9/10 December 2015 saw the second edition of the Oxford Symposium on Risk Management in Private Equity, organised by the Saïd Business School’s Private Equity Institute in cooperation with my company, LDS Partners. Nearly 70 academics and industry practitioners participated this time, an increase compared to last year and close to the lecture theatre’s maximum capacity. The Symposium was held under Chatham House Rules, where participants are free to use the information received, but neither the identity nor the affiliation of the speakers, nor that of any other participant, may be revealed.
One of the Symposium’s highlights was new and still preliminary research on the risk return profile of co-investments – a topic that in recent years grabbed great interest and is debated heavily in the industry. Still the bulk of private equity capital is intermediated through firms raising capital for funds that then make investments into companies. However, this is expensive and bypassing the middle man by investing in companies directly, i.e., doing ‘solo-deals’, or along-side funds in the form of so-called co-investments is one of the strongest trends in the asset class. Investors believe this to be the key to sustainably high private equity returns, but their enthusiasm may be misguided.
Beyond the ‘2 and 20’ model?
A fund’s standard compensation structure, comprising 2% management fees plus 20% carried interest for performance – the fabled ‘2 and 20’ model – implies a cumulative investment cost of 5 to 7 percentage points per year. What has probably confused the debate is the emergence of private capital as viable alternative to public listings. Nowadays companies have no difficulty in finding patient capital outside the public markets and many CEOs prefer private equity ownership where they need to worry less about regulation and unwelcome publicity and can even earn more.
The flipside of this development is that private equity firms often do not need to provide the operational improvements and other value add that is inherent in their ‘classical’ model. Under such circumstances the traditional compensation is likely to be misapplied and ‘2 and 20’ often translates into high fees for disappointing performance. So in theory the cost savings potential when cutting out the fund as intermediaries should be significant. No wonder, therefore, that prominent investors in recent years have been moving away from commitments to funds and shifting towards direct investments. Another driver of this development are the increasing costs related to compliance with newly emerging regulation, such as AIFMD, which institutional investors hope to avoid by going direct as they are subject to a different regulatory regime, i.e., a kind of regulatory arbitrage.
Cut out the middleman?
Previous research put forward in 2012 by INSEAD’s Lily H. Fang and Harvard’s Victoria Ivashina and Josh Lerner did not give an encouraging picture for co-investments.  They found that direct investments have actually significantly outperformed standard fund investment benchmarks and notably the co-investments. However, they described a passive approach, where only few institutional investors were looked at that only co-invested in deals offered to them, apparently as ‘lemons’. Co-investments being offered by the funds to their investors reflect a specific approach that is not necessarily representative. This outcome, therefore, should not come as a surprise: co-investment strategies are mainly touted as a ‘fee reduction thing’, so it is logical that investors who solely focus on costs had put less focus on the investment’s fundamentals. Practitioners point out that it is important that a fund’s investor generates his co-investment deal-flow like any other direct investor, and communicates to the fund managers his specific interest and follows-up through regular calls and meetings.
The new results presented at the Symposium relate to a much larger investor base and reflect different aspects of co-investments, for instance, where funds were facing capital constraints or situations were buyout firms wanted to curry favour with their institutional investor base. Preliminary findings suggest that there be substantial variance in returns from co-investments across investor categories, with advisors seeming to be offered/pick good assets in buyout and venture capital. However, at the deal level, gross performance of co-investments was not found to be significantly different from that of non-co-investments. Nevertheless, when taking into account savings from fee and carried interest, at least in the case of buyouts on average co-investments look as if they outperform the corresponding fund.
But with typically not more than 10% of private equity resources dedicated to co-investments and with illiquid assets typically also a relatively minor share within an institutional investor’s strategic asset allocation, it is the question whether this can make a dent and is the panacea everybody believes it to be now. Overall for most institutional investors the whole activity will have no meaningful impact but there are consequences related to administrative overheads, audit and reporting for the co-investments.
Private equity’s law of gravity
In fact, there is a well-documented concave relation between fund size and performance that suggests a decreasing return to scale of fund size (see Figure 1). On one hand there is a minimum feasible fund size, below which overhead costs have a strong negative impact on returns. With increasing fund size average returns start to climb and achieve a peak at a certain ‘optimum’ size. Beyond that point, however, diseconomies of scale set in and average performance tends to fall with further increases in fund size – mainly because private equity to a high degree is a people’s business and investment professionals can only deal with a limited number of portfolio companies.
It is difficult to see why investors doing solo-deals should be exempt from this private equity’s apparent ‘law of gravity’. Institutional investors will be subject to the same economies and diseconomies of scale as fund managers and for larger allocations to private equity therefore the advantages of going direct will be elusive.
A comparable concave relationship also holds for portfolios of funds. Because of the size limitation for funds, increasing assets under management also lead to increases of the portfolio of funds’ diversification. When looking at portfolios of funds, overheads for staffing for investment management, research, back office, financial control, office space, IT, expenses for conducting due diligence, and travel expenses for monitoring funds in the portfolio, need to factored in. The size of a portfolio (too many managers or too much capital) imposes a limit on diversification benefits: there are only a few top-tier funds, and these most desirable funds have capped their capital under management. However, for larger amounts of capital it seems to be possible to add funds beyond their optimum without suffering from adverse diversification effects (see Figure 2).
Are co-investments subject to the same diseconomies of scale as the solo deals? I do not know of such an analysis but I suspect not: they should perform in line with the portfolios of funds as they are actively managed by different fund management teams.
In any case, the potential for cost savings may be less than widely expected. Direct investing vs. fund vs. funds-of-funds is rather a question of outsourcing the activity vs. doing this in-house. There are clear limits for investing directly and with increasing capital allocated to private equity investors have no choice but to involve fund managers. As a rule of thumb, the right approach will depend on the size of the investment program. For a small investment program with focus on local and later-stage companies doing solo-deals will be advantageous. A medium sized investment program will require increasing the commitments to funds. For a large program, the solo deals need to be more and more replaced by co-investments, with third-party fund managers supporting sourcing, monitoring and existing portfolio companies.
Go solo – fail – repeat
We should not overlook that the debate is not new and institutional investors had tried direct investing before, for instance in Europe through captive structures until the end of the 1990s. The experiences raise questions regarding the applicability of this approach within a wider investor community and suggest that there be significant challenges, e.g., how to create a governance structure that gives investment authority and decision-making power to the staff involved and how to manage the difficult interaction of the private equity focused entities with the rest of the organisation. As a result of these issues, the share of institutional investors investing directly had decreased significantly before this debate started to again gain momentum since 2008 – led mainly by the private equity arms of the Canadian pension system. However, for such a long-term oriented asset class ten years is still ‘too early to tell’ and for judging whether this really works, the jury is still out. My personal bet will be that eventually the enthusiasm will die down, only to resurface in the next private equity cycle.
Finally, I am convinced that there is much more mileage in doing co-investments but it is not the free lunch many hope for. It remains an interesting strategy but co-investing requires stronger allocations to private equity overall as well as a significant investment in staff, systems, processes and procedures.
 The Disintermediation of Financial Markets: Direct Investing in Private Equity. Lily H. Fang, Victoria Ivashina, Josh Lerner, September 3, 2014, Journal of Financial Economics (JFE)