The California Public Employees’ Retirement System had unfortunate timing.
When markets crashed in March, CalPERS missed a payout of more than $1 billion after cutting its crash-hedging program as part of a cost-curtailment effort and because of a lack of understanding of how tail hedges work, according to sources familiar with the situation.
CalPERS was paying about 5 basis points, or 0.05 percent, for the externally managed part of a tail-risk insurance program that protected about $5 billion in assets. In October 2019, as investors worried about the inevitable end of the longest bull market ever, the pension plan decided to end the hedging initiative because it was too costly.
Universa Investments ran the largest portion of the mandate, while LongTail Alpha of Newport Beach, CA, managed another portion. CalPERS gave them the standard 90 days to unwind their positions.
Nassim Taleb, who wrote the popular book The Black Swan in 2007, advises Universa. That firm had disposed of CalPERS positions by January. LongTail’s hedge was in the unwinding process between January 1 and March 31 and may have generated a final distribution of around $150 million to $175 million. Universa and LongTail declined to comment.
CalPERS’ chief investment officer Ben Meng, who came aboard in January 2019, stands by the decision.
“We terminated explicit tail-risk hedging options strategies because of their high cost, lack of scalability, and the fact that there are better alternatives available to CalPERS,” Meng told Institutional Investor Thursday.
“At times like this, we need to strongly resist ‘resulting bias’ — looking at recent results and then using those results to judge the merits of a decision,” Meng cautioned. “We are a long-term investor. For the size and complexity of our portfolio, we need to think differently.”
Tail-risk hedging programs are similar to life insurance. With tail hedges, like any insurance policy, investors pay a small amount each year — a sunk cost — for a large potential payout if the event that is being insured against occurs. This was the crux of the problem at CalPERS, according to sources. There was a lack of understanding of that cost and thinking of it as a management fee of sorts. CalPERS “benchmarked” the tail-hedge program to the generic benchmark for the fund itself. It did not have a customized benchmark, so the 5 basis points just became another cost for CalPERS.
This isn’t the way tail hedges should be assessed, according to Universa, one of the former external managers. “Risk mitigation performance must of course be measured by its ‘portfolio effect’ — specifically, the impact it has on the compound annual growth rate (CAGR) of the entire portfolio whose risk it is trying to mitigate,” the firm told clients in a recent letter obtained by II.
For example, Universa recommends a hypothetical portfolio of a 3.33 percent allocation to its tail-risk product, coupled with a 96.67 percent position to the Standard & Poor’s 500 stock index, a proxy the firm uses for the systematic risk being mitigated.
In the month of March 2020, the hypothetical portfolio showed a compound annual growth rate of 0.4 percent. In March, the S&P 500 stock index lost 26.2 percent at its lowest point, and closed the month down 12.4 percent. For the year to date, Universa’s hypothetical portfolio had a CAGR of 16.2 percent, versus the S&P 500’s 4.5 percent. The model has produced a CAGR of 11.5 percent since March 2008 inception.
According to one source, CalPERS’ tail-hedge program “may have been swept up in a purge of many inefficient active manager programs. But the program was never meant to be under the mandate for these other programs. By design, it loses a little bit of money during good times with a huge benefit during a severe drawdown in equities.”
CalPERS’ decision to establish a crash-insurance policy goes back to 2016 under previous CIO Ted Eliopoulos. The pension system was trying to be cautious as it was still smarting from the global financial crisis, and started evaluating different tail risk managers that year. By August 2017, CalPERS hired Universa and LongTail Alpha. It set up a smaller in-house program to run related strategies, and planned to reassess the program every three to six months to see if it would meet expectations.
The initiative got a few tests, including the market selloff in February 2018, which it passed. CalPERS continued to increase its allocation. Then in the fall of 2019, with a new leader in place, CalPERS cut the program.