Let’s spend five days reviewing the latest annual report (pdf) from the Investment Office. Greatest hits commentary on related topics include here, here, here and here. Today is Day 2.

The College (and Investment Office) provide little of the necessary information that we would need to evaluate their performance when it comes to the management of the endowment. The Record ought to write an article about this. Start by asking the Administration this question:

Q: The endowment dropped 1.5% last year. It is impossible to understand performance without reference to a benchmark. What does the College use as a benchmark and who calculates its performance? Can you provide us with the details behind this calculation?

The Investment Report refers to an “internal benchmark” and to a “Policy Portfolio Benchmark” (presumably the same thing) but has declined, in the past, to provide any details. That is ridiculous. There is no way for anyone to know if Chilton and her staff are doing a good job if there is not some (public) benchmark to compare their results with. Every mutual fund in the US is required by the SEC to provide performance relative to a well-defined benchmark. Consider this chart from the Annual Report:


This chart is designed to make us cheer: Yeah Williams! Hooray for Collette Chilton! We/she beat the “Policy Portfolio Benchmark” over the last 1, 3, 5 and 10 year horizons. Pay Collette and her team more money! They are geniuses!

But Williams does not provide the necessary details for us to calculate those benchmark returns ourselves. Consider how a peer school does it:

Amherst’s strategic policy benchmark is a weighted average return derived by applying the target strategic portfolio weights of each asset class to the performance of the respective asset class benchmark.

Just so. And because Amherst is a serious and honest organization (!), it provides all the necessary details, e.g.,


Investment professionals might quibble with some of the choices here. But there is no excuse for Williams not to provide the same sorts of data. Instead, we get prose like this:


We need the exact benchmark that William is using to measure the performance of, for example, its global long equity managers. MSCI World, MSCI ACWI, even MSCI EAFE plus the S&P 500, would all be reasonable choices. Nor is Amherst an outlier when it compares to transparency in reporting benchmark performance. Consider Yale:


Again, we can quibble. Is the Wilshire 5,000 really a good benchmark for domestic equities? But that is precisely the discussion we should be having. Williams has no good excuse for not being at least as transparent as peer colleges like Amherst and Yale when it comes to providing the details behind endowment performance calculations.

To be fair to Williams, not every school is as transparent as Amherst and Yale. Dartmouth, for example, does not provide the details (pdf) behind the returns of its Policy Benchmark portfolio. Perhaps our friends at Dartblog have some thoughts?

Print  •  Email