Deleveraging and decline in revenue‐expense matching over time

• DOI: http://doi.org/10.1111/jbfa.12343
• Published date: 01 October 2018
• Author: Jeong‐Hoon Hyun,Hyungjin Cho
• Date: 01 October 2018

DOI: 10.1111/jbfa.12343

Deleveraging and decline in revenue-expense

matching over time

Jeong-Hoon Hyun1Hyungjin Cho2

1NEOMABusiness School

2UniversidadCarlos III de Madrid

Correspondence

HyungjinCho, Department of Business Adminis-

tration,Universidad Carlos III de Madrid, 28903

Getafe,Madrid, Spain.

Email:hcho@emp.uc3m.es

Fundinginformation

EuropeanRegional Development Fund,

Grant/AwardNumber: UNC315-EE-3636

JELClassification: G32, M41

Abstract

Accounting rules mandate that the cost of debt should be recorded

as an expense,while the cost of equity does not appear in the income

statement. Therefore, the amount of financing expense,and thus net

income,in the income statements depends on how firms finance their

business. Based on a clear, substantial trend of declining leverage

since the 1990s, we examine how changes in capital structure might

influence earnings attributes—the matching between revenues and

expenses. We find that the contemporaneous relation between rev-

enues and interest expense in US firms has decreased from 1972 to

2013, a result of both changes in leverageand the declining explana-

tory power of interest expense with respect to revenues. When

we construct the weighted average costs of capital based on the

costs of both debt and equity, we find the contemporaneous rela-

tion between revenues and the costs of capital has not significantly

changed. Our results indicate that differential accounting treatment

of the costs of debt and equity can affect earnings attributes through

change in capital structure.

KEYWORDS

earnings quality, financing decision, leverage, revenue-expense

matching

1INTRODUCTION

We examine whether capital structure (i.e., leverage) changes the interpretation of earnings attributes over time

by investigating how interest expense influences the contemporaneous relation between revenues and expenses.

Generally, we recognize expenses when we record the revenues they help generate. Accounting textbooks (e.g.,

Kieso, Weygandt, & Warfield, 2013; Spiceland, Thomas, & Herrmann, 2016), and surveys of company practice(e.g.,

Dichev,Graham, Harvey, & Rajgopal, 2013) argue for “matching” as a fundamental accounting convention. Therefore,

exploring the matching concept provides valuable help in understanding the properties of earnings. In particular,

revenue-expense matching makes earnings less volatile, more persistent, and more predictable than cash flows,

implying that matching influences earnings quality (Dechow, 1994; Dichev & Tang, 2008; Dichev et al., 2013). Several

studies suggest that recent changes in the economy and accounting standards affect the properties of accounting

earnings, including revenue-expensematching (Dichev & Tang, 2008; Donelson, Jennings, & McInnis, 2011; He& Shan,

J Bus Fin Acc. 2018;45:1031–1050. wileyonlinelibrary.com/journal/jbfa c

2018 John Wiley & Sons Ltd 1031

1032 HYUN ANDCHO

2016; Srivastava, 2014). One noticeable result in those studies is a decline in the contemporaneous relation between

revenues and expenses.

However, few attempts have been made to examinethe influence of changes in capital structure on the decline

in revenue-expense matching. Moreover,prior studies have not examined the influence of a key expense component,

interest expense, on matching. US companies have traditionallyused debt financing, the origin of interest expense, as

animportant source of investment funding (Armstrong, Guay, & Weber,2010; Graham, Leary, & Roberts, 2015). US cor-

porationsincreased their reliance on debt financing until the 1990s, but since then, this reliance has decreased steadily

(Graham et al., 2015; Michaely,Popadak, & Vincent, 2014; Strebulaev & Yang, 2013).1The steep decline in corporate

leverage has resulted in a decrease in interest expense. The fact that US corporationshave undergone huge changes

in capital structure in recent decades provides an interesting opportunity to investigate the influence of changes in

capital structure on earnings attributes. Specifically, we investigatewhether this change in capital structure leads to

time-series differences in revenue-expensematching.

The economy-wide decline in debt financing could influence revenue-expensematching since matching involves the

alignment of expenseswith the revenues they help generate. Matching can take three forms: (1) direct cause and effect

(e.g.,cost of sales), (2) indirect cost allocation (e.g., depreciation and amortization), or (3) immediate expensing of intan-

gible investments (e.g., advertisingand R&D expenses) (Zimmerman & Bloom, 2016). For interest expense,the match-

ing model engenders an aura of certainty that masks the arbitrariness of the allocation process because of the need

for periodic reporting. In other words, matching inherently deals with the short-run correlation between revenuesand

expenses using discrete periods. Therefore, when firms have the same revenue generatingmodels except for capital

structure, their revenue-expense matching can differ due to different accounting treatment of the costs of debt and

equity. If one firm finances its investment with debt, all of its cost of capital is recognized as an expense.In contrast,

if another firm finances its investment solely with equity,its cost of capital does not appear in the income statement.

Therefore, both the different accounting treatment of debt and equity financing and radical changes in leverage since

the 1990s likely influence the matching between revenuesand interest expense over time.

Studying US firms from 1972 to 2013, we find that firms with greater changes in leverageor interest expense expe-

rience a lower contemporaneous relation between revenuesand non-operating expenses. Further, when we follow the

classification of expensesproposed by Donelson et al. (2011),2we find a declining contemporaneous relation between

revenuesand other expenses (those other than cost of goods sold; selling, general and administration expenses; depre-

ciation and amortization; tax; and special items). We then explore the influence of financing activities on the decline

in revenue-expensematching by further dividing other expenses into interest expense and others, and investigate the

revenue-interest expense relation overtime. The decrease in matching between revenues and other expenses disap-

pears once we exclude interest expense from other expenses, whereas revenue-interestexpense matching declines

over the sample period. We also find that the deterioration in the contemporaneous relation between revenuesand

other expenses disappears when we focus on firms without large interest expensevalues and those that do not expe-

rience a significant fall in leverage. In sum, these results indicate that changes in interest expense, resulting from

intertemporal changes in leverage,are significantly associated with declining revenue-expense matching.

Next, we examine whether the influence of capital structure changes on matching results from accounting rule

issues. Differential treatment of the costs of debt and equity, together with the recent reduction in leverage, could

cause a deterioration in revenue-expensematching. Under current US accounting standards, interest expense, a real-

ized portion of debt financing costs, is recorded as a component of non-operating expense, while the cost of equity

financing is not recognized in the financial statements. Totest the influence of the differential treatment of the costs

of debt and equity on the matching between revenues and expenses,we replace interest expense with the overall cost

1Michaelyet al. (2014) observe that institutional investors induce the firm to maintain low leverage, whereas Strebulaev and Yang (2013) argue that manage-

rialcharacteristics explain the decision to maintain extremely low leverage.

2Donelsonet al. (2011) classify expenses into six components: cost of goods sold; selling, general, and administration expense; depreciation and amortization

expense; tax expense;special items; and other expenses. They argue that the deterioration in positive relation between revenues and expenses is primarily

attributableto an increase in the incidence of large special items.