Stumbling and Mumbling

The virtue of mediocrities

In the few days I’ve been blogging, I’ve already fulfilled one of bloggers’ main obligations – to slag off the Groan. But I’m reluctant to fulfill another – to blog about the US election.

One reason for this is that no-one can foresee the full consequences of a Bush or Kerry victory. It’s a cliché that no-one knows what Kerry really believes. And in 2000 no-one foresaw that a Bush victory would lead to a war on terror, humungous Budget deficits or rising tariff barriers. Very few presidencies are defined by a President enacting policies he had promised beforehand. Instead, they are dominated by image and how the President tackles unforeseeable contingencies. So we’ve had Carter’s well-meaning impotence in the face of the Iranian hostage crisis; Reagan making America feel good about itself, Clinton and Lewinsky. And so on.

This means elections are dominated by question of character, rather than policy substance. And this degenerates into tribalism. Are we members of the Bush tribe or the Kerry tribe?

Equally, though, I don’t like the clichéd de haut en bas approach of the newspapers. “Most unimpressive candidates in recent history”, lesser of two evils”, “fine judgment”, “on balance,” “grave misgivings.” Build your own leader column.

Also, like most people, I don’t like to confront dissonant facts. I prefer to think of the US as the last great hope of mankind, the home of Dar Williams and John Roemer, not as the country in which Bush or Kerry are popular.

Which raises the question. Could it be that the poor quality of the US’s politicians and its leadership in culture, business and academia are two sides of the same coin? One reason why the US has succeeded whilst Communist and many African states have failed is that in the US talented and creative people go into jobs where they can do good for others – this includes business as well as academia – whereas in failed states such people go into politics where corruption and rent-seeking can make them fortunes.

It’s a cliché of the Groan-reading class to bemoan the fact that the election is a contest between two rich men. But isn’t it so much better that politicians are rich before taking office than that they become rich whilst in office?

Perhaps a contest between rich mediocrities is just the price we must pay for the US being the great nation it is.

Britain's productivity slump

Why does productivity grow? Common sense says it’s because firms become more efficient. However, a paper in the latest Oxford Review of Economic Policy shows that this is only part of the story – and a shrinking part at that. (The paper’s only available to subscribers, but related papers are here.)

Jonathan Haskel and colleagues have estimated that in the late 1990s 40 per cent of UK manufacturing productivity growth came not from existing firms becoming more efficient, but from new firms entering the market and lame ducks leaving. In chemicals and pharmaceuticals, four-fifths of productivity growth was due to entry and exit.

In the late 80s, however, only 15.6 per cent of productivity growth came from entry and exit.

The authors attribute the increasing role for entry and exit to globalization and technical change. It’s often easier for new firms to exploit new technology, as old ones face costs of retooling and retraining.

There is, however, another story here. National Statistics estimate that aggregate productivity growth slumped from 23.5 per cent in 1985-90 to 14.3 per cent in 1995-00. With an increasing fraction of this slower growth due to entry and exit, it follows that the productivity growth of incumbent firms has collapsed.

This raises the question: could it be that traditional hierarchic firms are less and less able to increase efficiency?

Should we raise the minimum wage?

In today's Groan, Polly Toynbee calls for a rise in the minimum wage.

Let's clear up the factual errors in this piece.

Claim one:

the Gini coefficient increased between 1996 and 2003.

Table 27 in this article from National Statistics shows that it fell.

Claim two:

alarmist lobbying [by the CBI] caused the minimum wage to be brought in at such a low rate that it has benefited about half as many people as the government originally intended.

True, fewer people were covered by the minimum wage than intended. However, David Metcalf, who chaired the Low Pay Commission which recommended the level of the minimum wage, has blamed poor official statistics for this. The ONS over-estimated the numbers in low pay in 1999. "The [Low Pay] Commission has been badly let down by the ONS" he told the FT in November 2002.

Locke on Property

Today is the 300th anniversary of the death of John Locke. A good excuse - not that one should be needed - to celebrate liberalism and reaffirm some reasons for re-reading the Second Treatise of Government.

One reason for re-reading the treatise is that it shows why liberty has so often been infringed. The problem is that Locke's defence of liberty rests on a defence of property rights. As this defence is inadequate, governments find it easy - too easy - to justify infringements of liberty on the grounds that they protect non-property owners.

Regulating Gambling

The ususally admirable Civitas has surely conceded too much to the illiberals. In this post, David Green says:

It is a legitimate use of law to regulate organisations who exploit the weakness of other people for monetary gain

However, countless transactions exploit people's weakness. Tesco exploit my weakness for salt and vinegar Pringles, every investor who buys a share he believes to be cheap is exploiting the perceived intellectual weakness of others, WH Smith exploit people's weakness for reading the Spectator. And so on. If we allow regulation whenever an organisation exploits people's weakness, we will rapidly lose all economic freedom.

This is especially true because it is easy for us all - and especially policy-makers - to over-estimate people's weaknesses. Too many people agree with Homer Simpson, that "everyone is stupid except me."

Rather than debate the issue of gambling regulation from first principles - which invites a futile discussion about human nature - why don't we ask a more limited question: should gambling be more or less tightly regulated than alcohol?

I say less, for three reasons:

1) Alcohol use imposes a bigger external cost onto others. The Home Office estimates that there are 1.2 million cases of alcohol-related violence a year. Sure, many of these are on other drunks, and "related" doesn't imply causality. Even so, an innocent person is much more likely to be assaulted because someone has gotten drunk than because someone has gambled. And this is not to mention the possible role alcohol plays in increasing other crimes, such as burglary.

2) It's easier to slip into alcohol addiction than gambling addiction. A heavy social drinker is a good bloke, one of the lads. It's easy for some people to slip from this into alcoholism. There's no such social convention which encourages the first steps into gambling addiction.

3) Alcoholism is physiologically more damaging than gambling addiction. 6000 people a year die directly as a result of alcohol. And this is not to mention the cost to the NHS of treating alcohol-induced illnesses. Also, it's probably tougher to kick an alcohol addiction than a gambling addiction.

You could argue that these factors are a case for tougher regulation of alcohol. I prefer to think - given my Bayesian priors about the value of liberty and the freewill and intelligence of most people - that they are a case for lighter regulation of gambling.

Limits of arbitrage

Both fans and critics of the efficient market hypothesis should watch the price of PZ Cussons, makers of Imperial Leather.

The company has two classes of share - ordinary voting ones and non-voting A shares. Today, the premium on the voting shares hit a 17-month high of 20.5 per cent. In July, it was as low as 3 per cent. It's averaged 10-15 per cent in the last 16 months.

Now, you'd expect the premium to rise a little ahead of the AGM on Monday; there's a vote on a share buy-back then. But does this really justify a 20.5 per cent premium? And can anything justify the volatility of the premium? In the last 52 weeks, the correlation between weekly moves in the two classes of shares has been just 0.65. That's the correlation between two different companies.

Instead, this might show us that arbitrage has its limits. A rational investor, seeing an over-valued voting share, might be frightened of shorting that share and buying the A share, for fear that the over-valuation might persist. The upshot is that mispricings can persist.

Ifyou want evidence of this, just look at one of the world's biggest companies - Royal Dutch Shell. In the US the premium on Royal Dutch over Shell's ADR has ranged between 10 per cent and 35 per cent in the last seven years; it's now 16 per cent. If highly liquid shares in the same company can vary so much, doesn't this show how powerless the smart money is?

Now, this doesn't mean there's easy money to be made by shorting PZ Cussons voting shares. Anomalies exist and persist precisely because they are risky or expensive to eliminate.

Instead, it means two things. First, even the law of one price needn't always hold. Second, markets can be inefficient without yielding money-making opportunities.

We're all doomed...

...or so tomorrow's newspapers might tell us, in reporting these figures from the British Bankers' Association.

The two key points are:

The number of loans for house purchase approved in September were 30 per cent lower than last September.

The average value of those loans has risen just 5.1 per cent in the last year. This is much lower than the rise in house prices recorded by the Nationwide (17.8 per cent) or Halifax (19.8 per cent). Last year, there was no such discrepancy.

But will there be a hard or soft landing for the housing market?

One point that gets overlooked here, I think, is that many house buyers are momentum investors. You can't explain the big rise in prices in the last three years without reference to bandwagon effects.

And momentum can go both ways. Once prices start to fall, people might postpone buying, in the hope of getting bargains later. This could cause prices to fall below "fair value". At least it would unless this effect is offset by a reduction in supply as a result of potential home-sellers not putting their house on the market.

This doesn't mean there'll be a crash in the sense of us waking up tomorrow and finding house prices have fallen 10 per cent. But it does raise the possibility of a longer, and ultimately deeper, fall in prices than the "soft landing" forecasters would have us believe.

For me, though, the question of where house prices are going isn't anything like as important as newspapers would have us believe. But that's a post for another day...

Goodhart's law: a codicil

A new paper from the Bank of England puts a twist onto Goodhart's law.

Goodhart's law says that:

Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.

A codicil to this is that the regularity will re-emerge once policy-makers no longer rely upon it.

And this is just what's happened to the Phillips curve - the trade-off between unemployment and inflation. In the 1950s and 60s, when policy-makers (supposedly) tried to exploit this trade-off, the correlation between the two was unstable. It became horrendously so in the 1970s.

However, says Luca Benati, since inflation targeting began in 1992, a stable trade-off between unemployment and inflation has re-emerged - just when no-one's been been using it for control purposes.

This finding raises the possibility that some people will call for the inflation target to be raised, as a little more inflation would allow unemployment to fall.

I think this would be a mistake. It would invite Goodhart's law to work again. And at current low levels, unemployment is a microeconomic phenomenon, not a macro one.

What this does show, though, is one problem economists have that natural scientists don't. Scientists don't have to regularly re-do experiments to check whether established findings still hold (though this old story suggests that they should do so). Economists, however, have to keep checking their established ideas. Which is one reason why our job is so tough (and interesting).

A depressing fact

It's cheaper to go to Saturn than to Manchester, measured by pence per kilometre.
Cost of Cassini-Huygens mission = £1.81 billion
Distance travelled now = 1.329 billion kilometres.
Cost per kilometre = £1.36
Cost of first-class single ticket from London to Manchester for family of four = £420.
Distance = 264 kilometres
Cost per kilometre = £1.59.
Sure, it's a lot cheaper standard class and booking ahead. And on Saturn you can't throw soup at Scotchmen, but even so...

Sexism, potential and New Labour

Trade Secretary Patricia Hewitt has reminded us today that there really is a New in New Labour. She's said:

Career sexism limits opportunities for women of all ages and prevents them from achieving their full potential

What's new about this is the belief that the workplace is or should be a place where people achieve their potential. Four strands of thinking say it isn't.
The obvious one is classical Marxism. This says the (capitalist) workplace is a place of alienation, in which people are not the masters of objects, but their slaves.
Secondly, there's Harry Braverman's famous thesis in Labour and Monopoly Capital, where he showed that craft skills were being destroyed and degraded by managers' efforts to control the workplace.
Thirdly, there's Adam Smith's observation that the division of labour weakens intellect and character. The worker, he says:

generally becomes as stupid and ignorant as it is possible for a human creature to become. The torpor of his mind renders him not only incapable of relishing or bearing a part in any rational conversation, but of conceiving any generous, noble or tender sentiment, and consequently of forming any just judgment concerning many even of the ordinary duties of life.

Finally, there's the contention that the workplace today has been hijacked by a managerialist ideology in which measurable competence and hard skills are subordinated to softer unmeasurable attributes like interpersonal skills and the (perceived) ability to manipulate symbols and images. On this point, I can't improve on Robert Protherough and John Pick's Managing Britannia. Echoing Smith, they argue that managerialism drives out some desirable character attributes:

In the managerial lexicon, British citizens may no longer be commended for being gentle, amateur, scholarly, patriotic, modest, unambitious, dutiful, charitable or sportsmanlike. To be taken seriously modern Britons must be ambitious, entrepreneurial, hard-nosed, interventionist, focussed, driven, work-centred and, above all, committed to modern managerialism

There is, therefore, a long and strong tradition which says that the workplace necessarily stifles potential. Sexism is just one of many ways in which it does so.
This raises the question: why does Ms Hewitt assume that the workplace should develop potential? I suspect it's a reflection of a central aspect of New Labour ideology - the belief that there are no trade-offs, so we can have both economic efficiency and human development.
Adam Smith, however, thought there was a trade-off. What if he was right?