The government’s new white paper and the devolution settlement: pulling bricks from the Jenga tower

One central question for any federal or devolved system that gives wide law-making powers to states or provinces, is what to do about legislation in one state that throws up internal barriers with another. This could be law that affects trade, or makes it hard for businesses in other states to compete fairly. If one of the aims of that federal system is to allow businesses, consumers, and workers to trade and move freely between different states—which is what an “internal market” means—then the answer to the central question cannot be “nothing.”

But when the UK devolution settlement was drawn up, that question could safely be ducked. That was because EU law effectively did the job, regulating the extent to which the devolved governments could impose standards which excluded competing goods or services from elsewhere in the EU, and their ability to use subsidies to give a leg-up to domestic industry facing competition.

But EU law is now an outgoing tide, and the question of how the UK devolution settlement should deal with obstacles to the UK internal market can no longer be ducked.

It is important to appreciate just how wide the issues here go. To take some examples:

- Would a ban by state X on possessing more than a certain quantity of crayfish (because of concerns about crayfish stocks in that state) be acceptable, or would it be regarded as an impermissible restriction on the import of crayfish from state Y?

- Would a state X rule providing that blackberry spirit must be sold at a minimum price be acceptable, or would it be regarded as an impermissible restriction on the import of cheap blackberry spirit from state Y?

- Would a rule in state X preventing anyone from opening an office and calling themselves “solicitors” unless they were on the roll of solicitors in that state be acceptable, or would it be regarded as an impermissible restriction on the ability of solicitors from state Y to set up shop in state X?

- Would a rule in state X requiring a 50p deposit per bottle on sales of beer in non-refillable bottles be acceptable, in circumstances where most beer in that state was produced in refillable bottles but most beer in state Y was not, or would it be an impermissible restriction on imports of beer from state Y?

- Would a tax in state X on retailers providing for a much lower rate for smaller shops (mainly owned by state X residents) than large ones (often multinational or owned by people in state Y), be permissible, or an impermissible subsidy for small shops?

Those familiar with EU law will probably recognise the second, third and fifth examples (see the string of cases on minimum alcohol pricing—mixed; Gebhard on restriction on lawyers—impermissible; and Commission v Poland on lower taxes for small shops—permissible). But the first and fourth examples may be less familiar: they come from Australia, where they are relatively well-known cases under section 92 of the Australian Constitution, which provides that “trade, commerce, and intercourse among the States … shall be absolutely free.” That shows that the tension between the ability of states to regulate as they wish, and the aim of creating an internal market, is not just an EU issue—it is a systemic issue, to which different systems have given somewhat different answers.

What the examples given above all show, though, is that any restriction on states’ or devolved governments’ ability to legislate in the name of preserving an internal market is likely to be significant, potentially impacting a wide range of important policies where they want to be able to do their own thing (conservation, public health, consumer protection, tax, environmental protection and so on). The restriction cannot sensibly be limited to measures that expressly treat inter-state trade differently from domestic trade: such a restriction could too easily be circumvented by clever wheezes (just as the law of discrimination cannot limit itself to overt or direct discrimination but has to address measures that look neutral on their face but are discriminatory in effect).

But when it comes to measures that are ostensibly neutral but in practice make inter-state trade more difficult, there has to be a mechanism for deciding if the measures do sufficiently impact inter-state trade to be caught and, if so, whether they are nonetheless justified in policy terms (for example, whether other forms of regulation that would restrict inter-state trade less would be as effective, and whether the regulation is a reasonable way of achieving the policy objective). Precisely how that restriction is formulated and interpreted also matters: indeed, the Australian crayfish case—Cole v Whitfield (1988)—marked a sea change in the Australian courts’ approach to section 92, requiring claimants to show that the contested measure could be characterised as “protectionist,” resulting in a marked weakening of the impact of the restriction.

In proposing to restrict devolved governments’ powers to legislate in ways that restrict the UK internal market, the Internal Market White Paper is therefore sailing into choppy waters. The proposed restriction could well amount to a very major incursion into the areas over which the devolution governments currently have power to legislate.

The current government’s answer to that objection is to remind everyone that all the devolved governments were constrained—both as a matter of EU law and expressly under the devolution settlement—not to legislate in ways that infringed EU law on the four freedoms and state aid. The white paper, it is said, is not therefore taking away any real power that they currently have—and to the extent that the UK regime on, for example, subsidies is less restrictive than EU law, confers more powers on them.

That answer, though, misses basic points that flow from the asymmetric nature of the devolution settlement. That asymmetry is that the devolved governments are ultimately bound by the Westminster legislation that created them, including the Westminster legislation that will impose whatever internal market restrictions are formulated (overriding, if necessary, the non-binding convention that Westminster will not legislate to restrict the competence of the devolved legislatures without their consent). The “English” legislator, though, is the UK Parliament: and the UK Parliament (in which MPs from the devolved nations are only a small minority) is truly sovereign. Moreover, the “federal” UK government is not (as the EU institutions are or the Australian government is) in principle neutral between the different states: it is also the English government, and so, inevitably, speaks for England, and acts in the interests of England—especially when, as now, the majority party holds few seats outside England.

From the point of view of the devolved governments, therefore, the possibility of restrictions on their powers being imposed by Westminster looks like being told to play a game where the strongest player writes the rules, is free to ignore the rules when it wishes to, and is the umpire who decides whether the rules have been infringed. That does not look like an attractive game to play. In contrast, in the EU, the rules were written externally (albeit with some UK input), the UK/English government was equally bound by EU law, and the EU institutions were entirely independent of Westminster.

That fundamental problem bedevils both the main principles set out in the white paper. To be acceptable, mutual recognition—ie accepting that products can be sold, or businesses can trade, if they have received relevant regulatory approvals in the other jurisdiction—involves either enormous trust (trust which is somewhat lacking as between the devolved and Westminster governments) or a binding agreement on common minimum standards (but no agreement can bind Westminster, and Westminster can ultimately impose minimum standards if no agreement is reached).

Similarly, as I observed above, any prohibition of discrimination ultimately requires a binding mechanism for deciding whether a measure does in effect discriminate and, if it does, whether it is nonetheless acceptable for good policy reasons. Such a mechanism could be a court applying a test resembling that laid down in section 92 of the Australian constitution or Articles 28-34 of the Treaty on the Functioning of the EU, or it could be a legislative mechanism setting out what regulation in a particular area is permitted or required, such as is found in innumerable pieces of EU legislation. But, however it works, any such mechanism in the UK context suffers from the structural defect that the UK Parliament, acting as the sovereign legislature for both the UK and England, can override it all.

What, then, could be done? Under the current devolution settlement, it is impossible to avoid the point that, when the chips are down, the Westminster Parliament will have its way. But a lot could be done to ensure that the chips stay up. There could be a commitment that any general restriction on the powers of devolved governments will be agreed unanimously. The legislation could make it clear (perhaps by borrowing wording from section 2 of the European Communities Act 1972) that Westminster legislation was to have effect subject to the same restriction unless expressly stated to override it (or, at least, to be interpreted if at all possible so as to comply with it and with a power for the courts to declare it incompatible with the restriction, along the model of the Human Rights Act 1998).  A framework for unanimous agreement of any system of mutual recognition could be set up. On subsidies, if (which at the moment is entirely unclear) there is to be any regime with teeth after the end of 2020, the regulator could be appointed by, and policy frameworks within which the regulator is directed to act could be set by, all four governments acting jointly—a measure that would have the incidental but important benefit of strengthening the independence of the regulator.

In the end, though, all those possible ways forward depend on self-restraint and a willingness to build consensus by a government with a substantial majority in the House of Commons. Little in the white paper suggests either is likely: it looks more likely that the current government will impose a very substantial reduction in the powers of the devolved governments against their will—and in a rush in order to get the new system in place by 1st January 2021. In doing so, the current government is acting like the player of the game of Jenga who decides to have a go at pulling out a brick at the bottom of the tower: the risk is that the tower will collapse. If it does, the question will be whether that collapse takes the form of Scottish independence or a consensus that the current asymmetric model of devolution simply cannot stand, and needs to be replaced by a properly federal structure involving England as well as the other nations of the United Kingdom.