Question: How Does Investment Return Affect Pension Cost?
Answer: It doesn’t.
Yes, a higher return on plan assets reduces the funding requirements for the pension plan and the expense that the sponsor must report. But the plan’s true economic cost is independent of the investment performance of the plan assets.
To see why this is so, suppose that you establish a fund to pay for your child’s college education and I do the same for my child. We make equal contributions to our respective funds, and we both face the same tuition payments. But being a smarter, bolder, or luckier investor, you grow your college fund to twice the size of mine. Can we now say that your child’s education costs less than my child’s education? Surely not. Our tuition payments are the same; it’s just that you have a larger education fund available to help pay your child’s tuition.
Or think of it this way: Suppose that your college education fund performed miserably and a similar fund that you had set up to buy a small vacation home struck it rich. Would you now say that college tuition has become very expensive but vacation homes very cheap? Can you now afford to buy a vacation mansion — or private island — but not to send your child to college? Behavioral economics suggests that you might think along those lines, but common sense says, “Get over it.”
Similarly, a higher pension fund return does not lower the economic cost of the plan. The economic cost reflects solely the amount and timing of the pension payments, which are unaffected by the size or growth of the assets.
Does it matter that the pension assets are legally linked to the plan? It does not; the use of a US IRC 529 fund for college tuition expenses would not affect our scenario. The accounting convention of offsetting the cost of pensions with the return on plan assets is merely that — an accounting convention.
The distinction made here between the accounting cost and the true economic cost of the pension plan sounds like an argument over semantics, but it should guide the management of corporate pension funds. For example, suppose that a plan sponsor decides to eliminate its pension deficit. Lacking the necessary cash, the company seeks a higher pension fund return by upping the allocation to marketable equities and alternative investments.
But a broader view would suggest that the company not limit itself to the pension fund as a possible source of the needed additional return: Perhaps that return could be better sought outside the pension fund. The company has determined to take on more risk in the hope of generating returns that will erase its pension deficit. Where should it take that risk? Why not raise the risk profile of its overall business through a leveraged stock repurchase or by investing in projects that offer higher, though riskier, returns that can fund the pension deficit? Shouldn’t the company exploit its own particular franchise and expertise rather than compete in the capital markets, where it has no edge over other companies?
These considerations extend beyond deficit situations to suggest that a company should never take risks with its pension assets, particularly when promising business projects are available. Various rules — for example, the inaccessibility of surpluses and the Pension Benefit Guaranty Corporation premiums on deficits — make pension funds an inhospitable locale for risk taking. But the fundamental problem is that the danger of incurring financial distress limits a company’s risk capacity. Risk capacity should be reserved for projects in which the company has a competitive advantage. It should never be squandered on endeavors in which the company has no reasonable expectation of earning more than the market return available to all other investors.
“All other investors” notably include a company’s own shareholders. Shareholders invest in a company for its franchise value and business expertise. They do not look to the company to generate shareholder value by acting as or hiring investment managers for its pension funds; shareholders can engage their own investment managers if they wish. Rather, the company should take on risk only in its basic business while minimizing pension fund risk.
Pension fund risk management rests on understanding that the enemy is not asset risk per se but, rather, asset–liability mismatch. Asset returns that look satisfactory in isolation may be inadequate when falling interest rates send liabilities soaring. And asset losses are harmless if they are associated with rising interest rates that produce matching reductions in liabilities.
To minimize pension risk, a company should therefore invest its pension funds solely in high-quality fixed-income securities that match the projected cash flows or durations of the accrued pensions. Thus assured that the pension assets will track the liabilities, management can promote shareholder value by confining its attention — and its risk taking — to its basic business.
The guest editorial above was published in the September/October 2014 issue of the Financial Analysts Journal.
Please note that the content of this site should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute.
2 thoughts on “Question: How Does Investment Return Affect Pension Cost?”
I think the question is somewhat ambiguous, but more importantly, the answer provided is not correct. Let me explain.
First I believe the question to be ambiguous. It is ambigous because it does not make clear what is meant by “investment returns” and “Pension Cost”. To be clear: In what follows, by “investment return” I will mean the best guesestimate of anticipated return when a pension actuary is estimating the cost of a (given) pension plan. So, for example, currently the best estimate of return on (relatively) safe investments, say, AAA rated bonds of 10 year maturities is around 4 to 5%, whereas during 1980, a comparable return would be close to 8 to 9%.
Next, we have to define what is meant by “Pension Cost”: To make the discussion simple, let’s go with the example given in the answer, and assume we want to calculate the 4-year university cost of a newly born in 2014. We shall further assume that the newly born will go to university when he/she turns 19, in 2033.
“Pension Cost” can be defined in many ways, but, one actarially reasonable definition would be to say the (equal) monthly amount that (the parent) should set aside so that when the child in 2033 starts college will have sufficient fund to get four year college education.
Using our best estimate of college education inflation, we first estimate the amount of fund that the child will need in 2033 when he/she starts college. Let’s call this number $X. Clearly $X is not (at least directly) related to “investment rate of return”. For the sake of argument, let’s say the amount of college fund will need to be $350,000.
So the question becomes the following: Given interest rate r, what should be the monthly amount, M, that we should set aside in order to have $350,000 in 2033. The answer: if we assume r to be 4.00% (0.33% per month), the monthly outlay, M = $954.41, whereas if we assume r to be, say, 8.00%, the monthly outlay becomes =$594.21.
So I do not know why the person providing the answer has said that the cost will be the same.
Well-Guided Information on whether Investment can affect pension cost, and you have done a great job by explaining some alternative methods to minimize pension risk. Thanks for sharing this amazing post.