Wednesday, December 09, 2009

Malkiel on the transactions tax 

Thirty-two years ago, I sat in a lecture hall as a freshman at Princeton University and took in the wisdom if Prof. Burt Malkiel, as he explained the intricacies of inelastic demand curves, the relationship between savings and investment (S=I, in theory), and other important foundational concepts in his Economics 101 course. I am still predisposed to listen to what he has to say.

Prof. Malkiel co-authored an op-ed piece in the Wall Street Journal explaining why the "Let Wall Street Pay for the Restoration of Main Street Act" is a bad idea. The proposed act would impose a 0.25% transactions tax on all stock sales and some derivatives sales.
A small increase in trading costs would, according to supporters, be a manageable burden, and one borne by the speculators who the bill's authors apparently believe (to judge by the bill's name) created the financial mess. Across the pond, Prime Minister Gordon Brown of Britain has supported the idea as a way to take the burden off taxpayers during a time of financial crisis. In reality, the tax would deal a poorly-timed blow to long-term investors everywhere.

Proponents of a transactions tax misunderstand the way markets work. The bubble in home prices in the United States was not caused by the rapid buying and selling of individual family homes. The financial crisis was primarily a liquidity crisis and a credit crunch, and the major problem with collateralized mortgage-backed bonds was that they declined significantly in value and became illiquid. A transactions tax that would have reduced trading and made repurchase agreements more costly, could have made the problem even worse.

Moreover, "Wall Street" would not foot the bill for the presumed $150 billion tax. In fact, the tax would simply be added to the cost of doing business, burdening all investors, including 401(k) plans, IRAs and mutual funds.
Of course the tax would be passed along to the consumer; in this case the consumer is an investor engaging in a transaction. Wouldn't it be helpful to refrain from passing new taxes on investments, since, you know, we want to stimulate investment and liquidity so that the economy has a better chance of achieving robust growth? Prof. Malkiel and his co-author, Vanguard Group's George Sauter, conclude:
The transactions tax would gravely wound financial markets. It is hard to imagine a piece of legislation that would have more damaging unintended consequences.
I hope Prof. Malkiel is correct in his use of the word "unintended" -- it would be disturbing indeed if Rep. Fazio (D-OR), the lead House sponsor of the bill, wanted the consequences to occur.


By Blogger Purple Avenger, at Wed Dec 09, 02:42:00 AM:

1/4% on a $1,000,000 trade is large enough that it will start to pay to privately trade paper stock certificates again.

A large fund might save itself in the range of a million bucks a year moving paper certificates.

All that's needed is an escrow company to make the physical exchange of paper for money.  

By Anonymous Anonymous, at Wed Dec 09, 04:59:00 AM:

I'm not sure if this was the bill in question, but I recently saw a proposal in Congress that contained a provision outlawing bearer shares. I am certain that paper trading would run afoul of that bill, were it to become law.

One thing I haven't figured out yet: does the tax apply to shares that aren't publicly traded?  

By Blogger Andrew Hofer, at Wed Dec 09, 06:22:00 AM:

Paper shares can, and usually are, registered, but it obviously increases the frictional costs of transfer.  

By Anonymous Anonymous, at Wed Dec 09, 08:56:00 AM:

Malkiel is wrong, and his piece is nothing more than a shill for Wall Street, particularly Goldman Sachs.

Out here in the real world, where SOMEBODY has to pay to clean up the mess generated by the Wall Street bailout, a transaction tax is an excellent idea. It's primary result will be to kill, or at least seriously wound, the high frequency trading market. This is a good thing. At the very least, nobody has shown that this is a bad thing.

As referenced in the article, statistics indicate that HFT now accounts for 70% of the market activity. Quite literally, 70% of our stock market activity is now computer-based algorithmic trading between a small number of computers seeking arbitrage opportunities as they trade stocks among one another.

These players have significant unfair advantages over other market participants in both information access and timing. Further, there is significant evidence of HFT "pump and dump" activity in stocks, the perfect computer-based analogue to the pump and dump schemes of the 1920s. Finally, there is evidence for the conspiracy theory that the Fed is colluding with HFT traders to artificially drive up stock prices to give the appearance of economic recovery.

In short, HFT is unfair and prone to manipulation. The only defense HFT houses can offer is that they are providing much needed liquidity. Pardon me? I've been buying and selling stocks for 25 years. I witnessed no shortage of liquidity before the advent of HFT. At the very least, the burden should be on HFT houses to show that the markets were illiquid before they rode on to the scene.

In the last quarter, Goldman's HFT prop desk earned roughly $100 million PER DAY on HFT, and lost money only ONE trading day in the quarter. Ask anybody who has been in the trading business if this type of track record is possible in an honest market and they'll laugh in your face. The best traders hope to win 55-60% of the time.

Apply Occam's razor and the inescapable conclusion is that the HFT market is rigged. Taxing the HFT market may gravely wound Goldman's bottom line, but there is no evidence that it will harm the average investor.

I hope Congress moves forward with this tax, but I doubt they will. If they don't your taxes will go up even more. So it is quite literally you vs. Goldman Sachs on the hill. Guess who will win that one?  

By Blogger Dawnfire82, at Wed Dec 09, 09:10:00 AM:

Because the obvious response to success and innovation in the marketplace is to punish it.  

By Anonymous Anonymous, at Wed Dec 09, 09:16:00 AM:

As a follow up to my previous comment, this is an issue that is going to separate the Country Club Republican crowd from the Free Market Republican crowd. I'm solidly in the Free Market camp.

If the Republican party can't get behind a proposal like this then I see little future for the Republican party. This will confirm the conventional perception that the Republican party is party of the rich and powerful elite.  

By Blogger Dawnfire82, at Wed Dec 09, 12:58:00 PM:

Confirm, hmm? Because no one in the Democratic Party is rich or powerful.

"If the Republican party can't get behind a proposal like this then I see little future for the Republican party."

Hyperbole. Whether or not to slap a penalty on certain kinds of transactions is hardly a make-or-break issue for a political party, much less one in a first past the post electoral system, especially for one that is currently a superminority. As long as there is an establishment to hate, there will be an opposition.

It's also hard to sell (as I intimated before) a penalty on an innovation that gives a market advantage (which will simply get passed onto marketplace prices) as 'pro-free market.' I don't buy it, for instance. Free markets are supposed to encourage exactly that kind of innovation through the crucible of competition.

How could a policy which penalizes new ways of business (which apparently, according to your own account, make up 70% of trading now) and encourages going back to paper trading be described as anything other than regressive?

Punitive (in a literal sense; to act as a disincentive) taxation and regulation are enemies of free market movements. You'd have to make a damn fine argument to convince me (and others, I'm sure) that such a proposal, put forth by statist and regulation-happy Democrats by the way, is *really* a 'pro-free market policy that all Real Republicans should support, or else their party will die.'  

By Blogger Escort81, at Wed Dec 09, 01:16:00 PM:

Anon 8:56 & 9:16 - why would a senior Vanguard guy co-shill for GS? I get that you don't like HFTs, but what about the other part of the market that is not oriented that way? Why penalize that? If you don't like GS, maybe you can persuade Fazio to come up with a special tax on MDs levied against all bonus income over $1 mil. Wait, that's already part of the Senate health care bill for everybody over that income level, but it's only a 5.4% surtax. Maybe for GS people it could be 54%, but M&A people are exempted from the higher rate.

I respect that you are in the free market camp, but why not regulate the HFT market to "un-rig" it to your satisfaction (I realize that is pretty complicated to do), rather than taxing all stock transactions?  

By Anonymous Anonymous, at Wed Dec 09, 03:33:00 PM:

@Escort81: My comment was directed toward Malkiel, who is an academic. I am unaware that he is affiliated with Vanguard.

Without question a 0.25% transaction tax will increase costs for mutual fund companies, and by raising the transaction costs of trading would probably make the stock market slightly "sticker."

Noticably missing from the article was any estimate of how this tax would affect Vanguard or other mutual funds. I suspect this is because the increase in transaction costs would be minimal for most funds.

My post focused on HFT because their op-ed focused on HFT. This tax is targeted directly at the HFT market. My point was to rebut their argument, which appears to be that taxing HFT transactions will somehow "gravely wound financial markets." This is rubbish and they know it. The stock markets were perfectly liquid for decades before HFT came on the scene and they will remain perfectly liquid if HFT vanishes.

Anyone with even a cursory understanding of HFT gets that HFT does NOT, by its very nature, contribute in any significant way to market liquidity. HFT traders typically hold stocks for a period of milliseconds. They don't buy a stock unless they already have a seller on whom they can dump it. They just use preferred access to information, closed trading rooms, preferred trading positions (orders are NOT necessarily filled on a FIFO basis) and ultra-fast computers positioned across the street from the exchange to scoop up stocks and flip them before other exchange participants' computers can match buy/sell transactions.

100% of these transactions would clear anyway, usually within 1 second of the HFT transaction. However, to use an example, in that 1 second between the time in which your broker posts your order for 1000 shares of Acme at 10.50 and closes the transaction at 10.50, those 1000 shares might be passed from the original seller, who sold at 10.25 and round-robbined between a handful of traders 24 times, at a gain of a penny each time. Your single stock trade of 100- shares resulted in a volume of 25000 trades on the exchange.

The HFT transaction(s) ADD NO VALUE WHATSOEVER. It simply took the 25 cent spread from the seller and put it into the pockets of the HFT traders. Your transaction might close 100 milliseconds faster than it otherwise would have, but the seller lost $250 in the process.

Taxing derivatives is an even better idea. Hopefully it will force companies, banks, and funds to bring these liabilities onto their books and into the light of investor scrutiny. Much of financial mess we are in would have been avoided if risks from derivatives had been on the books.

(Note: the transaction tax could be targeted specifically at HFT by applying a minimum holding period. Anything longer than about 1 second would target the tax directly at HFT trading. I think it is telling that Makliel did not draw this distinction in the editorial.)

@Dawnfire82: it is apparent from your post that you know nothing about the HFT market. I'm a free market advocate. HFT is NOT free-market trading. Rather, it capitalizes on special trading positions granted only to certain market participants. Read up a bit and come back if you have something intelligent to contribute.  

By Blogger gs, at Wed Dec 09, 05:48:00 PM:

In the last quarter, Goldman's HFT prop desk earned roughly $100 million PER DAY on HFT, and lost money only ONE trading day in the quarter. Ask anybody who has been in the trading business if this type of track record is possible in an honest market and they'll laugh in your face. The best traders hope to win 55-60% of the time.

I'll provisionally accept the above at face value.

Suppose a trader makes or loses an equal amount of money on each trade. Take his winning percentage as 55%. Suppose he trades once a minute. Since the market is open from 0930 to 1600, that's about 400 trades per day.

I calculate his probability of a net loss on a given day as about 2%. About one day in fifty should show a net loss. Caveat: my algebra has always been wobbly. Trust me, but verify...

At (roughly) five trading days in a week, four weeks in a month, three months in a quarter, there are sixty trading days in a quarter.

Goldman's one losing day in a quarter is roughly consistent with what a very good trader would achieve by turning over a position once a minute: once a minute, not once a second or less.  

By Anonymous Anonymous, at Wed Dec 09, 08:53:00 PM:

Hopefully it will force companies, banks, and funds to bring these liabilities onto their books and into the light of investor scrutiny. Much of financial mess we are in would have been avoided if risks from derivatives had been on the books.

I agree, how dare those companies act like the US Government.  

Post a Comment

This page is powered by Blogger. Isn't yours?