Algo trading rapidly replacing humans, warns government paper
computerworlduk.com
computerworlduk.com
From the article "On the positive side of electronic trading, the panel said, liquidity has improved, transaction costs are lower, and market efficiency is generally better."
Relevant Monty Python: http://www.youtube.com/watch?v=ExWfh6sGyso
A position should be a position and not a matter of beating the other guy to a momentary, transitory state.
You are arguing to shut down the whole stock market then. Even value investors like Buffet are in the business of beating the other guy to understanding that given stock is over or under valued.
As for transitory state, the point of the stock market is to reflect what is thought to be the 'true' value of a company at any given time.
(And a new meaning for the term "jittery market". ;-)
Once a minute, or something else. That's something the big brains can sort out -- earn those salaries in a productive endeavor, for a change.
Nickle and diming retirement investments does not, in my perspective, contribute qualitatively to society. (A bit baldly, and narrowly, stated, but a significant part of my point.)
Of course, in an increasingly global marketplace of trading, this probably does not have legs.
I see the market as connecting investors with investments. Spending heavily to leverage momentary arbitrage does not, in my perspective, add significantly to the base function of the market.
Such a stepped clock might also allow/aid some monitoring and capping of flash changes -- the continuing dread of many, that an unbalanced algorithm or combination will take the market down.
I agree that it is immoral for Enron to reduce the supply of electricity, while towns are having brown-outs, just to raise the price. The same is true of hoarding much needed commodities, just to raise their price. I'm not even a big fan of speculatively buying up domain names.
However, stocks can not be consumed. For most people, there is no reason to transact in the financial market other than financial gain.
I don't see how this type of trading automated trading harms retirement investments. High frequency exists because exchanges now allow electronic transactions and automated matching. Algorithmic trading (which generally means automating the execution of large number of shares, such as institutional (retirement funds) orders) have also greatly benefited from same opportunities.
If the complaint is that when retirement funds buy or sell large number of shares, HF guys come in and pick them off -- this is no different from regular day traders (or even professional traders). In fact, large trades have no "right of way" in the stock market. If you are buying a million shares of some stock, other traders will react and raise the bid. This is not a subversion of how the market works, this is the essence of price discovery and how the market is supposed to work.
>I see the market as connecting investors with investments. Spending heavily to leverage momentary arbitrage does not, in my perspective, add significantly to the base function of the market.
But does it take away from the base function? Just because some rich guy is spending money on computers, routers, programmers and mathematicians does not mean the market is negatively effected. Surely the burden is on the critics to prove why HF is bad.
actually, study after study shows that algorithmic traders improve things for "retirement investors". more algo trading leads to lower spreads, which decreases trading costs for long term investors.
think about who value investors had to buy from before algo traders. they bought from human market makers who were ridiculous inefficient, and as a result, had to charge huge spreads. these spreads have a material (negative) impact on investment returns for value investors.
I'm not opposed to automated trading. I do, however, think (well, have the perhaps unqualified opinion) it has become excessively focused on near time outcomes and pushing transaction volume in preference to the actual performance/outcomes of investment properties.
Investing in companies, fine. The risk that you take with your invested money evolves over time, true. But at some point it just becomes ridiculous that people are making money from 0.0001 fluctuations in prices.
Also "oft-cited liquidity that they provide" >>
This argument is farcical to the nth degree. When the average time for HFTs to hold onto a new position is less than 30 seconds, I'd say they're merely claiming credit for liquidity that was already there.
"Algorithmic trading has sharply increased over the past decade. Does it improve market quality, and should it be encouraged? We provide the first analysis of this question. The NYSE automated quote dissemination in 2003, and we use this change in market structure that increases algorithmic trading as an exogenous instrument to measure the causal effect of algorithmic trading on liquidity. For large stocks in particular, algorithmic trading narrows spreads, reduces adverse selection, and reduces trade-related price discovery. The findings indicate that algorithmic trading improves liquidity and enhances the informativeness of quotes. "
To put it another way, how do we know the function of liquidity to net human benefit is monotonic (always increasing) rather than having a local maximum?
lower spreads means you pay less when you trade, simple.
Imagine it - "Earn cash in your spare time - estimate the future profits of Fortune 500 companies, and plug them into a structured database! Senior members represent us on boards of directors, telling CEOs what the computer thinks they are really worth! (payments contingent on the long-term value of your work)."
However, to suggest that the entire 100 are needed to liquify when 10 will do (less than 30 seconds later I might add) is farcical.
> For large stocks in particular, algorithmic trading narrows spreads, reduces adverse selection, and reduces trade-related price discovery.
its not just a question of who will buy, its a question of who will buy at what price. if you have more liquidity suppliers, spreads will thin and you'll pay less.
I freely admit that if you want to measure liquidity as the amount of buyers in the marketplace, then HFT definitely increases liquidity. However, to suggest that HFT is the only source of liquidity, as I've said twice before, is farcical.
Also, it is totally accurate that HFTers reduce spread in the marketplace. They also increase the spread when they are driving the price of a stock up. Also, HFTers aren't unique to affecting the spread in the market. Every large scale buyer can change the spread.
Not only that, but my original point regarding the provision of liquidity and HFTers claiming they provide the essential job of supplying it (ie: without HFT, there would be no buyers) and I used the short period of time of the average stock position holding to suggest that the liquidity exists regardless of the existence of HFT.
The issue is not whether a 30 second hold on a stock is better / worse than a 3 year hold but whether stock markets still serve their purpose of providing capital for companies. The vast majority of trades are for stock that does not capitalize the company.
This is factually incorrect. A lot of ultra-high frequency trading refers to a practice where traders post a bid or an offer--their orders sit in the 'book,' waiting for other to come and trade with them. The traders who have bids and offers sitting out in the market are even referred to as liquidity providers (as opposed to liquidity takers). There are whole exchanges out there which offer rebates to such traders for providing this liquidity (which means you actually get paid by an exchange to post liquidity).
These guys are not in the business of holding positions, they are in the business of providing trade-able liquidity. A car dealer may buy your car, then turn around and sell it the same day--that doesn't not mean they are harming the car market.
I realize the semantics of this can be argued ("existing liquidity doesn't preclude or prevent the existence HFT" for example), but not only do I think that is a wash argument but it also misses the point entirely on the utility of HFT.
HFTers don't have to justify their existence at all. They exist because they are able to profit and their existence is not against the law. As far ethics and morality, I guess that is what we are arguing here :)
Liquidity is not pregnancy, there can be a little of it.
Besides, a lot of what is called algorithmic trading is driven by human, "manual" decisions: say a fund manager wants to buy $10 million worth of stocks someone composing the S&P 500 today, without raising the price of ACME too much, so they use a piece of software that does that automatically.
That, or people just fear things they don't understand, and robot traders are even harder to understand than human traders, I guess.