Corporate finance research suggests that firms
approach risk in two different ways: risk management
or risk shifting. According to the risk-management
theory, firms minimize cash flow risks to ensure they have
sufficient funds to pursue profitable projects. In contrast,
the risk-shifting theory suggests firms invest in riskier assets
to increase the value of shareholders’ equity when the firm
faces potential bankruptcy. Even though the risk-shifting
theory is widely taught in finance courses, there are few real-world examples.
In the recent study “Risk Shifting versus Risk Management:
Investment Policy in Corporate Pension Plans,” University of Chicago Graduate School of Business professor Joshua D. Rauh studies the investment behavior of firms with defined benefit pension plans.
“Debt agreements don’t place restrictions on pension fund
investment policy, which is one reason why pensions are an
ideal context for studying whether risk management or risk
shifting strategies determine the way firms invest,” says Rauh.
A number of high profile bankruptcies—especially in the
airline industry—have required government bailouts to cover
the bankrupt firms’ pension plans. Rauh suggests that these
bankruptcies have contributed to the popular view that firms beat the system by engaging in risky investment strategies with pension fund assets.
However, Rauh finds that both administrative pension data
and survey data on pension fund asset allocation for a smaller sample of publicly traded firms support the role of risk management in pension fund investment behavior.
“One of the strengths of this study is that the data sources
are very complete,” says Rauh. “Consistent with the riskmanagement theories, firms consistently take less risk as they get closer to financial distress.”
Firms with poorly funded pension plans allocate a greater share of pension fund assets to safer securities such as government debt and cash, whereas well-funded plans invest more heavily in volatile assets such as equity. Among large
public firms, those with better credit ratings allocate greater shares of pension fund assets to equity and smaller shares to government debt and cash.
Rauh also considered the relationship between pension fund asset allocation and the firm’s estimated likelihood of bankruptcy and found that the higher the probability of bankruptcy, the safer the observed pension fund allocation. This is consistent with risk-management strategies.
“As the overall health of a firm deteriorates, it takes less and less risk in pension plan investments,” notes Rauh.
Overall, the incentive to limit costly financial distress plays a considerably larger role than risk shifting in explaining variation in pension fund investment policy among U.S. firms. Incentives to manage risk and avoid bankruptcy, debt overhang, or other underinvestment problems serve as constraints on risk taking.
Pensions
In defined benefit pension plans, a firm pledges retirement
benefits to employees according to a formula that is generally
a function of each employee’s age, years of employment, and
salary. The firm then funds these liabilities with dedicated
financial assets, the allocation of which is determined by officers
of the corporation often in consultation with investment
advisors. Firms often contribute the minimum required
amount to defined benefit plans, but can voluntarily
contribute more if the plan is underfunded.
Though defined benefit plans have been declining in
importance relative to defined contribution plans (in which retirement income is determined by employee contributions), assets in private sector defined benefit plans still totaled $1.9 trillion at the end of 2003.
Given that corporate boards and management control pension fund asset allocation, there are incentives for firms to underfund pension plans and invest the assets in risky securities. Many have argued that incentive conflicts in pension investing were exacerbated by the creation of the federal
Pension Benefit Guaranty Corporation (PBGC) in 1974. If a firm enters bankruptcy with insufficient pension assets to cover its liabilities, the PBGC takes over the pension plan and
guarantees that plan recipients receive their annual pensions up to the maximum set annually by law ($46,659 in 2006 for employees retiring at age 65). Employees who would be below this threshold, and the organizations representing them, therefore have a muted incentive to monitor the investment behavior of the pension sponsor before it gets into trouble.
Recent pension fund terminations have indeed strained
the financial position of the PBGC. For the fiscal year ending
in September 2000, the PBGC had a surplus of $9.7 billion.
For the fiscal year ending in 2005, the PBGC had a $22.7 billion
deficit. The recent termination of the pension plans of United
Airlines was the largest in history, with PBGC assuming $6.6
billion of United’s $9.8 billion in unfunded liabilities, which would shift the risk of poor pension performance onto the PBGC and the beneficiaries.
However, Rauh finds little evidence to support the claim
that the PBGC is the primary driver of risk taking in pension funds, since the firms for whom this insurance would be the most valuable generally take the least risk.
In the case of pension funding and investment policy, the
tension between risk shifting
and risk management is very
apparent. If a firm goes bankrupt,
the unfunded liability is
low in priority. This creates
incentives for firms to take
risks with the assets that back
that liability during the period
before a possible bankruptcy.
On the other hand, if a firm
survives and has to make large statutory contributions to its pension plan, it may face financial distress and lose the opportunity to undertake profitable investment projects, creating an incentive for firms to limit pension fund risk.
From this analysis, Rauh predicts that if risk shifting is an important determinant of risk taking in pension funds, one should observe financially distressed firms taking large risks. Within pension plans and within firms over time, risk taking should increase as funding and financial conditions deteriorate. If risk management is an important constraint on risk taking in pension funds, then risk taking should be smaller as funding and financial conditions deteriorate.
Patterns in Risk Taking
Rauh used two sources of asset allocation information: 1)
pension plan data from electronic IRS 5500 filings from the
Department of Labor covering 1990 to 2003; and 2) firmbased
survey data on the largest corporate pension sponsors
from Pensions and Investments, matched to Compustat data.
To determine which pension plans ended in bankruptcy,
Rauh matched data from IRS 5500 filings to data from the
PBGC website. The full sample consisted of 33,144 plans and
23,456 employees.
The Pensions and Investments survey covered the 1,000
largest U.S. pension sponsors from 1997 to 2004. On average,
firms have $3.5 billion in pension assets and liabilities. Over the sample period, firms allocated 30 percent of assets to debt and cash, and 63 percent of assets to equity.
One method of testing theories of risk shifting against risk management is to examine the funding status of the pension
plan. The level of assets to fund a given liability is a cumulative
result of all past investment returns, as well as past decisions
about how much to contribute to the plan with respect to benefits paid. Plans that are well funded are likely to reflect financially healthy sponsors.
Rauh documented the percentage that each pension fund
invested in equity against the plan characteristics, and found
that better funded plans invest more in equity securities. If the pension funding deteriorates within a firm over time, the percent of the fund invested in safe securities tends to increase.
Rauh also studied the relationship between firm credit
ratings and asset allocation. Relative to their observable characteristics,
the pension investment behavior of firms seems
best characterized by theories of risk management in which firms in worse financial condition limit risk rather than increase speculation.
For large firms that have accessed public debt markets,
the Standard & Poor’s credit rating is one of the best available
measures of a firm’s financial strength and likelihood of
defaulting on debt agreements. The study finds that firms
with better credit ratings invest their pension funds in riskier
assets. Furthermore, within a given firm over time, the higher
the credit rating, the larger the proportion of assets invested in riskier securities and the smaller the proportion of assets invested in safe securities.
In addition, the allocation of pension fund assets to equities
is positively correlated with the share of participants who are current employees. The duration of pension liabilities thus plays a role in pension fund investment strategy.
Did plans of firms that went bankrupt take on more investment
risk in the years before bankruptcy, as the theory of risk
shifting would predict? Rauh compared the asset allocation
of firms that went bankrupt with those that survive, and found
neither a statistically nor an economically significant relationship between risk taking in pension plans and indictors of bankruptcy.
Plans that ultimately terminated do not appear to have more volatile annual investment returns over the sample period than those that survived, nor were they more volatile than the PBGC’s own investments.
Dynamic Investing
“There has not been enough consideration of the relationship
between risk and investment opportunities, by virtue of the fact
that there have been so many firms with pension problems
that have impacted firm operations,” says Rauh. “Firms
should do more than wait until problems arise and manage
risk at that point. Shareholders want to see more dynamic risk
management that creates value for the firm.”
Rauh’s study also shows that there are some natural limits on the extent to which firms are going to take risk to expropriate creditors—an important consideration for policymaking.“Pension funds are not trying to game the system in terms of investments,” says Rauh. “One may not need large interventions to limit risk taking, because the desire to avoid costly financial distress serves as a natural constraint.”
“Risk Shifting versus Risk Management: Investment Policy in Corporate Pension Plans.” Joshua D. Rauh.