How to Predict Brexit - the Human Face of Market Risk

The Human Face of Risk

"In fact, it was as a journalist who played around with the facts that Mr. [Boris] Johnson first made his name. He was fired from his first reporting job, at The Times of London, for inventing a quote and attributing it to an Oxford professor (who happened to be is godfather)."
~ "Luck Runs Out for a Leader of 'Brexit" Campaign." New York Times, June 30, 2016.  

Boris Johnson is a former Mayor of London, Member of Parliament (MP), and was a leader of the Brexit campaign.  He was poised to become the next British Prime Minister until engaging in odd behavior last week.  

Johnson started his working life in the media.  In 1989 he was assigned to report on the EU from Brussels by the Daily Telegraph where he misrepresented the dysfunction of the EU in order to gain readership, and he enjoyed the public's shock at his stories:  “I was just chucking these rocks over the garden wall and listening to this amazing crash from the greenhouse next door over in England,” he told an interviewer.  Some of his misrepresentations about the EU became conventional wisdom in the UK, and they may have stoked the euroscepticism behind the Brexit vote (and the subsequent global market declines).

At MarketPsych we research how media information moves investors and market prices.  Some information that hits markets is well-defined, such as the impact of the nonfarm payrolls number described in this past newsletter.  When information fits an understood model, it is taken in stride, and the markets stay calm.  But sometimes events shake our fundamental beliefs about the world, and new frameworks for understanding events are needed.  Today’s newsletter examines human risk in markets - the urge to independence - and a unique media-based trading model that correctly predicted the market direction around the Brexit vote.
 

How Brexit Was Predicted

 
While most of the establishment (including many of us at MarketPsych) were surprised by the "Leave" outcome of the Brexit vote, the only active trading model we're involved with bet on it.

We've been collaborating with Stanford-trained data scientist Tayyab Tariq for about a year.  Tayyab develops machine learning models both for natural language processing and also for financial prediction.  We mentioned some of his findings in our book "Trading on Sentiment" (Wiley, 2016).  Tayyab developed a machine learning model that trades the S&P 500 directionally daily based on the tone of the media as quantifed by our Thomson Reuters MarketPsych Indices (TRMI).  Trading a small account, Tayyab’s model has been generating outstanding returns since its launch on February 22nd.  It correctly predicted S&P 500 activity around Brexit and is described in this LinkedIn post.  Tayyab's model is the only active trading model we're involved with.  That is to say, we're not cherry-picking here - we don't have 3 other models that did not predict Brexit.

Machine learning is a blunt tool when wielded in inexperienced hands.  Creating machine learning models is both art and science, and there are efficiencies that masters perfect over time in order to extract the most value from data (watch the online lectures by Andrew Ng to learn some of these).  Tayyab is a seventh-generation artist (literally) as well as a scientific genius who is truly a master of financial market prediction models.  We'll have more contributions from Tayyab in future newsletters.

The Urge to Independence

The Scottish vote to stay in the UK in 2014 (55% stay) was not the first nail-bitingly close independence vote in the past decades.  Quebec in 1995 decided to stay in Canada at a narrow 51% margin, and Catalonia has yet to vote (if at all).  More EU independence referenda are likely.  Why are so many willing to pay a personal economic cost for political independence?
 
Recently I saw a T-shirt emblazoned with the slogan: “Freedom Isn’t Free”, reflecting (I presume) the sense of sacrifice (of blood and treasure) and courage necessary to achieve independence.  Such a sentiment reflects why British voters would pay an economic cost to be free of EU regulations (or maybe racism drove the vote, but I'll take a high road and stick to the freedom argument here).  Perhaps 52% of Leave voters didn't value independence from the EU at 9.5% of national wealth (using the relative decline in the GBP/USD exchange rate after Brexit as a proxy for decline in wealth), but nonetheless the idea of independence was clearly valuable to Leave voters.
 
As Boris Johnson's own journalism showed, media stories can provoke public resentment against "the establishment" and foster a desire for independence.  Johnson isn't alone, in the New Yorker Magazine James Surowiecki speculates that Donald Trump is using exaggerated negativism compel voters to take a risk and vote for a politically unexperienced businessman.  And it is the narratives of skilled communicators like Boris Johnson and Donald Trump that help voters 1) Recognize how trapped they are, and 2) Give them the motivation to break free.

[W]arning of dire losses has been the core of Trump’s campaign. Free trade means that “we’re losing our jobs, we’re losing our money.” China’s trade practices amount to “the greatest theft in the history of the world.” We need a wall to stop illegal immigration because “we’re losing so much.” In Trump’s world, things are much worse than they seem, and it’s because American prosperity has been stolen: “We’re losing everything.”   James Surowiecki, New Yorker.

And the solution to these losses is the freedom from the rules and regulations that bind us into losing situations.

While voters' independence decisions have real world costs, the key question is whether the long term consequences are being adequately assessed.  Certainly short term there are real costs, as the declines in markets demonstrate after Brexit.  Similarly, US independence from England caused much short term pain. Longer term it was much better for the US economically.  In other situations the results are not so clear.  The costs of many countries' independence from the UK were probably negative in both the short and long term.  But if you poll residents of former British colonies, in nearly all cases the majority will agree that the psychological benefits of independence are worth far more than forgone economic gains.  Are global independence voters - those who want to overturn the establishment - simply pawns of skilled emotional manipulation, or is there some wise internal logic that guides their behavior?

Over the next decade one key to understanding markets will lie in understanding the psychological motivations for and the economic costs (and psychological benefits) of independence in a newly fragmenting Europe and the global community.

Changing Conceptions of Risk

That people would choose greater instability, independence, and less diversity - at an economic cost to themselves - may be baffling to economists, but not so to psychologists.  The science of understanding of how stories and feelings drive crowd behavior is reaching a new level of coherence with the advent of data such as our Thomson Reuters MarketPsych Indices (TRMI), Google Trends, and other crowd-sourced big data sets.  In future years, as Tayyab’s predictive models demonstrate, there promise to be many more exciting breakthoughs in the science of forecasting irrational crowd behavior.

We love to chat with our readers about their experience with psychology in the markets.  Please send us feedback on what you'd like to hear more about in this area.

Learn more about improving your investment returns with insights from sentiment analysis of the herd in our new book, “Trading on Sentiment:  The Power of Minds Over Markets.”

If you represent an institution, please contact us if you'd like to see into the mind of the market using our Thomson Reuters MarketPsych Indices to monitor real-time market psychology and macroeconomic trends for 30 currencies, 50 commodities, 130 countries, 50 equity sectors and indexes, and 8,000 global equities extracted in real-time from millions of social and news media articles daily.

Happy Investing!
The MarketPsych Team