Quants have affirmed what most allocators hold to be true: Investing in private equity funds is a skill unto itself.

Limited partners’ (LPs) returns tend to persist beyond what random distribution would indicate, an analysis of 12,043 investments by 630 LPs has found. One standard deviation in allocator skill translated to roughly 3 percentage points in added annual internal rate of return (IRR), according to the working paper, published in August by the Charles A. Dice Center for Research in Financial Economics at Ohio State University’s Fisher College of Business.

The four authors, who derived their 15-year sample (1991–2006) from data providers S&P Capital IQ and Thomson Economics’ VentureXpert, found it consistent with research firm Preqin’s industrywide performance figures.

Using a model designed to measure general partner (GP) skill, the researchers controlled for investments’ vintage year, strategy (buyout versus venture capital), and other factors to isolate LPs’ fund and manager selection ability.

Dartmouth College’s endowment added more value through skill than any other institution analyzed, followed by the William and Flora Hewlett Foundation. The Louisiana State Employees’ Retirement System, perhaps surprisingly, came in first among pension funds, handily topping the Canadian and Californian giants.

The greatest selection-based losses all came from U.S. public funds: the Teachers’ Retirement System of the State of Illinois, the New Hampshire Retirement System, and the Ohio Police and Fire Pension Fund. The New Jersey–based Meadows Foundation and the U.K.’s Wellcome Trust placed last among nonprofits.

"These estimates are relatively noisy," the authors warned, noting an average 2.5 percent standard of error. "It does appear, however, that the LPs we identify as being in the top [skill] group do have noticeably better performance than those in the bottom group."

The findings reflected investment performance for funds’ lifespan; they didn’t account for performance spreads on stakes bought or sold midway. Likewise, portfolio holdings were only as complete as reported in the two databases.

Individual LPs’ rankings ought to be taken with caution, but what’s clear is that "the ability of LPs to pick GPs is not random, and better LPs outperform less talented LPs," the researchers wrote.

Therefore, they concluded, "it makes sense for institutional investors to bid to acquire the best investment officers, and that high-quality investment officers can more than earn their relatively high salaries."

Even CIOs who would quibble with their skill score probably agree.

Leanna Orr is Global Content Director of Investor Intelligence Network (IIN), Institutional Investor’s private community for asset owners.