The idea for a generalized Tobin tax (financial transactions tax – FTT) on financial transactions continues, slowly but steadily, to gather supporters.

For a long time it was mainly backed by starry-eyed development NGOs who wanted to tax currency speculators (as in Tobin’s original proposal), but to fund a UN war on poverty. Tobin dissociated himself from this linkage. Either way, Tobin tax v.1 didn’t get anywhere.

It took the current financial crisis to revive it in the generalised v.2 form. Nicolas Sarkozy was SFIK the first political heavyweight to back an FTT. Gordon Brown supported him in 2009. The European Commission has tabled a proposal for an EU FTT tax. The IMF is sitting on the fence. Now Sarkozy has been pushing the idea at the G-20, though there’s no mention in the Cannes communique.

The latest converts, according to press reports from the G-20 conclave at Cannes, are Argentina, Brazil, Spain and South Africa. It’s unclear to me why any of them were against it in the first place, given the ideologies of their governments, no friends of Wall Street. Perhaps successful lobbying by local bankers on an issue their governments didn’t take seriously before, and now do.

What are Sarkozy’s chances of pulling this off? Not great, but not zero. The USA are opposed but not fighting it for others – it’ll just attract business to Wall Street. The UK, whose economic policy is now mainly run by Tory politicians whose non-political work experience has been as PR flacks to the City, is vehemently against, unless the tax is introduced worldwide, i.e. never. Japan, Canada, Australia, and China make up the club of noes.

Sarkozy’s best bet seems then to introduce it as a eurozone tax, marketed as a way of paying for the bailouts. It would need London’s cooperation on cross-border transactions. British opposition is so obviously a defence of the City’s interests rather than the country’s that it may be vulnerable to a shift in public opinion – in turn empowering Lib Dem coalition ministers like Vince Cable – and to street anti-banker protests.

Punishing bankers is such good fun and a necessary catharsis that it’s easy to forget that you need to be clear about the objectives. Tobin’s original idea, following Keynes, was a tax on foreign currency operations only, intended to reduce the volatility of exchange rates and the risk of speculative runs like Soros’ billion-dollar bet against sterling in 1992. With most exchange rates freely floating, and the disappearance of most European currencies into the euro, this specific problem has faded. Speculators like Soros are free to try their luck against the Bank of China’s administered exchange rate, backed by $3.2 trn in foreign exchange reserves; an opportunity which curiously appears unattractive.

The new FTT is mainly marketed today as a way of making the finance industry pay for the crisis it unleashed on us in 2008 and to insure against future ones. It’s not clear that a transaction tax is better here than a levy on assets or profits, which are much easier to collect. (To be fair to Cameron’s government, it has introduced a bank levy in the UK).

Joseph Stiglitz has endorsed a global FTT in statements to the press . I can’t find a formal paper or speech by him giving his reasoning – bleg anybody?

So here goes: bloggers rush where angels &c. The argument has to be that wholesale financial markets are too liquid for their own good. In pursuit of tiny arbitrage profits per dollar, financial firms generate colossal volumes of trades and pile up huge balance sheets of claims against each other. British banks, according to the Vickers report, have £6 trn in assets, of which only a third represents lending to British households and non-financial businesses. The rest is mostly lending in a circle to other banks and financial companies.

What for? This lending greatly exceeds the minimum liquidity required by say insurance companies. A lot, as we Main Streeters painfully discovered in 2008, goes to leverage both sides of huge zero-sum bets on derivatives within the finance club. At best that increases volatility, at worst increases counterparty risk when a player cannot meet a losing bet, and in any case reduces transparency, at some of these bets are incomprehensible even to those who make them.

The counterparty risks of this merry-go-round are not properly priced, and conceivably can’t be; they are assumed to be low until a chaotic change in sentiment makes for a sudden spike, and markets in particular securities cease to exist. Shifting risk into specialised CDSs and the like merely moves the counterparty problem, it doesn’t solve it.

In all these ways a hypertrophied financial sector doesn’t – as we’ve seen – reallocate risk efficiently as it claims, it manufactures risk and volatility. To contain systemic risk, regulators now insist on greater capital reserves – this is real capital which is sucked out of more genuinely productive sectors of the economy, and in the aftermath of crises, often comes directly from taxpayers.

If this deeply sceptical view of the current activities of financial corporations is correct, then raising their transactions costs would be a Pareto improvement. Compare the dermal denticles of sharks, the dimples on golf balls, and the fine grooves on fighter fuselages, which counter-intuitively improve their hydrodynamics. A little roughness often makes things go smoother. An FTT also represents a market alternative to heavy-handed regulation, always at risk of being captured or outmanoeuvred by financiers whose intelligence is as great as their scruples are low.

Since the plausible rate (a fraction of one percent) would be irrelevant to a transaction with a non-financial entity like you and me and Apple Corp., the tax would shift the incentives of financiers back to providing services to customers and away from incestuous deal-making. With luck it would make the industry less obscenely profitable, more boring, and less attractive to the best young minds coming out of our universities.

So for now I hope Sarkozy succeeds.

PS: another idea for banker-bashing, served half-baked for commenters to cook or bin. The enforcement of private contracts by the courts is seen as an unconditional right. Should it be? Why should secret contracts for large sums, often the mechanisms for Adam Smith’s conspiracies against the public, enjoy the privilege of enforcement by the state?

An efficient market requires transparency: participants need to know what deals others are making, so as to adjust their own behaviour rapidly. Well-regulated securities and commodities markets already approach this ideal: think of an open outcry trading pit. So my impeccably neoclassical free-market idea is to make any contract for over say $1 million only legally enforceable if it (or perhaps an accurate précis) is published. Anybody would still be free to make secret deals relying on non-binding private arbitration: we’d quickly find out how popular that would be. Vampire squid won’t like the idea; they rely on clouds of ink.