View RSS Feed

Bullitt

High Frequency Trading and the Market

Rating: 2 votes, 1.00 average.
High Frequency Traders (HFT’s) are the biggest player in global markets today and are responsible for the highly complex trading algorithms that companies like GS strive to keep secret. HFT’s claim to be “liquidity providers” to pick up a bid when needed while large institutions who trade large blocks of a stock claim they have had a negative impact on the overall financial markets. This post is a quick overview of one particular strategy HFT’s participate in called “Rebate Arbitrge”.

The reason your broker (Scottrade, Trade King, eTrade, etc) can charge a small commission compared to 20 years ago is because your order, when placed on the website, is sold to a HFT who will place the trade. HFT’s operate with a special privilege which allows them a rebate of ¼ penny per share per one way trade which is offered by exchanges. One can easily see already how there is an incentive for a HFT to buy and unload as many shares as possible during the trading day. HFT’s co-locate their servers right inside NYSE and NASDAQ for that millisecond advantage over competitors (low latency) which means the difference between them or their opposition making a profit. During the trading day, HFT’s ping an exchange repeatedly and then cancel the order before it gets filled just to get an idea of what the market book for a stock or ETF looks like. This is kind of like faking a shot or pass in basketball to see if your opponent takes the bait. Even if they do, you still have the ball, but the defender has played his hand. Some say 95% of HFT orders are cancelled before another side can take the trade. Before you try this at home, remember these guys have a millisecond advantage over the general market, so the general market can’t even react fast enough to make the trade when they want to.

How does an HFT make money and why are they controversial? Let’s say a high volume stock, INTC, is trading at 23.95. You run a mutual fund and want to buy 10,000 shares. Putting the bulk order out there would kill you because any trading desk would run up the price. Instead, your algorithms break the trade into smaller chunks and pass them on to your trader who will strive for something like a VWAP or “better than close” price. HFT’s have counterintelligence algo’s that spot things such as a steady stream of smaller trades at the VWAP and are able to interpret them as a big buyer. The HFT will step in and offer to buy at 23.96 driving the price up a penny which will get filled by a seller looking to sell higher. (Remember HFT’s get the ¼ penny rebate per share whether the trade makes money or not.) The HFT will instantly turn around and sell those shares purchased for 23.96, effectively making their trade a wash, but costing the institution a penny more than it should have. The HFT proceeds to collect a rebate of ½ penny per share (round trip) and in the meantime, has ‘provided liquidity’, and drove the price viable institutions must pay per share.

Multiply this by millions of shares a day and it’s pretty hard to have a bad day on wall street.

Submit "High Frequency Trading and the Market" to Digg Submit "High Frequency Trading and the Market" to del.icio.us Submit "High Frequency Trading and the Market" to StumbleUpon Submit "High Frequency Trading and the Market" to Google

Tags: hft
Categories
Uncategorized

Comments

  1. bmneveu's Avatar
    great read! very interesting
  2. Cactus's Avatar
    Thanx for the info, Bullitt. I had no idea. It never occured to me someone could get paid to trade. I thought all trades had to be paid for and that's how the exchange made its money.
  3. RealMoneyIssues's Avatar
    Where's the dislike button (about what the HFT's are doing, not the post)

    Great explanation Bullitt!!
  4. mufasa's Avatar
    Wonderful stuff. I have been dabbling with a python algo and a few API's. Some interesting results based on historical data. But like they say historical data does not future results lol...nice to see articles such as these.

S&P500 (C Fund) (delayed)

(Stockcharts.com Real-time)
DWCPF (S Fund) (delayed)

(Stockcharts.com Real-time)
EFA (I Fund) (delayed)

(Stockcharts.com Real-time)
BND (F Fund) (delayed)

(Stockcharts.com Real-time)

Yahoo Finance Realtime TSP Fund Tracking Index Quotes