Being a securities analyst is a good job. It pays well, it is intellectually interesting, has a bit of variety, and allows you to work indoors in the winter time. It would be a great job if only we didn’t have to deal with our clients. One of the irritating things customers do, in years such as 2009, is decide to sell out at the bottom. So they miss the next rally, blame you, and fire you. What were the fools thinking?

I spent many years trying to keep people from mishandling their money. Now I am spending my declining years on investment committees, trying to keep investment consultants and money managers from mishandling an endowment’s money. In the typical investment committee meeting, all the people in the room are honest, well-meaning, and knowledgeable. The number of bad decisions, and the level of animosity so engendered, seems hard to explain. I now think, however, that some elementary Behavioral Finance theory helps with the explanation.

Most investment committees believe that their main job is establishing asset allocation rules. They select what percentage of the money they should invest into stocks, bonds, cash and alternatives. To simplify the problem for this discussion, let us reduce the asset classes to two—a stock market index fund and a bond market index fund. Now the job of the committee is to ignore its consultant, guess a percentage to put into equities, and go to lunch. You are a distinguished member of this investment committee managing the endowment of Embraceable U. The market has been pretty good for the last year, so a 70% equity allocation sounds fine, and you vote for it. For the next year, the stock market goes up, and everyone feels proud of their investment skill. But then the market takes one of its unpredicted downward lurches. The committee commands the presence of the consultant and furiously demands that all equities be sold. How come?

If the market is fairly quiet, a 70/30 allocation works fairly well, because any fixed allocation acts as an automatic rebalancing program. The policy forces the endowment to sell stocks when the market goes up a bit, and then buy them back again when the market goes back down. So, most of the time, everyone is happy. Happy until the market takes a steep dive, as it did in the fall of 2008. Then the consultant is summoned to the investment committee. She says “This is a very unusual event that no one could have possibly forecast. Now it is time to rebalance, so at least you will be investing your cash at some very good prices.” The response to that is savage and unexpected “We cannot afford to take more risk at this time.” The committee chairman continues his outburst,“We want to reduce our equity allocation, so sell the pitiful remnant of our portfolio. By the way, we are starting a search for a new consultant, because you are fired!” Why did this pleasant, urbane, experienced trustee turn into a fierce maniac?

One’s first thought is usually betrayal. When we set up our 70% equity allocation, the market was moving along in a general uptrend, and investment committee meetings were cordial and self-congratulatory. The committee was “summer soldiers” who enjoyed their popularity with the board of trustees as long as they showed money-making acumen. Then, a bad market comes along, and at the first sound of hostile gunfire, the old fools panic, drop their muskets, and run. When I was a portfolio manager, I cursed their perfidy. However, now that I am one of the panicky old fools, I have to justify our volatile behavior.

Endowment assets come in different sizes, following a power law distribution, with a few giants, a lot of middle-size ones, and a great many small endowments. Let’s concentrate on small endowments, because they are the ones that you are most likely to run into. Small endowments will usually have a consultant suggesting investment strategy, and a committee that follows the advice of the consultant, since the organization cannot afford a professional investment staff. The investment committee will be made up of responsible trustees who often have considerable experience in finance, but any investment committee has defects.

One committee problem is a very long decision time. It is hard to get any change voted in less than three meetings, and some take five meetings, so if the committee meets quarterly that means it takes about a year to get anything done.

A second defect is a lack of institutional memory. If committee members serve for three years, and there are six members, then there are only two members around long enough to remember the last disaster. This perpetual inexperience promotes the leadership role of the consultant.

Will the investment committee use leverage?

The acceptance of a leveraged balance sheet for EU is worth thinking about. Suppose the school had only a $20 million endowment. A nicely dressed CFA comes to the President’s office and suggests a deal; his bank will lend EU $80 million in return for $80 million in tax-exempt bonds. The school can then invest its $100 million in assets at a positive carry, receiving high yield on its $100 million investment but paying a lower yield on its tax-exempt obligations and profiting on the spread. Would this leveraging of the balance sheet be approved by the trustees? I believe that in most cases the board would reject such a strategy, on the grounds that a drop in the market would put the endow-ment assets below their debts, triggering defaults on the bonds. The school would be bankrupt, fighting desperately to find some other college to take it over and allow EU to keep its doors open. The trustees would be humiliated and subject to lawsuits. The chances of such an ugly outcome would cause many trustees to vote against this leverage strategy.

However, if the EU had a $100 million endowment and had the same pressing need for $80 million to rehabilitate its physical plant, in my experience the school will in fact sell an $80 million bond issue, leveraging up the balance sheet, and hope for the best. This is clearly a risky strategy, for a 25% drop in endowment assets would trigger default on the debt, but the trustees will be willing to leverage up rather than liquidate most of the endowment.

The endowment fund is an asset of Embraceable U, but in many cases there are also liabilities. Perhaps the building inspector rated Ivy Hall three years ago and declared the old dorm uninhabitable. The board agreed to sell $60 million in tax-exempt bonds, $40 million for Ivy Hall repairs and $20 million for updating classrooms and labs, God knows they needed it. The $100 million endowment assets, of course, are pledged as collateral on the bonds. So, if you look at the balance sheet, it is now highly leveraged. This is not a hypothetical example; there are many organizations where the debts are 100% of the endowment assets. If the market drops, Embraceable U has a real problem.

The investment committee will act as if they do not regard their investment portfolio as a unitary thing. Instead it is divided into conceptual layers, as suggested by the Behavioral Finance gurus. For instance, they might regard the first $60 million in the endowment as a conservative reserve that will be 50% in equities. (Remember that the institution has borrowed money for deferred maintenance, and if the endowment goes below $60 million, we default on a loan covenant.) The next layer of $40 million is investable, so an asset allocation of 100% in equities would allow the funds to grow. If the total endowment is $100 million, you would end up with a blended equity rate of 70% (graph 1). Just what the committee voted. Again, time passes. If the stock market goes up, there will be some pressure on the committee to increase equity weightings, because there is more money in the second zone where 100% equity is the desirable allocation. This is a pleasant decision to make, because the endowment has been making money, and everyone is quite pleased with themselves. There will be a push to increase the equity percentage to 80%, in a fit of euphoria. This occurs about three months before the market peaks, and will cause tears later.

The crisis erupts when the market goes down sharply, so there is very little money in the 100% equity layer, and the $60 million reserve layer is threatened. Oh, my! If the market continues to dive, the fund could collapse down into the first layer. Panic time! We have just been reminded that we will be in default. Our loan will be called. We have no way to repay. That is why we call in the consultant and sell our stocks.

Another conceptual way to think about the endowment problem is that the investment policy is not simply a parameter, but is in fact a variable. The desired allocation percentage changes if the market makes a dramatic move in either direction. If you are trying to guide the endowment investment committee it would be wise to keep this in mind. You can try to control the reallocation process in order to make sure that the fund does not threaten to drop to the level one crisis zone. That is easy advice for me to give and doggone hard advice for you to follow.

Another way to describe the shift in investment policy with market results is to acknowledge that investment policy has a positive beta.

If investment policy is a function of the size of an endowment, this might explain data that I always found mysterious. NACUBO data show that past performance of college endowments was clearly related to the size of the endowment. Big endowments had higher returns than little endowments (chart 2). In principle, little endowments can hire consultants and investment managers who ought to be just as smart as the consultants and investment managers that big endowments can hire. One could even hypothesize that small endowments could do very well because of their ability to invest in small but enticing op-portunities that would not make sense for large, clumsy organizations. But the data shows otherwise. The most reasonable interpretation of the data is that large endowments take more risk than small ones. The giants have favorable long-term results because risky portfolios have higher returns. But why should small endowments act in a risk averse way? If as shown above, the investment policy of an endowment gets more conservative if the assets are smaller, then behavioral Finance may explain why the rich endowments keep getting richer.

The NACUBO figures for the periods ending June 30, 2008, show a perfect ranking between return and fund size. This time period rewarded risk-taking and alternative investments. The data suggests an obvious strategy; if two schools each have $100 million endowments, then they can do much better by simply merging the endowments. Some did this, by moving their assets into Commonfund or a similar structure, but most committees stay independent and keep trying to get good results from a small fund.

The 2009 NACUBO results are not out yet, but will show a much changed picture, possibly too gory to display in this dignified journal. The same high-risk, illiquid alternative investments that helped returns up to June 30, 2008, did poorly in 2009 and have forced EU and its fellow schools to make sharp cuts in expenditures. Being an endowment administrator has been a stressful and unpleasant job for the last year.

Ralph Wanger, CFA, is Senior Advisor to Wanger Investment Management, Inc.