Monetarism and Monetary Economics A Delayed Comment

Frisch, Helmut Article Monetarism and Monetary Economics A Delayed Comment Kredit und Kapital Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Frisch, Helmut (1977) : Monetarism and Monetary Economics A Delayed Comment, Kredit und Kapital, ISSN 0023-4591, Duncker & Humblot, Berlin, Vol. 10, Iss. 3, pp. 321-335, https://doi.org/10.3790/ccm.10.3.321 This Version is available at: https://hdl.handle.net/10419/292794 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Monetarism and Monetary Economics A Delayed Comment By Helmut Frisch, Vienna The strong reaction to Professor Mayer's paper "The Structure of Monetarism"1 indicates that he has touched upon a central theme of the contemporary theoretical discussion. Th. Mayer by using 12 propositions which are of varying importance has characterized present-day monetarism. In this note I concentrate on his first proposition: the predominance of the impact of monetary factors on nominal income (the Neo-quantity theory of money). My point in chap. I is that the Neoquantity theory is completely trivial as a theory of nominal income if a monetary impulse cannot be divided in a real effect (output and employment effect) and in a price effect. This is shown by appeal to the so-called accelerations-theorem, which is formulated by a merging of two Friedman models.2 The few empirical studies which exist make it, however, questionable, whether one can speak of a "dominance" of a monetary impulse on output and production. In chap. II a neglected aspect of the monetarist transmission process (proposition 2 in Th. Mayers list): the formation of inflationary expectations is considered. There I claim that the accelerations theorem is compatible with adaptive expectations, but not with the model of rational expectations. According to the latter a monetary impulse would only generate inflationary and no real effects. Newer empirical work conveys the impression that for the USA in the period after the II. World War, the acceleration or deceleration of the rate of monetary expansion has not been anticipated. Therefore the accelerations theorem seems to be more compatible with the empirical evidence than does the model of rational expectations. 1 Th. Mayer, The Structure of Monetarism, Kredit und Kapital, Vol.8 (1975) p. 191 - 215 and p. 293 - 313. 2 M. Friedman, A Theoretical Framework of Monetary Analysis, J. P. E. 78 (1970) p. 193 - 238. A Monetary Theory of Nominal Income, J. P. E. 79 (1971) p. 323 - 37. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 322 Helmut Frisch Further it is surprising that Th. Mayer neglects the "crowding-out" effect, which some authors (for example J. Stein3) consider as the main difference between Monetarists and Neo-Keynesians. The crowding-out effect claims that government spending not accompanied by monetary expansion, i. e. financed by taxes or borrowing from the public results in a crowding-out of private expenditure with little if any increase in total spending. The neglect of the "crowding-out" is an expression of optimism concerning the stabilization policy of the government, whereas stressing it emphasize the opposite. Chap. Ill discusses the "crowding-out'' effect as a noticeable difference between Monetarists and NeoKeynesians. The fundamental difference, however, between monetarism and monetary economics in general is to be found in the 'stability conjecture' according to which the private sector of the economy is inherently stable (Chap. IV). This postulate or as Th. Mayer often emphasizes "belief" belongs to the "presuppositions" (A. Leijonhufvud) of the monetarists and is always formulated as a contrast to the 'instability postulate' of Keynesian economics. After a discussion of a more operational concept of stability it is pointed out that the older monetarists such as K. Wicksel, G. Myrdal and F. A. Hayek used instead of the stability conjecture the concept of the 'cumulative' process which rests on the assumption that the monetary sector of the economy (in contrast to the real sector) is unstable, since a discrepancy between the real rate and the market rate of interest moves the system away from equilibrium by a sequence of expenditures and price changes. I. The Accelerations Theorem The recent reformulation of the quantity theory is the accelerations theorem. It implies that only an acceleration or a deceleration of the rate of money growth produces any real effects, i. e. employment and output effects, while a constant rate of growth of the quantity of money determines the rate of inflation. For example Laidler's formulation is: "The effects of a change in the rate of change of money supply are felt initially on the level of real income and the rate of inflation, but in the long run it is only the rate of inflation that is affected."4 3 J. L. Stein, Inside the Monetarist Blackbox, in J. L. Stein ed. "Monetarism", p. 183 - 232, Amsterdam (1976). 4 D. Laidler, An Elementary Monetarist Model of Simultaneous Fluctuations in Prices and Output, in H. Frisch ed., "Inflation in Small Countries", OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 Monetarism and Monetary Economics 323 In M. Friedman's (1970, 1971), D. Laidler's (1976) and K. Brunner's (1970)5 model the accelerations theorem appears with a further conjecture, namely that the impact of monetary acceleration (deceleration) has only a temporary effect on output and employment. Both propositions can be found in a particularly simple formulation in Friedman's model. Merging both his theoretical models (1970, 1971) the following theoretical sketch is obtained: (1) JZ = N* + <x(y -y*)+y (log X - log X*) (2) x = x* + (l -oc)(y-y*)-y (log X - logX*) (3) y = y*+ 1 (m - y*) List of Symbols: n ... rate of inflation x ... rate of growth of real income y ... rate of growth of nominal income X ... level of real income m ... rate of growth of money R ... rate of change of the velocity of money fi ... adjustment coefficient for inflationary expectations This system of three linear differential equations expresses the acceleration theorem very clearly. If in (3) the exogenous rate of growth of the quantity of money increases compared to the expected rate of growth of nominal income (m > y*), a positive difference (y — y*) arises. Equations (1) and (2) show how that deviation of the actual rate from the expected affects the rate of inflation n and the rate of growth of real income x. The parameters a and (1 — a) can be interpreted as price elasticities and production elasticities. The system demonstrates a causal direction. An increase in the growth rate of money supply produces real effects via equations (1) and (2), the magnitude of which is set by the ratio a/1 — a. An expectations adjustment process of the type: — {y*) = fi (y — y*) dt increases y* in the state of disequilibrium until y = y* = m. In the new Lecture Notes in Economics and Mathematical Systems, Springer-Verlag, Berlin-Heidelberg-New York (1976) p. 76. 5 K. Brunner, The Monetarist Revolution in Monetary Theory, Weltwirtschaftliches Archiv, 105 (1970). 21 Kredit und Kapital 3/1977 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 324 Helmut Frisch steady state the effect of an increase of the growth rate of the quantity of money on the real variables have disappeared and all actual rates of growth are equal to the anticipated. If it is intended to produce real effects again the rate of money growth m has to be raised again. A permanent effect on the real system can only be made possible by a permanent acceleration of the growth of the quantity of money. The accelerations theorem agrees with the first Mayer-proposition concerning the "predominance of the impact of monetary factors on nominal income" and it contains some elements of the monetarist transmission process. Two questions arise immediately: (1) The accelerations theorem is an empirical hypothesis and thereby examinable. The only study to my knowledge which has attempted to subject the accelerations theorem to a direct statistical test is from P. Korteweg (1976) and is based on the Dutch economy from 1955 to 1972. In this study the "monetary impulse hypothesis" competes with the "fiscal impulse hypothesis" and the "foreign impulse hypothesis". P. Korteweg concludes: "Not rejected are the weak foreign and monetary impulse hypothesis. That is: changes in output growth without foreign and monetary impulses are highly unlikely".6 The empirical results do not oppose the accelerations theorem; however they oppose its interpretation in a causal sense, in that an acceleration of the money supply always generates a change in real production.7 A change in the real rate of growth is always correlated with changes in the rate of growth of money supply; but not every change in the rate of money expansion induces a change in the real rate of growth. (2) The duration of the resd effects depends on the speed of adjustment of expectations and thereby leads to the question, which concept of expectation formation is compatible with the monetarist theory. 6 P. Korteweg, Inflation, Economic Activity and the Operation of Fiscal, Foreign and Monetary Impulses in the Netherlands — A Preliminary Analysis 1953 - 1973, De Economist (1975) p. 616. 7 According to the weak version of the impulse hypothesis, the phenomenon will not occur without the impulse; according to its strong version, the phenomenon will occur anytime the impulse specified occurs. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 Monetarism and Monetary Economics 325 II. Endogenous Expectations The central question concerning the process of expectation formation is not analysed in Th. Mayer's "Structures", although it is precisely the expectations which play a fundamental role in understanding the accelerations theorem, as well as the monetarists transmission process. The acceleration theorem is compatible with the adaptive-expectations model, which is explicitly or implicitly accepted by the majority of monetarists authors (M. Friedman 1970, 1971), (K. Brunner 1970), (D. Laidler 1976). If the economic agents behave according to the model of adaptive expectations, the real variables of the system (output and employment) can be changed through a change in the rate of change of money supply, because of the appearance of an unanticipated inflation. Let us consider the situation in the labor market, which is usually neglected by the monetarists (E. Claassen8). If we start with a steady state situation in which money wages {w) grow at the rate of increase of marginal productivity of labor g and the expected rate of inflation, 7i* we have: W = g + JZ* The rate of growth of the market real wage, however, is w — n, where n is the actual rate of inflation: w — n = g + (rt* — n) . If a non-anticipated inflation develops due to the acceleration of money supply, the market real wage drops below the marginal productivity of labor and (assuming profit maximizing firms) the rate of unemployment is lowered below its "natural" level u* (corresponding to the state of affairs in which inflation is fully anticipated n = JT*). This situation could be demonstrated by the following (linearized) adjustment process: The increase in employment is proportional to the rentability difference g — (w — ri). 8 E. Claassen, Short-Period Fluctuations in Nominal and Real Income: A Monetarist Model, in E. Claassen and P. Salin ed. "Stabilization Policies in Interdependent Economies", North-Holland (1972). 21* OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 326 Helmut Frisch However the adaptive expectations adjustment model implies that money wages rise as long as there is a positive difference (jt — n*). When (7i — 7i*) = 0 the previous real wage is reached again, the rate of unemployment is equal to u* again, but the rate of inflation is permanently higher. Any further attempt to lower u* below u requires a higher rate of expansion of the money supply. An acceleration of the rate of growth of the money supply induces real effects only when a non-anticipated inflation results, which temporarily lowers the market real wage below the marginal product of labor. Therefore we have the following lemma: The accelerations theorem is explained by the existence of a nonanticipated inflation which is a consequence of the change in the rate of money supply. While authors as M. Friedman (1970, 1971), D. Laidler (1976) or J. Stein (1976) (in his "synthetic" model) have formulated an adaptive expectations process, another group of authors, monetarists "in a broad sense", — such as Th. Sargent9, Th. J. Sargent and N. Wallace10 and R. E .Lucas11 reject adaptive expectations as a waste of information, preferring rather "rational" expectations. The central idea of the rational expectations hypothesis (REH) is that the expectation of an economic variable "depends in a proper way on the same things that economic theory says actually determine that variable" (Th. J. Sargent and W. Wallace, 1975). More precisely: rational expectations of inflation are unbiased estimators of the actual inflation rate nt, given all information at the beginning of the period. From a theoretical point of view "rational expectations" render the accelerations theorem invalid. Since any economic agent knows the model, each change in the rate of growth of money supply leads not only to a change in the actual inflation rate but also in the expected rate of in9 Th. J. Sargent, Rational Expectations, The Real Rate of Interest, and the Natural Rate of Unemployment, Brookings Paperson Economic Activity, 2 (1973). 10 Th. J. Sargent and N. Wallace, Rational Expectations, the Optimal Monetary Instrument, and the Optimal Money Supply Rule, J. P. E. 83 (1975), p. 241 - 54. 11 R. E. Lucas, Jr., Econometric Testing of the Natural Rate Hypothesis, in O. Eckstein, ed. "The Econometrics of Price Determination", Washington, D. C. (1972) p. 50 - 59. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 Monetarism and Monetary Economics 327 flation, so that an impact on the real variables of the system is excluded. In contrast, the econometric application of the concept of rational expectations seems to support the accelerations theorem. Especially informative is an econometric attempt by R. J. Barro12 for the USA (1976). The work can be considered as an econometric test for the accelerations theorem and the concept of rational expectations. The hypothesis that forms the basis of this study is that only unanticipated changes in money have effects on real economic variables like the unemployment rate and the level of output. That hypothesis was quantified by interpreting the "systematic part" (= anticipated money growth) of the money supply as the amount that could have been predicted based on a reduced form equation, where money growth is explained by the "normal federal budget", the lagged rate of money growth and the lagged rate of unemployment. Unanticipated money growth was measured as actual growth less the amount obtained from this predicted relation (mt — mt*). The current and two annual lag values of unanticipated money growth were shown to have considerable explanatory value for unemployment and output according to the accelerations theorem. The results show that from 1961 - 1967 in a period of relatively constant growth the "unanticipated" rate of money expansion was very small; however the acceleration of the money supply 1968 (+ 2.5 °/o) was not anticipated and this brought the rate of unemployment down to 3.5 % below the estimated "natural level". An unanticipated monetary contraction 1960 (— 3.9 •%) accounted for a sharp rise in the unemployment rate to 6.7 %>. The empirical verification speaks for the accelerations theorem and against the REH. R. Barro's result shows that for the USA for the period 1960 - 1975 an acceleration (deceleration) of the rate of money supply was regularly not (!) anticipated and it thereby led to changes in employment and output. Why does the empirical evidence speak for the accelerations theorem and not for the REH? There are several reasons: (1) Existing price agreements and wage contracts make short-run changes difficult, so that for parts of the priceand wage system adaptive behaviour again appears realistic.13 12 R. J. Barro, Unanticipated Money Growth and Unemployment in the United States, A. E. R. 67 (1977). 13 See W. Poole, Rational Expectations in the Macro Model, Brookings Papers on Economic Activity 2, (1976) p. 484 f. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 328 Helmut Frisch (2) Economic agents might form conditional mathematical expectations using an economic model and information about exogenous variables in t, but different agents might have different models. People are rational with respect to their model; but the same information might convey different meanings to different economic agents. (3) The regular appearance of a significant unanticipated rate of money supply, whenever the rate of money supply changes, shows that economic agents know the pre-determined variables of the model but not all exogenous variables at time t, when they make their predictions. (For example, the fiscal policy variable in t — 1, but not the value of that variable in t). (4) The accelerations theorem can be viewed as a special variant of the hypotheses that only the unanticipated part of changes in the rate of money expansion has effects on the real economic variables. III. The Crowding-out Effect It is interesting to note that Th. Mayer did not discuss the crowdingout effect as a point differentiating monetarism from Keynesian economics. The crowding-out effect deals with the different ways of financing a budget deficit. "Whether deficits produce inflation depends on how they are financed. If, as so often happens, they are financed by creating money, they unquestionably do produce inflationary pressure. If they are financed by borrowing from the public, at whatever interest rates are necessary, they may still exert some minor inflationary pressure. However, their major effect will be to make interest rates higher than they would otherwise be." (M. Friedman, 1972)14 The crowding-out effect emphasized by M. Friedman and the econometricians of the St. Louis15 model stresses the fact that government spending not accompanied by monetary expansion, that is financed by taxes or borrowing form the public results in a crowding-out of private expenditure with little if any net increase in total spending. 14 M. Friedman, Comments on the Critics, in R. J. Gordon, ed., "Milton Friedman's Monetary Framework", The University of Chicago Press (1974), p. 140. 15 L. C. Andersen and K. M. Carlsont A Monetarist Model for Economic Stabilization, Federal Reserve Bank, St. Louis Rev. (1970). OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04 Monetarism and Monetary Economics 335 "stability conjecture". This conjecture is not an operational concept and belongs to the "Weltanschauung" of the monetarist school of thought. After a discussion of a more operational concept of stability it is pointed out that older monetarists such as Wicksell and Hayek used instead of the stability conjecture the concept of the "cumulative" process, which implies that the monetary sector of the economy is inherently unstable. Résumé Monétarisme et théorie économique monétaire Dans son essai «The structure of monétarism», le Professeur Mayer a caractérisé l'actuel «monétarisme» par 12 propositions. La présente étude s'intéresse principalement à la nouvelle formulation de la théorie de la quantité (Première proposition dans l'essai de Mayer) ainsi qu'au « postulat de la stabilité » des monétaristes. La théorie néo-quantitative s'exprime dans le théorème de l'accélération, selon lequel l'activation de l'expansion de la masse monétaire engendre des effets réels, alors qu'en modèle statique (steady state), le taux d'expansion de cette masse n'influence que le taux d'inflation. L'auteur démontre que le théorème de l'accélération implique des expectatives adaptables; des expectatives rationelles écarteraient anticipativement les effets réels d'une modification de l'offre de monnaie.Aux EtatsUnis, des recherches empiriques ont établi qu'après la seconde guerre mondiale l'accélération ou la décélération de la masse monétaire n'était régulièrement pas anticipée, de sorte que le théoreme de l'accélération semble plus compatible avec l'évidence empirique qu'avec des expectatives rationnelles. La différence fondamentale entre le monétarisme et l'économie monétaire consiste en l'acceptation ou le rejet du «postulat de la stabilité» qui veut que l'économie de marché soit stable en elle-même. L'on définit le monétarisme comme étant la théorie économique monétaire complétée du «postulat de la stabilité ». Ce concept n'a aucun caractère opérationnel, mais participe plutôt de la philosophie de l'orientation monétariste. D'anciens « monétaristes » tels K. Wicksell et F. A. Hayek exploitent non pas le postulat de la stabilité mais le concept du processus cumulatif qui implique l'instabilité inhérente du domaine monétaire. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.321 | Generated on 2023-01-16 12:42:04