Abstract
Exchange rate targeting had fallen out of favour in the first half of the 1970s. The collapse of Bretton Woods signalled a global shift towards greater exchange rate flexibility and the European response in the form of the ‘snake’ common margins agreement did not offer a broadly acceptable alternative. Indeed, as participation dwindled, the snake symbolised how limited the common ground was on the strategic objectives of monetary policy and how pitifully little was left of Europe’s collective monetary ambitions. But, despite its shortcomings, the snake did constitute a nucleus of monetary stability in Europe. It was around this nucleus that new monetary initiatives could be built once greater like-mindedness in monetary policy objectives returned.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
Kruse (1980), Thygesen (1979) and Ungerer (1997). A political analysis of the motives leading to the EMS is provided in Ludlow (1982) and Szász (1999).
Thygesen (1979, p. 94-108)
Monticelli and Papi (1996) provide an extensive review of asymmetry in monetary policy co-ordination.
In an effort to increase the attractiveness of holding official ECUs, the yield was adjusted in 1985 to the average of national money market rates, typically about 1 percentage point higher than the average of national discount rates.
While the short-and medium-term credit facilities were accessible to all EC countries, regardless of whether they participated in the ERM, it was agreed that the enlarged quotas of the United Kingdom in the former facility were only to be used after this country had joined the ERM; see Ungerer (1997).
Although the size of these transfers was originally modest, they played a significant role in the negotiations on the creation of the EMS; see Kruse (1980) and Ludlow (1982). The issue featured particularly prominently in Ireland, where ERM participation involved severing the exchange rate parity with the British pound that had lasted for over fifty years; see Honohan (1997).
These issues were the objects of intensc empirical study during the first decade of the EMS. Evidence on the reduced volatility of bilateral exchange rates and lower variability in short-term interest rates is presented in Artis and Taylor (1988), Padoa Schioppa (1985), Ungerer, Evans and Nyberg (1983) and Ungerer et al. (1986). Evidence on the contribution of the EMS to disinflation is more tentative since disinflation was a world-wide phenomenon in the 1980s; see for example Collins (1988) and Giavazzi and Giovannini (1988). A formal analysis of the EMS as a disciplinary device fostering inflation convergence is set out in Giavazzi and Pagano (1988).
European Commission (1989), Giavazzi, Micossi and Miller (eds.) (1988), Gros and Thygesen (1992), Monticelli and Papi (1996), Ungerer (1997) and Ungerer et al. (1990)
This can be illustrated by comparing the average outcomes for these variables during the periods 1975–78 and 1979–1982 between the ERM countries (excluding Luxembourg which did not have its own currency), on the one hand, and a group of seven other industrial countries (Canada, Japan, Norway, Spain, Sweden, UK and US — selected on the basis of data availability for all four variables), on the other. In the case of the general budget balance, the standard deviation of the outcomes increased in the ERM countries from 2.6 during 1975–78 to 3.6 percent of GDP during 1979–82, against an increase from 2.4 to only 2.7 percent of GDP in the control group of other industrial countries. For the external current account, wage and consumer price changes, there was an increase in the standard deviation of the outcomes for the ERM countries (from 2.1 to 4.2 percent of GDP, from 3.6 to 5.4 per cent, and from 3.7 to 4.9 percent respectively) against a decrease in the other industrial countries (from 2.6 to 1.5 percent of GDP, from 7.6 to 4.2 percent and from 4.4 to 2.9 percent respectively). Substitution of Austria, Australia or Switzerland in the control group leads to broadly the same conclusion. The background data are drawn from Tables 13, 17, 19 and 31 in Ungerer et al. (1990).
Jurgensen(1983).
Gros and Thygesen (1992), Padoa Schioppa (1985) and Ungerer (1997).
An overview of the adjustment measures announced in connection with the realignments through 1990 is provided by Ungerer (1990).
Belgium and Denmark asked for devaluations of 12 and 7 percent, but ultimately agreed to 8½ and 3 percent respectively; see Gros and Thygesen (1992, p. 76).
The stabilisation package was supported by an ECU 4 billion loan through the EU medium-term facility.
This can be deduced from the evidence presented in De Grauwe (1990, p. 148).
Folkerts-Landau and Mathieson (1989).
Mastropasqua, Micossi and Rinaldi (1988) and Giavazzi and Giovannini (1989). Gros and Thygesen (1992, pp. 136-57) provide an overview of the findings in other asymmetry studies and conclude that asymmetry tests fare best in the 1983–86 period.
Gros and Thygesen (1992, p. 84) and Ungerer et al. (1990, p. 14-15).
Although the bilateral central rate for the lira was devalued by 3.75 percent when Italy moved from the wide (± 6 per cent) to the narrow (± 2.25 per cent) band in January 1990, this did not increase the scope for lira depreciation as the lower intervention points remained unchanged.
Norway, Sweden and Finland pegged their currencies to the ECU in October 1990, May 1991 and June 1991, with fluctuation margins of ± 2.25, ± 1.5 and ± 3 percent respectively.
The texts of the Basle-Nyborg communiqués are provided in Ungerer et al. (1990, Appendix II).
See Walters (1986, pp. 125-132).
The analysis in Eichengreen and Wyplosz (1993) that the persistently higher inflation and rising labour costs in these countries did not cause competitiveness and balance-of-payments problems, but reflected the Balassa-Samuelson effect, is unconvincing. With the benefit of hindsight, it is easy to refute this analysis by pointing at the massive subsequent devaluations by Portugal (12½ percent cumulatively between mid-1992 and end-1993) and Spain (19 percent) within the ERM framework, and the depreciations by Italy (32 percent between the mid-1992 Deutsche-mark parity and the end-1993 market rate) and the United Kingdom (15 percent) outside it. With the exception of the United Kingdom, the real exchange rates of these countries remained broadly around their newly found levels thereafter. If the Balassa-Samuelson effect had previously been a driving force behind the inflation differentials, the initially higher real exchange rate levels would have proved sustainable.
The term ‘New EMS’ originates from Giavazzi and Spaventa (1990).
In a study of monetary conditions in the EMS during this period, Clarida, Gali and Gertler (1997) show that Germany’s interest rates were indeed too high for the economic conditions of other countries in the system. By applying an estimated reaction function of the Bundesbank to France, the United Kingdom and Italy, they find that the inflationary pressures from German unification led to unreasonably high interest rates and thus to untenable economic stress, particularly in the first two of these countries.
Finland, Sweden and Norway severed their ECU-pegs on 8 September, 19 November and 19 December 1992, respectively.
At the time of the widening of the fluctuation bands, Germany and the Netherlands mutually agreed, outside the ERM, to maintain the ± 2.25 percent margin for their bilateral exchange rate. Szász (1999) describes the astute political manoeuvres by the Netherlands that elicited this exceptional deal, notwithstanding the opposition of countries such as France that were bent on avoiding a differentiated treatment.
Von Hagen (1999, p. 694).
Von Hagen (1989) shows that the empirical results for short-run and long-run sterilisation are very different and that the latter generally did not occur throughout the 1980s. The pattern of long-run sterilisation can be clearly linked to the exchange rate developments of the Deutsche mark against both the dollar and the EMS currencies.
Hochreiter and Winckler (1995).
Gnan (1994, p. 32).
Glück, Proske and Tatom (1992, p. 168).
Hochreiter and Winckler (1995) present evidence that asymmetric shocks dominated until the mid-1980s and that real wage flexibility allowed Austria to become part of an optimal currency area with Germany. In particular, they show that the variability of the schilling’s real exchange rate vis-à-vis the Deutsche mark was significantly higher than that of, for example, the Dutch guilder during 1980–87, but that it was clearly lower than that of all other EC and EFTA currencies during 1987–92.
In the words of then central bank official Boot, the main lesson of the unsatisfactory realignment outcome was the importance of creating “an undeniable political fact regarding the guilder in as far as this is possible: an unassailable position in the market, implying that next time a deviation from the German mark would not be credible”; see Szász (1999, p. 199).
In the Netherlands, monetary policy is the domain of the central bank, but exchange rate regime and realignment decisions are governmental responsibility. When such decisions are taken, the central bank provides a public recommendation.
Szász (1999, p. 201).
Using correlation coefficients for GDP, inflation, unit wages, and short-and long-term interest rates as broad indicators for the fulfilment of optimal currency area criteria, Wellink ( 1994) finds that the economy of the Netherlands is more tightly linked with Germany than any other EMS country — with Belgium a close second — and that this linkage was generally tighter for the period 1989–1993 than for 1983–93, suggesting increasing convergence.
Luxembourg did not have its own central bank. The exchange rate between the Belgian franc and Luxembourg franc was fixed in 1936 and rescaled to a rate of one-for-one in 1944. However, Luxembourg did give its opinion on the monetary and exchange rate policies pursued by Belgium, occasionally voicing its disapproval and raising the spectre of a break-up of the Belgium-Luxembourg Economic Union. This was, for instance, the case when Belgium requested a disproportionately large devaluation in 1982; see Gros and Thygesen (1993;p. 76).
Michielsen(1994, p. 66).
Quintyn and Vuchelen (1992). Monetary financing is defined as the total of direct and indirect borrowing by the Treasury from the central bank, including foreign borrowing whereby foreign exchange proceeds are sold directly to the central bank.
Denmark was continuously a member of the snake and the ERM, with the exception of a short interruption in 1972 when the snake was abandoned in the wake of the United Kingdom’s withdrawal. However, the krone was kept within the margins and Denmark formally rejoined the arrangement at its original parity within four months.
De Grauwe and Vanhaverbeke (1990) attribute the Danish success primarily to the impact of the one-off capital liberalisation, which allowed the external inflows that fuelled the domestic consumption and investment boom. They arrive at this conclusion in a comparison with Belgium and the Netherlands, which also initiated adjustment programs in the early 1980s, but could not count on such an impact as their capital flows were already essentially free. However, it seems the much more forceful and comprehensive nature of the Danish policy package was the key distinguishing factor.
Christensen and Topp (1997).
Denmark was the principal — and unsuccessful — advocate of a general return to more narrow bands. The August 1993 statements by the Prime Minister and the central bank Governor are reproduced in Christensen and Topp (1997, pp. 22-23).
The pre-announcement expressly mentioned the agreement of the euro area ministers and the ECB. Although the central rate around which the narrow fluctuation band would be positioned was not specified, the press communiqué did spell out that Denmark had had an unchanged parity within ERM since January 1987.
James (1996, p. 476-77) offers a colourful account of the heated debate on exchange rate strategy in France during 1982–83.
In an interview with Le Monde on 20 December 1989, Finance minister Bérégovoy indicated that the stability of the exchange rate vis-à-vis Germany had become the “cornerstone of policy-making”. From the perspective of the Banque de France, Icard (1994a and 1994b) recognises that the exchange rate has been given precedence in monetary policy setting since the mid-1980s, but conjectures that the money supply target continued to fulfil an important role in influencing expectations by stressing the authorities’ abiding commitment to price stability. However, it is clear that the prospect of exchange rate stability, which had to be upheld in daily monetary operations, has had a much more powerful influence on expectations than the annual announcement of a money target that was more often than not missed. Thus, the role of the money targets primarily seems to have been a political one: to provide a domestic coating to an externally oriented monetary policy.
The central bank of Ireland was set up in 1943, but domestic lending operations remained on a very modest scale through the 1970s. From 1971 on, adjusting the exchange rate no longer required legislative change, but could be decided by the minister of finance after consultation with the central bank. Honohan (1997) provides an insightful review of the gradual evolution of Ireland’s currency board arrangement.
Leddin and Walsh (1995).
The exchange rate strategy was apparently not credible in the labour market during the initial period of ERM membership. In 1979, following entry into the arrangement, determined efforts to have wage agreements based on a single digit inflation forecast failed and wages subsequently fuelled the inflationary spiral; see Walsh (1983).
Dornbusch (1989) assesses the first decade of Ireland’s EMS membership and concludes — given the costs in terms of unemployment, emigration and government debt-that the stabilisation program was a failure and that EMS participation did not furnish a credibility bonus. While he considers an overly tight exchange rate policy to have been a prime cause of this failure, a major shortcoming was that this policy, if anything, was initially unduly loose. His appraisal would most probably also have been quite different if he had known of the impressive, drawn-out Irish recovery that would be built on the painful stabilisation efforts of the 1980s. More specifically, Kremers (1990) finds twofold empirical evidence that ERM entry actually did influence policy credibility: first, in 1979 inflation expectations shifted away from the United Kingdom towards the rest of the ERM and, second, during the ERM period-and especially 1981–82-competitiveness became an important determinant of expected inflation, with competitiveness losses expected to be recouped by adjustments in Ireland’s inflation rather than by exchange rate correction.
Leddin and Walsh (1995) recount the authorities’ attempt to avoid devaluation based on considerations of adequate domestic competitiveness, the need to defend the ERM against speculators, the desirability of breaking the dependence on the United Kingdom, and the lack of durable benefits from devaluation. It is testimony to the authorities’ efforts that the Irish devaluation only occurred four months after the United Kingdom withdrawal from the mechanism.
The Finnish markka had previously been devalued by over 30 percent in both 1957 and 1967 and by about 10 percent in 1971–73. Korkman (1978), for instance, provides a sharp theoretical critique of the Finnish devaluation cycle; Halttunen and Korkman (1983) present econometric simulations of the detrimental impact of this cycle.
Lehmussaari, Suvanto and Vajanne (1992, p. 10).
Gylfason (1990) finds that the standard deviation of Finland’s real exchange rate (2.0, calculated on the basis of the IMF’s Multilateral Exchange Rate Model) during 1975–88 was notably lower than that of for instance Norway (6.4), the United Kingdom (10.0) and the US (15.4, over 1978–88).
At a technical level, the central bank did, however, strive to increase the transparency of the basket, for instance, from 1984 on, by publishing the index daily rather than as a monthly average, thereby facilitating public monitoring of the exchange rate within the band; see Puro (1984). In addition, the switch to the ECU basket in 1991 undoubtedly enhanced policy communicability.
Lehmussaari, Suvanto and Vajanne (1992) offer a sweeping account of Finland’s boom-bust cycle in the late 1980s and early 1990s.
Pikkarainen et al. (1997, p. 39).
Hörngren and Lindberg (1993) provide an elaborate and compelling account of Sweden’s endeavours to turn the krona into a hard currency.
Hörngren and Westman-Mårtensson (1991, p. 176).
In contrast with the common institutional set-up elsewhere in Europe, Swedish exchange rate policy-including the regime choice-has traditionally fallen exclusively under the central bank, although the government had to be consulted prior to major decisions; see Hörngren and Westman-Mårtensson (1991, pp. 173-174).
In interpreting budgetary policy in Sweden, it needs to be borne in mind that this country is estimated to have had the most cycle-sensitive budget of the current EU members; see European Commission (1995).
The intervention total is reported in Hörngren and Lindberg (1993).
In the advent to EMU, Italian interest rates were the last to converge to the commonly agreed level (3 per cent), only reaching that level in the last week of 1998.
Progress in each of these three fields was in fact identified in 1981 by the central bank governor as a prerequisite for a stable currency; see Passacantando (1996) and Visco (1995).
Ayuso and Escrivá (1998).
In an empirical study of the Bank of Spain’s reaction function, Escrivá and Santos (1991) find that, prior to 1986, the exchange rate was not a variable systematically affecting policy-making. They also establish that the exchange rate against other EC currencies had a consistent influence on the interest rate thereafter and that, from mid-1987 on, the authorities’ policy response was essentially governed by the behaviour of the peseta-Deutsche mark rate.
If it is assumed that the peseta would have been realigned in parallel with the French franc (implying devaluations of 6 and 3 percent against the Deutsche mark in April 1986 and January 1987), the peseta would have appreciated against its upper ceiling in late 1988; see Viñals (1990, pp. 209-210).
Escrivá and Malo de Molina (1991) underscore this signalling role of the money targets.
Ungerer et al. (1990, p. 66) show that the variability of the United Kingdom’s effective exchange rate, in both nominal and real terms, was higher than that of any other of the then EC countries and about twice their average during 1975–78 and 1979–89 (the ERM period before UK entry).
Particularly influential was the critique of the MTFS program by Buiter and Miller (1981), who modelled how the government’s tight money approach was leading to a transitory real appreciation and a decline in output below potential capacity. Their analysis of exchange rate overshooting was based on the standard proposition that, with liberal capital flows and a floating exchange regime, restrictive monetary policy is immediately reflected in the nominal exchange rate, whereas, owing to wage and price inertia, it only affects domestic costs with a lag.
Artis (1990) provides a verbatim chronicle of the increasing emphasis on the exchange rate in the evolving MTFS.
Thatcher (1993) and Lawson (1993) provide first hand accounts of their fundamental discord.
Townend (1991, p. 210).
George (1998, p. 180).
At this summit, the Delors Report was certified as a starting point for the architecture of EMU. The conditions to be met for ERM entry were specified as: a lower inflation rate in the United Kingdom; the removal of exchange controls within the ERM; sufficient progress towards the single EC market including the liberalisation of financial services; and agreement on competition policy.
The announcement to enter ERM in combination with a lowering of interest rates was politically timed on the eve of the annual Party Conferences. Prime Minister Thatcher insisted on the rate cut against the advice of the Treasury and the Bank of England, and placed heavy emphasis on it in presenting the decision to join the ERM. It is more than curious that a Prime Minister entering an exchange rate commitment should point at money supply figures as a sign that inflation will decline in the near future: “I for my part was determined to demonstrate that we would be looking more to monetary conditions than to the exchange rate in setting interest rates.” See Thatcher (1993, p. 724).
Buiter, Corsetti and Pesenti (1998) provide a detailed reconstruction of the 1992–93 ERM crises.
George (1998, p. 181) concedes that the Bank of England cannot explain the marked appreciation of sterling against European currencies in the second half of the 1990s, even when taking account of interest rate differentials. This seems to echo the conclusion of Hacche and Townend (1981) who dramatically failed to find a stable relationship between exchange rate developments during 1972–80 and the main (monetary and other) variables which theoretical considerations suggest may be its chief determinants.
The system of credit ceilings remained in place until 1990 and was mostly supplemented by administratively set interest rates. These combined regulations not only served a monetary function, but also levied an implicit tax on the banking system that represented a major source of financing for the large public sector deficits during this period.
On the basis of a relatively high ratio of (short-term and overall) capital flows to GDP as well as of onshore/offshore interest rate differentials, Torres (1990) concludes that Portugal’s capital controls were largely ineffective from mid-1986 on.
The new currency basket comprised the Deutsche mark, the Spanish peseta, the French franc, the Italian lira and the British pound; the weight of the Deutsche mark was adjusted upwards to reflect the importance of trade with the Benelux countries.
The comprehensive inflation assessment was not published as a separate inflation report but was included in the central bank’s quarterly Economic Bulletin and updated after a semi-annual interval; see the March and September issues of Banco de Portugal (1997).
On the favourable Greek attitude to European objectives for greater exchange rate stability, as well as the scant Greek contribution to these objectives owing to the dominant national political business cycle, see Papademos (1990).
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Houben, A.C.F.J. (2000). The Evolution of Exchange Rate Targeting. In: The Evolution of Monetary Policy Strategies in Europe. Financial and Monetary Policy Studies, vol 34. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-4471-5_6
Download citation
DOI: https://doi.org/10.1007/978-1-4615-4471-5_6
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-7014-7
Online ISBN: 978-1-4615-4471-5
eBook Packages: Springer Book Archive