You don't need a (tech) crunch to kill 50-80% of startups....

....they do it even in good markets.

Responding to a post on the top of Techmeme, where Jason Calacanis writes a long and winding article on Henry Blodget's (yes, that one) Silicon Alley Insider about company failure - the key hypothesis being:

It's my believe that the economic downturn will be much worse than it is today, and that 50-80% of the venture-backed startups currently operating will shut down or go on life-support (i.e. 3-4 folks working on them) within the next 18 months.

It kinda encapsulates the worst of Web 2.0 thinking for me, in that it is long on fluff and short on facts around the central hypothesis, which, while very plausible, is provably misleading with a bit of analysis. Not because there is no crisis, (nor that it won't have an impact) but because small companies fail in high proportions in any market - typically 2/3rds in the first 2-3 years* - so the 50-80% was a given anyway.

What is true however is that businesses being set up on FreeConomic principles (ie using other people's money) can only run as long as those others' pockets stay lined. So if they cannot mature into real business models (or sell themselves to Google etc) they will die. And there is a shortening of that runway in tighter times. (ie they will fail sooner). It is also true that there will probably be less money for me-too's now, as there is just less appetite for them.

The essay actually also has a lot of good sense at a tactical level, thus making it's central tenet very believable because much of the rest of it is good stuff for startup survival.

Because it is promoted by one of the Web 2.0 grandees (and don't get me wrong, Mr Calacanis is a very capable person, and is probably more skilled than lucky in my opinion - but even very smart people sometimes say daft things) it gets more uncritical acceptance than it otherwise would.

Except here of course

Update - the article has been pulled from SAI, sadly along with all the comments - c'mon guys, thats pretty poor form! And it wasn't that bad!

Update to the Update - its back on Jason Calacanis' blog - apart from the 50-80 stuff, the rest is tactical and rather more useful.

*I've never seen formal VC stats (wonder why ) but friends of mine in the industry used to tell me its 1/3rd die, 1/3rd live, (of which 1/10th make a decent return) and 1/3rd are "zombies" - living dead - that can stagger on for ages until either sold or put down. Hence the 2/3rd fail estimate