So far I've posed the question ‘Can the stock market be predicted?' Can economic forecasts, for example, be used to predict where the stock market is heading? I hope by now you've come to the same conclusion as me: they can't.

Now I want to turn things around. Rather than asking whether economics can predict movements in the stock market, let's ask whether movements in the stock market can predict the economy. That's right. Can what was previously being predicted deliver a prediction on what was being used as the vehicle for the forecasting? The logical question is, how is it possible that each could forecast the other? It sounds ludicrous but that's exactly what many people propose and believe.

Here's a comment I recently read in the investment journal Equity in 2012:

Technical analysis is particularly useful in providing underlying information about the market perception of the health and probable future price movement of a stock. This technical approach also applies to a national stock index which gives a forward glimpse in the economic activity of a country perhaps up to nine months into the future.

Let's consider whether there's any basis to this belief. Firstly, the stock market is nothing more than a mixing pot of the combined beliefs of the multitude of individuals interacting with it. They back their conviction with their chequebooks, and the weight their vote carries depends on the size of the cheque they write. Intuitively this questions how the market could be expected to have predictive powers. We know that the individual investors and those who influence the investors — the analysts, the fund managers, the brokers, the market commentators, and even the politicians and the economists who choose to deliver economic commentary — have little idea where the economy will be in six to nine months. So how can their combined consideration deliver what they can't individually? Surely individual ignorance can't be transformed into collective wisdom.

Yet the popular belief that the stock market is a forecaster of the economy has been around for a long time. I don't know who first proposed it, but Thomas Gibson had this to say back in 1910: ‘Security prices always move ahead of basic developments, never after or in connection with such developments.'¹⁶

It's not an unrealistic observation that stock market crashes tend to precede economic downturns, or that stock market recoveries tend to precede better economic times. But does this demonstrate a capacity to forecast, or is it simply an association? For example, could it be that a stock market crash is the initiator rather than the predictor of an economic downturn? Or maybe the stock market upturn initiates a sense of improved economic wellbeing, which then leads to an economic recovery.

This was the view held by author Henry Hall. Early last century Hall wrote that, while the stock market appeared to precede economic conditions, it could be that the stock market was causing the economic effect rather than predicting it. He stated: ‘It is to be noted, as a distinct and logical phenomenon, that a decline in stocks antecedes an actual turn downward in business, and sometimes brings about the crisis.'¹⁷

It seems Hall's conclusion that a decline in stocks ‘sometimes brings about the crisis' was nothing more than his personal opinion. Nevertheless it ran counter to the belief of many people in both his day and ours. In my mind Hall's view remains a highly credible one. Which comes first, the economy or the stock market? Which is the horse and which the cart?

Another who considered this issue was unemployed securities dealer turned author Julian Sherrod. Sherrod took to writing in the wake of the 1929 Crash after losing his job as a bond salesman for National City Bank. In his 1931 book Scapegoats, Sherrod lambasted the Wall Street excesses leading up to the 1929 Crash. He wrote: ‘To say that the Depression has caused the things which I have been discussing is not only putting the cart before the horse, but it is turning the horse around also and facing him north while driving south.'¹⁸

Sherrod felt the 1929 Crash and the Great Depression were the direct result of the excesses preceding the Crash. The Crash didn't predict anything. It was a catastrophe just waiting to happen.

In September 1930, around the time Sherrod was writing Scapegoats, Richard Whitney, the President of the New York Stock Exchange, delivered an address to the Merchants Association of New York. His topic was ‘Trade Depressions and Stock Panics'. Whitney blamed the unfolding business slump on the excessive production and artificially high prices prior to the Crash. Like Sherrod, he used the horse-and-cart analogy, stating: ‘To attribute business depressions to stock market panics is to place the cart before the horse.'^a

He proposed that stock prices were merely a barometer of business conditions and a response to, rather than a cause of, economic conditions. In his mind there was nothing predictive about the Crash since deteriorating business conditions preceded the Crash by five months. On that point Whitney was right. In the months leading up to the Crash, steel output and auto production were falling.

But if stock prices respond to the real economy rather than predict it or affect it, why can subtle shifts in the economy result in such dramatic shifts in stock prices? And why is the deterioration in economic conditions typically amplified in the wake of a significant crash, not before it? Why did the Long Depression^b of 1873 to 1879 follow the Panic of 1873? And why did the Great Depression of the 1930s follow the 1929 Crash?

Charles Dow had expressed his thoughts on this matter 30 years earlier. In his editorial of 10 May 1900 he said:

The stock market discounts tendencies. Stocks went up before the improvement in business became pronounced. Stocks will discount depression before depression actually exists, but this discounting quality in stocks makes them run to extremes. They discount shadows as well as substances and often anticipate that which does not occur.

Dow is suggesting here that the stock market reacts both to expected and to current alterations in the economy. But, equally, markets can overreact. So a very large correction can result from subtle shifts in economic data. I might add that an overreaction is more likely when the market is substantially overpriced to start with, as it so obviously was in 1929.

So we've now come full circle. The cart is chasing the horse. And so it goes around and around.

Let's see what Jesse Livermore, the cunning and perceptive reader of shifts in market sentiment, had to say on the matter. Livermore published his share trading secrets just before his death in 1940. In How to Trade in Stocks he wrote:

It is therefore foolish to try and anticipate the movement of the market based on current economic news and current events … these facts were not facts I could ever use to ‘PREDICT' the market.¹⁹

Prominent economist and US fund manager John Burr Williams was interested enough in the horse-and-cart issue to devote an entire chapter of his 1938 book, The Theory of Investment Value, to the subject. His views were similar to those of Dow and Whitney. He pointed out that the stock market did not issue the first signal. Shifts in the volume of orders for new goods in the real economy preceded the stock market's reaction. This meant the first investors to react were merely responding to what they sensed, read about or could see for themselves — that is, real shifts in economic activity. All the other investors then just responded to the price moves that resulted from their selling. These price moves began as a trickle but ultimately became a flood. The early economic signs from the real economy, and the downward drift in share prices from their peak, are effectively the ‘canary in the coalmine' preceding the main brunt of a crash. But it's the crash that grabs everyone's attention. Williams concluded the stock market had no capacity for prophecy at all.

This line of thinking is akin to that of modern-day investing marvel Warren Buffett. Buffett has often stated that the economic statistic he would choose if he were stranded on a desert island for a month and could get only one set of economic numbers would be railway freight car loadings. The Association of American Railroads reports the amount of raw materials, inputs and supplies moving around the country every week. And since the inputs transported by rail eventually get processed into inventory, final output and goods for sale, freight loadings should help flag early shifts in the direction of the overall economy.

So let's do a stock take at this stage. It's been observed by some that the stock market moves ahead of the economy. Others have said it doesn't. Some say the stock market is a predictor of economic change. Some suggest it's not a predictor of change but rather the catalyst. And some suggest that the stock market does no more than respond to current economic conditions, but in doing so can overshoot in either direction.

All clear now?

This whole issue has never really been put to bed. Many investors today still believe the market possesses predictive powers. I think a big part of the problem is the general inability of people to distinguish between association and cause. At university my epidemiology^c lecturers used to ram down my figurative throat that simple association does not prove cause and effect. To repeat an analogy I made earlier, bald men are more likely to wear hats but that doesn't mean hats cause baldness. And even if an association can be demonstrated, it could be a classic case of — dare I say it again — putting the cart before the horse.

So let me finish the discussion this way. The popular belief is that the stock market is capable of predicting the future. That's wrong. The stock market is people, and people can't predict beyond their next coffee break. The reality is that stock market participants attempt to predict the future. The difference is night and day.

The relationship between the real economy, stock market prices and participants in the stock market is far more complex than simple statements such as ‘the stock market forecasts the state of the economy six to nine months ahead'. How participants in the stock market respond to economic signals is complex. How the economy reacts to the signals issued from the stock market is complex. There are a multitude of variables. How significant is the economic news? How sensitive are participants to that news? How overstretched are stock prices?

Stock market crashes can trigger economic downturns. Stock market crashes very rapidly strip people of wealth, destroy confidence and seize up lines of credit. They lead to bankruptcies, which lead to job losses and factory closures. If that isn't enough to bring on an economic downturn then I don't know what is. I'm proposing that stock market crashes have the capacity to bring on significant economic downturns, but not to predict them.

The problem is that people observe that crashes often precede downturns, and they assume the crash ‘predicted' the economic downturn. But the dictionary shows us that the words preceding and predicting have very different meanings.