OPM's Paid Parental Leave

Analyzing the agency's benefit-cost analysis.
September 10, 2020

Download this Public Interest Comment (PDF)


Introduction

The Federal Employee Paid Leave Act, enacted in 2019, allows several categories of federal employees to substitute up to 12 weeks of paid parental leave for the unpaid leave they are entitled to under the Family and Medical Leave Act. The substitution is permitted in connection with the birth, adoption, or foster care placement of a child with the employee. OPM’s interim final rule implements the legislation. OPM estimates that the value of annual salary and benefits associated with this leave would likely range between $890 million and $1.3 billion, with a mean estimate of $1.1 billion.[1]

We write to express our concern about two elements of OPM’s economic analysis of the rule. First, OPM’s discussion of benefits should be clarified and made more complete.[2] Second, OPM classifies the value of salary and benefits associated with paid parental leave merely as a transfer from taxpayers. However, it is clear that the statute and the rule are intended to produce a real reallocation of resources toward child care and away from other activities. Therefore, an accurate assessment of the costs of the regulation should include these real resource costs, including the deadweight cost of taxation.  

Benefits of Parental Leave

The interim final rule notes that a rule requiring the provision paid parental leave would shift activities toward activities such as child care, and away from other activities.[3] At the same time, treating the monetary value of paid leave to federal employees as a transfer from taxpayers to the employees seems to implicitly assume that the main effect of the rule is to shift the financial burden from employees to taxpayers of taking unpaid leave that would already be taken.  If this is the case, the rule would not induce any change in behavior, and hence there would be no social benefits from implementing the rule.  If, however, the rule does induce federal employees to take more parental leave, it would induce a shift toward activities such as care of children, and away from other activities. In this situation, the net benefits from implementing the rule would equal the difference between the benefits attributed to the additional leave taken and the social opportunity cost of activities not undertaken.  

Although OPM acknowledges that benefits could result from a shift in activities caused by additional leave-taking, OPM does not attempt to quantify either the economic value of the benefits of paid parental leave or the social opportunity cost of activities that are foregone. 

As OPM notes, there is a considerable empirical literature on both the benefits and costs of paid leave.  The types of benefits to workers that may result from providing paid leave are well-summarized in a recent report from the American Enterprise Institute:

Paid parental leave is beneficial for at least three primary reasons. First, the composition of the workforce and the demands on working families have changed dramatically over recent decades, meaning far more parents are struggling to balance the competing demands of work and family. Second, a growing body of evidence shows children fare better when their parents have access to leave,  and individuals are more likely to take leave when it is paid.  Mothers’ leave-taking after childbirth can improve maternal health, and fathers’ access to paid parental leave can improve gender equity in the household, fathers’ involvement in child care, and outcomes for children. Finally, national economic growth depends on strong labor force participation by both men and women. Paid parental leave enables parents to remain attached to the labor force while they care for and bond with their new children, and it protects against the financial hardship of going without an income during leave.

By improving women’s ability to return to their previous employer after taking leave for the birth of a new child, state-level paid parental leave laws in California and New Jersey were associated with increased labor force attachment among women in the months surrounding childbirth. Lawrence Berger and Jane Waldfogel show that women with access to paid leave are more likely to take longer periods of leave following a birth than women without leave coverage are, but they are also 40 percent more likely to return to work after giving birth than those without access are.[4]

Although numerous studies present evidence on the labor market impacts of paid family leave, these studies stop short of estimating of the dollar value of the benefits to individuals of paid leave. Nonetheless, the empirical estimates of labor market outcomes could be used to estimate some of the benefits. 

For example, several studies have found that paid parental leave reduces cases of respiratory illness in young children.[5] These estimates could be combined with administrative data on the number of additional employees taking leave as a result of the interim rule to estimate the potential number of cases of respiratory childhood illnesses avoided. This number could then be multiplied by estimates of the cost per case of treating childhood respiratory illness to estimate the dollar benefits of implementing the interim rule. 

Similarly, studies cited in a report by the Council of Economic Advisers find that the cost of replacing a worker who has left averages one-fifth of that employee’s salary.[6] Such an estimate could be the basis for estimating that one component of the economic benefits of implementing the interim final rule could equal ∆N x S x .20, where  ∆N = the number of employees who otherwise would have left federal employment without the availability of paid family leave who are induced to remain because of paid family leave; and S equals the average salary of such employees.

Estimates such as these are beyond the scope of OPM’s analysis of the interim final rule, and certainly beyond the scope of this comment.  It would, however, be instructive for OPM to take advantage of its rich trove of administrative data to conduct such an analysis in the future.

Costs of the Rule

OPM’s analysis characterizes the monetary value of paid leave as a transfer from taxpayers to the employees who utilize the leave. Consequently, OPM estimates that the only costs associated with the rule are annualized administrative costs of $5.9 million.[7] This treatment contradicts basic economics and OMB guidance.

OMB Circular A-4 notes that “Benefit and cost estimates should reflect real resource use.”[8] It also instructs that benefit and cost estimates should include “private-sector compliance costs and savings.”[9]

The paid leave regulation is clearly intended to move resources from one use to another. OPM notes, “we estimate that this rule results in shifts in activity toward the care of young children by Federal employees, and away from other activities.”[10] OPM assumes that the opportunity cost in terms of forgone federal services will be zero, either because existing federal staff will perform the work that would have been performed by employees on leave, or because additional employees will be hired. Thus, total federal payments to employees will increase.[11]

If this were a regulation mandating paid leave by the private sector, the additional costs borne by the private sector would be considered part of the cost of the regulation. To the extent that businesses passed the cost on to customers through price increases, forgone producer and consumer surplus caused by output reductions induced by the price increases would also count as costs of the regulation. Similarly for the current regulation, the compliance costs passed on to taxpayers, plus the associated deadweight cost of taxation, should be counted as costs of paid parental leave.

OMB Circular A-94 prescribes a marginal excess burden of taxation equal to 25 percent.[12] However, most empirical research suggests that a figure in the neighborhood of 40 percent is more accurate.[13] Either of these figures is more accurate than zero. 

Conclusion

OPM’s analysis suggests that paid parental leave could create social benefits, but the only cost would be the associated administrative costs. Both claims cannot simultaneously be true.

If paid parental leave generates no change in behavior – because parents are already utilizing sick leave, annual leave, or unpaid parental leave[14] – then it makes sense for OPM treat the entire value of paid parental leave provided to employees as a transfer.  There will still be a cost (the marginal excess burden of taxation) associated with raising the revenues to fund the transfer, which should be included as a cost of the regulation, along with the administrative costs.  But with no change in behavior, there will be no benefits.

Alternatively, OPM might reasonably assume that the availability of paid leave will cause an increase in parental care of young children, and thereby generate benefits.  To the extent it does so, however, OPM must then also account for the additional costs associated with those same employees being absent from work.


[1]    Id. at 48085.

[2]    Id. at 48086.

[3]    OPM, supra note 4, at 48086.

[4]    Aparna Mathur, “Paid Family Leave: Understanding the Benefits and the Costs,”American Enterprise Institute (April 2019).

[5]    Ariel M. Pihl and Gaetano Basso, “Did California Paid Family Leave Impact Infant Health?” Journal of Policy Analysis and Management, 38(1) (2019), 155-180; Lindsey R. Bullinger, “The Effect of Paid Family Leave on Infant and Parental Health in the United States,” Journal of Health Economics, 56 (2019), 101-116; Maya Rossin-Slater, Christopher Ruhm, and Jane Waldfogel, “The Effects of California’s Paid Family Leave Program on Mothers’ Leave-Taking and Subsequent Labor Market Outcomes,” IZA Discussion Paper No. 6240 (2011).

[6] Council of Economic Advisers, “The Economics of Unpaid Leave,” Office of the President of the United States (June 2014).

[7] Id. at 48088.

[8] Office of Management and Budget, “Circular A;4: Regulatory Analysis” (Sept. 17, 2003), 37

[9] Id.

[10] OPM, supra note 4, at 48086.

[11] Id. at 48084.

[12] OMB Circular No. A-94 Revised, “Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs” (October 29, 1992), 13. Available at: https://obamawhitehouse.archives.gov/omb/circulars_a094#11

[13] Brian Mannix, “Public Interest Comment on The Office of Management and Budget’s Document, ‘Marginal Excess Tax Burden as a Potential Cost Under Executive Order 13771’” (February 20, 2020), 4. Available at: https://regulatorystudies.columbian.gwu.edu/sites/g/files/zaxdzs4751/files/downloads/PICs/GW%20Reg%20Studies%20-%20Marginal%20Excess%20Tax%20Burden%20and%20EO%2013771%20-%20BMannix.pdf. See also estimates cited in Jerry Hausman, “The Efficiency Effects on the U.S. Economy of Wireless Taxation,” National Tax Journal 53:3 (September 2000), Table 1.

[14] OPM, supra note 4, at 48084-85.