« Open mouth; insert foot II | Main | Putting the hype in "hyperbole" »

June 18, 2004


Tim Swift

This is not the only recent example of 'knee jerk' politics from your favourite party.

Their claim that 'it costs more to adminster pension credit than some people get' (see http://news.bbc.co.uk/1/hi/uk_politics/3817043.stm) only makes for a good headline if you fail to apply even a moment's thought to it.

Any means-tested benefit that is gradually withdrawn will generate a situation where some people qualify for very small amounts of income. And obviously, if you average the costs over the number of claimants, that means for some people the average cost of processing a claim will be more than the actual benefit they receive.

But what is the alternative? Are they suggesting that you should have a cut off, and that if people qualify for under say £5 a week, they don't get it?

It's all very well to argue that the basic state pension should be increased (I tend to agree). But there are still a huge number of people (mainly women) who do not receive the basic state pension; so there will still need to be a safety net scheme. Which will suffer from precisely the same 'problem' that they have highlighted with the pension credit.

Cheap and easy headline, but absolutely meaningless in terms of sensible policy making.

Peter Cuthbertson

Even without your apt quotation from Alison Wolf, I still find Peter Black's assertion baffling. What "shortage of graduates" is he talking about? Already, 20% of people with degrees do jobs that don't require them. Is there any objective basis for such a claim at all?


Peter Black is talking rubbish, but so is Alison Woolf. Have you read her trashy book? Load of right wing garbage.

Of course, Universities in Wales are now about to be crippled because the Assembly has declined to introduce varaible fees. Apparently it is more left wing to not make the rich pay more to subsidise the poor than vice versa.

Oliver Kamm

I have indeed read Alison Wolf's (not Woolf's) book, which is distinguished by its stress on careful empirical research as opposed to (see above) the usual currency of politicians' comments on this subject. Empirical research is neither right-wing nor left-wing: I am a left-winger, yet my high opinion of Professor Wolf's book is shared by the Socialist Workers' Party (see Socialist Review article by Nick Grant, September 2002), which, as I have documented, believes all manner of reactionary things.

John Thacker

Clearly it is to some extent in the interest of the state to subsidize education, and hence it does, even with tuition charges. However, considering that not everyone goes to college, and that going to college makes one wealthier than if one does not, it seems strange that the Left would insist that the poor subsidize the middle-class and well-off-- especially those who are well-off now and will be well-off in the future.

Chris Lightfoot

You need to be a little more careful. You cannot infer anything by attempting a correlation on two data points drawn from such a biased sample. If you consider otherwise similar polities -- not, for instance, the USA and Switzerland -- you are rather likely to discover a strong correlation between participation in higher education and income. For instance, among individual states of the USA, you will find that there is a highly statistically significant positive correlation having, as shown in this plot (data from the National Center for Education Statistics).

Now, it is quite correct to say that this does not indicate a causal link or indeed allow us to elicit the direction of causality should such a link exist, but the evidence which you quote certainly isn't enlightening on this point either. Wolf's book may be better, but the article you cite likewise contains little substantive information. Since I am not a citizen of Wales I'm in no position to comment on Peter Black's statements about Welsh education policy, but if you want to you might want to consider data from polities similar to Wales, as Switzerland and the United States are not.

(Your claim that "empirical research is neither right-wing nor left-wing" is a nice idealisation, but it is certainly true that empirical research can carry a bias. For instance, consider two hypothetical pieces of "econometric" research looking at means to reduce crime. One searches for correlations with, say, variables indicating gun ownership, the use of capital punishment, etc. etc.; another with funding for drug rehabilitation, education and so forth. Both are "empirical" by the standard commonly used; each carries a bias. I have not read Wolf's book, so I cannot defend or condemn it on this basis; but you have, so if it's that good you ought to be able to put up a better defence than you have!)

More broadly I'd say that your stuff about the war against Iraq and the "war on terror" is much more entertaining than these exercises where you quote two or three random bits of polemic out-of-context and try to use them to illustrate that (say) a Liberal Democrat, an environmentalist or a Stalinist has (in your view) made a fool of themselves. Since your trademark gag here is to claim to be "objective" you might want to try sampling from a wider field. For instance, it has been drawn to my attention that from time to time politicians of other persuasions say idiotic things; in fact, they do it quite often. I'm sure that your position on the left of politics will allow you to identify many ready targets for your invective in the US Republican party, the Home Office, the City of London, and so forth.

Oliver Kamm

I have read Professor Wolf's book, so I'm happy to correct your misapprehension of it. She argues not that there is a statistical relationship between certain variables (specifically, higher education and economic growth), but that no such relationship can be inferred from the data. I would honestly have thought that was clear from the passage I've quoted. On this, she is right. Not only can you not infer a correlation from so limited a sample, you can't infer causality from a correlation without both time-series' being joint-covariance stationary. (For that reason the hypothetical empirical studies you describe are biased principally because their methodology is flawed.) The value of Professor Wolf's criticism is in demonstrating that confident empirical assertions about a causal relationship between the number of graduates and economic growth rates are not justified. This contrasts with the approach of Mr Black, the Lib Dem spokesman on education in Wales, who takes a causal relationship between the number of graduates and economic growth as - in his word - 'given'. It's not 'given', of course: it's a speculative hypothesis for which he provides no substantiation, and which I suspect he has never questioned.

My objectivity is not a gag but a statement of fact. If you look back at, e.g., my series on Do-It-Yourself Economics, you'll find that I select targets from business (the Director-General of the CBI) as well as politics. I've also criticised Tony Blair, whom I admire as a politician, for his closeness to precisely that way of business thinking. It so happens that my research indicates that the most consistently disingenuous and fallacious argumentation in mainstream British politics comes from the Liberal Democrats.

Chris Lightfoot

Errm. You can't infer that causality is present only from statistical properties of (for instance) the time series in a regression study. You need actually to have a model with testable empirical predictions. This is somewhat beside the point, though, since none of the discussion of the area here is likely to reach that standard of detail.

Secondly, the quotation you give emphatically does not make it "clear" that a relationship of the type you describe cannot be drawn -- in particular, it quotes almost no data and the results of no studies. No doubt, as you say, her book is very good and does this properly; but that's not a conclusion which could reasonably be drawn from the passage you quote. Perhaps I was supposed to infer from Wolf's tone and language that, oh well, if this is how she writes in "On Line Opinion" magazine, well then her book must be good.

(As I say, I'm not in a position to comment on Welsh education policy, but perhaps Black is referring to the rhetoric of the DTI and its rhetoric that universities and graduates are vital for "economic success", which presumably means growth. It is interesting that the government is intending to widen access to universities -- with a target of, what, 50% participation among young people? -- while simultaneously increasing the financial barriers to doing so and therefore narrowing the population from whom that 50% can be drawn.)

"It so happens that my research indicates that the most consistently disingenuous and fallacious argumentation in mainstream British politics comes from the Liberal Democrats."

Got any data for that, or is all your evidence anecdotal?


The idea of Alison Wolf as some kind of Right-wing shill is a bit of a stretch. Her profile page throws up publications for the in (avowedly Left-of-centre) New Economy and in a recent Social Market Foundation collection; she works at the Institute of Education - hardly a seething hot-bed of Rightist reaction. Her conclusions in Does Education Matter? are focused on equity considerations - using education to drive social inclusion is not effective, and we would be better to subsidise low-paid jobs directly if we want to improve equality of opportunity (p.252). She goes on to say that making education an economic policy instrument narrows our view of education (pp.254-256).

I'm pretty much persuaded by Wolf's presentation in the book, but I think Chris touches on a relevant point. The Government's rhetoric is that increased quantities of graduates are needed for continued economic growth, and yet they're making attending university less attractive. Wolf's book is clear that the evidence does not support the idea that increased HE participation will drive economic growth - and that it can become counterproductive, as a Degree's signalling properties are blunted through oversupply.

The weird thing is that the Government seem to be pursuing the wrong objective (raising participation) and the Tories the right objective (raising quality) yet they've then each got the wrong policy to match. The Government want to increase participation by charging people to go; and the Tories want to raise quality but are against fees - which will increase funding, increase quality-raising competition and encourage students to make rational choices over their participation. The Lib Dems are at least (rarely, and probably not intentionally) intellectually consistent, even if they're working on the basis of factually incorrect 'givens'.


"My objectivity is not a gag but a statement of fact."

So reassuring to know you're infallible.


Objectivity is infallibility?


Probably not ... but Mr Kamm claims that at all times he is "objective". This strikes me as almost meaningless. How can anyone ever prove or disprove this proposition? To me, this seems to imply that none of Mr Kamm's arguments can ever be criticised because, according to himself, only Mr Kamm has some kind of purchase on objective truth, which thus negates any possibility of error on his part. It's just one of the most silly, self-serving sentences I've ever read.

Oliver Kamm

Yes, I know what's required to infer causality. Professor Wolf is not trying to infer causality; Peter Black is. He does so with no evidence and without apparently even having questioned his proposition, which he describes as a 'given'.

FredF contradicts himself and carries on regardless. Objectivity is indeed not infallibility; there is no difficulty in principle with refuting claims that are testable and made by someone who is objective. I am an objective commentator because I am willing to criticise error regardless of its source. I refer again to my DIY Economics series - you'll find I criticise business leaders, Conservative politicians and Government ministers, as well as Liberal Democrats, whereas you will not find a similarly catholic approach in, for example, the web log of Peter Black.

As for the evidence of peculiar Lib Dem partiality to disingenuousness and fallacy, it's something that could indeed be framed in quantitative terms - the number of times, for example, that a political leader has been caught lying in the service of his party.

Chris Lightfoot

"Yes, I know what's required to infer causality."

Jolly good.

You allude to the dishonesty (alleged?) of a leader of the Liberal Democrats. Given your previous claim, I assume, therefore, that you have comparitive data for leaders of the other major parties. That would be very interesting to see.

I had a brief look at your "Do-it-yourself economics" series, but it hasn't really had enough episodes to determine whether your picking of targets is biased or not; however, I note that in addition to a brief (and as I pointed out at the time, slightly eccentric) foray against financial journalism, you have attacked the Liberal Democrats twice as often as the other targets you mention. It seems to me that I need not repeat the analysis for your "Those Liberal Democrat predictions" series; I searched in vain for the corresponding "Those Labour Party predictions", "Those Conservative Party predictions", "Those Green Party predictions", "Those RESPECT Coalition predictions", "Those US Republican Party predictions", etc. series. Perhaps those other political groups never make predictions, or never make inaccurate predictions? Again, if true, this would be an important finding, which I'm sure you would wish, objectively, to report.

Oliver Kamm

It's surprising, given the closeness of your analysis of my posts, that you overlooked the rationale of my 'Those Liberal Democrats predictions' series. In the first week of this blog's life, the Liberal Democrat leader in the House of Lords, Baroness Williams of Crosby, remarked on the party's perspicuity in predictions about the Iraq war and the unfairness of those warnings' having been rubbished. I commented at the time that rubbishing them was kinder than repeating them, because almost everything uttered by the party on the Iraq war has turned out to be wrong. I have continued with that series ever since, on matters broadly related to Iraq (including western economic prospects, which the party deemed to be poor on account of the war), not out of blind predjudice towards that party but because the party itself specifically invited judgement on the matter. As for the persistence of Liberal Democrat entries in the DIY Economics series, I would attribute that to the generally poor grasp of economics evidenced by the previous Liberal Democrat Treasury team, notably the former principal Treasury spokesman Matthew Taylor. Whatever political judgement you make of Gordon Brown, he genuinely has a deep intellectual interest and competence in economics. I have moreover balanced my criticisms of Liberal Democrat economic utterances with consistent praise for the intellect and judgement of Matthew Taylor's successor, Vincent Cable, for whom I have a good deal of respect.

Because I believe that the number of claims made by Messrs Blair and Howard that are comparable to Charles Kennedy's invocation of the Institute for Fiscal Studies is zero, the presentation of comparative data in this context amounts to proving a negative - something that I cannot do. I am, however, happy to be corrected if a counterexample to my hypothesis is presented.


Howard made a number of speeches in 2003 rejecting Gordon Brown's growth forecasts, e.g in March.

"In addition, the Shadow Chancellor referred to the claims of independent experts that Mr Brown has got his growth forecasts wrong...",


"He avoided answering any questions on his economic forecasts, which many independent commentators believe he has got wrong – for a second time."

As you have shown in criticizing the Liberal Democrats for similar speeches, the growth forecasts proved to be broadly right.

Chris Lightfoot

"Because I believe that the number of claims made by Messrs Blair and Howard that are comparable to Charles Kennedy's invocation of the Institute for Fiscal Studies is zero, the presentation of comparative data in this context amounts to proving a negative - something that I cannot do."

That is a remarkable hypothesis, and I think you should test it properly, perhaps by sampling from the statements of the various parties and their leaders. That would be the objective thing to do, after all.

Oliver Kamm

You think it's remarkable? I wouldn't regard it as such. The Liberal Democrat claims that I alighted upon certainly were remarkable, however: they involved a misrepresentation that was both patent and extreme. I have sampled statements from other party leaders. I have found much that is for various reasons exceptionable - on this blog, I have described one speech of Tony Blair's as disgraceful, even though I admire Blair in many respects - but I have found nothing that is comparable to the Liberal Democrats' standards of veracity. That is my objective conclusion. If you consider it partial or unfair, then I should be glad to consider your evidence; till then, I stick with my working hypothesis - the fruit of disinterested research and sober reflection - about the Liberal Democrats.


Oliver do you claim "objectivity" in your coverage of American politics as well?


Well, despite the censorship that's gone on here, I'd just like to say that I have never claimed that "objectivity equals infallibility", Mr Kamm's assertion that I've contradicted myself notwithstanding. The only point I've tried to make is that someone who claims that "My objectivity is ... a statement of fact" seems to have departed from any notion of ... er ... objectivity.

Oliver Kamm

For the record: this blog does not practise censorship, but it is a moderated discussion. A poster who has form of the type that characterises one of the leading technical economists of the past century as a "kook", without bothering to read any of his works, is not welcome.

Please do not continue this thread with a discussion of that policy, which is fixed and unalterable.

The comments to this entry are closed.