Does Debt Slow Growth?

by: Scott Sumner

Maybe, but I'm not too clear on exactly how. Here's the Financial Times:

Others believe the People's Bank of China will retain its ability to ward off crisis. By flooding the banking system with cash, the PBoC can ensure that banks remain liquid, even if non-performing loans rise sharply. The greater risk from excess debt, they argue, is the Japan scenario: a "lost decade" of slow growth and deflation.

Michael Pettis, professor at Peking University's Guanghua School of Management, says rising debt inflicts "financial distress costs" on borrowers, which lead to reduced growth long before actual default.

"It is wrong to assume that 'too much debt' is bad only if it causes a crisis, and this is a typical assumption made by almost every economist," Prof Pettis wrote in a draft of an forthcoming paper shared with the Financial Times.

"The most obvious example is Japan after 1990. It had too much debt, all of which was domestic, and as a consequence its growth collapsed."

Distress costs include increased labour churn as employees migrate to financially stronger companies; higher financing costs to compensate for increased default risk; demands for immediate payment from jittery suppliers; and loss of customers who worry a company may not survive to provide aftersales service.

Many are now concerned that China's debt could lead to a so-called balance-sheet recession - a term coined by Richard Koo of Nomura to describe Japan's stagnation in the 1990s and 2000s. When corporate debt reaches very high levels, he observed, conventional monetary policy loses its effectiveness because companies focus on paying down debt and refuse to borrow even at rock-bottom interest rates.

The final paragraph discusses the ineffectiveness of monetary policy. Obviously I don't agree with that; monetary policy is always and everywhere highly effective. So if the mechanism is supposed to be "less AD," then I'd say debt is nothing to worry about.

The preceding paragraph discusses some possible supply-side mechanisms, but they don't seem powerful enough to have large macroeconomic effects. I suspect that Japan's slow growth has been a mixture of tight money (low NGDP growth), low population growth and low productivity growth. Only the productivity growth could be plausibly linked to debt, and even there the connection is tenuous.

The article also has a graph putting China's debt in perspective, total debt as a share of GDP:

Notice that China's debt ratio is almost identical to the US. But the FT also mentions two reasons to worry:

1. Their debt ratio is quite high for a developing country.

2. Their debt ratio is increasing rapidly.

I'm not too worried about the first point, as China is a very unusual developing country. But the second point does seem like a cause for concern. Will the economy be able to shift away from high levels of debt formation, without triggering a recession? I honestly have no idea. My best guess is that China will have a financial crisis and recession at some point in the next 20 years, but I have no idea when.

As always, NGDPLT would make the debt crisis (if it does occur) much less severe.

P.S. Note that China bears have been concerned about debt for many years now, and so far their predictions have not proved accurate.

P.P.S. Here's the debt breakdown by sector:

P.P.P.S. Over at Econlog I have a new post, explaining what would make me doubt the truth of market monetarism.