A Valuation Method for Private Equity

Anyone who has created valuation models knows that there are certain types of businesses that challenge traditional methods. One classic example is the private company, which has long posed problems for evaluators. But a new firm, FEV Analytics, has developed a proprietary method for valuing such entities and is directing its product at the private equity space. As this interview with cofounder Sheridan Porter indicates, FEV’s approach can inform the valuation of publicly traded firms, too.

CFA Institute: Tell us a little bit about FEV Analytics. What is the main idea behind the firm?

Sheridan Porter: FEV Analytics is an independent data science company located in Seattle, founded in 2013. We have developed the market’s first objective, accurate (R² of FEV to firm value is 0.813), and consistent measure of private firm performance and risk.

The main idea behind FEV Analytics is to use the FEV measure to improve systemic inefficiencies within the private equity industry.

Interesting. Tell us a bit more about how the sausage is made. For example, what is an FEV measure? What are some of the elements that go into yours?

FEV stands for fundamental economic value, and is the value of an asset’s economic engine. It is conceptually similar to an intrinsic value but differs in that it is produced using only an asset’s industry classification and financial fundamentals as inputs. The analysis involves no forward-looking inputs or assumptions, so it is 100% objective, repeatable, and automated.

Inside the sausage is a mathematical framework that draws from biological size and growth models to explain the multi-scale relationship between financial fundamentals and asset value by industry. Independent academic testing has confirmed that the FEV model is equally accurate for big, small, public, and private companies.

Why does FEV focus on the private equity space?

We decided to focus on private equity ahead of debt or public equities because PE is where the need is most urgent and where our products are uniquely impactful. The absence of a mathematically sound model and objective measure has resulted in inefficiencies that cost investors billions every year. FEV makes more frequent, more granular monitoring economically feasible. FEV’s model allows benchmarking to finally be put to work, made CFA Institute compliant contemporaneously, and synchronous with public markets. The right instrumentation is potentially transformative to the industry, but importantly to us (and why we got into this business), it can sharpen the focus on what actually matters — managing the growth of individual businesses.

What’s the weighted average time embedded in the data? Because you are using historical data, each company has differing amounts of historical data. So how far, on average, is that look back? How does this affect the valuations you develop?

Year 1 and Year 5 would be weighted exactly the same: zero. For each valuation, the model looks at one [corresponding] set of trailing 12 month financials. So today’s value of the hypothetical company would be computed from a set of trailing 12 month financials dated today.

How does your model deal with fluctuating premiums?

An excellent question. The FEV model has stripped out the premium, so theoretically FEV is separated from market price by market premium. Because the FEV model crosses the public/private boundary, we can take advantage of public company information to understand where the premium is at any given time. We can measure the premium of the market as a whole, or a subset — all the way down to a single asset. To increase the specificity of the premium to a given private asset, we create a custom public benchmark — typically comprising around 50 stocks — using the private asset’s FEV (size) and industry as the primary criteria. Then we measure the benchmark’s premium, synchronous with the date of the private asset’s financials, and map that information back to the private asset. So, the benchmark process that we use is instrumental in dealing with fluctuating premiums, applying the insight they bring. They also need to be created in an objective way, and we’ve invested in developing tools to optimize their robustness, preserve objectivity, and automate the entire process.

You quote an R² of 0.813. What is the confidence interval around that? Also, presumably that R² represents an underlying positive correlation, yes?

99% confidence interval:               0.81010 ≤ R² ≤ 0.81590

95% confidence interval:               0.81080 ≤ R² ≤ 0.81520

90% confidence interval:               0.81115 ≤ R² ≤ 0.81485

Where k=15, n = 90000, R² = 0.813 (for the independent validator’s data set)

Yes, the R² represents an underlying positive correlation.

A large part of your marketing emphasizes the bias-free nature of FEV Analytics. What does “bias free” mean to you?

To me, bias free has both a statistical meaning and a methodology meaning. FEV’s methodology excludes subjectivity. Data inputs are taken from financial statements, so there’s no setting or variable that requires judgment, for example, a discount rate or — even more nuanced — the purpose of the valuation. Statistically, bias free means that the residuals have mean approximately 0 and a symmetric random distribution. (This is true in log space). This means that error or noise isn’t compounded as the portfolio size increases. In fact, FEV becomes more accurate since the errors diversify away.

Importantly, FEV’s bias-free approach provides repeatability, which in turn makes comparison possible, even across traditional boundaries such as industry and vintage year.

Our last question, a two part one: Say a genie appears out of a magic lamp and grants FEV a wish about its relationship to the private equity space. For what does FEV wish? And tangent to that question, what evolution remains to occur with the quality of FEV’s offerings?

We’d wish for FEV to be the standard measure in the private equity space. Since measurement affects behavior, the industry would evolve to be a more efficient allocator of capital, more engaged in value creation, with less “dark space” between LPs and GPs. Because FEV can measure through the j-curve, we’d give the industry the tools to be strategic long-term investors — able to put capital to work on grand challenges as well as making returns.

Right now we’re seeing a lot of interest in measurement, particularly performance attribution, but the mechanism to tie that measurement into better decision making about future investments and risk management is not a mature process for most investors. We see our products evolving along with the industry to support a common framework that incorporates asset allocation and risk management smoothly across both liquid and illiquid assets, truly unifying the portfolio.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©iStockphoto.com/retrorocket