Last year, 15 prominent CIOs and investment industry leaders were brought together to discuss the most pressing problem facing allocators: how to fix the endowment model.
The group, which included investment chiefs from the Harvard, Columbia, and Cambridge University endowments, among others, spent three hours debating issues like: Will private equity continue to outperform? Should investors move away from rigid targets for asset allocation? What’s the best way to measure performance?
The takeaways from these discussions were then compiled into a new white paper published by True North Institute, an independent investment think tank started last year by Partners Capital founder Stan Miranda. Miranda told Institutional Investor that his goal was to get a group of thoughtful investors in a room together to discuss “the biggest elephants in the investment boardroom,” starting with: Should we challenge the endowment model?
“That’s what we all do for a living, so yes, let’s make sure we’re always challenging it,” Miranda said.
The new research follows a previous white paper published by True North Institute, or TNI, in May of last year which found that the alpha produced by elite endowments had all but disappeared in recent years. When comparing the performance of Ivy League and similar endowments against a multi-asset benchmark, TNI determined that the cohort delivered alpha of 1.2 percent over the ten-year period ending in June 2023. That figure then shrunk to 0.7 percent over 5 years and 0.1 percent over three years.
“In the last five years, the endowment model really hasn’t delivered better than the benchmark,” Miranda said. He noted that most of the outperformance has derived from large allocations to private equity and venture capital — a big concern for allocators, given the industry consensus that these asset classes will deliver lower returns going forward.
In the new paper, the institute examines various alternatives to the current endowment model, including the Canadian model, the McKinsey Investment Office’s portable alpha model, and the total portfolio approach now employed by allocators including Future Fund and GIC.
The Canadian model — which emphasizes direct investing by in-house talent — has been held up as a world-class model of pension investing. But according to TNI’s analysis of the last 10 years, “the Canadians would have performed better if they had outsourced to third party PE and VC firms and paid the fees, perhaps with a large amount of fee free co-investments.”
The actual best-in-class model, according to TNI? MIO’s portable alpha approach, which uses derivatives to “achieve high returns with relatively low market exposure,” according to the paper, with positive alpha almost every year since 1999, the lone exception being 2008.
“The McKinsey Investment Office may well be the most successful institutional investor in the history of institutional investing, as properly measured by how much beta they’re taking,” Miranda said. “If the measure is just consistent, high levels of alpha, MIO wins hands down.” (McKinsey is considering selling MIO, which has long faced criticism over conflicts of interest with McKinsey’s consulting business.)
The downsides: “It would take at least 10 years to build and scale,” according to the paper.
“It is also highly complex and difficult for lay members of the investment committee or trustee board to understand, especially given the leverage and heavy use of derivatives. Execution risk is also very high.”
Ultimately, TNI’s CIO forum found the total portfolio approach to be the most intriguing — at least as a modification on the current endowment model.
As Miranda explained, one of the criticisms of the endowment model is that it’s too structured around asset allocation. “It’s sacrificing the quality of managers by optimizing within each asset class,” he said. “To optimize within each one of those, you need a certain amount of managers. So you have too many managers.”
TPA solves for this by doing away with traditional asset class buckets and focusing instead on the underlying market risk factors, such as equity, credit, inflation, and rates. This way, manager selection drives asset allocation, instead of the other way around.
While the CIOs and other investment pros convened by Miranda disagreed about whether they could or should eliminate strategic asset allocation targets, most felt that the endowment model could benefit from more flexibility around the targets.
Ultimately, the forum concluded that “the endowment model is still robust and compelling for many truly long-term investors, but not all,” according to the paper. “Most, but not all, CIOs in the room felt PE and VC would continue to deliver outperformance vs. public equities, but not nearly as much as in the past.”
