Market and PPP measures of GDP

Ian and the for­mer head of the OECD sta­tis­ti­cal divi­sion, David Hen­der­son, were the authors of an influ­en­tial crit­i­cism of the IPCC’s pro­jec­tions for glob­al growth in 2004. They showed that the IPCC emis­sions sce­nar­ios, which used on MER esti­mates of the growth ‘gap’ between devel­oped and devel­op­ing coun­tries, implied growth in glob­al out­put (and hence, CO2emis­sions) that was far too large. The UK Par­lia­men­tary Com­mit­tee to whom they pre­sent­ed their case accept­ed their crit­i­cism of the IPCC sce­nar­ios. The IPCC was, pre­dictably, less respon­sive to their con­struc­tive crit­i­cism.

You may be won­der­ing why, if the MER basis is an inac­cu­rate way to mea­sure and com­pare out­put and growth, the World Bank (espe­cial­ly, but also the IMF and OECD in some pub­li­ca­tions) use it. The answer, as explained by the IMF staff, is that it is “too hard” (!) to esti­mate accu­rate PPP weights for the prices of goods and ser­vices, espe­cial­ly if they are not trad­ed. This makes it seem that some of the inter­na­tion­al finan­cial insti­tu­tions would rather be wrong than over-bud­get.

Update: In fact, it is very dif­fi­cult for some not-so-obscure rea­sons teased out by Bri­an Fish­er and Hom Pant in this ABARE Con­fer­ence Paper on the PPP ver­sus MER debate. Here’s the abstract: it’s a short paper and worth read­ing.

Is PPPE a bet­ter mea­sure than MER to com­pare real incomes across nations? Would emis­sions pro­jec­tions nec­es­sar­i­ly depend on which mea­sure of exchange rates we use to com­pare real incomes? This paper seeks to answer these ques­tions. It exam­ines the con­cep­tu­al and empir­i­cal basis of the PPP the­o­ry and con­cludes that PPPE is based on a set of very restric­tive assump­tions which (except one) are relaxed in MER. In fact, MER is a gen­er­al­i­sa­tion of PPPE. One can make a con­sis­tent com­par­i­son of real incomes by using the real exchange rate, which is a ratio of MER to PPPE, not by using just the PPPE or MER. MER, how­ev­er, can be used to con­vert nom­i­nal cur­rent incomes from dif­fer­ent cur­ren­cy units into units of a ref­er­ence cur­ren­cy and the prices in ref­er­ence cur­ren­cy units can be used to deflate the nom­i­nal incomes into con­stant price incomes to make valid real income com­par­isons across nations. Final­ly, it is argued that com­par­i­son of real incomes is not always nec­es­sary in mak­ing emis­sions pro­jec­tions. ”

One Comment

  • Peter,

    In my opin­ion, Hom Pant and Bri­an Fish­er are wrong to believe (if they still do) that

Leave a Reply

Your email is never shared.Required fields are marked *