In the wake of the LinkedIn IPO, and some recent comments from Marc Andreesen, a lot of people seem pretty obsessed with whether there is or isn't a "tech bubble" - and, specifically, what that all means. I'm on record as calling the "bubble" an "expansion" (I was holding off on the b-word for a bit), but as the recent jobs data comes in and the macro-picture seems a bit murky heading into 2012, I decided to devote 60 minutes of walking to thinking through scenarios for how the tech bubble ends.
A couple of caveats:
1) I'm not an economist. True, I worked as a stock broker once upon a time, and held various licenses, and consider myself fairly knowledgable about public markets. But I'm by no means an "expert" in any sense of the word.
2) Most of what I'm fascinated with here is the effect on the startup ecosystem -- mostly because so many startups these days have something approaching zero experience or memory of what it's like when a bubble REALLY pops (2008/2009 didn't count, despite Sequoia's "Good Times R.I.P." press). When a bubble goes pop, there is no funding. You'd better be self-sustaining, or have a damn good plan for how you're going to manage your remaining capital to slog it out for 2-3 years.
In short, part of my exploring these scenarios is for my own benefit (more on that in a second), and part is because I hope that startups plan *now* for the eventuality of this (and take it into their fund raising process planning).
My benefit is simple: I run businesses in the tech space, and I like to have a working 3 year thesis. My thesis can change constantly, but I prefer to have some macro-framework that I think/work within (while validating or invalidating it against the outside world). Please note that others (some VC's I know) think *exactly* the opposite of this. Neither is good nor bad, it just is.
That said, some context:
1) My cursory look at bubbles (tech, housing, tulips, etc) would say that bubbles tend to last 4-5 years. Of course, the trick is figuring out when they start, and then watching markers along the way.
2) There's often a "bear trap" in bubble dynamics -- which is to say, a steep run up that precedes a sharp pullback, only to be followed by the asymptotic chart that marks the final days of the bubble. Knowing that the bear trap exists can be a valuable planning tool.
3) My brain works on analogies and cycles (history rhymes). So, when I look at the Netscape IPO in 1995 and the NASDAQ peak in 2000, that provides me with a crude (almost childishly crude) analog.
4) The dot-com bubble/bust of 1995-2000 involved the *public* (non-tech community) in a significant and substantial way, and that has it's own effects (which I'll talk about below).
Enough dithering around…the 3 scenarios for the end of the tech bubble (if you assume we're in one at all):
The End Early Bubble Scenario (Late 2012): This scenario assumes a bubble start date of roughly fall of 2008 (the financial crisis), and measures out four years (the minimum, cursory look, bubble timespan). I'd assume that this early end scenario would be caused by a larger economic downturn (double-dip recession) that drags down all of tech. Beyond that, I'd also say this the mildest of bubble popping scenarios. For two reasons:
A) The public isn't involved (largely) -- which is to say that not every 401k in the U.S. will be heavily invested in the 2012 equivalent of Webvan (though, hey, I could be wrong on this; see the recent GroupOn kerfluffle)
B) The bubble pop is largely regional in nature -- which means that the angels and startups in the Bay Area are the ones that get wiped off of the map; NYC a little bit, but not much; places like Boulder and Austin even less so. Reason: the bubble really exists there, and when it pops, that's where it'll hit the hardest.
So, in scenario 1 (the end early scenario), this whole thing never gets too out of hand, and when it pops the angel and startup communities feel it the hardest -- especially in the Bay Area (ie, companies receiving a crapload of cash right now via Andreesen-Horowitz, etc will somehow find a way to muddle through).
…which bring us to…
The End Mid Bubble Scenario (2013/2014): The scenario assumes one of two things -- either A) a later "start date" for the bubble (2009), or B) a longer bubble blowing period (5 years not 4). In that context, it then takes Scenario 1 and makes it bigger, badder and broader. The longer the time that elapses, the more that the geographic areas outside of the Bay Area get dragged into the bubble, and the more widespread the damage to the startup, angel, VC, and tech ecosystem upon popping. I think of this as the mid-line scenario.
The End Late Scenario (Fall, 2015): Assume that LinkedIn's IPO marks the beginning just like Netscape's did and you get to this scenario. In this one, the broader economy gets more time to right itself (this scenario actually assumes we avoid the double-dip in 2012), and as such, the public markets allow for more and more IPOs. Throw in a bear trap (a head fake slowdown in 2012), and assets begin seeking out the classes with the most bang for the buck - namely, technology. This scenarios assumes full-on public market participation similar to 1999-2000, and as such, also assumes a similar nuclear winter that was the 3-5 years following the NASDAQ peak (though one could argue some of that length was due to 9/11).
So, you've got your three timeframes and scenarios (2012, 2013/14, and 2015). What's a startup to do?
1. Plan your rounds in advance. I'd urge folks to remember that cash is king, and often "scaling quickly" can be overrated (if it assumes bang-bang fund raising cycles at ever-higher valuations). If you're a startup now, you're wise to at least have a plan for how you can either last 2-3 years or make it to profitability if the fund-raising spigot goes to ZERO starting in October 2012. Of course, if Scenario 1 looks not to be happening, then you switch plans for 2013/14, etc.
2. Understand the consequences of *other* people's actions. Even in private markets, the actions of other players have huge effects on your business. Billion dollar valuations skew things. IPOs warp perceptions. Reality bends. Stay micro in so far as you drive as hard and fast as possible to a solid customer base and profitability. I'm not saying don't raise. I'm saying have a "worst case scenario" plan on hand at all times that includes how you'd *radically* slash your burn, fire most of the people you've hired that are now friends, and get the ship right.
3. Pick your VC firm wisely. You want someone who has a sterling reputation for sticking with their entrepreneurs when the going gets tough. Their advice and experience will mean more to you than any dollar. Experience matters.
3. In the face of all of this doom and gloom, be an optimist. There is a smart way to contingency plan for the bubble popping, and a way to get through it. There's also a way to take advantage of it (cautiously, with history as a guide).
That was my walk this morning. That's the game plan I'm using as I plan out several businesses over the next few years. Will it stop me from chasing down opportunities? Nope. Does it scare me? Nope. Can I change my mind tomorrow if the data invalidates my thesis? Yep. But at least I've got a workable contingency plan. You should too.