“Nobody, neither the experts nor us amateurs have the least idea what is going to happen with the economy in the future… I don’t think the economists really know”, argued former British Conservative member of parliament, Matthew Parris, in an interview with BBC World Business Report. “I don’t see why we should accord to them any more authority than we accord to palm readers or astrologers. They’re a bit of fun and they sometimes get it right, but it’s not a serious science.”

Though many economists would beg to differ that the complex mathematical models they employ in attempt to forecast the future are unscientific, sentiments such as these are common among those that see the futility in trying to predict the upticks and downturns of the economy. Though it has done little to prevent people from trying.

In the interview, the former MP did eventually conceded that economists and forecasters are proficient at predicting certain things, but that sadly for us, none of them tend to be very useful. “They’re very good at looking into the far distant future and predicting very long-term trends, you know, looking at the way China is growing looking at the way [the] European manufacturing industry has been struggling and predicting a change in balance over decades”, says Parris. “But the one thing where we most [need] economists is where they most regularly fail, and that is not saying what is going to happen tomorrow, but what is going to happen next week, next month, next year.”

In short, forecasters are very good at explaining why the proverbial horse has bolted after the fact, but not very useful in providing insights prior to the event in order to prevent it from happening in the first place. But why do they struggle to get it right where it counts, and will they ever be able to predict the future – or will they always be at the mercy of the market?

Different perspectives
One man that seemed to have succeeded in the business of financial fortune telling where others had failed was Jerome Levy. Back in 1929, the businessman and economist managed to see what so many did not; leading him to making the right call and pulling out of the positions he held in the market just ahead of the October crash. While you would be forgiven for thinking his decision that day was merely a fluke, he and the consultancy company that now bears his name would be more than happy to argue otherwise.

The profit perspective he developed in order to make that impressive judgement call at the beginning of the 20th Century is the same model used by his grandson David, and the company he now runs – The Jerome Levy Forecasting Center – that successfully predicted that the next recession would be caused by a deflating housing bubble.

The company then went on to correctly predict that shortcomings in the subprime-mortgage market would eventually infect “virtually all financial markets”, and be the catalyst for the global financial crisis that broke out in 2008. The company clearly boasts an impressive track record, which is why when it released a forecast predicting, with a 65 percent probability, that a worldwide recession would force contraction in the US by the end of 2015, it got a number of people’s attention.

But how does the profit perspective work, and why do their forecasts fly in the face of those made by investment banks, such as Morgan Stanley and Goldman Sachs, who believe the polar opposite? “We are always a little contrarian, but the most important thing to remember is that we look at the economy from a completely different perspective”, says Srinivas Thiruvadanthai, Director of Research at The Jerome Levy Forecasting Center. “Most Wall Street economists, in fact most economists in the forecasting business, » use conventional macroeconomic forecasting, which relies on large-scale macroeconomic models.

“The problem with them is that they are by their very nature, extremely linear and do not pick out the turning points.” Such models, he argues, tend to focus attention on deviations from trends or what is known as full capacity growth. This method tends to look at the world using a certain lens, which sees economies as either operating above or below full capacity.

This means if a particular economy is running under this benchmark (known as underproduction), it has room to expand, which will lead to a rise in GDP, but when full capacity is exceeded (known as overproduction), the opposite will occur. But it is this same lens that investment banks use, and which leads them down the wrong road, often restricting them from seeing the complete picture. “In reality, if you think about the economy, GDP is not what drives it; what drives the economy is profits, because corporate businesses making profits and their expectation of profits is what drives those same companies to hire new staff and to invest in the economy, which is what really promotes growth.”

This difference in perspective in how they view the economic landscape is what allows the half a dozen experts at the centre to draw more accurate conclusions from the vast amounts of data available and, therefore, make better predictions on the future state of the economy.

The profit perspective
The forecasting centre admits in its publication Where Profits Come From, which explains its models methodology, that it is no secret that profits are a huge motivation for many activities in a capitalist economy, with them playing a large role in the decision making process of individual companies on issues, such as whether to expand or contract production, employ more staff or let people go, as well as being the guiding force behind most investment choices.

Profits, therefore, are crucial in microeconomic models, which attempt to make predictions about the economy through the study of individual markets and the companies, workers and consumers that compose them. But the Levy Center takes this notion a step further. It believes that in the same way an individual company’s profits play a vital part in dictating its behaviour in decision making, aggregate profits – the combined profits of all corporations – have a huge influence on the direction and behaviour of the overall economy.

“Production, employment, and capital spending for the economy as a whole are strongly influenced by aggregate profitability, therefore understanding the determinants of aggregate profits leads to powerful insights into these activities and other economic phenomena, including inflation, unemployment, and business cycles”, claims the report. “Yet conventional macroeconomics, the study of the economy as a whole, rarely considers the role of total profits.”

The Profits Perspective model is one based on what the centre calls a direct flow-of-funds analysis, rather than on statistical approximations of the economy. Meaning that their forecasting model involves no outlandish mathematics, no unrealistic assumptions about the nature of human behaviour, and no unrealistic expectations about how companies, consumers, and investors operate.

Describing the Profits Perspective model in this manner makes it look rather simplistic, but that is not necessarily a bad thing. The 2008 crash occurred at a time when economists were using supercomputers and highly complex models, but they still managed to miss what David Levy and his team spotted. It would appear simple models are far better than complex ones. Though that has not made their forecasting framework completely immune from making miscalculations. Only four years ago the Levy Center made a prediction that was very much along the same lines as the one they released in October 2014.

In it, they predicted a 60 percent chance of a US recession occurring. Sound familiar? In the end they got it wrong, but issuing a probability of a US downturn with just over a 50/50 chance at a time when the American economy faced challenges, such as high unemployment and a financial industry in the midst of huge regulatory reform, is hardly an impressive call.

Prophesying the eventual downturn of the economy appears ever more pointless, when you take into account the fact that boom-and-bust cycles are built into the fundamental framework of the global financial system. This means that making predictions about its eventual demise are as impressive as foreshadowing someone’s death; it’s going to happen eventually, so why not keep firing out the forecasts, at some point one will hit the mark.

Predicting the unpredictable
The bust that inevitably follows the boom is brought on by the most chaotic of forces – basic human nature. It is this feature of mankind that contains within it one of the major factors for why economists’ attempts to predict the future are often unsuccessful. During the booming years when the economy is stable, businesses, governments and the individuals that comprise them easily become overly optimistic as the good times continue to roll on.

Such sentiments lead to overstretches on financial commitments, which leave them – and the larger economy that they are all a part of – more vulnerable to sliding into free fall. Stability, it seems, is inherently unstable. This paradoxical point is made clear when looking at government attempts to intervene in unpredictable markets in a bid to steer away from the cliff’s edge, only to have their best efforts eventually fail.

“Any kind of policy will work for period of time like a charm, but it will lay the seeds for its own demise, because if it is working the way it should then people will take that into consideration when making their plans, which in turn undermines the effectiveness of the policy”, explains Thiruvadanthai. “To give you an example, suppose you do a great job of stabilising the economy, as an investor I will think, ‘oh look; the economy is more stable; I can take on more risk’ because the possibility of a downturn is lower, but in doing so I will undermine that stability.”

This concept is explored within Hyman Minsky’s Financial Instability Hypothesis, which helps show that one of the fundamental issues when trying to navigate a complex financial system to safety, is that no matter what kind of policy you introduce, within a short space of time the system will incorporate it into its own behaviour, undercutting policy and rendering forecasting models useless over time.

“From time to time, capitalist economies exhibit inflations and debt deflations which seem to have the potential to spin out of control”, says Minsky in his thesis. “In such processes, the economic system’s reactions to a movement of the economy amplifies the movement – inflation feeds upon inflation and debt deflation feeds upon debt deflation.”

Paul McCulley, the economist and former managing director of investment management company PIMCO, called this process a ‘Minsky moment’, with the greater significance of the stability/instability hypothesis being that it demonstrates human beings’ propensity to make investments based on momentum, rather than value.

Back in 1929, entrepreneur and business theorist Roger Babson may have unwittingly played up to our penchant for making decisions in this manner when he said: “Sooner or later a crash is coming, and it may be terrific… factories will shut down… men will be thrown out of work… the vicious circle will get in full swing and the result will be a serious business depression.” His remarks came just before the crash and may have provided the necessary momentum to tip the system over the edge.

The fact that human beings make decisions in this manner poses a real challenge for economists when trying to develop forecasting models that are capable of making sense of these irrational forces. But there are some that believe a solution might be found through the application of computer software, enabling us to unlock the potential of big data. Unlike many other fields, financial forecasting is one that is never short on data. There is plenty of information available of people making actual decisions to buy and sell, as well as a host of other information for economists to base their models upon. The real problem they face is making some kind of sense out of all the vast reams of information they have at their disposal. Big data, therefore, is both a blessing and a burden.

Big data
Data sets this large and complex offer those that study them the opportunity to gather new insights, but also have the potential to overwhelm. As any economist knows, successful forecasting relies heavily on one’s ability to separate the signal from the noise. Google Trends (see Fig. 1) has been purported to have the potential to do just that, helping economists to unlock the underlying power of big data in order to assist them in making more precise predictions about market moves before they happen – something that, if achieved, would impress even the likes of Matthew Parris.

Google Trends has been in the news a lot recently as it has been hailed as having crystal ball-like properties, with success in tracking Influenza throughout the US, as well as being able to predict the next platinum album before its even been released. And now, a study carried out by Warwick Business School argues that the technology, which uses complex computer algorithms to pick out patterns in peoples search terms on Google, could be used to identify falls in stock market prices, acting as an early warning device for the next financial crash.

In their paper, the researchers explain how technology has grown to become “deeply interwoven into the fabric of our society”, with the internet being the “central source of information for people when making day-to-day decisions.” What they found was that, by running analysis of search terms made by users on Google and online encyclopaedia Wikipedia, they were able to find “evidence of links between internet searches relating to politics or business and subsequent stock market moves.”

Most interestingly of all, they found that when a spike in these searches was observed, it often preceded a fall in market value. In simple terms, their study showed that when people lack confidence or are unsure about something, they tend to go online in order to make sense of things. Their data provides that, statistically when you see a rise in these types of searches, it implies uncertainty, and that this lack of confidence then correlates to subsequent dips in the market.

Their method worked, but just like instances of market intervention through monetary policy, positive results were short lived. “We [found] that the predictive value of these search terms has recently diminished, potentially reflecting increasing incorporation of internet data into automated trading strategies”, the paper concluded. The market appears to want to remain unpredictable.

Cause and effect
In the same way that governments may introduce new fiscal policies in order to improve the economy, investors weaken its effectiveness, in this case by High Frequency Trading (HFT) platforms, incorporating it into their own investment strategy. When a method is devised to observe potential falls in the market, the market eventually incorporates those same strategies, transforming the original application into something more akin to of a magic eight ball, rather than one made of crystal.

In their conclusion, the researchers propose that their forecasting model’s ‘predictive value’ may have diminished, as a result of automated trading platforms incorporating similar techniques into their investment strategies. What this shows is that their model worked while investors were in the dark about the methods they were using in order to make predictions on the market, but that once investors observed the techniques and then incorporated similar practices into their own strategies, their forecasting model became compromised. But why? An answer for its sudden inability to draw conclusive results could be found in different interpretations of quantum mechanics.

In 1957, Hugh Everett came up with the many-worlds theory, which argues that observation changes the outcome or rather, the probability of that outcome. How this relates to forecasting, is that while it may be theoretically possible through quantum computing to filter out noise in the markets, there is a catch, which is it could never be made public knowledge, because once observed the odds would have changed, compromising the forecasting model.

In his book, The Future of Everything, economist David Orrell argues that economic systems are unpredictable as they are too complex. “One of the lessons of complexity science is that systems can be what Stephen Wolfram calls computationally irreducible.” Wolfram contends that the behaviour of complex systems – though composed of simple structures – is capable of throwing up a wide range of behavioural variables, making it impossible to predict the outcome with any degree of certainty, but despite all this it will do little to deter people from the path of prediction.