supply disruption and rising global prices, businesses potentially have a strong incentive to consider whether more reliance should be placed on the domestic production of some items, including medical supplies, foodstuffs, and information and communications technologies. At the margin, this might involve bringing back onshore some forms of employment from countries where labour costs are lower. The government's new migration policy, together with the impact of Brexit, could well create large numbers of job opportunities in those sectors which have been heavily reliant on EU migrant labour. In particular there may be significant increases in the demand for UK workers to fill relatively unskilled jobs in agriculture and food processing. Similarly, the reduced hospitality and retail sectors may have to become more reliant on local labour. But it is not just a matter of finding work for those at the lower end of workforce. There will also be demand for more skilled employees. The issue is whether the skills available will match the skills needed. Only some of the largest companies appear to have the will and the resources to shift skillsets en masse: at least one UK-based oil major is retraining large numbers of engineers in artificial intelligence, but this looks like an investment that most other companies would find difficult to pull off. The UK entered this crisis with a shortage of STEM (science, technology, engineering and mathematics) workers at all levels, from highly qualified university graduates to technicians at level 3 or 4the equivalent of A levels and just above (Caselli and Manning, 2019) . Progress had been extremely slow to address these problems; not least because attempts to reinvigorate the apprenticeship system appeared not to be working. To make matters worse, the COVID crisis is threatening the intake and the training of new apprentices, whilst the university sector is in financial disarray. In particular, the heavy reliance of many institutions on foreign students, who will arrive in much smaller numbers this autumn, is intensifying cash-flow problems. In terms of the medical and scientific reactions it is clear that some countries have tackled the COVID-19 crisis more competently than others. This will have implications for its longerterm economic impact. Those countries which have done well will be able to return to something akin to economic normality more rapidly. Sadly, the UK is not among this group of countries. Therefore, the emergence from full lockdown is going to be slower than it might otherwise have been. Early reaction to the Government's so far very tentative suggestions for moderating the lockdown illustrate some of the difficulties. The TUC and a number of individual unions are voicing concern that arrangements to maintain social distancing are inadequate and therefore threaten the health of the workers involved. Against this background, we highlight some of the near-term policy issues.  The policy, via the JRS, of attempting to keep as many people as possible attached to their current employer has seen early success. It should be kept in place as long as possible. However, it is costly for the government finances as compared to the net cost of paying out even enhanced social security benefits and the loss of tax revenue. So, some early reduction in the 80% subsidy and the £2,500 upper limit is inevitable.  Furthermore, the longer various forms of lockdown remain, the more companies will decide on permanent reductions in the workforce or, at the extreme, go out of business. To an extent this can be mitigated by financial help from the government. At present the support available varies somewhat from devolved government to devolved government, but largely takes the form of tide-over help for the depths of the crisisloans, small grants, rent holidays, postponement of VAT payments etc. The Chancellor has recently announced a modest 'bounce-back' loan scheme for small businesses. How much further the Government will be willing and able to go is a matter of speculation. A possible alternative to JRS would be a much more targeted scheme to help employers. One candidate would be a version of the German working time accounts, described earlier. The government could provide loans to companies to cover cash-flow issues consequent on the averaging procedures. Interestingly, the Chancellor's 12 May announcement indicated that employees would be able to work part time under JRS. It could be a fairly easy step from this to some form of working time account arrangement.  Now that a V-shaped recovery of the real economy is not going to happen, unemployment will rise. It is simply a matter of how far it will rise. In an already unequal country, income inequality will increase. At the very least, one would hope that the government maintains the enhanced levels of social security benefits. Beyond that there is a need to consider policies to diminish the extent of unemployment scarring. Welfare-to-work type policies may get people rapidly back into jobs but too often they are badly paid, dead-end jobs. Historically this has been the essence of scarring, particularly for young people (O'Higgins, 2012) . A whole new look at active labour market policy is required. This would involve not just matching people to the available jobs, but also trying to influence the quality of those jobs and ensuring that individuals are given the training necessary to obtain them. This depends upon the capacity within government and its agencies to deliver it. Our history is not reassuring in this respect. The Department for Work and Pensions and Department for Education have not worked well together. JobCentre Plus has little resource to engage with employers or those seeking work, let alone to keep them in work. The Office for Budget Responsibility (OBR, 2018) has recently commented scathingly on the role and pay of the Work Coaches who are central to the workings of Universal Credit. In other words, huge capacity building is necessary, as is a degree of interventionism (see below) that has become almost alien to British policy thinking, but unusual times require unusual reactions. Beyond securing a viable active manpower policy, capacity building will be needed simply to secure the efficient handling of the vastly increased number of Universal Credit and other social security claimants (OBR, 2018).  Work-based training had declined even before COVID-19. Radical action will be needed to give the best possible chances to those leaving education. The UK's Apprenticeship Levy (a tax on employers to fund training) has not worked as intended and it should be suspended. Instead, there should be a subsidy to firms offering apprenticeships to young people. Much tighter controls than we have seen thus far will be neededcontrols on the content and quality of the apprenticeship and on the extent of training provided. These apprenticeships could be at level 3 (equivalent of A levels) or even level 2 (equivalent of GCSEs) and could provide an important pathway for a significant proportion of youngsters who might otherwise experience extreme scarring (see Goodhart (2020) for a fuller version of a similar proposal).  Meanwhile the country's higher education sector is in serious trouble. The Government has already offered some emergency funding. But the sector could play an important role in active manpower policy, as has recently been suggested by Goodhart (2020) . This policy would require better labour market intelligence than is currently available. Information could be gathered on the employer demand for individual occupations/skills and on the range of pay offered for these occupations/skills. This information would be made widely available to workers looking for a job and to workers contemplating a change of job. Some of these occupations would require quite short courses, others longer ones. Further education (FE) and higher education (HE) institutions could be encouraged to offer these courses and be paid a fee directly by the government, whilst individuals would receive a maintenance loan to cover their living expenses for the duration of the course. This would bring much needed funding to the FE sector, which has been starved of money for many years. It would also bring funding to cash-strapped HE institutions as long as they were prepared to change the nature of their offer. They would concentrate less on 3-year bachelor's courses and more on shorter sub-degree courses and, in a way, some of them would be moving back towards their old functions as polytechnics or specialist institutions (Goodhart, 2020). The spread of the COVID-19 virus and the public health response have created a shock to employment patterns that has been deeper and more rapid than any experienced since the first half of the twentieth century. Incomes and employment have been partially protected in the short run, but long-term impacts will be significant. Unemployment will riseit is just a question of how far. Government action will be needed to get people back into jobs quickly, to minimise scarring and to achieve optimal matching between the skills that individuals possess and the skills that employers want. Young people are at particular risk, especially those leaving education. In the UK, recovery from the last recession was accompanied initially by falling, and then only slowly rising real wagesreal wages getting back to their pre-2008 levels not long before the COVID-19 crisis struck. Worryingly, evidence is beginning to emerge that this crisis may accelerate the secular rise of zero-hours contracts and other low-quality forms of work, in the absence of further legal protections and other forms of government intervention. If the average employee turns out to be no better off after the next 10 years than they were at the beginning of this year, then effectively we will have seen 20 years of stagnant real income. In an already highly unequal society this would be a disaster. These are unprecedented times and they call for unprecedented measures. A Post-Great Recession Overview of Labor Market Trends in the United States and California EU Labour Market Behaviour During the Great Recession, Economic Papers Is unemployment really scarring: effects of unemployment experiences on wages The German labour market after the financial crisis: miracle or just a good policy mix What Explains the German Labour Market Miracle in the Great Recession Robot Arithmetic: New Technology and Wages The UK labour market and the 'great' recession The scarring effect of unemployment throughout adulthood on psychological distress at age 50: estimates controlling for early adulthood distress and childhood psychological factors The Labour Market Policy Response to COVID 19 must Save Aggregate Matching Capital, VOX CEPR Policy Portal Will COVID-19 fiscal recovery packages accelerate or retard progress on climate change? Monitor 1 st Edition: Covid 19 and the World of Work: Impact and Policy Responses Real unemployment and the US has now lost 26.5 million jobs The scarring effects of early career unemployment This time it's different? Youth labour markets during 'the Great Recession Resolution Foundation (2020) Doing More of What it Takes: Next Steps in the Economic Response to Coronavirus Global value chains after COVID 19 and Brexit: is this the end of the world as we know it?