Study: co-invests likely to underperform

Even reasonably-sized co-investment portfolios probably won't make any money, according to Altius.

LPs may be increasingly keen on co-investment opportunities, but the results are unlikely to live up to expectations, fresh research claims.

Co-investment portfolios – even if they are reasonably sized – have a substantial risk of generating poor or even negative returns, according to a study by advisory firm Altius Associates.

Using a sample of 886 realised US buyout and growth investments made between 1979 and 2010, the study suggested there was a substantial probability that a co-investment portfolio consisting of 10 assets would generate an IRR below 0 percent.

Furthermore, even with a 20-company co-investment portfolio, it is still possible to lose significant capital as measured by either IRR or money multiples, according to Altius.

The study also found that a 10-company co-invest portfolio based on growth funds did not generate an attractive return profile under any circumstances, with an average IRR of-8.1 percent. On the other hand, a 10-company buyout fund co-investment portfolio produced an average IRR of 14.2 percent.

On the plus side, these growth deals had a substantially higher maximum multiple of 14.2x, compared to the a maximum of 8.7x for buyout deals. However, the distribution of returns indicated that there was a high probability of poor returns, even across a full portfolio, Altius warned.

Many institutional investors are keen on establishing or expanding co-investment programmes for a number of reasons including reducing overall management fees, capturing a greater share ofl upside by not paying carried interest, mitigating the J-curve, deploying capital more quickly and taking more control of their investments.

But as well as these benefits, there are also a number of risks, the study claimed. Adverse selection may occur when funds offer co-investment in deals that exceed fund capacity; this could put the GP in a market segment for which past returns are not representative and are more uncertain.

Equally, as LPs usually only invest in a subset of a GP’s portfolios, co-investing may lead to an over-concentration of risk in a small number of companies, the study claimed.

Furthermore, GPs may have an incentive to keep the highest expected return investments wholly within the fund structure, as it increases their personal benefit by increasing the return that is subject to carried interest, Altius said.

“I can completely understand the argument for reducing fees,” William Charlton, head of Americas Investment and author of the research paper, told Private Equity International. “But if funds that are not in high demand offer co-investments, you may be picking from the lower end of the distribution and it may hurt your returns ultimately. You don’t want to trade fees for performance – that’s the key.”

He admitted however that it is difficult for GPs to know in advance which deals in the fund will perform the best. “On some cases, they may have a good indication, but there may be surprises in the portfolio.”

LPs can also reduce these risks by only co-investing with existing managers, he said. “If I think [the GP] treated me poorly in the co-investment, I have retribution in the fact that I may not participate in their next fund.”

The Altius study echoes a recent study by Harvard University which also concluded that co-investments underperform fund investments. Poor performance appeared to be driven by selection as institutional investors could only co-invest in deals that were available to them, the study concluded.

The Altius research looked at realised transactions, Charlton pointed out. “This would have been an ideal portfolio, where LPs would have access to every deal. But [our results] show that even if they have access to all the deals, it’s still a risky portfolio.”

“Personally, these findings surprise me,” Paul Newsome, executive director, head of investment selection & monitoring at Unigestion, told Private Equity International in response to the Altius study. “GPs tend to offer co-investments based on a desire to keep a certain diversification within their funds and to reward loyal LPs, not to sell down their riskier investments.”

Most of the GPs Unigestion works with have great track-records on the co-investment side, he insisted. “Indeed, for some of these GPs, the co-investments have performed in aggregate better than their fund investment track record.”

Of course, a co-investment fund is riskier than a well-diversified fund investment program, Newsome admitted. “However, that risk can be managed through a measured diversification across different GPs, different geographies and different sectors.”