The influence of other comprehensive income on discretionary expenditures

• Author: K.C. Lin,Roger C. Graham
• Published date: 01 January 2018
• DOI: http://doi.org/10.1111/jbfa.12284
• Date: 01 January 2018

DOI: 10.1111/jbfa.12284

The influence of other comprehensive income

on discretionary expenditures

Roger C. Graham1K.C. Lin2

1OregonState University, 424 Austin Hall, Cor-

vallis,Oregon, USA

2OregonState University, 430 Austin Hall, Cor-

vallis,Oregon, USA

Correspondence

RogerC. Graham, Oregon State University, 424

AustinHall, Corvallis, OR 97331, USA

Email:roger.graham@bus.oregonstate.edu

Abstract

Other comprehensive income items (OCI) increase and decrease

book value and therefore indicate more or less firm value. It follows

that OCI items, albeit transitory, may contribute to a wealth effect

that influences expenditure decisions. In support, our regression

results indicate an association between current year OCI and future

discretionary financing, investing, and operatingexpenditures. How-

ever, we also find that OCI-influenced expendituresare not associ-

ated with future profitability, suggesting such expenditures are not

value creating. In further tests, we find that future discretionary

expenditures are associated with both positive OCI and negative

OCI for higher leveragedfirms but only associated with positive OCI

forlower leveraged firms. These results suggest that, for highly lever-

aged firms, positive OCI loosens debt constraints on future expendi-

tures while negative OCI tightens debt constraints on future expen-

ditures. Forfirms without debt constraints the results are suggestive

of possible wealth transfers from debtholders to shareholders.

KEYWORDS

debt constraints, discretionary investing, financing and operating

expenditures, other comprehensive income, wealth effect, wealth

transfers

1INTRODUCTION

The information value of other comprehensive income relative to earnings is not well understood. Other comprehen-

sive income (OCI) and earnings appear to havesimilar components. OCI and earnings both contain gains that increase

owners’ equity and indicate more firm value. Similarly, OCI and earnings both contain losses that decrease owners’

equity and indicate less firm value. The wealth effect of gains and losses on firm value suggests, all things being equal,

that managers with firm value increases will tend to spend more and managers with firm value decreases will tend to

spend less. However, OCI items and earnings items differ fundamentally. Unrealized gains and losses are likely to be

presented, at least temporarily, in owners’ equity as OCI items while realized gains and losses appear on the income

statement as components of earnings. In part because OCI gains and losses are unrealized, they are largely considered

transitory thereby potentially limiting their usefulness for predicting future cash flows (Jones & Smith, 2011), income

72 c

2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2018;45:72–91.

GRAHAM ANDLIN 73

(Rees & Shane, 2012), and market values (Barker,2004).1It follows that unrealized OCI gains and losses may not be

indicative of wealth realizations, and related spending decisions will be unwarranted.

At first, the transitory nature of OCI would suggest little to link OCI with future expenditures. However,

Chambers, Linsmeier,Shakespeare, and Sougiannis (2007) report a response coefficient greater than one for the asso-

ciation between OCI unrealized holding gains and losses from available for sale securities and stock returns. Although

inexplicable if such gains and losses are transitory,the greater than one response coefficient could i ndicatethat the

wealth effect from OCI gains and losses is viewed by market participants as having consequence. In this study, we

address the question of whether OCI gains and losses have consequence (i.e., a wealth effect) by looking for links

between changes in OCI and following-year discretionary financing, investing, and operatingexpenditures. We define

discretionary financing expenditures as changes in net payouts to shareholders, discretionary investingexpenditures

as the sum of changes in capital and acquisition-related expenditures, and discretionary operating expenditures as

changesin advertising, research and development, and selling and administrative expenditures. Regression results indi-

cate positive associations between OCI and these one-year ahead discretionary expenditurevariables.

We then test for associations between the OCI-influenced discretionary expenditures and future performance

as represented by two-year ahead return on assets (ROA). Regression results indicate negative relations between

the OCI-influenced discretionary expenditures and future ROA,suggesting that OCI-influenced expenditures are not

directly value creating. We hypothesize that, despite being non-value creating, OCI may influence expenditures indi-

rectly because of the inherent conflict betweenowners and c reditors. In the contextof the conflict, OCI could influence

discretionaryexpenditures under two settings. The first setting involves debt contracts when financial statements have

a direct bearing on contractual outcomes (DeFond& Jiambalvo, 1994; Press & Weintrop, 1990; Watts & Zimmerman,

1986). Specifically,debt contracts often contain covenants based on numbers reported in financial statements (Begley

& Freedman,2004). OCI gains and losses impact current balance sheets when the OCI items are recorded in equity but

impact future income statements when the OCI items recycle out of equity onto the income statement.2When debt

covenants are tied to the balance sheet, OCI gains recorded in equity will cause covenants to be less restrictive and

managerial discretion will be loosened. Similarly, OCI losses will cause covenantsto be more restrictive and manage-

rial discretion will be constrained. Thus, a firm with debt covenants will haveincreased flexibility when OCI is positive

and less flexibility when OCI is negative.

The second setting where OCI could influence discretionary spending involves potential wealth transfers from

creditors to owners (Jensen & Meckling, 1976; Modigliani & Miller, 1958). Wealth transfersoccur because creditors

and owners share disproportionally in firm outcomes and because owners, acting through management, make current

expendituredecisions that are ultimately dependent on future cash flow realizations. Relative to the OCI wealth effect,

the owners may makeexpenditure decisions based on unrealized gains before cash flows from the gains are realized. If

the cash flows are realized, anyincremental benefits from the expenditures will accrue to the shareholders because the

creditors'returns are fixed.However, if the cash flows are not realized, the creditors may be forced to share in any cash

shortfalls. Thus, shareholders can transfer some of the cash flow risk inherent in unrealized gain recognition to cred-

itors. In this context, we can assume expenditures will be more likelywith the positive wealth effect of OCI increases

than with the negative wealth effect of OCI decreases.

We look for evidence on whether OCI influences discretionary expenditures because of restrictions related to

debt and/or because of potential wealth transfers. Forthese tests we focus on differences for higher debt and lower

debt firms in associations between positive and negative OCI and discretionary expenditures. We predict that debt-

related OCI-influenced expenditures will be inverselyassociated with both positive and negative OCI changes as debt

1The unrealized nature of OCI items has been described as transitory because OCI items are more related to uncontrollable market-widefluctuations and

lessrelated to firm-specific performance and growth opportunities (Barker 2004; Chambers et al., 2007; Jones & Smith, 2011; and Rees & Shane, 2012). Rees

andShane (2012) refer to an inherent uncertainty between OCI items and subsequent cash flows as the basis for the provisions requiring unrealized gains and

losses to bypass the income statement. One implication of the inherent uncertainty is that OCI items should be less representativeof changes in firm value

and thus less useful for managers in their future planning. Barker(2004) argues that because OCI items are related to unrealized market value changes (i.e.,

changesin the market value of available for sale securities), OCI items cannot be expected to predict future market value changes.

2Our focus is on OCI as a balance sheet item. OCI recyclesout of shareholders’ equity onto the income statement when items are realized. Recycling should

haveno effect on total shareholders’ equity as recycling in effect moves OCI items into retained earnings by way of the income statement.