Government contracts and US bond yield spreads: A study on costs and benefits of materialized political connections

• Author: Jurica Susnjara,Thanh Ngo
• Date: 01 July 2020
• Published date: 01 July 2020
• DOI: http://doi.org/10.1111/jbfa.12440

DOI: 10.1111/jbfa.12440

Government contracts and US bond yield spreads:

A study on costs and benefits of materialized

political connections

Thanh Ngo1Jurica Susnjara2

1East Carolina University, 3127 Bate Building,

Greenville, NC 27858

2Barry University, 11300 NE 2nd Ave,Miami

Shores, FL 33161

Correspondence

JuricaSusnjara, Barry University, 11300 NE 2nd

Ave,Miami Shores, Florida 33161, USA.

Email:jsusnjara@barry.edu

Abstract

In a 1991–2013 sample of bonds issued by US public firms, we

find that the cost of debt (yield spread relative to comparable

Treasuries)of suppliers to government agencies is contingent on the

strategic importance of the supplier’s industry.The yield spreads for

strategically unimportant government suppliers are higher than for

firms that are not government suppliers. If government contracts

serve as tangible evidence of political connections, these higher

yield spreads indicate that weaker corporategovernance as a cost of

political connections outweighs the benefits of said connections. For

the subsample of government suppliers from strategicallyimportant

industries, where the benefits of implicit bailout guarantees and

revenue stability outweigh the corporate governance problems, the

cost of debt is lower than for firms that are not government sup-

pliers. The higher (lower) cost of debt for strategically unimportant

(strategicallyimportant) suppliers is confined to contracting with the

federal government. Our findings are robust to alternative variable

and sample specifications, and to endogeneity concerns.

KEYWORDS

cost of debt, government contracts, political connections, yield

spread

JEL CLASSIFICATION

G12, G23

1INTRODUCTION

In 1991, US public firms reported US$ 169.3 billion worth of government contracts, of which US$ 13.4 billion (7.9%)

came from bond issuers.1In 2013, US$ 563 billion of government contracts were reported, of which US$ 82 billion

1Datafrom Compustat Segment Customer and SDC New Issues Databases. Restrictions apply to the availability of these data, which were used under license

forthis study. Data are available from the authors with the permission of the appropriate vendors.

J Bus Fin Acc. 2020;47:1059–1085. wileyonlinelibrary.com/journal/jbfa c

2020 John Wiley & Sons Ltd 1059

1060 NGO ANDSUSNJARA

(14.6%) came from bond issuers. Motivated by both the increased economic importance of contractswith government

agencies in the United States and bythose contracts’ increasing overlap with bond markets, we explore whether having

government contracts has an effect on a firm’s access to public credit. Insofar as governmentcontracts can be viewed

as proxies for political connections, both benefits and costs associated with political connections come into play. A

growing body of finance literature associated with politically connected firms (PCFs henceforth) offers possible expla-

nations of both benefits for and costs to a firm’s cost of capital, without conclusively stating which of the two effects

dominates. Governments can directly extractrents by forcing firms into inefficient investments that do not maximize

shareholder wealth (Brockman, Rui, & Zou, 2013). They can also indirectly hurt the firms they are connected to by

allowing managers to become entrenched (Boubakri, El Ghoul, & Saffar, 2013) and subject to moralhazard. Together,

these inefficiencies can lead to a higher cost of debt, as Borisova, Fotak,Holland, and Megginson (2015) find in a global

sample. Alternatively, most single-country studies on the financialeffects of political connections find greater access

to credit markets, in particular during financial crises (Johnson and Mitton (2003) in Malaysia, and Fisman (2001) and

Leuz and Oberholzer-Gee (2006) in Indonesia during the Asian Financial Crisis).

Financial costs and benefits to PCFs have not been studied as extensively in the United States, since political con-

nections as defined byprevious studies are more widespread in countries with weaker legal systems (Faccio, 2006) and

lower judicial independence (Boubakri, Cosset, & Saffar,2008). In the United States, both significant government own-

ership(Borisova et al., 2015; Houston, Lin, & Ma, 2011) and bank lending under government influence (Beck, Demirgüç-

Kunt, & Levine, 2006; La Porta, Lopez-de-Silanes, & Shleifer, 2002; Serdar Dinç, 2005) are less prevalent. Recently,

studies have found quicker and higher bailouts (Tahoun & Van Lent, 2018) and lower costs of bank loans (Houston,

Jiang, Lin, & Ma, 2014) for PCFs.

Mostinternational studies rely on the Faccio (2006) definitionof political connections: having a large shareholder or

top officer who is a top government official or closely connected to a top governmentofficial. With stricter conflict-of-

interest regulation in the United States, this results in very small samples of US PCFs. We add to this body of literature

by specifically examining whether governmentcontracts, as both tangible evidence and consequence of political con-

nections (Faccio, 2006; Goldman, Rocholl, & So, 2013; Tahoun, 2014), are related to the cost of borrowing through

bond markets. Using the Compustat Segment Customer Database, we identify whether a firm serves as a supplier to

government agencies (i.e., has governmentagencies as its customers).

This approach allows us to identify political connections via government contractsin a sample of US publicly traded

firms, while prior studies on political connections focus primarily on international firms. Some prior political connec-

tions that result in the attainment of government contracts maynot be captured by the Faccio (2006) definition. Even

political connections captured by this definition may be considered stronger once they have materialized as govern-

ment contracts. In other instances, a government contractobtained without a prior political connection may create an

ongoing business association and a de facto political connection for future use, also not captured by the Faccio (2006)

definition. For those reasons, a government contract may send a previouslyunexplored powerful signal to the credit

markets.

We examine a sample of US public bonds issued in 1991–2013, and their yield spreads relative to closest-in-

maturity Treasury securities. Government contractors from strategically important and strategically unimportant

industries are examined separately.2Our results show that the yield spread is lower for governmentsupplier firms

from strategically important industries. This suggests that the positive signal regarding political importance sent by

contracts with government agencies outweighs corporate governanceconcerns for firms more likely to be important

based on their industry. Among strategically unimportant industries, the yield spread is higher for firms that supply

government agencies; the spread increases as the proportion of sales coming from government contracts increases.

The government contracts in these strategically unimportant industries appear to emphasize the negatives of politi-

cal connections and influence, both direct (governmentalrent extraction through inefficient investment decisions) and

2We derive our definition of strategically important industries from Manzetti (1994) and similar studies: financial, mining, oil, steel, telecommunications,

transportation,utilities, and military-related production. We individually identify the SIC codes that match those industries, and list them in Appendix A.