Comment by strangedynamics | Hacker News Reader

Comment by strangedynamics | Hacker News Reader

Parent Full thread

strangedynamics

My experience is limited, so take with a grain of salt: my sense so far is that it helps stamp out any statistical/numerical inefficiencies in the markets that humans would otherwise be inefficient/slow to adjust to. For instance, if a straight up arbitrage opportunity exists across exchanges, automated strategies ensure that it goes away very quickly, much more quickly than discretionary human traders would. If there are proven statistical trends to the markets, then those should be acted on as well, because it means we'll arrive at the fair price of an asset sooner rather than later. A simple example: if a stock closely tracks the price of oil, and there's a quick uptick in the price of oil, it's better for the price of the stock to get immediately adjusted upwards rather than after a delay (and purely quantitative, automated strategies make this happen). I think the name of the game is that you have a bunch of people acting in their best interest, and thereby giving us an efficient market, which benefits the rest of the world by ensuring low slippage, fair prices, high liquidity, etc.

Basically this. I'll slightly elaborate on why an efficient market is a good thing.

It means that anyone can change money into a different currency without having to worry about being ripped off. There is one exchange rate and you'll be paying that plus a small margin.

It means that anyone can invest in publicly traded companies without having to spend an inordinate amount of time and effort trying to value them. Pension funds would be next to impossible without this. To the extent that the market is efficient, you don't have to worry about the imminent collapse of IBM. If anybody knew, the prices will reflect it even if the rest of the world has no idea why the prices are as they are. (The further markets get from being efficient, the further this gets from being true. Compare investing in the S&P 500 to the shenanigans happening with penny stocks. Be glad that your pension fund gets to invest in the former so it can pass on the latter.)

It means that companies can raise money without too much difficulty. They'll sell a part of themselves, and a legion of quants will give them as fair a price as humanly possible. If the price were unfair, someone could exploit it to make money , correcting the price in the process. Ergo the price will be fair. While this sounds far removed from the welfare of the people, this is what greases the engine that runs the jobs of virtually everyone in the developed world. It also lets people pool risk to try risky but worthwhile things. The impact this has on innovation cannot be overstated.

Of course, I agree with the above poster that finance probably isn't the best place for smart people to spend their lives. I'm just claiming that finance people aren't overpaid, everyone else is ludicrously underpaid.

This is a really good case for quantitative finance, and it's also really clear, so thanks for both. Consider my perspective changed for the better.

There's still a bit of dissonance in this for me, but it's abstract and strays a bit what you're saying, so don't feel a need to respond: I think, for example, about the friends and acquaintances I have in finance who I've often heard lament about the lack of meaning, utility, and social value in their work despite the good pay. They're smart men and women, every single one, so I'm sure they know they case you're making - so what gives? (This is clearly more a question for them than for you.)

I also wonder idly what proportion of the explosion in the finance industry since the 1970's is taken up by the sorts of plainly beneficial and useful efforts you describe, and what proportion is simply craven, greedy bullshit. Clearly the latter is more salient in the cultural and moral imagination, in no small part because of the Madoffs and the Beskys and the Milkens of the world, not to mention everyone involved in the 08 crisis. That the former isn't as salient makes me think, like, how many Jack Bogles are there for every Gordon Gekko? I know a ton of Gekkos - I went to college with them! - but few Bogles.

I feel both of your questions and have no answer for either :(

Reply on news.ycombinator.com