Monkeys Fly in China
In the very early days of this blog, I told the story about what Chief Economic Advisor Beryl Sprinkel said on Black Monday, 1987, when a panicked Treasury Secretary James Baker wanted to close the stock market: “We’ll close these markets when monkeys fly out of my ass.” No monkeys flew, and the markets stayed open, eventually stabilized, and then recovered.
But many monkeys are flying out of many asses in China. Although the authorities have not closed the stock markets, individual companies have halted trading in their stocks: trading in more than one-half of the listings in China is currently suspended.
Halting trading more than for a short interval in order to resolve information asymmetries and permit the flow of liquidity to stocks that have just experienced an information event (as during a temporary stock halt in the US) is in general a bad idea. (Post-87, Greenwald and Stein wrote a paper published in the JOB laying out this argument.) An uncoordinated and extended halt of many stocks is a really horrible idea, because of the negative externalities. That is, uncoordinated flying monkeys wreak even more havoc than coordinated ones.
Halting trading in a large number of stocks increases selling pressure on stocks that are still trading. This happens for at least a couple of reasons. First, individuals who need to raise cash (e.g., to meet margin calls) are forced to concentrate their sales in the stocks that keep trading. This tends to concentrate selling pressure, rather than diffuse it. Second, individuals who want to rebalance their portfolios away from equity into cash or bonds have to concentrate their sales in the stocks that continue to trade. Again, this concentrates selling pressure.
This creates a vicious feedback loop. A number of companies halt trading, which forces selling pressure to spill over with greater force on other stocks, which leads some of these companies to halt trading, which intensifies selling pressure on other companies, and so on. The ultimate likely outcome is a protracted lockdown of the entire market. Protracted because who is going to be the firm to restart trading first, and risk having everyone sell the hell out of them?
The vaunted Chinese economic managers (ha!) have well and truly bungled this one. They should have prevented open-ended trading halts, or had a coordinated stoppage and restarting of trading. The coordination failure at work now is manifest.
Again, I believe that the sharp selloff is more of a symptom of a deeper economic problem than a potential direct cause of such a problem. The main adverse spillover that the stock selloff could cause is through the margin debt channel. Margin calls could lead to fire sales of illiquid assets. Again, the more stocks that are not trading, the more severe these fire sales in non-equity assets will be: this is another adverse consequence of uncoordinated monkey launches. Moreover, failures to meet margin calls will saddle the lenders (themselves often highly leveraged) with losses. Both of these channels could have adverse consequences in the brokerage, banking and shadow banking sectors. Their balance sheets are not that hale and hearty to begin with, and this kind of shock could spark broader financial distress throughout the sector.*
In other words, the stock market decline is less of a crisis in itself, than a potential catalyst to a crisis via informational and fire sale channels. And perversely, uncoordinated trading halts in the stock market are more likely to intensify than mitigate any such catalytic effect.
But the Mandarins know everything, so I’m sure it will turn out swell.
In the meantime, the Mandarins have a message for all investors in China. Good luck with that!
* Perhaps one could argue, as Michael Brennan did when trying to explain price limits in futures markets in the JFE in 1986, that halting trading could ease pressure on margin credit. I am skeptical though. Even if stocks stop trading, margin lenders are likely to demand additional security in current conditions. Indeed, trading halts that reduce the informational content of stock prices create a source of uncertainty to margin lenders which they are likely to compensate for by demanding additional margin based on their estimate of the stock price once trading recommences, plus a premium to compensate for the uncertainty.