I'm all for transparency and mark to market. As Roger Ehrenberg has blogged about consistently, investors need to know what the underlying securities are worth inside banks, brokerage firms, insurance companies, hedge funds, private equity funds, and yes venture capital funds.
The FASB (which governs accounting standards in this country) recently issued a new rule called FAS157. I'm not going to get technical on you and the accountants in the community can weigh in via the comments if they feel the need for more rigor in this discussion.
In layman's terms, FAS157 says that you have to value your investments at 'fair market value'.
I've been dealing with valuing venture investments every quarter since I got into the VC business in the mid 80s. One of the big roles I had at my first VC job was the help manage the annual audit process and I'd prepare the schedules of investment for our firm's funds every quarter
Back then, we did it the old school way. We'd carry our investments at the 'lower of cost or market' and rarely write them up without a new financing event. Even with a new financing event, we'd often continue to carry the investments at cost if we felt that the markup to the next round price was 'shaky'. That could be because it was an internal round, a round led by strategic investors, or just a wildly inflated valuation that we were uncomfortable with.
On the other hand, we were quick to mark things down when the investments soured. We'd often mark an investment down to one dollar to signify that it was still unrealized but we felt it had no upside potential.
The net result of this approach is that the fund would always be undervalued until the investments were realized and the distributions had been made in full. And everyone was mostly fine with that approach because almost all LPs at that time held their fund investments to maturity and didn't care about interim valuations. They cared about cash out/cash in and not much else.
All that has changed and the venture business is a different place now. Interim valuations (and the associated returns) matter quite a bit when you go out to raise another fund. Also, the LPs are now selling their positions in the secondary market and this year we'll likely see more of that than ever before. So the carrying value of the investments is starting to matter.
And then comes these new rules from FASB. Our auditors, and I suspect every VC firm's auditors, want us to be rigorous about our valuation methodology and try, each quarter, but particulalry at year end, to come up with a calculation of 'fair value' for every company we've invested in and every underlying security we own in those companies.
Over the past week, as the audit season hits, I've had a number of conversations with other VCs about how we are valuing companies we share an interest in. And from those conversations, its clear to me that there are a lot of different approaches being taken out there.
So I thought it would be a good idea to blog about how we do it. First, we start with two crack members of our team, Andrew and Eric, who do most of this work. I hope they'll weigh in with comments and correct me where I've mis-stated something.
They have created a spreadsheet template that we use for each company. We then go out and find comparable private and, ideally, public companies for each of our investments. We like to have at least three 'comps' and often use four or five.
Then we calculate multiples of revenues, ebitda, and sometimes other measures of 'traction' like size of audience (UVs per month) for all the comps. We then take an average and get a baseline comp for each portfolio company. We back out excess cash less debt from the market prices when we calculate these comps.
When we use private companies and M&A transaction values that happened some time ago (we don't use comps that are more than 18-24 months old) we'll apply a market factor to them. For example, we'd take the Bebo sale price of $850mm and cut it roughly in half because the market has dropped in half since Bebo was sold. We do the same with comps for venture deals like the latest private round for WordPress.
We then collect the financial and other important data points for our companies for the current and next fiscal year and apply the relevant comp to that company and calculate an enterprise value. If the portfolio company has a significant amount of excess cash on its balance sheet we will add that minus any debt to the enterprise value to get a market value.
Then we assume we sold each and every company in an M&A transaction at that market price and do a liquidation analysis and determine the value in that scenario of each underlying security. And that's what we carry them at each quarter.
This is a tortuous process and produces some very non-intuitive results. For example, we are going to mark down the unrealized gain of the two companies in our portfolio that had the best fourth quarters of all of our companies. Seems strange that we be marking down the carrying values of our two best companies doesn't it? But the reason is that even though they both significantly beat their plan in Q4, the comps we use for both came way down in the fourth quarter.
So we take the adjustments and explain it to our investors in the reports and on the investors calls we do every quarter.
We also have the opposite problem that sometimes we are forced to write our investments up to levels we are not comfortable with. When their comps have big and probably temporary increases in stock price, our companies get written up. Which inevitably leads to a markdown in a future quarter.
The public market investors who are reading this are probably laughing right about now and thinking that it's about time the private equity and VC firms had to deal with vagaries of the market.
But I'm not sure this is all good news for the investors in VC funds. We are going to see a lot more volatility in fund valuations and the ways of the past where you had slow but steady increases in fund value are gone. VC interim returns are going to get more correlated with the market and we'll all end up spending countless hours explaining away certain numbers that are on the books but make no sense.
And, like has happened to me over the past couple weeks, VCs are going to have to talk more to our colleagues about how we are valuing our common investments so we can take the phone calls from our common LPs and explain why we've marked something up but our colleagues have marked it down.
There's a silver lining in all of this, including the IRS 409a pronouncement of a few years ago that has created a whole industry of private company valuation firms and, if anything, even lower stock option grant prices, and that is that we are starting to collect a huge data set of private company valuations over time.
This, combined with the efforts of a few brave souls to create secondary markets for private company stock is going to lead to more data, more transparency, and more liquidity without having to register and do an IPO or sell your company.
But right now, it all feels like major pain in the butt.