Executive compensation and cash contributions to defined benefit pension plans

• Published date: 01 October 2018
• Date: 01 October 2018
• Author: Laura Swenson,Qiang Cheng
• DOI: http://doi.org/10.1111/jbfa.12339

DOI: 10.1111/jbfa.12339

Executive compensation and cash contributions

to defined benefit pension plans

Qiang Cheng1Laura Swenson2

1SingaporeManagement University, Singapore

2Universityof Wisconsin - Milwaukee, Milwau-

kee,Wisconsin, USA

Correspondence

LauraSwenson, P.O.Box 742, 3202 N. Maryland

Avenue,Milwaukee, WI 53201-0742, USA.

Email:swensola@uwm.edu

JELClassification: M41, M48, J32, J33

Abstract

Cash contributions to defined benefit pension (DB) plans reduce

cash flows from operations without directly affecting the current

year's net income. We utilize this unique setting to investigate how

managerial incentives to report higher cash flows from operations,

executive compensation in particular, affect contributions. Using a

comprehensive dataset of DB plan contributions, we find that firms

with higher chief executive officer (CEO) compensation contribute

less to DB plans, consistent with managers benefiting from lower

pension contributions. Results are stronger when CEO compensa-

tion is directly linked to cash flows from operations or when CEO

compensation is more sensitive to cash flows from operations. We

also find enhanced disclosure and more transparent reporting under

the recent pension accounting regime mitigates the negative associ-

ation between executivecompensation and cash contributions.

KEYWORDS

defined benefit pension plan, executive compensation, pension

accounting, pension contribution

1INTRODUCTION

In this paper,we examine the impact of managerial incentives on companies’ contributions to their defined benefit pen-

sion (DB) plans. In particular,we investigate whether firms with higher CEO compensation contribute less to DB plans.

The motivation for this research question is two-fold. First, in the United States, the funding status of DB plans and

companies’ abilities to contribute to their pension plans have decreased significantly over the last couple of decades.

This is due to an unprecedented number of corporate bankruptcies, which has dramatically increased the financial

burden of the Pension Benefit Guaranty Corporation (PBGC).1,2 For example,on July 23, 2009, the PBGC agreed to

take on US$6.2 billion of pension liabilities from Delphi Corporation, a bankrupt auto supplier.This is just one of the

companies the PBGC recently assumed responsibilities from. Other notable examples include Lehman Brothers and

1Under a DB plan, the retirement benefit that an employee will receiveis a function of the employee's years of service and compensation level in the years

approachingretirement. Companies make periodic contributions to pension plans to ensure there are enough assets to cover the promised future benefits.

2The PBGC is a federalcorporation that insures the retirement future of nearly 44 million people in over 26,000 private sector DB plans (www.pbgc.gov).In

theevent that a company goes bankrupt and does not have sufficient assets to cover its pension obligations, the PBGC will provide the retirement benefits to

theparticipants of the plan. The PBGC's main income source is the premium paid by all companies with insured DB plans.

1224 c

2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2018;45:1224–1259.

CHENG ANDSWENSON 1225

Circuit City. Because of these and subsequent events(e.g., lackluster pension asset returns and low interest rates), as

of September 30, 2015, the PBGC had a deficit of US$76.35 billion, almost seven times the deficit as of September 30,

2008. If in the long run the PBGC does not have sufficient funding to cover its obligations, (1) Congress will have to

increase the premium paid by all companies with DB plans, (2) the government, and ultimately taxpayers, will haveto

bail out the PBGC, and/or (3) insured participants will suffer the losses. As such, it is important to understand compa-

nies’ funding decisions relating to their DB plans.

Second, under the current accounting rules, a contribution to a DB plan is simply a transfer on the balance sheet,

from cash to net pension asset/liability.A DB plan contribution has no direct impact on current year's net income, but it

reduces cash flows from operations. Severalrecent studies demonstrate an increased demand for information relating

to firms’ cash flows from operations (e.g.,DeFond & Hung, 2003; Givoly, Hayn, & Lehavy,2009) and the adverse conse-

quences of missing analysts’ cash flow forecasts (e.g., Edmonds, Edmonds, & Mahor,2011; McInnis & Collins, 2011). As

such, the board of directors and shareholders likely incentivize CEOs to increase reported cash flows from operations

by rewarding CEOs reporting higher cash flows from operations with higher compensation. As a result, we argue that

managers likely delay pension contributions when theycan personally benefit by receiving higher compensation from

reporting higher cash flows from operations.

Despite its importance, there is limited research on companies’ contribution decisions concerning DB plans, and

to our knowledge, there is no research on how managerial incentives affect DB plan contributions. This is largely

attributed to the fact that most DB plans in the United States were well funded in the 1980s and 1990s, thanks to

the high returns from pension assets and high interest rates during that period. After the burst of the Internet bubble

in 2000 and 2001, the returns from pension assets were lowered, leading to lower pension assets. During the same

period, interest rates decreased, resulting in increased pension obligations. The recent financial crisis has exacerbated

the problem. In the period 2001–2012, 70–95% of DB plans were underfunded, up from around 30% in the late 1980s

and 40% in the 1990s, as shown in Panel A of Figure 1. PanelB of Figure 1 shows that the magnitude of unfunded pen-

sion obligations has also increased from roughly zero in the 1980s and 1990s to over US$600 billion for S&P500 firms

and over US$1.1 trillion for the Compustat population in 2012.

Another reason for the lack of research on the contribution decision is data availability. The contribution data

reported in Compustat and the pension footnotes to the annual reports often do not separate contributions to DB

plans from those to defined contribution plans. We solve this problem by obtaining DB plan contribution data from

companies’ Form 5500 filingswith the Internal Revenue Service (IRS).

Afirm's decision to contribute to a DB plan reflects the currentperiod trade-off between the benefit, increasing pen-

sion assets and reducing the net pension liabilities for underfunded DB plans, and the cost, reducing cash flows from

operations and cash available for other purposes. Prior research shows that cash flows from operations are an impor-

tant determinant of CEO compensation (e.g., Nwaeze, Yang, & Yin, 2006). Wepredict that managers have incentives

to contribute less to DB plans in order to report higher cash flows from operations and increase managers’ compen-

sation.3However,delaying DB plan contributions leads to higher net pension liabilities in the current period, which in

turn leads to higher pension expensesand lower net income in future periods.4Lower net income in the future can lead

to lower compensation in the future. Thus, whether executive compensation considerations affect pension contribu-

tions is an empirical question.

Our initial sample consists of 5,221 firm-year observations from S&P 1500 firms in the period 1994–2007 (the data

on DB pension contributions ends in 2007). Consistent with our prediction, we find a negative association between

3One keyassumption is that various stakeholders do not expect managers to delay contributions to report higher cash flows from operations. Even if these

stakeholdersdo and react accordingly, the model in Stein (1989) implies that in equilibrium managers still delay pension contributions and report higher cash

flowsfrom operations. When outsiders think reported cash flows from operations are higher due to lower pension contributions, they will ‘discount’ reported

cashflows from operations. To offset this discount, managers reduce pension contributions and report higher cash flows from operations.

4Contributionsto DB plans increase pension assets. Under the current accounting standards, the expected return (not the actual return) from the beginning-

of-period pension assets reduces pension expense. Thus, cash contributions to pension assets reduce pension expense in the futureperiod by the expected

returnfrom such contribution.

1226 CHENG ANDSWENSON

FIGURE 1 Fundedstatus of defined pension plans overtime [Colour figure can be viewed at wileyonlinelibrary.com]

This figure presents the percentage of firms with underfunded defined benefit pension plans (Panel A) and the

magnitude of unfunded pension obligations (Panel B) in the period 1986–2012. Wepresent the chart for both

S&P500 firms and all firms in Compustat population with data on the fair value of pension assets and projected

benefit obligations (PBO).

pension contributions and CEO compensation. Firms with higher CEO compensation contribute less to pension plans,

indicating that managers likely seek personal benefits at the expenseof employees’ retirement benefits.

Since the argument for managers to delay pension contributions in order to report higher cash flows from oper-

ations rests on the importance of cash flows from operations for CEO compensation, within our full sample, we use

two approaches to identify a group of firms for which cash flows from operations is an important determinant of

CEO compensation. First, we identify a subsample of firms that explicitly mention cash flows from operations when

discussing CEO compensation in their proxy statements. Consistent with our prediction, we find that the negative

association between executive compensation and pension contributions is more pronounced for these firms than for