The Employment Effects of Faster Payment: Evidence from the Federal Quickpay Reform

• Author: JEAN‐NOËL BARROT,RAMANA NANDA
• Published date: 01 December 2020
• Date: 01 December 2020
• DOI: http://doi.org/10.1111/jofi.12955

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 6 •DECEMBER 2020

The Employment Effects of Faster Payment:

Evidence from the Federal Quickpay Reform

JEAN-NOËL BARROT and RAMANA NANDA∗

ABSTRACT

We study the impact of Quickpay, a reform that permanently accelerated payments

to small business contractors of the U.S. government. We find a strong direct effect

of the reform on employment growth at the firm level. However, we document sub-

stantial crowding out of nontreated firms’ employment within local labor markets.

While the overall net employment effect is positive, it is close to zero in tight labor

markets. Our results highlight an important channel for alleviating financing con-

straints in small firms, but emphasize the general-equilibrium effects of large-scale

interventions, which can lead to lower aggregate outcomes depending on labor mar-

ket conditions.

FOLLOWING THE 2008 FINANCIAL CRISIS and subsequent slow recovery, the

contraction in credit supplied by financial intermediaries to nonfinancial firms

had a substantial impact on the real economy, particularly so on smaller firms

(Ivashina and Scharfstein (2010), Iyer et al. (2014), Paravisini et al. (2015)).

Policy makers interested in stimulating aggregate employment have thus fo-

cused on increasing small firms’ access to bank credit to alleviate financing

constraints (Bernanke (2010), Yellen (2013), Mills and McCarthy (2014)).

Beyond facilitating access to bank credit, however, government can impact

small firms’ financing more directly through its role as their customer. In the

∗Jean-Noël Barrot is at HEC Paris. Ramana Nanda is at Harvard Business School and a Vis-

iting Professor at Imperial College London. We are indebted to Editor Amit Seru and two anony-

mous referees for helpful comments. We are grateful to Manuel Adelino; Oriana Bandiera; Nittai

Bergman; Emily Breza; Hui Chen; Erik Hurst; Mauricio Larrain; Erik Loualiche; Karen Mills;

Ben Pugsley; David Robinson; Antoinette Schoar; Scott Stern; John Van Reenen; David Thes-

mar; Chris Woodruff; Liu Yang; Eric Zwick; and participants at the NBER Entrepreneurship and

Economic Growth Conference, Toulouse School of Economics, MIT Finance lunch,HEC, INSEAD,

Toulouse School of Economics, LSE, Maryland Junior Finance Conference, Georgia Tech, NBER

Corporate Finance, American Finance Associations meetings, Sciences Po, NBER Summer Insti-

tute Entrepreneurship meeting, and NYU for helpful feedback. We are also grateful to the U.S.

Department of Defense for sharing data on the timing of payments from their MOCAS accounting

system and to Paynet for providing us data on loan delinquencies. Barrot recognizes support from

the Kauffman Foundation Junior Faculty Fellowship and MIT Sloan. Nanda thanks the Division

of Research and Faculty Development at HBS for financial support. All errors are our own. We

have read The Journal of Finance disclosure policy and have no conflicts of interest to disclose.

Correspondence: Jean-Noël Barrot, Department of Finance, HEC Paris, 1, rue de la Libération,

78350 Jouy-en-Josas, France; e-mail: barrot@hec.fr.

DOI: 10.1111/jofi.12955

© 2020 the American Finance Association

3139

3140 The Journal of Finance®

United States, federal government procurement amounts to 4% of GDP with

$100 billion of goods and services purchased directly from small firms. Gov-

ernment contracts typically require payment one to two months following the

approval of an invoice. Thus government contractors are effectively lending

to the government while simultaneously having to finance their payroll and

working capital through the production process or by borrowing from banks.

Can paying small business contractors faster have a meaningful effect on their

cash flows, facilitate hiring, and ultimately stimulate aggregate employment?

Theoretically, complementarity between capital and labor imply that em-

ployment is likely to be depressed when firm-level investment is held back by

financing constraints. In addition, if there is a mismatch between the timing

of cash-flow generation and payments to labor, firms need to finance their pay-

roll through the production process (Benmelech, Bergman, and Seru (2011),

Jermann and Quadrini (2012)). A positive cash-flow shock from accelerated

payments could therefore have direct effects on employment for firms looking

to grow, independent of the indirect effects through firm-level investment. Con-

sistent with these arguments, recent studies have documented that firm-level

employment seems to respond to the intensity of financing frictions faced by

firms (Benmelech, Bergman, and Seru (2011), Chodorow-Reich (2014), Green-

stone, Mas, and Nguyen (2020)).

Nevertheless, there are reasons to also believe that faster payment might not

affect aggregate employment by much, or at all: payment acceleration should

have negligible effects in the absence of financing frictions. Furthermore, as

Acemoglu (2010) points out, employment growth measured at the firm level

may be offset in general equilibrium due to “business stealing” effects. That

is, even if affected firms grow their workforce in response to the payment ac-

celeration, this might impact the employment decisions of other firms hiring

from common local labor markets. The overall effect of payment acceleration

on aggregate employment, if any, therefore depends on both the intensity of fi-

nancing frictions and the direction and magnitude of the effect on other firms.

We study the impact of payment acceleration on firm-level employment in

the United States in the context of the federal Quickpay reform of 2011. Quick-

pay indefinitely accelerated payments to a subset of small business contractors

of the U.S. federal government, cutting the time between invoice approval and

payment by half, from 30 days to 15 days. For treated firms, the reform perma-

nently reduced the working capital needed to sustain a dollar of sales with the

government. A total of $70 billion in annual contract value was accelerated,

with small businesses impacted across virtually every industry sector and U.S.

county due to the massive footprint of federal government procurement.

We analyze the effects of the Quickpay reform in two steps. We first es-

timate the direct effect of this policy using the National Employment Time

Series (NETS) data set, which comprises establishment-level panel data that

come from the Dunn & Bradstreet (D&B) registers. Our establishment-level

results provide strong evidence of greater employment growth in treated firms

after the reform, an increase that remained statistically significant and eco-

nomically meaningful for at least three years following the reform, when our

The Employment Effects of Faster Payment 3141

establishment-level data end. In addition, and consistent with a shorter cash

conversion cycle1driving employment growth, we document that treated firms

also begin paying their own suppliers in a more timely manner, leading to

improvements in their own payment-related credit score within the D&B reg-

isters. Placebo regressions using government contractors not exposed to the

treatment show no such employment or credit score effect, providing reassur-

ance that our results are not driven by unobserved heterogeneity related to

being a government contractor. The magnitude of the employment effects we

estimate from our establishment-level regressions using NETS data mirror

those using the more comprehensive public use Census data at the county-by-

sector level; based on the elasticity of the employment response, we estimate

that the implied cost of external finance for treated firms is approximately

40%, which is comparable to the cost of trade credit and of other sources of

financing available to small businesses in the wake of the financial crisis.

We next aggregate the results up to the level of local labor markets, to study

whether these establishment-level results flow through to increases in aggre-

gate employment measured at the commuting zone level. We find that aggre-

gate employment increases, but only in areas where unemployment is high

relative to the number of vacancies. In tight labor markets, where vacancies

are high relative to unemployment, we find no increase in aggregate employ-

ment. These findings suggest the presence of a crowding out effect in tight

labor markets, where the employment growth among treated firms comes at

the expense of those who do not benefit from the improvement in cash flows

stemming from accelerated payment. In additional analysis, we provide evi-

dence of this crowding out effect, as well as evidence of employment flows from

low- to high-treatment sectors.

Taken together, our results document substantial financing frictions facing

the small businesses in our sample, but also highlight the importance of ac-

counting for equilibrium effects when studying the real effects of large-scale

reforms aimed at relaxing firm-level financing constraints.

Our study is related to several strands of the literature. First, our work con-

tributes to the literature on financing constraints among small, private firms.

These firms account for a substantial portion of employment and output, but

have received relatively little attention due to the paucity of data on their fi-

nancing. In particular, our findings point to important constraints on working

capital finance for such businesses, a question that has only recently begun

to be examined in detail. By being paid weeks after the sale of a good or a

service, firms—many of which are small businesses—effectively provide short-

term corporate financing to their—often large business—customers. Such

interfirm financing is referred to as trade credit and, in aggregate, is three

times as large as bank loans and 15 times as large as commercial paper in

1“Cash conversion cycle” refers to the number of days of working capital that need to be fi-

nanced. For example, if the firm has to pay cash on delivery of inputs,which sit in inventory for an

average of 15 days, and on average, the firm is paid by its customers 30 days after the sale, then

the cash conversion cycle is 45 days.