Yale’s Risk-Adjusted Returns Not So ‘Superior,’ Firm Argues

[truetype]

Yale’s Risk-Adjusted Returns Not So ‘Superior,’ Firm Argues

Ivy league endowments, including Yale University’s, are taking a lot more risk than their publicly reported numbers suggest, according to an independent study from research firm Markov Processes International.

MPI found that Yale and the other Ivy endowments displayed some skill at finding the best asset managers and they did outperform popular benchmarks.

Still, “we find that Yale’s superior performance as compared to other Ivies and a 60/40 proxy could be explained primarily by its much higher estimated risk,” wrote the authors of an MPI paper expected to be released later this week...

 

[Daal]

"MPI finds that the Sharpe Ratio, a measure of returns relative to the risk taken, of all the endowments are very close to each other as well as to a 60/40 portfolio."

I need to read the research in full but if thats what they used to measure risk adjusted returns, then the research sucks. Measures like the Sortino, MAR ratio and others are much better than the Sharpe, which punishes upside volatility

 

[truetype]

Those measures are all highly correlated in real life, except for oddball portolios such as short-vol or tail-hedge. And an institution such as Yale, with loads of illiquid holdings, can't be benchmarked accurately anyway except over very long time horizons.

 

[Daal]

I read the research, and it doesnt seem to be all that great They used 2003-2017 data to measure the Ivys performance. 2003 was the bottom for the bear market, there is only 1 bear market in their sample. So, they are talking about risk adjusted returns but there is not much risk in their data(other than 2008). Also, as mentioned, they use the Sharpe ratio which is not that great compared to the Sortino or the MAR ratio
But to make the matters worse, they did find that the Ivy funds produced better risk adjusted returns than a typical 60/40 benchmark

Not by a lot but its better. They claim
"When properly measured, such risks indicate that the alpha achieved is not high enough to result in noticeably different Sharpe ratios to a 60-40 portfolio."
I would call a 0.07 improvement from the Sharpe noticibly different, its an 11% jump
The Ivys also beat the 60/40 on actual return

So the Ivys are looking great right? Well, not so fast. In the paper they got a table which shows the Ivy returns going back to 1991

I extracted that (by eyeballing the chart) and put in my AA excel spreadsheet to compare its performance to a 60/40 portfolio. And its not pretty what comes out (these are all inflation adjusted returns):

Winning months are actually winning years (im using yearly data). There is some issues with that ivy league data as their fiscal year doesnt yend in December but rather in June but I dont think it matters a whole lot.
Ivys made more money but they did so by taking into way more risk, they lose quite heavily on the Sortino and MAR ratios. If I limit the data to the 2003-2017 period (which is not a great idea but just to compare apple to apples), things dont improve either. Ivys still take a beating on the MAR and Sharpe ratios (so my tests are not matching that ones from the paper)
Bottom line is that the Ivy League Endowments don't seem to be all that good at producing great risk adjusted returns, which ironically, that paper research didn't quite show to its readers