Monday, 25 July 2016

Having more money helps the poor? Go figure!

The link between income and happiness has been one of the most hotly contested relationships among social scientists. Many researchers from various branches of social sciences have repeatedly denied the existence of a causal relationship between income and happiness (despite overwhelming evidence on how the great rise in living standards was led by the significant increase of post-Industrial Revolution incomes per capita), with the obvious normative conclusion being that having more money is not that important to us after all. 

Well it just happens that it is important.

In an article published last month in the New Yorker entitled "The Case for Free Money", the author James Surowiecki (the Wisdom of Crowds guy) makes a compelling case in favor of universal basic income - a policy proposal (rejected in a referendum in Switzerland earlier this year) where every adult citizen of a country would each year receive a guaranteed basic income. In the US this would be equivalent to $10,000. While there are many potential benefits to this idea (as well as some major obstacles - such as it being too expensive), I do not intend to contemplate on the basic income proposal in this post. I will leave that for another time. What caught my attention was the opening paragraph where Surowiecki calls upon a research paper from a Canadian economist Evelyn Forget who examined the effect of a field experiment being done in Canada in the 1970s. Here's the excerpt:
"In the mid-nineteen-seventies, the Canadian province of Manitoba ran an unusual experiment: it started just handing out money to some of its citizens. The town of Dauphin, for instance, sent checks to thousands of residents every month, in order to guarantee that all of them received a basic income. The goal of the project, called Mincome, was to see what happened. Did people stop working? Did poor people spend foolishly and stay in poverty? But, after a Conservative government ended the project, in 1979, Mincome was buried. Decades later, Evelyn Forget, an economist at the University of Manitoba, dug up the numbers. And what she found was that life in Dauphin improved markedly. Hospitalization rates fell. More teen-agers stayed in school. And researchers who looked at Mincome’s impact on work rates discovered that they had barely dropped at all. The program had worked about as well as anyone could have hoped."
I went through the paper and it was indeed a proper field experiment (lasting a total of 4 years): they had random selection of participants, paired control group from the same community, isolated families, etc. In other words, it was legit, making the conclusions of the paper correct. Having more studies and field experiments just like this one would make the case for the universal basic income even stronger (although, the main problem was the lack of money to carry out the experiment further than 4 years).

However, you can't tell any of this from the paragraph above as the author makes an inference mistake. He should have said that hospitalization and schooling outcomes improved compared to the relevant control group. This way a reader might suspect that the research was not done properly if it only looked at the pre- and post-basic income outcomes in that city alone (these alone would not tell you much about the causal relation). I, for example, had immediate doubts about the validity of the research, which is why I had to read the paper myself. But as I said, the paper is legit. 

For me the most interesting finding this experiment suggests is that by handing out more money to people, the policy entailed positive effects on their quality of life. The basic income improved peoples' health and social outcomes. In other words, it made them happier. So without any unnecessary quibbles about the existence or non-existence of the correlation between money and happiness, by using basic income experiments we can finally empirically establish a positive relationship between money and happiness. Not directly, as few people enjoy money just for the sake of money, but indirectly, as money helps us attain a better living standard. 

My point is, if you advocate universal basic income, don't contradict yourself by saying that money doesn't make people happy. I certainly does. If not directly, than indirectly (just look at the leftist propaganda meme below). For one thing having money means less worries about it. That fact alone improves happiness.  

Tuesday, 12 July 2016

Why is the US white middle class going rouge? ... and voting for Trump

A recent paper published in PNAS by Princeton professors Anne Case and Angus Deaton (last year's Nobel prize winner) reported a stunning finding: there has been a significant increase in mortality and morbidity among middle-aged non-Hispanic white men and women in the US over the past 15 years. The causes have been attributed to suicide, and drug and alcohol abuse. The graphs below summarize the findings:
Mortality in selected countries, 1990-2013.
USW stands for US white. USH stands for US hispanics.
Source: Case & Deaton (2015) "Rising morbidity and mortality in
midlife among white non-Hispanic Americans in the 21st century."
Proceedings of the National Academy of Sciences Vol. 112(49)
December 8th 2015.
Mortality by cause, US white non-Hispanics, 49-54. Source: Case & Deaton
(2015) "Rising morbidity and mortality in midlife among white non-Hispanic
Americans in the 21st century." Proceedings of the National
Academy of Sciences
112(49) December 8th 2015.
Every comparable developed country has experienced a steady decrease in midlife mortality over the past two decades, while the trend for whites in the US has suddenly switched upwards in 1999 and has been steadily increasing ever since. The second graph illustrates why this is so. Deaths from poisonings have gone way up (over 300%) over a time span of only 15 years. Liver diseases and suicides have also gone up. Furthermore, the turnaround in mortality has been driven primarily by whites with a high school degree or less. Those with college education or less remained the same, while those with BA or higher have had death rates decrease. All this suggests that mid-life low-skilled workers, after experiencing job losses, have descended into personal depression and are trying to cure this with drug and alcohol abuse. And if you think that drugs are too expensive for an average unemployed person in the US, think again. The popular TV series Breaking Bad has showed that getting your hands on things like crystal meth is becoming widespread across the low and middle classes in the US (I know it's a fictional TV show, but it does draw implications from the real world). 

Why are these people descending to such states? Why are they killing themselves with alcohol and drugs? A simple explanation is economic, although not necessarily tied to the financial crisis since the trend started during favorable economic times. However the productivity slowdown and the outsourcing trend (a trend that accelerated in the 1990s and particularly in the 2000s), have made their own dents among the low-skilled population. These are not just job losses we are talking about. It's also the prospects of finding a new job for the low-skilled workers (which is perceived to be difficult given immigrants which are prepared to work for much less), as well as their lagging incomes, widening inequality, and disturbed social mobility. However most developed countries have experienced the same problems in recent decades, but these problems haven't manifested themselves into higher death rates for the mid-life low-skilled workers. 

So what gives? Perhaps the fault lies within the US health care system that is out of reach for the poor and middle classes (particularly if unemployed), making them more likely to die from the conditions they've imposed on themselves. In other words, perhaps European low-skilled workers are reacting the same way to globalization-induced job losses (via drugs and alcohol) but are simply more likely to be cured by their free-of-charge public health care systems. That still, however, doesn't explain the increase in suicide rates in the US. Perhaps the inequality and social mobility problems are even worse than we think which is forcing people over the edge. Financial insecurity could also be the issue, as pensions are traditionally tied to stock market earnings. But it's always been this way, and it wasn't a problem before. On the other hand, there has never been such a prolonged period of stagnating low and middle class incomes, so once a low-skilled worker losses his/her job, not having any significant savings accumulated (recall that the savings rate in the US is in steady decline since the 1980s), and possibly overburdened with a home loan, there isn't really much else he or she might hope for. Finally, perhaps the US cultural values emphasizing individual success are taking a psychological toll on its population for the first time in history (or since the Great Depression) as they are unprepared to deal with failure. I'm tapping in the dark here, but one thing is certain - something has definitely gone wrong in the US. 

It is therefore perfectly understandable why the white mid-aged US low and middle classes are switching to populism and are grouping behind Donald Trump. This has been long time coming. Populist messages blaming immigrants on job losses and a wrecked benefits system is exactly what the low-skilled workers have been craving for. Looking at these graphs it is easy to understand why Trump's support has been so high despite all of his political incorrectness and even his billionaire stature. Trump is delivering loud and clear messages that resonate with a part of the electorate that seems to be fighting with various forms of psychological depression and which represent clear losers of globalization (note: globalization certainly has a lot of benefits, but it also has its inevitable losers - for developed countries these are the holders of the scarce resource - the low-skilled workers). Furthermore, the political establishment has failed to hear these voices over the past decades (not only in the US). In such a state, where you lose your job and blame immigrants for it, where you have no savings and no long-term prospects, when you're faced with foreclosure on your home loan, when you realize your children will probably be facing the same bleak future, and finally when the politicians you never really liked anyway are not even hearing your messages, you very easily descend into populism and accept the first loud enough rhetoric that's expressing everything you're concerned about. It is no wonder these people "want to make America great again". For them, America has failed. Their intrinsic voter motivation is in this case actually quite rational. 

Europe is faced with the exact same problem of rising populism resonating to low-skilled losers of globalization, despite its lower mid-life death toll. The most dangerous way of fixing these problems facing low-skilled workers is descending into populism. Let's hope the establishment will realize this and offer realistic solutions before it's too late. 

Tuesday, 28 June 2016

BREXIT, THE REACTION: democratic deficit, the falling elites, and the future of the EU

After hearing the results of the Brexit referendum on early Friday morning, my initial reaction was this comment on Facebook:

After cooling down over the next few days and reading about Brexit from a number of perspectives, I have to say that everything I said initially - still holds. I will carefully explain each point.

First, the democratic deficit problem. This is, sort of, my PhD topic, meaning that I'll be writing quite a lot about it over the next few years. Before explaining why this outcome is a deficit of democracy we must first define the concept of democratic deficit (or perhaps even - democracy failure). Basically, the democratic deficit concept concerns the interaction between politicians and partial interest groups, a legitimate consequence of electoral competition and political freedom to express and fight for one’s interests, and whether or not this interaction results in adverse economic outcomes (in my PhD I will be focusing on linking the failures of democracy to the rise in income and wealth inequality). The point is that in some cases democracies fail to erect the necessary institutions to prevent the favouritism of partial interests, the consequence of which are usually cronyism, corruption, nepotism, clientelism, etc. To make this a bit more relevant, consider the following excerpt from an earlier essay of mine
"The recent economic crisis has exposed all of democracy’s deficits. Dysfunctionality and political gridlocks that only worsened the crisis became a standard in the United States and Europe. Government bank bailouts and rapid accumulation of debt stroke a huge blow to the positive perception of Western democracy and capitalism itself. In addition, the Western model of democracy is facing a serious problem with rising inequality and to some extent the lack of social mobility. Various interest groups dominate the political spectrum in biasing budgetary expenditures towards their preferred goals leaving relatively less money for redistribution programs aimed at the poorer ends of the society, particularly in terms of education and health care. Politicians themselves engage in direct or indirect vote buying (either through gerrymandering or by giving direct concessions to their support groups), budget-maximizing bureaucrats add to the rise in government spending which isn't targeted towards the general population, while political campaigns are financed heavily by the corporate sector desiring prone legislation. All of this adds concerns over a poor image of the Western-style democracy. It has failed to become fully robust to cronyism."
So how is this linked to the Brexit? In two ways: first, the 'angry' Leave voters which consider themselves the losers of globalization (the 'immigrants taking our jobs argument'). Globalization always has its winners and losers, this is inevitable. As the theory of international trade teaches us the losers are usually the holders of the deficient resource within a country - in the case of the West: the blue-collar working class. Naturally as the working class keeps losing their jobs they have a tendency to blame both immigrants and their domestic political elites which have failed to protect them from these misfortunes. The European Union is a natural enemy of anti-globalizationalists. It represents everything they fear - in particular their intrinsic loss of competitiveness as the market attracts the better skilled workers to replace them. This is true for every country in the West, without exception. There is growing discontent from the working classes regarding the benefits of globalization and its prime manifestation - the European Union. They feel alienated as they did not reap any visible nor direct benefits from it. Naturally, they will rebel and oppose globalization and will support any political platform (far-right or far-left) that delivers the same bold criticism. 

However, the working class voters often fail to realize that the reason for their economic misfortunes lie beyond globalization itself and can be traced within their domestic institutions, in particular within their domestic political elites. How exactly? 

I've written about this before as well. Globalization, in addition to bringing enormous opportunities for wealth creation, in some instances led to the abrupt rise of new powerful elites - banking, political, and media - which threatened the sustainability of the system. The principles of competition were replaced by the rapid accumulation of power within a handful of selected business groups (recall the Leveson inquiry or the LIBOR scandal). British political elites forgot the distinction between being pro-business (favoring monopolies, oligopolies and picking winners) and pro-market (supporting competition and equal opportunity). This was not, as many falsely believe, a consequence of Thatcher's reign as she vehemently opposed any partial big business interests that undermined the interest of the customer, as well as the rapid accumulation of power within any industry, particularly within banking or politics. The post-Thatcher political leadership forgot those lessons and set the country on a path to increasing cronyism, rising inequality and declining social mobility, all of which was further emphasized by the financial crisis, the post-crisis austerity approach, and the subsequent long (double-dip) recovery. Luckily, none of these undermined Britain's vast accumulated wealth, but they had certainly limited its growth and split its population into the winners and losers of globalization. 

The second way in which Brexit represents a democratic failure is the huge political (and even constitutional) instability and economic uncertainty arising as a consequence. The markets' reaction is only a natural response to the huge uncertainty surrounding Britain and the EU. So the paradox here is that the legitimate (and direct) decision of the electorate (the very essence of democracy) has undermined the economic and political stability of the country. Some say this won't last too long and that the long term consequences are likely to be positive for Britain, however it's hard to see how that will be achieved given that the very country, the United Kingdom, may fall apart itself (if the Scottish and Northern Irish enact their independence referendums). 

So, the failure of elected political elites to suppress cronyism and the subsequent decision of the electorate that undermines the country's political and economic stability can both be considered good examples of how democracies need not always yield optimal outcomes. 

Second, the decline of trust in establishment elites, and the rise of populism. This is closely linked to the previous points. The failure of the political establishment to prevent cronyism and the rise of new powerful elites has severely deteriorated the people's trust in institutions and expert opinion. Whether domestic (the Treasury, Bank of England, prominent UK Universities,  and think tanks), or international (the IMF, World Bank, any EU-related institution) - they were all considered to be untrustworthy during the Brexit debate. Why? One reason was the use of populist ideology to unsettle the electorate accusing the experts (particularly the economists) of having a vested interest as they tend to receive a lot of money from the EU. This argument worked well having in mind the anger against the establishment, as expert opinion was seen as a justification of establishment policies and was therefore considered to be untrue. Whatever argument was being thrown into the debate, not a single one had the allure of impartiality. On the other hand the establishment (and the experts to some extent) made their own mistakes of ignoring the concerns of their electorate, in particular the low and middle classes. This led to the further deterioration in the relationship between the elites and the "non-elites", implying even lower levels of trust among the non-elites. Unfortunately, there seems to be no signs of this damaged relationship to be improved any time soon. 

Third, Cameron's political legacy. Boy, did he mess up! This time last year he was in heaven. He had just secured his second mandate by a landslide electoral victory, removing his Conservative party of the LibDem coalition shackles, and was on course to bring a truly positive political legacy to Britain. After winning two general elections, and securing a major political win with the Scottish referendum, he decided to tie his political career, not to mention the future of his country, to the Brexit referendum. Just like Blair destroyed his positive legacy with the Iraq war, Cameron did perhaps even worse with the EU referendum. A man who had it all he ever wanted, just lost everything in a crazy gamble with his own party and what is mostly his own electorate. Not unlucky, just plain stupid.

Fourth, Boris Johnson as the new PM, the spread of populism, and the tectonic changes that await us. Boris Johnson is in a very difficult position currently. He appears to have gotten exactly what he wanted - Leave won and now he's the first favorite to take over the party and the Prime-ministership in October. Which is four years earlier than he'd hoped. However his victory seems awfully bitter to him as well (judging from his reactions and, well, his body language). It is as he had thought: "there is no way Leave will win, but I will make myself the leading figure of the campaign, and use this to take over the party with its looming euro-skeptics after Cameron quits" (which he announced to do prior to 2020, and said he was not seeking a third term). Sounds like a perfect plan, particularly as Boris would have no problem winning the generals in 2020 from the lackluster Labour leader Jeremy Corbyn (who is also likely to face a leadership challenge soon). Except that it backfired - Leave won! So Boris got what he wanted, he's becoming the PM, why isn't he happy? Perhaps because he realizes he probably won't be PM for too long. Time for another Batman quote (this time from the Joker's conversation with Harvey Dent):

As for the consequences and the spread of populism, my immediate fears are not with the EU, but on the upcoming Presidential elections in the US in November. There you have another typical establishment candidate, Hilary Clinton, which would in normal times, given her enormous experience (First Lady, Senator, Secretary of State, etc.) be a shoe-in to lock the victory. However she is going up against the worst possible manifestation of cheap populism, lies, and low-class appeal - Donald Trump. Who would have figured that an 80s style billionaire would speak to the mind of the poorest better than any socialist out there (including Bernie Sanders). Trump's messages are almost the same as those of the Leave campaign, as is his likely voting population. Perhaps the Democrats will learn from the mistakes of the Remainers by not having a negative campaign and not trying to scare the voters as to what will happen if Trump wins. A positive message is necessary. 

Fifth, the EU devolution, the reaction of EU policymakers, and the potential separation of Scotland and Northern Ireland from Britain (this last one I failed to touch upon in my initial comment). Despite the initial reaction of Europe's far-right and far-left parties demanding EU membership referendums in other countries, I am still confident that the EU will survive this shock. As it survived the sovereign debt crisis, the boiling point of November 2011 (which was several times worse in its destabilizing effect than the current referendum result), not to mention several highly likely Grexit possibilities. Europe's strength has been tested on a numerous occasions in the past 6-7 years, and even though this might seem as the decisive blow, I highly doubt it. Particularly given the reactions not only of Europe's leaders, but many others across the EU (this could however be selection bias, as only the most vocal express their concerns about the EU). 

This is not to forget that Europe needs a deep, deep reform of its institutions. I've personally been emphasizing this since I started the blog. The EU has completely alienated itself from the people. Even more so than any political leadership in any of its members. No one sees the EU as the convergence machine for prosperity anymore, while the benefits of the free trade area and the guarantee of peace on the continent are mostly being taken for granted. Instead the EU is, quite rightly, being portrayed as the bureaucratic leviathan with endless regulatory and legal requirements that stifle business and innovation, and that are to the common EU citizen nothing but an unnecessary burden. The EU's bureaucratic regime is turning into the worst manifestation of Kafka's and Orwell's novels. It is no wonder the people have an urge to fight against it. 

I sincerely hope the EU political elites get the message from Brexit. That could be the biggest positive that comes out of it - that the EU reforms and starts re-emphasizing innovation and trade, and starts to focus on lowering within-country (as well as between-country) inequality and increasing the living standards of its people. But that's certainly not all. EU law needs to be altered as the biggest objection against the EU is that it is being run by unelected technocrats. This is true, and it need to change. EU Parliament elections are not enough since Parliament has very little political clout in the EU. On the other hand the EU budget itself has a purely developmental goal (agricultural subsidies, EU funds, education and skills, etc.) and is not really a mechanism for economic policy. So despite being overreaching and intrusive in its regulatory patterns, the EU at the same time signals vast incompetence and inability to deal with the people's problems. Just recall its responsiveness to the sovereign debt crisis, or its reactions regarding the Ukraine crisis. 

In order to do all this, perhaps more federalism is needed. Now that the UK, the biggest opponent to "an ever closer Union", is out of the picture, this may very well be achieved. However, the danger is again the same - will this new federalism enable the unelected bureaucrats with more power which would imply more of the same or will it force them into promoting the true goals of Europe, as envisioned back in the 1950s? Given that the EU project is still evolving, meaning that it is still in the phase of trial and error, we can think of most efforts in the past 6-7 years as being an example of error. But that doesn't mean we should give up on it quite yet. 

Finally, what is to be left of the UK after this whole thing boils down? Possible just England and Wales. The biggest problem in the whole post-referendum debris is the vast political uncertainty. Literary no one knows what comes next: will Article 50 be invoked?; when and how?; what will the negotiations bring?; is the referendum outcome fully binding and will the "Bregret" crowd succeed in overturning it?; will Scotland block the Brexit vote?; will they have their own referendum which will bring them back into the EU?; will Northern Ireland do the same? - these are all questions no one has an answer to yet. One thing is certain though - in the short run, Britain will suffer. Political instability always gives rise to economic instability and possibly a recession, depending on how long it takes to resolve the situation. And from the signals we're currently getting, it will take quite some time to resolve it. 

Friday, 24 June 2016

Brexit: The analysis of results and predictions

On yesterday’s historic referendum, Britain has voted Leave. It was decided by a small margin, 51.9% to 48.1% in favour of Leave, with turnout at a high 72.2% (highest since the 1990s). The outcome dealt a decisive blow against PM David Cameron who announced his resignation in the morning. The markets have had a strong negative reaction, with the markets plummeting, and the pound sharply declining to its 30-year low against the dollar. It was an outcome the markets failed to anticipate (or were hoping to avoid), which explains the investors’ abrupt reactions.

Read the initial reactions: The Economist is in state of disbelief, trying to find a solution, and describing what happens next (invoking Article 50 of the Lisbon Treaty). They also have this interesting piece on the fallen legacy of David Cameron. The FT dreads about "Britain's leap into the dark", and keeps warning on the negative economic consequences. Martin Wolf also had a good comment. The BBC brings reactions from abroad, discusses the possibility of another Scottish referendum, and sums up eight reasons as to why the Leave campaign won. Other reactions are in the same direction: "a split nation", "what will the uncertain future bring", "what have we done?", and of course - the celebrations of the Brexiters. 

How did we do with our predictions?

Even though our prediction of the most likely outcome was a narrow victory for Remain (50.5 to 49.5), our model correctly anticipated that Leave has almost the same probability of winning. We gave the Leave option a 47.7% chance, admittedly more than any other model, expressing clearly that our prediction was nothing short of a coin toss.

As could be seen from our probability distribution graph below, the highest probability for the exact result of 49.5% for Leave (the one we decided to go with) was 7.53%, while the probability for the actual outcome of 51.9% for Leave was a close 6.91%, according to the model. This is a painfully small difference that comes down to pure luck in the end. Or as we said – a coin toss.

Source: Oraclum Intelligence Systems Ltd.
Turns out – the coin fell on the other side. Nevertheless, we stayed within our margin of error and can honestly say that we came really close (off by 2.4%; see the graph below). We knew that the last few days have been hectic and that the Remain campaign was catching up (high turnout suggests so), but it was obviously not enough to overturn the result. Leave started to lead two weeks before the referendum, and just as our model was showing an increasing chance of Leave over the weekend, a new flock of polls switched some voters’ opinions towards a likely Remain victory by Wednesday. In addition to the trend switch in our model we also failed to receive a larger sample, which proved to be decisive in the end.

Our results in greater detail are available in the graph below. It represents the comparison of our predictions to the actual results for the UK as a whole, and for each region (in other words, the calibration of the model). It shows that most of our predictions fall within the 3% confidence interval, and almost all of them (except Northern Ireland) fall within the 5% confidence interval. The conclusion is that we have a well calibrated model.

Model calibration (click to enlarge) Source: Oraclum Intelligence Systems Ltd.
This is even more impressive given our very small overall sample size (N=350). However even with such a small sample we were able to come really close to the actual prediction, beating a significant amount of other prediction models. Obviously the small sample size induced larger errors when it came down to certain regions (e.g. Northern Ireland or Yorkshire and Humberside), but it was remarkable how well the model performed even with so few survey respondents. Even if it did eventually predict the wrong outcome.

This was a model in its experimental phase (it still is), and the entire process is a learning curve for us. We will adapt and adjust, attempting to make our prediction method arguably the best one out there. It certainly has the potential to do that.

How did the benchmarks do?

It appears that the simplest model turned out to be the best one. The Adjusted polling average (APA), taking only the value of the polls two weeks prior to the referendum gave Leave 51% and Remain close 48.9%. This doesn’t mean individual pollsters did good, but that pollsters as a group did good (remember, polls are not predictions, they are merely representations of preferences at a given point in time). The problem with the individual pollsters was still a lot of uncertainty, such as double digits for undecided voters, even the day before the referendum. This is hardly their fault of course, but it tells us that looking at pollsters as a group is somewhat better than looking at a single individual pollster, no matter when they publish their results.

However, the Poll of polls (taking only the six last ones) was off, as it was 52:48 in favour of Remain (they’ve updated that yesterday just after I published the post, so I didn’t have time to change it). And the expert forecasting models from Number Cruncher Politics and Elections Etc both failed by 4% and 5% respectively.

Most surprisingly, the prediction markets and the betting markets have all failed significantly! As have the superforecasters. It turns out that putting your money where your mouth is still is not enough for good predictions. At least not when it comes to Britain. Prediction markets in some cases were giving an over 80% chance to Remain at the day of the referendum. In this case ours was the only model predicting a much more uncertain outcome.

Thursday, 23 June 2016

Brexit: the final prediction!

Our final prediction is a close victory for Remain. According to our BAFS method Remain is expected to receive a vote share of  50.5%, giving it a 52.3% chance of winning.

Click to enlarge. Source: Oraclum Intelligence Systems Ltd. 

Our prediction produces a probability distribution shown on the graph above (see explanation to the right), presenting a range of likely scenarios for the given vote shares. Over the past week we have consistently been providing estimates of the final vote share and the likelihood of each outcome. Daily changes and close results simply reflect the high levels of uncertainty and ambiguity following the EU referendum. However our prediction survey (the BAFS) has noticed a slight change of trend in favour of Remain in the past two days.

This is why our final prediction gives a slight edge towards Remain, and predicts a vote share of 50.5% for Remain, and 49.5% for Leave (the graph below represents the vote share of Leave, denoted as ‘votes for Brexit’ – a higher expected vote share for Brexit decreases the probability of Remain as the final outcome). The probabilities for both outcomes are also quite close, standing at 52.3% for Remain, and 47.7% for Leave. This means that 52% of the time when polling is so close and when the people themselves expect and predict a very close result, the Remain outcome would win. 48% of the time it wouldn’t.
Click to enlarge. Source: Oraclum Intelligence Systems Ltd. 
Vote share for Leave (votes for Brexit). The grey area describes the average error. As the sample size grew, the average error decreased.
Click to enlarge. Source: Oraclum Intelligence Systems Ltd. 
A timeline of probabilities for both outcomes since the start of our survey.

Why such low probabilities?

Due to a relatively high margin of error (± 5.3%). However given that this is not a standard survey with a representative sample, the error term does not mean much in this case (there is a whole debate about the controversy behind the margin of error – read it here).

Nevertheless, why is the error so high? Primarily because of very high levels of uncertainty among the actual polls, as well as among the predictions our respondents gave us. Also our sample size was relatively low (more on that below). If the error was around 1%, then the probabilities would have been much higher in favour of Remain (above 70%). This is closer to what the prediction markets and the superforecasters are saying.

But this means the prediction is as good as a coin toss?

Indeed. As it stands, the race is nothing short of a coin toss.

The problem in predicting such close outcomes is the measure of relative success of the prediction method. Usually being correct within a 3% margin is considered to be quite precise. In this case nothing short of a 1% margin will be permissible, which is essentially ridiculous and extremely difficult to guessestimate.

Having said that, we do hope our prediction method will be correct within its margin of error, but more importantly that it has correctly predicted the final outcome.

How does the method work?
The BAFS method (Bayesian Adjusted Facebook Survey) is a prediction method based on its own unique poll where we ask the people not only to express their preferences, but also who they think will win and how they feel about who other people think will win. This makes it different than regular polls which are simply an expression of voter preferences at a given point in time.

The obvious difference between standard polling and our method was noticeable during our initial predictions where we had a very small sample (around a 100 respondents) which was obviously biased towards one option (it gave Remain a 66% vote share), but we were still able to produce very reliable and realistic forecasts (see the graph below, the first results pointed to a slight victory for Remain, even with very high margins of error – initially over 10%). The later variations in our predictions were small even as the sample size increased threefold.

We follow here the logic of Murr’s (2011, 2015, 2016) citizen forecaster models where even a small sample within each constituency (21 average respondents per constituency for group forecasts) is enough to provide viable estimates of the final outcome across the constituencies.

The BAFS method, similar to the citizen forecaster model, is therefore relatively robust to sample size, as well as the self-selection problem (all of our respondents voluntarily participated in the survey). Both of these issues undermine the quality of standard polling, but in this case it was shown to have little or no effect. The BAFS method, utilizing the wisdom of crowds approach (group level forecasting), benefited from a diverse, decentralized, and independent group of respondents (seeSurowiecki, 2004) which gave us very realistic estimates of the final outcome. This implies that our prediction is likely to be quite close to the actual outcome on 23rd June.

How do we compare to other methods?

As we announced last month, in addition to our central prediction method we will use a series of benchmarks for comparison with our BAFS method. In the following tables we have summarized the relevant methods. For more about each method please read here. (Note: We have decided to introduce two new methods, fromNumber Cruncher Politics and from Elections Etc., both of which have proven track records in previous elections).

* For the adjusted polling average, the regular polling average, and for the forecasting polls we have factored in the undecided voters as well.

As it stands, we tend to be quite close to the predictions for the vote share (polls are slightly in favour of Leave, while other prediction methods are slightly in favour of Remain), but we tend to be a bit far from the probability estimates (the reasons for which are described above – if our error was lower, our probabilities would have also been around 70:30 in favour of Remain).

Mapping the results

Finally, here is how the map of the UK is supposed to look like if our predictions are correct:

Click to enlarge. Source: Oraclum Intelligence Systems Ltd. 
And here is the table by regions:

Source: Oraclum Intelligence Systems Ltd. 
Copyright for all visuals: Oraclum Intelligence Systems Ltd.