Supra-national authorities, such as the EU, should do two things. They should restrain governments from over-stepping their natural authority. Second, they should provide ‘public goods’ that might not be efficiently provided by nation states. These might include defence or dealing with cross-border environmental problems. The second of these functions appears to be generally better performed through international agreements such as NATO. Certainly, the EU appears to contribute little to solving international foreign policy problems.

So, let’s concentrate on the first function of the EU. By promoting the free movement of goods, services, people and capital, the EU locks governments into a semi-permanent agreement, which prevents them undermining the freedoms of their own citizens. For example, there have been huge benefits from the free movement of labour, both in the form of permanent migration and also in relation to people moving around to do business. This is much more in evidence on continental Europe than in Britain, of course.

Some Conservatives might argue that this undermines countries’ democratic rights. However, all governments operate under constraints designed to prevent them from wielding unreasonable power over their citizens. But, there is a problem. Many saw the creation of the single market as an important milestone on the way to promoting free trade. In fact, Ted Heath called this correctly when he argued that the single market was never meant to be a free market. The single market is best thought of as a ‘single regulatory zone’. And, once a layer of government (the EU in this case) is given the authority to regulate business in the name of promoting free trade, almost any transfer of regulatory authority is deemed legitimate. This is how we have created the regulatory monster.

There is no intrinsic reason why companies operating under different regulatory systems cannot trade financial services freely. Similarly, there is no reason why all companies need to follow the same accounting standards throughout the EU or issue the same forms of prospectus when they come to market, or why stock markets should have to follow EU directives such as the Markets in Financial Instruments Directive. Indeed, many of these things were hardly regulated at all by government in the UK until 1986. However, the argument goes that, in order to have a ‘single’ market, all financial regulation must be the same: this lowers transactions costs and creates a level playing field. But centralising regulation at the EU level also prevents regulatory competition and centralises authority at a level of government that is totally unaccountable; it fossilises regulation; it ensures that processes can be captured by interest groups (mainly big firms and regulators); and it prevents the evolution of new ways of doing things. Furthermore, centralising regulation at the EU level increases systemic risk by promoting herding – and if that regulation promotes perverse behaviour across the EU as a whole (Solvency II promotes investment in risky government bonds, for example) risks are multiplied.

No matter how damaging, it seems impossible to turn back the regulatory ratchet in Brussels. Whether it is costs loaded on firms leading to lower productivity or labour market regulation leading to the tragedy of youth unemployment, the EU elite is unmoved. That is how it appears from a British point of view.

So what would a good reform programme look like? When it comes to the regulation of labour markets, Brussels should have no powers. The Common Agricultural Policy is a disaster and should be scrapped. Fishing should not be regulated at the EU level. With regard to banking, insurance and other financial market regulation, the EU should have one power only – to prohibit member states from bringing in regulations that restrict trade. If the French government wishes to burden its consumers with regulation, so be it. But if a UK insurance firm establishes a subsidiary in France regulated under French law and buys services from the UK subsidiary, the French government should not be able to stop this by implicit or explicit means. The internationalisation of insurance regulation – if regarded as desirable – should happen by inter-governmental agreement between countries with similar markets and systems, such as the UK, Ireland, Canada and the Netherlands.

This should be David Cameron’s renegotiation agenda, but it will not happen. For these reasons, most of my allies in Westminster think-tank land are convinced Britain should leave. But, at the moment, I am a fence-sitter. I look at Westminster and do not like what I see there either. It was the UK government that drove through Solvency II in the EU; it was the UK government that promoted the most expensive way of reducing carbon outputs and ensured that it became EU policy. And its record at home is just as bad.

A government that claims to be in favour of deregulation is introducing charge caps for pensions; Harold Wilson-style training levies; and a huge increase in the minimum wage. Eurosceptics tend to suffer from nirvana fallacy – they compare the EU with all its faults to a perfect policy environment in the UK. On the positive side, since 1986, the EU has done much to promote the free movement of capital and I very much value the free movement of labour. Indeed, should we leave, my fear is we will still end up with all the EU regulation and, in addition, more restrictions on migration. At the moment, I cannot migrate firmly to one side of the fence.

Prof Philip Booth is the IEA’s Editorial and Programme Director, and a professor at St Mary’s University, Twickenham. This article first appeared in The Actuary.