Multiples vs IRR

by Matt McCall

I was at a business school today helping judge several business plans. As group after group presented, I saw each make the same mistake as the previous. When they tried to justify the investment from the perspective of the VC, they kept telling us that this was a 40% IRR deal or a 25% IRR as if we had magical IRR thresholds.

The reality is that the venture world is all about multiples and the IRR's are the results. I don't know what the original legacy behind this was, but from a practical perspective, it is driven mostly by the fact that we live in a boolean world. Some is also based upon the high net worth legacy of our business. Originally, because pension law did not permit the large institutional investors in, our business was funded by family offices, endowments and foundations. Multi-generational families, while they want high IRR's, are really looking to double or triple their invested capital.

From a portfolio perspective, if we invest in 10 deals, 4 are tube shots, 2 we fight to get our money back on, 2-3 we get 2-5x on and the 10th deal drives the return (hopefully north of 10x). If we doubled our money in 1 year (100% IRR) but lost all our money on the next deal over 6 years, we aren't happy (net gain is $0). We don't care that we made 2x in 1 versus 3 years or lost all of our money over 6 years versus 4 years (this impacts IRR), because we earned 1x on the capital.

We often see complex financial models with discounted cash flows, hurdle rates and such. These are useless. I have never seen a set of financials in an early stage company that ever reflect what Darwin will allow to happen in reality. So, you start your modeling with unreliable numbers.

Secondly, what is the beta for an early stage biotech deal, a semi-conductor start-up, etc? Can you assess the risk associated with a given management team? How about a new market space?
Perhaps we are too lazy to try and figure this out, but after decades of effort, the only method that seems to work in the venture world is to target 10x on each early stage deal (3-5x on later stage plays). They all look like the next Microsoft, but eventually, the portfolio of these settles down to the profile above. In the early stage world, if you target, say a 40% IRR, through assuming a number of 5x wins in a compressed period of time, you will likely be out of the business. Your 5x wins, while possibly generating high IRR's, don't return enough multiple to pay for the 4 tube shots and 2 break-even deals. Your winners need to deliver 10x.

So, next time you are trying to convince a VC about the merits of your firm, show them how they can make 10x capital on a realistic exit scenario (not how to get a 40% IRR).