Defined benefit pension math still works

The question of whether defined benefit pension plans are large contributors to the deteriorating financial condition of cities and counties and other public-sector sponsors of these plans, especially within the lower growth and low-interest-rate environment that characterizes the current economy, continues to be at the forefront of public policy discussion. Unrealistic promises of high returns from long-term pools of invested capital have been identified as a culprit in the widened spread between the dollars necessary to pay benefits and the assets available for that purpose.

In assessing the validity of these claims, let us analyze the investment environment and the likelihood of achieving the goal of a 7% to 8% return on a pool of assets invested in a diversified manner over an extended investment horizon.

We will evaluate four historic periods, three by decade plus the trailing 10 years ended June 30, 2013. We will then contrast those results with forward-looking assumptions as developed through a rigorous process by the beta research group at Segal Rogerscasey. Noting that each plan sponsor's asset allocation should be a byproduct of an intensive evaluation of their objectives and risk tolerances, we used this as a baseline approach to investing with a modest allocation to real estate and private equity along with market-like weightings to subclasses such as small-cap stocks. For the fixed-income component, where a U.S.-centric core approach is particularly problematic, we also applied a diversified approach for both credit quality and geography. Although one could certainly argue for a different methodology, we believe the allocations are consistent with reasonable practice and involve asset classes and subclasses that are accessible and investible by most defined benefit plan sponsors.

We then calculated the average annual return and standard deviation of this portfolio assuming annual rebalancing for the periods described. Given the relevance of real returns for sponsors of open defined benefit plans, we also included inflation and real returns comparisons.

Given that the foregoing represents history that might not be repeatable, it is appropriate to compare the results with forward-looking assumptions based upon the Segal Rogerscasey beta team's assessment of long-term expected returns, standard deviations, inflation rates and correlations developed for use by our clients in establishing asset allocations within the context of their investment objectives and risk tolerances. It is important to note these are long-term assumptions, contemplating 20-year forward-looking estimates, and actual returns will vary over time. There is the possibility of other so-called “lost decades,” but within the time horizon of most defined benefit plan investors, we expect normalization consistent with our expected returns and the long-term history of the capital markets as shown above. Applying these forward-looking 20-year estimates to the same allocations for the historical period's results in the following return and volatility:

There are several observations from this data:

The simple average return for the three decades, from Jan. 1, 1980, to Dec. 31, 2009, was 11.1% — an excellent result despite ending with the historic bear market of October 2007 to March 2009.

The most recent 10-year period — the one which, given the natural bias to recent events, should carry the most weight in our perception of return — exceeds the 7% to 8% goal, or investment assumption, that is so important to most defined benefit plans. It is also notable that the forward-looking expectations are in that same range, despite the low-interest-rate environment and more conservative expectations for equity returns in anticipation of a less-robust growth rate for the developed world in the near term.

It is remarkable how stable volatility, as measured by standard deviation, has been throughout the historic periods as well as on a projected basis. No doubt, the 1990s were golden years for investors — it was possible to earn double-digit returns on a diversified portfolio while experiencing abnormally low volatility.

This is a very basic analysis with assumptions designed to provide a minimalist approach to the results. There was no additional increment for active management even though while excess return is not always achievable, many of the associated asset classes or subclasses have been very conducive to successfully adding value over benchmarks that are somewhat simplistic in their construction. Conversely, we have subtracted no asset management fees or transactions costs for rebalancing, but point out that for most of the allocations those costs are very modest for institutional investors and presume inexpensive passive implementation. We have also not included several asset classes or investment types, such as infrastructure or hedge funds that can improve the return-vs.-risk profile and outcome, although we realize alternative investments play an important role in many defined benefit pension plan portfolios, including those of our clients. Nor have we sought to improve the results through any active rebalancing strategy that could serve as an overlay to moderate tail risk.

To conclude, on average, over the last 30-plus years, the investment side of the equation for defined benefit pension math has delivered on its promise. Applying reasonable expectations for a diversified global portfolio, it appears that this is also achievable going forward — despite current economic realities. While our analysis suggests investors should not necessarily expect the outsize results of the 1980s and 1990s, a goal of 7% to 8% appears reasonable and responsible.

The classic relationship describing the financial status of pension plans continues to hold true: Investment earnings + contributions – expenses = benefits. In addition, investment returns have been, on average and over a very long time, consistent with expectations. When these truths are combined with a strong conviction that calculations designed to assure that the equation remains in balance, there is one conclusion: Many plan sponsors, for policy, fiscal and regulatory reasons that might or might not be valid based upon the circumstances prevailing, elected to forgo contributions to their plans. For many, this was in the hope that investment returns of the 1980s and 1990s were sustainable; for others it might have been the stark reality of the demands upon the jurisdiction's or organization's ability to balance revenue and expenditure; and for still others a maximum taxation issue was the culprit. There are certainly funds that did not achieve the average investment results shown earlier, either due to suboptimal asset allocation and diversification or through poor experience in manager selection or investment structure. But, future returns in the 7% to 8% range should not be dismissed as unreasonable, given what we know about the behavior of the markets.

Since 2008, many have said, and many continue to say, that we have a “new normal” and future long-term returns will be lower. If they are referring to the double-digit returns of the 1980s and 1990s, we would wholeheartedly agree. Return expectations since the financial crisis have been consistently lowered by numerous plans, as they had been consistently increased during the boom times. While we would agree with the market observations that growth prospects, and returns, globally will not likely repeat those years, we continue to believe that, depending upon asset allocation and risk preferences, the pension math from the investment perspective will work going forward. Perhaps the new normal is really just back to normal.

Joseph A. LoCicero is New York-based president and CEO of The Segal Group; Timothy R. Barron is Darien, Conn.-based senior vice president and chief investment officer, Segal Rogerscasey, a Segal Group unit.