[nostrademons]

FWIW most of the folks who work with C++ in the financial industry are not really "messing with people's money" except in the sense of trying to exploit market inefficiencies to take it. In a typical quant hedge fund no actual money changes hands; instead, the C++ parts are there to perform analytics extremely quickly, which then feeds into an execution engine (usually also in C++, or sometimes Rust or assembly) that's sending trades to the exchange. And they're usually trading their own capital or a small set of very wealthy limited partners.

The retail financial industry, the folks like Fidelity and Vanguard that manage money for ordinary people or folks like Bloomberg that supply data for this, largely runs on Java. There are fewer foot-guns here, and you really don't want a security vulnerability that loses your customer's funds or creates an inaccurate statement.

[perlgeek]

> except in the sense of trying to exploit market inefficiencies to take it

that's precisely the bad part about it

[nostrademons]

Well yes, but the gaming industry is all about exploiting psychological efficiencies to take your money. Big Tech (in its advertising/retail/OS forms) is all about exploiting psychological inefficiencies to help other people take your money, and then taking a cut of it.

[klik99]

If you mean "gaming industry" meaning gambling machines, then yes. If you mean "games industry" as in video games then it's just not true that it's "all about exploiting psychological inefficiencies". There are many monetization strategies that are not exploitative that are common and even dominate in the market. Ubisoft, DLC, or Candy Crush freemium models are not representative of the whole industry.

[nostrademons]

AAA games with one-time payments are still exploiting psychological inefficiencies; there is a reason why we are evolutionarily hard-wired to pay attention to fast-moving visual settings where action is required from us.

For those who say "but I like to play video games" - yes, that's the point. Day traders like to speculate, and gamblers like to go to the casino. Your downside is capped with one-time payments, but a gambler who goes into the casino with a $20 and no other money also has a capped downside.

[zeroonetwothree]

Arguing that one-time payments with up-front pricing are exploitative is pretty silly. That's the basis for the entire economy since we invented money thousands of years ago. If it's exploitative then you're basically lumping all economic transactions into the same bucket, which makes it fairly meaningless.

[jcelerier]

> That's the basis for the entire economy

this close to getting it

[sshnuke]

so gaming addiction is also a silly argument?

[klik99]

Unless you're selling me living in a cave foraging for nuts and berries, you're trying to exploit some psychological inefficiencies.

I do get what you're saying, but if you take it this far then literally everything sold is considered exploiting inefficacies, psychological or biological, and it's kind of a useless statement in the context of comparing markets, only useful as part of an overall critique of capitalism.

All I meant was the games industry gets a very bad rap for being dopamine driven skinner box design and it simply isn't true, and being in the industry for 15+ years I have met very few people who are comfortable exploiting psychological effects for gain. If I was in a different part of the industry I probably would have met a lot of people who ARE comfortable with that. It is there. But it is far from "running solely on" that attitude.

[lopatin]

Sorry what’s bad about making profit from market inefficiencies?

[saagarjha]

It’s not something that drives obvious value for people.

[isbvhodnvemrwvn]

Don't throw rocks in a glass house, a lot of shit software engineers do does not stove obvious value for people.

[saagarjha]

Hey, I'm not giving software engineers a free pass either.

[dahfizz]

How does your employer pay you if they don't take money from anyone?

[perlgeek]

Of course they take money, but they hopefully also create real value.

Providing liquidity a fraction of a microsecond faster than a competitor doesn't seem like real value to me, more like a cover story so that politics doesn't outright forbid high-speed quantitative trading.

[mr_00ff00]

Ahh classic. Obviously only these financial firms are the only ones not creating value.

Those social media companies, start-ups people do for venture capital cash grabs, Web3, and addictive mobile games all create such value.

EDIT: Okay, as the replies mentioned, OP did not make this argument. I still feel it’s valid, as a lot of big tech is neutral/worse for society, yet they are some of the loudest critics of low latency liquidity.

[bobsomers]

This is a strawman argument. GP never claimed what you stated.

[saagarjha]

I don’t see that being argued.

[stouset]

By providing a service to customers, rather than finding ways to skim money off the top of markets?

[nostrademons]

If you ask most financial institutions they would say the service they provide is "liquidity" - they let you put your money in or take your money out at any time (in exchange for securities), as well as make decisions about where you want to allocate your capital based on your personal view of the world.

This always fell a little flat to me (hence why I left the financial industry early in my career), but it's pretty analogous to merchants or retail. What service does a storefront provide, other than marking up the wholesale price of a good by 3x? They offer convenience - you can buy anything you want at the time you want it just by picking it up and swiping a credit card, vs. having to contract with a wholesaler for delivery, which might be months in the future, at a place that's convenient for the wholesaler, and needing to buy in quantities that no individual really needs. So it is with brokerages & exchanges - they let you buy any security with a couple clicks of a button, vs. needing to get an ISDA and pay a lawyer to write a contract, plus find someone else that wants to trade with you, plus take on the risk that your counterparty goes insolvent. And so it is with market-makers like Citadel or Jane Street: they ensure that you can always find a buyer for your securities (for a price), and that you don't have to worry that about simply not being able to sell stocks at any price.

[bobsomers]

> If you ask most financial institutions they would say the service they provide is "liquidity".

I've honestly never understood this. Economics 101 teaches us that artificially manipulating supply (liquidity) is at direct odds with a free market's ability to do price discovery. If there is effectively infinite supply of any security (because all these high frequency firms are instantly hedging out the risk against their entire portfolio), why should anyone believe the NBBO? It's just a made up fantasy number at that point. It doesn't represent what the real security is worth because there are an infinite number of buyers and sellers willing to transact at any moment.

It really does feel like an enormous grift.

[nostrademons]

HFTs and other market makers are time- and risk-shifting supply & demand. In a functioning market, you can always find someone else to take the other side of the trade if you wait long enough, but you may not be able to find them now. Particularly in market panics, where all the buyers tend to disappear at once.

What a market maker does is say "Sure, I'll take the other side of the trade - right now, at this price, even though I don't actually need or want the security. And I'll take on the risk that I might not be able to find a counterparty later. I trust my knowledge of the market enough to believe that the price I've offered you is one where I can profitably unload it later." And if they're wrong about gauging future supply & demand, they go bankrupt. Markets function on survivorship bias; repeat this cycle for long enough and the only market makers left are those that are extremely good at gauging future supply & demand and using it to set spot prices.

On some level, they're arbitraging risk & rationality. A retail investor makes a trade that is locally rational - perhaps they need to sell stocks to pay taxes, or purchase a home, or they've lost their job and need a savings cushion. And many of these events affect multiple people all at once (eg. everyone paying taxes on Apr 15, home sales peaking in the spring and declining in December, a recession causing people to lose their jobs all at once), which leads to temporary declines in demand that don't at all reflect the discounted value of all future cash flows of the stock market. Market makers smooth out these fluctuations, buying & selling stocks when it's globally rational based on the fundamentals of the companies and holding inventory when short-term supply does not match long-term supply.

[bostik]

> artificially manipulating supply (liquidity) is at direct odds with a free market's ability to do price discovery. [...] there are an infinite number of buyers and sellers willing to transact at any moment.

No, there are not. There is a limited (and not really that large) number of active participants in any given market segment, and the 80/20 rule holds among them pretty well too. Thanks to Money Stuff it dawned on me recently that market makers are not really providing liquidity. They provide immediacy. Buyers and sellers have to meet not just in price, but in time too.

The spread paid by market participants should be seen as their baked in commission for providing a very specific service: reduced wait time. We can certainly disagree about the societal value of what that service has morphed into, but the reality is that for other than the most liquid[ß] assets, participants are willing to pay extra for the ability to transact NOW, and not in some unspecified time in the future.

My personal opinion is that NBBO is an imperfect mechanism to set upper bounds to these commissions. In other words, it limits how much retail transactions can be fleeced for.

I am without a doubt a retail investor: I generate maybe three transactions a year, and I'm happy to pay something like 0.15% to 0.25% transaction fee each time, in the knowledge that I am getting a fair price at the time. To me that is a reasonable cost of convenience. My transactions are so small compared to the trading volume that they have zero price impact: in purely financial terms I could be paying a smaller commission if I set up the limit order thresholds myself - but that would take more time and be less convenient.

[ß] Read: continuously traded in very large quantities

[stouset]

The average retail investor just does not need that much liquidity, especially if they're not buying individual stocks (which by and large they shouldn't be).

[nostrademons]

And the average retail investor pays pennies to market-makers - if a typical spread is 1-2 cents and you make 5 trades a year, you've paid between a nickel and a dime.

As with advertising, when you sum up dimes across hundreds of millions of people, it becomes an appreciable amount. Particularly when you also include trades made by institutions on behalf of retail investors. Your 0.5-1.5% management fee on an actively-managed mutual fund is partially going to pay salaries for the fund manager, and partially toward brokerage commissions, spreads, etc. for the trading itself. And mutual funds care a great deal about getting liquidity when they need it, since they're in breach of contract if a flood of redemptions come in and they can't liquidate assets to pay them.

(Market makers are also analogous to the credit card industry in that a small fraction of dumb people subsidize a large number of fiscally responsible people. If you get a rewards card with no annual fee and always pay it off on time, you turn a profit, paid for partially by merchant fees and partially by idiots who carry a huge balance at 25% interest rates. Similarly, if you buy-and-hold a good stock or index fund, you're making profits far in excess off what the financial industry can skim off you. The suckers are largely made up of day traders, wage slaves who can't save anything, and underfunded pension funds with principal agent problems.)

[Gigachad]

Levelling prices between markets is a useful service. Eventually the amount being skimmed is tiny and regular consumers are much more likely to be getting a fair price without having to shop around.

[whiddershins]

Just try imagining in what ways the world would be different if financial markets were very slow and inefficient.

[jeffreygoesto]

Exploiting information asymmetries to funnel other people's money into your pocket before or without them knowing it. No value had been generated, it's just extraction.

[jleahy]

Bloomberg is actually mostly C++.