Conference on Entrepreneurship and Economic Growth
Manuel Adelino and David T. Robinson, Organizers
October 14-15, 2016
Can Paying Firms Quicker Affect Aggregate Employment?
Governments around the world are major consumers of goods and services produced by private sector firms. In the U.S., federal government procurement amounts to 4 percent of GDP and includes $100 billion in goods and services purchased directly from small businesses. Typical contracts between firms and the government as with most buyers and sellers only require payment one to two months following the approval of an invoice, while payroll is usually made weekly or bimonthly. Small businesses contractors therefore effectively lend to the government while simultaneously having to borrow from banks to finance their payroll and working capital.
The researchers study the impact of the federal Quickpay reform on firms' employment. The reform, announced in September 2011 by President Obama, accelerated payments to a subset of small business contractors of the U.S. federal government, reducing the time taken between invoice approval and payment by 15 days. Annual contract value of $64 billion was accelerated and impacted a set of small businesses across virtually every U.S. county due to the massive footprint of federal government procurement. Despite being extensive in coverage, the reform was focused on firms that were small business contractors to the federal government who were not already being paid in 15 days. Moreover, since payment terms were changed at the level of the federal government, they were unrelated to the specific labor market or financing conditions in a given firm's county or industry. This feature of the setting not only provides useful cross-sectional variation in the intensity of treatment across county-industries, enabling study of the causal impact of the reform, but also allows for contrasting the effect of treatment on different local labor markets. For example, the researchers can examine how it varied across counties where unemployment rates were particularly high at the time the reform was implemented, or where bank credit was particularly scarce. They therefore can go beyond most studies looking at the labor market effects of financing frictions, as most such studies exploit regional variation in financing frictions to identify causal effects and hence are unable to examine heterogeneity in response across counties or industries.
They find a substantial impact of accelerated payment on small firms, despite the acceleration being only 15 days. By permanently reducing the assets needed to sustain a dollar of sales to the government, the policy freed up substantial cash flow that could be directed towards firm growth. Each additional accelerated dollar of sales led to almost 10 cents estimated additional increase in payroll over three years, with two-thirds of the increase coming from new hires and the other third from increased earnings per worker, the researchers estimate. To put this magnitude in perspective, a firm that sold 100 percent of its output to the government would have increased payroll 30 percent and employment 20 percent over a three-year period, relative to a firm that did not sell to the government at all. Aggregating these relative effects across firms that were treated, their estimates suggest that the direct effect of the policy was to increase annual payroll by an additional $6 billion, and to create just over 75,000 additional jobs over the three years following the reform. They show that there were no prior trends, and that the effects were not driven by county-sector exposure to government contracts or to small businesses, but by exposure to accelerated contracts only. Consistent with the acceleration of the cash conversion cycle driving payroll growth, they find the effect to be strongest in sectors where receivables account for a larger share of assets and in counties where financing frictions are more severe for small firms.