Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. Professor Wright on Methodology-A Rejoinder Author(s): Nicholas Kaldor Source: The Economic Journal, Vol. 65, No. 257 (Mar., 1955), pp. 157-159 Published by: Blackwell Publishing for the Royal Economic Society Stable URL: http://www.jstor.org/stable/2227473 Accessed: 26/10/2010 16:43 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=black. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Royal Economic Society and Blackwell Publishing are collaborating with JSTOR to digitize, preserve and extend access to The Economic Journal. http://www.jstor.org
  2. 2. 1955] PROFESSOR WRIGHT ON METHODOLOGY-A REJOINDER 157 without harming the rest. Or it might be said that those who suffer from the removal of the social organization from the maximum of ophelimity could, if they were allowed to reach the maximum position, pay a sum such that everyone would be better off with the new or- ganization. It was thus that, in the past, the redemption of certain seigneurialrights was to the advantage of both serfsand landlords." The idea is introduced in a parenthetical, almost underhand way, with the result that it has escaped attention for many years. Pareto offers no solution to the problems of interpretationwhich immediately arise, nor did he incorporatethe idea into his general formulation. Whatever the reasons for this, they can hardly have been that he was unaware of the interest and importance of this novel criterionof economic policy. DAVID POLE Cardif. PROFESSOR WRIGHT ON METHODOLOGY-A REJOINDER IN the September issue of the ECONOMIC JOURNAL 1 ProfessorMcCord Wright takes me to task " for not distinguishing sufficientlyin his paper 2 between a possibleand a truestory. I submit, however, that one of the most fundamental principles of scientific business-cycleanalysis is that it is never enough merely to show ways in which a cyclical movement couldoccur- granted certain assumptions. The problem is to find out empiricallyand historicallywhich one of these possible storieshas empiricallyoccurredand is mostlikely to recur" (pp. 622-33. Italics in the original, except the second " empirically"). If ProfessorWright can "'find out empirically" which of a seriesof alternativehypotheses" has empiricallyoccurred" and thusselect the " true story " and not merely a " possiblestory" he is a better man than the restof us. No scientist (certainlynot in the naturalsciences)would claim to be able to provethe correctnessof an hypothesisby empiricalobservation. The most that empirical observation can do is to disprove a theory, by showing that it fails to account for the observedphenomena, or else showing that it is based on demonstrablyfalse assumptions. An hypothesis can be rejectedalso in favourof a simplerone, if the latter accountsfor the observed phenomena equally well. But in neither case could it be claimed that the hypothesisfinally chosen represents" thetruth " or " reality." The purposeof my cyclical model was to show that given elastic expecta- tions-i.e., a state of affairswhere " the facts of the existingsituation enterin a sensedisproportionatelyinto the formation of long term expectations" 3- a cyclical movementis bound to follow, at any rateif thereis net savingin the community when production is at its maximum level. I further intended to show that the same assumptionsmay,but not necessarilywill, generate a trend. The model was also intended to throw light on the factors which 1 Pp. 622-5. 2 ECONOMICJOURNAL, March 1954, PP. 53-71. 3 Keynes, GeneralTheory,p. 148.
  3. 3. 158 THE ECONOMIC JOURNAL [MARCH determine the length of the cycle and of its various phases. I hoped it was obvious that the examples chosen were of a purely expositorycharacterand that the necessaryassumptionswere stated in an extreme form in order to emphasisethe essentialfeatures. I can readily believe that it is alwayspossibleto introducesuch additional assumptionsinto any particular " model " as would destroy the "a priori necessity" of its conclusions. But if these assumptionsare merely chosen at random (without any regard to their relevance to actual situations) they do not prove anything for or against an hypothesis. Scientific hypotheses are invented in order to account for the phenomena actually observed, not in order to demonstrate that there " ain't no such animal." Of course I would agree with ProfessorWright (if this is what he really intended to say) that such simple analytical models should merely be regarded as a starting point; and they shouldserveas a guide to, and not a substitutefor, empirical research.' When ProfessorWright says that the model suffersfrom a "logical weak- ness" in that business men's expectations would not be elastic (in the above sense) in a purely static or stationaryworld, he shows a complete mis- understandingof the purposeof an analytical tool. A " static " model of the trade cycle is not intended to " explain " non-existingphenomenaof a purely imaginary world: the purposeof the static abstraction,here and elsewhere, is to enable us to isolate the relevant factors from the irrelevant ones in the worldas it exists. If it can be shown that it is the existence of elastic expecta- tions, and not economic growth as such, or technical innovation, etc., which is responsiblefor the generation of cyclical movements, it is perfectly legitimate from a methodological point of view to do this by assuming a particular state of expectationsin a static framework-without being called upon to say whether that assumption is " realistic " in that particular imaginarysituation. (Realism is only a meaningful attribute when applied to reality, not unreality.)2 The same kind of misunderstanding underlies also his criticism of my numerical examples. I would agree, of course,that under the strictassump- tions of the numerical example, it is only as a resultof a chance constellation of the various parameters that the cyclical movement takes the form of a " short cycle " of a simple pattern (meaning by a " short cycle " one where only a fraction, and not the whole, of the capital stock is renewed in each boom). But since, as I show later, more realistic assumptions concerning service-liferemove this particularproblem, I cannot see the inconsistencyin 1 My remark that it would not be worth while to introduce certain additional assumptions into the model " at the present state of knowledge " (which seems to have been misunderstood by Wright) referred to our knowledge of the facts of the real world, not to our knowledge of altemative trade-cycle models. It merely intended to say that it is not worth while to introduce " complica- tions " at random until we know which particular " complications " are relevant. 2 That I am myself very much aware of the connection between the elasticity of expectations and " dynamic " change is shown by the more extended treatment of the problem of expectations in my " Speculation and Economic Stability," Review of EconomicStudies, Vol. VII (October 1939), particularly pp. 9-12.
  4. 4. 1955] DEPRECIATION, REPLACEMENTAND REGULAR GROWTH 159 using-deliberately-the rightfigures to produce the required results in an expositoryexample.' ProfessorWright'sstricturesthusfailto reveal (tome at anyrate) the " in- ternalinconsistency" orthe "logical weakness"whichhe claims-which do-es not exclude, of course, the possibilitythat hisown model of the trade cycle is far superiorto mine, or anyone else's. I can assurehim that it was not my intention " to omit all rival accounts " on purpose, or to claim " exclusive validity" 2 for my theory; and if I had not taken account of his own work on the subject this possibly may only have been because I was not familiar with his books-a defect which I can see I must speedily remedy. NICHOLAS KALDOR King'sCollege, Cambridge. DEPRECIATION, REPLACEMENT AND REGULAR GROWTH THEeconomic professioncertainlywill be gratefulto ProfessorDomar for having drawn attention (in " Depreciation, Replacement and Growth," ECONOMIC JOURNAL, March 1953) to the consequences of the replacement lag, which were well known to practice but rather neglected in economic theory except in a famous passage in the GeneralTheory. As Domar remarkshimself,the subjectneed not be tied to modelssuchasthosepresented in his paper. Indeed, in the revision of the Harrod-Domar model of a regularly progressiveeconomy, in Sections III and IV of the paper, certain shortcomingsof Domar's approach become apparent; I referto the implicit assumptionthat, asin the oldermodel, the regularlyprogressiveeconomy, in which replacement lags behind depreciation, will necessarily follow an exponential time path. Formally, the problem is not attackedby Domar in its full generality; and from the economic angle a discussionof a most im- 1 I thought I had met ProfessorWright'sstricturesconcerning the initial build-up of the capital stock in the footnote to p. 59 of my paper, where I pointed out that it is only when over-lapping between building and scrapping periods is completely avoided that there will be a " short cycle " which, in the example chosen, meant booms and slumps alternating at five-yearly intervals (though I failed to make explicit that over-lapping could notbe avoided if the initial build-up of the stock was a con- tinuous process). Otherwise the assumption of a rigidly fixed and uniform service-life will lead to " long cycles " which imply, in terms of the same example, alternating booms and slumps of fifteen years each, when the service-life is twenty-five years; fifteen years of boom followed by twenty years of slump, when the service-life is thirty years and so on. The reason for selecting " the short cycle " of the example was that the fluctuations in the actual world are undoubtedly short cycles (in this particular sense) since only a fraction of the outstanding stock of capital is renewed during each boom, whilst the inevitable over-lapping of scrapping and building periods (as a result of the " echo ") which prevents cycles from being " short " is only an acute problem (as Wright agrees) under the artificial assumption of a fixed and uniform service-life. (I completely fail to follow Professor Wright's own arithmetic either in the case of my original example or in the case of his assumption of a thirty-year service-life-but since the correctness or otherwise of his arithmetic is without significance to the points at issue, I do not wish to pursue this further here.) 2 Italics in the original.