PhillyDeals: PSERS report makes not-so-careful comparison

Why don't they just hire Vanguard

?

The Pennsylvania State Employees' Retirement System thought some explaining was in order.

PSERS needed $1.4 billion from state and local property taxpayers last year, and it expects to need $2.7 billion next year, it told legislators in February in its yearly report.

That's after paying $552 million to hundreds of private investment firms - more than half the total for private equity, private debt, hedge, venture capital, commodity, and other investments you can't buy from a broker - to keep its assets from sliding farther below its liabilities.

Isn't there a better way?

"Arguments have been made that private equity is just an expensive form of investing in public equities and can be replicated by passively managed public equity indexes," PSERS concedes in an unusual one-page statement almost halfway through the 116-page report.

But actually, it really works, PSERS adds, promising to show "what the returns would have been if, instead of investing in private equity, the cash that flows into/out of private-equity investments were made into/out of a very low-cost mutual fund," the Vanguard Total Stock Market Index, for the period December 1998 through September 2013.

That 15-year graph shows a PSERS private-equity profit line that runs higher than the Vanguard fund's line, indicating higher profit from high-paid managers vs. cheap index-fund algorithms. Next to the chart, PSERS lists "10 Year, Net of Fee" results showing PSERS private equity returned 13.95 percent a year in those years, beating the S&P fund's 8.38 percent.

Ergo, it concludes, "PSERS has been able to generate in excess of $5 billion in incremental value vs. the passive, low-cost approach to index investing."

Now, forget that past returns are no guarantee of future performance. (Or that Vanguard funds have become so popular the Malvern company is now one-fifth of the mutual-fund industry.)

Consider instead that, also in the report, PSERS shows the system's private investments, in each category, have not yet returned the billions of dollars PSERS has invested in each category, let alone net profit.

How can it claim to be beating Vanguard and the S&P? PSERS spokeswoman Evelyn Tatkovski said Tuesday staffers would be too busy for the rest of the week to talk. So I called some experts.

It's a good idea to carefully compare private investments to Vanguard funds, so too bad that's not in the graph, Chris Geczy, academic director of the Wharton School's Jacob Levy Equity Management Center for Quantitative Research, told me:

Private-equity investments aren't committed quickly, like stock money, but over years. So full private-equity results would be dragged down by lower cash returns that don't seem to be reported here.

Private-equity is illiquid and tough to give away when the market is down. You've tied up your cash, and that has costs the graph doesn't show.

Separately, private-equity boosters often combine the actual profit they collect with managers' estimates of what their remaining investments might be worth some day, and call this hybrid the "internal rate of return."

But because unsold investments may also turn up dead, "internal rates of return don't mean anything," Geczy said. Given the academic evidence, "I don't much believe in the ability to beat the market" by private or public managers.

"Why would [PSERS] graph 15 years?" Because stock returns for that period are the lowest number they could come up with, asserted Daniel P. Wiener, editor of the Independent Adviser for Vanguard Investors newsletter. And why, he asks, put a 15-year graph next to 10-year data? Why compare a U.S. stock index to multinational private-equity results? And why just private equity, not other private investments?

Liz Headley, a CPA in Paoli who recently made a presentation at Stanford University on PSERS accounting, and who met last week with state senators from both parties to review her findings, also questioned PSERS's Vanguard comparison.

It's "spin or cherry-picking," she told me. "A more appropriate benchmark would be to break their private-equity portfolio up into comparable segments, then make comparisons with the appropriate Vanguard funds."