Mandatory Paid Sick Leave and Epidemics
This week, congress and the president clashed over the details of a coronavirus relief package (I believe a deal has now been reached). One source of the conflict was congressional Democrats' efforts to include a mandate that employers give paid leave to employees. The mandate, it is worth noting, would be permanent. I don't know whether the bill that stands to pass mandates paid sick leave for all employees; news sources are rather ambiguous at the moment, as the term "sick leave" is used to refer to the provision of paid sick leave to health care workers specifically, or to federal employees, or payment by the federal government to sick workers. Originally at least, though, there was a proposed provision to mandate private employers generally to provide paid sick leave.
This, I argue, is a bad idea, particularly during an epidemic. The reasoning behind why this policy is specifically necessary or useful during an infectious disease epidemic is, as I understand it, twofold. 1) we want to discourage people from trying to work despite being sick, as this exacerbates the spread of the virus; and 2) An unusual number of employees are going to be sick, and not only for short time but even for weeks on end, and thus employees economic straits will be particularly dire during this period without paid sick leave.
The first reason is a valid concern, but mandatory sick leave is probably the wrong way to accomplish it. It is, in essence, an externality problem: the main cost of people showing up to work sick is borne by the other people they cause to get sick. For highly contagious diseases like covid-19, the costs are probably mostly borne by the general population, as the illness spreads not just to other employees at the sick person's employer, but their families, their friends, fellow commuters, etc. It is therefore arguably the case that sick leave should be either paid by the state, or that employers should be subsidized to provide it to employees. Alternatively (or in addition) people who show up to work sick could be taxed for doing so, or companies could be taxed according to the extent that illnesses are spread by their workforce. Of course, identifying which employee or even company is responsible for the spread of an infections disease to others is almost certainly implausible.
The issue with the second reason is that the unusual economic hardship caused by an epidemic does not fall only on employees, but also on businesses. Just as employees lose income from having to take time off, employers produce less with fewer employers. Mandatory sick leave would effectively place the economic burden of the inevitable decline in productivity squarely on businesses. This is not something we should want to do, as the heightened costs caused by so many sick workers still collecting full paychecks are likely to increase the number of employers that go out of business or downsize during the course of the epidemic. It also would likely disincentivize hiring new employers.
Incidentally, recent research (https://www.nber.org/papers/w26832) suggests paid sick leave doesn't seem to crowd out other benefits (no mention of whether it crowds out wages). If sick pay mandates don't crowd out wages or benefits, however, this suggests either that the cost of the benefit is small enough to not matter much to businesses (the authors' mention that sick leave only slightly increases the number of sick days taken, so it's possible paid sick leave is a minor cost*) or, for whatever reason, sick leave costs are borne almost entirely by the employers (or by customers). During normal times, in fact, both may be true. However, if sick leave costs are indeed borne almost entirely by employers, this means during health crises, mandatory sick leave will render businesses particularly vulnerable to skyrocketing costs.
In the long run, we may pay for such policies by seeing more businesses shutter, downsize, or cut back on hiring, and thereby weaken the economic recovery from the crisis. To the extent that businesses fail during the crisis, workers will lose their jobs entirely. Because it is not optimal to place the entire burden of taking care of workers temporarily removed from the workforce on their employers, the state paying sick workers without the means to provide for themselves directly would be preferable. Ultimately, though, if it is taken for granted (I don't think it is, but I may be in the minority) that everyone should want to work in a job with paid sick leave, the best policy would be to require, by law, that all employed individuals (unless they work at companies that voluntarily have paid sick leave) purchase 'sick leave insurance' or temporary unemployment insurance, and pay the market rate premium (which would be based in part on how often one takes off sick). This would more appropriately distribute the cost of illness among the relevant parties, as well as spread the cost over time, so health emergencies don't cause personal financial crises. Alternatively, we could mandate (with a fine being the likely penalty) that everyone save a certain percent of their income until they have at least a certain amount saved up in case of a crisis. Neither of these policies of course are particularly libertarian in their treatment of individual choices, but I think they are both fairer and more effective at accomplishing the desired result than mandatory paid sick leave.
*The linked study, as far as I can tell, looked at fairly 'short run' effects of sick leave mandates, and only in the US. In western European countries where mandatory sick leave and disability benefits are more widespread and generous, there's good reason to suspect that these policies drive up temporary unemployment to very high levels compared to the US. In Norway, for example, a quarter of workers are absent on any given day (https://www.nytimes.com/2004/07/25/world/norway-looks-for-ways-to-keep-its-workers-on-the-job.html). My suspicion is that, as with other benefits like social security disability insurance or negative income tax, whose negative effects on workforce participation to materialize gradually rather than immediately and become more pronounced in the long run, the costs of mandatory paid sick leave would increase gradually as people habituate to the availability of the benefit.
This, I argue, is a bad idea, particularly during an epidemic. The reasoning behind why this policy is specifically necessary or useful during an infectious disease epidemic is, as I understand it, twofold. 1) we want to discourage people from trying to work despite being sick, as this exacerbates the spread of the virus; and 2) An unusual number of employees are going to be sick, and not only for short time but even for weeks on end, and thus employees economic straits will be particularly dire during this period without paid sick leave.
The first reason is a valid concern, but mandatory sick leave is probably the wrong way to accomplish it. It is, in essence, an externality problem: the main cost of people showing up to work sick is borne by the other people they cause to get sick. For highly contagious diseases like covid-19, the costs are probably mostly borne by the general population, as the illness spreads not just to other employees at the sick person's employer, but their families, their friends, fellow commuters, etc. It is therefore arguably the case that sick leave should be either paid by the state, or that employers should be subsidized to provide it to employees. Alternatively (or in addition) people who show up to work sick could be taxed for doing so, or companies could be taxed according to the extent that illnesses are spread by their workforce. Of course, identifying which employee or even company is responsible for the spread of an infections disease to others is almost certainly implausible.
The issue with the second reason is that the unusual economic hardship caused by an epidemic does not fall only on employees, but also on businesses. Just as employees lose income from having to take time off, employers produce less with fewer employers. Mandatory sick leave would effectively place the economic burden of the inevitable decline in productivity squarely on businesses. This is not something we should want to do, as the heightened costs caused by so many sick workers still collecting full paychecks are likely to increase the number of employers that go out of business or downsize during the course of the epidemic. It also would likely disincentivize hiring new employers.
Incidentally, recent research (https://www.nber.org/papers/w26832) suggests paid sick leave doesn't seem to crowd out other benefits (no mention of whether it crowds out wages). If sick pay mandates don't crowd out wages or benefits, however, this suggests either that the cost of the benefit is small enough to not matter much to businesses (the authors' mention that sick leave only slightly increases the number of sick days taken, so it's possible paid sick leave is a minor cost*) or, for whatever reason, sick leave costs are borne almost entirely by the employers (or by customers). During normal times, in fact, both may be true. However, if sick leave costs are indeed borne almost entirely by employers, this means during health crises, mandatory sick leave will render businesses particularly vulnerable to skyrocketing costs.
In the long run, we may pay for such policies by seeing more businesses shutter, downsize, or cut back on hiring, and thereby weaken the economic recovery from the crisis. To the extent that businesses fail during the crisis, workers will lose their jobs entirely. Because it is not optimal to place the entire burden of taking care of workers temporarily removed from the workforce on their employers, the state paying sick workers without the means to provide for themselves directly would be preferable. Ultimately, though, if it is taken for granted (I don't think it is, but I may be in the minority) that everyone should want to work in a job with paid sick leave, the best policy would be to require, by law, that all employed individuals (unless they work at companies that voluntarily have paid sick leave) purchase 'sick leave insurance' or temporary unemployment insurance, and pay the market rate premium (which would be based in part on how often one takes off sick). This would more appropriately distribute the cost of illness among the relevant parties, as well as spread the cost over time, so health emergencies don't cause personal financial crises. Alternatively, we could mandate (with a fine being the likely penalty) that everyone save a certain percent of their income until they have at least a certain amount saved up in case of a crisis. Neither of these policies of course are particularly libertarian in their treatment of individual choices, but I think they are both fairer and more effective at accomplishing the desired result than mandatory paid sick leave.
*The linked study, as far as I can tell, looked at fairly 'short run' effects of sick leave mandates, and only in the US. In western European countries where mandatory sick leave and disability benefits are more widespread and generous, there's good reason to suspect that these policies drive up temporary unemployment to very high levels compared to the US. In Norway, for example, a quarter of workers are absent on any given day (https://www.nytimes.com/2004/07/25/world/norway-looks-for-ways-to-keep-its-workers-on-the-job.html). My suspicion is that, as with other benefits like social security disability insurance or negative income tax, whose negative effects on workforce participation to materialize gradually rather than immediately and become more pronounced in the long run, the costs of mandatory paid sick leave would increase gradually as people habituate to the availability of the benefit.
Comments
Post a Comment