At the outset of this blog I pointed out that if we want an economy that produces enough for everyone then we’re concerned with the productivity of the economy. Sustaining and improving productivity requires investment, and in this section of the blog I have been exploring the implications of the relationship between saving and investment.
Economic textbooks will tell you that financial markets intermediate saving to investment, and I have been pointing out that most activity in these markets does not, in fact, channel saving to investment, and that this leads to the paradox of thrift (slowing down the economy) and asset price inflation (bubbles).
But there is a theory in economics that states that asset markets are ‘efficient’ – that the prices in these markets represent the ‘fundamental’ or ’intrinsic’ value of that asset (a concept explained last week). As this is such a direct contradiction of the ideas presented in the blog, it would be intellectually dishonest not to point out this theory’s existence.
The theory is often attributed to Eugene Fama, particularly his 1970 paper “Efficient Markets: A Review of Theory and Empirical Work”, but as the title suggests, this paper was itself building on a large body of on-going work by himself and others. Fama was awarded the nobel prize for economics for his work in this area in 2013.
The efficient market hypothesis (EMH) states that asset prices in financial markets will reflect all available information, and that therefore it is not possible to beat the market.
There is a slight paradox at the heart of this theory, in that this does not mean that traders who analyse and track markets in detail are wasting their time. It is precisely because they do so that they can spot and leap on any imperfection in the market as soon as it appears, making that imperfection disappear. If an asset is under-valued they will rush in to buy it, pushing the price up. If the price is too high they will sell it, causing the price to fall. As soon as any new information becomes available the professional traders will act on it and prices will changes almost immediately.
Economists often talk about markets and prices conveying information, and this is what they mean – as soon as information becomes available that would affect an assessment of its fundamental value, traders buy or sell the asset and price changes. And hence the EMH states that asset markets reflect all available information.
Marcia Stigum gives a wonderful description of this frenetic activity, as traders jump onto every opportunity to profit by moving their funds from one market to another, in her classic Stigum’s Money Market (page 900 – cited by Perry Mehrling):
“Money market swaps occur in what could rightly be called arbitrageland. Traders arbitrage swaps against futures, swaps against cash, swaps against forward rate agreements (FRAs), FRAs against futures, and so on. Arbitrage opportunities keep arising because these related markets are constantly affected by many different events. Maybe an Asian bank does a big cash and swap arbitrage, which drives up the swap market. This creates profit in the swap-FRA arbitrage, so someone does the swap-FRA arbitrage, which drives up FRAs. This creates profit in the FRA-futures arbitrage, so someone does the FRA-futures arbitrage, which drives up futures. An event that moves one rate causes a rate ripple that creates some basis points for every play, except maybe the futures player, if he is an unhedged spec [speculative trader]. Clearly someone loses, usually the spec player in the future’s pit.”
While not written to explain the EMH, the description of trader activity helps us understand how the process would work. It also highlights a point made 3 weeks ago, that when a trader sells an asset it is very often not to release funds back into the real economy, but just so that they can buy a different asset.
The implication of this hypothesis is that no-one can beat the market, and research shows that this is almost always true. Because professional traders react so quickly to new information, prices change rapidly. Even these traders are not making a killing because they all jump on any profit-making opportunities so quickly. Study after study has shown that you can pick stocks at random and perform as well as most managed investment funds. In fact, these studies directly led to the creation of index funds that track entire markets, described three weeks ago.
This doesn’t necessarily mean that markets are right – that prices reflect fundamental values – but that prices reflect all available information. Yet even in this weak form the theory can be taken to imply that there should be no Government intervention or regulation in financial markets. Markets are responding to information far more swiftly and efficiently than clunky, bureaucratic government institutions ever could. This is exactly the point that Alan Greenspan, Chairman of the Federal Reserve from 1987-2006, was famous for making to explain his attitude to avoiding intervening in markets – that even if markets weren’t perfect, the Government wasn’t smarter than the market and couldn’t improve its functioning.
But the strong form of the theory argues that prices are, in fact, correct – that the wonderful workings of the market mean that prices will reflect fundamental values. The prediction made by EMH is that markets will follow a “random walk” (this is a very famous phrase associated with the theory). Prices will move up and down randomly around the fundamental value, with small adjustments occurring as new information causes assessments of that value to be tweaked. And much of the time this is how markets behave.
But, of course, crashes are not “random walks” but dramatic plunges. You might think that the existence of periodic crashes disproves this theory, and basically you would be right! But justifying this statement, presenting the arguments on both sides, and explaining why so many economists, politicians and others have clung onto the EMH in the face of this obvious evidence would take way too long.
But very briefly, when prices fall so rapidly this suggests that the price before the crash was wildly inflated. If it wasn’t inflated, this means that the crash itself is the result of irrational selling, and that the new price is markedly below the fundamental value. Either the pre-crash or the post-crash price must have been wrong, disproving the strong form of the hypothesis.
Even the weak form fails in the face of crashes. If prices reflect all available information, then a crash should only happen because new information becomes available that would indicate that prices were not in fact correct (in the weak form, prices might not be correct, but they will be the most rational assessment of fundamental value that could be made based on available information). But no crash in history was triggered by the sudden availability of such information.
Therefore the fact of periodic crashes certainly points to huge problems in the hypothesis. We will discuss these further next week.