*spoiler alert this post includes references to accounting, which I know is boring. I tried to spice it up.*
I first started writing today’s blog post as an excuse. You see, chances are that if you emailed me or called me in the 10 days I haven’t gotten back to you. You might be worried that I fell into a well and Lassie is searching for help, and in a proverbial sense that may be true, but the real truth is that I am stuck working on FAS 157 valuation reports. This happens every year in January. I promise to resurface by Friday. And in no event later than next Monday.
While I was working away on these reports I got the idea that I should shed some light on how VCs value private companies. This isn’t rocket science but it might make interested reading. Or maybe I am just delusional from spending too many hours in spreadsheets. Back in 2006, venture capital firms had a very simple approach to their accounting. We used to hold everything at our investment cost unless one of two things happened: 1.) an outside investor came along and bought some shares at a new value. In that instance we would mark our holdings to the new price. 2.) The company substantially missed its budget. If that happened we either wrote the investment down or wrote it off altogether. This old system had flaws for sure, but it was workable and produced generally good valuations for over thirty years.
The problem is that the old rules I described above applied to all sorts of securities beyond the VC business. And some of those other securities happened to blow up the economy in 2008. For instance, many banks had billions of dollars of assets, like all of the CDO stuff I wrote about in my earlier post on “The Big Short”, for which there was no market. They didn’t write them down because they were convinced that this paper was worth a lot of money, and the auditors didn’t make them prove the underlying value. This type of decision making caused problems because lenders believed they had more money than they actually had, which meant they made more bad loans, which caused a death spiral. This is super simplified but you get the point about why valuation is important for banks.
In 2007 the regulators came to town. The Federal Accounting Standards Board passed new guidelines about the valuation of illiquid securities, which is documented in FAS 157. This gem of legislation has added a few hundred hours of work every year for venture capitalists like me. Instead of holding an investment at cost and calling it a day, we now hold it at cost and then provide four pages of charts, graphs, and analysis to explain why we are holding it at cost. Typical reports at Greycroft are six pages long for each portfolio company, and we do one for each of the forty-five investments. This year we will write approximately 270 pages of documentation, and the final product is a three ring binder that gets put on a shelf in our office. Our accountants review and approve this work prior to signing off on our financial statements, but aside from that brief inspection it is a vampire project – it never sees the light of day, it can’t be killed, and it sucks the life out of you. The contents of these reports include discounted cash flow models, multiple analysis, and pages of explanation about what accounting hierarchy (Level I, II, or III) applies to our businesses.
The FAS rules were not designed for the venture capital business, but it had the good intention of making valuations more accurate and transparent. I am concerned that what you get is actually more opaque and convoluted as a result of the new rules.
When I was in business school I was the teaching assistant for a class called Entrepreneurial Finance. The crux of the class is that you can use tools to value private companies – option value, comps, discounted cash flow – but at the end of the day you need to look at hundreds of companies before you acquire the pattern recognition to accurately price start-ups. And even then the most experienced investors get it wrong a good percentage of the time. In fact there is a case to be made that angel investors are getting this math wrong almost all of the time.
I have a lot of problems with FAS 157. One of my biggest issues is that FAS assumes that public company tools, like discounted cash flow and multiple-based pricing, work when valuing start-up companies. To put this in layman’s terms, the accountants are now asking a start-up company, which may not even have a product yet, to predict what their profits will look like five or ten years in the future, and then discount those profits back to the present. This method of valuation is meant for blue chip companies like IBM, not for five guys sitting in a loft in Tribeca. It has about as much relevance to my industry as the man in the moon.
Using comparable companies to derive valuations is equally inaccurate. We saw how well this strategy worked in the housing market – when you link your price to what other people pay you risk getting stuck in an asset bubble. In the venture capital business we have a unique challenge because there is very little pricing data for private companies, and public companies are imperfect comparables for too many reasons to discuss here.
The lesson from all of this is that valuation is more of an art than science. Venture capitalists spend their lives learning a very specific skill on how to value small companies and raw teams of people. A lot of it is based on a single factor – an idea of what the final outcome will look like if the company is successful. That outcome is a combination of market size, capital intensity, team, competitive dynamics in industry, pricing, and then macro factors like the IPO market and strategic fit with large acquirers. If you think you can distill that knowledge into a few charts you are mistaken.