I’ve been reading the book “The Black Swan” recently on the recommendation of my two partners.  I had heard about the book for years, but it never made it off my “to-read” list until now.

One of the concepts that the book discusses is the way we think of risk differently when we are generating profits vs. when we are minimizing losses.  The simple illustration goes something like this:

If someone gave you the offer of $100, no strings attached, vs. flipping a coin for the chance of winning $200, what would you choose?  Although both options are mathematically equivalent, most folks would choose the $100.

On the flip side, if things were reversed, and you could either lose $100 for sure, or have a 50% chance of losing $200 or nothing, what would you choose?  Most people in this situation tend to prefer the possibility of losing nothing, even though there is the 50% chance of a larger loss.  

This illustrates a simple point that we tend to be irrationally risk tolerant in protecting capital.  Social scientists call this loss aversion.

This has major implications for the venture business in the realm of follow-on investment decisions.  It’s a part of the business that doesn’t get much attention, but consider this:  I think it’s safe to say that well over 50% of a typical venture firm’s capital actually comes in after the initial investment round of financing for a company.  So even if a fund is supposed to be “early stage” focused, the reality is that the bulk of their capital is going into the follow-on investments in the B, C, D and later rounds. 

I didn’t realize this before I went into VC, but most VC firms are lifecycle investors, meaning that they have large reserves and expect to participate in most of the follow on rounds for companies that are doing reasonably well.  One would think that the follow-on investing decision for VC’s would be an easy one.  After all, no one has more information on a company than the existing investors and board directors.  Therefore, they should be very well equipped in figuring out which companies deserve follow-on capital, and which ones don’t.  Even though the follow-on capital is usually at a higher cost base than the earlier investments, this should be concentrated in the “best” companies, and should perform very well from a risk adjusted basis (even before considering the protection from being higher up in the preference stack).

Case closed right? Wrong.  There are a lot of reasons why follow-on financings might happen when they shouldn’t, causing VC’s  to “pour in good money after bad”.   

• Loss Aversion.  As discussed above, the uber-reason this happens is that one is irrationally risk tolerant when trying to preserve capital.  Or put another way, once you have a vested interest (time or money) into a company, you are willing to take irrational risks to protect your investment.   

• Delayed Gratification.  No investor wants to see a “zero” on their track record, and no investor wants to report “zeros” to LP’s.  This is true even though a small -100% return today might be much much better than a big -80% return in 5 years.  The pressure of needing to raise a future fund, looking good in front of your partners, trying to get promoted, trying to look like a clever guy in the twitterverse, etc leads to unnecessary risk-taking in follow-on financing decisions.  Even though almost every firm says they evaluate follow-on rounds like “new deals”,  I think this is actually far from reality.

• The Signaling Death Spiral.  Let’s take the hypothetical case of a company raising a series B that is doing ok, but not great.  The existing investors will often say they will support the company but have an outside lead price the round.  The new investor will ask the existing investors if they are “in” for their pro rata as a signal that it’s worth investing.  If an outside lead is willing to price and lead a round, it’s very very hard for the existing investor to say “you know what, I don’t believe in this.  I’m going to pass on this investment and risk that the whole deal blows up” (note that this is different than the follow-on dynamics of VC led seeds, where the investor will have a much smaller % of capital at risk and knows that they are buying 5 options to make 1 true investment.)  So in this scenario, a follow-on round gets done, and both parties are heavily influenced by the fact that the other is investing.  Puzzling no?

• Confirmation Bias.  This is the tendency for people to favor information that confirms their preconceptions regardless of whether that information is true or complete.  When layered in with Loss Aversion, it creates a deadly combination.  Because an investor is averse to losses, he/she is biased against any data that suggests that the initial investment decision was a mistake and will gravitate towards information that supports a follow-on investment.  

• The Bridge to Nowhere.  Even if a company is really struggling, the following logic is very appealing: wouldn’t you be willing to spend $2M to save the last $8M?  Because investors usually buy preferred stock, they get paid first and so they only need the company to sell for the value of the preferred stock to get their money back.  As a result, you often see struggling companies raise inside rounds under this logic (often crushing the employee’s equity in the process).  But many times, this round of “bridge” financing ends up being a bridge to nowhere.  

So, follow-on investing ends up being a much more complicated endeavor than it would first appear.  Clearly, there are some firms out there that have a great deal of discipline about follow-on financing and have been very successful.  But I think that this is a very very easy way to falter as an investor because it’s so natural to fall prey to these pitfalls.  As some super-angel funds increase in size, it will be interesting to see how they deal with these hurdles as well.  It’s easy to say that one will “pile in on their winners”, but the ability to do so will cut both ways.