JOBS NOTES Issue No. 15 PROTECTING WORKERS, FIRMS, AND WORKER-FIRM ATTACHMENT DURING COVID -19: ECONOMIC CONSIDERATIONS FOR THE ASSESSMENT OF POLICY MEASURES The economic crisis triggered by the COVID-19 pandemic The purpose of this note is to provide an overview of has been of a different nature than typical recessions. some of the basic economic considerations for the It was not driven by a lack of aggregate demand design and assessment of these policy measures, with resulting from monetary policy mistakes or credit special attention to emerging economies. We start by market frictions. Instead, activity in significant parts of sketching the economic context of the response to the the economy was restricted intentionally—whether by COVID-19 crisis (section 1), highlighting the impacts the government or by private actors—to combat the of the pandemic on firms and workers, and the value public health crisis. As a consequence, businesses that of the ties that link firms and workers. We outline a would otherwise be in perfectly fine shape saw their simple framework for policy assessment that accounts continued existence threatened. for the mechanisms that transmit COVID-19 shocks through the economy. To preserve organizational knowledge and other forms of intangible capital, employer-employee links, and We consider the implications of two key distinguish- firm-specific human capital and to facilitate the recov- ing features shared by many emerging economies ery, governments provided various types of support to on policy making. First, they typically have a larger businesses affected by the crisis and their employees. informal sector than advanced economies, which Globally, policy makers adopted a wide range of policies dilutes the value of firm-worker ties and makes the to address the consequences of this public health crisis distribution of support through firms a less natural (Gentilini et al. 2022; ILO 2022a). option for policy makers. Second, they operate under fiscal constraints that both bind policy makers’ hands Significant efforts have been dedicated to document directly and make generous support programs riskier the patterns of labor-related policy responses across in the medium term. countries by income level, the number of COVID-19 cases, the extent of the output and employment shocks, We then apply this framework to analyze an array and the level of public debt (Contreras et al. 2023). of policies that have been deployed to prevent and But much of the analysis of these programs has been address business failures and job losses in sectors focused on Organisation for Economic Co-operation directly or indirectly affected by the pandemic. We and Development (OECD) countries, particularly on classify these polices as labor market policies meant to the United States (see, for example, the symposium on support workers in their capacity as employees, firms Macro Policy in the Pandemic in the Spring 2022 issue in their capacity as employers, and firms as businesses, of the Journal of Economic Perspectives or the essays as well as generalized income support policies. In in Strain and Veuger, 2023). conclusion, we draw lessons for policy makers. Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 1. THE IMPACTS OF COVID-19 and, by implication, of the value of preserving them. Employer-employee ties are not lost or have low value among informal jobs, which most frequently are in 1.1. The Impact on the Macro Economy self-employment and small establishments with rela- tively low productivity. The COVID-19 pandemic triggered reductions in output not registered in decades: global gross domestic prod- 1.2. The Impact on Firms uct (GDP) shrank by over 3 percent in 2020 (IMF 2021). Unlike the fallout from the global financial crisis, these The negative shocks caused by the COVID-19 pan- reductions in output were in large part the direct con- demic had markedly different effects on different types sequence of intentional decisions made by households, of firms. Firms in businesses directly affected by public firms, and governments to navigate public health risks. health measures and concerns were hit the hardest, in many cases overwhelming firm-specific productiv- Macroeconomists have suggested a variety of models ity levels or size, in clear contrast to typical patterns to assess this unusual shock. associated with recessionary periods. A particularly distinctive result is that at-risk firms were not neces- Prominently, Guerrieri et al. (2022) propose a straight- sarily, or even generally, low-productivity firms or firms forward model that treats the impact of the pandemic, that would not have been able to compete during a at least initially, as a supply shock. Limitations on the ‘normal’ period of depressed aggregate demand. As a ability of firms and workers in certain industries to result, significant amounts of intangible capital—that provide their usual services reduce output in those is, human, social, organizational, informational, and so sectors. This output reduction is accompanied by a on—were at risk of destruction (Corrado et al. 2022). drop in compensation for workers, entrepreneurs, and the owners of firms in the affected sectors as well as In addition to important sectoral differences, some decreased demand for the goods and services produced general patterns have been noted in works such as that by their suppliers. of Bartik et al. (2020). Preserving employer-employee ties through a period of negative shocks is generally Through these channels, what started out as some- easier for larger and more productive firms, as they are thing reasonably conceptualized as a supply shock will less likely to become liquidity constrained and more trigger negative demand shocks affecting the rest of likely to remain solvent. This fact influenced the design the economy, leading to even greater contractions in of many business support programs, including the output. The presence of economies of scale in affected American Paycheck Protection Program, a detailed firms (Céspedes, Chang, and Velasco 2020) or credit description and preliminary assessment of which can market imperfections in an environment of reduced found in Autor et al. (2022a, 2022b) and Hubbard and collateral values (Céspedes, Chang, and Velasco 2020; Strain (2020). Fornaro and Wolf 2020) can exacerbate these problems even further. In contrast, without government intervention, the crisis would also have accelerated the closure of low-pro- Alfaro, Becerra, and Eslava (2020) analyze the impact of ductivity firms through the usual, and in many ways, these shocks on employment, particularly in countries desirable process of creative destruction. These firms with significant informal employment. They find that may have had lower liquidity at the moment of the job losses because of the COVID-19 shocks were con- shock and may have struggled to adjust to the new centrated in industries that are heavily reliant on direct environment even without liquidity constraints. Low- contact with customers and cannot rely on telework performing and smaller businesses may also have a as easily, as one might expect. Informal jobs in those more difficult time gaining access to government sup- sectors and in the wider economy were at particular port programs. In developing countries, in particular, risk. However, while informal employment was hit hard they tend to be informal and thus do not qualify or initially, it also bounced back relatively rapidly. The are hard to reach, or they may find it burdensome to authors interpret this as evidence of the cost associated claim benefits for which they qualify. with (re-)establishing formal employer-employee ties 2 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Furthermore, even conditional on survival, without Even without full employment, such ‘mobile’ workers government support, small firms would more likely be will be able to maintain their consumption levels in a forced to lay off essential workers than large firms. They way that workers who cannot move cannot, due to would therefore be more likely to disappear and face a borrowing constraints. The overall reduction in output steeper road back if they intend to restart operations. without government intervention will thus be greater Small formal firms would, by definition, incur greater if cross-sectoral labor mobility is lower (Gibbons and losses of organizational and firm-specific human cap- Katz 1992). ital than their informal counterparts (Alfaro, Becerra, and Eslava 2020), potentially leading to a process of Employment losses have dramatic negative conse- informalization of the economy. quences for workers and their families as well as the communities in which they live. Cohorts of college 1.3. The Impact on Workers students who graduate into a recession earn persistently and substantially lower wages (Kahn 2010; Schwandt Containment measures and consumer hesitancy were and von Wachter 2019). Individuals who lose their primary reasons for reduced output and employment jobs do not just face large earnings losses but con- in sectors directly affected by the public health crisis. sequences that go well beyond the immediate loss of Occupations that were disproportionately affected wages (Jarosch 2021). They face mental health problems were those that required in-person contact, especially (Farré, Fasani, and Mueller 2018) and are more likely indoors, and are nonessential in the sense that work- to struggle with substance abuse and commit suicide arounds are not overly costly. Overrepresented among (Autor, Dorn, and Hanson 2019; Pierce and Schott those occupations are those of relatively low-skilled, 2020). Negative employment shocks also deter marriage low-income workers who are particularly vulnerable to formation and fertility and increase the risk of children income shocks and who are often liquidity constrained. living in poverty (Autor, Dorn, and Hanson 2019). For example, Bartik et al. (2020) find that drops in These and other problems have immediate as well as employment were greatest in the services sector, espe- long-term consequences for the wider economy and cially in the low-wage segment. Most affected were society as a whole. leisure and hospitality services, including restaurants and hotels, other services, and retail trade. Systematic Less time spent in employment limits workers’ ability differences were also observed in the probability of to gain experience and build human capital of the job loss and labor force exit across sociodemographic types that are hard to acquire in a formal education groups. For instance, the workers most likely to exit setting. Reduced earnings, consumption, and output the labor market, at least in the US, were those above have a direct fiscal impact by both bringing down tax the age of 65 and those without high school degrees. revenue and triggering increased spending on various Around the world, particularly in emerging economies, social insurance and welfare programs. This in turn is women and youth between the ages of 15 and 24 were likely to either trigger tax increases or crowd out the hit hardest by the job losses (ILO 2022b). provision of public goods, harming long-term growth and prosperity. Employment losses in sectors that suffer the conse- quences of reduced demand from firms and workers Increases in crime harm the victims directly; incar- in directly affected industries, on the other hand, will ceration and other forms of punishment are extraor- more closely resemble those we observe in a typical dinarily costly both to the offenders and to society recession. In fact, where full employment was restored as a whole. These costs include but are not limited before the end of the pandemic, unaffected sectors to expenditures on the criminal justice system as absorbed workers from affected sectors and the full well as the health and labor market consequences employment level of output and consumption in these of incarceration and the marker of a criminal record sectors was higher. (Shoag and Veuger 2019). 3 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Children who grow up in poverty are less likely to firm- and job-specific human capital reduces the former complete high school and college and find employ- employee’s earning potential. ment after their formal education ends (Bastian and Michelmore 2018). This leads to intergenerational Preserving employer-employee links may thus allow transmission of the various negative effects of lack of for a faster recovery while protecting both employ- employment discussed here. ers and employees from incurring unnecessary costs. These considerations are of particular importance in Workers who lose their jobs at an older age, on the environments characterized by long tenures and high other hand, have typically accumulated dispropor- levels of firm- and job-specific human capital. Their tionate amounts of firm-specific and industry-specific importance is exacerbated in environments where labor human capital and are, partially because of that, less market regulations make layoffs costly, as they raise likely to find matches that allow them to be as produc- the option value of not hiring after the downturn has tive as they were before the job loss. passed. All these factors combine to reduce the value of employer-employee links significantly in low-skill and These considerations highlight the importance of informal areas of the economy. avoiding mass layoffs, the urgency of supporting those workers whose employment prospects were harmed by the pandemic, and the potential value of supporting 2. EVALUATING POLICY OPTIONS employer-employee links. While the setting of the COVID-19 pandemic was a new one for policy makers, the criteria by which we might 1.4. The Impact on Employer‑Employee Links judge policy choices are not necessarily different from what they would otherwise be. We can classify these While the textbook response to a negative demand criteria in three general categories. shock is for firms to implement layoffs, there are costs associated with finding and hiring the right workers. First, how effective were different policy options? Did Without obvious post-pandemic shifts in demand, it is they achieve the set goals? Did they let households con- suboptimal to incur these costs again after the public sume and did they help firms survive? Did they reduce health crisis is abated to reconstitute the same employ- or at least not increase the spread of the disease? er-employee links that existed before the pandemic. Second, how efficient were different policy options? However, in the presence of borrowing constraints or What was their fiscal impact? How did they affect persistently suppressed demand, it is difficult for firms incentives to work, save, and innovate? Minimizing to sustain a maximum-employment equilibrium and the fiscal burden per business or job saved or for firm-worker links will be lost. This will make it more a given level of consumption or growth potential difficult for firms to recover productivity after the shock preserved is particularly important for emerging econ- passes, which amplifies the effect of the negative shock. omies, which are often more liquidity constrained and more exposed to sudden changes in the international The consequences of worker turnover also vary financial context. depending on the substitutability between incumbent and outside workers. 1 When a firm’s production Third, how equitable were different policy options? process depends on firm- or even job-specific human capital, outsiders are imperfect substitutes for insiders Were households, firms, and workers in similar situ- (Heining and Jäger 2019). Hiring a replacement ations treated similarly? Were vulnerable members of worker post-downturn forces the firm to incur new society protected? Were the effort and risk-taking by frontline workers rewarded appropriately? hiring and training costs, while the effective loss of 1 See Schmutte (forthcoming) for a more extensive theoretical treatment and for an overview of the empirical evidence on the value of employer-employee links. 4 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 We provide a broad assessment of different policy businesses; and generalized income support policies. options based on these criteria. They are grouped Different policy makers will, of course, attach different in four categories: supply-side labor market policies, weights to the pros and cons of various policies. We focused on individuals in their capacity as workers; focus here on a period when efforts to suppress the demand-side labor market policies, focused on firms pandemic were under way. as employers; labor market policies to support firms as BOX A. COVID-19 LABOR MARKET AND JOBS POLICY MEASURES More than two years since the onset of the COVID pandemic, countries worldwide have introduced an unprecedented number of social protection and jobs policies to mitigate the economic effects of the COVID-19 crisis. According to the World Bank COVID-19 Social Protection and Jobs Policy (SPJ) Inventory that tracks policy responses across 224 countries, over 5,700 social assistance and labor market policies were launched or announced between January 2020 and January 2022. More than half (3,436) of the responses are pro- grams and policies that affect the labor market (Kamran et al. 2023). On the supply side of the labor mar- ket, they include policies that help workers maintain their income such as income tax reduction, wage subsidies, employee-side payroll tax cuts, and unemployment benefits. On the demand side of the labor market, they cover policies that help firms retain workers, such as relief from social security contributions; employer-side payroll tax cuts; and wage subsidies, tax credits, or grant payments associated with hiring or retention of workers. Furthermore, the inventory includes firm liquidity support policies that help busi- nesses survive, including tax relief for firms, credit facilities, credit guarantees, and corporate tax reduc- tions, as well as generalized policies that support aggregate demand such as public works for the unemployed and cash transfers to economically active persons. Information on other policies such as labor regulations and active labor market policies such as entrepreneurship support, training, and placement assistance is also collected. Figure A Percentage of countries with at least one labor market policy, by policy category 95 implementing at least one policy 100 Percentage of countries 80 60 63 60 52 39 40 40 34 35 21 20 0 pp hip n or y rs lly s fit t an d ns ks ie ne en pp it io ist an t t s s ce or pe ca io or su uid on id su urs ct be ym at w e mi ss g bs du liq t a nin ne ul o ic su tiv no re pl eg bl re rm en rai e ac co m x Pu ep ag rr ta Fi ne em T e bo tr W to e U En m La CT co ac In pl Income support policies Active labor policies Labor regulations Firm liquidity support policies Source: Global Database on Social Protection and Jobs Responses to COVID-19 (2021). Own elaboration, based on Kamran et al. (2023). 5 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Among countries implementing these policies, all but one in the sample, or 98 percent of countries, intro- duced at least one labor market policy (Figure A), with an average of 13 labor market policies each. Most labor market policies, about 60 percent, are newly adopted by governments. This is true irrespective of countries’ income level and may stem from the lack of preexisting policies in developing countries (Contreras et al. 2023). The remaining are adaptations of existing programs such as expansion of the number of beneficiaries. Although expenditure data are incomplete, recorded expenditures on labor market initiatives were substantive, amounting to an average of 3.59 percent of GDP (US$32.6 billion) across countries (Kamran et al. 2023). It is worth emphasizing that all the policies discussed the other hand. The fundamental difference between here can be targeted more or less stringently at different these two sets of policies is how we condition on firms (by sector, by size, on the basis of revenue losses) (continued) employment. or households (by employment status or by income). Targeting requires more administrative capability but Effectiveness reduces the fiscal cost at similar levels of effectiveness. A concern about targeting, especially on the basis of Unemployment benefits are particularly effective at characteristics that recipients have control over and channeling funds to workers who would otherwise future characteristics, is that it can produce perverse struggle to maintain their consumption levels. This was incentives and unintended consequences. We highlight particularly important in the context of the COVID-19 these where they are of particular concern. crisis, where we saw dramatic job loss in numerous countries. Employee-side payroll tax cuts were much 2.1. Support for Workers less effective in this situation, as they channel funds to workers who are still employed and not to those A commonly deployed set of policies aimed to provide who have lost their employment. A disadvantage financial support directly to individuals on the basis of both approaches is that the link to employment of their place in the workforce. Some of these are makes it less likely that support will reach workers in countercyclical by nature, such as unemployment the informal sector. insurance benefits, and target workers who have lost market earnings. Other policies supplement market Both types of programs are effective in sustaining earnings or reduce the tax burden on labor income, demand for industries that were not directly affected at least in partial equilibrium: examples are earned by the pandemic, though the propensity to consume income tax credits, wage subsidies, and employee-side will be greater for unemployment benefits, especially payroll tax cuts. given the concentration of job losses among low-in- come workers. A major disadvantage of unemployment All these policies have one goal in common: to allow benefits is that they are necessarily preceded by the workers to maintain their consumption levels in the severance of employer-employee links. face of reduced market earnings. This is particularly clear in the case of unemployment benefits, where the Finally, unemployment benefits were more attractive policy objective is to reduce workers’ earning losses from a public health perspective, as they do not after a separation. Beyond that shared goal, they have incentivize workers to engage in more work activity different effects. (and in fact do the opposite; see Holzer, Hubbard, and Strain 2021). Employee-side payroll tax cuts, by making Two policies that illustrate some of the tradeoffs work more attractive, will contribute at the margin to involved are unemployment insurance benefits on the spread of the disease unless the additional work the one hand and employee-side payroll tax cuts on performed crowds out risky leisure activities. 6 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Efficiency There are countervailing forces here, though. Valuable employer-employee links had to be broken for workers To the extent that consumption smoothing was the to qualify for unemployment benefits, destroying their primary objective of policy makers, unemployment firm-specific human capital and generating search costs benefits came at a low fiscal cost, as they reached down the road. Workers’ non-firm-specific human workers who lost significant market income. Employee- capital will also have deteriorated during their unem- side payroll tax cuts, on the other hand, can only do so ployment spell. While some have argued that more to the extent that workers saw their hours reduced or generous unemployment benefits increase the quality were part of a household that contained workers who of workers’ next match, the relevant counterfactual lost their employment. here was continuation of the original match. BOX B. POLICY MEASURES SUPPORTING INDIVIDUALS IN THEIR CAPACITY AS WORKERS About one-third of countries sought to help workers directly by boosting unemployment benefits, offering wage subsidies, or reducing or postponing income taxes (Figure B). Lower-income countries (LICs) tended to implement only one of these policies at a time, while lower-middle-income countries (LMICs) and upper- middle-income countries (UMICs) introduced on average two of these policies at a time. LICs prioritized public works programs and income tax reductions. LMICs were able to implement more unemployment benefit programs, in some cases accompanied by wage subsidies. UMICs, in turn, prioritized wage subsidies, accompanied by either unemployment benefits or income tax reductions. Figure B Percentage of countries with at least one worker support policy, by country income level A. LIC B. LMIC Unemployment benefits 0 Unemployment benefits 28 Wage subsidies 10 Wage subsidies 43 Income tax reduction 24 Income tax reduction 21 0 20 40 60 80 100 0 20 40 60 80 100 Percentage of countries Percentage of countries C. UMIC D. HIC Unemployment benefits 41 Unemployment benefits 57 Wage subsidies 53 Wage subsidies 71 Income tax reduction 51 Income tax reduction 35 0 20 40 60 80 100 0 20 40 60 80 100 Percentage of countries Percentage of countries Source: Global Database on Social Protection and Jobs Responses to COVID-19 (2021). Own elaboration, based on Kamran et al. (2023). 7 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Most programs related to wage subsidies and unemployment benefits were newly created in response to the economic downturn. Wage subsidies were introduced by 52 percent of countries. In the case of unem- ployment benefits, while most new policies were implemented in higher-income countries (63 percent) compared to LICs (0 percent), their scaled-up adoption is explained by the existence of unemployment insurance schemes—an estimated 94.5 percentage points more likely (Contreras et al. 2023). On average, the budget allocated to cash transfers for workers is US$714 million, larger than the combined value of all other traditional labor market policies (US$686 million), including unemployment benefits and wage subsidies. This is especially true for LMICs and UMICs and might be explained by LMICs’ and UMICs’ increased ability to target specific types of workers. Equity workers, employer-side payroll tax cuts, grant payments associated with maintaining employment during a As emphasized earlier, unemployment benefits were downturn, or schemes that facilitate short-time work attractive because they reached workers who have or temporary layoffs. recently lost significant market income. In the pan- demic context, job losses were concentrated among All these policies shared one common purpose: they low-income workers, who are particularly vulnerable directly rewarded companies that maintain employ- financially, which makes them attractive targets for ment levels or links to pre-crisis employees. This is support. This was more so the case in the context of particularly clear in the case of employer-side payroll tax supply restrictions that forced businesses to close or cuts, which provide a direct cash benefit in exchange operate at reduced capacity. for avoiding layoffs. On the other hand, while the so-called essential work- There are differences between the different policies ers continued to receive their regular pay, they and their in this category as well: some approaches envision an families were exposed to greater health risks during the unchanged number of hours worked, while others pandemic. Employee-side payroll tax cuts functioned focus on maintaining employer-employee links even as compensation for increased risk, especially when during a period of reduced hours. Here we compare targeted appropriately. That said, the incidence of employer-side payroll tax cuts to short-term work sub- employee-side payroll tax cuts is less clear than that sidies along the lines of the German Kurzarbeit system, of unemployment benefits, especially beyond the very described in more detail in Krebs (2023). short run. Effectiveness Unemployment benefits and employee-side payroll tax cuts can in principle be targeted across the income dis- This set of policies explicitly combined two objectives: tribution as one prefers, through phase-ins, phase-outs, to facilitate consumption smoothing for workers and and variation in replacement rates. A disadvantage is to preserve employer-employee ties. Relative to pro- that they will usually be tied to formal employment grams that provided direct support to workers, the within a firm, leaving self-employed workers and work- incidence of the support provided here landed more ers in the informal economy to fend for themselves. on employers. At the margin, this is precisely what will have motivated them to keep employer-employee 2.2. Support for Firms as Employers links intact. An alternative way to provide support to workers was The incidence of inframarginal payments is unclear and to help their employers and firms maintain employment they presumably benefited owners of capital and other levels. This has taken the form of wage subsidies or stakeholders in addition to employees. If governments tax credits associated with the hiring or retention of were simply trying to replace the market income of employees, the previous category of measures was 8 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 presumably preferable. And neither type of program 2.3. Firm Liquidity Support will have been particularly effective in reaching the informal sector. Apart from policies that supported businesses in their capacity as employers, there is a suite of policies that Which links to preserve and how is where the policies supported businesses to help them deal with the in this category differ. Kurzarbeitergeld and other types downturn, whether they maintain employment levels of financial support for temporary reductions in hours or not. The idea here was to ensure the survival of the per worker focus on preserving the number of workers economy’s supply side, particularly in industries where tied to each employer, while they allow for reductions public health measures and consumer hesitation make in total hours worked. Employer-side payroll tax cuts profitable operations impossible. typically do not embody a preference for the number of workers who continue to be employed over total The approaches in this category share a common theme hours worked. All else being equal, they will lead to in that they were explicitly designed to help businesses more hours worked, while short-term subsidies will lead and not just those employed by the businesses. This to a smaller number of layoffs. broader approach can be justified by considering the value of organizational knowledge and other intangi- Efficiency ble capital, the network effects of continued business operations, and a desire to preserve entrepreneurs’ Support for short-term work approaches is particularly (access to) capital. attractive when employer-employee ties are valuable and costly to reestablish through renewed matching, The two broad subcategories of support here are as they maximize the number of links preserved. Labor grants and loans. Grants can include tax relief or new markets with greater fixed costs of hiring and firing are spending that may be conditional on certain levels of likely to be particularly fertile ground for this type of revenue loss or that can be industry specific. Loans can approach. Employer-side payroll tax cuts, on the other be explicitly or implicitly subsidized, available only for hand, will be more attractive if there are high levels of certain industries or firm sizes, and be provided by fiscal variation in the value of employer-employee ties across or monetary authorities. workers within the same firm. Effectiveness To the extent that the formal sector is where valuable employer-employee ties were concentrated, these In terms of pure effectiveness, it is difficult to beat approaches will have been relatively efficient and likely generous grants to a broad universe of firms, though more so than approaches that targeted the universe of firms in the informal sector may remain difficult to firms, both formal and informal. reach. While providing cheap credit helped some firms survive, from the perspective of the firm, not having Equity to return the money made survival strictly easier. This in turn helped preserve employment at these firms Policies in this category were unlikely to help workers in as well. the informal sector and did not help workers who end up losing their jobs. This, in addition to the opaque inci- Survival of firms that received government support has dence discussed previously, made these policies relatively had implications beyond directly affected firms: for regressive. There are differences between employer-side example, it helped preserve formal-informal links and payroll tax cuts and short-term work programs as well. through that channel provided relief to parts of the Facilitating short-term work spreads the loss of income economy that are otherwise difficult to reach. Note and experience across all workers at a given firm, while that even at inframarginal firms, these types of support employer-side payroll tax cuts were of most help to served to avert capital shallowing. those workers who were harder to replace. 9 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 BOX C. POLICIES SUPPORTING FIRMS IN THEIR CAPACITY AS EMPLOYERS AND BUSINESSES Firm liquidity support programs, aimed at helping businesses survive and keep their employees, were the most widely used labor market policies. They comprised 77 percent of all labor market policies introduced. Most countries prioritized giving firms tax relief, followed by credit facilities and guarantees, loan payment facilities, deferral or reductions in social security contributions, and utility and rent support. Almost all countries adapted existing programs to improve firm liquidity, but most governments that introduced access to credit guarantees developed these funds in direct response to the pandemic. Figure C Policies supporting firms in their capacity as employers and businesses Social security 8 contributions for firms Utilities and rent 8 support Loan payment facilities 11 Other special measures 18 Credit 24 facilities/guarantees Tax relief 30 0 10 20 30 Percentage share of firm liquidity programs Source: Global Database on Social Protection and Jobs Responses to COVID-19 (2021). Own elaboration, based on Kamran et al. (2023). Tax relief measures such as the postponement of corporate tax payments, extension of value added tax (VAT) credits, additional deductions for tax credits, or the extension of the deadline for settling tax liabilities were taken by 76 percent of countries. Around 35 percent of countries allowed firms to postpone the payment of their part of social security contributions. These policies have likely been an easy way for governments to help businesses without having to make additional transactions or reallocating resources. Many governments also introduced more active policy options to inject firms with liquidity. Around 67 percent of the countries created credit guarantee funds, allocating an average of US$1.2 billion. While UMICs allocated the biggest budgets to these funds, LICs and LMICs also implemented this policy but with a smaller budget. Meanwhile, 51 percent of the countries also offered loan repayment loan repayment facilities to support businesses’ liquidity, by reducing the interest on their loans or allowing firms to restructure or postpone their debts. 10 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Efficiency 2.4. Generalized Income Support Policies A focus on efficiency dramatically shifts this picture. Just The final set of policies we discuss is special in that they as grant support is preferable from the firm’s perspec- are not targeted in particular at the situation in the tive, credit support is cheaper from the government’s labor market or at preserving any of these links that we perspective. In the worst-case credit scenario—the have emphasized so far. It is a broad category of macro entire principal was lost—the government’s total loss instruments that range from looser monetary policy still did not exceed that associated with a grant. from direct payments to households. Cash transfers, in particular, experienced a breakthrough moment In addition to this obvious difference on the expenditure during the COVID-19 crisis, as documented in detail side, a number of other efficiency considerations are in Gentilini (2022). For a country-level overview of the important. Support that was too generous will have broader macroeconomic response to the pandemic, led to the survival of what are often called zombie see IMF (2022). firms—firms that would have gone under in a normal business environment. Keeping such firms afloat delays Effectiveness the reallocation of resources toward more productive ends. Employer-employee ties and organizational capital These generalized policies are particularly useful in at such firms are effectively of negative social value. reaching workers in the informal sector and for coun- Support that is overly generous can also have crowded tries where large numbers of workers who have lost out the normal operations of the financial sector and led market income would fall through the cracks of the to problems or unnecessary adjustments in that sector. safety net. They can be effective in supporting house- holds economically while reducing public health risk; Of course, support that is not generous enough helped see, for example, Karlan et al. (2022) for evidence from inframarginal firms survive. The key tradeoff here, from an experiment involving mobile transfers to low-income an efficiency standpoint, was to target aid at firms Ghanaians during the pandemic. that genuinely need support, without reaching too many firms that would have gone under even in the Efficiency absence of the COVID-19 pandemic. This targeting was particularly important for governments working under Generalized policies are by nature not efficient in pre- stringent fiscal constraints. serving employer-employee links or in helping affected workers smooth their consumption. To the extent that Equity they reduce aggregate demand shortfalls, they may be relatively efficient for macroeconomic stabilization. To the extent that direct grants or credits to businesses This will be the case if they produce a greater stimu- benefited the owners and customers of businesses, lative effect than other policies discussed here, which this type of support was likely regressive (Autor et al. is plausible given their ability to reach low-income 2022b). This may speak in favor of credit support over households and participants in the informal sector. In grants, as a less generous regressive program will have countries that face significant budgetary constraints, had desirable redistributive properties. In more practical such considerations will inform the choice between political terms, it may have spoken in favor of tying either generalized cash transfers or programs more general business support to employment, even if most directly targeted at the supply side. support is in effect targeted at firm survival as opposed to the preservation of employment levels per se (cf. the At the same time, large parts of the global economy US Paycheck Protection Program). have faced significant inflationary pressure since at 11 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 least mid-2021. In some of those places, especially in Equity advanced economies, the more targeted instruments described earlier delivered effective, efficient, and Especially in economies with an extensive informal equitable aid to businesses and households without sector, targeting support at all households as opposed producing unnecessary overheating. to only those that include formal sector workers may be quite beneficial from an equity perspective. BOX D. GENERALIZED INCOME SUPPORT POLICIES Governments directly targeted self-employed or informal wage workers, rather than households, with cash transfers. Cash transfers were largely adaptations of existing programs. Over 38 percent of cash transfer programs were directed to workers, commonly in informal settings. Public works were also adapted to the crisis, to help informal workers who lack income protection. Cash transfers were particularly prevalent among LMICs and UMICs, possibly explained by their increased ability to target specific types of workers. Meanwhile, LICs prioritized public works in their policy response, which highlights the role of safety nets for informal workers in countries that have no other social security mechanisms. Figure D Percentage of countries with at least one cash transfer and public works program, by country income level A. LIC B. LMIC CT to economically CT to economically active persons 28 active persons 60 Public works 41 Public works 34 0 20 40 60 80 100 0 20 40 60 80 100 Percentage of countries Percentage of countries C. UMIC D. HIC CT to economically CT to economically active persons 78 active persons 65 Public works 22 Public works 8 0 20 40 60 80 100 0 20 40 60 80 100 Percentage of countries Percentage of countries Source: Global Database on Social Protection and Jobs Responses to COVID-19 (2021). Own elaboration, based on Kamran et al. (2023). On average, budgets allocated to cash transfers for workers were roughly equal to the combined value of all other labor market policies, including public works, unemployment benefits, and wage subsidies. Moreover, the budget allocated to cash transfers for workers was twice as large as the amount allocated to non-labor cash transfers. Governments have allocated on average US$686 million for increasing workers’ income through traditional labor market policies, while the average budget for cash transfers directed to workers is US$714 million, compared to US$304 million for non-labor-related cash transfers. 12 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 3. CONCLUSIONS Workers who nevertheless lost their jobs in the for- mal sector have been supported with unemployment The policy response to the COVID-19 pandemic fea- benefits. These efforts have been accompanied by tured dramatic interventions by governments all around generalized fiscal and monetary support, sometimes the world. Many of these efforts were focused on in the form of cash transfers, to stabilize the macro businesses and their workers and the links between economy and aid households without regard for their them. We have given a broad overview here of some of position in the labor market. the key questions policy makers faced as they navigated the uncertain and rapidly changing economic landscape The latter aspect of the policy response, generalized shaped by the public health crisis. aid to households without regard for their position in the labor market, was of particular importance in Perhaps the key takeaway from our analysis is that set- countries with large informal sectors. It may have been ting up support programs in a manner that facilitates counterproductive in advanced economies, where the recovery was feasible and in fact accomplished overshooting by fiscal and monetary policy makers by numerous governments. Preserving high-value has contributed to significant inflationary pressures. employer-employee links and other forms of intangi- ble capital was accomplished through aid in the form of short-term work subsidies, payroll tax cuts, and direct support to companies. REFERENCES Alfaro, Laura, Oscar Becerra, and Marcela Eslava. 2020. “EMEs and COVID-19: Shutting Down in a World of Informal and Tiny Firms.” National Bureau of Economic Research (NBER) Working Paper 27360. Autor, David, David Cho, Leland D. Crane, Mita Goldar, Byron Lutz, Joshua K. Montes, William B. Peterman, David D. Ratner, Daniel Villar Vallenas, and Ahu Yildirmaz. 2022a. “An Evaluation of the Paycheck Protection Program Using Administrative Payroll Microdata.” Journal of Public Economics, 211 (2022): 1-9. Autor, David, David Cho, Leland D. Crane, Mita Goldar, Byron Lutz, Joshua K. Montes, William B. Peterman, David D. Ratner, Daniel Villar Vallenas, and Ahu Yildirmaz. 2022b. “The $800 Billion Paycheck Protection Program: Where Did the Money Go and Why Did it Go There?” Journal of Economic Perspectives 36 (2): 55–80. Autor, David, David Dorn, and Gordon Hanson. 2019. “When Work Disappears: Manufacturing Decline and the Falling Marriage Market Value of Young Men.” American Economic Review: Insights 1 (2): 161–178. Bartik, Alexander W., Marianne Bertrand, Zoë B. Cullen, Edward L. Glaeser, Michael Luca, and Christopher T. Stanton. 2020. “How Are Small Businesses Adjusting to COVID-19? Early Evidence from a Survey.” Proceedings of the Natural Academy of Sciences 117 (30): 17656–17666. Bastian, Jacob, and Katherine Michelmore. 2018. “The Long-Term Impact of the Earned Income Tax Credit on Children’s Education and Employment Outcomes.” Journal of Labor Economics 36 (4): 1127–1163. Céspedes, Luis Felipe, Roberto Chang, and Andrés Velasco. 2020. “The Macroeconomics of a Pandemic: A Minimalist Model.” NBER Working Paper 27228. Contreras, Ivette, Melanie Khamis, David Newhouse, and Michael Weber. 2023. “How Did Countries Respond to the COVID-19 Crisis? Emerging Patterns on Jobs-Related Policies.” JobsWatch COVID-19 Brief, World Bank. Corrado, Carol, Jonathan Haskel, Cecilia Jona-Lasinio, and Massimiliano Iommi. 2022. “Intangible Capital and Modern Economies.” Journal of Economic Perspectives 36 (3): 3–28. 13 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Farré, Lídia, Francesco Fasani, and Hannes Mueller. 2018. “Feeling Useless: The Effect of Unemployment on Mental Health in the Great Recession.” IZA Journal of Labor Economics 7: 8. Fornaro, Luca, and Martin Wolf. 2020. “Covid-19 Coronavirus and Macroeconomic Policy.” Barcelona School of Economics (BSE) Working Paper 1168. Gentilini, Ugo. 2022. “Cash Transfers in Pandemic Times: Evidence, Practices, and Implications from the Largest Scale Up in History.” World Bank. Gentilini, Ugo, Mohamed Almenfi, Hrishikesh TMM Iyengar, Yuko Okamura, John Austin Downes, Pamela Dale, Michael Weber, David Newhouse, Claudia Rodriguez Alas, Mareeha Kamran, Ingrid Veronica Mujica, Maria Belen Fontenez, Muhammad Ezzat, Sandra Asieduah, Vikesh Ramesh Mahboobani Martinez, Gonzalo Javier Reyes Hartley, Gustavo Demarco, Miglena Abels, Usama Zafar, Emilio Raul Urteaga, Giorgia Valleriani, Jimmy Vulembera Muhindo, and Sheraz Aziz. 2022. “Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures.” Version 16. World Bank: February 2. Gibbons, Robert, and Lawrence Katz. 1992. “Does Unmeasured Ability Explain Inter-Industry Wage Differentials?” Review of Economic Studies 59 (3): 515–535. Guerrieri, Veronica, Guido Lorenzoni, Ludwig Straub, and Iván Werning. 2022. “Macroeconomic Implications of COVID-19: Can Negative Supply Shocks Cause Demand Shortages?” American Economic Review 112 (5): 1437–1474. Heining, Jörg, and Simon Jäger. 2022. “How Substitutable Are Workers?” Evidence from Worker Deaths.” MIT. Holzer, Harry J., R. Glenn Hubbard, and Michael R. Strain. 2021. “Did Pandemic Unemployment Benefits Reduce Employment? Evidence from Early State-Level Expirations in June 2021.” NBER Working Paper 29575. Hubbard, Glenn, and Michael R. Strain. 2020. “Has the Paycheck Protection Program Succeeded?” Brookings Papers on Economic Activity Fall 2020. IMF (International Monetary Fund). 2021. “World Economic Outlook: Managing Divergent Recoveries.” International Monetary Fund: April. IMF. 2022. “Policy Tracker: Economic Responses to COVID-19 by 197 Countries.” International Monetary Fund: August 23. ILO (International Labor Organization). 2022a. “COVID-19 and the World of Work: Country Policy Responses.” Geneva, ILO. ILO. 2022b. “Global Employment Trends for Youth 2022: Investing in Transforming Futures for Young People.” Geneva, ILO. Jarosch, Gregor. 2021. “Searching for Job Security and the Consequences of Job Loss.” NBER Working Paper 28481. Kahn, Lisa B. 2010. “The Long-Term Labor Market Consequences of Graduating from College in a Bad Economy.” Labour Economics 17 (2): 303–316. Kamran, Mareeha, Ingrid Mujica, María Belén Fonteñez, David Newhouse, Claudia Rodriguez Alas, and Michael Weber. 2023.“Exploring Two Years of Labor Market Policy Responses to COVID-19: A Global Effort to Protect Workers and Jobs.” JobsWatch COVID-19 Brief, World Bank. Karlan, Dean, Matt Lowe, Robert Darko Osei, Isaac Osei-Akoto, Benjamin N. Roth, and Christopher R. Udry. 2022. “Social Protection and Social Distancing During the Pandemic: Mobile Money Transfers in Ghana.” NBER Working Paper 30309. 14 Protecting Workers, Firms, and Worker-Firm Attachment During COVID-19: Economic Considerations for the Assessment of Policy Measures JUNE 2023 Krebs, Tom. 2023. “German Labor Market Policy during and after the COVID-19 Recession.” In Preserving Links in the Pandemic: Policies to Maintain Worker-Firm Attachment in the OECD, edited by Michael R. Strain and Stan Veuger. Pp. 115-135. AEI Press. Pierce, Justin R., and Peter K. Schott. 2020. “Trade Liberalization and Mortality: Evidence from US Counties.” American Economic Review: Insights 2 (1): 47–64. Schmutte, Ian M. 2023. “The Value of Job Matches.” In Preserving Links in the Pandemic: Policies to Maintain Worker-Firm Attachment in the OECD, edited by Michael R. Strain and Stan Veuger. Pp. 9-31. AEI Press. Schwandt, Hannes, and Till von Wachter. 2019. “Unlucky Cohorts: Estimating the Long-Term Effects of Entering the Labor Market in a Recession in Large Cross-Sectional Data Sets.” Journal of Labor Economics 37 (S1): S161–S198. Shoag, Daniel, and Stan Veuger. 2019. “The Economics of Prisoner Reentry.” In Education for Liberation: The Politics of Promise and Reform Inside and Beyond America’s Prisons, edited by Elizabeth English and Gerard Robinson. Lanham, Maryland: Rowan and Littlefield. Strain, Michael R., and Stan Veuger. 2023. Preserving Links in the Pandemic: Policies to Maintain Worker-Firm Attachment in the OECD. AEI Press. 15 The policy note was prepared by Eliana Carranza (World Bank, Social Protection and Jobs), Stan Veuger (American Enterprise Institute), Michael Weber (World Bank, Human Capital Project). All Jobs Group’s publications are available for free and can be accessed through the World Bank or the Jobs and Development Partnership website. Please send all queries or feedback to Jobs Group. Join the conversation on Twitter: @WBG_Jobs #Jobs4Dev.