Do Private Firms Invest Differently than Public Firms? Taking Cues from the Natural Gas Industry

• Author: JEROME P. TAILLARD,ERIK P. GILJE
• Date: 01 August 2016
• Published date: 01 August 2016
• DOI: http://doi.org/10.1111/jofi.12417

THE JOURNAL OF FINANCE •VOL. LXXI, NO. 4 •AUGUST 2016

Do Private Firms Invest Differently than Public

Firms? Taking Cues from the Natural

Gas Industry

ERIK P. GILJE and JEROME P. TAILLARD∗

ABSTRACT

We study how listing status affects investment behavior.Theory offers competing hy-

potheses on how listing-related frictions affect investment decisions. We use detailed

data on 74,670 individual projects in the U.S. natural gas industry to show that pri-

vate firms respond less than public firms to changes in investment opportunities.

Private firms adjust drilling activity for low capital-intensity investments. However,

they do not increase drilling in response to new capital-intensive growth opportu-

nities. Instead, they sell these projects to public firms. Our evidence suggests that

differences in access to external capital are important in explaining the investment

behavior of public and private firms.

UNDERSTANDING WHEN AND WHY LISTING status affects investment decisions is

important due to the large role of private firms in the U.S. economy.1Public

firms have better access to external financing than private firms.2This im-

proved access to capital markets could make public firms more responsive to

new investment opportunities. However, once publicly traded, the separation of

ownership and control can make a firm more prone to agency conflicts, whereby

management’s incentives can lead to significant distortions in investment

∗Erik P. Gilje is with The Wharton School, University of Pennsylvania. J´

erˆ

ome P. Taillard is

with Babson College. For their helpful comments and suggestions, we thank Zahi Ben-David, Joan

Farre-Mensa, Laurent Fr´

esard, Xavier Giroud, Itay Goldstein, Todd Henderson, Yelena Larkin,

Nadya Malenko, Gordon Phillips, Lee Pinkowitz, Jon Reuter, Berk Sensoy, Andrei Shleifer, Phil

Strahan, Per Str¨

omberg, Ren´

e Stulz, Michael Weisbach, Luigi Zingales, the Editor, the Associate

Editor, two anonymous referees, and seminar participants at Boston College, the Federal Reserve

Bank of Boston, the NBER Spring 2012 Corporate Finance conference, The Ohio State University,

the 9th Annual Conference on Corporate Finance at Washington University in St. Louis, Rutgers

University,the 2013 SFS Cavalcade, the 2013 Adam Smith Workshops in Asset Pricing and Corpo-

rate Finance, the 2013 Western Finance Association Meetings, and the Tenth Annual Penn/NYU

Conference on Law and Finance. We also thank Saeid Hozeinade for his research assistance. All

remaining errors are our own. We have read the Journal of Finance’s disclosure policy and have

no conflicts of interest to disclose.

1For 2008, we estimate that in the United States at least 94.3% of business entities were

privately held and 53.8% of aggregate business net income was from privately held firms. These

calculations are based on data reported by the Internal Revenue Service in its Integrated Business

Dataset.

2See Pagano, Panetta, and Zingales (1998), Brav (2009), Schenone (2010), Saunders and Steffen

(2011), and Maksimovic, Phillips, and Yang (2013).

DOI: 10.1111/jofi.12417

1733

1734 The Journal of Finance R

policies.3In this study, we investigate how these forces affect the investment

policies of public and private firms.

Identifying the influence of listing-related frictions on investment decisions

presents several challenges. First, data on private firms’ capital projects are

often unavailable. One therefore needs to focus on a setting in which detailed

data on project-level investment decisions and asset ownership are available

for all firms. Second, it can be difficult to accurately measure investment oppor-

tunities (e.g., Erickson and Whited (2000), Alti (2003)). One therefore needs to

observe responses to well-defined measures of changes in investment opportu-

nities for both public and private firms. Lastly, firms may differ on important,

non-listing-related, dimensions that could directly affect their investment de-

cisions; for example, firms may differ in cost structure, pricing power, or tech-

nology. One therefore needs a setting in which the impact of these differences

is limited.

This study addresses the above challenges by focusing on project-level invest-

ment decisions in the U.S. natural gas industry. Our setting is attractive for

three reasons. First, we observe detailed characteristics for 74,670 individual

capital projects made by public and private firms. Second, we can evaluate the

investment response of firms to two different measures of investment opportu-

nities: (1) natural gas prices and (2) shale discoveries. Both measures affect the

investment opportunities of public and private firms similarly. Third, we show

that, for any given project, both public and private firms have similar costs,

output prices, and technology. This allows us to rule out many potential alter-

native explanations and confounding effects. We then use detailed information

on each project’s capital requirements, location, and ownership over time to

isolate the impact of different listing-related frictions.

We find that drilling by private firms is 60% less responsive to changes in

natural gas prices relative to public firms. This difference is driven by public

firms drilling relatively more than private firms when natural gas prices are

high, whereas both types of firms drill with similar intensity in low price en-

vironments. Given that the profitability of drilling new wells is directly tied

to natural gas prices (see Lake et al. (2013)), this evidence is consistent with

public firms being better able to increase drilling activity when more wells be-

come economically viable. The response of public firms is linked to their capital

market activity as public firms raise significant amounts of external capital

when natural gas prices are high.4This result suggests that access to capital

markets is important for the difference in investment responses we observe.

Investment responses are related to project capital requirements. Using a

quasi-natural experiment, we analyze investments for capital-intensive shale

projects in 102 counties. Using a difference-in-differences framework, we find

that public firms respond to shale discoveries with an 80.5% increase in

3See, for example, Jensen and Meckling (1976), Jensen (1986), Stein (1989), and Bertrand and

Schoar (2003).

4The average public firm in our sample raises the equivalent of 15% of its total assets in high

natural gas price years, compared with just 6% in low natural gas price environments.

Do Private Firms Invest Differently than Public Firms 1735

county-level drilling activity, whereas private firms do not show any increase

on average. Given the larger capital requirements of shale projects, this result

is consistent with private firms facing a higher cost of external capital relative

to public firms.

By tracking asset ownership over time, we show that private firms sell

capital-intensive projects to public firms. After shale discoveries, we find that

private firms sell their shale drilling tracts to publicly traded firms 63% of

the time, and only 5% of the time to other privately held firms. In contrast,

we find that private firms are just as likely as public firms to buy privately

held less capital-intensive nonshale assets. This evidence indicates that, de-

spite having access to the same projects, private firms do not pursue the more

capital-intensive ones. These asset sales provide a mechanism for private firms

to capture a portion of the projects’ benefits without having to make the capital

commitment.

The greater drilling response by public firms to both changes in natural gas

prices and shale discoveries could be consistent with overinvestment caused

by the separation of ownership and control (e.g., Jensen (1986), Stulz (1990)).

This hypothesis can be ruled out empirically, however, by looking at several

pieces of the evidence. First, public firms drill more than private firms only

when natural gas prices are high, that is, when investment returns are higher.

Additionally, we find a positive stock market reaction for public firms that

acquire natural gas assets from private firms. Markets recognize a wealth

transfer from private to public firms in these transactions, and provide the

necessary capital for public firms to develop these assets. Using the Rajan

and Zingales (1998) measure of external capital, we find that the average

public natural gas producer raises 34% of its capital expenditures from external

capital markets during our sample period. By accessing capital markets on a

regular basis, public firms are more likely to face a disciplinary effect from

financial markets similar to that induced by product market competition (e.g.,

Jensen (1986), Giroud and Mueller (2010)).

Given the capital needs required for shale development and the greater ac-

cess to external capital that public firms have, one might expect a significant

increase in the number of private firms seeking a public listing during our

sample period. There are no IPOs in the first four years of our sample period,

while 12 private firms go public following shale discoveries. All but one state

the need for better access to external capital to pursue shale drilling as the

motivation for their IPO. Although the rate of IPOs following shale discoveries

places our sample firms in the top 10% of all four-digit SIC codes during these

years, many private firms in our sample choose to remain private.

So why might a firm remain private? First, we find that only private firms

with the most significant shale opportunities will benefit from a public listing

once we factor in the large fixed costs related to a public listing (see Ritter

(1987)). Second, beyond transaction costs and disclosure considerations, Brav

(2009) and Brau and Fawcett (2006) highlight control and ownership rights

as important factors in the decision to remain private. One novelty of our

study is to highlight an alternative channel through which projects initially