I plan to post a new entry very soon but before doing so I wanted to say a few things about the stock markets, which continue to be insane (but not unexpectedly so) and then repost a blog entry that is nearly five years old. By the time I published my latest (July 17) blog entry Beijing had managed to stop the panic with the use of what I called “brute force”, by which I meant that there was never likely to be much impact from interest rate moves, regulatory changes, margin relaxation, and so on. This is because there had been such a remarkable convergence among investors, almost all of who were purely speculative, on how to interpret information, and because any interpretation was likely to be self-consciously skeptical, that any regulatory response had to be completely unambiguous.

There is nothing less ambiguous than actually buying or selling large amounts of shares. In a July 8 message to my clients, I argued that

… the only way to create a credible floor, or to create credible expectations of rising prices, is by “brute force”. Beijing must force entities under its control, or entities it can influence, to buy shares until all uncertainty is removed.

The panic could only be stopped, in other words, by very visibly forcing institutions under state control to buy heavily, and to prevent them from selling. Other forms of signaling would not work. This is indeed what the regulators did, and they did it powerfully enough that by July 9 they had arrested the panic and set the market off on another surge.

The problem with the surge was that the various “unorthodox” measures used to stop the panic created all sorts of strange convexities and implied options that could either interrupt or speed up the surge, and in a market in which there has been both a convergence of strategies and, what's worse, a convergence in the way information is interpreted, any interruption in the surge was likely to be brutal. In this market, either we collectively agree that we have decided to buy, or we sell.

What worries me most is that many of the measures employed by the regulators to halt the panic are unorthodox enough, to put it mildly, that they can introduce all kinds of new convexities and implied options that we don't fully understand. As we begin to recognize and understand them, however, these might be enough to undermine confidence in our widespread agreement about having reached a consensus.

There was one measure about which there is some disagreement as to its size and its importance, but this might be more than counterbalanced by the very simple and clear signal it gives:

I realize this is very abstract, but it might help make things a little clearer to consider one of the best-known of the measures employed over the desperate weekend of July 4-5. This measure is described in a recent article in Caixin, which describes a meeting held by the CSRC involving the heads of China's 21 largest brokers: “The firms announced in a joint statement that to stabilize the stock market they would spend at least 120 billion yuan combined to buy exchange-traded funds linked to blue-chip stocks listed on the Shenzhen and Shanghai bourses. Moreover, the firms pledged to hold all stock that had been bought with their own money until the index reached at least 4,500 points.”

Why would this matter? Because if brokers are holding large amounts of shares that they are eager to sell, but cannot do so until the index hits 4,500, this creates a barrier, or at least a speed bump, at around 4,500 whose impact as the market races up is hard to determine. This acts effectively as a kind of call option that investors must give away any time they buy stocks while the index is below 4,500.

My worry, as I discussed with my clients, was that as the index approached 4,100 or higher, the threat of intense selling by capital-tight brokers at 4,500 meant that anyone buying shares was implicitly giving away a free call at 4,500, and the higher prices went, the less upside there was and the more downside.

Writing a synthetic put option

By the way remember that if you are long the underlying asset and short a call option, you are effectively short a synthetic put option struck at the same price as the call option. This means that anyone who owns shares might in fact be short a complex synthetic put option on the market, and as the put becomes increasingly less out of the money (i.e. as the index approaches 4,500), the value of the put rises, and so the loss to the implicit “writer” of the put would also rise.

If the writer of the put can cancel the option at no cost, the rational thing for him to do would be to cancel it. In fact he can do so simply by closing out his long position and selling his shares. By the way the fact that most investors do not understand option theory is irrelevant. The option framework predicts how investors will behave as long as they understand that a lot of selling puts downward pressure on prices and a lot of buying upward pressure, and in a purely speculative market, this is pretty much the only thing investors have to understand.

I don't know if this is indeed what triggered the selling, but on Thursday, July 23, following a string of uninterrupted up days, the market closed at a new recent high of 4,124, after which it dropped sharply every day for the next three days to close down by just over 11%, at 3,663. This is exactly what you would have expected if traders believed that the threat of significant sales when the index hit 4,500 was substantial, or at least if they believed that the market believed it.

This expectation is what matters – or, more accurately, what matters is that everyone knows that everyone else is focusing on 4,500 as a break point. There may or may not be a great deal of selling likely to occur once the index hits 4,500 as brokers are forced by their weak capital positions to sell shares, and opinions vary on this point, but in this kind of market as long as investors believe that this is enough of a possibility for a consensus to form around it, they will act as if it were true. This means, then, that they will act like they have written a put option struck at 4,500, and unless they are absolutely certain that stocks will trade up right through 4,500, they are in the position of being able to cancel their short put any time they like simply by selling their shares.

The market may or may not be thinking this way, but it has certainly acted like it might be. After those three bad days the market traded up on Wednesday by 3.4% to close at 3,789, but then lost faith again Thursday and traded up by 2.2% to close at 3,706. Today, Friday (July 31), the market opened 1.3% lower and except for a few minutes in the early afternoon it was in the red all day to finish at 3,664, down 1.1%. I expect it might make several runs at 4,500 before it breaks through.

If this model is appropriate (i.e. if enough investors believe that there is a consensus around brokers selling substantially once the index breaks 4,500), there are two obvious implications:

But this is all wild speculation on my part. My main point is that the structure of investment strategies in the Chinese stock markets had always guaranteed that this would be a brutally volatile market that trades almost exclusively on “the consensus about the consensus”, and therefore prices will reflect very rapid shifts in this consensus, in exactly the way Keynes explained in his description of beauty contest strategies. The market's mood will rise or fall rapidly, buffeted on the one hand by “brute force” buying and on the other by unexpected speed bumps structured around complicated forms of regulatory intervention, and in a speculative market it is the mood, and not fundamentals, that determines prices.

Recalling 2011 forecasts

To move away from the stock markets and onto a different topic altogether, I thought it might be helpful to reprint my blog entry from August 29, 2011. I was reading though it last week, and one of the points I tried to make at the time was that these predictions were not a set of independent predictions but were rather part of a unified set of outcomes based on the model I use for thinking about the global economy. They were, in other words, all likely to be true or all likely to be false.

Several of these predictions were extremely controversial at the time. I thought it might be useful to re-read these predictions and see which had been accurate and which hadn't, and what this suggests about modifying the model I use to understand the global economy. Before doing so, however, because the concept of rebalancing is fundamental to understanding the adjustment China is undertaking, I thought first I would reproduce a series of short comments I made earlier this week to clients about rebalancing in China:

Does China's demographics help rebalancing?

A client recently asked me whether there were any rebalancing benefits to the Chinese economy as ageing Chinese retirees continue to consume, or would the economy be worse off if it reduces the amount of savings that go into investment. Are the demographics favorable or unfavorable?

I think there are two useful points. First, your savings are equal to what you produce less what you consume. Once you are no longer working, your production drops to zero, and because your consumption doesn't drop to zero, your savings become negative. Consumption is broadly a function of the size of the total population whereas production is broadly a function of the size of the working population, so that anything that reduces the working share of the total population – retirement, unemployment, children – tends to increase the consumption share of GDP.

Whether the economy is better off or worse off depends on many things. If you want maximum current welfare, the higher the consumption share
the better off you are. If you want maximum growth, the higher the investment share, the better off you are, although of course the growth represents an increase in wealth only if investment is productive – i.e. the increase in future consumption caused by the higher productivity the investment generates should be greater than the reduction in current consumption that paid for the cost of the investment (this, of course, has been one of China's big problems).