In many disputes about social or political issues – whether globalisation is a good thing,
for example – the accumulation of evidence does not seem to be sufficient to bring
people of divergent views to a consensus. New research by Professor Wing Suen,
published in the April Economic Journal, explains why.
He argues that most people do not have the time or expertise to look at the evidence on
big issues for themselves – so they rely on others for information to guide their views.
The problem is that they generally seek advice from like-minded individuals, which means
that the advice they get will typically confirm their existing beliefs. This means that initial
biases becomes even more biased – bias tends to beget bias – and pre-existing
divergences of opinion become even more polarised.
Professor Suen cites the example of international trade. Some economists bemoan the
fact that the rest of the world fails to see it through the same lens as they do. The theory
of comparative advantage has been around for more than a century. If truth and reason
ultimately prevail, they believe, there should have accumulated enough evidence to
convert everybody into a free trader.
Alas, the rest of the world still talks about trade issues in terms of trade deficits, job losses
or cheap labour – and wonders why economists never abandon their arcane theories.
Most people do not have the time or expertise to look at the evidence on such issues for
themselves. They rely on third parties – journalists, friends, consultants and specialists –
to provide information to guide their decisions.
But the inherent limits of the communication process means that information cannot be
conveyed in its entirety. Instead of portraying the world in different shades of grey, third
parties often have to provide information in the form of binary ‘yes’ or ‘no’
recommendations.
Such filtering of information serves a useful function as it simplifies the task of decisionmaking.
But it also opens the question about selecting an appropriate standard of
evidence.
It is commonly observed that people tend to seek advice from like-minded individuals.
Psychologists use the concept of ‘cognitive dissonance’ to explain such behaviour. They
argue that people feel discomfort about confronting evidence that are contrary to their
pre-existing beliefs.
Hence, Conservative-leaning voters choose to read right-wing newspapers to avoid the
discomfort. Professor Suen's analysis, on the other hand, suggests that such a pattern of
choice could be due to rational calculations rather than self-delusion.
To overcome one's prior beliefs, strong evidence to the contrary is needed. Since a leftwing
newspaper will endorse the Labour party based on weak evidence that Labour is
superior, a recommendation by this newspaper to vote for Labour is not sufficient to move
a Conservative voter's beliefs sufficiently to alter his vote.
If the information provided by the left-wing newspaper has no effect on the Conservative
voter's decision, he or she is better off reading a right-wing newspaper.
When people seek advice from like-minded individuals, the advice they receive is likely to
confirm their existing beliefs most of the time. This means that a person who begins with
some initial bias has a large chance to become even more biased after learning the
evidence that has been filtered by third parties.
In other words, bias tend to beget bias and pre-existing divergence of opinion will become
more polarised given the selective choice of evidence.
Professor Suen's research shows that ‘rational’ learning is still possible – in the sense
that people will learn the truth in the long run – but the learning process can be made very
slow.
ENDS
Notes for Editors: ‘The Self-perpetuation of Biased Beliefs’ by Wing Suen is published in
the April 2004 issue of the Economic Journal.
Professor Suen is in the School of Economics and Finance at the University of Hong
Kong.
For Further Information: contact RES Media Consultant Romesh Vaitilingam on 0117-
983-9770 or 07768-661095 (email: romesh@compuserve.com); or Wing Suen on +852-
2859-1052 (email: hrneswc@hkusua.hku.hk).