By William Dupor, Assistant Vice President and Economist
In February 2009, the federal government allocated more than $64 billion to local school districts through the Department of Education as part of the $840 billion American Recovery and Reinvestment Act. The Office of the President wrote that the funding “helped fill gaps and avert layoffs of essential personnel” at schools across the nation.1 As part of the Recovery Act, the federal government tracked the number of jobs payrolled by these funds via surveys of recipient organizations. According to those reports, Department of Education dollars payrolled more than 750,000 jobs during the first two school years following its passage.2
While informative, the payroll approach does not answer the question: What would have happened to schools in absence of the Recovery Act? For example, school districts may have found other ways to pay employees in absence of the act's funding.
Professor Saif Mehkari of the University of Richmond and I answered the question by examining the spending, staffing and Recovery Act grants of over 6,700 U.S. school districts during the Recovery Act period.3 Professor Mehkari and I use the heterogeneity in how Recovery Act funds were distributed across districts. By comparing how school districts that received relatively little grant money adjusted their staff levels relative to districts that received larger amounts, we inferred what districts would have done had those funds not been available.
We found that for every $1 million in grants to a district, the district employed roughly 1.5 additional staff members and increased expenditures by $570,000. Moreover, the jobs effect estimate was not statistically different from zero. We also found that the additional staff members were nonteachers and that nearly 70 percent of the increased expenditures came in the form of capital outlays on things such as construction and equipment and existing structures purchases.
According to our estimates, the total impact on education employment was 42,600 workers, much smaller than the 750,000 payroll number discussed above. The difference between our approach and the payroll approach is that the latter did not account for the possibility that districts would have found other ways to pay employees to stay on the job had Recovery Act funds not been available.
Why might the education employment effect, according to our estimates, have been small? One potential explanation may be the districts’ own preference for maintaining student-teacher and student-staff ratios. If this was a top priority, districts with relatively small ARRA grants and large budget gaps may have found ways to meet budget shortfalls other than laying off teachers, such as implementing furloughs or reducing high-cost course offerings. Moreover, districts that received relatively generous grants may have not hired additional workers since, once the short-lived grants were spent, the new staff would likely be let go.
Note that our data do not include the employment numbers for the companies that sold equipment or did the construction work for the school districts. Since some of the grant spending took the form of capital outlays, employment in these companies and industries is the next logical place to look for more job creation from the Recovery Act’s education grants.
1 “Educational Impact of the American Recovery and Reinvestment Act,” Domestic Policy Council, Executive Office of the President and the Department of Education, Oct. 19, 2009.
2 The numbers that make up this total appear in a sequence of quarterly reports from the Council of Economic Advisers titled “The Economic Impact of the American Recovery and Reinvestment Act of 2009” published between 2009 and 2011. These reports are available here: https://www.whitehouse.gov/administration/eop/cea/factsheets-reports.
3 Dupor, William; and Mehkari, M. Saif. “Schools and Stimulus.” Federal Reserve Bank of St. Louis Working Paper, 2015.