Before HFT existed those risk free trades were taken by NYSE floor brokers. If anything the amount taken is smaller since spreads are so thin now - they used to be much wider in those days. There is always going to be some way to profit by doing a really obvious thing the fastest. Like trading on news or company earnings, etc.

And as the sibling commenter said I suggest reading other sources besides Flash Boys, which I have read before. There is a lot of misinformation in that book. And by the way, IEX has been a public exchange now for some time and still has a tiny share of the volume.

They are not real spreads. They are fake spreads made by HFT bots using small orders. Once a real investor will try to execute on such an order, the HFT bot will either withdraw it (or it will simply get filled because it's a minimum order for the minimum allowed amount), and then the bot will immedately put in another order which will make more money for the HFT and less money for the investor. HFTs will use their fast connections to other markets to make sure that the new order is such that the investor will have no choice but to pay the price, because by the time the investor will try to go to another market to execute the trade cheaper, HFTs will already have been on that other market have front-runned the investor's order.

This is not "providing liquidity". This is parasitic behavior. Not all HFT strategies do this, but many do.

You have a skewed perspective of what HFT is - Flash Boys probably contributed to this. That is a sensationalist novel and widely derided within the actual market making community. For example, fiber optic cables were outdated by microwave feeds several years before the book came out.

In fact, the most common form of HFT is low-latency market making, which directly increases liquidity and lowers spreads for market participants.

> For example, fiber optic cables were outdated by microwave feeds several years before the book came out.

That's the issue you have with it? Of all the possible issues, that's the one you think underlines the bad quality of the book?

The fiber lines or microwave feeds or any other technology used is only a minor detail. The meat of the issue is that there are many HFT strategies that don't provide any real tangible service to the market. HFTs race clients to different exchanges, buy assets cheaper and sell them the next millisecond to the same client at higher price, because they can (because the client is slower). They put in fake small orders in order to read market information before anyone else so they can front-run legitimate investors. How is that providing service to the market?

There seems to be a lot of disinformation in the market, and a lot of things that are meant to sound much more complicated than they are. But the bottomline is that if there is a certain strategy X, that complies with the following conditions: 1) It doesn't take positions in the market, doesn't take risk, so it doesn't evaluate underlying assets and does not help price discovery. 2) It gets in the way between two other participants, simplifyingly a buyer and a seller (where the buyer will end up holding an actual invested position for a longer time) 3) Both the buyer and the seller will end up paying more for their trade when the HFTs are executing the strategy X, compared to a case where noone would be executing the strategy X.

Then yes, it's objective to say that the strategy X is a parasite on the market, and does not provide any real value to the society. The value of markets is providing liquidity, distribution of investment value, risk hedging and price discovery. These HFTs (which is a bulk of all HFT strategies) don't do any of that. No real investor will notice a difference of 50 microseconds when executing their trades. No investor is interested in that.