The hedging benefits of industrial and global diversification: Evidence from economic downturns

• DOI: http://doi.org/10.1111/jbfa.12354
• Author: Yilei Zhang,Yixin Liu,David C. Mauer
• Date: 01 October 2018
• Published date: 01 October 2018

DOI: 10.1111/jbfa.12354

The hedging benefits of industrial and global

diversification: Evidence from economic

downturns

Yixin Liu1David C. Mauer2Yilei Zhang2

1PeterT. Paul College of Business and Economics,

Universityof New Hampshire, 10 Garrison

Avenue,Durham, NH

2Departmentof Finance, Belk College of

Business,University of North Carolina at

Charlotte,9201 University City Blvd.,

Charlotte,NC

Correspondence

DavidC. Mauer, Department of Finance, Belk

Collegeof Business, University of North

Carolinaat Charlotte, 9201 University City

Blvd.,Charlotte, NC 28223.

Email:dmauer@uncc.edu

Abstract

In a large panel of US firms, we find that discounts for industrial

and global diversification significantly decrease in economic down-

turns, suggesting a valuable hedging benefit of diversification. This

finding is robust to a wide variety of business cycle and economic

uncertainty variables and persists when we account for endogene-

ity of industrial and global diversification. We further analyze the

channels through which diversification is more valuable in economic

downturns. We find that the value improvement is largely explained

by increases in investmentand product market performance relative

to focused firms. There is no evidence that improved internal capi-

tal market efficiency or access to external finance contribute to the

hedging value of diversification.

KEYWORDS

diversification, firm valuation, globalization, hedging

1INTRODUCTION

Despite mixed views in the literature about the benefits of corporate diversification,it is an important strategic deci-

sion for firms. Firms pursue diversification in two primary dimensions – diversify across industries and/or diversify

across borders. For the sample of firms used in this paper over the period 1984 to 2014, we see a steady decrease in

industrial diversification with the fraction of industrially diversified firms stabilizing around 3% entering 2000. Over

the same time period, a growing number of firms have become globally diversified. From1984 to 2014, the fraction of

firms with foreign sales (i.e., sales of goods and services produced outside the US) increased from 33% to 52% and for

these firms, foreign sales to total firm sales increased from 22% to 43%. Despite a rich literature in economics explain-

ing why firms expand abroad (see Helpman, 2006 and Syverson, 2011 for reviews), empirical evidence is mixed on

whether global diversification increases firm value. Some studies find that global diversification increases firm value

(Bodnar,Tang, & Weintrop, 1999; Gande, Schenzler,& Senbet, 2009; and Sturgess, 2016), while other studies find that

globally diversified firms, like domestically diversified firms, have significantly lower marketvalues than comparable

portfolios of specialized firms (Denis, Denis, & Yost,2002).1

1Startingwith Lang and Stulz (1994) and Berger and Ofek (1995), a large literature documents a diversification discount in industrially diversified firms.

1322 c

2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2018;45:1322–1351.

LIU ET AL.1323

The finance literature focuses on three primary explanationsfor industrial and global diversification discounts. The

first explanation is that diversified firms allocate capital inefficiently and this decreases firm value. Rajan, Servaes,

and Zingales (2000) and Scharfstein and Stein (2000) argue that agency problems in internal capital markets result

in an inefficient allocation of resources between business segments relative to the allocations that would result if

the segments were standalone entities seeking financing in the external capital market. The evidence in support of

this view is mixed. Lamont (1997), Rajan et al. (2000), Shin and Stulz (1998), and Ozbas and Scharfstein (2010) find

evidence in support of inefficient allocations. In contrast, Billett and Mauer (2003), Hovakimian (2011), Maksimovic

and Phillips (2002), and Kuppuswamy and Villalonga (2016) provide evidence of efficient allocations.2The second

explanationis that a desire on the part of managers to empire-build and/or diversify their human capital induces value-

reducing industrial and global diversification strategies. In support of this explanation, Denis, Denis, and Sarin (1997)

find that diversified firms have low managerial share ownership and equity holdings by large outside blockholders.

They further find that decreases in diversification tend to occur after corporate control threats. Hoechle, Schmid,

Walter,and Yermack (2012) find that poor corporate governance explains 16% to 21%of the diversification discount.

Finally, Campa and Kedia (2002) and Villalonga (2004b) argue that estimates of the discount are biased because the

decision to diversify is endogenous and unobserved factors influence firm diversification and value.3However,work

by Ammann, Hoechle, and Schmid (2012), Laeven and Levine (2007), Schmid and Walter (2009), and Hoechle et al.

(2012) find a robust diversification discount after accounting for endogeneity and omitted variables bias.4

This paper contributes to the literature by examiningwhether the state of the economy influences the relationship

between diversification and firm value. In principle, when markets are incomplete or firms and investors face signifi-

cant financial market frictions, diversification provides hedging value when the fortunes of one business segment are

imperfectly correlated with the fortunes of another. For example, in a US domestic recession that affects primarily

US firms, a globally diversified firm may be better positioned to weather a decline in US demand for its products and

services if it has significant sales from foreign operations. This hedging effect should be particularly valuable during

economic downturns as firms struggle to maintain profitability. To test if this is the case, we first examine whether

discounts for industrial and global diversification narrow during economic downturns when the hedging benefits of

diversification are larger.We directly estimate the reduction of diversification discounts in economic downturns, and

the hedging value of industrial and global diversification over our sample period. We then examine the financial and

real determinants of the hedging value of diversification and assess the explanatorycontribution of each determinant.

We use the Compustat Business and Geographic segment databases to construct a panel dataset of US firms over

the period from 1984 to 2014. Firm-years are classified as single-segment domestic, multi-segment domestic, single-

segment global, or multi-segment global based on the number of business segments and whether a firm reports foreign

sales.5Following the literature, we denote multi-segment firms as industrially diversified and firms with foreign sales

(whether single- or multi-segment) as globally diversified. Global firms are further categorized by the intensity and

location of foreign operations.

Following Berger and Ofek (1995), for each firm-year we compute excessvalue as the logarithm of the ratio of a

firm's actual value to its imputed value, where imputed value is the sum of the imputed values of its segments using

sales multiples for same industry-year single-segment US domestic firms. We estimate the value of diversification in

regressions of excess value on dummy variables for industrial and global diversification, control variables, and firm

and time fixed effects to account for possible omitted variables bias. The coefficient estimates on the diversification

2SeeStein (1997) and Matsusaka and Nanda (2002) for models predicting efficient allocations in internal capital markets.

3Chevalier (2004), Graham,Lemmon, and Wolf (2002), Hyland (2003), Hyland and Diltz (2002), Mansi and Reeb (2002), Villalonga (2004a), Whited (2001),

andCreal, Robinson, Rogers, and Zechman (2014) point to other problems with the estimation of industrial and global diversification discount estimates.

4Creal et al. (2014) argue that estimates of the global diversification discount are biased if the value of a US multinational firm is compared to a benchmark

portfolio of US same industry standalone firms. However,as originally noted by Denis, Denis, and Yost (2002) and more recently quantified by Lee, Naranjo,

and Sirmans (2016), the problem with comparing a US multinational to foreign country counterparts is that differences in legal, regulatory,tax and political

institutionsinfluence the comparison and thereby bias the estimate of the value of global diversification.

5Foreign sales result from the sale of goods and services produced outside the US (e.g., a US firm with a manufacturing facility in Europe). The goods and

servicesmay be sold in the US (i.e., domestically), in the foreign country where they are produced, or in another country.