More evidence for voter irrationality

Whenever some of us claim that voters are irrational, we’re met with the accusation that we’re elitists who haven’t come to terms with the possibility that crowds are wiser than we are. In this context, therefore, I welcome some new research by Erik Eyster and Ernesto and Pedro Dal Bo who provide experimental evidence (pdf) that voters are irrational.

They got subjects to play a prisoners’ dilemma game in which there were big pay-offs to defecting with the result that subjects did not cooperate even though the aggregate rewards to doing so were high. After a few rounds of that game, subjects were offered a vote on a tax which would have disincentivized defection and so encouraged cooperation and the achievement of those higher rewards. However, the majority of subjects voted against the proposal even though it was in their interests.

This suggests that people look too much at the obvious cost of a policy and under-estimate the extent to which it can have the benefits of changing behaviour. The authors say:

Voters will tend to focus on the direct effects of the policy change and underappreciate the indirect effects. As a result, voters will favor reforms with positive direct effects, even when undone by negative indirect effects, and reject reforms with negative direct effects, even when more than compensated by positive indirect effects.

This is consistent with work by David Leiser and Zeev Kril, who have shown that laymen are terrible at identifying causal connections in economics.

Does this have real-world relevance? What sort of policies fit this sort of pattern?

Plenty.

The authors give the example of pollution taxes: maybe voters overweight their direct cost and underweight their indirect effect in curbing pollution. They also suggest that price or rent controls fit this pattern. Voters see their direct benefit, but under-rate their adverse impact upon supply.

Some of you might add minimum wage laws to this category: voters might overweight their direct benefit of raising wages but underweight the indirect effect of curbing labour demand.

We might add that policies that disincentivize arms races also fall into this category. This might explain hostility to Robert Frank’s call for a progressive consumption tax to deter the excessive spending that results from expenditure cascades.

Perhaps immigration controls fit the pattern too. Voters see the direct benefit: fewer immigrants mean less competition for jobs and hence higher wages. But they don’t see that if their wages rise without any increase in productivity then inflation and interest rates will rise, which means that people will lose their jobs because of weaker demand.

The Brexit vote might also conform to this. Maybe voters over-rated the direct benefit of Brexit (“we’ll save £350m a week”) and under-rated the indirect cost, that Brexit would weaken the economy and hence the public finances.

Franklin Roosevelt once said that “democracy cannot succeed unless those who express their choice are prepared to choose wisely.” And it seems voters don’t choose wisely.

It would, however, be wrong to blame voters alone here. They aren’t expert economists, any more than I’m an expert plumber or electrician.

The fault also lies with politicians – with the “elite” if you like. They don’t seem to care much about the quality of voters’ decision-making, as they as they support the right side. And the media are also at fault, in taking little effort to improve voters’ wisdom. If Roosevelt was right, such carelessness is positively dangerous.  

Another thing: we must distinguish between being right and being rational. It's possible that Brexit or high minimum wages are the right policies even if voters choose them for irrational reasons. I find it hard to believe, however, that irrational voters will always or even often choose the right policies for the wrong reasons.

Comments

[a random eman]

It's interesting you have picked minimum wage laws as an example of irrationality. You are surely aware of the recent evidence from America that they do no harm to labour demand?

That said, your general point is spot on.

[Ben]

Might this not be a rational response to another, arguably higher order, cognitive frailty that people largely understand quite well - ie - the planning fallacy?

Voters quite rightly put high error bars around the 2nd order effects of policy changes - and understand that there are many unintended consequences which could affect them either way.

[Doug]

'They aren’t expert economists...'

You mean all the experts who saw 2008 coming or all the experts who who have spent their lives ideologically bolstering capitalism because it's a good career move.

[Blissex]

«fewer immigrants mean less competition for jobs and hence higher wages. But they don’t see that if their wages rise without any increase in productivity then inflation and interest rates will rise, which means that people will lose their jobs because of weaker demand.»

It is amusing to see in this time an age such a hearty endorsement of the Iron Law Of Wages, a corollary of which is that the "lump of wages" and the profit share of value added are fixed and higher wages necessarily reduce employment.

[Blissex]

«fewer immigrants mean less competition for jobs and hence higher wages. But they don’t see that if their wages rise without any increase in productivity then inflation and interest rates will rise, which means that people will lose their jobs because of weaker demand. [ ... ] And it seems voters don’t choose wisely.»

Also it is not just voters: the trade federations of food and entertainment businesses share the same lack of wisdom as voters because they are afraid that lower immigration from very poor countries would raise wages:

https://www.theguardian.com/business/2016/dec/08/uk-food-prices-to-rise-without-eu-workers-warn-trade-groups

While IIRC out blogger has proven abundantly that immigration from very poor countries immigration raises wages.

We are fortunate that employer lobby groups misunderstand so massively the effect of a large increase in the labour supply.

:-)

[Luis Enrique]

a random eman

you are overstating it a bit. this is an area where short and long-run responses may differ, the long run coming as (in this case) new restaurants are established with more automation, more capital intensive production, even if in short run already established restaurants do not reduce hiring: http://voxeu.org/article/long-run-employment-effects-minimum-wage

[Patrick Kirk]

Democracy is an alternative to settling disputes with civil war. Given that civil wars are terrible, democracy is always the rational choice. As for the decisions taken by voters, there is no reason to think that allowing decision be taken by the results of a civil war would be any better.

[B.L. Zebub]

@Chris,

From the study you quote, you conclude,

"This suggests that PEOPLE look too much at the obvious cost of a POLICY CHANGE and under-estimate the extent to which it can have the benefits of changing behaviour."

This could easily be redacted thus,

This suggests that EXPERTS look too much at the obvious cost of NREXIT and under-estimate the extent to which it can have the benefits of changing behaviour.

The experimental subjects (which you and the researchers identify as analogues or proxies for the voters) were primed by the first exercise. Economists are primed by their education.

Presented with the second exercise, experimental subjects did not consider the situation in its own merits, but extrapoloated from the first exercise.

Presented with BREXIT, experts did not consider the situation in its own merits, but extrapolated from their first exercise.

You know, the key is to ask oneself who the experimental subjects actually represent. You and the researchers take it for granted they represent the voters only.

But, correct me if I'm mistaken, you and the researchers are also human, yes?

[Keith]

It is off course the job of politicians to debate and explore the finer implications of policy ideas as part of the political process. What we are seeing in many countries seems to be a bad failure by those whose profession is politics to do the job that political theory assigns to them.

P Kirk has a very low bar as to how decisions should be taken. Bad political decisions off course can result in civil wars among other evils. Which is why we should try harder to make good decisions.

[aragon]

"we’re met with the accusation that we’re elitists"

See the people are right!

https://static1.squarespace.com/static/56afbfd58a65e261b705f1f9/t/56b507fbe3214049a5dafecb/1454704635507/eurojournal.pdf

"Development agencies often ignore the rights of the poor by supporting the development plans of
authoritarian governments. This is the main argument of The Tyranny of Experts, and few observers of development cooperation will disagree."

i Blank Slate vs Learning from History
ii Well being of nations vs individuals
iii Conscious design vs spontaneous outcomes.

What conclusions can be draw from an contrived experiment, that excludes the complex factors and experiences that:

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1880926

"First, the dominance of Traditional Risk Analysis, with its emphasis on statistics and cost-benefit analysis, has downplayed the role of values and subjectivity in risk management. The result has been that risk decisions have been based upon the erroneous assumption that empirical data and mathematical calculations alone are adequate bases for risk decisions. There has been virtually no acknowledgment that ex ante consideration of the convictions and passions of the public is valuable. In fact, a number of scholars have argued that because individuals become emotional about potential harms, scientific experts should make all risk decisions without any public involvement at all."

Government requires the consent of the governed, not the "Tyranny of Experts or even Light"

http://www.sciencedirect.com/science/article/pii/S0016328797000359

"More information may lead to less understanding; more information may undermine trust; and more information may make society less rationally governable."

[Bob]

Good to see Chris endorses macro policy to sustain pool of unemployed and micro policy to blame unemployed.

[jonny bakho]

MinWage is a bad example.
Most workers make wages higher than the minimum, so MinWage is NOT a direct benefit.

[Calgacus]

This is consistent with work by David Leiser and Zeev Kril, who have shown that laymen are terrible at identifying causal connections in economics.

They are a lot better than economists. Laymans heuristics makes some logical sense and have some empirical validity. The economists theories are rather worse on both counts.

Their complaint is the opposite of Keynes's - who said the problem was that people believed neoclassical economics (from Leiser & Kril the only kind there is) far too much & distrusted their own more accurate common sense.

But the last few decades have seen far more people than ever having their minds "fuddled with nonsense for years and years."

[rogerh]

Voters are irrational - well maybe. Firstly there seems the curious phenomena that most democratic societies seem to divide roughly 50/50 between rightish and leftish politics. This seems more a feature of the human mind rather than any obvious evidence that leftish is any better or worse than rightish. There seems no obvious way to decide and little unbiassed evidence.

So your average voter is left rather worse informed than someone buying a washing machine or telly. The techniques of advertising and publicity combine with the legendary untruthfulness of polticians and their henchpersons to appeal to tribalism and prejudice. In the end it probably comes down to brand names and whether the voter likes or loathes the salesperson and what body language the saleperson exudes.

The experts - economists and technocrats - may have some competence but those of them who get listened to by politics and the media seem a very mixed bag and equally tribal in their support for this or that policy. Worse, their recommended nostrums seem to change like the wind and seem to have little or no predictive value. So are voters irrational - yes, they cannot be otherwise in the circumstances.

[aragon]

Government requires the consent of the governed it's called self-deterination.
https://en.wikipedia.org/wiki/Self-determination

As self-determination is recognised as a fundamental human right, it might be a little difficult to side step the voters. The predictive value comes from the collective experience and judgement.

Simon Wren Lewis may have coined Media Macro, for the false view of the media for state of the art in Economics. But Macro economics suffers from Macro Myopia (visible to the non-short sighted).

For economists, if it is not in their model it doesn't exist or is not significant.

However facts are not enough, as they embed value judgments, and as William Eastman argues often target intermediate solutions.

For example the standard Portes defense of immigration as economically beneficial.

Social Cohesion is not a factor, or can be mitigated by a small economic gain to the nation.

Simon Wren Lewis
http://www.newstatesman.com/politics/economy/2016/12/how-media-misled-us-over-brexit-and-donald-trump

"So the idea that immigration adds to pressure of public services is a myth perpetuated by politicians trying to deflect attention from their own actions, and a myth that the media does little to counteract. It is another politicised truth."

We could go around the houses (yes, immigrants need houses, and want cars - public infrastructure like roads). And wages may be a poor measure of value.

But these are not in the model? Simion Wren Lewis states that Young Migrants don't use public services like the NHS. Consistent with the pattern is the fact that this is a intermediate argument. What else do Young Migrants do? They have children and children are exclusive uses of school education and Mothers and children of the NHS.

https://www.migrationwatchuk.org/briefing-paper/15.3

"So with no migration at all, the zero net migration projection from the ONS shows that our population would continue to grow for the next three decades and then decline gradually, returning to the current level towards the end of the century."

[...]

"On those assumptions, it is clear that all or almost all UK population growth in the longer term would arise from immigration, directly or indirectly."

Population growth and therefore increasing pressure on ther UK environment and public services is due almost exclusively to migration!

I find economists appeals to authority very unconvincing as for their predictive power, that looks like a random walk.

Black-Scholes-Merton anyone? Long Term Capital Management? Remove risk from both sides of the equation? (To mention one economic failure)

Perhaps a cobbler should stick to his last?