Valuations and Sources of Volatility: Part 2 – Option Pricing Models

This is the second post in our series discussing new sources of volatility in our valuations. In this post, we want to discuss option-pricing models, how their inclusion in our valuation process introduces new sources of volatility and the impact the terms of new rounds of financing have within these models, particularly liquidation preferences.

Let’s first discuss what are option pricing models (OPMs). OPMs use call option theory to derive the value of sets of classes of securities taking into account the financial rights and preferences of classes of securities such as liquidation preferences, redemption rights and dividends. This method treats common and preferred stock (the classes of stock that we most commonly invest in) as call options on the company’s enterprise value. It derives breakpoints based on liquidation preferences of the preferred stock and then calculates the values of those liquidation preferences and the company as a whole using Black-Scholes-Merton equations. The sum of these values yields the estimated enterprise value of the portfolio company. This method of derivation is often referred to as “backsolve” as it uses the price per share of the most recent round of financing to backsolve for the values of the other classes of outstanding securities of the company.

The OPM uses following inputs in its calculations:

  • Last Round Price per Share
  • Liquidation Preferences (including dividends and redemptions, if any)
  • Estimated Time to Exit
  • Estimated Volatility
  • Risk-Free Interest Rate
  • Outstanding Capitalization of the Company

This method contrasts with the common-stock equivalent (CSE) method. CSE derives the enterprise value of a company by multiplying the last price per share by the fully diluted capitalization of a company. While the CSE method was historically used to derive value, industry practice has shifted to the use of OPM in cases where other methods such as multiples to revenue or discounted cash flows are not appropriate.

So why is this shift important to understand? Simply put, OPM is highly influenced by the inputs listed above, particularly the estimates for time to exit and volatility and the structure of liquidation preferences. The estimate of time to exit is derived based on historical times to exit for our portfolio and factors related to each company on an individual basis. Estimates for volatility are derived from the volatilities of stock prices of publicly traded comparable companies on an individual portfolio company basis. Liquidation preferences are common in most venture capital-backed companies. They are generally negotiated to provide downside protection for the risk that a company is not successful. In some cases, it also provides for a way to increase the return on invested capital by having the initial capital returned followed by participation in the distribution of proceeds after the payout of the liquidation preferences. These liquidation preferences can be stacked, meaning the last money in gets paid out first, pari passu, meaning all investors participate at the same time regardless of who invested last, or a mixture of stacked and pari passu depending on each round of investment. Therefore, new rounds of financing at higher share prices may not result in increases in the value of our ownership in those companies depending on how these liquidation preferences are structured.

As of December 31, 2014, 16 of our portfolio companies have multiple classes of preferred stock outstanding with various sets of liquidation preferences. Within these companies, 14 have at least one set of preferences that are senior to another class of preferred stock. Within this subset of portfolio companies, the most recent round of financing of 10 companies included a senior liquidation preference. Additionally, in these 16 companies at least one class of preferred stock included a greater than 1x multiple preference in 5 of these companies.

Here is an illustration of what we mean. Let’s take a hypothetical portfolio company that raises three rounds of capital at progressively higher valuations. Let’s assume the capital is invested in preferred stock with a 1-time liquidation preference. Let’s also assume that the option pool and common stock start with combined ownership of 50 percent and get diluted over time. Let’s then calculate the enterprise value of the company after each round of financing under OPM if the liquidation preferences are stacked or are pari passu and compare those valuations with the valuation of the company on a CSE basis. The summary of these scenarios is shown in the table below:

As you can see, liquidation preferences yield very different outcomes for enterprise value under OPM depending on whether those liquidation preferences are stacked or pari passu. Now, let’s dig a little deeper and look at how the valuations of each class of stock change at the end of each round of financing under the same set of assumptions.

The chart above illustrates that even if successive rounds of financing are completed at increasing prices per share, earlier rounds of financing will not necessarily increase in value at that time depending on whether the liquidation preferences are stacked or pari passu.

It is important to understand that liquidation preferences generally do not survive if a company completes a public listing. In a public listing, all of a company’s preferred stock converts to common stock, as it did with Enumeral Biomedical, Solazyme, NeoPhotonics and other IPOs within our portfolio. OPM does not model these types of events. Additionally as companies mature, we may shift from using OPM methodologies to derive enterprise values to ones based on discounted cash flows or market comparables. We encourage those interested in learning more about our company and our portfolio companies to look at the announcements regarding our portfolio companies to understand the potential for these companies to generate returns on investment. We note that the ultimate return on any investment may be materially different than the fair value derived as of a particular date of valuation.