Peer Comparison: Why Endowments Are Doing It Wrong

Peer comparison has long been a staple of endowment culture – but just how funds are compared could be detrimental to both endowments and the universities they serve.

In a white paper published this month by the TIAA Institute, Cristian Ioan Tiu, a finance professor with the University at Buffalo, argued that grouping endowments by size makes for poor benchmarks of endowment performance. The practice, he said, disregards key differences in goals and objectives among similarly-sized endowments – and can even result in worse performance outcomes.

“A small liberal arts college with a large endowment is perceptively different from a large public university, yet their endowments may be similar in size,” Tiu wrote in the paper.

Beyond assets under management, endowment operations are impacted by factors such as the flow of donations and the degree to which the university budget relies on endowment payouts. These characteristics, rather than size alone, should determine how endowment investors set asset allocation strategies and risk budgets, Tiu wrote.

For example, while smaller endowments may “seek to emulate” the “superior performance” of large endowments like that of Yale, Tiu warned against the use of indexes tracking endowments from an “aspiring size” category.

“Larger endowments owe their higher returns partly to taking more risk,” he added. “A university whose budget relies a great deal on payouts from the endowment may find itself lacking resources if excessive risk taking leads to sharp declines in endowment value.”

To prove his point, Tiu constructed “organic” benchmarks based on characteristics such as budget contributions and university enrollment, using data from the National Association of College and University Business Officers and the National Center of Education Statistics.

By comparing the endowment performance of 898 universities between 2002 and 2014, he found that endowments which matched or outperformed their size benchmark but underperformed organic benchmarks saw lower growth in assets over time. Moreover, the universities associated with these funds suffered declines in enrollment and endowment-per-student ratios, and the endowments paid out less in proportion to their size.

“The construction of performance benchmarks that are based on university characteristics ... helps avoid implicitly assuming endowments follow ad hoc performance indices while focusing at the same time on the particularities of their institutions,” Tiu concluded. “An endowment following these benchmarks is more likely to align itself with its university’s goals.”