This page has been archived and commenting is disabled.
Whitney Tilson And High Frequency Trading v3
Whitney Tilson continues investigating High Frequency Trading
From his Sunday letter to clients:
1) An interesting comment from a friend:
The reason these firms make so much more money in fixed income is because the market making and prop trader is the same person. The distressed desk flow traders (the area that I know best) at the big firms have balance sheets up to a couple billion per group and they have all the information about who is buying and who is selling. For a big buy side account to get out of a position usually takes up to a week and can move a bond 25-30% easily, so once a desk gets the first sell order from a Pimco or the like, they immediately start reducing their own exposure or get short because they know that a huge amount of supply is coming. So on top of the fact bond bid/ask has moved from 10bp 2 years ago to 200bp now, these guys take huge prop positions and front run their clients. The weird part is that every single person in the market knows what is going on and nobody does anything about it.
2) An excerpt from another friend's comment (full text below):
My business partner passed on your piece on algorithmic trading and
so-called "toxic equity trading." We are fundamental value investors
with long holding periods, but I worked with and know well a number of
algorithmic traders and the strategy.
I think it’s very, very
important to understand that the anti-high frequency trading arguments
in the commentary are poorly informed and/or the very biased views of
market dinosaurs. As I’m sure you’ve seen, news and manufactured
controversy over high frequency trading has somehow exploded in the
last couple of weeks. This campaign is an invitation to regulators to
wipe away an *enormously valuable* source of liquidity and price
competition in the equity markets.
High-frequency trading benefits
*all* of us long-term investors by making it easier to enter and exit
positions, narrowing spreads, and reducing the trading costs paid to
market intermediaries such as broker-dealers, market makers, exchanges
and specialists. There are a few key points that need to be understood:
1)
Algorithmic traders are in the stock market to make money, plain and
simple. That is what we all do. The market is set up for people to
trade with each other and try to make money. Calling one type of
trading "toxic" and another type "good" is just positioning.
High-frequency
trading has in fact reduced the enormous costs that American investors
pay to financial intermediaries, which Buffett and Munger so rightly
rail against, so you may as well call HFT "good" vs the old high-cost
voice market-making world.
2) High-frequency trading creates
liquidity across the stock market, primarily (but not exclusively) in
larger cap stocks. This is because more volume now trades in these
stocks. Any long-term investor who wants to build or exit a position
can do it more easily and with less market impact. If 10,000 shares of
MSFT trade per day vs 1000, then it is easier to get in and out.
3)
High-frequency trading narrows the bid-ask spreads on stocks, in the
biggest stocks down to usually a penny a share (and it would be lower
except trading in subpennies was barred by the SEC in Rule NMS, at the
urging of the NYSE and others). Narrower spreads are good for all
investors. You pay less to the person on the other side of the trade in
a market order, and it makes limit orders easier to execute.
(One of
your commenters cites data from October 2008 to argue that spreads are
higher now. No one can claim that the chaos of October 2008 is
representative of how markets normally function).
4) High-frequency
trading, which has grown hand-in-hand with the rise of low-cost ECNs
and electronic broker-dealers, has significantly lowered the fees
broker-dealers, market-makers, NYSE specialists, and the exchanges can
charge us for commissions, spreads, exchange fees and other costs that
these financial intermediaries can impose.
Just ask any traditional
broker-dealer, market-maker, or old-school exchange (NYSE, NASDAQ and
the Amex) about how the rise of ECNs and high-volume trading have
dramatically reduced the costs they can charge investors, and increased
how much they have to compete for business. This is why the floor of
the NYSE is shrinking and specialists are watching movies during the
day, instead of earning huge incomes by getting a private look at
customer trades before they execute them at wide spreads. That’s good,
not bad. [TD: Please read this for just how good slippage costs courtesy of HFT are.]
5) HFT opponents claim high frequency trading moves prices,
changes the P/E ratios of stocks and create an inefficient market. Yes,
high-frequency trading is a high percentage of volume. That does not
mean that there are artificial price changes caused by this type of
trading. Opponents are citing anecdotes (in fact, making up anecdotes)
and making conclusory statements, as opposed to demonstrating this by
any real analysis of trading data over the last 5 years or so in which
HFT has become a significant percentage of market volume.
This is
not just about "flash" trading which is a tiny % of volume and which
big HFT traders like GETCO publicly oppose. "Flash" trading can be done
away with by a simple rule tweak. Your commenters instead are fighting
a Luddite rear-guard action against a much bigger
efficiency-improvement that threatens their existence, or they are just
poorly informed about how computerized markets work. They need to
accept the world has changed. The markets are run by computers,
software makes trades instead of oligopoly gatekeepers at the
exchanges, people with science and technology backgrounds make money
instead of people with backslapping skills, and there aren’t any more
steak dinners and father-to-son specialist firms on the floor. That’s
life, and it’s never perfect.
You’re welcome to quote some or all of
this email if you wish, without using our names or our company name.
Thanks! I appreciate you looking at this as it’s a very important issue
for market participants to understand, and we value investors are huge
beneficiaries of this.
3) A third friend's comments:
I have been doing a lot of thinking on the HFT issue. I am probably considered an expert on the HFT issue with respect to equity derivatives. I have built several equity derivative HFT systems. The standard equity world is mathematically less sophisticated so I believe some of my thoughts may apply.
There are roughly three types of HFT systems:
1) Correlation Systems—These systems constantly look across sectors and the market to determine if a transaction in one security should be reflected in other parts of the market. For example, if someone lifts my offer in the Educator STRA, I will attempt to lift someone else’s offer in COCO or CECO as soon as possible. This is a pure arms computer and mathematics arms race. Goldman is a big player here as it fits nicely into their execution business.
2) Dark Pools—Currently the most favored of the HFT traders, these “liquidity enhancements” allow market makers to get a look into the order before it hits the market. Depending on the “Dark Pool” they allow between 1-3 milliseconds for certain players to see the order before the general market. Our experience is that Dark Pools do not enhance execution in any way and may in fact be negative. Knight plays in this space.
3) Momentum Guys—These guys try to quickly figure out what is happening in the market and move join the Gold Rush. As opposed to the Correlation systems, they are attempting to load up the risk as much as possible. The most famous is SAC Capital.
Many of the market makers are playing in all three places, but I tried to indicate the strength of player.
It is only the “Dark Pool” that I find to be effectively illegal. It does two things that allow the market maker to peer into the order. I know that if I can see part of the order, I can take advantage of the market and move my quotes effectively. In a correlation system, I can in fact move all of the related quotes. Lose money on filling a small part of the first order to make a killing on the rest of the sector.
The second is the source of the order also means something. When I call up a stock order, Goldman knows that I am market maker and I will not run them over. If SAC calls, they know that they are getting only part of the order and that there is a lot of momentum behind it. (In fact, SAC’s order is so valuable that they will pay for it in order to take advantage of the information.) Therefore, getting orders from multiple sources can allow be to differentiate as to the value of the source.
For a pure HFT player, obfuscation, complexity of rules, shifting regulatory environment, etc are all boons because they create profit opportunity and they create barriers to entry from other players. Market fragmentation and lack of transparency mean that the people who facilitate execution can demand a bigger vig for their participation.
Philosophically, I believe that there is an important role for market makers in our capitalistic system. They provide liquidity into markets allowing for the free trade of financial markets. They deserve to be rewarded for some of the risks they take.
Historically, we have developed a system of balance to create a reward based system. Market Makers would receive certain advantages for making markets (Location and Time advantage of the floor, exemption from Reg T funding requirements, etc.) in exchange for meeting certain obligations (minimum quote size and spread)
It is very clear that the balance is out of whack. I would go further and state that the increasing proportion of transactions being done by fewer and fewer institutions is also incredibly unhealthy for the market.
4) I'm quoted in this story from this week's Economist:
Asymmetric information is nothing new. Even its critics concede that most HFT is perfectly legal. But some of the advantages that accrue to high-frequency traders look unfair. Flash orders, a type of order displayed on certain exchanges for less than 500 microseconds, expose information that is only valuable to those with the fastest computers. By locating their servers at exchanges or in adjacent data centres traders can maximise speed. “It appears exchanges are conspiring with a privileged group of high-frequency traders in a massive fraud,” says Whitney Tilson, a fund manager. Requiring orders to be posted for at least a second would nullify the value of flash orders and of probing the market.
5) This NYT Op Ed raises some very good points:
That’s rather how I feel when people talk about the latest fashion among investment banks and hedge funds: high-frequency algorithmic trading. On top of an already dangerously influential and morally suspect financial minefield is now being added the unthinking power of the machine.
...So, is trading faster than any human can react truly worrisome? The answers that come back from high-frequency proponents, also rather too quickly, are “No, we are adding liquidity to the market” or “It’s perfectly safe and it speeds up price discovery.” In other words, the traders say, the practice makes it easier for stocks to be bought and sold quickly across exchanges, and it more efficiently sets the value of shares.
Those responses disturb me. Whenever the reply to a complex question is a stock and unconsidered one, it makes me worry all the more. Leaving aside the question of whether or not liquidity is necessarily a great idea (perhaps not being able to get out of a trade might make people think twice before entering it), or whether there is such a thing as a price that must be discovered (just watch the price of unpopular goods fall in your local supermarket — that’s plenty fast enough for me), l want to address the question of whether high-frequency algorithm trading will distort the underlying markets and perhaps the economy.
It has been said that the October 1987 stock market crash was caused in part by something called dynamic portfolio insurance, another approach based on algorithms...
This is the sort of feedback that occurs between a popular strategy and the underlying market, with a long-lasting effect on the broader economy. A rise in price begets a rise. (Think bubbles.) And a fall begets a fall. (Think crashes.) Volatility rises and the market is destabilized. All that’s needed is for a large number of people to be following the same type of strategy. And if we’ve learned only one lesson from the recent financial crisis it is that people do like to copy each other when they see a profitable idea....Thus the problem with the sudden popularity of high-frequency trading is that it may increasingly destabilize the market. Hedge funds won’t necessarily care whether the increased volatility causes stocks to rise or fall, as long as they can get in and out quickly with a profit. But the rest of the economy will care.
Buying stocks used to be about long-term value, doing your research and finding the company that you thought had good prospects. Maybe it had a product that you liked the look of, or perhaps a solid management team. Increasingly such real value is becoming irrelevant. The contest is now between the machines — and they’re playing games with real businesses and real people.
6) An article in Saturday's WSJ, answering questions about HFT:
High-frequency trading, long an obscure corner of the market, has leapt into the spotlight this year. Wildly successful in 2008, high-frequency traders are the talk of Wall Street, attracting big bucks and some unwanted attention. Concerns that some traders are taking advantage of less fleet-footed investors has drawn the attention of regulators and members of Congress. The following is an explanation of the core issues, based on interviews with industry participants and regulators.
7) An article about BATS:
Each trading day, as a bell atop the M&I Bank building next door chimes gently, BATS quietly conducts about 25 times the volume of the venerable American Stock Exchange. Here, 1,200 miles from the financial center of the world, a few dozen employees pad around in shorts and polo shirts, amid green cubicles and whiteboards. On any given day, its servers off in New Jersey will process about 12 percent of the trades made in the vast U.S. markets. In less than 36 months, BATS (it stands for Better Alternative Trading System) has evolved from a start-up into an international stock exchange with powerful partners and a nine-figure valuation.
8) An FT article about HFT:
Both BATS and Direct Edge say any investor can participate in flashing prices or receiving them on their trading venues.
Any move to ban the practice, says Direct Edge, is seen as creating a two-tier market as it is likely to push business away from the main electronic platforms towards “dark pools”.
Still, there are concerns that market makers, such as high-frequency traders, cancel many of their flash orders before other investors can execute a trade. This can enable the market maker to come back and offer shares for sale at a higher price, or place a buy order at a lower level.
“Certain black-box models have cancellation rates as high as 99 per cent on orders,” says Paul Zubulake, senior analyst at Aite Group.Mr Arnuk believes liquidity rebates and flashing should be stopped in order to level the playing field for all investors.
9) Joe Saluzzi was on Bloomberg, clarifying his views on HFT: www.youtube.com/watch?v=iTkCNsnDgws
»
- 6890 reads
- Printer-friendly version
- Send to friend
- advertisements -
Summary:
The Fed is openly encouraging and enabling massive front-running of their own very large orders.
Let's put High Frequency Trading on the back burner, this article from the Financial Times points to a much bigger scandal (of which HFT can be a part (trading very liquid markets highly correlated to the markets for which "they" have information)).
From the FT article:
Wall Street profits from trades with the Fed
"However, the Fed is not a typical market player. In the interests of transparency, it often announces its intention to buy particular securities in advance. A former Fed official said this strategy enables banks to sell these securities to the Fed at an inflated price."
In other words, the Fed, the biggest player in the history of the financial markets (Alphabet soup of programs, Quantitative Easing, Regular FOMC actions, etc....), is transferring $10s of billions to Wall Street (and BlackRock and PIMCO, BTW), without this support hitting the bailout totals.
Right on #1 (in the original post)
We can't lose sight of the fact that HFT is a very small percentage of bank profits (maybe hundreds of millions, but relatively small).
Making fixed income trading more transparent would save taxpayers (pension funds, municipalities) billions of dollars in transaction costs.
Frontrunning in OTC fixed income markets makes the HFT 'scandal' look trivial.
I totally agree...this is why GS makes so much money in their fixed income trading unit. WHAT EVER HAPPENED TO THE CHINESE WALL? Prop and market making were supposed to be kept totally separate, not even on the same trading floor. This is the reason GS makes so much and this is what should be looked into......and yes, also, why is their no nasdaq type posting for every fixed income trade and bid/offer rather than going through a dealer making what is now a huge spread.
Maverick
I only care about HFT because its momo traders -- in commodities, FX, rates, fixed income, equities -- are driving up the price of risk assets, and if something very big and very bad happens, they'll try to exit at the same time, and prices will have to fall to the level at which vultures will buy. Wanna see things like senior loans with YTM and PTPs with distribution yields of over 20%?
Transparency in the bond market is important in order to prevent the Fed giving handouts. Ideally, we could shift as much as possible onto exchanges. For example, putting most corporate bond trading back onto the NYSE, like it was until the mid 1940's. Here is a good article tracing the history of how OTC took over bond trading.
http://idei.fr/doc/wp/2007/bondmarket.pdf
In addition to the history lesson, the paper also points out that institutional investors do well in an OTC market, and retail investors get screwed.
"In addition, on the OTC market, and especially for large blocks, institutions could negotiate the compensation of the intermediary. In contrast, on the exchange, commissions were regulated, and could not be negotiated. Furthermore, the professionalized management and relatively frequent presence in the market of institutions makes transparency less important to them than to less sophisticated small investors who trade infrequently. The repeated interaction that dealers and institutions have with each other renders them less vulnerable to the opportunities which a lack of transparency affords other participants to profit at their expense on a one-time basis. Smaller institutions and individuals, for the opposite reasons, will tend to fare better in an exchange-based trading regime."
Cry wolf too many times and everyone eventually becomes deaf.
thx.
I am posting an excerpt of a post by James Anderson from Minyanville's Buzz and Banter....
"The attached table and chart are the New York Stock Exchange Tick Indicator. Tick is defined in the following manner. If a stock trades up from its last trade, that is a positive tick. Down, it would be a negative tick. The Tick Indicator simply sums up all the positive and negative ticks on all the NYSE stocks and reports that number every 10 seconds. (At least my software reports it every 10 seconds.)
The table includes all reported ticks from 10:00 to 10:06 this morning. The column on the right is the difference from the tick reported 10 seconds earlier. The chart is the difference column. As you can see in this 7 minute period, the tick swings wildly back and forth from a high of 580 to a low of -318. Remember, those are 10 second swings.
Folks, It's not people doing this. This is HFT at work. Is it just noise? Does it really matter? Obviously it matters to the people doing it because they are extracting profits from somebody." .....
https://admin.minyanville.com/assets/FCK_May2009/File/00%20twoo/ja1.jpg
Now Tilson is going to "investigate" HFT? Wow, first he copied and parroted everything Einhorn and Ackman did, then he took everything Amherst Securities produced and made a book off it, and now he is covering HFT? Guess actually investing is why he's only up about 15-20% this year when other value funds are killing it.
Now Tilson is going to "investigate" HFT? Wow, first he copied and parroted everything Einhorn and Ackman did, then he took everything Amherst Securities produced and made a book off it, and now he is covering HFT? Guess actually investing is why he's only up about 15-20% this year when other value funds are killing it.
-----
Whitney has often been late to the show. In 2000, he was touting buy and hold investing at Motley Fool, and in 2003 and 2004, he was touting shorting tech at Motley Fool.
2003 re shorting tech: http://www.fool.com/investing/general/2003/11/07/how-to-bet-against-tech.aspx
2000 re buy and hold: http://www.fool.com/BoringPort/1999/BoringPort991227.htm
his buys in that 2000 article aren't bad. BRK.A has more than doubled since then, and APCC got taken over.
if your point is that sometimes WT is wrong, he is. good thing he's in a job where one only has to be right 55% of the time to make a killing.
The guy manages $100MM, a value oriented guy that is a disciple of WEB investing in BRK when u runs $100MM adn can invest in anything? he just copies Einhorn, Ackman, etc, every pick he has is a copy of their picks. He lost a nice chunk following Ackman into TGT and BGP and BKS. But when Whitney is so busy selling your newsletters and the value investment congress tickets and writing for Kiplingers about dog stocks the actual investing gets short shrift. But the good thing is he does have his defenders like you.
First the Dolly Parton cover, now this. When is Whitney Tilson going to get her career back on track? I blame Bobby Brown.
Regarding your 1st point about the fixed income market with issues moving 25%: if there was HFT in fixed income this problem would go away. Mean reversion quants would supply liquidity whenever large orders moved the price, effectively reducing the price swings. As more players compete they have to settle for smaller and smaller profits which means supplying the liquidity closer and closer to the original price which means smaller and smaller swings.
That's what has happened in the equity markets.
> "I only care about HFT because its momo traders -- in commodities, FX, rates, fixed
> income, equities -- are driving up the price of risk assets, and if something very
> big and very bad happens, they'll try to exit at the same time, and prices will have to fall
> to the level at which vultures will buy."
HFT is almost by definition different than the momentum behavior you're concerned about. Let's say a HFT algo places 1000 trades per second. The only way for it to contribute to upward momentum would be if they were all buy orders. If the algo places little $2K buy orders at that rate for a 60 seconds it will have consumed $120M of capital. Very quickly you run out of capital and trading has to stop until you start selling things (at which point the trades would cause downward pressure). The point is that over the long term and even the medium term, HFT algos are pretty market neutral and thus don't contribute to momentum.
So refreshing to have a bit of common sense on the subject.
you mean there will be no liquidity if "something bad" happens ? brilliant, that's what we actually anticipated for so long
I said nothing in my previous comment to imply liquidity drying up. But to address this canard:
Because the HFT algos are relatively market neutral, if "something bad" happens they would not be pushed to the sidelines by massive losses.
In theory and in practice (so far), these algos do better when volatility is high. So in the case of a market dispruption these traders would want to keep their algos running and adding liquidity (while watching their screens *very* closely though).
And by the way, "something bad" did just happen - worst market crash since the depression, and liquidity did not dry up.
Right on about Tilson. The guy hasn't had an original thought since....well, I'm not actually sure when. He's great at jumping on the bandwagon and has a big audience, but there is nothing original there.