According to popular wisdom the days of unprofitable technology companies raising money from the public to pursue their growth dreams died in 2000, along with the Millennium Bug and the Macarena.
Yet one can’t help but notice how many of today’s thriving companies were unprofitable at the time of their IPOs. Not just ones from the dot.com years. Consider Compuware (1992), Concur (1998) and SuccessFactors (2007). All three companies were unprofitable at the time of IPO. All three are thriving today.
We decided to ask a simple question: Does IPO profitability matter? Specifically, do companies that are profitable at the time of IPO outperform the ones that are not? Rather than guess, why not look at the data? The result was unexpected:
We examined a large sample of successful technology companies, specifically the top 100 public software companies. We discovered some interesting things:
Unprofitable Companies Actually Did Better, At First
The unprofitable group outperformed the profitable group during the first two years after IPO. The difference was stark: 86% average return for the unprofitable group versus 39% for the profitable group at the end year 1. And 110% versus 89%, respectively, by the end year 2. Indeed you can see that the unprofitable group outperforms the profitable group for nearly the entire two year period after IPO.
IPO Profitability Mattered Greatly Over the Long Term
Even more interesting, however, is the fact that the profitable group clearly triumphed over the longer term (153% average return versus 36% by the end of year 3). This doesn’t simply mean that company profitability matters over the long term: that would be a trite finding and isn’t the subject of this study. This study focuses on whether IPO profitability matters over the long term. Here is what I mean: The “not profitable” group consists of companies that were unprofitable at the time of IPO. This makes the finding unexpected. At the time of an IPO, investors know full well whether a company is profitable. So if IPO profitability is an important determinant of future success, you wouldn’t expect a dramatic difference in the average returns of the two groups. If the effect was known, investors would adjust the price they are willing to pay for a company in either group accordingly, and the effect would be eliminated.
Are investors duped by the promises of unprofitable technology companies? Are they swayed by the grand growth stories of unprofitable IPOs only to be punished years later when reality strikes? Perhaps so.
What about Market Conditions?
One could argue that it isn’t fair to evaluate returns without factoring in the performance of the general market conditions each company experienced. For instance, if the technology industry as a whole returned 55% during the first year after a particular IPO (like it did for unprofitable Verisign) while it fell 40% in the year after another IPO (like it did for profitable OPNET), then perhaps it’s unfair to draw conclusions about relative company performance. For this reason, all of the returns in the study have been Nasdaq adjusted. This means that the returns shown above are those in excess of the return on the Nasdaq index. This is a way to control for general technology market conditions.
What about the Variance?
Do the average returns tell the whole story? Don’t we need to look at the spread of different outcomes in order to have an accurate understanding of the risk of each group? When we do this, the plot thickens. In the visualization above, hover over some points on each line. You can see the standard deviation of the returns of each group at each point in time. The standard deviation (a measure of spread) is large for both groups. For instance, after the first year, the average market-adjusted return of an unprofitable company was 86%, but the standard deviation was 250%. The average market-adjusted return of a profitable company was 39%, but the standard deviation was 97%. The profitable group had a much lower standard deviation for nearly the entire period in fact. This means the profitable group was performing poorer on average but was less risky.
{ 1 trackback }
{ 3 comments… read them below or add one }
Good post although your link in article to top-100 sample did not work. It makes hard to justify the numbers to make sure the theory is correct
Sorry about that! Link fixed now. Thanks for reading.
I’m not sure if you can make the conclusion “Unprofitable Companies Actually Did Better, At First” when you take include the insanity that was going on during the tech boom of the late 90s. At that time, ‘everything was going to be different on the web’ and you could IPO being uprofitable and get tremendous returns at first (until the crash came). If you remove the 90s and just look at the 80s and 00s, unprofitable IPOs were only similar to profitable IPOs for about 6 months, and then got blown away.