Against a financial transactions tax

Maria Coelho, job market candidate at UC Berkeley, studied this topic and came up with this:

This paper analyzes the effect of the introduction of financial transaction taxes in equity markets in France and Italy in 2012 and 2013, respectively, on asset returns, trading volume and market volatility. Using two natural experiments in a difference-in-differences design, I identify bounds on elasticity estimates for three categories of avoidance channels: real substitution away from taxed assets, retiming (anticipation of transaction realizations and portfolio lock-in), and tax arbitrage (cross-platform and financial instrument shifting). I find large responses on all margins, that account for significantly lower revenues than projected. By far the strongest behavioral response comes from high-frequency trading lock-in on regulated exchanges, with a high tax elasticity of this type of turnover in the order of -9. The results shed light on overlooked features of optimal FTT design, suggesting they may be poor instruments for both revenue-raising and Pigouvian objectives.

The paper is called “Dodging Robin Hood.”  This is consistent with earlier findings on Sweden’s transactions tax, and that proposal continues to be one of the more overrated ideas in American Progressive political discourse.


Generally, reducing transaction volume of electronic traders is the win - raising revenues is kind of an ancillary perk.

Why is that good?

Because it's something that rich people do and I don't do; also I do not understand it. Therefore it is bad.

One often cited feature of the market is that it allocates capital in a "correct" distribution. This has gone bye bye with the destruction of the market as formerly understood. What is replacing it?


We could cartoon and equal and opposite position, that since I think I understand it and since "profit!" it is good.

Question: should Ponzi schemes be illegal?

After all we understand them, and there is money to be made.

Not impressed by this angle. Fear of the unknown is a real thing; blanket approval of the known is not. It's my opinion that Occupy was a Woodstock of well-meaning naifs who have since moved on to Bernie Sanders. Twenty-dollar minimum wages and luxury taxes and wealth taxes and trading taxes and a 90% marginal rate all sound great if you get your economic ideas from Facebook memes and the NYT Nobelist-in-residence.

What is the value to society of HFT? Isn't it just traders sneaking in between transactions to snag a fraction of a penny a billion times?

Why do you want to decrease market depth?

Why is additional market depth an unalloyed good?

Do we want significant resources devoted to profiting from small frictions in the capital markets that have no practical bearing on the functioning of the economy?

To play devils advocate, Because the rewards are funding strategies that can then be deployed on the rest of the economy once they're perfected and made more efficient. See: Jeff Bezos going from DE Shaw, to building a company vastly reliant upon (very) advanced statistical analysis applied to market and supply chain data... Wall Street has been doing things for awhile that are just now beginning to hit the real economy and be considered revolutionary. Furthermore, couldn't we argue that the type of telecom networks HFT firms are building out will become widespread and possibly an integral feature by the time they're perfected by these companies, of which no other activity had the ability to "self fund?" Could their activities become important for dealing with huge amounts of high frequency data in the future? I'm not sure but I feel like you're not taking this into account.

+1 the Tobin Tax is good, even if people try and avoid it, because electronic trading dries up during panics so no liquidity in fact provided, and the traders are parasites that legitimate buyers and sellers try and avoid (see Michael Lewis' Flash book)

Further, the paper is flawed since it assumes another exchange, Tobin-tax free, exists. However, if nations can get together, as they have with other taxes, (see Switzerland - US tax treaties), and impose a 'uniform' worldwide Tobin tax, this paper's conclusions will be moot.

+Why are you guys so pro-Wall Street? Everybody I know from there is an a-hole.

Weird - everyone I know who is anti-Wall Street is an ahole. My anecdote trumps yours.

I doubt it's necessary for new ideas on running businesses to originate on Wall Street. They didn't invent statistics or sophisticated data-gathering techniques.

Sure, using advanced techniques helps with market analysis and supply chain management, but I don't accept that they would not have been adopted unless they originated with investment firms.

B/c depth helps transactions occur

As to the second part of your statement - why not? If there's an inefficiency, why restrict opportunity to take it? Why should we have significant resources devoted to Instagramming white chicks and their pumpkin spice lattes?

"Why should we have significant resources devoted to Instagramming white chicks and their pumpkin spice lattes?" - we should not. If this is clogging up the internet, we should tax it.

"Depth helps transactions occur" - is false. During panics depth dies. When no panic plenty of depth but not needed, so can be safely Tobin-taxed if countries work in harmony to harmonize their taxes.

Sing after me: "Wall Street is bad music"

Tobin came to this conclusion years ago, and he favored the tax. Of course, as every intervention to avoid catastrophe are determined to be impossible to implement, we eventually come down to what's left, the market correction. Do you have tenure?

Why is pump and dump rent seking asset churn a good thing?

A FTT simply taxes out the rent seeking in asset trading leaving neither tax revenue nor pump and dump.

Capitalism is build on building assets, not on rent seeking requiring restrictions on building assets to generate monopoly profits from depreciating assets. No transaction tax is due when you build your assets instead of merely trading old decaying assets.

If a FTT increases the bid/ask spread, it creates a huge deadweight loss in the economy.

In the old days, stock brokers took a 1% commission on every sale and the minimum bid/ask spread on a highly traded security was 1/8 of a dollar. That increased the cost of capital substantially.

Now commissions are basically zero and bid/ask spreads rarely hit a penny.

One reason not to care about HFT is that at most is stealing pennies, but some tiny increased spread is only stealing pennies as well.

I agree that the best argument is the long term one, about effective or mal-investment.

Any reason to think a small transaction tax would keep people from growth investments?

Of course there is! Don't you remember what blog you are commenting on? Think at the margin!

OK, I don't think the mainstream would be affected by the marginal argument. They only buy stocks through two levels of abstraction (mutual funds in 401K accounts). I also don't think Warren Buffet is going to be put off an investment in railroads. So who are we left with? Theoretical rational investors operating at the margin, and trading furiously to create rational investments for year+10 economies?

What do we mean by "small"?

If the idea is to raise money from the financial industry -- and that could be a very laudable goal -- then a 1% additional income tax seems vastly more straightforward.

According to Wikipedia, the French "FTT levies a 0.2 percent tax on stock purchases of French publicly traded companies with a market value over €1 billion." While "Italy levies financial transaction tax on qualified equity transactions of up to 0.2% (0.22% in 2013) of the value of the trade"

Seriously? Mine the effect of a 0.2% tax out of a few years data?

Pull the other one.

also note that the average trading commission in the late 1980s was $45, today it is $7.


Financial transactions taxes are too clever by half. If we want to raise revenue, raise revenue! But the idea that you can change the behavior of fin mkts in useful ways by this kind of tax is not well thought out.

"Now commissions are basically zero and bid/ask spreads rarely hit a penny."
Is this why America's GDP is growingbfaster than in the 50's and 60's, right?

Also why America was so much more racist and sexist in the 50's and 60's.

And why General Eisenhower liked to play folf.

And we are so well-off thanks to all the speculation...

So you're saying we'd be better off w/o capital markets?

(This non sequitor game is fun!)

I did not know we did not have capital markets in the 50's (i.e. well before "now commissions are basically zero and bid/ask spreads rarely hit a penny) when the GDP grew faster and the whole population prospered. I did not realizd that taxing financial transactions (i.e. taxing the richest among us, but not taxing them nearly as much as they were taxed back then) would make the capital markets vanish. Do you have any other words of wisdom to share?

These are distinct problems.

We know how to do prudential financial regulation; we're just not doing it (despite the last decade). Financial transactions taxes are no substitute for that.

You want to tax the wealthiest, I'm with you, but keep it simple. Tax income and inheritance.

I would rather tax finance than productive work(ers).

Ever notice how everyone wants to tax the rich, but they never offer to end the tax exempt muni bonds?

And yet the 30s and 60s and 70s were so likely totally awesome - even with the same commission structure.

Do you have any logic to share - b/c you haven't shared any yet.

By your logic, we should restrict women from the workplace - in the 50s, few women worked professionally - the economy like totally rocked. QED. Amirite?

Why would a job market candidate argue otherwise?

Overall this seems a very small data set from which to draw conclusions.

This whole argument comes down to priors - whether one sees finance as parasitic or as productive. The parasitic view must hold two contrasting points of view simultaneously - that trading volume consists of (1) wily insiders who trade ahead of dumb money and (2) dumb money that is constantly nudged by finance insiders to overtrade in search of unattainable excess profits. The framing of "finance" as the insider making profit in either case is definitional. Finance is parasitic, by assumption. When capital takes a loss, as in the "dumb money" here, it is simply redefined as "not finance".

When moving outside this dualism, sophisticated folks tend to agree that active trading, on net, is not profitable after costs, and that passive investing is the preferred way to invest. Another way to say this is that all net aggregate profits from active trading are captured as a positive externality by passive investors. Taxing an activity whose entire output is captured as a positive externality is probably, shall we say, suboptimal.

Why stop with financial markets? Why not a financial transaction tax on all financial transactions - taking money from an ATM, buying a tank of gas, selling a painting at an auction, refi on your house...

Interesting. We like it when VISA does it, as opposed to a national digital currency. Rent seeking?

Perhaps - it'd certainly be one way to "go after the rich" - just tax all money movement the same - why should we make distinctions between financial transactions. Should be enough money moved around to have the tax be quite low - and think of all the monies we'd save by getting rid of most accounting departments.

I think you missed my point. Conservatives are generally happy with the Visa skim because they see it as the strength of the market. But if transaction taxes are bad we should seek to eliminate it. But whoops we can't trust the government to do that, because what do they have to do with money?

Basically it is hard for anyone who is for arcane financial systems to seriously oppose a 0.2% tax.

Buying a tank of gas and selling a painting are already taxable transactions (excise and capital gains tax, respectively). In countries with a VAT system, just about any sale of goods or services gets taxed.

In finance, the idea is to tax transactions over and above capital gains in order to discourage high-volume, short-term churning and speculation. Ever since the failure of Long Term Capital Management, we have had very strong evidence that highly liquid markets tend to be destabilizing and do not always drive prices toward fundamentals.

This would replace sales, VAT and all other taxes.

You're somewhat incorrect wrt LTCM - it's not the liquidity per se as much as it is overwhelming available liquidity. When everyone runs for the doors at once, the doors had better be pretty big. Bigger doors help more than smaller doors - but when the crowds trying to get thru them, there'll still be a bottleneck.

Additionally, prices do return to fundamentals - sometimes it just takes longer - that's all. Thus the old saying "the markets can remain irrational longer than you can remain solvent". Reference point for today - the shorts in KBIO

Indeed, why is it that this idea - taxing each and every transaction of any monetary kind - apparently earns so little wonkish attention? It's super simple and transparent, and "fair" while still being progressive in its way. It's easily understood as a social "service charge". is one advocacy site, but I've not seen it discussed much overall, even though it seems like a really good idea to me, with obvious benefits.

If people here know of "obvious" drawbacks, I'd like to hear about them, or even a diagnosis of why it doesn't receive more attention among tax proposals.

Comments for this post are closed