On The Difference Between Investing,
Saving, and Speculation Dr. Frank A. Sortino 2/03/04 The first part of this article will appear in the next issue of Pensions & Investments Magazine
In the last decade, pension fund sponsors have been awash with all or nothing theories of investing. * As the stock market was rising to its peak in the 1990s, we were presented with theories that provided a rationale for investing largely in stocks. Jeremy Siegel’s 1994 book titled, "Stocks for the Long Run," encouraged pension funds to invest substantial allocations in equities because they have a long-term liability and stocks have always beat bonds in the long run. * In his March 2000 commentary in Pensions & Investments Charles Munger, Warren Buffet’s associate, wrote commentary titled "Diversification is bunk.". He went so far as to say that putting all your money in one stock, like the Woodruff Foundation had with Coca-Cola Co., could be a wise strategy. * More recently, we have been presented with theories to support investing solely in bonds. Remember the risk premium argument that was popular in recent years? This analysis claimed stocks should be shunned by pension funds because the risk premium of equities over bonds was considered nonexistent. Well, guess what? The broad stock market is up more than 30% from the low in 2003 and the Nasdaq is up more than 70% from its low. Hello! *In 2001, John Ralfe, then CFO at Boots PLC in London, sold all the stocks in its pension fund and invested solely in long-term bonds. He now has a consulting firm that promotes this strategy. Peter Bernstein supports his idea, saying pension sponsors should "figure out a mix with the highest probability of being able to pay for the groceries." In my opinion, putting all your money in one asset category constitutes a breach of fiduciary responsibility. If you are not diversified across asset categories you are not investing, you are speculating. You are behaving as if you know bonds (or stocks) will produce better results than stocks (or bonds). You are bound to be caught out.
Almost all Possibly the most popular topic at conferences today is "risk management" or "risk budgeting." These are catchy phrases that appeal to the emotions of an investment community that is still suffering from the trauma of the worst bear market it has ever witnessed. Proponents sum up risk management strategy as: match the duration of the liabilities with the duration of fixed income instruments and if there is any money left over, invest with equity managers with a high information ratio. This can be justified if the duration matching strategy creates a risk-free asset. Unfortunately, there is no such thing as an asset that is free of investment risk. It may be free of interest rate risk for an intermediate period of time but that is not what is at stake.
Case for diversification Another camp includes Nobel economics prize laureates Harry Markowitz and Bill Sharpe. They believe that the way to reduce risk is through diversification. Furthermore, it is not just a matter of finding the minimum risk portfolio; it is finding the portfolio with the best risk-return trade off. I will argue for this camp. The global stock market decline since 2000 has illuminated some serious problems in the pension community affecting both public and private pension plans. All parties are calling for change. We need to make sure the changes correct the problems. We need to keep in mind that the pension fund is an investment problem not an accounting problem. Accountants and actuaries have a different focus than investment officers. They focus on meeting the bills today where investments are made with an eye to the future. The pension liability is a future value problem not a present value problem. Yes, Mr. Bernstein, we need to make sure we have grocery money for the next year or two, but we don’ t need to have the cash now for the groceries 20 years from now. There is a difference between saving and investing. One should invest for long term needs and save for short term needs: Use cash flow matching to insure the firm will not have to sell stocks in order to meet retirement payouts in the next two or three years, but invest for the long term needs.
Relevant risk No one would argue in favor of 401(k) participants putting all their money in bonds instead of a well diversified portfolio of stocks and bonds. How then can they make this claim for DB plans? That’s because 401(k) plans don’t have actuaries. Instead of accountants and actuaries arbitrarily picking an interest rate that discounts the future benefit payments to present value, they should solve for it. In other words, solve for the rate of return that must be earned on the assets and contributions in order to meet the future payouts. This is the minimal acceptable rate of return, or MAR, that separates good outcomes from bad outcomes. Returns above the MAR will reduce the cost of capital to the firm and thereby increase the value of the firm. Returns below the MAR incur risk to the firm that it will have to increase its contributions and risk to the Pension Benefit Guaranty Corp. that it may have to assume the responsibility for the firm’s mismanagement. The deviation of returns below the MAR is the relevant measure of risk, not interest-rate risk. Interest-rate risk focuses on what is going on out there in the financial markets, not what’s going on internally in the management of the firm’s assets and liabilities. The MAR is the return that must be earned on average over a long period of time. That means a return of 30% in one year does not justify the pension sponsor treating it as an addition to current earnings or labor demanding an increase in benefits. Also, high returns for a short period of time does not justify the government taxing the excess or forbidding any new contributions. Negative returns should be anticipated after a period of returns that are substantially above average. Therefore, the federal government and Financial Accounting Standards Board should rescind accounting rules and tax laws that discourage long range planning and penalize diversification. Over funded and under funded are misleading terms when compared to the present value of future liabilities instead of the MAR. Using MAR has some limitations. MAR is not known with certainty and it implies a forecast of the liabilities in order to obtain an estimate. I am willing to accept the actuaries’ estimate of the future value of the benefits. MAR also does not take into consideration the cash flows needed to make payments at any point in time. We operate in a world of uncertainty. The question is: Which approach is less flawed and offers the best tools for managing the assets and liabilities of the organization under conditions of uncertainty? The argument should be about how to diversify, not whether to diversify. ------------------------------------------------------------------------------ Part II The above was written for an article in Pensions & Investments. The following enlarges on the concepts previously mentioned.
Finance 101 Finance is not accounting. Accounting focuses on the here and now. What are the income and expenses this month, this quarter, this year. If income exceeds expenses they call it profit. CFO’s know they must earn their cost of capital (COC) to be profitable. If the firm’s COC is 8% and their accounting profits are 6% the value of the firm will decline. Black ink does not make a firm profitable. The corollary is to minimize the COC to increase the value of the firm. At one time the pension fund was treated just like any other corporate debt, but repeated abuses of beneficiaries rights led to the creation of the Employees Retirement Income Securities Act, which mandated the pension obligation be funded. Furthermore, fiduciary standards were imposed that required the funds to be managed for the sole benefit of the beneficiaries and for the exclusive purpose of providing benefits. Nevertheless, pension management does affect the value of the firm in this way: Profits in the pension plan lead to a surplus which leads to a lower MAR which leads to taking less downside risk which leads to lower contributions which leads to a lower COC which leads to an increase in the value of the firm. Profits in the pension fund due to management skills do not belong to labor. Employees are entitled to a salary and promised retirement income. The employee is not entitled to profits due to management skill. Corporations do not exist for the purpose of providing jobs or pension benefits. They exist to provide a reward to those who took the risk and those who managed the risk successfully, i.e., shareholders and management. A side benefit is that people are gainfully employed and governments get taxes to provide societal needs. The more profitable the firm the better off are all the parties, shareholders, management, employees, and government. Higher profits mean higher dividends and capital gains to shareholders, bonuses to managers, secure jobs to employees, and assured compliance with government regulations.
Basic Risk Management Strategies According to the Goldman Sachs articles, the core of the strategy lies in identifying three kinds of risk: interest rate risk, market risk, and active risk. These, they claim, are the only risks that matter. Hmmm, what about the risk of not earning the return necessary to fund your plan. Doesn’t it matter? Interest rate risk, they claim, is an uncompensated risk and should therefore be hedged away by matching durations. Isn’t this the old immunization strategy of the 70’s? Yes, but this time they do it with derivatives. Another variation on this theme is to match the cash outflows from the pension liabilities with the cash inflows from the assets, which may be bonds, collars, or other exotic instruments. This used to be called the dedicated bond portfolio strategy. Exhibit 1
If you believe there is such a thing as a risk free asset then you might want some linear combination of the risk free asset (A) in Exhibit 1 and the one risky asset (1) which supposedly would be on the efficient frontier. What risk would portfolio A be free of? It appears to be mismatch risk. Whether matching cash flows or durations one could possibly get a little over 5% for the next ten years. But most pension plans have to earn at least 8% forever in order to fund their plans. Therefore, in a downside risk framework the A portfolio would be very risky and there would be no leine A to 1. Furthermore, if one were interested in upside potential, Asset A has none. What we have here is a strategy that, if it works, will eliminate uncertainty. You know you will not earn enough to fund your plan unless you increase your contributions. The fundamental flaw as I see it is the focus on what is going on out there…in the market place with interest rates and the risk of being in the market. What do interest rate risk, market risk, and tracking error have to do with the risk of not earning the rate of return necessary to fund the plan? In my opinion…NOTHING! Therefore, risk management strategies are based on irrelevant measures of risk. Pension fund management is not an appropriate problem for the Capital Asset Pricing Model (CAPM). CAPM explains how all assets should be priced in equilibrium so that, on a risk-adjusted basis, all returns are equal. CAPM ignores liabilities! The pension fund is an asset-liability management problem! The operative question is: what rate of return must be earned on the assets in order to meet the liabilities in the future? I agree with John Ralfe on this point, the task is to manage the assets and liabilities in your own organization and not get sidetracked with beating the market. This is the position Dr. David Hopelain and I took in an article for Financial Executive magazine in August 1980. However, we disagree with his conclusion.
Exhibit 2
Exhibit 2 shows projected returns for ten years for three portfolios. Let’s suppose you must earn at least 8% in order to fund your plan within your cost constraints. Would you prefer portfolio A to either B or C? Well, one might say, "I know what I’m going to get for the next ten years with A and I don’t know what I’m going to get with B or C. Yes, you know with certainty you will fail to achieve the rate of return you need at minimum (MAR) in order to accomplish your goal. Since performance is measured ex post let’s suppose Exhibit 2 accurately portrays the pattern of returns. Portfolio A will be shown to have the worst performance if measured in terms of upside potential and downside risk. If the mean and standard deviation are used instead, A would no doubt rank higher than C. Does that make sense? Most of the volatility of C is above 8%. From a corporate officer’s point of view, if A and C are in the same business and everything but the pension fund is held constant, wouldn’t company C end up with a lower cost of capital and therefore higher profits, and therefore a higher value of their shares? I agree with the basic premise of these risk management strategies, that one should not manage the pension assets without considering the liabilities. It seems to me that the MAR provides a nice link between assets and liabilities. It is simply the rate of return that must be earned on the assets in order to meet the liabilities. Risk is measured as deviations below the MAR. Upside potential is measured as the difference between the MAR and returns above the MAR. This allows one to manage the assets and liabilities of the portfolio simultaneously instead of managing the liabilities and then managing the assets as a second step. The Role of Government The Dutch government will soon pass a law that will treat pension liabilities as though they were due today and officials are recommending pension funds should therefore invest in long term bonds. I believe the opposite is true. Interest rates are at a 45 year low due to a global recession. As economies recover interest rates will rise and bonds will decline. The Dutch government strategy will result in selling stocks that are still down substantially from their highs in order to buy bonds that will decline if interest rates rise. Treating the pension like a current liability is leading them to the wrong conclusion at the wrong time. It is the government’s job to establish laws to protect the rights of pension beneficiaries and the public interest against corporate practices that adversely affect those rights and interests. Therefore, the government should pass laws that protect beneficiaries from corporate practices that breach their fiduciary responsibility to beneficiaries. But they should refrain from making asset allocation decisions. Politics makes for poor investment policy. Fix the problem, don’t institutionalize it The global stock market melt down since 2000 has illuminated some serious problems in the pension community affecting public and private pension plans as well as government organizations. All parties are now calling for change. We need to make sure the changes correct the problems. The Financial Accounting Standards Board (FASB) has established rules that prevent pension investment officers from treating pension assets as a long term investment to offset a long term liability. FASB should not be setting the valuation rules as though the entire pension liability is due today. It’s not! Instead, they should require investment officers to inform all interested parties how they plan to manage the current assets in the fund and future contributions in order to meet the future liabilities. The way to do this is to require all pension sponsors to calculate their MAR and publish it in their annual report. Organizations with a high MAR should be required to increase their contributions to bring the MAR to a reasonable level within a reasonable period of time. I leave it to them to determine what is reasonable. Organizations with low MAR’s should be allowed to reduce their contributions. No one should be allowed to withdraw money from the pension fund for any reason, including mergers and acquisitions. This strategy would lead to reducing equity exposure after a substantial rise and increasing equity exposure after a substantial decline. Buy low, sell high. What a concept.
Home | Published Papers | Work in Progress
| Past Conferences | Future
Conferences |
