The Evolution of Money Targeting, 1975–1998

  • Aerdt C. F. J. Houben
Part of the Financial and Monetary Policy Studies book series (FMPS, volume 34)


The rise of money targeting has its roots in the 1960s, when ‘monetarism’ gained academic prominence and the preoccupation of policy-makers started to shift gradually away from interest rates towards monetary aggregates.1 But the eventual adoption of money targets for the most part reflected a pragmatic policy reaction to the changed global economic circumstances of the early 1970s, rather than any strong ‘monetarist’ convictions. Two developments notably spurred the advance of money-based strategies in Europe. First, the collapse of Bretton Woods and the limited, brittle scope of the snake granted monetary authorities greater control of domestic monetary developments and, by implication, also bestowed them with greater responsibility for monetary policy outcomes. Whereas the exchange rate had previously served as the primary (intermediate) monetary policy target, the new international monetary environment left the central banks of the large European countries without a nominal anchor at which to gear their instruments. Second, the first oil crisis and the concomitant near quadrupling of oil prices — representing the largest supply-side price shock in modern history — sparked a global inflationary momentum. This shock occurred at a time when monetary conditions were already loose, partly on account of the massive exchange market intervention in the terminal phase of the Bretton Woods system. Given unabated money growth in 1974–75, direct price effects were accommodated and inflationary pressures continued to build. In itself, this wrenching experience contributed to wide acceptance that rapid growth in the money stock and in prices go hand in hand.


Monetary Policy Central Bank Money Demand Money Growth Liquidity Ratio 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    On the former development see M. Friedman (1968), on the latter OECD (1979).Google Scholar
  2. 2.
    During the brief period in 1971 that the German mark and the Dutch guilder floated (from shortly before the end of gold convertibility until the Smithsonian Agreement), the Bundesbank intervened to generate an appreciation of the mark and thereby to signal the need for wage and price restraint; see Solomon (1982, p. 181). In a similar vein, Austria’s official revaluation against the Deutsche Mark in September 1979, at the beginning of the autumn wage bargaining round, was expressly aimed at preventing a lag between revaluation and wage response; see Hochreiter and Winckler (1995, p. 93).Google Scholar
  3. 3.
    On the differences of opinion between the Bundesbank and the German government regarding currency revaluation during the Bretton Woods era, see for example, Berger and De Haan (1999) and Giavazzi and Giovannini (1989, p. 23-24). In this context, it should be noted that the views within both the Government and the Bundesbank were generally divided.Google Scholar
  4. 4.
    Schlesinger as quoted in Issing (1992, p. 291).Google Scholar
  5. 5.
    For the early use of money targets in Greece, see Cottarelli and Giannini (1997, p. 25); for the Netherlands, see De Greef, Hilbers and Hoogduin (1998). Dutch monetarism has its roots in the interbellum and was translated into operational terms in the 1950s; it had a strong influence on the subsequent development of the monetary approach to the balance of payments. Kessler (1958) sets out the main concepts underlying this method of monetary analysis. Fase (1987) provides an academic review of Dutch monetarism.Google Scholar
  6. 6.
    Next to the specific country cases mentioned, money targets also started playing a role at the supranational level at a relatively early stage. In 1972, the European Council recommended that the member states’ central banks adopt such targets; see Von Hagen (1999, p. 688). More specifically, in late 1973 the EC Economic and Finance Ministers agreed that member countries should strive gradually to lower money growth to the targeted rate of nominal GNP expansion; see McClam (1978, p. 9).Google Scholar
  7. 7.
    The primacy granted to the money growth target ranges was subject to key money market rates (chiefly the Federal funds rate) not moving outside a stated range. See Wallich and Keir (1978) and Kole and Meade (1995).Google Scholar
  8. 8.
    Foot (1981, p.15) and Richardson (1978, p. 33).Google Scholar
  9. 9.
    For France, see Cobham and Serre (1986, p. 24); for Spain, Escrivá and Malo de Molina (1991, pp. 161-162).Google Scholar
  10. 10.
    Kloten(1992).Google Scholar
  11. 11.
    OECD (1979).Google Scholar
  12. 12.
    Schlesinger (1983).Google Scholar
  13. l3.
    Fforde(1983, p. 53).Google Scholar
  14. 14.
    Lawson (1993, pp.45 and 66-87) provides a vivid account of how seriously the Government initially took its money targets, expressly opting for a new monetary policy framework with less discretion.Google Scholar
  15. 15.
    Cobham and Serre (1986, p. 27).Google Scholar
  16. 16.
    See for instance the views of Banque de France official J-P. Patat set out in Eijffinger (1993, p. 174).Google Scholar
  17. 17.
    Fase (1985) presents a comprehensive account of the Dutch experience with monetary control aimed at the liquidity ratio.Google Scholar
  18. 18.
    A conceptual description of how the Netherlands’ exchange rate and liquidity ratio targets were combined is provided by Wellink (1989).Google Scholar
  19. 19.
    Ayuso and Escrivá (1998).Google Scholar
  20. 20.
    Pringle(1995).Google Scholar
  21. 21.
    For a more elaborate discussion of Italy’s long road towards money targetry, see Angeloni and Passacantando (1991), Argy (1990), Caranza and Fazio (1983), and Passacantando (1996).Google Scholar
  22. 22.
    The public sector share in total annual domestic credit expansion soared from around 35 percent in the early 1970s to more than 65 percent a decade later; see Caranza and Fazio (1983).Google Scholar
  23. 23.
    The Danmarks Nationalbank Monetary Review of February 1991 refers to the new target as follows: “As an element of strengthening co-operation between the central banks of the EEC countries on the prior co-ordination of national monetary-policy objectives, the Nationalbank has decided to use as its indicator the domestic money creation. … A range for growth in domestic money creation has been set at 4–7 percent from the 4th quarter of 1990 to the 4th quarter of 1991.” While the bank’s Annual Report over 1991 (p. 38) expressly links the outcome of “just over 1 per cent” to the target (4–7 per cent), the Annual Report over 1992 no longer relates the outcome to the target. This is not surprising given that performance under the target (3–6 per cent) would have been difficult to explain: domestic money creation fell in absolute terms by 9 percent during that year.Google Scholar
  24. 24.
    Although the statistical differences between central bank money and the monetary base are small (mainly on account of the latter’s inclusion of banks’ excess reserves and reserve requirements on external liabilities), the Bundesbank repeatedly emphasised the major conceptual differences between these two aggregates in a targeting framework. Specifically, rather than advocate a strategy of direct money base control, the German central bank has underscored the intermediate character of its central bank money target, to be achieved indirectly by influencing money market conditions. On the motives behind the selection of central bank money as target variable, see also Schlesinger (1979 and 1983) and Kloten (1992).Google Scholar
  25. 25.
    Central bank money consisted of currency in circulation, plus 16.6 percent of sight deposits, 12.4 percent of time deposits, and 8.1 percent of savings deposits below four years. These weights corresponded to the reserve ratios applicable in January 1974 and sought to do justice to the varying liquidity of the different components of M3. Currency effectively accounted for about half of the central bank money stock and roughly ten percent of M3.Google Scholar
  26. 26.
    Later data revisions indicate that only two of the three targets had in fact been met. More importantly, the success in 1982/83 was partly cosmetic, as it was met through overfunding. With this technique, the Treasury sold more gilt-edged securities than needed to finance the public deficit in order to attract money that would otherwise have been left with banks. Over the period 1981/82 through 1984/85, overfunding is estimated to have reduced £M3 growth by nearly 4 percent a year; see Lawson (1993, pp. 458-460).Google Scholar
  27. 27.
    Fforde (1983) and Lawson (1993) provide animated chronicles of the United Kingdom’s struggle to find an appropriate money aggregate to target. See also Cobham and Serre (1986) and Townend (1991).Google Scholar
  28. 28.
    In the words of Prime Minister Thatcher’s personal advisor Alan Walters: “M3 was a dog that barked too often to be taken seriously”; as quoted in Lawson (1993, p. 482).Google Scholar
  29. 29.
    The decision to abandon monitoring ranges for MO and M4 is set out in the Bank of England’s November 1997 Inflation Report (pp. 8-9).Google Scholar
  30. 30.
    Raymond (1983) and Argy (1983, 1990).Google Scholar
  31. 31.
    Fournier and Aileron (1991).Google Scholar
  32. 32.
    Zijlstra (1985).Google Scholar
  33. 33.
    The liquidity ratio figures were actually subject to enormous data revisions and substantive definition changes. In fact, the Annual Reports of the Nederlandsche Bank for the period 1976 through 1980 contain different historical figures each year, whereby the magnitude of the statistical adjustments at times even exceed the size of the targeted change.Google Scholar
  34. 34.
    Rojo and Ariztegui (1984, p. 182).Google Scholar
  35. 35.
    OECD (1979), Caranza and Fazio (1983), and Vaciago (1985).Google Scholar
  36. 36.
    Angeloni and Passacantando (1991). As an illustration of the apparent changes in the stability of money demand, an OECD (1979) study had earlier concluded that it could not find one conventionally defined money aggregate that could be considered a reliable guide for monetary policy in Italy.Google Scholar
  37. 37.
    Goodhart and Viñals (1994).Google Scholar
  38. 38.
    In practice, the Bundesbank fudges at the margins in order to improve its targeting record. Specifically, by rounding actual growth figures to the nearest percentage point (notwithstanding targets sometimes expressed in half percentage points), outcomes that otherwise fall outside the announced target range can be classified as having been kept within it. This was the case in 1980 (outturn 4.9 per cent, target 5–8 per cent), 1981 (outturn 3.6 per cent, target 4–7 per cent) and 1991 (outturn 5.2 per cent, target 3–5 per cent), years in which the target was met according to Deutsche Bundesbank (1995, p. 79) and Issing (1997, p. 69). Since this rounding practice effectively widens the target range by half a percentage point at both the lower and upper limit (implying bandwidths of three or four percentage points), it is inconsistent with the statement (Deutsche Bundesbank p. 84, Issing, p. 71) that the target ranges are typically two or three percentage points. With this elastic interpretation of its money targets, the Bundesbank improves its success rate in meeting money targets from 42 percent to 54 per cent.Google Scholar
  39. 39.
    Anticipation of this withholding tax had prompted large shifts into cash in the previous half year; the tax was rescinded in mid-1989. See Kole and Meade (1995, p. 924).Google Scholar
  40. 40.
    The increasingly prominent role of inflation targets in Italy, particularly after 1993, is set out in Visco(1995).Google Scholar
  41. 41.
    The Bundesbank rejects using only multi-year targets, as this risks weakening policy discipline by creating a wait-and-see attitude to target deviations; see König (1996, p. 123).Google Scholar
  42. 42.
    An exceptional position is taken in by the United Kingdom where, in line with the different coverage of the fiscal year, targets have generally extended from March to March. As a further peculiarity, the United Kingdom had money targets during the first half of the 1980s that referred to a fourteen-month period (February to April) thereby reducing the impact of irregularities linked to the end of the fiscal year.Google Scholar
  43. 43.
    Bank of England Governor Richardson (1978, p. 37) was a firm advocate of six-month rolling targets, but considered quarterly reassessments — such as those then conducted by the US Federal Reserve — as too much of a good thing.Google Scholar
  44. 44.
    Escrivá and Malo de Molina (1991).Google Scholar
  45. 45.
    Issing (1997, p. 71).Google Scholar
  46. 46.
    In the words of Thatcher (1993, p. 97): “The MTFS would only influence expectations insofar as people believed in our determination to stick to it: its credibility depended on that of the Government — and ultimately, therefore on the quality of my own commitment, about which I would leave no one in doubt.”Google Scholar
  47. 47.
    Raymond (1983) provides a detailed description of the decision-making procedure for the monetary target, at the time involving the Prime Minister, the Minister of the Economy and Finance, and the Governor of the Banque de France.Google Scholar
  48. 48.
    This presentation includes an address by the Governor of the Bank of Italy. More generally, the Italian central bank is intimately involved in the preparation of the Ministry of the Budget’s macro-economic projections. It is also noteworthy that although all financial targets fell under the final responsibility of the Government, the Governor customarily proposed the monetary policy targets and these targets were never rejected; see Angeloni and Passacantando (1991).Google Scholar
  49. 49.
    On this basis, Von Hagen (1995) suggests that the strategic orientation of the Bundesbank should not be classified as monetary targeting, but rather as “monetary and inflation targeting”.Google Scholar
  50. 50.
    In discussing monetary targets in its Annual Report for 1976, the Netherlands Bank noted that “it is desirable that published targets as to the liquidity ratio should be accompanied by their implicit targets with respect to money supply growth, given the expected rise in net national income in real and nominal terms.”Google Scholar
  51. 51.
    Townend(1991, p. 206).Google Scholar
  52. 52.
    For the group of industrialised countries as a whole, the use of money targets surged in the second half of the 1970s, stabilised at close to 50 percent in the first half of the 1980s and declined to below 25 percent over the course of the next decade. Monetary targetry has remained relatively rare among developing countries, although a slight increase has been apparent in recent years; see Cottarelli and Giannini (1997).Google Scholar
  53. 53.
    Shigehara(1996, p. 11).Google Scholar
  54. 54.
    Goodhart (1997) provides a colourful account of the Bank of England’s vainglorious attempts at finding a stable money demand function and of the warnings to the Government, including by himself, that inordinate emphasis was being placed on the money targets.Google Scholar
  55. 55.
    Lawson (1993, p. 456).Google Scholar
  56. 56.
    Monetary authorities initially sought to cover this risk with minimal references to uncertainty. For instance, the United Kingdom’s first MTFS opened with a footnote stating “The way in which the money supply is defined for target purposes may need to be adjusted from time to time as circumstances change.” The fact that Germany originally only adopted money targets on an experimental basis could also be seen as conveying the prevailing uncertainty, albeit more implicitly. However, this dimension of targeting frameworks was not given much prominence. Indeed, highlighting the uncertainty governing a target does not sit well with the desire to strengthen its signalling role. Anyhow, as recognised by Lawson (1993, p. 68) on the basis of first-hand experience, such nuances about uncertainty are unlikely to be picked up by the public, including academic commentators.Google Scholar
  57. 57.
    Early studies already indicated the superior money demand stability in Germany: see for example OECD (1979), Atkinson et al. (1984), and Isard and Rojas-Suarez (1986).Google Scholar
  58. 58.
    The fact that Germany has combined the lowest variability in interest rates and in money growth illustrates that, contrary to the intuitive view, there is not necessarily a trade-off between these two variables; see Bernanke and Mishkin (1992, p. 202).Google Scholar
  59. 59.
    An overview of the recent studies is provided in Kole and Meade (1995). König (1996) presents further evidence that the long-run demand for money function has remained stable and predictable even after reunification.Google Scholar
  60. 60.
    In a study of money demand stability in Europe over the period 1971–1989, Fase and Winder (1993) find the United Kingdom to have the least stable money demand function for each of the aggregates Ml, M2 or M3 of any of the then eleven EC currencies (with the singular exception of greater Ml demand volatility in Greece).Google Scholar
  61. 61.
    Lawson (1993, p. 80) and Goodhart (1997, p. 407). The Bank of England had previously made a rough estimate that the abolition of the ‘corset’ direct credit rationing scheme (which penalised banks’ growth of interest-bearing deposits above a certain allowable rate) would raise £M3 by somewhat over 2 per cent; in reality, the contribution to money growth was nearly 5 per cent.Google Scholar
  62. 62.
    Vega (1994) analyses the deterioration of ALP money demand stability in Spain, suggesting a structural break in 1989 related primarily to the lifting of restrictions on capital movements. Although incorporating foreign interest rates in the long-run ALP demand function restores an appearance of stability, this also implies a reduced controllability, thereby bleakening prospects for a money targeting strategy.Google Scholar
  63. 63.
    The study by Fase and Winder (1993) confirms that money demand stability in individual EC countries has generally been inversely related to the size of the country.Google Scholar
  64. 64.
    There is a subtle distinction between credit ceilings as instruments towards achieving another intermediate money target (such as with the credit ceilings supporting the liquidity ratio targets in the period 1977–80) and credit ceilings as an intermediate monetary target in their own right: the strategic focus of the central bank switches from the liability side to the (domestic) asset side of the balance sheet. Hilbers (1998) discusses the changing design of the various credit controls in the Netherlands.Google Scholar
  65. 65.
    The shift away from a dual strategy combining an exchange rate target, on the one hand, with a liquidity ratio or domestic money growth ceiling, on the other, and towards singular exchange rate targeting is explicitly acknowledged in the Nederlandsche Bank’s Annual Report over 1991.Google Scholar
  66. 66.
    A prime illustration of the detrimental effect of excessive changes in a money targeting framework is provided by the United Kingdom’s announcement of novel targets for M0 in 1984. After having recently targeted £M3, Ml and PSL2, the new target failed to have a real impact. In the words of the then Chancellor Lawson (1993, p. 457): “The financial markets [were] inclined to believe that any departure from the original rules was politically expedient and therefore economically unsound. … In time it became clear that the real problem with MO was that it lacked street credibility. … The achievement of the MO target-or even its undershooting — had no great effect on inflationary expectations.” The irony is that M0 subsequently proved to be more reliable and stable than any other aggregate targeted in Europe (and probably the world).Google Scholar
  67. 67.
    Fase and Winder (1993) present evidence on the greater stability of broad aggregates for each of the EC member countries. Monticelli and Strauss-Kahn (1992) discuss the preference for broad aggregates in Europe.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2000

Authors and Affiliations

  • Aerdt C. F. J. Houben
    • 1
  1. 1.De Nederlandsche BankAmsterdamThe Netherlands

Personalised recommendations