Chris Dillow's Blog, page 165

Nadine Dorries: some cognitive biases

Why has Nadine Dorries' decided to appear on I'm a Celebrity? She claims it's to get her message across to millions of people MPs can't reach. Others mention vanity and £40,000. In the long tradition of seeing what we want to see in any event, I suspect some cognitive biases are at work:

1. Neglect of base rates. Nadine isn't the first MP to apear on a reality show. Lembit Opik was on I'm a Celebrity two years ago.His career has not conspicuously thrived since. And George Galloway's appearance on CBB achieved the astonishing feat of making him look an even bigger arse than we thought. If Ms Dorries were a good Bayesian - and something makes me think she might not be - she'd attach a low prior probability to her chances of success. And two things mean she she's probably overweighting the posterior probability too...

2. Projection bias (pdf). People are bad at foreseeing their future tastes, exagerrating the degree to which they'll resemble their present tastes. I fear hope that Ms Dorries is doing this. She might now believe she can cope with having buckets of cockroaches tipped over her or having to eat kangaroo's bollocks. But when she actually has to do so, she might think otherwise.

3. Optimism bias. Even in the improbable event that Ms Dorries does a fantastic job in the jungle, does she really think this will come across? She seems to be forgetting that TV shows are edited. If she spends 95% of her time speaking eloquently on abortion and being a great team player and the other 5% whining and looking dog-rough, which 5% will be broadcast?

I suspect, therefore, that what we're seeing here is further corroboration that MPs are surprisingly bad at making judgments about probability.

Childhood's legacy

An unhappy childhood has long-lasting adverse effects. A new paper by Nattavudh Powdthavee shows that people who reported a high fear of being bullied between the ages of 11 and 15 not only have lower happiness, lower incomes and worse health in adulthood, but also suffer more if they become unemployed. They are less resilient to adverse events.

Now, there could be an endogeneity issue here. If kids pick on kids they sense are vulnerable, then bullying in childhood might be correlated with a lack of resilience in adulthood, without bullying playing a causal role.

But there's another possibility. It could be that bullying weakens individuals' hedonic capital (pdf) - for example by lowering their self-esteem and ability to make friends or trust people - and this increases their psychological vulnerability in later life. 

Insofar as this is the case, it adds to a large body of evidence which shows that circumstances in our childhood - or earlier - help determine our success in adulthood. For example, we know that a healthy birthweight, good health (pdf) in childhood, being taught in small classes and having a father who stays in work (pdf) are all associated with better outcomes in later life. And as James Heckman has shown, non-cognitive skills acquired (or not) very early in life matter (pdf) enormously for adult success.

One inference I draw from this is that mere formal equality of opportunity is a harmful myth. By the time an individual is mature enough to stand a chance of being able to take decisions herself, her life chances have already been heavily shaped by circumstances beyond her control. None of us is a "self-made man".

For me, this is one argument for strongly redistributive policies. They help to compensate people for misfortunes beyond their control - such as a disadvantageous upbringing - which blight them in later life.

The Chigley politics of the living wage

Readers of a certain age will remember the children's TV series, Chigley. At the end of each episode, workers would leave the factory to dance in Lord Belborough's garden. All conflicts were thus ended, and there was happy harmony between aristocrats, factory owners and workers.

Reading the two Milibands on the living wage reminded me of this. Here's Ed:

Businesses which have introduced it tell us that it saves them money.
In reduced turnover of staff.
And lower sickness absence...
So it makes business sense.

And here's David with Dave Prentis:

Paying the living wage cuts absenteeism and labour turnover and raises productivity.

This is not so much "pony politics" as Chigley politics - something that, like Lord Belborough's organ, makes every happy.

This the fallacy of composition. It's perfectly likely that if one or two firms pay higher wages than their rivals they'll attract more motivated staff and enjoy lower staff turnover than their rivals. But these advantages would disappear if all firms pay reasonable wages.And it's possible that the higher labour costs would lead - maybe marginally - to fewer jobs and shorter hours. Efficiency wage theory, remember, is a theory of unemployment.

You might reply here that higher wages would boost demand as workers spent their higher wages. Not necessarily, For one thing - as Ed nearly said - workers would lose around 50p in lower tax credits and suchlike for each £1 of pay rise, so to some extent paying a living wage is partly a form of fiscal tightening. And for another, it's possible that firms won't react passively to higher costs by running down cashpiles, but might instead cut spending on jobs or investment.

I fear, then, that the Milibands are doing what Kristian calls motivated reasoning - they want to believe that something that is morally desirable is also economically efficient.

Such a motivation, however, doesn't arise from the particularities of the living wage. Instead, it's a product of social democratic ideology. The defining feature of social democracy is that there are some positive-sum policies that benefit both capitalists and workers.

Sometimes, this is true. For example, in the 50s and 60s the full employment policies that benefited workers also delivered high profit rates.

But just because something is true sometimes doesn't mean it is always true. Here's an alternative possibility:

Factors such as globalization and (power-biased) technical change have massively depressed demand in the west for low-skilled workers, leading to unemployment and low wages.Policies to rectify this impose costs upon others. These might take the form of higher taxes if the low-skilled get in-work benefits. Or they might take the form of lower profit margins, if a living wage is imposed upon firms. And capitalists' investment function might well be such that lower profit margins lead to lower investment. In this sense, policies to help the low-skilled might well be zero-sum ones, or perhaps even negative-sum ones.

Such a view flatly rejects the Miliband's Chigley politics. But I suspect there's some truth in it. And if there is, then we cannot escape the reality that there is a class conflict.

The rationality of voting

John Gibson points to some neat research showing that voting is sensitive to the opportunity cost of doing so and infers that not voting might well be rational.I don't think it's as clear as this.

John says: 

If voting is rational it comes down to a comparison of the costs with the consumption benefits.

This, however, is true only if we identify rationality with instrumental rationality. But there are, as Robert Nozick pointed out in his best book, two other conceptions of rationality which better justify voting.

One, which arises from Newcomb's problem, is evidential expected utility. You might believe that your decision to vote will affect the outcome not because it will be the decisive vote, but because your choosing to vote will be evidence that people like you will also choose to vote.Your voting then brings thousands of votes out.

Voting is like a Schelling focal point; it's an action people undertake in the belief that others will do so. In the days before mobile phones, people in Leicester would often meet under the clock tower; the clock tower was a focal point. And often, the real world solution to the classic prisoners' dilemma is to keep quiet; not grassing is a focal point. (Social norms are focal points.)

The second conception is symbolic utility. Some things we do not because they make sense in selfish cost-benefit terms, but because they symbolize - maybe just to ourselves but maybe to others - the type of person we are. For many people, voting is rational not because it changes the outcome of the election, but because it's a way of reminding ourselves that we are socially-engaged citizens.

Why do people go on protest marches? Why do so many countries ban drugs? Why don't we euthanize unhealthy old people? The answer lies not in cost-benefit thinking, but in symbolic utility.

I don't say this to say that voting is or is not rational. I do so to point out that rationality has more conceptions than Gradgrindian economists sometimes think.

Immigration: let's not be reasonable

Sarah Mulley has written an intelligent and reasonable piece on immigration. And this is my problem; this is an issue on which intelligent people should not be reasonable.

She says that "immigration is important for growth" and that the coalition's cap on immigration "appears to make no economic sense". She's right. Jonathan Wadsworth's brief survey (pdf) says there's "little evidence of overall adverse effects of immigration on wages and employment for people born in the UK." A new study by Francine Blau and Lawrence Kahn says "most research does not find quantitatively important effects of immigration on native wage levels or the wage distribution." And Danny Blanchflower has said that A8 migrants might have helped reduce the Nairu.

So, if immigration has roughly zero effects on average on natives, and is positive for the immigrants themselves - immigrants are humans too, remember - why not just argue for freer migration?

Sarah gives three reasons, none of which convince me.

First, she says, the average zero effect "does not rule out more significant impacts on specific groups of workers (for example in some sectors in particular local areas)." Again, she's right, to some extent. But so what? Pretty much every decent policy has some losers: free speech is bad for lawyers and bigots; counter-cyclical monetary policy is bad for savers, and so on. We don't - or shouldn't - let minorities of losers block other good policies. Why should migration policy be different?

It's certainly not because politicians care especially about the welfare of low-skilled workers. For years, their prospects have been blighted by numerous developments, ranging from poor education through to the rise of China and India and power-biased technical change. There's something nastily hypocritical about a political class which has been indifferent (at best) to the well-being of the unskilled suddenly caring when it comes to immigration.

Insofar as some natives lose from immigration - and they do - the solution is higher in-work and unemployment benefits - policies which I think desirable on other grounds. It is not to stop immigration.

Secondly, Sarah says "migration (including skilled migration) has been part of an economic model that has seen wages at the top end of the labour market become disconnected from those at the bottom." Again, this is true. But it's irrelevant; it's the association fallacy. The fact that a good policy has been associated with bad ones is a reason for ditching the bad and keeping the good - not for doing the opposite, which seems present policy.

Thirdly, Sarah says:

Migration also poses a range of complex policy challenges beyond labour markets and the economy, particularly at the local level – the rapid population change that can result does affect housing, public services, and community cohesion.

However, we have reasonable evidence that migrants aren't a burden on public services.For example, Jonathan Wadsworth has found they are no more likely to use health services than natives, and Christian Dustmann and colleagues have concluded (pdf):

A8 immigrants’ receipt of government expenditures, in terms of benefits and other transfers, is substantially lower than their share of population, so, on balance, A8 immigrants have made a significant net contribution to the UK fiscal system.

There's a reason why I say all this. The "debate" on immigration is currently skewed against the evidence, and against freedom. Personally, I think the role of decent informed people such as Sarah should be not to acquiesce in this bias, but to fight it. We should try to shift the Overton window towards good policy. And this means being less tolerant of those who'd like to sympathize with "concerns" about immigration.  

Friedman's failure

Douglas Carswell has a nice quote from Milton Friedman:

The important thing is to establish a climate of opinion which will make .. the wrong people do the right thing.

By this standard, however, Friedman and his followers have failed in some respects. My chart shows one.It shows that there has recently been a close correlation between inflation expectations and the S&P 500, with higher inflation expectations associated with higher share prices. This suggests that the stock market disagrees with Friedman. He thought the optimal inflation rate was slightly negative. Stock markets clearly think it's positive. They - rightly or wrongly - don't want sound money.

This is not the only way in which business is anti-Friedmanite. Tim points out, as I have, that businessmen believe in central planning; it is, after all, what they do. And you'll not find many enthusiasts for a minimal state on the boards of Capita or BAe Systems.

I don't say this merely to remind you that there's a big difference between pro-market policies and pro-business ones. I've often complained that there's no constituency to press for sensible leftist policies. But there's no constituency to press for Friedmanite ones either.

Happiness, & statistical fetishism

Are measures of happiness prone to Goodhart's law - that they become useless if used for policy purposes? A new paper by Bruno Frey claims so. He contends that if happiness indices become a benchmark for judging policy success, they will become distorted.

One reason for this is that respondents will mis-report their well-being. Supporters of the government will overstate their happiness and opponents understate, and there's no reason to suppose the two will cancel out.

What's more, he says, governments can manipulate the results of happiness surveys more than they can orthodox economic indicators. For example, you can increase reported happiness by surveying people on Friday (pdf) rather than Monday, and by not preceding questions on happiness with questions on politics. And this is not to mention more other less professional methods: "here's a picture of a kitten: how happy are you?"

I would add another danger - that if governments focus more on happiness, they'll have even stronger incentives to promulgate ideologies that diminish people's expectations. If you can convince people that poverty is unavoidable, you might reduce discontent even if injustice increases.

However, I don't think this is an argument against happiness policy.

We know there are some things governments can do to improve well-being such as: reduce unemployment; give people more control over their lives; and - credit Ed Miliband here - do more to reduce mental illness. We don't need a happiness index to do any of this.I suspect the desire to use such an index is a form of statistical fetishism - a wish for the mere appearance of useful knowledge. But as Keynes said, it is sometimes better to be roughly right than precisely wrong.

Cuts vs efficiency

Reading Paul's call for increased spending on childcare might suggest that he's conflating two separate issues. He says:

So how will the extra money be found? I’m a sensible post-Keynesian who enjoys the benefit of Modern Monetary Theory insight, and so I understand better than IPPR appear to that the best way to invest is through deficit spending. That would, well, just work.

Now, there is a case for deficit spending. And there is a case for increased spending on good childcare. But they are two different things.

The case for spending on childcare is that, as James Heckman has shown (pdf), it is a great investment. Better childcare leads to more educational achievement and better health and thus higher economic activity and lower public spending in the future. Even if the Austerians and "in the black Labour" group were wholly correct and we do have to curb public spending overall, there'd still be a case for higher spending on childcare. And if there isn't, you should be arguing against Heckman, not Keynes.

So, is Paul just mistaken to conjoin the two logically separate issues? Maybe not. The problem here is one of public choice.

Ed Balls has claimed that a Labour government's "zero-based spending review" will "[assess] every pound of taxpayer’s money including for its impact on growth and fairness."

Given its high social return, childcare spending should emerge from such an assessment not only intact but with a higher budget.

If, that is, the assessment were based only upon disinterested and far-sighted rational thinking.

But it might not be. As Roger says, zero-based budgeting usually becomes a way for consultants and managers to press their own interests. There's a danger that, by the time a government has pandered to powerful bureaucrats, corporate lobbyists, the Daily Mail and unions, zero-budgeting will end up cutting the sort of pre-school childcare that has good returns because it doesn't have  powerful interests to speak up for it. Who's got the most power over politicians: unions, big business, military chiefs, the media - or nursery teachers?

From this perspective, I fear that Paul might be right to link deficit spending with childcare - because, given political pressures, only deficit spending can protect sensible long-term investments.

Mr Balls might want to make intelligent cuts in public spending. But what if his desire exceeds his ability? There then emerges a trade-off between smaller government and efficient public spending.

The low pay problem

KPMG says that one in five workers is paid less than a "living wage", of £7.20 an hour outside London and £8.30 in it. This is consistent with official figures which last year found that 20% of employees earn less than £7.31 per hour.

This reflects a fundamental problem with the global economy - that the rise of China and India has contributed (pdf) to a collapse in demand for low-skilled workers in the west.

What can be done about this?

Paul Sellars points to large firms hoarding cash and claims that transferring some of this to pay packets would "revive consumer spending and business confidence, thus helping to kick-start the recovery." I'm not convinced. I suspect many of the firms paying low wages are small firms struggling to get by, not the mega-corporations stock-piling cash.

Paul also claims that:

Many of these companies could actually gain from adopting the living wage. First, their reputation for corporate social responsibility would be enhanced – being “nice” can be a valuable selling point. Second, they would derive solid personnel benefits in terms of recruiting, retaining and motivating high-quality staff.

This, I fear, is the fallacy of composition. It's true that any individual firm paying good wages would get reputational gains, the pick of staff and lower turnover costs.But this wouldn't be true if all firms did so.

Nor am I convinced by the obvious statist policies.

There are two problems with raising the National Minimum Wage:

- It wouldn't hugely benefit the low-paid to the extent that a higher wage causes the withdrawal of tax credits and higher income tax payments. Such withdrawal rates mean that many low-paid might see only 30p of a £1 rise (pdf) in wage income.

- It would (perhaps?) reduce labour demand - in terms of jobs or hours. If we assume the average low-paid worker outside London gets £6.70ph, raising the NMW from £6.19 to £7.20 implies a 7.5% rise in the wage bill. Assuming a price elasticity of demand of 0.5, this implies a 3.75% reduction in demand for labour - equivalent to 160,000 job losses.

However, the alternative - more generous tax credits - also have problems, even assuming (generously) that there is popular demand for such redistribution:

- Employers might respond to higher tax credits by cutting wages. One reason why a minimum wage was introduced was precisely to stop this happening when tax credits were introduced.

- More generous support for the low-paid must be withdrawn as wages rise. But this means even higher marginal withdrawal rates and thus reduced incentives to work longer hours or get better jobs. This is an unavoidable trade-off with any system of in-work benefits.

So, what are the alternatives? I'd like to consider ways of increasing the bargaining power of the low-paid, perhaps through some combination of stronger unions, the state acting as employer of last resort and a basic income.

But let's be clear. There's a bigger problem here than all our political parties are willing to acknowledge: predistribution, in the form its currently envisaged, is not sufficient.

Savile: remember Bayes theorem?

The Monday morning quarterbacks are busy today. The Observer accuses the BBC of "institutional incuriosity" about Jimmy Savile and Chris Patten asks:

Can it really be the case that no one knew what he was doing?
Did some turn a blind eye to criminality? Did some  prefer not to follow up their suspicions because of this criminal’s popularity and place in the schedules? Were reports of criminality put aside or buried?

This misses something - Bayes theorem. This allows us to quantify the probability that someone will be a pervert, in light of an allegation against them. To do so, we need three numbers:

- What is the chance of someone being a pervert, in the absence of evidence against them? This is our prior probability.

- What is the chance of rumours against them, if they were a pervert? This is the chance of a true positive.

- What is the chance of such rumours if they are innocent? This is the false positive rate.

The maths is well explained here. Let's stick some numbers onto this. Let's say we think there's a 10% chance of someone being a pervert, in the absence of evidence. Let's say that, if he were a pervert there's a 100% chance of talk about him; no fire without smoke. And let's say that if he were innocent there's a 20% chance of there being rumours about him. Bayes theorem then tells us that there is a 35.7% chance of him being guilty.

My numbers are arbitary. But they show the point - that even if we give significant credence to the allegations - only a 20% chance of them being false - we should still think there's only a small chance of the accused being a pervert. This is because if there is a good chance someone is innocent, the chance of him being falsely accused is greater than the chance of him being guilty.

Now, I'm not saying BBC bosses consiously applied Bayes theorem; I suspect they were as ignorant of it as their accusers are today. What is possible, though, is that they applied it intuitively, believing that the prior probability of Savile being a pervert was so low that it survived contact with the doubtful allegations against him.

A belief in Savile's probable innocence need not therefore have arisen from turning a blind eye or from incuriosity. It might instead have been the result of maths and logic.

What is possible, though, is that the inputs into this logical process arose from irrational processes. BBC bosses might have attached an unduly low prior probability to Savile being a pervert because of wishful thinking or reframing: "he's not a pervert, just eccentric." And they might have exaggerated the false positive rate, thinking they came from silly girls.

With hindsight, this seems likely. But is it really reasonable to condemn people with hindsight? In the 70s, there was less alertness (panic?) about paedophiles, which generated a lower prior probability then than we'd have now. And would you really have given great credence to the word of the sort of people who screamed and fainted at the sight of the Bay City Rollers?

Yes, it now seems as if BBC bosses were wrong. But Bayes theorem tells us that, given the paucity of hard evidence against Savile in the 70s and 80s, they might not have been as culpably irrational as they seem.

But then, I've missed the point, haven't I?