The "New Era" of Controlling Monetary Aggregates

Duwendag, Dieter Article The "New Era" of Controlling Monetary Aggregates Kredit und Kapital Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Duwendag, Dieter (1977) : The "New Era" of Controlling Monetary Aggregates, Kredit und Kapital, ISSN 0023-4591, Duncker & Humblot, Berlin, Vol. 10, Iss. 3, pp. 391-402, https://doi.org/10.3790/ccm.10.3.391 This Version is available at: https://hdl.handle.net/10419/292799 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/ Berichte The "New Era" of Controlling Monetary Aggregates* Since about 1975 a marked shift towards controlling monetary aggregates has taken place in several countries. Among others, the central banks of Switzerland and Germany (at the beginning of 1975), of the United States (in March 1975), of Canada (in early 1976), and of the United Kingdom (in July 1976) adopted a policy of monetary growth targets. What distinguishes the new strategies of the central banks mentioned is the stringency in pursuing monetary growth targets. So far the Swiss National Bank (SNB), the Deutsche Bundesbank, and the Federal Reserve (Fed) have realized the strictest approach to controlling monetary aggregates. Therefore, this report concentrates on these central banks, roughly outlining the features which are common to their strategies and evaluating some special problems of their new approaches. I. Characteristic features of the "New Era" What's new about the "New Era"? Is it really a "new" era? No doubt, that one has to be very careful in using such a lofty qualification. Thus, it is perhaps less ambitious to simply speak of "experiments" in monetary policy as the central banks themselves have repeatedly emphasized. The essential characteristics of the "new era" or "experiment" can be briefly summarized in the following six points: 1. Most obvious, the central banks committed themselves to publicly announce and to pursue monetary growth targets. In comparison with the previous practice this strategy is a courageous step on a new ground, in particular if we consider the general reluctance of central banks in making experiments, in setting up and in publicly announcing quantitative targets. * Summary of papers prepared for 1) the "Committee on Monetary Theory and Monetary Policy" of the Gesellschaft für Wirtschaftsund Sozialwissenschaften (Verein für Socialpolitik), and 2) the 1977 "Konstanzer Seminar on Monetary Theory and Monetary Policy". The detailed and empirically supported version will be published in a volume of the „Schriften des Vereins für Socialpolitik", forthcoming 1978. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 392 Berichte 2. The adoption of monetary growth targets was accompanied with a transition to broader monetary aggregates. This notion covers a large spectrum of definitions of the money stock in the various countries, implying that a policy of monetary growth targets is not tied to a specific definition of the money stock. 3. The underlying rationale of the new policy orientation — explicitly focused by all central banks — is to reduce short-run cyclical instability in economic activity. Monetary growth targets are supposed to restore a sense of greater stability and confidence in economic performance. Thus, "stability" or "stabilization" is the crucial and almost magic word of the new strategy. The argument goes that the expectations and decisions of economic agents could be stabilized by giving them in advance a monetary framework (a sort of "monetary budgeting"). No doubt, that the bad experiences with stop-and-go practices, with time lags and the ensuing danger of procyclical effects of past monetary policy have facilitated the move towards monetary growth targets. 4. In computing the monetary growth target the longer-run perspective is explicitly aimed at gradually reducing the growth rate of the money stock to a level compatible with the longer-term growth rate of potential physical output. Being still a long way from that objective the computation of monetary growth targets has to take into account also some cyclical components, such as predicted changes in capacity utilization, in income velocity of money, and in the rate of inflation. 5. The tactical element of the new strategies lies in a certain shift of responsibility for inflationary price shocks. By setting a monetary framework in advance and by publicly announcing the decomposition of monetary growth targets — the so-called "unavoidable" rate of inflation on the one hand, and the expected growth rate of real GNP on the other — the monetary authorities seek to set some disciplinary data in order to affect the price and wage policies of both private parties and the Government. Whether or not such a disciplinary effect will be attained ultimately, is another question. In any case, however, the convenient excuse of an overly monetary alimentation of inflation falls away. Thus, the group of those who are responsible for inflationary price shocks is narrowed. 6. Last but not least, the "New Era" seems to be deeply imbedded in the monetarist doctrine, even though most central banks do not exOPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 Berichte 393 plicitly profess to monetarism. Actually, the shaping and practical implementation of monetary policy diverge more or less from the policy recommendations of the monetarists. Their basic ideas, however, recur time and again. So the "New Era" may be perhaps qualified as "experiments in practical monetarism", as Paul A. Volcker, President of the Federal Reserve Bank of New York, put it most recently. II. Elements of the strategic approaches 1. In evaluating the new monetary strategies three elements are to be analyzed: (1) The instruments, (2) the monetary indicators which are simultaneously control variables, (3) the intermediate targets. Control variables bundle up the numerous effects emitted by the central bank's monetary impulses in only one or two magnitudes. Because of this property control variables are also best suited to serve as monetary indicators. Changes in control variables are supposed to bring or to keep the intermediate target on track. A problem of monetary semantics is that the central banks' "official" terminology normally does not speak about "intermediate targets", "control variables" and "monetary indicators". Thus, for analytical purposes it is necessary to interpret as objectively as possible the published reports of the Fed, the SNB and the Bundesbank in terms of the elements mentioned. These elements are summarized in the following juxtaposition: Monetary indicators/ Control variables Intermediate targets Bundesbank: FLR (iCB) CBM SNB: Ba (iCB) Mt Fed: RA (FFR) MU2,3 (longer-run), M1%V (short-run), FFR (?) 2. With respect to the instruments of monetary policy, by and large, nothing has changed as compared to the situation prior to the "New Era". Similar holds for the monetary indicators and control variables, respectively. Common to all central banks' concepts is that these variables include at least one component, i.e. excess reserves (ER), which signal the future expansion potential of banks. Within the Bundesbank's approach the "free liquid reserves" of banks (FLR) are still employed as monetary indicator and, at the same time, as control variable. The OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 394 Berichte FLR consist exclusively of banks' future expansion potential, comprising their ER and — as "potential" central bank money — their open rediscount quotas and their holdings of money market papers returnable to the Bundesbank at any time. Thus, in the Bundesbank's concept the aged "free reserves doctrine" seems to be pretty well alive. In the Fed's approach the indicator and control variable functions are fulfilled, now as before, in particular by various reserve aggregates (RA), such as the montary base, total or nonborrowed reserves, and the RPDs. Only in Switzerland a marked shift ocurred in so far as the "adjusted monetary base" (Ba = monetary base minus SNB's refinancing credits granted to banks) is now designed to serve as monetary indicator and control variable. icB (those interest rates which the central bank is able to fix directly) and FFR (Federal funds rate) are set in parentheses. This is to indicate that these variables also function to a certain degree as monetary indicator and control variable in the pertinent cases, partially as supplements, partially as reflexes of changes in the liquidity and reserve aggregates. Due to the relative unimportant role of icB in the Fed's conception the FFR replaces icB. 3. With respect to the intermediate targets each central bank has its own speciality: Within the Bundesbank's concept it is the CBM ("central bank money" in the specific definition of the Bundesbank). The CBM is a money stock definition sui generis, from the uses side comprising all currency in circulation and vault cash of the banks plus required reserves calculated at constant reserve ratios. The CBM consists exclusively of "used up" actual central bank money. Moreover, due to the composition of the CBM there is (empirically) a significant pars-pro-toto-identity between the M3-aggregate and the CBM. Thus, the CBM is a monetary aggregate and not a reserve aggregate, as it is often interpreted. The CBM is therefore not an indicator of whether monetary impulses have been emitted, but of whether they have shown up — together with all the other influences besides monetary policy — in the aggregates Mi, 2,3. The SNB has chosen the Mi-aggregate in the usual definition as its intermediate target. Unlike the Swiss and the German central bank the Fed specifies tolerance ranges and announces target values of four variables: (1) for the 12-months growth of the Mi,e,3-aggregates and the credit proxy; (2) for the 2-months growth of the aggregates Mi and Mg; (3) for the monthly evolution of the FFR, and (4) — until March 1976 — for the monthly growth of the RPDreserve aggregate. Whereas the Mi,2,3-aggregates are employed as OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 Berichte 395 intermediate targets, the role of the FFR remains somewhat unclear (see IV). 4. The fact that single number targets (SNB, Bundesbank) simulate a precision of monetary control which actually does not exist, is often mentioned as a disadvantage. On the other hand, however, the Fed's procedure gives the very same impression when the upper and the lower limits of the Mi,g,3-tolerance ranges are changed by 0.5 percentage points. III. Techniques of controlling monetary aggregates Common to the monetary control techniques of all three central banks are the following features and problems, respectively: 1. On principle, the adoption of monetary growth targets does not mean that ad hoc-measures have been given up. Rather, the very control policy continues to be discretionary, that is to say a policy of "ad hoccery". 2. The technique of controlling monetary aggregates may be denoted as "indirect control mechanism". Discretionary injections or absorptions of actual or potential central bank money are supposed to change the control variables, thereby affecting the money market rates and subsequently radiating out from the money market on the whole interest rate structure of the various financial markets. The transmission of changes in the control variables to the intermediate targets is viewed by all central banks as a portfolio adjustment process generated through interest rate impulses which ultimately lead to changes in the growth rates of monetary aggregates. The control devices of all central banks are thus based on price respectively interest rate-theoretical grounds. To judge from the central banks' statements the transmission of monetary impulses to the intermediate targets primarily runs through the banking sector with credit interest rates and the demand for and the supply of bank credit as key variables. 3. The main problems of controlling monetary aggregates are the tuning and the timing of control variables, and in addition, the time lags. The crucial point is: When and by how much should RA (B®, FLR) be changed to bring or to hold Mi, 2,3 (Mi, CBM) "on target"? And how long does it take until changes in the control variables affect the composition of banks' and nonbanks' portfolios? The plain fact is that, up OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 396 Berichte to now, a reliable basis is missing for calculating the appropriate change in control variables required to achieve a certain monetary growth target. What would be needed are empirically supported and predictable experiences about the portfolio behavior of banks and nonbanks to escape from the "trial and error" method used so far. 4. Along with this problem, fluctuations of considerable size in the velocity of money and in the money multipliers have proved to be a serious disturbing factor for the control policy of all three central banks. These fluctuations reflect, although not exclusively, also changes in the portfolio behavior. IV. The perfomance of the new strategies On the basis of the short experience it would be downright a speculative endeavor to judge whether the "New Era" was a success or not. With respect to the implementation procedure and the performance of the new strategies, however, some inferences can already be drawn. 1. The Bundesbank was faced in particular with the following problems: a) Its main problem were extremely fluctuating FLR-multipliers (mFLR = CBM/FLR) which made the control of the CBM-growth rate enormously difficult. As recent experience shows, mFLR fluctuated between the extreme values of 45 and 7 within a period of only 6 quarters (Q3/1974 - Q4/1975) without affecting the speed of expansion of the CBM very much. Considering these wide fluctuations it is yet somewhat amazing that the CBM-growth targets of each 8 % in both past years were only slightly overshot. b) According to the Bundesbank the CBM is designed to serve as an "indicator of monetary expansion", a confusing notion which has never been defined precisely, but which is nevertheless time and again stressed by the Bundesbank. Actually, there is no room left for the so-called "monetary expansion" in the Bundesbank's approach. Since the CBM largely is a reduced copy of the Ms-aggregate the CBM measures itself — to exaggerate a little bit. c) The analyses of the Bundesbank do not show on what theoretical and empirical basis the composition and weighing of the CBMcomponents rest or whether there is such a basis at all. A priori the CBM-definition seems arbitrary. Its obvious advantage is only that the CBM is largely independent of shifts among deposits. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 Berichte 397 d) Even though this aspect is of minor relevance for the construction of an intermediate target, the performance of the CBM as intermediate target over the past five years was by far superior as compared with other monetary aggregates — in the meaning of the smallest variances between the growth rates of both CBM and nominal GNP. e) The Bundesbank explicitly includes expected changes in the velocity of CBM in the computation of the CBM-growth rate, thereby bridging a gap to a magnitude (nominal GNP) which is about 10 times as great as the CBM. Bad miscalculations of the velocity of CBM might result in short-run revisions of the CBM-growth target, thereby casting considerable doubt on the desired stabilizing effect and on the credibility of the new strategy. 2. Some major deficiencies of the SNB's approach were: a) As experience shows so far, the SNB has chosen with Mi an intermediate target which is a variable too narrow and volatile because of shifts in nonbank deposits. Similar holds for Ba as control variable. Extremely wide fluctuations of the BMnultiplier and a considerable over-respectively undershooting of the Mi-growth target — specified at a level of each 6°/o in both past years — have been the consequence. To judge from recent experience the combination of not adjusted monetary base and a broader monetary aggregate, say or M3, showed a far better correlation — again in the meaning of the smallest deviations between the growth rates of both variables, thus possibly suggesting an adoption of that combination. Even though Ba has analytical advantages as monetary indicator, with respect to the practice of controlling Mi, however, the elimination of refinancing credits seems to be a somewhat artificial separation of determinants of Mi which belong together. b) M2 as intermediate target (instead of Mi) would have had the advantage of partially identical growth rates of both M2 and nominal GNP over the past five years with correspondingly small fluctuations of the velocity of Mg. In contrast, the growth rates of both nominal GNP and Mi did not show any connection whatsoever. c) The SNB's procedure to set the same target value (or a target value at all) for both the control variable and Mi is very questionable, if the respective money multiplier is highly unstable (as it actually OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19 398 Berichte was during the past five years). For 1977 the SNB has, presumably therefore, renounced to publicly specify a target value for Ba. d) Foreign reserve transactions of the SNB have so far been the most important lever to affect the monetary base. This practice is subject to constraints in so far as the attempt to keep the Swiss exchange rate stable and, at the same time, the domestic monetary growth rate on target is not always compatible. In those cases other instruments of monetary policy need to be intensified. 3. The Fed's approach was impaired by the following problems: a) A serious defect was that, with respect to the setting of monetary growth targets, no consultations with the Federal Government took place. Furthermore, the Fed did not make public the components of monetary growth targets, for instance, what is considered as "unavoidable" rate of inflation. Since the new approach is to a great extent based on psychological grounds, more public relations by means of publicly announcing the decomposition of monetary growth targets could possibly improve the performance of the new strategy. b) It is very difficult to judge whether the Fed's approach has been a success or not, especially due to the simultaneous announcement of target values for four different monetary variables, moreover, due to the extremely wide tolerance ranges, and also due to the changing bases for the computation of target values for the longer-run Mi.g.uaggregates. c) The simultaneous setting of interest rate and quantity targets sharply raises the problem of consistency, because only a small number of interdependences between the monetary variables can be built into an empirical model. d) The wide tolerance ranges may be considered by the Fed as "political advantage" and "monetary flexibility", especially in the light of House Concurrent Resolution 133. From an economic point of view, however, this practice is questionable because the pursuit of either the lower or the upper limit of those wide tolerance ranges may have very different effects on economic activity. Thus, as long as the ranges for the annual Mi,2,3-growth targets are not considerably reduced, it is hard to believe that a policy of controlling the money stock is really seriously treated by the Fed. OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/ccm.10.3.391 | Generated on 2023-01-16 12:42:19