The last global trade deal was the Uruguay round, finally agreed in 1994 after seven years of negotiations. The deal included the setting up of the WTO, a stronger organization than GATT, which it replaced. But no further global trade deal has been agreed. The WTO launched the Doha round  in 2001, but it has fittingly run into the sand.

Trade negotiators are nothing if not obstinate, and tried a new tack. If a global deal is too difficult, why not try regional ones? So TTIP,  the transatlantic deal, and TPP,  the Pacific one, were born. Well, conceived.

Both are moribund. Hollande has declared France’s opposition to TTIP in its current form, which is also under sustained attack in the European Parliament, especially over ISDS.  [Update 30/8: the French trade minister has called for the talks to be suspended. If this is a negotiating tactic, it’s reckless hardball - it would be very hard to walk back.] TPP is opposed by both Clinton and Trump. Obama still officially hopes to get TPP through the Senate in the lame duck session. (See supportive comment from Harold Pollack.) Do you credit this? McConnell has not shifted from his policy of Adullamite obstruction of every Obama proposal. Even if he allowed a vote, would senators really vote against the platforms of their parties, which accurately reflect a hostile public opinion?

This widespread failure of the trade liberalisation agenda is usually put down to a widespread turn in public opinion against free trade, now seen by many on both left and the populist right as a callous neoliberal plot to enrich capitalists at the expense of workers. (It is true that the compensatory support for workers who lost their jobs as a result of past agreements like NAFTA somehow failed to materialise.) Some trade advocates resort to the absurd argument that the failure of TTIP and TPP would put existing trade at risk. But there is very little support for proposals to roll back existing trade agreements, from NAFTA to Uruguay to the European single market. There is something in the trade negotiation process of these new deals that gets voters’ goat.

Let me nail up a thesis to the trade church door. Modern trade negotiations are illegitimate. In their current form they cannot possibly lead to a democratically acceptable result. That is why they are doomed to fail.

The argument has two parts.

First, diplomacy and its embodiment in international law has a problem of democratic legitimacy in general. Ever since Grotius, the international community has been seen as a society of sovereign states. (You and I are mere objects of international law, with the odd exception of pirates and the more recent exception of human rights.) When these states were represented by absolute monarchs like the Kings of Spain or France, there was no disjunction between domestic and international law: both were built on a free promise by the autocratic monarch. But when the state is less monolithic, and especially if it is democratic, tensions can and do arise between the domestic and international orders. (Since the United Provinces of the Netherlands were a notorious example of a fragmented, pluralistic state, I wonder if Grotius considered the problem. Enlightenment welcome.)

Bilateral diplomacy is still conducted today by diplomats and foreign ministers, representing the executive, and largely in secret as in 1648. Multilateral diplomacy – as with the Iran nuclear deal – is a bit more open, but not much. Making the results legitimate depends on national mechanisms that allow an approximation to democratic norms of governance. There are in broad terms three methods:

• political accountability (public statements of foreign ministers, legislative approval of treaties)
• transparency of process (publication of drafts, procedures defined in advance)
• participation by stakeholders.

I’m not trying to work out a full theory here, just to establish a presumption that a diplomatic process lacking these features is likely to lead to illegitimate results, while one that has them is likely to lead to legitimate ones. Now, how does this framework apply to the process of trade negotiations?

Today these are no longer about tariffs. After the Uruguay round, tariffs on manufactured goods fell to 3%-10%, though higher on agricultural produce.

So attention has shifted to regulatory barriers. TTIP and TPP do include clauses on the tariffs that are left, which blurs the argument; but all the fuss is about regulations that make access difficult, or otherwise hinder foreign-owned businesses.

Some of these regulations are narrowly directed at imports (an onerous and discriminatory régime of inspection in Asia for American beef, say). It is fair to assimilate these to covert tariffs. More are standard measures of public policy, like patent protection for biological pharmaceuticals (an issue in TPP) or restrictions on private education (this has come up in TTIP). They reflect a balance drawn up within each country between the competing interests of consumers, incumbent providers, and potential providers. The balance is informed by general political preferences, say on the place of private versus public education. It is reached following a reasonably transparent and uniform political process, allowing stakeholders a voice and identifying decision-makers for accountability. All these factors contribute to making the apparatus of national regulation politically legitimate.

Legitimate does not mean sensible, let alone optimal. Regulatory capture is real. The recent growth of occupational licensing in many US states, extending to hairdressers and florists, offers a good example. A more serious one is the laxity of European standards on diesel engine emissions, revealed in the VW scandal – by American investigators. However, the stigmatising of regulations as such as obstacles to business is merely conservative propaganda. It disappears once the issue is the extension of state-created monopolies, as in intellectual property law.

Even in the case of hairdressers, there is a genuine though weak argument from the benefits to consumers: a failed hairdo is a social embarrassment, and nobody wants to sit in a chair in front of Sweeney Todd. It will be generally agreed that occupational regulation is essential for professions where incompetence can be fatal: doctors, gas installers, pilots, handlers of explosives. The spectrum from these to manicurists is long and there are many hard cases. Just how much regulation is needed for child-carers or roofers? The lower cost of solar panel installation in Germany and Australia than in the USA may well be linked to stronger systems of accreditation: picking a firm at random from the yellow pages is much less of a risk.

Schematically, the regulatory process for access to occupations has to deal with three interest groups, with six interests potentially in conflict:

• consumers, who want both low prices (low regulation) and an expectation of decent quality (higher regulation),
• existing providers, who want both low compliance burdens (low regulation) and high barriers to entry by competitors (high regulation),
• potential new providers, who generally want low barriers to entry (low regulation) – but if they are prepared to make the investment in training and accreditation, they may align their views with those of existing providers.

This complexity extends, I suggest, to most regulations on business. It is also typically impossible for third parties such as policymakers to assess the weight of these interests a priori. They have to find out, and this is normally done by allowing representatives of the affected groups to state their case.

We would therefore expect trade negotiations over regulations to be more problematic than those about tariffs. And so they are.

When trade negotiations were about tariffs, there was only one variable per commodity – the tariff rate, a price. Consider the simplest case: a bilateral negotiation between England and Portugal over the tariffs on cloth and wine (the example is a hat-tip to David Ricardo’s classic text on comparative advantage). Basically the negotiation is a bargain: a lower English tariff on wine in exchange for a lower Portuguese tariff on cloth. The interests involved are:

• Portuguese cloth producers and English wine producers (there are some today), who will lose from cheaper imports,
• Portuguese wine producers and cloth importers, and English cloth producers and wine importers, who will gain from more trade,
• consumers of both products in both countries, who will gain from lower prices and more choice.

Suppose, not unrealistically, that the trade negotiators on both sides are incorruptible Benthamite officials. They know in general terms about the consumer benefits. Their problem is to ensure that for their national capitalists and their workers as a whole, gains will exceed losses. For this, they will need to learn the elasticities, and the only source of this information is the businessmen. So they have to be invited to have a say. There is no need to bring in consumers; they benefit whatever the deal, and detailed information about consumer surplus is unnecessary. The interests of English wine drinkers are adequately represented by Portuguese wine exporters. The interests of workers are aligned here with those of their employing capitalists, and do not need separate representation. The process thus reflects adequate input from stakeholders. As far as transparency goes, it’s a haggle – a monopolistic seller against a monopsonistic buyer. Keeping your real intentions and reserve price secret is the only plausible strategy on both sides.

Trade negotiations have elaborated on this model but not changed it fundamentally. Trade negotiators are card players who keep their hands secret. They consult affected businesses systematically, and nobody else. The resulting deals are packages, and can be accepted or rejected en bloc by the principals, but not changed.

Now change the negotiation from one in which tariffs are exchanged for tariffs, to one in which regulations are swapped for regulations. (Modern trade deals like TPP and TTIP are giant bundles of such swaps.) As we have seen, regulations involve stakeholders in a different and more complex way.

Suppose (counterfactually) that one of the regulatory trades on offer is access to US job markets by Mexican plumbers, in exchange for better IP protection for US pharmaceuticals. At the very least, the consumers and safety watchdogs involved in the domestic US regulation of plumbers should have a seat at the table. Worse, the whole idea of the swap upsets the domestic legal and political order. Why should Texan plumbers face more job competition purely so as to benefit the stockholders of US pharmaceutical companies? The regulation – previously a state or local responsibility - has been federalised, and the secrecy has lowered transparency and accountability.

My contention is that conducting trade negotiations over business regulations in the way developed for tariffs is fatally flawed. It would need a radically different system, with true participation of stakeholders, transparency of process, and better political accountability.

You may say: this is Utopian. These requirements would make trade negotiations impossible.

I have two responses to this. First: freezing international trade rules in their current state would not be a disaster. The tariff reductions under the Uruguay round and previous negotiations have led to an enormous expansion of world trade, including substantial and contentious shifts in employment and wages. There is no pressing need for yet more trade. There are diminishing returns in gains from trade, as with most things.

Second, there is one example of a politically legitimate market-opening process. It is the European Single Market. When the newly democratic countries of Eastern Europe applied to join the EU, they were presented with a huge pile of the acquis communautaire, previously adopted legislation. At the time this was often cited as 50,000 pages, but other estimates come out much higher - up to 170,000 pages.  The applicants were not given the choice – all of it it had to be swallowed unchanged. Much of this was regulations for the single market. There aren’t British and French and German standards for washing machines or food labels (etc. x 1000) any more, just European ones.

In many ways, the process was radically different from standard trade negotiations.
1. Everybody understood from the outset, including Mrs. Thatcher, that the single market involved a wide transfer of sovereignty over business regulation. The scope of the transfer was pretty much agreed in advance.
2. The process was uniform and (just) comprehensible, with opportunities for input from stakeholders. These were better for businesses than for workers and consumers, but that’s true of national regulation too.
3. The process was not designed to facilitate arbitrage and bundling across different areas. I may be wrong, but I don’t think much took place. The regulations were adopted separately and at different times.
4. There was real and visible political accountability at two points: approval by the European Parliament, and by national governments in the Council. Most of the dumb ideas that fed the appetite of the British press for Euro-idiocies were proposals that died young, like the proposal by olive-oil producers to ban open bottles of olive oil in restaurants in favour of sealed containers.

So it can be done elsewhere. Will it? No. Regulation is an exercise of sovereignty. The European Union is a quasi-confederation, with its own political institutions, reflecting both democracy and a democratic deficit, to which elements of state sovereignty have been ceded by members. I can’t see any other regional grouping of states heading down the same path in the near future.

My guess is that regulation trading will either go back to being bilateral, or become sectoral on a global scale. The WTO could draw up treaties just on IP or food inspection or on education. With successive drafts, published for comment by all and sundry. That’s how the Paris Agreement was done, and the huge Law of the Sea treaty. Westphalian diplomacy can be carried out in a reasonably democratic way.

* * * * * *

Where did I go wrong? Feedback really welcome – I’m not temperamentally radical and am rather shocked by my own argument.

I have not emphasised ISDS in these trade deals, the lightning-rod for much criticism. It strikes me as a try-on by the well-paid little club of international commercial lawyers. It’s obvious overreach, and the whole wretched scheme could be jettisoned without damage to the rest of the trade negotiations, so it’s not really part of the wider failure of legitimacy. Companies can put compulsory arbitration clauses in contracts without ISDS. If they don’t trust a foreign court to respect these, they should stay out of the kitchen. Hatchet jobs on ISDS by Chief Justice Robert French of Australia and Professor Siegfried Broß, former judge on the German Constitutional Court. The argument from authority saves time.