International regulation - institutionalising systemic risk

I am currently attending a conference on the future of financial regulation being held by the Irish central bank in Dublin. We are hearing from the regulators about the importance of global regulation. This blog post is the second half of my remarks in the final panel discussion. They address the notion that, as the world becomes more complex, we need more and more regulation at a higher and higher level. In discussions about the history of regulation it is often suggested that we cannot go back to the ways of the 1980s because the world is more complex today. It is argued that the networks of trust and so on do not exist to allow market-developed mechanisms of regulation to thrive.

There are two responses to this. The first is that regulators themselves can be blamed – at least to some extent – for making the world a more complex place. This is so directly: Fannie Mae and Freddie Mac encouraging securitisation; the Basel Accords encouraging complex models and gaming through the use of more complex financial structures; and so on. It is also true indirectly because of the way in which regulation makes it less important that we know the bona fides of those with whom we are dealing. The academic work, for example, suggests that deposit insurance and regulation led to the decline of simple mutual businesses in both insurance and banking because their comparative advantage – the trustworthiness that such organisations signalled to customers – lost its value.

However, it is worth noting that regulators’ argument that the more complex the world becomes the more we need regulation was exactly the argument that Mussolini made in favour of central planning. And it is wrong for the same reasons that Mussolini was wrong. The more complex the world becomes, the more difficult it becomes to improve markets through regulation because the necessary knowledge becomes harder to centralise.

Thus, when it comes to international regulation, we have a choice. We can continue along the path of the last 20 years. In 2011, worldwide, there were 14,200 new banking regulations. The US Dodd Frank Act will probably have 30,000 pages of associated regulations. Regulators are paid to write rules and no one can doubt their exceptional productivity. Or we can build legal and regulatory systems that assist and support those dispersed centres of knowledge within markets that can regulate behaviour – self-regulating mechanisms such as exchanges; professional bodies; financial intermediaries; and so on.

It should be a sobering thought for any regulator that the institutions that did not create economic chaos during the financial crash were the unregulated offshore banks and hedge funds. But don’t worry: the regulators are on their case too.

Let me finish with one constructive suggestion. The most important development in the last few years is the development of what is called a ‘resolution mechanism’ for banks – this is better termed a ‘safe bankruptcy mechanism’. No bank should have a banking licence unless the authorities – probably the central bank – can be sure that it can be wound up safely. The elephant in the room, however, is that any bank can operate across the European Economic Area through branches – that is without a separate legal personality. The EU’s own regulations require that. I have no principled objections to banks operating through branches and I have no objections to multi-lateral agreements between countries to unify legal frameworks etc. But, it is extremely dangerous to have a situation where deposit insurance, legal personality and place of operation are not aligned – as a general principle.

The primary focus of international effort should be on international co-ordination of resolution mechanisms. But, if this is to be effective, it has certain implications for EU single market regulation – especially the requirement on all countries to allow no-questions-asked branching. Such a requirement is not necessary to promote free trade in financial services.

More generally, rather than centralising regulation at ever-higher levels, we should develop systems of regulation that use the dispersed sources of knowledge within the market whilst ensuring that firms that fail can be wound up safely. There are fine lessons from history here. The 1870 Insurance Companies Act which was more or less unchanged for a century, simply required information to be published to the market through the regulator. It also had an excellent bankruptcy mechanism for insolvent companies. In 100 years, there were only two life insurance company failures and nobody can name them because they were handled so well. The regulatory mechanisms used all the dispersed sources of knowledge and information in the market. Regulators did not believe that they were themselves the source of all knowledge and information. Such humility would be a welcome development in the twenty-first century regulatory environment.

There is one final problem with increasing centralised EU and global financial regulation. Global regulation is fine if regulators never get anything wrong – but, of course, they will get things wrong. The objective of banking regulation has not generally been to prevent failure but to ensure that failure does not bring the system down. If we impose global uniformity of regulation on the banking sector, we create the opposite dynamic. We will make it less likely that a bank will fail but almost guarantee that, when regulators get it wrong, the whole system will collapse at the same time.

Yes, Philip makes an important point that is too often completely overlooked. Central government planning and state monopolies (as well as over-powerful regulators) are high risk. If (I am tempted to say 'when') governments get things wrong it is often difficult for anyone to escape the consequences. So, for example, if the government runs a state monopoly education system (and now with a 'national curriculum' too!) and chooses to require teachers to employ an inadequate system of teaching children to read, many children's schooling will suffer. If instead we were to enjoy a genuinely competitive school system, freedom to compete would reveal which methods of teaching children to read worked and which didn't. Indeed there might even be something to be said for requiring ALL children to be able to read 'adequately' before they were permitted to leave school, though it would seem obviously preferable that they were actually able to read during their schooldays!.

Post new comment

The content of this field is kept private and will not be shown publicly.
Type the characters you see in this picture. (verify using audio)
Type the characters you see in the picture above; if you can't read them, submit the form and a new image will be generated. Not case sensitive.