How will the largest and best-run corporations manage their pension funds in 2000 and beyond? That is what United Technologies Corp., Hartford, Conn., sought to learn when it undertook a study of "best practices" at dozens of leading U.S. companies.
The study, "Benchmarking Pension Asset Management for United Technologies Corp.," sought to compare pension investment practices among U.S. corporations in an effort to identify best practices. The sample of 47 companies comprised the corporate sponsors of the country's largest defined benefit funds, as well as a number of companies widely regarded as well managed. Through comprehensive questionnaires and follow-up interviews with several of the 34 firms that participated, the study revealed a number of important insights in the areas of investment policy, governance, staffing, compensation, investment management and performance evaluation.
Following are findings of general interest:
The clearest pattern of anticipated activity is expanding international common stock diversification. Virtually every participant endorses this theme. Recent statistics from consultant Greenwich Associates bear out the trend. Last year, foreign stock investments of the typical corporate fund were 10.1%, up from 9% two years earlier. Corporate funds project their foreign stock investments to rise to 12.2% by 1998.
A fairly strong indication exists that corporate funds will be "globalizing" fixed-income investments, expanding the mandates of domestic managers to invest in Eurodollar, Yankee and non-dollar bonds on a tactical basis. There is no indication that funds will establish and maintain significant explicit policy allocations to non-dollar bonds.
Two out of three corporations expect to remain ambivalent about foreign exchange exposure for pension fund investments, at least with respect to common stocks. Accounting for their ambivalence are: the vagaries of foreign exchange management; hedging costs; and a belief that currency effects will "wash" over time. Half of the respondents state foreign exchange exposure might even be desirable to the extent it hedges dollar weakness. A relative few believe concern over foreign exchange will lead plan sponsors increasingly to hedge it or retain or encourage managers to do so.
Another factor accounting for the ambivalence might be the still relatively small allocations these funds have to foreign stocks. As the percentage grows, expect to see greater concern with the potential impact of foreign exchange.
Seventy-five percent of the companies anticipate investing a growing percentage of assets in private-market investments - real estate, venture capital or other vehicles.
The overwhelming reason is to increase return, which I view as a positive development. Pension funds' generally unprofitable foray into venture capital and real estate investment in the 1980s occurred largely in the name of diversification. Investors in these less liquid, less efficient markets should realize their purpose must be "winning," and not merely "playing," and should be confident they have a comparative advantage.
Look for private market investment activity to occur primarily in the form of non-marketable corporate securities or partnership interests in those securities. Even future real estate investments are likely to occur largely in corporate form through public and private real estate investment trusts.
Tactical vs. strategic allocation? No consensus exists as to whether corporate funds will maintain relatively constant asset allocation policies through time or employ managers that tactically shift allocations opportunely. Interestingly, there seems to be no middle ground here: Chief investment officers are squarely in one camp or the other on the issue.
Structure and implementation
Reducing the cost of pension asset management is the dominant theme in the area of structure and implementation. Virtually every participant in the survey is making a "concerted effort" to reduce its costs.
The second related theme is reducing the number of investment managers. The two principal motivations are to make the investment program more manageable through simplification and to reduce costs. Again, the Greenwich data lend support: Last year the average number of managers employed by large corporate sponsors declined to 17.2 from 18.1.
Roughly half of the survey participants believe establishing a pension asset management subsidiary would be beneficial; about 20% disagree and 30% have no opinion. Several reasons are cited for a separate subsidiary: to facilitate compensation objectives for key people; for limited liability or other legal reasons; to facilitate internal management to lower costs; and to establish a higher profile for pension asset management.
The interesting issue is that CIOs' enthusiasm for establishing separate asset management subsidiaries often is not shared by top management. In more than one instance, we discovered management had quashed the idea of a separate subsidiary. Chief financial and chief executive officers' concerns range from controlling costs to preserving the managerial culture of the firm.
With respect to indexing, there is a wide range of practice. A minority of funds, about 30%, do no indexing, while most funds embrace the concept to some extent. (Passive investment percentages among the participants range from zero to about 50%.) As to whether passive investment management percentages will grow, there is no consensus: Approximately one-third each agree, disagree or have no opinion.
There is no groundswell of interest in establishing performance-based fees for managers, with as many CIOs opposed to them as favoring them and many others expressing no opinion. Some are opposed on philosophical grounds, others because of the added administrative complexity in an era in which corporate staffs are shrinking.
Three out of four participants identify a policy portfolio - i.e., using market index returns for various asset classes, weighted by normal investment policy percentages - as the most useful performance evaluation benchmark for a pension fund's total portfolio. Such a benchmark provides a valid comparison of returns incurred with those available in the capital markets.
Real return objectives as well as the actuarial interest assumption were overwhelmingly rejected as useful benchmarks. Although such returns might be achievable over the very long term and select shorter time periods, normal market fluctuations virtually guarantee such benchmarks will be operationally useless in evaluating the success of an investment program.
Three out of four participants believe best practices will include placing greater emphasis on designing performance benchmarks for investment managers. Most participants now use either a market index or a style-specific index as the primary benchmark in evaluating investments at the manager level.
Participants unanimously agree best practices dictate evaluating manager performance net of fees. This makes a great deal of sense. It also reveals a great irony in the management of corporate funds -while corporate sponsors universally believe in evaluating manager returns net of fees, only about half of the corporate pension funds consistently report their own results net of fees.
This is as striking for its implications as for its apparent inconsistency.
Best practices depends
on accurate measurement
Management processes must be measured without bias if the measurements are to be useful. When bias is introduced into the process, decision-makers are led to suboptimal solutions.
The total cost of operating large corporate pension funds, including the cost of investment management, custody, internal staff and consultants, is about 50 basis points on average. In mutual fund jargon, this would be a typical "expense ratio" of large corporate funds. And it is a large expense, eating up about 10% of what funds ordinarily would expect to earn in excess of the Treasury bill return.
In evaluating a fund's results, ignoring these costs leads to favoring high-cost strategies over low-cost strategies. This might explain a seeming anomaly: Plan sponsors want to reduce costs and the number of managers, and yet many appear reluctant to increase significantly the use of low-cost indexing. It also might explain in part the appeal of international investing and private-market investments, both of which are considerably more expensive than investment in public securities in the United States, and passive investment in particular.
Investment executives' interest in identifying "best practices" is laudable. Analyzing their own investment returns net of the cost of realizing them will surely speed the task.
Richard M. Ennis, CFA, is a principal with Ennis, Knupp & Associates, Chicago, which conducted the benchmarking study on behalf of United Technologies.