There Is an Incubator Bubble—And It Will Pop

Bubbles are scary when they fill one criteria - investment in the bottom of the pyramid is feeding the top, making the whole thing look more profitable than it is.

For example, in the tech bubble, the new companies were buying Sun servers, Oracle databases, and Yahoo! ads. This made Sun, Oracle, and Yahoo very profitable, which encouraged more tech investment. But that snake can't eat itself forever, and when investment slowed it all went bust.

Likewise, housing bubbles affect (effect?) so much of the economy they can't help but feed off themselves.

And in a classic Ponzi scheme, the mechanism is the same - the bottom feeds the top, so people think that there are outsize gains to be made. Those gains evaporate once new entrants slow.

Now, there's some new startups like Heroku who are "selling picks in a gold rush". But there's plenty of others like Groupon and AirBnB who are selling to the wider economy. The successful internet start-ups don't usually feed off new entrants, so I don't think a really scary bubble can form.

That said, if a bunch of random airheads (who are these 60+ new incubators? top tier VC? bottom tier VC? government? universities? there will be winners and losers here) think they can attract the same talent as Paul Graham, and vet business and technology plans as successfully, they may stand to lose their shirts.

Now, in the case of incubators, they report that there's a 3x increase 2 years. This may be healthy development filling a need or it may that there are a lot unnecessary incubators. You can't tell at this stage, since the optimum number of incubators determined by market conditions is hard to tell.

The more you think about the harder it seems to judge the success of incubators, but in the end it comes down to money earned: how much it invested - values of current holdings is one simple way of doing it, I think (although this doesn't take into account other criteria, such as attracting companies to a region, which may important for some incubators).

In the case of a company like WUFOO, YC & PG did a follow on round of ~$100k bringing YC's total equity stake to say 10% and with an exit of $35m YC made off with $3.5.

The point being the initial $15k investment is to gain access to great companies at the beginning so that investors can make follow on rounds in the best companies in the class. This is where the real money is made with incubators and how this differs from the privies bubble where VC's would invest $10m in an idea for a company with zero revenue. Looks like capital efficiency to me...

Investing in early stage is cheaper and riskier. Investing in later stages (when a company has traction/growth - VC/PE stages) becomes more expensive, but is less riskier. The incubator model has the advantage of managing some of that early stage risk that angel investors traditionally didn't control by having experienced entrepreneurs on-board/within a network and in a collegial environment.

Any successful model will attract competitors, so you now have more incubators. He's equating the rise in incubators as a sign of a bubble. It would only be a bubble if you had tons of incubators sitting on cash and not finding any place to invest it. His second mistake is equating failed incubators, that result from poor investment decisions/leadership, with some kind of market correction ("the bubble bursts").

You could make the argument that this creates wonderful incentives for the startup environment in general, thereby attracting much more talent to the space. But in large part, this talent is poached or drawn away from existing tech companies. Increasingly, you're even seeing talent bounce from one pre-exit startup (!) to another. In addition, there may come a point where the end-stage acquisition market dries up, because the price of acquisitions gets too high for positive payback/ROI on would-be-acquirers' balance sheets. Once that happens, the bubble officially pops.

The idea of trying the incubator model in other areas such as health care and clean tech is an interesting idea and I'm interested to see if it will have the same effect on start ups that are much more expensive to start.

After being in 3 different incubators (2 government run, and 1 private in the US) I've learned a great deal about what works and what doesn't. The government doesn't work. If the folks who run the incubator have no stake in the companies and no interest as to whether they'll eventually be successful other than bragging rights, they will work as little as possible to not get fired.

Now when the incubator is run as a start up itself, the whole incentive system changes for the better. They want to work just as hard as you to make sure everyone is successful. Go YC!

An accelerator that speeds up the the rate of 'new business' failure.

The upside is that resources can be pooled in to a pinpointed funnel of repeatable goals. The downside is the mechanics of the model is scalable, but the quality is not.

There are competitors in every market space, to think that this wouldn't happen to 'incubators' is insane.

This is not a bubble, its business.

Not to mention that what the economy needs are good performing companies, I tend to think incubators and accelerators provide this.

I think he's overlooking the element of disruption here. While it's true that certain markets will need to expand to fit these new companies, many of these startups are entering markets with the hopes of exploiting inefficiencies in more traditional companies, or simply doing things better.

My understanding is that incubators are relatively new. It is not really reasonable to say the lack of IPOs indicate if incubators are successful or not...I believe the first yc class was like 6 years ago. LinkedIn, for example, took 9 years before it IPO'd.

Seriously, why is this getting upvotes?