Strategies for a discontinuous future.

Wednesday, April 05, 2006

Is Google the New LTCM?

Ed Sim compares Google to LTCM.

Now, I think Google sucks as much as you do.

But, IMHO, this is a (very) tenuous comparison.

Here's why. LTCM made essentially a single, enormous, and, as it turned out, very wrong bet - that global interest rates would converge. Now, this convergence was, in fact, taking place.

But LTCM's models were, as Nassim Taleb puts it, fooled by randomness: they didn't account for extremely rare but extremely catastrophic events. LTCM was betting on essentially normal (lognormal, if you like) volatility - when the power law of volatility in real markets is, in fact, much greater.

The rare but catastrophic event in this case was basically Russia defaulting on it's notes, causing a currency crisis - which caused more or less a global liquidity crunch, especially in the markets LTCM was betting on. That is, exactly those spreads LTCM were betting would converge began to diverge, as everyone rushed out of risky assets like Russian notes, and into safer assets.

Hence, a huge blow up - massively exacerbated by the fact that LTCM was leveraged by about 3000%ish.

Now, where's the similarity? LTCM had arrogant rocket scientists like Google, apparently.

But that's about as far as it goes. Google isn't leveraged, it's not making betting the farm on a single concept backed up by pure theory, etc.

So while the LTCM story is a great one, I don't think it's at all comparable to Google. A better comparison for Google might be a more mundane player - GM, perhaps, or GE.

-- umair // 11:21 PM //


I went and read the piece to see how Google might bring down global finance and found something completely different. Maybe you read a different post, but the one I read was about corporate, technophiliac hubris more than anything else, and the bad effects that has on a company (rather than the wider market).

A quick overview of what went wrong at LTCM at the financial level is always welcome, but doesn't seem to have a lot to do with Ed Sim's post.
// Anonymous // 11:39 AM

Hey Anon,

My point was that beyond hubris, not much is common between the two companies.

Thx for the comment.
// umair // 11:55 AM

Thanks for the link but I believe you misread my post. It has nothing to do with financial leverage but more about culture, hubris, and arrogance.

// Ed // 12:41 PM

As an aside, it wasn't meant to be an anonymous comment. I put my name in initially, but I now realise it's not such a good idea to leave comments using Netnewswire. It seems that its lack of Javascript does odd things to Blogger's comment function.

// Chris Edwards // 2:16 PM

As long as both Google or anyone for that matter realize that

"All models are wrong while some modeles are useful"

one can stay away from making blunders.

LTCM blundered because it made beautiful nobel prize winning models
but which was rendered wrong by randomness.

Google has simillar kind of people who fantasize about making such models all day.

That is the commonality I think was pointed out. As Ed mentioned it is culture. Agreed to your point that they have not made bets which have huuuge downsides like LTCM did but they have culture which can lead to such bets.

// Rajan // 5:48 PM
Post a Comment



due diligence
a vc
tj's weblog
venture chronicles
the big picture
bill burnham
babak nivi
n-c thoughts
london gsb

chicago fed
dallas fed
ny fed
world bank
nouriel roubini


uhaque (dot) mba2003 (at) london (dot) edu


atom feed

technorati profile

blog archives