Hacker News Reader

Comment by FlimGrimoire | Hacker News Reader

  Metacomment: geeks who believe they have outmathed a financial firm should ask themselves "Are financial firms likely to be bad at math?" and "Are financial firms incapable of hiring their own geeks?"

Ah yes, because as 2000, 2008, and 2019 have shown us, financial firms are infallible when it comes to maths and economics.

View on HN

2000/2008 didn't happen because someone did basic accounting math incorrectly.

I think you could argue that the rating agencies did the math incorrectly, at least in 2008.

They did the math correctly, it was just different math than what they were supposed to be using.

The incorrect ratings were intentional and designed to look like they were correct so that no one would find out. It's easier to make sure no one finds out the number is wrong if you know what the correct number is and how to tweak factors here and there to influence it.

This was an incentives issue. They were paid by the companies they rate. So in order to not lose market share they had to certify junk.

I would argue that they happened exactly because the brokers who made up those financial institutions understood the math perfectly well: They make huge commissions on CDO and other deals. They sell the bad debt to some other shmuck and walk away with cash in their pocket and someone else holding the risk.

That's why it keeps happening. They get rich and get away with it every time.

At least 2008 had nothing to do with being good or bad at math. In fact, it kind of happened because of some rather fancy math. This article has been posted on HN multiple times:

https://www.wired.com/2009/02/wp-quant/

I would call misapplying mathematical tools being bad at math personally. If you said that because you are intergrating revenue back into investments and are showing nx growth you'll get quadratic growth in exchange for cutting growth in half I would call that being bad at math. An uncommon form of it but still bad due to it being a clear failure to model the underlying domain. They did the same thing really by not taking common cause failures into account at all, thinking one can build low risk out of high risks, and more that an ounce of critical thinking could have saved them from.

I think that's a pretty broad definition that wouldn't find much support. I think the words you used later, "critical thinking" or lack thereof, are much more apt.

> Ah yes, because as 2000, 2008, and 2019 have shown us

Were you trying to sneak in a comment about an impending explosion? :)

Also, if you've read any of the stories about digitization in financial firms, they are totally incapable of hiring their own geeks because it would displace very, very rich people who would have to do the hiring.

Do you have a link to any of those stories? My experience with financial firms is completely contrary to what you're saying here.

Those examples happened because of people with the wrong incentives doing math correctly, not because of anyone doing math incorrectly.

2019?

Reply on news.ycombinator.com