The last
few years has seen quite a few surprising results that, together, show a very
clear trend. I would like to call this trend the “Me and my people” trend. In
this blog I will argue that this is not a new phenomemon, but that given the
change in the environment, it is getting more pronounced. I try not to pass any
value judgement; my aim in this blog is to try and help some of us who are
unable to understand what the trickle that became a big wave, so if we choose
to, we can possibly divert the wave, diminish it before it becomes a tsunami
that wipes us all off. What does that have to do with data? Well I will use
data to support what I am claiming. The only bush I may be beating around in
this blog is called George W.
15 to 20
years ago, someone made a proposal that now may seem far-sighted, he may have
overestimated the pace of change in the world, but let’s not jump the gun.
1 Donald Trump is elected president of the USA
To many
people, the election of Donald Trump came as a surprise, a shock even. How
could a “grab them by the p*ssy”(1) become president?
It is very
simple; the white voters made him win. I was still surprised not much has been
made of this, until recently. First, let’s take a look at the data (2) (3).
With Hilary
Clinton being a woman and the strong support of famous women, including “there
is a special place in hell” Albright (4), and his own “p*ssy grabbing”
declarations (1), it could be expected that Donald Trump would not do so well
among women.
Around 55%
of voters were women, but while she won among women, this was not enough for
Hilary Clinton to win overall.
Race seems
a much clearer driver of voting patterns. White americans, representing 70% of
voters, strongly preferred Donald Trump, while other races were overwhelmingly
in favour of his opponent.
Combining
race and gender:
Donald
Trump won even among white women.
In fact:
Virtually
the only demographic that Hilary Clinton won among white voters was the
white-college-grad (or higher) women.
In
conclusion:
It can quite easily be said that Donald Trump was elected president of the USA based on the votes of white voters.
It can quite easily be said that Donald Trump was elected president of the USA based on the votes of white voters.
2 Brexit
Another big
change decided by the ballot box was Brexit: the UK voted to leave the EU.
Again however, the characteristics that propelled “Leave” to victory are quite
clear (5).
The relationship
between age and the Brexit vote is very clear, older people tended to want to
leave the EU while younger people preferred to remain.
This is
reflected across gender; while women are slightly more likely to prefer remain,
age is crucial.
Similarly,
while social class has some impact, age is crucial. The classes are arranged in 'descending' order from upper managerial and professional to supervisory, clerical and administrative roles, to skilled manual occupations, to semi-skilled and unskilled manual occupations and unemployed. (6)
In conclusion:
Older people preferred to leave the EU, while
younger people who are more likely to live with the consequences of the vote
preferred to remain.
The fact that people who more likely to live
with the consequence of the vote (the young people) see that people who will
live through them less (the old) basically won the vote for the “leave” camp,
has create some inter-generational conflicts (7).
So was Lee Kuan Yew
right?
Interestingly, Lee Kuan Yew, spoke about the
idea of going beyond 1 person, 1 vote, in order to allow people who had more at
stake to have more say than by pure 1 person 1 vote. In fact, he said that
people who are married with kids (ok, unmarried with kids wouldn’t have
benefited, in some areas he was behind time) should have two votes instead of 1
because they would also think of the impact on the next generation (8). “…during
those critical years, 35-60, people who carry twice as much responsibility
should have two votes. This will make for a more viable system and a more
stable society.”
It seems
that Lee Kuan Yew understood that society was changing rapidly, and that people
would focus more on themselves than on others.
This kind
of makes sense when people are living longer, hence the weight they put on
their own preference as opposed to what would be better for their children or
the next generation.
But are we
under-estimating how much people value their own immediate needs and
preferences over that of society/their children/the next generation?
This is not
a new question.
1975
A very
insightful article, written in 1975 by Robert Heilbroner (9), asks a few
questions (10), but the crux of it all is: why should anyone care in mankind
survives 1000 years? Everyone’s children and even grandchildren would have been
long dead by then. Basically 1000 years is too far out in time for someone to
sacrifice something today, however trivial.
1759
The article
also refers to another source, in 1759, from the economist Adam Smith, he
theory of mral sentiments (11). In this book Adam Smith created a simple
thought experiment. Imagine, he argued, (the reader is assumed to be European)
that am earthquake strikes China and millions of people die. The reader would
feel ‘melancholy’, if he/she was business minded, may be invest in goods there
would be a shortage of. But eventually, the reader would simply go to bed and
sleep well. On the other hand, if the reader was told he/she would lose a
little finger the next day, the night would likely be sleepless. Millions of
people you don’t know, lose to your little finger?
Adam Smith
then argues for the “man within the beast” who will make the ‘right decision’,
to sacrifice the finger for the millions of people, “It is the love of what is
honorable and noble, of the grandeur and dignity, and superiority of our own
characters”
Economists,
Heilbroner argued, had a neat way of modeling (in the sense of a thought model
or a mathematical representation of a situation), and I would agree.
Economists
tend to think of people as making decisions in their own interest. As I argued
with my mum many years ago, she behaves ‘nicely’ because she believes that this
behavior brings benefits to her (good karma, heaven, whatever you believe in),
hence in a way that ‘nice behaviour’ is also for self-interest. Not that I am
saying she should not be nice, I am saying that people being nice can be in
their own self-interest, not that there is anything wrong in that. When
people’s incentives are aligned with that of ‘society’ then society will do
quite well; similarly when interests of ‘employees’ are aligned with that of
the organization, then the organization will not suffer from ‘principal-agent
problem’ (12).
This brings
me to 2 things I wanted to blog about. (Yes that was a loooong intro, so will
try to keep the arguments short)
Bias and Migration
One of the
ideas that people have been putting forward for the ‘move towards the centre’
(13) in politics in the 90s was that that’s how elections could be won. The
idea was that most people were not extreme, in fact the idea was that people on
the left, or on the right tended to be not so extreme left or right.
Imagine a
simple election where, in a population, the left and the right are balanced.
Assume the
2 candidates are Mr X and Ms Y. People will vote for the candidate closer to
their views. In this case, from their starting positions, Mr X and Ms Y are
stuck in a finely balanced election each getting around 50% of votes, people to
the left of the red line vote for Mr X and those to the right vote for Mr Y.
But what if
Mr X decides to move closer to the centre?
If Mr X
changes her position and moves towards the centre, he will eat into the share
who favoured Ms Y, the new split will be at the red line, and Mr X now would
win the election, like Mr Blair.
Ok you may
say, that may explain the moves to the centre, but the 2 cases above, Mr Trump
and Brexit are certainly not moves to the centre, on the contrary.
And you’d
be right.
So what is
going on?
Blame it on the internet
The first assumption of the models above is
that the population can be represented by 2 bell shaped curves, one for the
left and one for the right; we can argue about the thickness of the tails, but
that is not the point.
Pre-internet, people with extreme views and who
lived in small towns basically were faced with their own small bell-shaped
curves and a choice: to remain extreme and be isolated, social pariahs, or to
move towards the centre of their own town’s bell shaped curve. Humans are
social creatures and there is a need to belong. Hence towns would become either
single bell-shaped or double bell shaped with short tails, extremes moving to
the mainstream.
In some cases, extreme cases move slightly more
mainstream, “thickening the tails”, or they could move even deeper, squeezing
the distributions.
Now, as these individual bells are collected to
make up the population, we are likely to end up with 2 nice large bell-shaped
curves one for people on the left, and one for people on the right.
In a recent post (14), Azman Osman-Rani argued
that to make people change their minds, you need to make them understand that
changing their minds will not leave them alone, but that they can join up with
a group, a social group, since humans are social animals.
While this served to make people move to the
centre pre-internet, now this allows people to stay on the fringes, and even
make people gravitate to the fringes.
Basically, instead of being forced to gravitate
to the centre to belong to their towns, the people at the fringes realise that
there is a whole community of people on the internet like them, and they form
their social groups in cyberspace.
In the
illustration above, the internet makes people who are at the extreme left aware
of other similar people geographically further but ideologically closer, and
these people create join and form a new community from the original bell curve.
The people to the extreme left join the new blue bell curve and some people
closer to the mainstream also join up since the new ideology is closer to their
starting point that the mainstream. This is shown by the dark gray area from
the original curve that moves to the new community (online).
In the illustration above, the internet makes
people who are at the extreme left aware of other similar people geographically
further but ideologically closer, and these people create join and form a new
community from the original bell curve. The people to the extreme left join the
new blue bell curve and some people closer to the mainstream also join up since
the new ideology is closer to their starting point that the mainstream. This is
shown by the dark gray area from the original curve that moves to the new
community (online).
So what does that have
to do with Mr Trump or Brexit? Or even Mr Lee Kuan Yew?
Simply, it’s ok to vote as a community based on
skin colour/race, because these people are now part of a community of similarly
thinking people and feel no need to move to the mainstream. Similarly, it’s ok
for the old people in the UK to focus on their own preferences, even if the
young have more to lose, and vote for Brexit, simply because the Brexit
community is large.
Being extreme is not that lonely anymore.
On the other hand, was Mr Lee Kuan Yew wrong?
His underlying assumption was that people with children would take the future
of the children (and thus the next generation) into account when making
decisions.
Mr Lee Kuan Yew counted on people with children
not focusing purely on their own objectives, but lengthening their time
horizons, maximizing over more than their own lifetimes, whereas the people who
voted for Brexit (sometimes contrary to what their children wanted), focused
more their own present (or their past, some sort of nostalgia for the past)
rather than on a future they would not be here to enjoy.
On the other hand, the recent events in Hong
Kong where a large portion of the population, young and old, male and female…
came together as one to protest a change in rules that would have affected the
future of Hong Kong (15), Even the rich (16) although the latter may have more
to fear of their past and the present.
To sum up
This blog is a little bit longer than usual, so
I’ll stop here, and may be continue with some implications in the field of
analytics in my next blog.
People are more likely
to express, hold-on to, and even grow into more extreme opinions, and this is something we should
expect, understand, and act accordingly.
Expression
One of the reasons the victory of Mr Trump was
surprising was that even until polling day, his opponent thought she would win
quite easily (17). People basically lied to pollsters, not wanting to appear
extreme/racist…
Holding on and even growing into extremism -
radicalisation
Given that most people visit only a few
websites regularly (18), those websites and social media that occupy people’s
attention are likely to have large impacts on people. Hence there is a real
threat of self- radicalization (19) anywhere in the world.
5 https://www.ipsos.com/ipsos-mori/en-uk/how-britain-voted-2016-eu-referendum
6 https://www.ukgeographics.co.uk/blog/social-grade-a-b-c1-c2-d-e
6 https://www.ukgeographics.co.uk/blog/social-grade-a-b-c1-c2-d-e
19 https://www.todayonline.com/singapore/youth-self-radicalisation-concern-shanmugam
