Sunday, September 20, 2009

Are flash orders really so bad?

I confess I don't know much about flash orders, not being one of the Big Boys on the Street, until I read that the SEC is banning them. (For a clear diagrammatic explanation of flash orders, see here. For a refutation of some of the myths and misunderstanding surrounding flash orders, see here.)

It seems to me that flash orders can be understood as "request for liquidity" issued to various potential market makers/liquidity providers, not unlike the usual "request for quotes" (RFQ) common in other industries. They are issued when there is not enough liquidity on a specific exchange to satisfy an investor's need, and they ultimately benefit investors by lowering their transaction costs. The fact that high frequency traders are able to make lots of money by providing this liquidity is besides the point. Liquidity providers are supposed to make money by providing liquidity!

Some people, including Senator Charles Schumer and this New York Times op-ed, believe that flash orders are akin to front-running, a clearly illegal trading activity. But they are wrong. Front-running means that if you know someone is going buy a stock, you step in front of them
and buy it cheaply first, hoping to sell it to this slower buyer at a higher price. In the case of flash orders, the high frequency traders are instead selling this stock to the original investor, often at a lower price than available elsewhere and thus benefiting this investor, hoping that the prices will come down in the future after this liquidity need subsides. This is manifestly not illegal. This is what a market is built for!

Another way to understand that flash orders are not at all front running is that anybody, including you and me, are free to put in limit orders at the same price as those of the high frequency traders, way ahead of time, in a specific exchange, and become liquidity providers ourselves. You don't have to wait for a "request for liquidity" before doing so. And presumably you will reap the same benefits as the high frequency traders. You are not taking any additional risks over the HF traders either, since if no requests for liquidity ultimately arrive, you are not any worse off for wear. You cannot begrudge the profits of the HF traders just because you didn't put the limit orders in place beforehand!

Maybe there are some other angles which I miss which can convince me that flash orders are evil. But until my kind readers convince me otherwise in the comments section, I will regard this piece of legislation as another SEC attempt at demagoguery.


Anonymous said...

The reason flash orders are bad is that they allow some market participants to look at the order book and then decide if they will fill the order or not. So when a lot of large orders start arriving, they can join them and trade in front of them (while the orders are still being flashed), but when small orders arrive, they can fill them. They can jump in front of limit orders when non market-moving orders arrive and then leave only the market-moving orders to fill the limit orders. It is kind of like getting to put a limit order out there, but then being able to cancel it after you see how it would have filled.

The problem with flash orders is that it allows some market participants to provide liquidity when there is little risk in doing so without requiring them to stay there when the risk of providing liquidity increases. So they get to collect all the income from providing liquidity without taking on the risks. In other words, it allows them to provide liquidity when it is profitable and not very necessary without requiring them to stay there when the market is one-sided and the liquidity is actually needed. There is no way that a true market maker can compete with that business model.


Ernie Chan said...

Thank you for your good explanation. However, I don't necessarily agree with the "trade in front of them" part, since by the time the HF traders see the flash orders, they can only snap up the liquidity in other markets where there are better prices. And if there are really better prices in other markets, why didn't the first trader route the order to those markets in the first place?

I agree with the part about not providing liquidity when they see the flash order is large. But like Joe Ratterman said, this is only a problem because SEC regulation does not allow locked market when the bid price for a large buy order can be higher than the best ask price of a small order size. If such locked market is allowed, then there won't be any need for flash orders, and everybody can see the complete order on a certain market and see whether they want to fill it.

In that sense, yes, I agree that given the current regulations, flash orders are unfair. But the way to fix the problem is not just to abolish them ... it is to change the regulation concerning locked markets.


Anonymous said...

Themis Trading has a pretty good explanation...


Ernie Chan said...

Thanks for your link to the However, there are some details in that blog that I don't necessarily agree with.
For e.g. that article says:

"1-The market on XYZ is 21.05 bid for 500 shares, 500 shares offered at 21.08.

2- Say you want to buy 500 shares of XYZ at 21.08. You place a buy order with a $21.08 limit through your online broker .

3- Your broker routes to a market center which utilizes the flash method. Before your order gets routed to the market center which has the $21.08 offer, it is flashed to a subset of market participants.

4- A flash subscriber see this buy order coming and decides to trade in front of it and buy the 500 shares at $21.08 for their own account.

5-You get nothing done and 500 shares trades at $21.08."

But wait! Why should your online broker route your order to the market center that does NOT have the $21.08 offer? Why not directly route it to the one that HAS that offer and you can be filled immediately without all that flashing? It seems to me it is the fault of your broker that you got "front-run", not the fault of the market center with the flash orders.


Christian Gross said...

Didn't you know that Wallstreet is bad and you trying to defend it was bad?


A question I have is what about the pits? Remember those? The pits where people give hints, winks, nods to each in a quest to sell and buy things?

I wonder if a pit is not the ultimate frontrunning? Could not the argument be made that when big orders from clueless people come to the pit the pit just gives them the worst price?

Anonymous said...


Let me try my part to explain the problem with flash order (front running):

1. Let us say a mutual fund submits a buy order for 10,000 shares of XYZ at 10.00 at BATS.

2. BATS has a offer for 4,000 shares at 10.00 but now other "best" offer. However, BATS sees that NASDAQ has an offer for 6000 shares @ 10.02. So, in accordance with NMS rules, BATS should route the remaining 6,000 shares to NASDAQ for execution.

3. However, BATS supports "flash" orders. So, before routing the remaining 6000 shares to NASDAQ, BATS creates a "flash" order for 6000 shares @ 10.01 (a better price than NASDAQ's 10.02).

4. Now, if a "flash" subscriber firm sees the order and accepts the order, then the mutual fund benefits, and also BATS collects its commission.

5. However, the problem occurs when a HF/IB sees this order, and then sees that NASDAQ has offers for 6000 @ 10.02 and another 3000 @ 10.04, and decides to front run the mutual fund's order by buying NASDAQ's 6000 @ 10.02 and posting an offer for 6000 @ 10.03 fully knowing that within the next 500 ms the mutual fund's order will be hitting NASDAQ's new offer at 10.03. Now, what do you call this but for blatant front running?

Your comparison of "flash" order to RFQs is not correct. RFQs dont expose orders, but is just a request for quotes. The mechanism is different and you could learn more about this mechanism, should you wish, from CME group.

Hope this helps.


Ernie Chan said...

Thank you for your detailed example. However, there is a problem with it:

If the buy order for 10000 shares has a limit price of $10.00, BATS won't route the order to NASDAQ since NASDAQ doesn't have an offer at $10.00. Furthermore, it won't flash the order, since the bid price is now lower than the offer price (after filling the first 4000 shares). As a result, there will be a displayed bid for 6000 shares at $10.00 on BATS.

Flash orders only occur in situations where the bid price is higher than the offer price, yet the offer size is not sufficient to fill the entire order. (Or vice versa, of course.)

Perhaps you should reconstruct the example?


Anonymous said...


I saw my "mistake" after I posted (result of my editing on the run!) and didn't get a chance to follow-up as I had to run.

Anyway, point 1 should have read:

1. Let us say a mutual fund submits a market buy order for 10,000 shares of XYZ to BATS.

Everything else should remain the same in my previous example.


Ernie Chan said...

Thank you for your correction.
Yes, in your new example, there is indeed an opportunity for the HF traders to front-run your market order, and in this sense flash orders are not fair. But as I responded to Steve's comment (or rather, his reference article's comments) above, it is really the fault of your broker to allow this front-running opportunity. If your broker sees that there are certain depths of liquidity on both BATS and NASDAQ, the smart thing to do would be to break up the order and route to both markets. This is what a broker's smart routing supposed to do, and I think many brokers offer that facility (I know Goldman and Interactive Brokers both do.)

Anonymous said...

Ernie: I understand your point about broker routing, but there are two problems with your line of thinking:

1. There is always a possibility that when the broker routed the order (going back to my example), BATS had 10,000 @ 10.00 (so, the broker did the right thing routing the order to BATS). But when the order was in transit, another order (from somewhere else) took out 6000 of 10,000 @ 10.00 leaving 4,000 @ 10.00. This is a classical distributed system problem, and no broker routing algo can solve this problem. Now, this scenario leaves my previous example of front-running intact, with no fault attributable to broker's routing.

2. When an order is submitted to the market, there is an implied expectation that the market is fair. I do not expect my broker to make sure that the "executing" venue is fair (or device schemes to overcome unfair procedures used at executing venues); I expect executing venues, in-by-themselves, to be fair. The potential for misuse of "flash" orders breaks that expectation.

Have a great day.


Ernie Chan said...

Yes, I agree that the scenario you described does seem unfair to the original investor. You have convinced me that flash orders do present front-running opportunities.
Thanks for the detailed explanation!

Anonymous said...

A Question,
I'm working on backtesting pairs trading on US market.
Can you give me an idea of how much sould I count in transacton costs. If I calculate my returns based on 100% long+100% short (gross of 200%) taking 10 bps for one way is a resnable figure?

Ernie Chan said...

I typically assume 5bp one way. Of course, if your brokerage charge higher commissions, you should use a higher figure.

Anonymous said...

Golden Networking ( is organizing the High-Frequency Trading Experts Workshop 2010 (, Practical Implementation of High-Frequency Trading Strategies, on December 9th and 10th in New York City. It has been recommended for executives in finance and investments who work at Investment Banks, Hedge Funds, Pension Funds, Broker Dealers, Consultancy Groups, Prime Brokers, Solution Providers and Exchanges and wish to gain a thorough understanding and practical knowledge of high-frequency trading.