My Blog

May 6, 2008

TUE 6 MAY

Perceptions of market support can add shocks By Michael Pettis

The Chinese stock market continued to bounce around today, driven down largely by the poor performance of bank stocks.  There were fears among investors that we may see more tightening measures in the form of hikes in interest rates or minimum reserve requirements, or both, and these fears have hurt bank stocks in particular.  Shanghai opened the day down about 1% and quickly dropped another 1% or so in the first few minutes, before trading up in the late morning and bouncing in and out of positive territory all day, until by late afternoon it was up 0.6%.  In the last 30 minutes, however, it gave up all its gains and then some to close down 0.73%.  That doesn’t bode well for tomorrow, but either way I don’t think there was a lot of conviction.

 

A couple of my Peking University students who trade regularly and who keep track of the market gossip tell me that there is a real sense among investors that the government is in control of the market and won’t allow it to fall much further – 3000 seems to be the magic number below which it can’t fall (the SSE Composite closed at 3681 today).  This has buoyed market sentiment and has kept investors in the market.  Needless to say, this kind of belief can cause damage to the capital allocation mechanism by distorting the market clearing mechanism.  

 

There is something else here that may be of interest to those curious about the mechanics of the markets – and the rest of this post is not really about the Chinese financial markets.  It is just some speculation on ways in which markets can adjust after distortions have been introduced, so probably of very little interest to most of the regular readers of this blog.

 

It is widely known that the perception of a minimum trading level, enforced by some credible agency, can distort the actual trading level in a predictable way – keeping it above whatever the fundamental level supply and demand would have naturally created.  I try to show this in the graph below:

 

  

 

In the graph, assume that the hard horizontal line is the perceived minimum trading level of the SSE Composite permitted by the government (which the market perhaps assumes to be around 3000 or just below).  If we assume that normal supply and demand in the market would have caused the index, absent government support, to move up and down the upward sloping straight line labeled “fundamental value”, the implicit belief in the government support level will actually drive the market along the curved “trading value” line, so that its price will lie directly above where it should have traded without the perception of government intervention.

 

The problem with this kind of distortion is that if and when the government is no longer able or willing to support the market, or more importantly the perception of government support evaporates, probably at some point where the “fundamental value” is well-below the perceived minimum trading level (the horizontal line), there is a risk of a sudden and sharp drop in price from the “trading value” line to the “fundamental value” line as real market supply and demand are forced to clear. 

 

This is what Paul Krugman predicted would happen to the value of the euro many years ago when it traded down shortly after its launch but hovered above $1 – largely on the perception that the European governments would not want it, for political reasons, to trade below $1.  He said that when the euro broke $1, it would not do so gradually.  Instead it would fall very sharply. 

 

That is in fact what happened.  For many weeks the euro stayed above $1 dollar, trading up and down in a narrow range.  But the supposedly temporary support extended by intervening governments in the hope that the markets would eventually “get it” (i.e. understand that the euro really was worth a lot more than $1) was not able to turn market sentiment – perhaps because the creation of the euro itself caused a one-time liquidity adjustment, as Robert Mundell predicted it would – that would result in an excess supply. 

 

The attempts to change market dynamics by changing underlying sentiment, in other words, failed.  Eventually, after extended intervention was clearly unable to turn sentiment solidly around, European governments were forced to stop intervening.  Shortly afterwards the euro broke $1, and just as Krugman predicted, it immediately dropped very sharply by nearly 10 percent, before resuming its gradual drift downwards to below 80 cents.  

 

If the SSE Composite trades within a narrow band above 3000 for several weeks or months, with new administrative measures (or rumors of such) emerging every time it trades too close to 3000, we may find ourselves in the sort of situation Krugman posited for the euro.  In that case the model would predict that if and when it broke 3000 (or, more probably, some psychological support level below 3000), it would break sharply.  In that case we might see the index lose 5-10% or more within a day or two.

 

P.S. I know Cui Enze, Liu Bing and Shang Ning, as well perhaps as some of my other students who are fascinated by the dynamics of trading, are going to be all over this blog entry.