Stumbling and Mumbling

Costs of the "middle England" error

The TUC’s Nigel Stanley points out that “middle Britain” is a lot poorer than the right-wing media make out. This tendency to exaggerate the incomes of “middle Britain” has some pernicious effects.
The obvious one is that it allows the rich to deceive others (and themselves) into believing that they are ordinary people, and victims of “NuLabour attacks on the middle class” rather than what they really are - a privileged elite;  a salary of £50,000 puts one in the top 10% of earners.
However, I suspect the tendency to over-estimate “middle England” incomes has (at least) four adverse effects upon those on truly middle incomes, namely:
1. Lower self-esteem. Our estimates of self-worth are connected to our income - they call them “earnings” for a reason. And we measure this worth relative to other people; it’s for this reason that income is a positional (pdf) good (pdf). This means that if we over-estimate median incomes, we’ll under-estimate how well we are doing, and so feel worth less.
2. Over-ambition. If workers feel they are earning less than others, they’ll make more effort to climb the greasy pole. This isn’t just aesthetically unpleasant; as Mill said, “I am not charmed with the ideal of life held out by those who think that the normal state of human beings is that of struggling to get on.” It could also be counter-productive in aggregate. This paper shows that workers who seek promotion naturally invest in those skills which their managers can more easily evaluate. This can lead to over-investment in general managerial or salesmanship skills, and under-investment in technical geeky skills which bosses can’t understand. The upshot can be lower aggregate productivity.
3. Excessive debt and consumption. Robert H. Frank has written extensively about how perceptions of others’ wealth cause people to spend more on conspicuous consumption - bigger cars and houses - and so get into bigger debt. Spending is contagious. The more we over-estimate the Jones’ incomes, the more we’ll struggle to keep up with them.
4. Crime. When people think there’s money around, they become more selfish. So if they over-estimate others’ incomes they might become more mean-spirited and even prone to crime. Those MPs who fiddled their expenses did so in part because they thought (wrongly) that an income of £60,000 wasn’t much.
Now, I’m not saying these are large effects - just that they might be non-zero. If so, the persistent over-estimate of “middle England” incomes might be nastier than generally supposed.

Do we really want fewer MPs?

There’s much to like about David Cameron’s speech yesterday: the demand for “a massive, sweeping, radical redistribution of power“; the recognition that powerlessness creates irresponsibility; the bundling together of bankers and politicians as part of the same over-mighty elite, and suchlike. Curmudgeon that I am, though, I want to raise some doubts about one of his proposals - to reduce the number of MPs.
I fear there are (at least) three problems here:
1. How do we redraw the boundaries? Take my area, Leicestershire and Rutland, as an example. The natural way to reduce the number of constituencies is to merge the three city constituencies and merge some of the county seats. This, though, could just create a few safe seats - one or two safe Labour ones in the city, and - depending on what happens to Loughborough - safe Tory ones in the county.  
Alternatively, we could create more marginals by combining bits of the City with the outlying county.
The latter would make politics healthier. But it would give us constituency boundaries that have no clear basis in geography or local affiliations.
2. Which MPs do we lose? Under current arrangements, there’s a danger that the MPs who get to fight the new, larger, constituency elections will be the party loyalists. We could therefore lose the mavericks, and end up with an even more homogenous parliament than we already have.
3. What to do about MPs’ constituency work? Bigger constituencies mean MPs would have even more local case-work to do. That would leave them with even less time for the “proper, impartial, effective scrutiny” of legislation that Cameron rightly wants.
These problems are not insuperable - Cameron’s proposal for open primaries, for example, might mitigate the second problem, though I fear it might bring other difficulties. But they do suggest that, on its own, the call for fewer MPs might not a great idea.
* Apologies for the lack of posting recently. This was because of a family disaster - they came to  visit.

Redwood's anti-conservativism

What kind of a conservative is John Redwood? I ask because his continued attack upon the Bank of England seems to rest upon some unconservative assumptions.
He says:

Some see it as commendable humility and honesty to tell us on many occasions they do not know what will happen next.
It should lead instead to questions about why they employ so many highly paid economists and forecasters if they do not have a well informed opinion. Why issue forecasts at all, if you they are as wrong as the Bank’s have been in recent years, and if even the boss has little confidence in them?

However, the Bank’s inability to foresee the future is not a failing in the Bank, but a fact about the economy.
Put it this way. How much money are you going to spend in May 2010? You don’t know. And you don’t know because you haven’t decided yet. If you’re a consumer, you don’t know what whims or needs you’ll have, or what new products will be in the shops. More importantly, if you’re a businessman, you won’t know what investment opportunities you’ll perceive, or what new ideas you‘ll have.
In other words, the economy is unforecastable because people have free will - they choose; G.L.S Shackle makes this point  in this interview.
But surely, conservatives should be well aware of this. I thought key parts of conservative thinking were that free will exists, and that this generates a complex society which is hard for policy-makers to control or forecast.
We could only perfectly forecast the economy if we could anticipate the billions of decisions that dispersed individuals make. If we could do this, central planning would become feasible. But this is not the world we live in. And I'd have expected free market conservatives who are supposed to be acquainted with Hayekian thinking to know this. Such people should be welcoming Mr King's acknowledgement of ineradicable uncertainty, not attacking him for it.
Second, Redwood says:

This economy needs more savings. That requires higher interest rates than the Bank’s guideline rate.

But why do we need more savings? It can‘t be because he thinks we need to increase capital spending - because higher official interest rates are no way to achieve this. Another possibility is that our low savings rate has left the economy more vulnerable to shocks; people without savings must cut their spending if their income falls whereas those with savings can use them to smooth spending in the fact of shocks.
However, retail sales have held up well so far, which suggests households have been able to smooth spending in the face of shocks to house prices and incomes.
And again, the presumption that people save too little seems a very anti-conservative one. The idea that people are poor judges of their own self-interest - how much to save - and that politicians know better than they do seems inconsistent with conservatism, especially in its Thatcherite form.
I say all this from confusion.  How can the presumptions behind Redwood’s attack on the Bank be reconciled with any sort of  free(ish) market conservatism? I think I'm missing something, but what?

MPs' pay in historic context

How are MPs paid compared to average workers, by historic standards? My chart tries to answer this question. 
I constructed it as follows. First, I got the history of MPs’ pay - before allowances - from appendix A of this pdf: I divide this by 52 to get a weekly wage. Then, the Annual Survey of Hours and Earnings told me that the average worker was paid £471.9 a week in 2008; the average is well above the median. This means that in 2008, MPs were paid 2.58 times as much as the average worker. Latest labour market data (pdf) update earnings for 2009. The history of average earnings comes from here.
Putting it all together shows that MPs’ wages have trended upwards, relative to the average worker, since the late 70s. Because wage inequality has risen during this time, I suspect that they’ve risen even more relative to the median worker.
Even now, though, they aren’t brilliantly well paid by historic standards.
The best times for being an MP were just after a salary was introduced in 1911, as it came in at more than six times the average wage. The late 30s were also a good time, as were the mid 60s. Indeed, even throwing in the maximum second homes allowance still leaves MPs worse off now, relative to the average worker, than they were in the mid-60s before expenses.
Whether such high wages in the 30s and 60s attracted a better quality of MP than we have now is a question I’ll leave for others.

Immigration & GDP

With the BNP expecting to do well in the European elections, immigration will become more of an issue. Two recent papers should help us here.

First, this one (pdf) estimates, from a sample of 63 countries, that immigration actually raises GDP per head; a 10% rise in the stock of migrants raises it by 2.2%. This controls, as far as possible for the reverse causality - the tendency for rich countries to attract migrants - and for the correlation between openness to migrants and openness to trade.

There are two possible mechanisms generating this. One is simply that an increased supply of labour reduces inflationary pressures, and so allows real interest rates to stay lower for longer, which encourages economic activity*. The other is that migrant workers might be complements for native ones. If Polish builders cause more houses to get built, British plumbers and electricians can get more work done and earn more. If Latvians cause more fruit to be picked, British market stall holders can sell more. Or if you can see a doctor quicker because of migrant GPs, you can get back to work sooner.

This finding is entirely consistent with the notion that some unskilled native workers lose out from immigration. What it means, though, is that migration is a potential Pareto improvement, in the sense that winners from it can in theory compensate the losers with everyone becoming better off. This provides a justification for the Left’s response to fears about migration - that the way to protect unskilled native workers is not to restrict migration, but to have better redistributive mechanisms.

Our second paper (pdf), however,  is slightly more pessimistic. It estimates, from 14 OECD countries, that immigration simply increases raw GDP; native per capita GDP is unaffected by migrants.

This effect, though,  is the average of good times and bad times. In normal economic times (when GDP is within 1% of its trend level) immigration actually increases native employment. In recessions, however, it depresses it slightly; for every 10 immigrants, employment expands by only 9, implying that  either one immigrant is unemployed, or that he finds work at the expense of a native.

Whether this is an argument for restricting immigration in recessions, or for better counter-cyclical policies, is, however, questionable even in narrow economic terms.

In one important sense, though, both papers agree - that, on average, immigration does no harm to the native economy, although you could easily argue that 2.2% is a small gain.

The problem with the BNP, then, is not merely that they are vicious thugs. It’s that they are plain, factually, wrong.

* I suppose you could use this mechanism to argue that immigrants caused the banking crisis - but only in the same way that rope-makers cause people to hang themselves.

MPs' wages

It's insufficiently appreciated just how well paid MPs are. Their annual salary is £64,766. According to the IFS, a single person on this income is in the top 3% of earners. If he has a partner earning £30,000 a year, this couple is also in the top 3% of household incomes.

If a single MP claims £20,000 for waxing his drawbridge on top of this, he enters the top 1% of earners.

Even if an MP has a non-working partner and four children, his salary, plus £20,000 of second home allowance, puts him in the richest one-fifth of households.

Even if we concede that an MP works longer hours than average and look at hourly pay, MPs are still better off than most - in the top quartile for single persons.

However you cut it, then, MPs are unrepresentative of their constituents in terms of earnings.
What could justify this?

The Times argues that MPs should be paid £90,000 a year - similarly to GPs or head-teachers. This, I think, is mistaken. It takes years of experience and training to become a GP or head-teacher. No such qualifications are necessary for MPs. And GPs and head-teachers must exercise considerable judgment. MPs don’t. As Alice Miles writes:

With honourable exceptions — Labour’s Frank Field particularly stands out– MPs do little to apply any individual thought to the legislation before them...Watch an MP go off to vote: he will ask colleagues en route how they are supposed to be voting on this one.

So how should we pay MPs? Much of their time is spent dealing with constituents’ housing or immigration issues. Which suggests they should be paid like social workers. And the average social worker earns less than £30,000.

Yes, MPs should get more than this to reflect the fact that most of them must incur the expense of a second home. But the idea that they are under-paid now reflects, I suspect, the media’s “middle England” fallacy - the mistaken belief that high incomes are far more common than they in fact are.

Expenses and rule fetishism

Several of you have expressed dissatisfaction with MPs’ argument that their expenses claims were “within the rules.” What’s not sufficiently appreciated is just how deeply inadequate this defence is.

The problem is not just that it’s easy to obey rules if you make them yourself - - if indeed MPs have done this. It’s that the claim “I didn’t break any rules” is a category mistake. It confounds good, or merely acceptable, behaviour, with following rules. It’s a form of fetishism. Just as a foot fetishist mistakes a foot for a sexual object, or a primitive tribe mistakes a totem for supernatural properties, so MPs mistake adherence to rules for virtuous conduct.

I say this is a category error because following rules under-determines good behaviour. One can follow rules and behave abominably - “I was only obeying orders” is no defence -  and one can  behave well by breaking even quite serious rules; some assisted suicides, I think, fall into this class. To merely follow rules is to suspend one’s own moral judgment, to behave not as a human being but as an automaton.

The MP who checks every item on a till receipt against the expenses rulebook (pdf) is acting within the rules. But he is not behaving with dignity, nobility or virtue; as Libby Purves points out, thousands of people often under-claim expenses.

The contrast here is between “rule fetishism” and what the ancient Greeks called phronesis - the wisdom, or prudence, that allows people to judge what is virtuous conduct without needing to read a rule book. The man with phronesis would not have even tried to claim for Remembrance Day wreaths on expenses as he would know that only someone "without personal integrity, sans honour, sans virtue" would do such a thing.

This contrast, of course, extends way beyond the expenses row. The fact that New Labour has created 3000-plus new criminal offences since 1997 is testament to a belief that rules can determine conduct, and that people are incapable of exercising their own judgment as to what is right or virtuous.

In this sense, MPs' grossly defective defence of their ignoble greed and their illiberalism have a common root.

But why is it that MPs have supplanted phronesis with this rule fetishism? Is it because they are simply unaware of the possibility of virtue ethics as an alternative to priggish adherence to ever-accumulating rules? Or is it that they believe that they and we are incapable of exercising phronesis?

Worse still, might they be actually right to believe so?

Confusion about QE

Fraser Nelson is hopelessly confused here. He says:

The massive Quantitative Easing programme is making it harder for companies to raise money, because the government is flooding the market with its own IOU notes. The Bank of England today confirmed that less than 1% of the £44.5bn it has printed has gone to buy company loans – it had indicated that as much as a third of the £150bn pool would go to companies. Instead, it is a mechanism to help the government issue the £240bn of gilts it’s issuing this year.

He is correct that QE has overwhelmingly been used to buy gilts. But you can’t complain both about this and about the government crowding out the corporate sector. If the Bank of England were to buy all the gilts issued this year, the government wouldn’t have to tap the private sector for cash at all, leaving it free to fund companies. Bank of England gilt purchases are the solution to the crowding out problem. 

If the markets think QE is actually a way of one department of the government printing money for the other departments to spend (a la Weimar Germany), then confidence in the currency collapses.

And if my aunt had bollocks she’d be my uncle. The fact is that confidence in the currency has held up since the Bank announced the start of QE in early March. Sterling’s trade-weighted index barely budged. Sure, this could change. But, so far, it’s irrelevant to the relationship between QE and corporate financing.
Fraser then points to the collapse  in lending to non-financial companies, and says:

One explanation is that British companies are simply paying down debt. Another is that the government’s debt issuance is so extreme that it is crowding out other forms [of borrowing].

There is, though, a third explanation - that the credit crunch has made banks unable or unwilling to lend to companies. If we look at the monthly path of bank lending to non-financial firms, we see that this fell in the autumn, after the collapse of Lehmans, but has recovered a little since. If it was gilt issuance that was denying firms’ financing, you’d expect the opposite pattern.
Now, I say all this not to defend QE; I have big doubts about its effectiveness. However, to blame it for companies’ lack of finance is just silly. 

Doubts about democracy

Coming from different perspectives, Will Wilkinson and Henry at Crooked Timber make a similar point. Henry says:

Human beings may not want the kinds of autonomy that liberals presume they do

Will says:

The public does not want libertarianism. Which means that the public does not want a system that respects fundamental rights.

There’s a conflict between liberty and democracy.
The problem here, though, is that the public’s preferences might be systematically irrational; contrary to the optimism of wisdom of crowds thinking, their errors might not cancel out.
There are at least three different reasons for this:
1. We adapt our preferences to our circumstances; part of our psychological mechanism for coping with hard times is to reduce our expectations. The upshot of this, as Amartya Sen and John Roemer have pointed out, is that democracy - in the sense of giving the public what they want - gives too little to those who, in Henry’s words, “know their place.” The broken-spirited poor get too little whilst the rich, with their over-inflated sense of entitlement, get too much.
2. A basic cognitive bias - the availability heuristic - can lead to us under-estimating the value of freedom. This is because the gain from any restriction of freedom always seems clear, whilst the losses are more obscure. Hayek expressed this thus:

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. (Law Legislation and Liberty, Vol I, p56-57)

3. Other cognitive biases depress demand for radical change, be in it a leftist or libertarian direction; the status quo bias, endowment effect and loss aversion all conspire to lead people to prefer the devil they know.
This raises the question: why should leftists and/or libertarians respect democratic outcomes when these are founded upon preferences that might be irrational? As Will Kymlicka said:

[The aggregative or ’vote-centric’ conception of democracy] provides no opportunity for citizens to distinguish claims based on self interest, prejudice or fleeting whims from those grounded in principles of justice or fundamental needs…the outcome of the aggregative model has only the thinnest veneer of legitimacy. (Contemporary Political Philosophy, p290)

I’m surprised this issue isn’t more prominent. The growing interest in cognitive biases within centrist politics - be it Cameron’s embracing of “Nudge” or the “libertarian paternalism” of Julian Le Grand also suggest a scepticism about conventional democracy. After all, if people are poor judges of their best interests in health or pensions, why should they be good judges of their interest at the ballot box?
Now, if you’re with me so far, the question arises: how to respond to this tension between democracy and substantive outcomes?
One possibility is the retreat into Guido-style cynicism. Another is a vanguardist Leninist reaction - which I fear is a version of New Labour's position. But there is a third - to recognise that what’s valuable about democracy is not the aggregation of preferences - Kymlicka’s “vote-centrism” - but rather the opportunities we have to influence each other’s preferences. It’s the processes of deliberation that matter, not the cross in the ballot box.

Inflation targeting: some questions

Martin Wolf thinks this crisis represents a “disaster” for monetary policy, and that central banks in future should do more than just target inflation. I’m not sure. I’ve got six questions here.
1. Is it really necessary for a central bank to target asset prices? The argument for doing so is that asset price collapses have huge macroeconomic effects, and it’s easier for a central bank to stop a bubble before it’s grown too big than it is to clean up the mess afterwards.
But not all bursts do have calamitous effects. The bursting of the tech bubble in 2000 led only to the mildest recession in history. And looking at UK retail sales in the last 18 months, you’d never guess that house prices have fallen sharply.
Obviously, the bursting of the US housing bubble did have disastrous effects. But was this a failure of macroeconomic policy, or rather a failure of industrial organization - of a banking system that was mismanaged?
2. Is asset price targeting feasible? If people expect house prices to rise 20% a year, a percentage point on mortgage rates here or there will do little to stop their speculating. The interest rate rises necessary to stop a bubble might be so large as to be politically difficult. Can you imagine how the CBI or unions would have complained if the Bank had targeted house prices in the mid-00s; “you’re crippling manufacturers just to stop some southern speculators.”
If central banks are to target asset prices, they probably need some tool besides interest rates.
3. How are things complicated in a small open economy? Martin cites a point made by John Taylor:

The autonomy of the Bank of England was much smaller than most imagined: the wider the interest rate gap vis-a-vis the US, the more “hot money” flowed in.

In one sense, though, this makes inflation targeting easier: as hot money comes in, sterling rises which helps reduce inflation. Such speculative flows are more of a problem for a Bank trying to control asset prices than inflation. Which raises a question: could it be that a small open economy such as the UK can target inflation much more effectively than it can asset prices? Might it be that asset prices are best targeted - if at all - at a global level?
4. How should policy-makers respond to risks? Martin blames the Fed for keeping rates too low in 2002-04, thus stoking up a credit bubble. But at the time, there was thought to be a risk of deflation, which the Fed was averting*.
In hindsight, it looks as if the Fed was wrong. But trying to avert the small risk of big trouble made sense at the time, in expected utility terms.
There is sometimes a tension between targeting most-likely outcomes - the central part of the Bank’s fan charts - and targeting low probability/nasty pay-off outcomes - the extremes of those charts.
5. Are the recent policy failures a failure of the monetary policy framework, or merely of its application? In retrospect, we know that policy was too loose simply because inflation has been above target. But this, in itself, doesn’t invalidate the theory of inflation targeting. It merely shows how hard it is to get right.
6. How much can monetary policy do? The fact is that the future is unknowable so inflation targeting will often go awry. Policy errors - setting rates too high or too low - are not a departure from an ideal, but rather an unavoidable fact (at least under a fiat money system?) We must have low expectations for what policy-makers can do.
In this context, my view of inflation targeting is the opposite of Martin’s. He says:

Most of us – I was one – thought we had at last found the holy grail. Now we know it was a mirage.

However, I was always sceptical of inflation targeting; I thought it owed much of its apparent success to the fact that the economy was unusually stable, and wondered how it would cope with an adverse supply shock.  Recent experience has if anything diminished my doubts: inflation targeting can cope with adverse demand shocks of the sort we’ve had.
In my list of economic policy priorities, changing the monetary policy framework would come a long way down.
* Though remember, the Fed was not targeting inflation then.