Investment Policy for Defined-Contribution Pension Scheme Members Close to Retirement

Abstract

This paper considers the investment decision facing a defined-contribution pension scheme investor close to retirement. Specifically, it investigates the lifestyle strategy whereby investors automatically switch investment policy in the years before retirement. The argument for switching is that investors may become more risk averse as they approach retirement and will wish to prevent unnecessary volatility of their fund. This argument is intuitively attractive. However there are counterarguments. During the switching period, investors will not be able to benefit from possible excess returns from equities; if investment markets are inefficient, investors may benefit from keeping investment discretion; and the nature of the risk in a defined-contribution plan may be more complex than many plan holders anticipate. It is important to define risk criteria before determining optimal investment policy. Movement into cash before retirement may stabilize the cash value of the fund but will put the investor at risk of falling interest rates and rising annuity prices; movement into conventional bonds before retirement may protect the investor from changes in interest rates but put the investor at risk from a change in inflation expectations if an index-linked annuity (priced from index-linked bond yields) is required at retirement; even movement into index-linked bonds puts the investor at risk from changes in real yields unless the duration of the index-linked bonds in the investment fund is equal to the duration of the required annuity. The methodology of the research involved finding the distribution of the accumulated cash fund, fixed nominal annuities and fixed real annuities from different investment strategies in a defined-contribution pension fund, using different investment strategies. All available postwar data is used. The paper finds that there is no evidence to suggest that a lifestyle investment strategy, which involves moving into cash, bonds, or index-linked bonds, is beneficial. These results are found when using either empirical data or stochastic modeling to find distributions of outcomes. A diversified portfolio of real assets (including international equities) seems to give the investor reasonable risk protection for two reasons. First, diversification protects the individual from falls in particular markets. Second, equities are themselves interest-rate (and, particularly in the U.K., real-interest) sensitive. This means that part of the explanation for falling equity prices lies in changes in real investment yields. If real interest rates change, equity values can change but there can be a partially offsetting change in index-linked annuity prices. Also, there can be hidden risks in investing in cash, conventional bonds, or index-linked bonds immediately before retirement. These investments can be subject to “real shocks” and can underperform salary growth significantly: specific examples of this phenomenon are given. Additionally, bond investment only eliminates risk if the duration of the bond fund is equal to the duration of the required annuity at retirement. Such matching by duration is likely to prove difficult in practice except for investors with very large funds. Specific examples of this problem are also given.