Are markets efficient? And if so, in what way? Some will find this question irritating. This week saw the world’s biggest stock markets hit fresh all-time highs, despite widespread perceptions that the economy is in miserable shape. Stocks have trundled upwards in an almost straight line all year, oblivious to the alarm in the world. Efficient?
And there is the rise of bitcoin, the digital currency. It has no intrinsic value. And yet the price of one bitcoin has passed $10,000 this week, having risen more than 1,000 per cent in 12 months. It then went on to pass $11,000, before taking a dive towards $9,000. In what possible way is this efficient?
There is also irritation with academics’ efficient markets hypothesis (EMH). Easily made to sound ridiculous, EMH holds in its strongest form that markets go on a “random walk”. The market swiftly assimilates all known information about a stock, so that share prices simply walk randomly in response to news. The price is always right. Friday’s very sharp response across global markets to a report that the former national security advisor Michael Flynn was prepared to testify against President Donald Trump rammed home that markets move very swiftly to discount any news.
In its hard form, EMH does not pass muster. The big US tech companies that have led the world for years dipped by almost 4 per cent on Wednesday, when there was no new news. The price could not have been right at all times. And nobody could possibly claim that the market for bitcoin is efficient.
In the absence of strong fundamental anchoring forces, investors tend to under-react to news and/or take cues from past price changes
But market efficiency has more to be said for it than that. First, it implies that it is impossible to beat the market. And experience shows that beating the market is indeed very difficult.
Second, in the very long term markets do get the price right. Historians have shown that over the decades, the returns on stock markets move roughly in line with the growth in the economy.
And if markets are so inefficient, why is it that the invisible hand seems to do a better job than government planners have yet managed to do?
New research by a team led by Jean-Philippe Bouchaud at Capital Fund Management, France’s biggest hedge fund, may reconcile all of this. It confirms a conjecture made 30 years ago by the financial theorist Fisher Black that markets were efficient to within a factor of two. He meant that they would move in a range from half the correct value to double the correct value. Any market that stayed within these wide bounds could be called efficient.
“The factor of two is arbitrary, of course,” Black said. "Intuitively, though, it seems reasonable to me, in the light of sources of uncertainty about value and the strength of the forces tending to cause price to return to value. By this definition, I think almost all markets are efficient almost all of the time.”
This sounded accurate to M. Bouchaud and his colleagues because “humans are pretty much clueless about the ‘fundamental’ value of anything they trade, except in relative terms”. And when they crunched a number of markets, including stock indices, bonds, currencies and commodity futures, using data going back more than three centuries in some cases, for the US, Australia, Canada, Germany, Switzerland, Japan and the UK, this is what they found. Markets wobble in a range between half and double a fair value. They also found another application for the number two.
Behaviour in the market follows two streams. One is “trend following”. Once a stock is in motion, people tend to get on board, so it gathers momentum. Modern algorithms have made trend-following very sophisticated. They exploit and exacerbate what has always been there as a human tendency. These days, this tendency is often referred to as the “momentum” factor.
After a while, fundamentalist investors buy securities that look cheap compared to fundamental value, and avoid those that look expensive. This is generally known as the value factor. The point at which fundamental forces begin to beat the trend comes at about two years.
“In the absence of strong fundamental anchoring forces, investors tend to under-react to news and/or take cues from past price changes,” according to Capital Fund Management, “before eventually following the (very) long term fate of fundamental value”. The mathematical term for this is not a “random” walk, but a “biased geometric random walk with a non-stationary drift”.
So it turns out that markets have always been led by chartists who are checked by fundamentalists. Substitute the words “momentum” and “value” , to both of which most people should be deliberately exposed, and you have what many of today’s “smart beta” funds are offering .‘Twas ever thus.
Between them they perpetuate a market that is not random, but never quite loses its link with fundamental value, that is hard to beat, and does a better job than socialist planners.
john.authers@ft.com
Copyright The Financial Times Limited . All rights reserved. Please don't copy articles from FT.com and redistribute by email or post to the web.