A random walk up downtown Almaty

Can you get rich from speculating on stocks in Kazakhstan? Yes! Er, no….

Few markets are actually as competitive as simple economic theory recommends. One may be the stock market. With many buyers and sellers of shares, none is large enough to influence the price of a given stock, much less an index of all stocks.

In theory, then, all agents in the stock market will take the stock price for granted when deciding what to do. When buyers think a stock undervalued, then they will keep buying it until its price rises to the point of reflecting the issuer’s true value. Similarly, they will sell overvalued stocks. Buying and selling shares is simple, so the new information that prompts movements in the stock market will be absorbed by new prices quickly. The market is “efficient.”

This is not quite the usual economic definition of efficiency – i.e., that we maximize the net value of all inputs into production. But it does imply a saving of the most valuable input of all – time. To judge whether oil is becoming more valuable, an observer need not study assiduously the characteristics of the industry, such as the number and locations of oil wells and pipelines, the purity of the crude oil, and the demand of airlines for fuel refined from oil. He need look only at whether the price of oil stocks is rising.

But many are the slips twixt theory and fact. Stock markets may not be as efficient as they seem. Over the Nineties, U.S. stock prices soared for computer firms issuing shares for the first time. The index for NASDAQ, the U.S. stock market that attracted computer start-ups in droves, rose tenfold over the decade. This overvaluation led to the “dot.com crash” of 2000, in which the NASDAQ index would eventually fall by half, noted the financial economist Frederic Mishkin. As paper wealth melted away, American consumers cut back on spending. A mild recession ensued in 2001. Can such market corrections occur fast enough to avoid wild fluctuations in stock prices, wealth and income?

For stock markets in developing countries like Kazakhstan, the question is complicated by doubts about the markets’ basic efficiency. Trading is often thin in these embryonic exchanges. A seller or buyer of stock may find it easy to influence the share price by dominating the volume of trade – that is, by cornering the market.

How can we tell whether a stock market is efficient? If it processes information at the speed of light, then we should observe share prices changing only for unpredictable reasons; all predictable adjustments have already occurred.

Suppose, for example, that the central bank in the United States, the Federal Reserve, announces an apparent slowdown of the national economy. Demand will drop for oil, a primary input. Oil prices will fall as well, pulling down the value of oil extractors. Their share prices will plummet. This reasoning, of course, is almost too obvious to be stated. And that’s the point: Millions of investors will anticipate a fall in oil stock prices and so will cash in their oil shares or sell them short (i.e., sell a borrowed share today and buy a cheaper share tomorrow in order to close the loan). The prophecy will fulfill itself; oil prices will fall immediately. In short, yesterday’s news will affect yesterday’s stock prices. Today’s prices can react only to today’s news – and only the unpredictable news at that.

The Bourbon statistician

This suggests a simple pattern for stock prices: Today’s price will equal yesterday’s, plus a random change. That’s a “random walk.” In New Orleans, once Party Central for most of the known universe, native statisticians liken a random walk to the path of a drunk on Bourbon Street – one step forward, maybe one step back, as the party pro staggers on a whim.

This summer, a graduate student in KIMEP’s economics program, Assel Prmanova, looked at whether stock prices in Almaty followed a random walk. She gathered daily values of the index for the Kazakhstan Stock Exchange (KASE) for the past decade. She concluded that today’s index value does indeed incorporate yesterday’s.

But the daily change in the index was not random. It followed a negative pattern: Usually the index rose one day and fell on the next. (Econometricians call this “negative serial correlation.”) Evidently, KASE is not completely efficient: The price changes follow a pattern that speculators should exploit but don’t. If we saw the market rising today, then we should sell short in order to cash in on tomorrow’s likely decline. Our short sales would lower today’s prices, eliminating the pattern of rise and fall. In other words, the very presence of a price trend tells us that speculators are not acting on all information available.

There is a golden lining in these statistical clouds. If the stock market is fully efficient, then a speculator cannot hope to grow rich over time; his rivals have already exhausted all visible opportunities for profit. To make money at all, you must be lucky or -- what amounts to the same thing -- quick. (Or well connected.) But KASE does not seem that efficient. So get out your checkbook.

Some technical issues remain. Dana Stevens, KIMEP’s vice president of academic affairs and an economist trained by Stanford, points out that the KASE index changed little from 2000 to 2003. If the index was 100 on one day, it was 100 on the next. This inertia may mislead us into believing that today’s index incorporates yesterday’s more often than it actually does.

The key issue is whether KASE changed fundamentally in 2003. A constant stock price need not connote inefficiency; we would be rational to avoid a market that kept us from taking profits. But if, in fact, KASE took on a new shape in 2003, then we will need a new model to explain it for that year and forward.

Moral: No economic model is ever perfect. That’s why economists are guaranteed long and gainful employment. – Leon Taylor tayloralmaty@gmail.com

Full disclosure: I advised Ms. Prmanova’s thesis, as did Drs. Aleksandr Vashchilko and Eldar Madumarov, economics professors at KIMEP, and Dr. Stevens, a finance professor at KIMEP.

Good reading

Fama, Eugene F. 1970. Efficient capital markets: A review of empirical work. Journal of Finance 25:2, May, pages 833-417. A famous survey of early tests of the Efficient Market Hypothesis, which were generally favorable.

MacKay, Charles. 1996. Extraordinary popular delusions and the madness of crowds. New York: John Wiley. This 19th-century classic on financial hysteria is fun to read.

Malkiel, Burton. 2007. A random walk down Wall Street: The time-tested strategy for successful investing. New York: W. W. Norton. A classy defense of market efficiency, with a few words of advice on how to play the stock exchange. (The main word is: “Don’t.”)

Marsh, Terry A. and Robert C. Merton. 1986. Dividend variability and variance bounds tests for the rationality of stock market prices, American Economic Review 76:3, June, pages 483-498. Takes a tack differing from Shiller’s but also rejects the Efficient Market Hypothesis.

Mishkin, Frederic S. 2007. The economics of money, banking and financial markets. Eighth edition. Boston: Pearson Addison Wesley. Chapter 7, The stock market, the theory of rational expectations, and the Efficient Market Hypothesis. An easy-to-read overview.

Mishkin, Frederic S. 1978. Efficient markets theory: Implications for monetary policy. Brookings Papers on Economic Activity 1978: 3, pages 707-752. This survey is more detailed and sophisticated than Mishkin’s textbook chapter, but it’s still readable.

Prmanova, Assel. 2010. Factors affecting the Kazakhstan Stock Exchange Index. Almaty, Kazakhstan: KIMEP Master’s of Economics thesis.

Shiller, Robert J. 2003. From efficient markets theory to behavioral finance. Journal of Economic Perspectives 17:1, winter, pages 83-104. Develops a novel test that rejects the Efficient Market Hypothesis.

Revised on August 21, 2010