Europe’s robust financial-transactions tax

By Felix Salmon
January 30, 2013
Alex Barker has seen a draft, and it looks impressively robust.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

The details of Europe’s new financial transactions tax won’t be made public for a few weeks, but the FT’s Alex Barker has seen a draft, and it looks impressively robust. The tax is being implemented by 11 countries, including most importantly Germany and France, and it’s going to be levied at two levels: 0.1% on securities trades, and 0.01% on derivatives trades. It’s also going to be very difficult to dodge: any trader whose institutional headquarters is in one of the 11 countries will have to pay the tax, as will all transactions taking place in those countries, and all transactions involving securities issued in those countries.

The tax will have two main purposes. The first is to raise substantial tax revenues on the order of $45 billion per year; the second is to discourage financial speculation. I’m hopeful on the former, but less so on the latter.

As Robert Peston and Avinash Persaud pointed out back in 2011, financial transactions taxes work pretty well: even the UK, which is implacably opposed to the European tax and which won’t ever join such a scheme, levies a surprisingly large 0.5% tax whenever anybody — anywhere in the world — trades a UK stock. And yet, somehow, London remains the first choice for international companies looking for a place to list their shares.

The European tax, which is much smaller than UK stamp duty, will similarly have little effect on how and where financial markets operate. The “if you tax me, I’ll just move elsewhere” threat is a pretty empty one, in practice, especially if you have a carefully-drafted law which makes tax avoidance difficult, and if you’re talking about established financial institutions rather than individuals. Count me on the opposite side of Steve Slater: large banks won’t avoid this tax, because doing so would be both politically suicidal and practically incredibly complex. Especially in a world where tax laws can be changed quite quickly, if any obvious avoidance is noticed.

So I think that the financial transactions tax will actually be very good at raising money — possibly even good enough, over time, that the rest of the EU will come around to it. Maybe even London could decide to swap out its stamp duty and join a unified European system instead, especially if a less City-friendly government is elected in 2015. That’s one reason I like the idea of half the EU going ahead with this scheme while the other half stays out: it will provide a proof of concept to persuade the nay-sayers.

On the other hand, I doubt that speculators will find this tax particularly off-putting. Europe doesn’t suffer from the high-frequency trading that has overtaken the U.S. stock market, and these taxes are low enough that any remotely sensible financial transaction will remain sensible on a post-tax basis. It’s possible that total trading volume might decline a little bit in some markets, and that would be fine: no one thinks it’s too low at the moment, and in the derivatives markets especially, increase in volumes generally just translates into increased rents being paid to big sell-side banks. But I’m not someone who believes that speculators are causing a noticeable amount of harm in European markets: as far as they’re concerned, the financial transactions tax is likely to make very little difference to a group of people who are not much of a problem in the first place.

One of the themes in Davos this year was a series of EU politicians pushing pretty hard for a big EU-U.S. trade deal, while the U.S. seemed to feel much less urgency. The financial-transactions tax won’t help on that front: it will widen the gap even further with respect to the treatment of the crucial financial-services sector. Still, it’s good to see real leadership here from France and Germany. They’re going to implement a sensible tax, which will raise much-needed revenues at minimal societal cost. What’s more, if you pierce corporate veils to find out which individuals will end up paying the tax, it’s going to look a lot like a wealth tax, rather than an income tax. That’s good news, in a world where the wealthy tend to pay much lower taxes than those with high incomes.

So let’s hope that this tax gets introduced; that it works; and that the rest of the world, seeing the costs and the benefits, starts to follow suit and sign on too. The area covered by the initial 11 countries is big enough that the tax will work well at inception, but as more and more countries join the scheme, the tax will become increasingly efficient and effective. Maybe, eventually, it could even incorporate the U.S.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
7 comments so far

It’s not a tax on the financial sector.

“UK…levies a surprisingly large 0.5%” 71 percent of transactions in the UK are exempt from the tax. Only ignorant retail investors pay stamp duty, which is not the same as the new transaction tax. That 0.5% tax is also diluted down to approximately 5 basis points per transaction amongst all investors, even if they do not own shares…everyone pays, even money market funds.

The following Fun Facts, complete with sources, are from

The end-investor and consumer will pay the most of this tax. Note cumulative and cascading. The IMF’s FTT Final Report For The G-20, June 2010, “Its real burden may fall largely on final consumers rather than, as often seems to be supposed, earnings in the financial sector. Because it is levied on every transaction, the cumulative, ‘cascading’ effects of an FTT—tax being charged on values that reflect the payment of tax at earlier stages—can be significant and non-transparent.”

If the tax was Euro-wide: UK Parliament Economic Sub-Committee of the House of Lords, “The FTT is likely to induce a loss in GDP between five and 20 times larger than the revenues raised from the tax.” That means negative revenue for those that are counting. UK Parliament European Scrutiny Committee citing the EU Commission’s FTT Impact Assessment, before including negative relocation effects, “a 3.43% fall in EU GDP equates to a fall in economic output worth €421 (£362) billion and a 0.34% fall in employment equates to a loss of 812,000 jobs.”

“The DCB study showed that 1.7 billion euros, over 42% of the annual FTT cost in the Netherlands, would be borne by pensions

“this new tax would disproportionately impact pension funds and other institutions which provide retirement income.

“The Wellcome Trust, a charitable foundation with a £14bn ($22bn) investment portfolio, calculates an FTT would cost it £32m a year, equivalent to its 600-person strong programme in Kenya.

“The more we do to you, the less you seem to believe we are doing it.”
-Dr Josef Mengele

Posted by rhone | Report as abusive

>And yet, somehow, London remains the first choice for international companies looking for a place to list their shares.

That’s because they don’t care about prices being accurate.

What, exactly, qualifies as a sensible financial transaction in your mind? A portfolio with a 100% annual turnover will lose 40bp annually JUST to the increased spreads (no exemptions for market makers because that’d defeat the point). For comparison’s sake, my Schwab S&P500 ETF has annual fees of 4bp. A market order to buy/sell it, assuming no price impact, would cost me about 4bp, including commissions.

A 10bp FTT on SPY trades would increase its average spread by about 3000%. In what alternate universe does this seem like a good idea to anyone?

European markets already have very high transaction costs. Piling additional taxes, especially of such magnitude, on top of them will only serve to degrade market quality further. And it really only hits the most vulnerable people (by design of course, hence the lower taxes on derivatives).

Arbitrageurs can just take their money elsewhere, or stop trading. The only losers are the poor schmucks who are either forced to invest in these markets or don’t know better: the average European retail investor.

Posted by qusma | Report as abusive

This is a great idea and the U.S. should consider a variation. Specifically, how about a tax on business interest paid to financial institutions? That’s right, a tax on an expense, rather than on income.

Outrageous? Radical? Not exactly: For decades the U.S. has taxed certain business expenses. In particular, payroll tax is nothing but a tax on a specific business expense: labor.

Think about it: businesses are taxed on labor expense, but not on capital expense. Why not both?

Lest you argue that this would disincentivize (is that a word?) businesses from borrowing money and collapse the financial industry, think again. Does payroll tax make businesses decide not to hire people? Of course not.

You can imagine a proposal like this getting the financial lobby on K Street out in force. But here’s the question nobody has answered yet: why is it OK to tax labor expense, but not capital expense?

There is one huge benefit from a proposal like this: companies will be unable to park billions of dollars overseas to avoid the tax. And corporations might actually pay more in U.S. taxes than to their CEOs. How radical is that? :)

Posted by WilliamCowie | Report as abusive

“levies a surprisingly large 0.5% tax whenever anybody — anywhere in the world — trades a UK stock.”.

This really isn’t true without massive caveats. “Anybody” in this context doesn’t include market-makers and other recognised intermediaries. Also, “anywhere in the world” isn’t really true – trading in ADRs and other depositary receipts outside the UK isn’t subject to SDRT, which is why there’s a one-off charge of 1.5% on transferring a stock into a depositary receipt program. And “UK stock” in context means “stock where title is held in the UK”. If I incorporate my UK company in Delaware and list it on the NASDAQ, then transactions aren’t subject to SDRT, even if the entire actual business is UK based.

Posted by dsquared | Report as abusive

“If the tax was Euro-wide: UK Parliament Economic Sub-Committee of the House of Lords, “The FTT is likely to induce a loss in GDP between five and 20 times larger than the revenues raised from the tax.” That means negative revenue for those that are counting.”

It’s nice to see that that committee actually took note of my evidence to them.

And the original source is the EU Commission’s own report into the FTT.

And I do wish that people would grasp the effects of stamp duty before praising it. The IFS looked into the incidence of it and found that the two major groups actually bearing the tax were pensioners (reduced returns on pension funds) and the workers (the tax raises the cost of capital to corporations, reduces investment and thus reduces wages).

Finally, what makes Felix so sure that the FTT will “work”?

One of the major claims is that increased speculation increases price volatility. This is something we can actually check over the next few years. Will price volatility in instruments subject to the tax increase or decrease?

Posted by TimWorstall | Report as abusive

Insofar as there’s a problem with high-frequency trading that I’d like to address, a lot of it doesn’t result in executions, and therefore wouldn’t be directly affected by this. We need a microtax on orders, or perhaps better yet on order cancellations.

Posted by dWj | Report as abusive

Thinking that this wouldn’t help the big banks is optimistic. Assuming the law would be changed to close such a loophole is just naive.

Posted by MorgantownJoe | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see