Most people are convinced by now that there is a fixed-asset bubble in China that is on the verge of bursting. The question is, what is next? The answer will depend on how much debt is in the system. The more debt infests the system, the further and steeper asset price declines will be and the longer and more painful recovery will be. The official Chinese debt-to-GDP ratio is an unalarming 20 percent or so. But the true number is likely to be a lot greater, as it is hiding in — and off — the balance sheets of state-owned enterprises (SOEs) and municipal governments, explicitly or implicitly guaranteed by the Chinese government.
I tend to ignore the guarantees, or the lack thereof, for one simple reason: Most of the debt is underwritten by Chinese banks, which are majority-owned by the government and for all practical purposes guaranteed by it.
The Financial Times had an interesting article in June titled "China Warns on Lending to Steel Plants." When you read the headline, you think, "Well, there is overcapacity in the Chinese steel industry, and the government is trying to curb it by cutting off lending." Though this is correct — there is significant overcapacity in the steel industry, especially as China runs out of money building ghost towns and empty skyscrapers — the article is not about that. It talks instead about how steel SOEs used loans from banks to speculate on property and stocks.
"The China Banking Regulatory Commission said in its directive that steel trading companies had borrowed from banks for steel-related activities — sometimes then using the same collateral to get multiple loans from different financial institutions. Then, with little demand for steel, they had used the money for risky investments, such as stocks and property."
Remember all the talk about how China had no derivatives, no collateralized debt obligations–squared? Well, Chinese SOEs seem to have invented their own version of a CDO-squared: They used the same collateral to get multiple loans, and Wall Street didn't even have a hand in it. So when you hear an argument that loan-to-value levels are much lower in China than in the U.S., keep in mind that when you give out a loan secured by an asset that is pledged on another loan, loan-to-value has no meaning.
Steel companies are not the only SOEs that have embarked on real estate adventures in China. A few years ago the New York Times wrote, "Giant state-owned oil, chemical, military, telecom and highway groups are bidding up prices on sprawling plots of land for big real estate projects unrelated to their core businesses."
The paradox of the Chinese economy is that what is good for consumers is bad for the SOEs, and vice versa. Inflation is consumers' biggest enemy, eating away at their savings because the interest rate on bank deposits is set well below the inflation rate.
SOEs, however, are the primary borrowers and thus the beneficiaries of depressed deposit rates. In their paper "A Real Picture of State-Owned Enterprises," researchers at the Unirule Institute of Economics in Beijing showed that SOEs paid, on average, a real interest rate of 1.6 percent from 2001 to 2009, while the average market rate was 4.7 percent. The paper also showed that if interest rates and other subsidies SOEs received from the government were taken out, they would have earned a negative return on capital; in other words, they would have lost a lot of money.
So if SOEs had a hard time making unsubsidized profits when the Chinese economy was roaring — when they could build and flip towns as if they were playing FarmVille — what will happen to their profits and their ability to service their debt when conditions are less favorable?
Things may change. Consumers may revolt against interest rate oppression, SOEs may be forced to pay market rates, and inflation may be replaced by deflation. This point is important. Overcapacity in fixed assets is deflationary in the short and medium terms. However, overcapacity is usually inflationary in the long run, as the government has to print money to pay for all the debt it has taken on to bail out the economy. SOEs will have a hard time surviving the short and medium terms.
The Chinese debt-to-GDP ratio doesn't worry me; it's the bad-debt-to-GDP that causes me concern. In life I follow the principle spelled out by John Maynard Keynes: I'd rather be vaguely right than precisely wrong. Taking into consideration that the Chinese government uses its banks as a political tool to stimulate growth, and adding on top of that the insidious corruption and the little fact that freewheeling SOEs are 30 percent of the economy, I'd say Chinese bad-debt-to-GDP might be somewhere between enormous and enormous-squared, and I am gravitating toward the latter.