Happy New Year! This will be a shorter post than normal due to the holidays.
How do I evaluate my own investing performance?
DPI (the ratio of Distributions to Paid-In-Capital) = cash on cash returns from exits, and the only thing that really matters in the long run (but the long run is a long time, at least 7 years)
TVPI (the ratio of current Total Value to Paid-In-Capital) = DPI plus the estimated current value of investments without exits yet, which is the primary way to keep score along the way
IRR - the annualized rate of return of the TVPI based on when capital was committed - this is how to benchmark performance over time against other asset classes such as the stock market, bonds, etc.
Here’s a more detailed piece about these metrics, with some industry benchmarks.
For an angel investor, this is simple compared to a VC fund, which has to contend with gross vs net calculations (net of management fees, carry, etc.) for TVPI and IRR, and the timing of LP contributions.
It’s an oddity that hardly anybody posts online about how to calculate the unrealized portion of TVPI. Here are a few issues to think about:
The markup on equity after a portfolio company raises a new round is based on the share price appreciation — the new share price compared to the share price you paid. And this is determined by the ratio of the new pre-money valuation to the old post-money valuation. I’ve seen a fair number of people calculating against the new post-money valuation (or some other combination which leads to overstating the TVPI). You can go to any of the free equity calculators online and play around with some examples for five minutes and look at the share prices to see how this works. The way it works is not intuitive because companies issue new stock for new investors, which means my initial investment dilutes pre-existing investors, while subsequent investment dilutes me.
Some investors, as well as the tracker on AngelList, won’t show any markup on TVPI if a subsequent round is a SAFE at a higher cap. I see the logic in this because no new share price has been determined yet (see my last post for more on SAFEs), and the new cap would appear to only set an upper bound on the share price appreciation. However, I would also argue that this approach leads to understating TVPI. The issue I have with this is that even the markup between two priced rounds does not actually represent what you could currently sell the investment for on an illiquid secondary market. The error vs “fair market value” can be in either direction. I’ve seen plenty of shares of well known startups posted for sale on secondary markets (e.g. SharesPost, EquityZen) at share prices substantially less than the share price of the last round (or substantially more if a lot of time has passed and the company has continued to grow quickly). There’s going to be some error in the TVPI whether using SAFEs or priced rounds, and I prefer to document an estimate of the error rather than ignore a new SAFE round altogether. Again, in the end, cash on cash returns are all that matters.
An acquisition could lead to receiving stock in another company. If that company is private, there will be some fair market value used at the time of the acquisition, but that could change over time. If that company is public and I decide to hold the stock, do I allocate any value to DPI, or keep it all in TVPI?
There are a few angel investors who have published these metrics on their Medium posts, and many more who share them with the LPs in their syndicates, but it’s still a relatively opaque data set, which makes benchmarking rather challenging. Hopefully this will change in the future.
Thanks for the thoughtful post as always. I’ve been looking at this kind of data by deal source to try to get insight by source of deal flow. Curious if you have any insights on that point.