I do not know enough to assess the views of Paul Romer, the chief economist for the World Bank, when it comes to his specialism. I need no special knowledge to assess his views as reported in the Times on restoring the standing of his profession. He gets it.
Economists need to stop acting as if they own the moral high ground and start behaving with more humility if they are to win back the public’s trust after Brexit, according to the World Bank’s chief economist.
Paul Romer said that a popular backlash against experts needed to be taken seriously and that Brexit had been partly a reaction to the perceived hypocrisy of economists who claimed to be making unbiased judgments but were actually taking political positions.
Dr Romer, one of the leading economists of his generation, is known for speaking out against his profession. Last September he published a paper, The Trouble with Macroeconomics, in which he accused colleagues of practising a “pseudoscience” underpinned by an “honour code” that prohibits challenge to figures of authority even when their facts are wrong.
Dr Romer said: “To me, Brexit was a vote against the expert advice of economists. We have to earn back our credibility as professionals who will give an unbiased answer. In political discourse, activists often claim that their position is morally superior and no one seems to care, but when economists did so, voters reacted very negatively, perhaps because they are alert to even a whiff of hypocrisy and they sensed that economists were behaving like activists yet invoking the authority of science.
And if any smartarse wants to bring up Michael Gove’s remark about the British people having “had enough of experts”, tell them to listen to his actual words before he was shouted down. He wasn’t talking about any expert on any subject; he was referring specifically to those who said their predictions of Brexit disaster should be believed on grounds of their business and economic expertise, yet who had egregiously got their predictions wrong on the Euro and failed to predict the 2008 crisis at all.
These are “economists” who know nothing about economics – the European Union is an extra layer of government, endless regulations are what it is about. This is not good for economic life, it is bad for economic life.
We are suffering through a 3rd or 4th generation of economists who know nothing about economics. They were reared on the patent irrationalities and utter idiocies of Keynes and his disciples, and know nothing else. I can’t see how that’s ever going to change, as these people now control all the institutions in which young economists are bred. But until those Keynesian nostrums are returned to the dustbin of discarded economic theory to which classical economists had relegated them (which is precisely where Keynes found them) nothing is going to change. It will probably take a worldwide economic collapse to force a radical rethinking of what today passes for mainstream economic “wisdom”. It will not be pretty.
Romer may indeed “get it” when it comes to understanding why his profession is today so reviled. But I seriously doubt that he understands why mainstream economic thought has brought us to the current state of affairs. He and his ilk will continue applying the same leeches; he merely advocates displaying a little more humility while doing so. Personally, I’m underwhelmed.
Who was the prat interviewing Michael Gove ?
“To Hayek with Keynes!” or “Tell Keynesians to go to Hayek!” would make great slogans on T-shirts! Even Labour supporters would understand it.
Very witty, Nick U.J. ;>))!!
Thanks, Julie near Chicago. Can you think of a good slogan? (I also came up with, “No, I’m NOT Mr. Universe! Don’t feel too bad- everyone makes that mistake!”)
It is difficult to see how the reputation of the economics profession can be restored when it was never that high in the first place and so many of them believe Keynesian nonsense.
Lee More,
The interviewer was Faisal Islam, Political Editor of Sky News. Actually he is generally quite a bit better than he appears in that clip, IMHO.
The underlying problem is more serious, and closer, than even Mr Romer acknowledges. This is the problem: the data upon which economists (et al) rely is built using techniques based on assumptions which may have approximated to the truth 70-80 yrs ago, but are now smuggle in major misconceptions about how economies work. As Big Data gets a grip on what’s actually happening, those misconceptions will emerge. Indeed, they are beginning to emerge, and are quite startling. Example: China’s Alibaba produced two sets of inflation calculations based on their own massive databank of what people purchase. One was calculated according to the usual CPI techniques, in which various items are specified, weighted and deflated according to the usual hedonics techniques. This produced an inflation rate of 0.2% yoy. They then used the same data, but used the gross average price per item actually spent: ie, how much people were paying for a PC, or a decent phone, or whatever. This produced an inflation rate of 9.3%!
This is/was a very substantial difference, which Alibaba explained by saying that what the traditional CPI technique was capturing was overwhelmingly the price decay, which has accelerated sharply in response to changes in the way people shop for things, and consequently the way suppliers supply things. Product times, in every way, become shorter and more vicious. But merely tracking this, as CPI increasingly does, actually masks changes in the market dynamics between supplier and consumer, which is what CPI is meant to do.
The bigger point is this: so why, when you’ve got the data available, would you even bother to calculate a CPI which is no reflection of market characteristics.
Note that the point isn’t that CPI is wrong, but that the techniques by which it is calculated quietly (and unknown to almost everyone, including the economists and statisticians themselves) but devastatingly smuggle in completely erroneous assumptions about the structure of markets today.
Now multiply that example by however many you want. Maybe we could start with the assumptions underlying employment/unemployment numbers? Or maybe we could think about capital stock (hint: don’t go there, actually).
The wider point is that traditional economics, which has gradually collapsed into techniques of time-analysis, is genuinely in its death-throes – it just doesn’t know it yet. It is like going bankrupt: it happens very slowly, then very fast. The time is clearly coming.
I’m sure I’m not the first reader of Natalie’s post to think that the work ‘economists’ could be substituted with several supposedly hard-science alternatives and any soft science (or ‘science’).
Has his being an economist merely misled him into noticing more? Do all the same voters who despise the experts also equally despise the virtue-signallers? Is the latters’ asserted superior morality now as thoroughly seen through as the experts’ asserted superior understanding? Or is the quotee right? Is there an electorate that still credit the activists for caring while seeing through the experts? Indeed, are they even more sceptical of the activists than of the experts? Is science increasingly brought in to bolster the latters’ diminishing authority?
(My own answer FWIW is yes and no and maybe: the difference is shrinking – Romer’s remark is not precisely accurate – but the various kinds of lingering credulity mingle diversely through the public.
Economics and Meteorology
Every time some self-professed expert, who “learned” from another self-professed expert, decides they can manipulate the outcome for improved results, then things REALLY get buggered up, and somebody ELSE has to sweep up the mess with an old school broom.
But, you know…”It just hasn’t been done RIGHT yet! It’ll be different THIS time!”
I’ve recognized a pattern here.
The road to hell is paved with academic “Here, hold my
beercognac , Watch THIS….!”Romer is mostly famous for his graduate textbook on Macro economics. It was a while ago I went to university now, but at the time at least it was ‘the’ graduate textbook and I have one sitting on my bookshelf gathering dust. Unless he agreed to some spectacularly bad terms with his publisher he should have become a very rich man from this book.
His wife was an Obama appointee to the Council of Economic Advisors.
I have noticed that the media often interview ‘economists’ who work for large financial institutions, and I can’t help thinking that these people are not there to forecast events for the harder-nosed traders, or even to help the institution to make money, but are rather more like a Zampolit in a Soviet Army regiment, there to make the right noises and keep an eye on things, and to put out reassuring words when the PTB come nosing around.
The attitude that a lot of people took (and still take) to Brexit appears to me to have been flawed in two ways.
Firstly, it was and is assumed that economists can predict the economic impact of a unique event with absolute accuracy. I’m still pretty pessimistic about how Brexit will go economically, but the ‘immediate recession’ part of Project Fear is now demonstrably wrong. The claim during the referendum that leaving would make us poorer than we would otherwise have been always had the whiff of non-falsifiability about it in any case.
More than that, though, as Romer talks about, the arguments of the Economist / FT mob (for example that clot Tim Harford) suffer from the McNamara fallacy. It’s not possible to measure the value of being a sovereign country rather than being a German satrapy. In discounting the value of such things, economists are making an unacknowledged and unconsidered value judgement.
Possibly it’s because I have no background in statistics myself, but there are few things that annoy me quite so much as people pretending their irrationalities are either non-existent or are in fact rational. Give me a pundit who admits he’s guessing any day.
Ah, Dr. Romer is not related to Christina Romer, the chair of Obama’s Council of Economic Advisors whose public predictions of the efficacy of the stimulus plan were so far off that she had to fall on her sword and resign. Funny, though – he criticizes the “experts” while her unfortunate situation embodied the hubris of said experts.
I don’t disagree with Michael Taylor’s comment about the calculation of the CPI (other than to note, once again [he sighs resignedly], that CPI is merely a measure of certain price changes and thus is merely a proxy for inflation but is not actually a measure of inflation itself); people have been complaining for years that the government routinely manipulates the calculation for its own purposes by changing the components as well as doctoring the adjustment factors. It excludes the costs of food and fuel, for pete’s sake, two things which have a significant effect on the ordinary person’s purchasing power. But that is only one tiny part of the structural errors upon which the entire edifice of modern mainstream economics rests. Another is calculation of the unemployment rate, also noted by Michael, which is similarly manipulated by government.
But I maintain that the biggest problem is a fundamental error in the calculation (in the very definition) of Gross Domestic Product (GDP). The classic definition of GDP is the monetary value of all the finished goods and services produced within a country’s borders during a specific time period. That’s all well and good, but the actual calculation is “the sum of all private and public consumption, government outlays, investments, and exports minus imports.” A very large percentage of government outlays is simple transfer payments, which are mostly spent currently by the recipient. In other words, a significant portion of those government outlays ends up being double-counted in the calculation of GDP. If this double-counting were to be eliminated, calculated GDP would be quite a bit smaller than the reported figure. That in itself wouldn’t be a problem (assuming consistency in the calculation), but it also means that the rate of economic growth is likely much smaller than is generally reported; in fact, in some periods it could be negative. But it’s difficult to find the disaggregated numbers, or any calculation of “private” GDP (excluding government).
Macroeconomics isn’t even a thing as far as I’m concerned. Sure, it’s possible to add things together. (And make a career out of it.) But the only purpose this ultimately serves is to give some numerical credibility to the central planning fallacy.
Err, CPI excludes good and fuel? Not in the UK it doesn’t: https://www.ons.gov.uk/economy/inflationandpriceindices/articles/consumerpriceinflationbasketofgoodsandservices/2016
Food and non-booze drink accounts for 10.3% of the index and both household and transport fuels are included in the household and transport costs sections respectively.
If Economists actually knew anything they would all be running hedge funds
To James g. Whilst I’m in general agreement, I do think there are some things which are rescuable from national accounting (provided one takes the nominal number, not the technically impossible ‘real’ calculation). But not a great deal.
To Laird. I’m not entirely sure you’ve got the point. There is no mileage to be had in being pernickety about the details of how CPI is calculated – the point is that the evidence we now have from Big Data means a) the underlying assumptions upon which the idea is based are demonstrably wrong, and not just a bit, but a whole great lot; and b) there’s no longer a point in manufacturing a (massively wrong) proxy because we have now have access to the real thing, in near-real time.
The consequences for macro-economics as we know it are utterly revolutionary. The very notation which it manipulates is. . . wrong. In many cases, the variables it manipulates are simply fictive. A very radical collapse is underway.
Michael, I don’t (entirely) agree. There could be value in the CPI, however it is measured, as long as it is calculated consistently over a long period of time. That would at least provide a measure of the rate of change in the prices of certain items. Unfortunately, our government routinely tinkers with the components of CPI and their weightings, so valid comparison between time periods is impossible. (And the exclusion, in some widely-reported figures, of certain items which comprise a large portion of most people’s spending makes the number less relevant to the ordinary than the purveyors of those numbers would have you believe.) But the Big Data to which you refer isn’t inherently any better. It merely shows the change in prices of a different group of items. Yes, it shows that prices for those items have increased much faster than for those included in the CPI, but that’s all it does. You could argue that those particular items are more representative of normal spending practices and are thus more relevant to the ordinary person, and I might agree. But it’s not inherently more “correct” than the CPI; it’s just a different measure. It’s simply another illustration of the seven blind men trying to describe an elephant. I don’t really care which measure is used; probably there should be many. But I wish it would be made clear exactly what it is that the CPI (or the Alibaba calculation or any other index) actually measures and, even more importantly, what it does not. Because what it does not represent is precisely what it is routinely touted as being: the “rate of inflation”.
jim jones, that idea has actually been tried. Look up Long Term Capital Management, which was started and run by two Nobel Prize-winning economists. It crashed and burned in 1998, so spectacularly that it forced the US Treasury to intervene and (combined with the Russian bond default at about the same time) caused the largest economic meltdown in history until the 2007-08 financial collapse. If that doesn’t make you think twice about the quality of economics Nobel Prizes (I’m thinking Paul Krugman here), nothing will. Roger Lowenstein’s When Genius Failed, the story of that particular debacle, makes for good reading.
Dr. David Romer is the husband of Christina Romer and the noted macroeconomicist. He is from Massachusetts
Dr. Paul Romer is the fellow at the World Bank and a pioneer of endogenous growth theory. He is the son of former Colorado Governor Roy Romer.
In the reign of World-Emperor Kim, all economists will be required to wear a wizard’s hat in public.
(I used to say the same about psychologists, but I’ve recently changed my remedy for them to gallows trees.)
Laird,
I take your point, but the crucial thing is that if you can have an Alibaba index of the state of play between suppliers and consumers, then why would you actually want to construct a CPI in the first place? The answer is that ‘it is needed to satisfy the models of the economy upon which Economics is built.’ But that, honestly, is no answer at all. It seems very likely that what CPI (even without govt tinkering) is calculating is the number of angels on a pin-head. The point is not that it is badly calculated, so much as it is calculating something that doesn’t exist. That’s the radical thing that the Alibaba indexes are showing.
PS. It’s v interesting that just as people finally got interested in the lessons from the Alibaba indexes, they stopped publication! Make of that what you will.
Michael, I know nothing about the Alibaba index beyond what you wrote here. But your original post said nothing about “the state of play between supplier and consumers”; all you wrote was that this index measures “the gross average price per item actually spent”. From that description in inferred that it was just a different “basket” of goods than that used by the CPI. If that’s wrong, could you please elaborate?
It calculates two indexes.
First, it replicates CPI methodologies and weightings down to detailed sub-group levels, and uses its data to calculate. This is the one which put CPI at 0.2%.
Second, using exactly the same data, it takes the gross average prices actually spent on the product at a slightly less specified level. It was this which produced 9.3%.
The difference is this: if you ask ‘what is the price of a computer with 3G of RAM, a 14″ screen and a 3G HDD, the price you’ll get will be an amalagam of all the computers fitting that specification, almost all of which will be ‘old’ product, with prices increasingly discounted. And then, of course, they’ll adjust further for the impact of hedonics.
By contrast, Alibaba’s second index merely answers the question ‘what are people paying, on average, for a notebook computer, or a tablet computer, or ‘a decent phone.’ And when you answer that question, you find they are spending 9.3% more than last year for products which, whilst maybe not fitting the same technical description, do the same job to the same levels of satisfaction (accepting that ‘the same job’ may also have changed). Given that suppliers are also competing to figure out what that job is, so that they can charge a premium for satisfying it, the average price actually paid also reflects their success. Hence, you end up with a CPI of 9.3%, based on the same data. My point is that the 9.3% number gives you a whole lot more information than the CPI data gives you, and that, in fact, the CPI is not only wrong, but so structurally flawed as to be pointless.