Deleveraging and Dalio

Here, with a little delay, is an article in the Economist about hedge fund boss Ray Dalio's views about deleveraging. His fund, Bridgewater, is possibly the most successful in markets at present. And that means it is worth giving a hearing to the ideas that help make it so (or at least those that he reveals).

Bridgewater also has a reputation for one of the strangest and most savage cultures in any company anywhere. But they do seem smart.

The Economist article links to some longer writing by Dalio on his own site here. He says:

The long-term debt cycle top occurs when 1) debt and debt service levels are high relative to incomesand/or 2) monetary policy doesn’t produce credit growth. From that point on, debt can’t rise relative toincomes, net worth and money supply. That is when deleveraging – i.e. bringing down these debt ratios –begins. All deleveragings start because there is a shortage of money relative to debtors’ needs for it. Thisleads to large numbers of businesses, households and financial institutions defaulting on their debts andcutting costs, which leads to higher unemployment and other problems. While these debt problems canoccur for many reasons, most classically they occur because investment assets are bought at high prices andwith leverage – i.e., because debt levels are set on the basis of overly optimistic assumptions about futurecash flows. As a result of this, actual cash flows fall short of what’s required for debtors to service theirdebts.

The economy follows a leverage cycle which can last decades. Serious problems really start when the central bank cannot reduce debt service ratios directly any more because interest rates hit zero.

It is an attractive and essentially quite simple analysis.

The most interesting thing is that he obviously starts with economic history, rather than with standard academics. The papers on the site are littered with examples of how the UK economy rose and fell over a period of more than a century, and a long view of the US economy over the same period of time.

This just has to be the right way to do things - bedding things in actual history rather than abstract rational choice and optimizing, i.e. the framework of academic economics. And maybe that is his secret - it is poring over history for multiple cross-cutting lessons, which he embeds in computer rules.

Give me the man credit (literally). But surely such outperformance can't go on forever, just by the law of averages.