Q&A: HEC's Oliver Gottschalg

Q: You have developed a performance assessment system - PERACS - to replace IRR. What is wrong with IRR?

A: IRR is polluted by two things. First, was the manager lucky or unlucky in catching a good macro environment? If you catch the upswing then the IRR will be very large because you were operating during a boom period and got the sector right. Second, if you have a very early, very successful exit it just biases the IRR. If a manager has a bunch of winners in the early 2000s, IRR might be high, but there may not be more winners in the late 2000s because with the first bunch the manager got lucky with some quick hits in a strong external market environment. This dynamic pollutes the persistence mechanism and therefore the likelihood of finding a winner in the next cycle. A cash in-cash out multiple isn't a good measure of performance persistence either. You might have two GPs with very different longevity in their deals because of different approaches to how they transform businesses. If both double their money but one takes three years and the other takes six years the multiple looks the same but it's an imperfect performance measure because it ignores the time factor.

Q: You are also critical of IRR being compared against public market index performance...

A: Assessing PE relative to the public markets is the right thought but you have to approach it with the appropriate methodology. What people usually do is take the long-term IRR on PE and compare it to long-term buy and hold on the stock market. That's overly simplistic. Not only do you fail to capture any difference in risk, but you also ignore the specific timing of investments in private equity. Say you invest in a PE fund, your capital is drawn in 2003 and you get double your money back in 2006. That's fine, but you have to see what the stock markets are doing over the same period of time. If you would otherwise have been long on the stock market, there is an opportunity cost to PE. You have to take into account not only dividends, but more importantly, dollar appreciation. If the stock index doubled over those three years, you haven't gained anything in relative terms.

Q: What other factors come into play?

A: First, the performance of a sector. If you want to be more exact, don't compare performance against the entire public market, but the specific sector in which the private equity fund is investing. It is the same when looking at the use of debt. If you have completed a buyout in the retail sector then you need to calculate the difference between the debt on your deal and the average debt of the retail sector index constituents. Then there is the impact of different economic climates. In a boom period everyone does well and so private equity adds little in terms of performance. However, during a downturn private equity generates alpha that protects values or allows moderate value appreciation while everything else is depreciating.

Q: How does the PERACS multiple address the public market comparison problem?

A: For the normal multiple, you calculate the stream of cash flows, and the multiple is distributions over contributions. We do the same thing but take into account the present value of the distributions against the MSCI Global Index as our standard proxy for LPs' opportunity cost. We discount the returns available on the index at specific times during the period under analysis, matching the timing of each cash flow. It gives you a present value of all contributions and distributions and through this you get a ratio in net present value terms of money out and money in.

Q: Instead of IRRs and multiples, you document the persistence of five other indicators - alpha, holding period, loss ratio and return dispersion. Performance persists when measured in alpha (i.e. the annual return achieved minus the impact of extenuating circumstances, such as the return that could have been earned on the stock market, timing and the use of leverage). Why is persistence in the other factors important?

A: The holding period is a manifestation of the GP's strategy. Some guys roll up their sleeves and do very long transformation deals; others spot relatively quick turnaround opportunities. If holding periods persist, it indicates consistency in the type of deals a manager does and therefore makes it easier to look at past performance and get an idea of future performance. With loss ratio and performance dispersion, again it is a question of discipline. Is a manager taking the venture capital approach and doing two deals that are home runs while another six are lost? That would be one profile. Alternatively, a GP could do eight deals and each one is 2x. A low loss ratio and minimal returns dispersion point to a stable GP.

Q: What about the impact of significant economic change? In emerging Asia the opportunity set might be very different from five years ago, and this would have implications for performance assessment

A: This is a big caveat. All the persistence research has been done in relatively stable markets in Europe and North America. You would expect persistence to be lower in Asia. You have to be very careful applying this logic. The objective is to match the indicator of past performance with an understanding of how value has been created - and that is the much more substantive component of what we do with PERACS. We measure alpha: you start with the PERACS multiple, which tells you by how the capital has grown after market swings, and then the PERACS alpha looks at this growth on an annual basis. But then we do lots of things that illustrate value creation so the LP has enough data points to say, ‘This is how they did it. Now let me match it with my understanding of what the world will look like in the future and we will have enough information to decide whether or not we want to back these guys.' It is all about reaching a qualitative understanding of the relative strengths and weaknesses of a GP.

Q: Aren't some LPs already doing this kind of analysis?

A: The more sophisticated LPs are already doing it, but perhaps only after 10 weeks of initial due diligence work. We allow people to look at these issues very early on in the process. This allows LPs to do a first screening, before starting due diligence, and decide whether or not a fund is worth their attention.' If an LP has scare resources and it looking at 1,000 funds or so, a first screening can be very valuable. At the same time, at the other end of the spectrum, a small family office might treat this analysis not only as a first screening but also as a pretty good final decision-making tool that replaces in house efforts.

Q: How do you deal with LPs that are accustomed to using IRR and don't want to change?

A: It is really challenging to go against the received wisdom of the entire industry. Many people have been using the potentially biased IRR measure for a long time and there is a large amount of inertia. Even if you convince them that there is a better approach, they still have board members or other constituents who want to continue doing things in a particular way. My business card should really say chief evangelist. The good news is that investors who bet on these measures have a real advantage in the marketplace. My back-testing demonstrates there might be 300-400 basis points in performance to be gained by switching to the smarter measures.

Q: How many GP clients do you have?

A: We have a good double-digit number of GP clients. Managers using PERACS have between them raised more than $65 billion over the past two years. The client base is split between North America and Europe and initially there was a bias towards the larger groups, simply because that is where we started with the marketing. If the biggest PE firms feel PERACS has value then we have a better chance of going against IRR. Now we are entering the middle market segments in North America and Europe, and we have recruited our first client specializing in emerging markets. We have yet to get our first Asian client but I don't think it will be long.

• Tweet  
• Facebook  
• LinkedIn  
• Google plus  
• Save this article  
• Send to  

• Topics
• Performance
• Advisory
• Asia
• Performance
• professional services