The Stern Review on MER/PPP

One of the issues that’s been debated at length here is the choice of exchange rates to use in converting different currencies for projections of future economic growth and energy demand. The scenarios developed by the IPCC have used market exchange rates (MER) Ian Castles has argued, in very strong terms, that it’s crucial to use exchange rates adjusted so as to exhibit Purchasing Power Parity (PPP). In my submission, I made a couple of points. First, that there is no uniquely satisfactory method of obtaining PPP exchange rates. Second, and more importantly, the choice doesn’t make much difference to projections of energy use or CO2 emissions, as long as the same values are used consistently. A method like MER, which tends to overstate income differences between poor and rich countries relative to PPP will yield a lower income elasticity of demand for energy. And since MER data have been, until recently more readily available for more countries, there are some practical arguments in favour of using them.

That said, there are a couple of reasons to favour a move to PPP-based scenarios. First, since these are now becoming the norm, continued use of MER numbers is likely to cause confusion. Second, while the crucial numbers regarding emissions aren’t much affected (and any error may be either up or down) other variables, particularly those used in calculations of economic welfare, might be significantly affected.

In this context, it’s unfortunate that the debate has been seized upon by denialists as a basis for attacking the whole IPCC process. The energy that’s gone into pointless disputes could have better been used in a constructive attempt to improve things.

Where does the Stern report come out on all this? Pretty much right in my view. Key quote

efforts are under way to improve the provision of PPP data. The International Comparison Programme (ICP), launched by the World Bank when Nicholas Stern was Chief Economist, is the world’s largest statistical initiative, involving 107 countries and collaboration with the OECD, Eurostat and National Statistical Offices. It produces internationally comparable price levels, economic aggregates in real terms, and Purchasing Power Parity (PPP) estimates that inform users about the relative sizes of markets, the size and structure of economies, and the relative purchasing power of currencies.

In the IPCC SRES scenarios that use MER conversions, it is not clear that the use of MERs biases upwards the projected rates of emissions growth, as the SRES calibration of the past relationship between emissions per head and GDP per head also used GDPs converted at MERs as the metric for economic activity (Holtsmark and Alfsen (2003)). Hence the scenarios are based on a lower estimate of the elasticity of emissions growth per head with respect to (the incorrectly measured) GDP growth per head. As Nakicenovic et al (2003) have argued, the use of MERs in many of the IPCC SRES scenarios is unlikely to have distorted the emissions trajectories much.

I should point out that the World Bank ICP is a successor to the earlier ICP work of Heston and Summers who initiated the idea of systematic PPP comparisons and produced the well-known Penn World Tables. Still, as the quote makes clear, Stern can speak with authority on this topic.

34 thoughts on “The Stern Review on MER/PPP

  1. John, I’ve got no idea where you got the notion that statisticians “tend to prefer axiomatic approaches rather than those based on welfare theory”, but your statement that I suggested this is simply wrong. This is from my first major paper with David Henderson:

    “… our arguments for PPP-based comparisons and aggregates are in line with those of other professionals who are working on issues and problems relating to the process of economic growth. Thus the editors of a recent volume entitled ‘International and Interarea Comparisons of Income, Output and Prices’ note that the UN International Comparisons Programme (ICP), which is the prime source of PPP estimates, has now been running for over 30 years; that ‘measures derived from it are increasingly used in economic research in place of the distorted values derived from exchange rates’; and that ‘Almost the entire recent literature on the determinants of economic growth that covers large numbers of countries is dependent on these [PPP-based] data” (Ian Castles and David Henderson, “Economics, Emissions Scenarios and the IPCC”, “Energy & Environment”, vol. 14, nos. 2 & 3, p. 419).

    The volume that we cited, which was published in 1998, includes many illuminating papers, but I’ll mention just two:

    * “The World Distribution of Well-Being Dissected�, by two professors of economics and the University of Pennsylvania, Robert Summers and Alan Heston. The first sub-heading of the paper is “What Should Enter an Empirical Social Welfare Function?�, so these two eminent economists apparently believe that the extensive empirical data that they present in the paper (all of which is based on the Penn World Tables) have some grounding in welfare theory.�

    * “Axiomatic and Economic Approaches to International Comparisonsâ€? by Erwin Diewert, Professor of Economics at the University of British Columbia. The citation for a recent prestigious award to Professor Diewert states that has “developed, and adapted for implementation, … theoretically improved measurement methods in direct support of statistical agencies in the United States, Canada, Australia, the United Kingdom, France, Germany, Sweden and New Zealand.â€? Why do you suppose that all of these national statistics agencies are seeking Erwin’s expertise, if they “tend to prefer axiomatic approaches rather than those based on welfare theoryâ€??

    You say that “It’s surprising that someone with Henderson’s impressive qualifications would get this wrong, but the economic theory of index numbers is a rather specialised field and not one in which he has worked as far as I know.” I really find it rather a stretch to believe that the Head of the Economics and Statistics Division of the OECD during a period of rapid development of the PPP programme is not reasonably well acquainted with the economic theory of index numbers. But in any case, before concluding that the statement that I cited is wrong, I think that I should let you know that Professor Henderson sent the paper in which the statement was made to Erwin Diewert for his comment. Erwin thought the paper was excellent, and said that he had no suggestions to make.

    Could I suggest, if you believe that David Henderson’s statement is wrong, that you set down your reasons for this view and submit your comment to the Editor of the journal in which it appeared, Dr. Sonja Boehmer-Christiansen? That is the normal academic practice and, if your paper is accepted, there would be yet another Australian or part-Australian contribution to the MER vs. PPP debate in the pages of “Energy & Environment”. In addition to several by Castles and/or Henderson, this journal has published papers by Warwick McKibbin of ANU, David Pearce and Alison Stegman (now at Macquarie); and Peter Dixon and Maureen Rimmer of Monash.

  2. Ernestine, I apologise for not responding to your interesting post. I will do so as soon as I can, but felt an obligation to reply without delay to John’s claim that David Henderson was in error.

  3. “Could I suggest, if you believe that David Henderson’s statement is wrong, that you set down your reasons for this view and submit your comment to the Editor of the journal in which it appeared, Dr. Sonja Boehmer-Christiansen? That is the normal academic practice”

    A brief response on this. Comments on passing remarks like the one you quoted are not normally considered for publication. The normal practice when a minor error is pointed out is for the author(s) to submit a correction. As I’ve point out, in the welfare-theoretic interpretation of index numbers, the PPP measure is the cost of buying a standard bundle. So, in the interests of correctness, you and David Henderson might want to write and clarify that you should have said “this is not the only interpretation” or something similar. But I’ll leave this up to you.

  4. Ernestine, Yes, I agree with the arguments in David Henderson’s article: I was the co-author of several of the papers in which most of them were initially advanced. And I certainly agree with David’s view that the UK Government’s response to the House of Lords Committee Report on The Economics of Climate Change reflects little credit on the responsible Department (the Department of the Environment, Food and Rural Affairs: Defra).

    Let me give an example. Following the publication of the Report, I wrote to Henry Derwent, the senior Defra officer who had given evidence to the Committee, to complain about what I considered to be the misrepresentation of the Castles and Henderson position in Defra’s Memorandum to the Committee. In that letter I reminded Mr. Derwent that David Henderson had said in his first letter to the IPCC Chairman, Dr. Pachauri (28 October 2002), that “Ian Castles has suggested that in the next IPCC Assessment national and international statistical agencies should be brought in and represented�, and that he (David) agreed with my view; and that in my letter of 17 December 2002 to Dr. Pachauri I had said that I believed that the National Statistician of the United Kingdom, Mr Len Cook, would be very willing to assist the IPCC in the AR4, and that Mr Cook had subsequently written to me to confirm that this was indeed the case. Here is an extract from the reply I subsequently received from Defra:

    “… You state that national statistical offices and the International Statistical Institute should be involved in the emissions scenario work. I would comment that if they produce any papers for peer review then their views should be considered by the IPCC. Similarly governments can elect individuals from the statistical sources you mention to serve on the IPCC Bureaux if their credentials justify it. The requirement for peer reviewed articles (and the credentials these provide) may be more conducive to academics contributing to the IPCC process but they also provide a quality control that is important in order to maintain credibility in the findings.â€?

    Len Cook was appointed UK National Statistician and Director of Office for National Statistics in 2000 following an open competition, and served in those offices until 2005. He had been Government Statistician of New Zealand from 1992 to 2000, and had been elected a Chartered Statistician of the Royal Statistical Society in 1973. He is currently serving as one of the three Vice Presidents of the International Statistical Institute, and is a Companion of the Royal Society of New Zealand.

    Defra submitted a detailed review of the Castles and Henderson critique to the House of Lords Committee, included an Appendix entitled “Empirical Studies relating to Henderson and Castle (sic) Critiqueâ€?. But they apparently couldn’t see the need to consult Castles or Henderson on the subject, or Cook or his colleagues at the Office of National Statistics. Given the information I’ve just provided about Len, and your agreement that “it is the statisticians that are the best people to exercise informed judgement on which method is least unsuitable for the purpose at hand”, I think that you should also agree that Defra should have cleared a government response on a technical statistical matter with the national statistics agency.

    John, if “the PPP measure IS the cost of buying a standard bundle of goods and servicesâ€?, the statement in your original post that “Heston and Summers … initiated the idea of systematic PPP comparisonsâ€? is even more in error than I originally thought. I’d wondered why you left out the systematic PPP comparisons under the ICP rubric by Kravis et al in the 1970s, and those under the OECD rubric by Gilbert and Kravis in the 1950s, and those done single-handedly by Colin Clark when he was Government Statistician of Queensland in the 1940s. But if systematic comparisons of the cost of buying a STANDARD BUNDLE of goods and services in different cities and countries qualify as PPP comparisons, you’d have to include studies by the League of Nations and the ILO in the inter-war period and studies by the UK Board of Trade before World War I.

  5. Ian, I should certainly have mentioned Kravis as well as his co-authors Heston and Summers. As to whether you would call the earlier efforts “systematic”, I don’t have a strong view. The purpose of the remark was not to give a detailed history of the concept but to correct any possible impression that the ICP had been started by the World Bank, a point you have stressed yourself.

    None of this affects the point that the claim quoted from your article with Henderson is incorrect. As I say, it’s a minor point and does not justify a comment, but given your concern with accuracy in these matters, a correction would be in order.

  6. The claim I quoted is not from my article with Henderson, John: it’s from a sole-authored paper by Henderson, and it’s right. I don’t understand how you can assert that “the PPP measure IS the cost of buying a standard bundle of goods and services”, and then question whether the inter-war studies by the Geneva-based organisations were “systematic”: these were the only studies, of all of those that I cited, that attempted to calculate the cost of buying a standard bundle of goods and services in different countries. The ICP doesn’t do this. As you didn’t comment on my earlier point that your argument could at best only relate to the private consumption component of PPPs derived from the expenditure side, let me quote again from the advice I received from Erwin Diewert:

    ‘… when we are doing international comparisons of GDP, it is obvious that consumer theory cannot be the “right” economic approach to use, since I+G+X-M are not directly determined by consumer preferences. The axiomatic or test approach, the stochastic approach or the symmetric basket approaches can be used in this GDP context but the domain of definition for the admissible transactions that go into the aggregate are of course much bigger than just the C transactions! For these alternative approaches to index numbers for GDP, see the IMF Producer Price Manual and the following chapters {three chapters by Diewert cited].’

  7. Ian,

    I assume you haven’t had time to reply to what you called my ‘interesting response’. For house-keeping purposes I call this my question #1.

    I then asked a question regarding the article (my question #2). I understand from your reply that you agree with the argument in this reference. I don’t agree because, as far as I am concerned, the argument in this article is an argument for procrastination on controlling CO2 emissions:

    “Rather than pursuing as a matter of urgency ambitious targets for curbing emissions, governments should take steps, the sooner the better, to ensure that they are more fully and more objectively informed and advised in matters relating to climate change. This requires action on two fronts: first, to improve the IPCC process by making it more professionally watertight; and second, to bring to an end the IPCC’s monopoly status by providing for other sources of information and ideas.”

    In the context of my question #2, you quote my comment “it is the statisticians that are the best people to exercise informed judgement on which method is least unsuitable for the purpose at handâ€?. But my comment is irrelevant in the context of my question #2 because my comment relates to the first reference you gave in one of your earlier posts. This reference consists of a document on the index number problems in statistics.

    Negative feedback effects of C02 emissions on the natural environment are, IMHO, not an index number problem in statistics, but an important problem in Economics.

  8. Ernestine, I confirm that I haven’t had time to answer your question (i). I apologise for having misunderstood your point about the relevance of statistics. If there was an implication in my comments that your later point was not in line with what you’d said earlier, I withdraw it. But I’d adhere to the position that the projection of future levels of emissions is not for scientists but for economists (including, of course, index number theorists) and demographers.

    I also confirm my agreement with David Henderson’s position as quoted in the article you quoted. Last week David took the chair at a talk given in London by Dr. Dieter Helm of New College Oxford in the Beesley Lectures series on problems of regulation. The procedure for these Lectures provides for a personal contribution by the chairman, to be made after the talk and before the discussion is thrown open. The text that formed the basis for David’s remarks on that occasion included the following:

    “Let me however mention that a group of us, comprising both scientists and economists, hope to publish before long an assessment [of the Stern Review] which will be as extensive as we can make it. What we have in mind is two linked review articles, one focusing on scientific and the other on economic aspects. Though authorship would be largely or wholly separate, the two articles are being prepared in conjunction: they will be cross-referenced and mutually supporting. These twin contributions are scheduled to appear in a coming issue of the journal World Economics, which has already carried, in its summer issue, some exchanges between Sir Nicholas and the nine economists who are members of the group.�

    I am one of the co-authors of the economic review article that is in course of preparation, and this paper must now make the priority claim on my time. I must therefore ask you to excuse me for not having the time to continue our discussion. I hope that the article will provide answers, or at least explanations, for my position on some of the points that have been raised on this thread and in earlier discussions on John’s blog.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s