You can’t regulate with nostalgia

By Felix Salmon
January 3, 2013

The theme of the day, today, is nostalgia for the simple banking systems of yore, where the Bailey Building & Loan was run by simple, honorable men who had no problems complying with Basel I or its predecessors. If you have a large chunk of time today, you can start with the 9,500-word cover story in the Atlantic by Jesse Eisinger and Frank Partnoy, and then for dessert follow it up with Yalman Onaran’s 2,600-word explanation, for Bloomberg, that bank regulation these days is really complicated.

These are genuine problems. Once you’ve read the Atlantic article, which takes a deep dive into Wells Fargo’s 10-K and comes out convinced that it’s impossible to really know anything about the risks and assets in any big bank, you’ll understand why that’s a huge systemic problem:

As trust diminishes, the likelihood of another crisis grows larger. The next big storm might blow the weakened house down. Elite investors—those who move markets and control the flow of money—will flee, out of worry that the roof will collapse. The less they trust the banks, the faster and more decisively they will beat that path—disinvesting, freezing bank credit, and weakening the structure even more. In this way, fear becomes reality, and troubles that might once have been weathered become existential.

This paragraph is the heart and soul of the piece. (You’ve gotta love any article where the nut graf comes more than 2,000 words in.) We can’t trust the banks; but unless we can trust the banks, another major financial crisis is inevitable, since banks are built on trust, and without trust they are nothing.

This dynamic was central to what went wrong during the financial crisis, and a large part of the problem was the global system of bank regulation known as Basel II, which basically allowed the world’s biggest banks to simply make their own determination of what their risks were and how much capital it made sense to carry against those risks. Obviously, that didn’t work out very well. So, what can be done about this problem?

The answer of the global regulatory regime was something called Basel III. It was pushed through in something of a rush, and so it built on Basel II as a base: my metaphor is that it’s a bit like the way Windows was built on DOS. As a result, although it’s a clear improvement over Basel II, it is necessarily at least as complex as Basel II. And when complexity itself is part of the problem, extra layers of regulation are unlikely to constitute much of a solution.

That’s Onaran’s point, but I think he pushes it a bit too far. His headline is “Basel Becomes Babel as Conflicting Rules Undermine Safety”, and he talks at the very top about how “conflicting laws, divergent accounting standards and clashing rules” have “created new risks” in the banking system. But the shoe never drops: he doesn’t actually explain what these new risks are, or how Basel III undermines the safety of the baking system.

Partly, it’s a baseline game. Both the Atlantic and Bloomberg are essentially comparing Basel III to Basel I, and saying that everything is still far too complicated. By contrast, if you compare Basel III to Basel II, it’s a clear improvement. Onaran’s article is full of quotes from people saying that we should go back to a much simpler system. And the Atlantic actually lays out what such a system might look like:

Congress and regulators could have written a simple rule: “Banks are not permitted to engage in proprietary trading.” Period. Then, regulators, prosecutors, and the courts could have set about defining what proprietary trading meant. They could have established reasonable and limited exceptions in individual cases. Meanwhile, bankers considering engaging in practices that might be labeled proprietary trading would have been forced to consider the law in the sense Oliver Wendell Holmes Jr. advocated.

Legislators could adopt similarly broad disclosure rules, as Congress originally did in the Securities Exchange Act of 1934. The idea would be to require banks to disclose all material facts, without specifying how. Bankers would know that whatever they chose to put in their annual reports might be assessed at some future date by a judge who would ask one simple question: Was the report complete, clear, and accurate?

This is basically principles-based regulation, as opposed to rules-based regulation. Rather than forcing banks to comply with thousands of pages of abstruse regulation, keep things simple and deliberately vague: that’ll keep them on their toes, goes the argument, and force them to err on the side of caution.

If there were a real chance of doing this, I’d be all in favor. Principles-based regulation doesn’t always work: just look at what happened to the City of London. Banks ran rampant, committed massive Libor fraud, and required enormous bailouts; London is also, not coincidentally, the home of JP Morgan’s Chief Investment Office. Eisinger and Partnoy rightly use banks’ price-to-book ratio as an indicator of the degree to which anybody in the market understands or trusts what they’re doing; what they don’t say is that it’s hard to find a lower price-to-book ratio than Royal Bank of Scotland, which was regulated in the UK rather than the US, and which is owned not by out-of-control risk-loving capitalists, but rather by the much safer and much more risk-averse UK government.

And what neither article really admits is that regulators are painfully aware of all the problems they lay out — and many more. That’s why the UK government spent so much effort trying to lure Mark Carney over from Canada to run the Bank of England: he was one of the few regulators who managed to ensure that his national banking system didn’t implode during the crisis, or require any kind of bailout.

But the fact is that all regulation is, by its nature, path-dependent. In 1932 it was easy to install a simple system of bank regulation, because there was no existing system of bank regulation to replace or build on. Since then, as Eisinger and Partnoy write, “accounting rules have proliferated as banks, and the assets and liabilities they contain, have become more complex. Yet the rules have not kept pace with changes in the financial system.” That’s just the nature of things: complexity breeds further complexity, and with it much higher levels of endemic systemic risk.

The genius of Canada, and other countries with safe banking systems like India, is that they never allowed their banks to become highly complex in the first place. There are financial capitals like London, New York, and Frankfurt; those countries will have lots of capital flows and complexity and systemic danger. And then there are second-tier places like Toronto and Delhi, where financing can be much simpler. The problem is that you can’t turn New York into Toronto, or London into Delhi. Nor would any of the politicians in the US or the UK particularly want to do so: they get too much precious tax revenue from being global financial centers.

And it’s equally hard to take a multi-layered rules-based system, complete with a large and powerful and entrenched regulatory infrastructure, and tear it down to build something smaller and simpler. Accounting rules proliferated even during the era of deregulatory zeal in the 1990s; accounting rules will always proliferate, and it’s pretty much impossible to find an example of a regulatory system which has ever managed to go in the other direction, losing complexity, gaining constructive ambiguity, and reducing systemic risk in the process.

So while Eisinger and Partnoy and Onaran are absolutely right when it comes to diagnosing the problem, I think they’re either naive or way too optimistic when it comes to suggesting that all we need to do in terms of a solution is press some magic button and find ourselves with the banking system of the 1950s. We can’t — which is exactly why complexity and systemic risk are here to stay.

Basel III isn’t perfect, but it’s as good as we’re going to get, and is actually significantly better than most people dared hope when it first started being negotiated. And the technocrats who put Basel III together are not some group of knaves, deluding themselves that they’ve magically fixed all the problems with the banking system. They’re smart and well-intentioned regulators, who know full well what the problems are, and who are implementing the best set of patches and solutions that can be implemented in reality. Or if not the very best, then something damn close. They too would love to tear everything up and start from scratch with a much simpler system featuring much smaller banks. But, unfortunately, they can’t.


Comments are closed.