Governments and central banks in developed countries are hastily deploying aggressive macroeconomic policies of different shapes and types to fend off a global economic catastrophe caused by the COVID-19 pandemic. Despite the urgency, whether real or perceived, of these measures, it is essential to design these policies carefully. We have, at best, one shot to get this right.
Our research on the aftermath of the Great Recession of 2008-2009 in the US indicates a very strong propagation mechanism of that large shock to labour markets, which contributed to the ensuing slow recovery. In our view, it is now essential to design policy with this lesson in mind. We propose guidelines to target government support in the next few weeks and months. We do not discuss the equally important issue of how to pay for this support, but only where it should go.
We need to avoid blanket, untargeted help, but rather aim to create the conditions to preserve worker attachment to their old jobs and, once the crisis subsides, quickly restore the pre-existing allocation of employment to sectors and companies.
In a nutshell: we need to avoid blanket, untargeted help, but rather aim to create the conditions to preserve worker attachment to their old jobs and, once the crisis subsides, quickly restore the pre-existing allocation of employment to sectors and companies. The pandemic per se should not represent a permanent shock to preferences and technology, except possibly for public health demands. Any policies that would be conducive to a major and persistent reallocation of employment following such a deep but temporary and unusual shock are likely to result in sudden destruction of firm-specific human capital and customer base, slowly accumulated over many years of investment, experimentation and selection, an unnecessarily prolonged stagnation in productivity, and an anaemic recovery. Instead of a probably unavoidable, and painful, V-shaped recession, poorly designed policy could cause another lost decade, or two. Policy should provide insurance, not fundamentally alter the fabric of the world economy. At the same time, significant temporary employment reallocation is needed to face the emergency.
The policy principles that we propose aim for a dual objective: to facilitate quick but temporary emergency reallocation of employment, while preserving at all costs workers’ attachment to their previous jobs.
After the Great Recession, much debate revolved around the dramatic fall and slow recovery in employment rates and wages in the US in 2008-2019. We want to highlight two different sets of facts, which should clearly resonate in the minds of many workers right now.
The facts from the Great Recession aftermath
The first set of facts is presented in Fujita and Moscarini (2017).
- Workers who lose their jobs, but are eventually recalled by their last employer, almost always return to the same occupation and experience no earning loss, whether they expected to be recalled (were on temporary layoff) or not.
Even among job losses that are perceived as permanent by firm and employer, one in five end up in a recall, with noticeably better outcomes.
- In contrast, workers who are permanently separated from their employer and are not recalled suffer a very significant loss in monthly earnings, ranging between 2% and 12% in impact depending on the ensuing duration of unemployment, and often change occupation and career.
Longer-tenured workers are more likely to be recalled than more recent hires, and recalled workers stay longer at the firm thereafter than new hires.
We take this set of facts as evidence of sizable firm-specific human capital, which contributes to a worker’s welfare and productivity. This capital can take many forms: knowledge about details of a company’s operations, colleagues, commute to work, or proficiency in just the right combination of general skills for that particular job.
Much research has been devoted to the impact of a job loss on a worker, through no fault of their own. Studies of job separations – especially plant closing (Jacobson et al. 1993) or mass layoff (Davis and von Wachter 2011) episodes, when recall is typically not an option – estimate dramatic short-run and long-run losses in earnings and employment, especially when these episodes occur in a downturn. Our work extends beyond plant closings and mass layoffs, which are relatively rare events, and emphasises the importance of worker tenure.
What this means for the COVID Crisis
The COVID-19 pandemic is arguably as exogenous a shock as it gets, but it will probably not be nearly as persistent as the housing collapse, technological change, or globalisation. Therefore, closings and permanent separations are not inevitable.
We contend that it is an absolute priority to maintain a sense of attachment of workers to their previous jobs, by preventing closings of previously viable businesses which are likely to rebound once the health crisis is resolved.
We contend that it is an absolute priority to maintain a sense of attachment of workers to their previous jobs, by preventing closings of previously viable businesses which are likely to rebound once the health crisis is resolved. Furman (2020) also makes it one of the four priorities for the immediate response. This goal can be accomplished by providing ailing businesses with interest-free loans to cover their fixed costs and converting the loan into a grant, should the business offer to rehire former employees first, when it resumes normal operations.
Any form of subsidies has moral hazard ramifications; in this case, employers might lay off even more workers now, to then rehire them and receive the subsidies. But these incentives would not be stronger than those deriving from unconditional bailouts currently discussed. We just propose to add a ‘recall clause’, which would greatly relieve workers’ anxiety and make them more willing to take the stopgap jobs that are required to face the health crisis, without fear of seeing their career derailed permanently. The goal is to prevent a massive destruction of specific human capital, which took very many years to build, and would take many to rebuild.
Firm-specific human capital is unlikely to be priced correctly, especially when firm and worker are temporarily separated and cannot contract and commit to reunite, although that specific human capital is still viable and would be lost anywhere else. So, policy intervention is indeed called for: government can facilitate that commitment. The policy needs to be designed so that workers who can afford to just wait out the crisis will do so without the fear of losing their place in line for their old jobs. This allows us to avoid potentially large mismatch after this crisis is over.
The goal is to prevent a massive destruction of specific human capital, which took very many years to build, and would take many to rebuild.
The employer-to-employer rate after the Great Recession
This brings us to our second point. After the Great Recession, as is well known, in the US the job-finding rate from unemployment to employment collapsed, causing a surge in both the rate and the duration of unemployment. This steep drop mirrored that of the vacancy/unemployment ratio, a measure of excess labour demand. Simply put, in 2009-2010 there were very few new jobs available, as is likely to be the case in the near future. Yet, as we estimate in Fujita et al. (2019), the rate at which workers moved from employer to employer declined by only about half as much as the unemployment-to-employment rate proportionally.
While job openings almost vanished, employed workers were still able, somehow, to change employer at reasonable pace. The employer-to-employer rate then recovered and was back to pre-recession levels by 2015. At that point, while the economy kept on expanding and unemployment fell, the employer-to-employer rate stalled and started declining. We interpret these very slow-moving dynamics as follows (Moscarini and Postel-Vinay, 2016, 2019).
After the Great Recession, many workers were permanently separated from their jobs. The financial crisis immediately appeared to be extremely persistent for households, many of which had to deleverage. They also expected declining business entry and productivity growth due to their own weak demand, a vicious circle (Greaney and Walsh 2019). Desperate to find employment, workers flocked to the first jobs they could find. As their careers went off the rails, they were happy to accept other job offers, even if few were available. This willingness to take a job, any job, relieved any pressure on wages, which remained stagnant for years, further contributing to weak demand. Only after 2015, a full six years into the recovery, did employer-employee mismatch clear, easy reallocation between companies slow down, and real wages start to rise noticeably.
Policymakers should resist the temptation to throw money around and, instead, aim to protect those who are more likely to suffer in the short run and to gain from waiting out the end of the crisis.
This time is different
An aggressive public health campaign, a breakthrough in treatment, a mutation of the virus, social distancing, or the warmer season can all contribute to bring the pandemic under control within a few months. Policymakers should resist the temptation to throw money around and, instead, aim to protect those who are more likely to suffer in the short run and to gain from waiting out the end of the crisis.
More generous unemployment insurance (UI), both in terms of eligibility and replacement rates, and grants to businesses to cover part of their fixed costs (such as rent, loan interest, and indirect taxes) can both help the more vulnerable and preserve aggregate matching capital, much more efficiently than an equally expensive untargeted programme of cheques mailed to all households, which is part of the discussion in the US Congress. The transfers could be tied to monthly employment losses in that company’s sector as a whole, but, as mentioned above, they should only be forgiven under the condition of rehiring old employees first.
Many other useful measures have been implemented.
- On both sides of the Atlantic, tax deadlines have been pushed back.
- The UK government is further putting in place government-backed, subsidised loans of up to £5 million to help small businesses weather the storm.
- The French government is extending its ‘chômage partiel’ (temporary unemployment) program, effectively covering 85% of wages.
- Many US states waived the one week ‘waiting period’ before receiving UI benefits and the job search requirement, and expanded eligibility to include those who need to stay home to take care of either a child (due to daycare and school closures) or other dependent, who may be sick/quarantined, and those who are themselves sick or quarantined due to suspicion of being sick.
The U.S. Congress, as part of the $2 trillion rescue package, is about to add $600/week to state UI benefits to all workers, for up to four months, which will raise the replacement rate above 100% for many workers and is likely to discourage the desired temporary reallocation of employment. Loans will be extended to businesses who furlough their workers. This is in line with our goal of maintaining attachment. But each single State will have to waive rules that restrict furloughed workers from taking other full-time jobs. The same applies to Italy’s Cassa Integrazione, which has been deployed wholesale.
- Governments in the UK and Denmark are introducing generous wage subsidies of any employee finding themselves unable to work (80% of wages up to a limit of £2,500/month and 75% of wages, respectively).
These subsidies support both employers and employees, as well as maintain their attachment. Therefore, these policies share the very goals that we prioritise. But, again, furloughed workers (at least in the UK) will not be able to take other jobs without losing these benefits.
What we find lacking are provisions that will facilitate both temporary reallocation and a prompt return to the status quo, and that could avoid a permanent reallocation of economic activity and the resulting massive permanent distortions introduced in the face of what is, in essence, a temporary crisis.
Thus far, what we find lacking are provisions that will facilitate both temporary reallocation and a prompt return to the status quo, and that could avoid a permanent reallocation of economic activity and the resulting massive permanent distortions introduced in the face of what is, in essence, a temporary crisis.
Workers are aware that losing their old jobs will fundamentally change their lives. Any policy that makes that prospect less likely in their minds will now both reassure them and provide them with the right incentives to either stay home, as directed, or to work temporarily elsewhere, where needed.
Our proposed policies would also provide idle workers, who would be technically unemployed and UI recipients, more freedom to search and to explore temporary alternative activities, without fear of losing their ‘place in the line’ to rejoin their previous career. Furthermore, by conditioning forgivable loans on an actual reduction in payroll and to cover fixed costs only, they would reduce the incentives of still thriving businesses to take them up.
Davis, S J and T von Wachter (2011), “Recessions and the Costs of Job Loss”, Brookings Papers on Economic Activity 42(2): 1-72.
Fujita, S and G Moscarini (2017), “Recall and Unemployment”, American Economic Review 102(7): 3875-3916.
Fujita S, G Moscarini, and F Postel-Vinay (2019), “Measuring Employer-to-Employer Reallocation”, manuscript, Yale University and University College London.
Furman, J (2020), “Protecting people now, helping the economy rebound later”, in R Baldwin and B Weder di Mauro (2020) (eds) Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, Vox eBook.
Greaney, B and C Walsh (2019), “Deleveraging, Demand and Growth”, manuscript, Yale University.
Jacobson, L S, R J LaLonde and D G Sullivan (1993), “Earnings Losses of Displaced Workers”, American Economic Review 83(4): 685-709.
Moscarini, G and F Postel-Vinay (2016), “Did the Job Ladder Fail After the Great Recession?”, Journal of Labor Economics 34(S1): S55-S93.
Moscarini, G and F Postel-Vinay (2019), “The Job Ladder: Inflation vs Reallocation”, manuscript, Yale University and University College London.