Stumbling and Mumbling

Geroski on recession

The death of Paul Geroski deprives us of one of the authors of one of the most interesting books on macroeconomics written in recent years. I refer to Coping with Recession, a study of the 1990-91 recession, which he co-authored with Paul Gregg.
They showed that:

Performance differences between firms increase markedly in recessions. While mean profitability is procyclical, the variance in profitability across firms at any time rises sharply in cyclical downturns.

They found that just 10% of companies accounted for 83% of the gross fall in profits in the 1989-91 recession, with 20% accounting for 92% of the gross fall. The worst-hit 10% of firms also accounted for 85% of the gross fall in employment. (Looks a bit like a Pareto distribution, doesn't it?)
40% of firms saw their profits rise in the recession, and half raised employment.
This has (at least) two interesting implications.
1. It could lend credence to Fischer Black's view that recessions are due to mismatches between tastes and technology, rather than to downturns in aggregate demand. Black's view suggests a minority of firms - those with the worst mismatches - would suffer in recession. The aggregate demand view, on the other hand, is less easy to reconcile with the skewed distribution of firm performance . (I'd welcome a Keynesian-type rebuttal here).
2. It suggests there should be scope to insure ourselves against recessions. Insurance works if people are in different situations, with a minority suffering whilst others thrive. This is just what happens. This deepens the puzzle of why Robert Shiller's proposals for macro markets have not caught on.
What's more, found Geroski and Gregg, it's "very hard" to predict which firms will suffer in a recession. Pre-recession profits and sales were only weak predictors of suffering. And many firms in badly affected industries thrive whilst other firms in growing sectors suffer.
This should be a reminder  to stock-pickers not to be too confident about their abilities.
They also cast doubt on the view that recessions can boost long-term growth by weeding out lame ducks or by stimulating efficiency:

Save for those firms who fail, it seems hard to see any permanent effects of recession.

It would be a good tribute to Professor Geroski, as well as a good way to improve thinking about the economy, if people would read or re-read this fine book.

Hayek, Porn and Irrationality

Violent pornography is the latest target in New Labour's war on freedom. This seems another example of how New Labour responds to emotional bribery more than to  rationality.
The Daily Mail reports:

The proposals were welcomed by the family of 31-year-old Jane Longhurst who was murdered in 2003 by a male friend obsessed with violent sexual pornography. Her mother, Liz, from Reading, Berkshire, said: "Over the last 18 months I have worked hard to change the climate of public opinion through the Jane Longhurst Campaign by raising public awareness of the issue....[MP Martin Salter] said he "couldn't be more thrilled" with the news that 18 months of "hard work" was paying off. He said he was proud of Mrs Longhurst's courage and motivation, adding that she was motivated "by the desire to ensure that other patients did not experience her dreadful loss when her lovely daughter Jane was murdered by a sick, self-confessed addict of extreme internet porn".
"Now at last the loophole in the law caused by the creation of the internet will finally be closed when legislation is introduced, as I am sure it will be, to make it illegal to possess computer images which contravene the Obscene Publications Act," he continued.

It is grossly illogical - an example of the representativeness heuristic - to infer from this one terrible case that violent pornography causes violent sexual crime. To know the true link, we need to know how many viewers of such pornography did not commit crimes, and how many such criminals did not view such pornography. Such data are hard to get. As the Home Office itself says:

As to evidence of harm, conducting research in this area is complex. We do not yet have sufficient evidence from which to draw any definite conclusions as to the likely long-term impact of this kind of material on individuals generally, or on those who may be already predisposed to violent or aberrant sexual behaviour.

The representative heuristic is not the only cognitive bias underpinning the desire to ban violent porn. Another bias is our tendency to be over-confident about identifying beneficial effects of our actions, whilst under-estimating possible, but less foreseeable, costs. Hayek was clear that this bias was a cause of many infringements on our freedom. In Law Legislation and Liberty (Vol I, p56-57) he wrote:

Since the value of freedom rests on the opportunities it provides for unforeseeable and unpredictable actions, we will rarely know what we lose through a particular restriction of freedom. Any such restriction, any coercion other than the enforcement of general rules, will aim at the achievement of some foreseeable particular result, but what is prevented by it will usually not be known....And so, when we decide each issue solely on what appear to be its individual merits, we always over-estimate the advantages of central direction.

An unnecessary disclaimer: Despite the abbreviation of this blog, I have no personal interest in violent porn. My tastes are plain vanilla.

Discrediting egalitarianism

Polly Toynbee, as usual, suggests some ways to reduce inequality:

The minimum wage could safely become a living wage at £6.20 an hour...The state itself should blaze a trail and set a far higher minimum rate for all its employees and contractors. Taxes would have to rise to pay a higher wage bill.

This just won't do at all. As the Department for Work and Pensions and I pointed out yesterday (big pdf), marginal withdrawal rates are so high that any rise in the minimum wage has only small effects on household incomes.
Take a single parent working 30 hours a week for the  minimum wage of £5.05 (as of October this month.) If she can continue working 30 hours a week after the NMW has risen to £6.20, her weekly net income would rise by just £3.38. A 22.7 per cent rise in the NMW would lead to a rise in income of just 2 per cent. And this assumes - heroically - that a 22.7 per cent rise in labour costs will lead to no cut in the demand.
This is, of course, not a new finding. As Holly Sutherland pointed out in this paper (pdf):

The direct effect of the NMW on poverty rates and the intensity of poverty is relatively small. 

There are two things I find offensive about Ms Toynbee's claim. First, it comes wrapped in blather about how "this week's research" bolsters her argument - when a vital piece of that research, from the DWP, in fact undermines it.
Second, there is the  glib social democratic prejudice that we can achieve greater equality merely by apllying a greater dose of current reformist policies.
We can't. Serious egalitarians must be more radical. Ms Toynbee is discrediting egalitarianism with the feebleness of her arguments.

Hollowed out

Larry Elliott describes, without endorsing, the case for protectionism:

There are those, though, who see textiles as merely the thin edge of a very large wedge. What will happen, they say, when China can flood the world with everything from pharmaceuticals to financial services?...From this perspective, the whole of western manufacturing and much of its service sector will be hollowed out unless there are more vigorous measures to meet the threat of China.

But what would an economy look like with its manufacturing "hollowed out"?
It'd look like my street, that's what. Almost no-one in Belsize Park Gardens, to my knowledge, works in manufacturing. And there are loads of service industries where no-one's employed either - and certainly, no agricultural jobs (though some do grow their own herbs).
This isn't jsut true of my street. It's true of most of the area. The people round my way work pretty much in just two industries: entertainment/meeja and financial services.
And are we poor? No. Sure, we have problems with single parents and drug users. But we're pretty rich, despite a catastrophic hollowing out of manufacturing, and despite the fact that almost all the food and clothing we buy is imported, sometimes from as far away as Camden Town.
What, then, is the difference between  my area and a country that makes hollowing out a problem in the latter but not in the former?
It's certainly not that only a few people have the brains to work in the meeja or banking whereas a whole nation cannot. Such a view misunderstands both these industries and the nature of comparative advantage.
Larry commits the latter error when he says:

Countries specialise in what they are good at.

This is wrong. People - countries if you must - merely specialize in what they are relatively least bad at.  Even if China  were to become absolutely better than us at producing everything, we could still earn a living by specializing in what we were less bad than them at producing.
Think of it this way. I am a worse economics writer than Larry Elliott*. But I can make a living by specializing in a field - financial writing - where my inferiority to him is smaller.
So, here's my question. If I, and my neighbours, can do OK without manufacturing or agriculture, why can't larger geographical units?
I suspect this is one of many ways in which thinking about countries rather than people merely  clouds our judgment.
And: Tim is good on this.
* This isn't false modesty. I lack two talents necessary for any columnist: the ability to pander to people's prejudices and to repeat myself week after week.

Absurd

The Department for Work and Pensions has today released a report which strongly supports the principle of a flat tax and citizen’s basic income.

Well, it hasn’t done so explicitly. But its latest Tax Benefit Model Tables (big pdf) are the next best thing.

Here’s a question: Take a married couple with two children under 11 and pre-tax earnings of £200 a week. If they get a better job, raising their earnings to £300 a week, by how much does their net income rise?

£60? £50? £40?

Nope. £8.52.

Yes. £8.52. That’s a marginal deduction rate of 91.5 per cent.

The extra £100 this couple earns before taxes are swallowed up by higher income tax and National Insurance Contributions (£33); lower Working Tax Credits (£37) and less Housing Benefit (£19.50).

Higher earners face only slightly lower deduction rates. If a couple’s income rises from £500 to £600, they get to keep all of £23.45.

In the range £630-£710, the marginal tax rate drops to 23 per cent. But beyond then, it rises to 41 per cent.

Of course, things are different for different family types. But the fact remains. Marginal deduction rates for many people on lower earnings are horrifically high; 89.5 per cent for a single parent with one child earning less than £400 a week, for example.

For millions of people, it’s a bad idea to work harder.

Is this either economically efficient or a good thing for low earners?

Is this really the best tax system anyone can think of?

Woman: know your place

What, I wonder, is the point of this?

Men are more intelligent than women by about five IQ points on average, making them better suited for “tasks of high complexity.”

First, this is not a new story. Richard Lynn’s website tells us that he published a similar claim back in 1999.

Second, five IQ points is only one-third of a standard deviation. That means around 37 per cent of women have a higher IQ than the average man. Millions of women, then, are “better suited for tasks of higher complexity” than millions of men.

Third, IQ is only a very narrow measure of ability. For this reason, it is not very highly correlated with productivity and earnings. (Though for a different view, try this.) Countless things affect an individual’s productivity other than IQ: concentration, discipline, motivation, ability to get on with others, and so on. Everyone knows high IQ people who are economic failures and low IQ people who are successes.

Fourth – and this should go without saying – IQ is no measure at all of the value of an individual.

So goes the PC version.

Now for the non-PC version. This story corroborates my experience. The women I’ve met have, generally speaking, been less intelligent than the men; a disproportionate number of the exceptional women have come from eastern Europe.

Is there a selection bias here, with intelligent women disproportionately deterred from entering the City and journalism? Or is it that intelligent women are smart enough to know not to have anything to do with me?

Oil and the stock market

There's lots of good analysis of the likely impact of high oil prices on the US economy by James Hamilton, Nouriel Roubini and Barry Ritholtz. Can I add a couple of things?
First, even if we leave aside the oil shocks of the 1970s, there was for years been a close negative relationship between the real oil price (West Texas deflated by core CPI) and the price-earnings ratio on the S&P 500. Between 1983 and 2003 the correlation between the two was minus 0.5, with an R-squared of 24.7%.
A regression relationship based on 1983-2003 data predicts that the PE ratio should now be just 12.4 - implying the S&P would be 40 per cent below current levels.
This suggests stock market investors are strongly buying into the story that the economy will remain resilient to high oil prices - which suggests there's some potential for a nasty surprise.
Second, there's this paper by Ben Jacobsen and colleagues:

Investors seem to underreact to information in the oil price: a rise in oil prices drastically lowers future stock market returns. These predictability results are economically significant and robust with respect to sample periods, different kinds of oil prices we consider and the usual variables known to forecast stock returns.

Love and hate

From Clive Davis comes this story of what happened when one of my greatest loves met one of my greatest hates.
Not all financial advisors are worthless scumbags. But it's a good idea to act on the assumption that they are.

A new use for Saddam Hussein

The closure of the Darley Oaks guinea pig farm is a setback for scientific research. But there's an obvious solution - to use humans instead of animals.
I don't mean any humans - merely persistent or serious criminals. It's about time Saddam Hussein did something useful.
To see the case for doing so, consider why we don't (openly) experiment on humans now.  It is, presumably, because  humans have rights that animals don't.
But what is the basis for these rights? It can't be because humans suffer, because  animals do too. Nor can it be because humans have some functionings like higher consciousness that animals don't - because we presume that the severely mentally handicapped or people in persistent vegetative states have rights.
No. The strongest case for believing that humans have rights is that we have all entered - actually, implicitly or potentially - into some kind of social contract with each other. As David Gauthier (just one of many contract theorists of course) put it in Morals by Agreement:

The moral claims that each of us makes on others, and that are expressed in our rights, depend, neither on our affections for each other, nor on our rational or purposive capacities, as if these commanded inherent respect, but on our actual or potential partnership in activities that bring mutual benefit.

This view implies that some of the higher animals - cats, dogs, horses - can acquire rights. Surely, we are in partnership with them in "activities that bring mutual benefit." And don't think scientists don't experiment on these; these statistics suggest otherwise.
Equally, it suggests serious or persistent criminals can lose (some?) of their rights, by breaching the terms of the contract. So why not use them for vital medical experiments?*
It's tempting to think we don't do so out of mere squeamishness. But my instincts here differ. I would far sooner inflict great pain upon Saddam Hussein - to name but one of many - than hurt a cat.
So, what am I missing? What is it about "humanity" that generates rights unique to humans? (These guys have one intriguing answer). And why can't people who choose to act in less than human ways forfeit these rights?
* There's a good utilitarian argument for doing this as well. The medical gains, plus the deterrent to criminal activity, could easily outweigh the suffering of the criminal.

Euro effects

Ken Clarke has always struck me as an oaf, even by the demanding standards of the Stupid Party. His latest utterances on the euro - if understood in narrow economic terms - do not not change my mind. He says:

I thought [the euro] would lead to increased productivity, efficiency and living standards and stimulate policy reforms. On that front so far it has been a failure.

But surely, no intelligent person would have expected any boost to productivity to show up yet.
The main way in which a single currency would raise productivity is by cutting the transactions costs of cross-border trading. This could raise productivity in two ways. It could raise profits, thus encouraging investment in productivity-enhancing equipment. And/or it could encourage aglommeration - the concentration of industries in particular regions - which in turn raises productivity through either economies of scale or network effects (whereby companies that are closer together learn from each other).
However, these mechanisms   take years to show up (If that is, they exist at all - the classical dichotomy should remind us to be sceptical of whether monetary arrangements can have permanent real effects). Any reasonable supporter of the euro before 1999 should therefore not regard the experience of the last five years as strong evidence against the productivity-enhancing effects of the euro.
Sure, the euro zone's economic performance has been poor. But this is due to (among other things) competition from low-wage economies and over-regulated markets, rather than to any effect of the euro.
If you want a more rigorous analysis, try this paper (pdf) by Hashem Pesaran and colleagues. It concludes:

UK entry to the euro in 1999Q1 at 1998Q4 exchange rates would probably have reduced UK GDP in the short term and raised it in the long term, but the effects are small...the welfare effect is ambiguous, depending on the discount rate used, and on the relative weights given to output and inflation.

There's nothing in this to justify a supporter (or opponent) of UK entry into the euro  changing their mind. That Mr Clarke has done so cannot be justified by the economic evidence.
Note: I write as an opponent of the euro, for narrow economic reasons that aren't relevant now.