CBO Updates Its Interactive Tool for Analyzing the Effects of Federal Minimum-Wage Increases

Posted by on

April 5, 2021

Today, CBO updated its interactive tool-initially released in November 2019-that allows users to create custom policy options to examine how different approaches to changing the minimum wage would affect earnings, employment, family income, and poverty.

The estimates shown in the tool were generated using the same methods underlying CBO's most recent reports on minimum-wage increases: The Budgetary Effects of the Raise the Wage Act of 2021, published in February 2021, and The Effects on Employment and Family Income of Increasing the Federal Minimum Wage, published in July 2019. This blog post provides some additional information about the tool's estimates and the methods used to generate them.

How Do the Policy Options in the Tool Compare to the Raise the Wage Act of 2021?

The default policy in the tool is the most similar to the Raise the Wage Act (if enacted in October 2021):

• The minimum wage would rise to $15 per hour at full implementation;
• The date of full implementation would be January 1, 2026;
• The subminimum wage for tipped workers would rise from $2.13 per hour to $15 per hour, though it would not reach that level until 2028; and
• Both the minimum wage and the subminimum wage for tipped workers would be adjusted, or indexed, for changes in median hourly wages after full implementation.

All of the other policy options are patterned after that structure, allowing for several variations:

• The minimum wage at full implementation could be $10, $11, $12, $13, $14, or $15;
• The tipped subminimum at full implementation could be unchanged, 50 percent of the regular minimum, or equal to the regular minimum;
• The date of the policy's full implementation could be January 1 of 2025, 2026, or 2027; and
• The minimum wage could be fixed, indexed to consumer prices, or indexed to the median wage after full implementation.

For an illustration of where a minimum wage that reaches $15 at full implementation in 2026 would fall in the projected distribution of wages, see the figure below.

What Is the Basis of CBO's Estimates of How Raising the Minimum Wage Would Affect Employment?

In CBO's analyses, the size of the effects depends on the number of workers affected by the increase in the minimum wage, the changes in wages induced by the higher minimum, and the responsiveness of employment to those changes in wages. Employment would generally be more responsive (or elastic) under the following conditions:

• If the minimum-wage change affected more workers,
• If it mandated larger increases for directly affected workers (those who would earn less than the new minimum in the absence of the policy),
• If firms had more time to respond (for example, because the change was phased in over a longer period), or
• If the minimum wage was indexed to inflation or wage growth.

For more details, see Appendix A of CBO's July 2019 report.

How Responsive Is Employment to Changes in Wages in CBO's Analysis?

The responsiveness of employment to a change in the minimum wage is generally represented by an employment elasticity. As a part of its July 2019 report, CBO examined the elasticities reported in the literature for teenagers and for all workers. (Studies of minimum wages often focus on teenagers because they are disproportionately likely to earn wages at or near the minimum.) For example, in CBO's assessment, for the average change in the minimum wage studied in the research literature, the employment elasticity for teenage workers is −0.07, which implies that teenage employment declines by 0.7 percent, on average, for every 10 percent increase in the minimum wage. That value is CBO's median estimate of the elasticity that would apply to teenage employment in the short term (for one year) following a minimum-wage change that was close in magnitude to the average of past changes. CBO relied on other research to estimate the elasticity that would apply to adult employment.

To analyze the proposed policies, CBO then adjusted those elasticities to account for the following factors:

• The scope and size of the minimum-wage change under consideration-that is, the number of workers affected and the changes in their wages-relative to the average historical change to the minimum wage,
• Whether and how the new minimum wage would be indexed in the years after it reached its target amount, and
• How long employers would have to adjust to the new minimum wage.

In Table A-1 of its July 2019 report, which is reproduced below, CBO reported those estimates from the literature but adjusted them for policy options that would increase the minimum wage to $15, $12, or $10. CBO reported its estimates of both overall employment elasticities (which measure how employment for all workers would respond to changes in the minimum wage) and own-wage elasticities (which measure how the employment of workers directly affected by a given policy would respond to changes in their wages). Those elasticities measure the net effect on future employment for people who are working today and people who are not working today but could be working in the future.

Computer code showing exactly how those employment elasticities were calculated was published as supplemental information for CBO's interactive tool in November 2019.

What Research Did CBO Consult When Forming Its Estimates?

In updating its analysis in 2021, CBO reviewed recent research on how minimum wages affect employment; also, to account for declines in employment caused by the 2020-2021 coronavirus pandemic, the agency reviewed studies assessing whether those effects would be different during a period of high unemployment. Although the pandemic and associated increases in unemployment affected CBO's baseline projections of the budget and economy for the 2021-2031 period, they did not lead the agency to change its 2019 estimates of the employment elasticity for directly affected workers. (For details, see the section on 'CBO's Economic Modeling Approach' in the February 2021 report.)

The research that CBO used when forming its estimates was listed in Appendix B of CBO's July 2019 report. CBO examined review articles that synthesized information about the effects of the minimum wage on employment:

• 6 analyses of published estimates of employment effects (which accounted for journals' tendency to publish studies showing significant effects),
• 3 reviews of methods and data used to estimate employment effects,
• 2 reviews of employment effects in other countries, and
• 4 early reviews of the research literature.

CBO also considered various original studies, again listed in Appendix B of the July 2019 report, most of which were too recent to have been covered by the reviews:

• 6 studies focused on employment among teenagers,
• 2 studies of the influence of different economic conditions,
• 7 studies focused on industries in which low wages are prevalent,
• 4 studies focused on groups that tend to earn low wages,
• 12 studies focused on most affected workers,
• 10 studies examining long-term effects, and
• 7 other studies of employment effects.

As key inputs into its assessment of the distribution of potential elasticities, CBO used 11 studies-listed in Table A-2 of the July 2019 report, which is reproduced below-to calculate short-run own-wage employment elasticities for all or most directly affected workers (whether teenagers or adults). Next, by applying resampling methods to those studies and others listed in Appendix B of the same report, CBO estimated a distribution of short-run own-wage employment elasticities for all workers directly affected by a historically representative change in the minimum wage. The median estimate from that distribution was −0.25. CBO reported the median estimate to account for potential asymmetry in the distribution of responses: The agency concluded that there was a one-third chance that the own-wage elasticity would be between about zero and −0.25, a one-third chance that it would be between −0.25 and −0.75, and otherwise, roughly equal chances that it would be either positive or more negative than −0.75. CBO estimated the mean of the own-wage elasticity distribution to be −0.29.

Evidence on longer-run effects-applicable for three years after a minimum-wage increase-was available in only a subset of studies that CBO reviewed. To obtain a median estimate of the long-run elasticity that reflected studies both with and without long-run estimates, CBO first calculated the ratios between long-run and short-run elasticities for the few studies that reported both. CBO found considerable variation in those ratios: Some were as low as 1 (implying no difference between short- and long-run elasticities) and others as high as 2, with the median being 1.5. As with the short-run elasticities, CBO used resampling methods to gauge the methods' sensitivity to the inclusion of any given study; that analysis supported the use of a value of 1.5.

How Does CBO's Current Assessment of the Literature Compare With Other Recent Reviews?

As part of its July 2019 report, CBO conducted an exhaustive review of the literature on the effects of minimum wages on the employment of teenagers and adults. As part of its February 2021 report, CBO reexamined the literature, paying particular attention to studies published since July 2019, though this new review did not lead CBO to change its views on the range of elasticities from those used in its 2019 report.

In a January 2021 working paper published by the National Bureau of Economic Research and updated in March 2021, David Neumark and Peter Shirley summarized extant studies providing estimates of short-term employment elasticities. Using eight estimates of how the employment of directly affected workers responds to changes in the minimum wage, they reported a median estimate of −0.130 and a mean estimate of −0.270. To make those estimates comparable to the own-wage elasticities reported in Table A-2 of its July 2019 report, CBO multiplied them by 1.66-the historical ratio of minimum-wage increases to subsequent changes in wages. After that conversion, Neumark and Shirley's median estimate of the short-term own-wage elasticity is about the same as CBO's, and their mean estimate is larger than CBO's.

In a November 2019 study for the U.K. government, Arindrajit Dube summarized own-wage elasticities from 10 estimates for all directly affected workers in the United States and from 26 estimates for narrow subgroups (such as teenagers, restaurant or retail workers, and immigrants with low levels of education). In his assessment, the estimates for short-term and long-term elasticities were about the same. He reported a median estimate of −0.170, about two-thirds the size of CBO's estimate for the short term.

How Did CBO Incorporate Findings Related to Seattle's Minimum-Wage Increase?

Two studies examining a recent increase in the minimum wage in Seattle were released as National Bureau of Economic Research working papers: one in May 2018 and one in October 2018. The May 2018 paper addressed changes in the employment of low-wage workers and potential new entrants to the labor market. An estimated elasticity of −1.7 stemming from that paper was included in Table A-2 of CBO's July 2019 report; like all the papers listed in that table, it continued to influence CBO's analysis in 2021.

The October 2018 paper examined the employment, hours, and wage trajectories of people who were working at the time of the minimum-wage increase but not of new entrants. After reviewing the paper and consulting with one of the authors, CBO concluded that its findings could not be used to construct an elasticity comparable to those reported in Table A-2-that is, one reflecting changes in employment for new entrants to the labor market as well as people who were already working-because reductions in the hiring of new employees would be an important channel through which joblessness could increase. However, the paper was among those cited in CBO's July 2019 report as contributing to the broader base of evidence that CBO considered. Drawing on that broader base, CBO estimated that the chances of elasticity values at the ends of the range shown in Table A-2 of the July 2019 report (0.4 and −1.7) were very low. (CBO examined new studies for its February 2021 report, but that new review did not lead CBO to change its estimates of the likely range of elasticity values.)

Why Does the Tool Present Mean Rather Than Median Estimates?

The tool focuses on the mean estimated change in employment because it is the most important contributor to the projections of budgetary effects.

The responsiveness of employment to changes in the minimum wage is uncertain, and that uncertainty is asymmetric. According to the agency's assessment of the research literature, responsiveness is unlikely to be much lower than CBO's median estimate (which is equally likely to be too high or too low), but it could be much higher. That asymmetry implies that CBO's median estimate of the economic and budgetary effects is generally closer to zero than is its mean estimate.

The budgetary effects in CBO's analysis depend on whether the hourly wages of affected workers would be below or near the proposed minimum wage in the absence of the policy. Where affected workers' wages would be positioned in relation to the proposed minimum in turn depends on the intervening growth in wages, which is uncertain. In CBO's assessment, that uncertainty about wage growth is symmetric, meaning that wage growth is equally likely to be higher or lower, by a fixed amount, than CBO's mean estimate. Yet although the uncertainty is symmetric, the budgetary and economic effects of that uncertainty are asymmetric-that is, they would not be the same size if wage growth was faster than in the baseline projections or equally slower than in the baseline projections. The asymmetry arises because, if baseline wages rise fast enough, the minimum wage ceases to affect wages. (Similar issues are discussed further in CBO's October 2020 report, Estimating the Cost of One-Sided Bets: How CBO Analyzes the Effects of Spending Triggers.)

How Have CBO's Assessments of Employment Elasticities Changed Over Time?

CBO's first report on the effects of minimum-wage increases, published in 2014, was The Effects of a Minimum-Wage Increase on Employment and Family Income. In CBO's July 2019 report, the employment elasticities for teenage workers were similar, but the agency's estimates of the elasticities for adults had increased substantially. For example, the median own-wage elasticity for all directly affected workers in the short term, given a change in the minimum wage equal to the average historical change, was −0.25 in the 2019 report. It was −0.16 in the 2014 report for the option most similar to the average historical change (a $9 option). Also, in the 2019 report, that median elasticity was lower than the mean; in the 2014 report, the two were equal.

The increases stemmed from CBO's assessment of the research literature-of the studies listed in Table A-2 (including the five studies with elasticities more negative than −0.25, which ranged from −0.9 to −1.7) and of those listed in Appendix B of the 2019 report. Additional details are provided in Appendix A of that report.

In both the 2014 and 2019 reports, the elasticity for a given option depended on several factors: the number of workers affected, the increase in wages required to bring employers into compliance with the policy, whether or not the new minimum wage was indexed for wage growth, and the time that would elapse between when the policy was initially implemented and when the employment response was being estimated. That last factor was the primary reason that CBO used a larger own-wage elasticity (−0.38) for the $15 option in 2019 than for the options in 2014.

For both reports, CBO analyzed the effects of the options after the target minimum wage would first be reached, but the time elapsed since initial implementation was much longer for the options in the 2019 report. The options in the 2014 report, because of their more rapid implementation, were evaluated two years after the first scheduled increase. By contrast, the policy options in the 2019 report were evaluated five years after the first scheduled increase. The longer period between the start of implementation and the evaluation date led CBO to use the larger long-run elasticity in its 2019 analysis.

How Has CBO's Assessment of Effects on Poverty Varied Over Time?

In all of its analyses since 2014, CBO has found that increasing the minimum wage would reduce the number of people in poverty. For example, CBO estimated in its February 2021 report that the Raise the Wage Act of 2021 would reduce the number of people with income below the poverty threshold in 2025 by 0.9 million. That estimate reflected the net effect of some people's income moving above the poverty threshold and other people's income moving below that threshold. Some people near the poverty threshold would not be employed for at least some of the year because of the policy, and their income would fall below the poverty threshold. But a larger number of people would have jobs at higher wages, and their income would rise above the poverty threshold.

Unlike CBO's earlier reports, the interactive tool provides a range of effects on poverty, not just a central estimate. That range of effects was constructed using a simulation analysis, which entails independent sampling from CBO's estimated distributions of employment elasticities and wage growth-the same method CBO has used to estimate a range of effects on employment in earlier reports.

Nabeel Alsalam, William Carrington, Justin Falk, and Brooks Pierce are analysts at the Congressional Budget Office.