Institutional Investor Ownership and Corporate Pension Transparency

• Author: Abhishek Varma,John R. Nofsinger,Tim V. Eaton
• DOI: http://doi.org/10.1111/fima.12045
• Published date: 01 September 2014
• Date: 01 September 2014

Institutional Investor Ownership and

Corporate Pension Transparency

Tim V. Eaton, John R. Nofsinger, and Abhishek Varma∗

We examine the association between institutional ownership and defined benefit (DB) pension

decisions. Wefind that institutional ownership is negatively associated with pension underfunding,

opportunistic increases in the expected rate of return assumption in the presence of underfund-

ing, and significant ownership of the firm’s own stock in the DB plan portfolio. Furthermore,

these relations are stronger when institutional ownership is concentrated, when institutions are

nontransient investors, or when institutions are relatively large. These results suggest that in-

stitutional investors are monitoring firm pension decisions, particularly those institutions with

stronger monitoring incentives or resources.

Shareholder activism has been a popular investor activity for many decades. Gillan and Starks

(2007) detail the evolution of this activism. They note that one important factor of shareholder

activism is the accumulation of large blocks of ownership in the portfolios of institutional in-

vestors, such as public pension funds, mutual funds, insurance companies, and hedge funds. As

such, many scholarly studies investigatewhether these institutional investors monitor managerial

decisions. Most of these studies focus on the institutional ownership impact on highly visible el-

ements of corporate governance structures, such as executive compensation (Hartzell and Starks,

2003; Almazan, Hartzell, and Starks, 2005), chief executive officer (CEO) turnover (Parrino,

Sias, and Starks, 2003), and governance proxy proposals (Gillan and Starks, 2000). Institutional

ownership and other managerial decisions have also been investigated including disclosure prac-

tices (Bushee and Noe, 2000), earnings forecasts (Ajinkya, Bhojrag, and Sengupta, 2005), insider

trading (Sias and Whidbee, 2010), research and development (R&D) investment (Bushee, 1998),

and financial reporting (Burns, Kedia, and Lipson, 2010).

We examine the impact of institutional ownership on pension plans in the context of under-

funding, the expected rate of return (ERR) assumption, and ownership of a firm’s own common

stock. Underfunding of defined benefit (DB) pension plans increases the uncertainty of fulfilling

future pension obligations and adversely impacts future cash flow and earnings. In addition, an

opportunistic increase in the ERR assumption in the presence of underfunding leads to reduced

pension expense and could reflect increased risk taking with plan assets. Lower pension expenses

lead to earnings manipulation, while riskier investments place an already underfunded plan at

greater risk. Also, significant ownership of a firm’s own securities may represent a managerial

attempt to gain corporate control. Given these scenarios, institutional investors, which generally

hold large stakes in firms, should have significant incentives to monitor DB pension plans.

Dr.Eaton gratefully acknowledges funding from the Farmer School of Business.

∗Tim V. Eaton is the EY Teaching Scholar and Associate Professor in the Department of Accountancy at the Richard T.

Farmer School of Business at Miami University in Oxford, OH. John R. Nofsinger is the William H. Seward Endowed

Chair in Finance and Professorin the Business Administration Department at the College of Business and Public Policy

at the University of Alaska in Anchorage, AK. Abhishek Varma is the Dixie L. Mills Scholar in Finance and Assistant

Professor in the Department of Finance, Insurance, and Law at the College of Business at Illinois State University in

Normal, IL.

Financial Management •Fall 2014 •pages 603 - 630

604 Financial Management rFall 2014

Lara, Osma, and Penalva(2009) f ind that strong corporate governance can providea monitoring

role by producing more conservative discretionary accruals. Burns et al. (2010) provide mixed

results concerning the role of institutional ownership in monitoring financial reporting practices.

They examine the impact of both aggregate and concentrated institutional ownership. Aggregate

institutional ownership is positively related to the likelihood of financial misreporting, a result

that is driven by short-horizon investors. In contrast, they also report that concentrated institu-

tional holdings have an offsetting behaviorand conclude that the nature of institutional ownership

influences monitoring outcomes. While institutional investors appear to playan effective monitor-

ing role in relation to visible governance mechanisms, such as executive compensation (Hartzell

and Starks, 2003), their success in monitoring less transparent managerial activities remains to

be determined. Agency costs could occur in low visibility decisions that must be inferred from

accounting statements and various statutory Securities Exchange Commission (SEC) filings. A

corporation’s DB plan is a good example of this. Asthana (1999) suggests that managers may

manipulate the actuarial assumptions of the plan to manage the required cash flow of the plan

and to avoid visibility costs. Lower cash flows in the short term can increase short-term earnings,

but may leave the plan underfunded. This could result in a future liability that could require

higher payments to fund the pension plan (and, as such, lower earnings). Bergstresser, Desai, and

Rauh (2006) find that managers opportunistically manage ERR on pension assets (PA) to impact

earnings. They determine that the least constrained executives (measured through a corporate

governance index) tend to manipulate pension plans assumptions most aggressively. They con-

clude that pension accounting has many opportunities for earnings management. Furthermore,

Picconi (2006) concludes that investors and analysts fail to fully incorporate pension accounting

information into prices and forecasts on a timely basis.

This study makes a unique contribution to the literature by examining institutional ownership

and three progressivelyless transparent aspects of the f irm’s DB pension plan. Wef irst investigate

the ramifications of pension decisions as revealed by the funding level of the plan (underfunded

vs. fully funded). The examination of funding levels is important as significant underfunding of

pension plans could impact the future solvency of the firm. Using data from 1988 to 2008, we

explore the association between lagged institutional ownership and pension plan underfunding.

Consistent with the monitoring incentives of institutional investors,we find a negative relationship

between institutional presence and the probability of pension underfunding. That is, higher

institutional ownership is associated with better funded plans.

Next, we explore the ERR, an important pension accounting assumption. ERR refers to the

assumed long-term rate of return on PA, which may differ from the actual short-term returns. It

impacts the earnings statements through pension expense and is reflective of the plan’s investment

policy.Companies may have more leeway in setting their ERR than theydo in setting the discount

rate (or the salary and benefit progression rates) for pension liabilities as accounting rules for

the latter are more prescriptive (Bergstresser et al., 2006). Significantly underfunded pension

plans require additional cash contributions and the amortization of pension losses, tempting

managers to opportunistically increase the ERR. An increase in ERR lowers pension expenses to

the pension plans, resulting in an increase in near-term earnings of the firm. Also, an increase

in ERR in the presence of significant underfunding could reflect concerns about the pursuit of

a riskier/aggressive investment strategy aimed at reducing the funding shortfall that could place

the pension plan at greater risk in the future.

Prior studies indicate opportunistic management of the ERR assumption. Amir and Benartzi

(1998) find that the cross-sectional variance in ERR across companies cannot be adequately

explained by differences in pension fund investment strategies (explained by differences in the

percentage of equity holdings). Bergstresser et al. (2006) determine that managers increase

the ERRs as they prepare to acquire companies and prepare to exercise stock options, further