Swiss monetary policy as viewed by the European Central Bank

Speech by Jean-Claude Trichet,
President of the ECB
on the occasion of the 100th anniversary
of the Swiss National Bank on 22 June 2007 in Zurich

Madame la Présidente de la Confédération suisse,

Monsieur le Président du Gouvernement du Liechtenstein,

Ministers, Excellencies,

Dear President of the Bank Council of the Swiss National Bank,

Dear Jean-Pierre Roth, Dear Chairman of the Governing Board of the Swiss National Bank,

Dear fellow Governors,

Ladies and gentlemen,

It is an immense pleasure for me to be here today with such a distinguished audience to celebrate the 100th anniversary of the foundation of the Swiss National Bank (SNB). The SNB is to be especially congratulated because, in spite of the turbulences which have characterised the period since 1907 – two world wars, the Great Depression, and the Great Inflation – it has succeeded in delivering, on average, a remarkably low inflation rate. I regard a historical and international perspective on Swiss price stability as being especially appropriate.

A historical and international perspective: a low-inflation country

It has been noted that, notwithstanding the Great Inflation episode, and the temporary inflationary outbursts corresponding to the two world wars, the United States should correctly be characterised, when seen from a very long-run perspective, as a low-inflation country. This is a claim made, in particular, by Bradford DeLong in his analysis of the US inflationary experience since the time of the Civil War. [1]

Historical experience suggests that Switzerland deserves such a characterisation to a significantly greater extent. First, since 1880 Swiss annual inflation has been, on average, just 2.2%, to be compared to 2.6% in the US. Second, in an international comparison with 13 other OECD countries since 1880, and excluding from the computation the years corresponding to the First and Second World Wars and their immediate aftermath (in order to avoid a distortion of the results through the extraordinary turbulence associated with the two world wars), Switzerland comes out on top once again, with an annual average inflation rate of just 1.5%. Finally, focusing on the period following the creation of the Swiss National Bank, and excluding, once again, the years of the two world wars and their immediate aftermath, Switzerland, with an average annual inflation rate of 2.1% is, together with Germany, again the best performer.

Switzerland’s consistent “credible currency, low-inflation” strategy clearly emerged during the episode which can reasonably be characterised as the Swiss National Bank’s “finest hour”: the Great Inflation of the 1970s. As pointed out by Edward Nelson [2],

“[i]n Switzerland, […] while a debate on the nature of inflation did take place in the early 1970s, it was resolved quite rapidly in favor of the monetary view, leading to an eschewing of non-monetary approaches to inflation control”.

In subscribing to the ultimately monetary nature of inflation, the Swiss National Bank was accepting the lesson of Milton Friedman as remarkably exposed in the Monetary History of the United States [3], the single piece of research which – more than any other – contributed to changing policy making. And indeed, as emphasised by Robert Lucas in the remarks he delivered at Otmar Issing’s Festschrift one year ago [4],

“[t]he first central banks to deal effectively with the 1970s inflation were the Bundesbank and the Swiss National Bank, and both did so in the mid-1970s by deliberate reductions in money growth rates. In October 1979 the US Federal Reserve followed suit […].”

Monetary policy and monetary aggregates

As stressed by Lucas, recalling the Swiss National Bank’s – and the Bundesbank’s, as well as a number of European central banks’, including Banque de France’s –acceptance, back in the 1970s, of the ultimately monetary nature of inflation, and their use of monetary means to bring it under control, is especially important today, when the relevance of monetary analysis for monetary policy is sometimes questioned. As Charles Goodhart sarcastically put it [5]

“[m]onetarism, in any of its guises, has become somewhat unfashionable; and the New Keynesian three-equation synthesis […] rides high. […] Deviate from this, and you are not a member of the In Crowd.”

The reasons why the role of monetary aggregates in monetary policy analysis is today questioned are well known. On the conceptual front, the ascendancy of Neo-Wicksellian theory [6] which assigns to monetary aggregates a purely residual role. At the practical level, the breakdown, in several countries, of the empirical relationship between money growth and inflation, which led the former Governor of the Bank of Canada, Gerald Bouey, to famously quip that “ We didn’t abandon monetary aggregates, they abandoned us”. I have three comments on this.

First, empirical evidence [7] shows that – since the first half of the nineteenth century, and in a number of countries – fluctuations in trend money growth have almost always led to fluctuations in trend inflation, with upswings or downswings in trend money growth having been followed, a few years later, by upswings or downswings in trend inflation. The crucial point to stress here is that the resilience of the lead of trend money growth over trend inflation over time, and across radically different monetary regimes, clearly suggests that such a lead is a structural feature of the economy – that is, embedded in the structure of the system so as to be invariant to occasional changes in the monetary regimes. This empirical finding represents a bedrock fact that any model – including those built on New-Keynesian synthesis principles – has to confront and seek to reproduce. Since, over the last two centuries, fluctuations in trend money growth have almost always been followed by fluctuations in trend inflation in the same direction, an upswing in trend money growth points to the existence of upward medium to long term risks to the inflation profile. In the light of the evidence coming from the last two centuries of data, totally ignoring trend movements in money growth would therefore entail – in my view – excessive and unreasonable risks. The practice of the SNB – and of the ECB – of closely monitoring monetary trends in order to identify long-term risks to price stability is therefore validated by the monetary history of the last 200 years.

Second, as recently stressed, in particular, by Philippe Hildebrand at a central banking conference at the ECB [8], prolonged periods of low and stable inflation – like those that OECD countries have been experiencing for several years now – can give rise to distortions in empirical analyses, and can make it extremely difficult to correctly assess the informational content of a specific indicator of future inflation. To put it differently, in an equilibrium in which the central bank is successful at keeping inflation low and within a narrow range, the empirical relationship between inflation and any indicator tends to disappear. This is precisely why the experience of the Great Inflation – which can be regarded as the ultimate off-equilibrium event – is important. That persistent and prolonged deviation from equilibrium revealed, in a fundamental sense, the authentic nature of reality, which today, in a stable, low-inflation world, has become more difficult to extract from the data. And so, in the same way as the experience of the Weimar Republic hyperinflation was the key “founding memory” for the Bundesbank, or the dramatic experience of the revolutionary “assignats” for Banque de France, [9] the experience of the Great Inflation can be regarded by today’s central banks as the cornerstone of their institutional architecture. Indeed, as pointed out by the Vice Chairman of the Federal Reserve, [10] “[g]reater support for independent central banks […] was one of the results of the lessons of the great inflation of the 1970s.”

Third, it is important to note that the link between money growth and inflation can be blurred by systematic fluctuations in the velocity of money. Recent work carried out at the Swiss National Bank [11] has shown that, after controlling for systematic shifts in the equilibrium component of money velocity caused by Fisherian movements in the opportunity cost of money, the long-run relationship between money growth and subsequent inflation is still in the data, and not only for the euro area, but also for the United States, where the stability of such a relationship has been most strongly questioned in the past. This finding suggests that monetary indicators should not be hastily discarded when their link with inflation appears to have weakened. Rather, it suggests that monetary data should be subjected to a thorough and broad-based analysis, in order to filter out the noise and to fully understand movements in trend velocity which distort their low-frequency signal for long-term risks to price stability.

The new Swiss monetary policy concept: an ECB perspective

From 1974 until the end of the century, the Swiss National Bank consistently pursued a strategy of targeting monetary aggregates, announcing targets first for M1, and then, from 1980, for the monetary base. (Only during a very brief period in 1978-1979 did the SNB abandon monetary targeting in favour of an exchange rate target.) Owing to the increasing instability of the monetary base over the second half of the 1990s, however, the SNB decided, at the end of 1999, to announce a new monetary policy framework. The new framework is based on three key elements:

  • An explicit definition of price stability “[…] expressed in terms of the headline CPI. [The SNB] defines price stability in the same manner as the European Central Bank, [as] a CPI inflation rate of less than 2 per cent per year.” [12]

  • An inflation forecast as the main communication tool in order to explain to the public the rationale behind monetary policy decisions.

  • From an operational point of view, a range for the three-month LIBOR rate as the way to actually implement policy decisions in practice.

I find from time to time statements suggesting that, with its new monetary policy strategy, the SNB has essentially renounced its monetarist past. I believe such a position to be incorrect, and in what follows I will make two main points. First, I will stress the significant elements of continuity with the pre-2000 monetary policy strategy. Second, I will argue that, if one looks carefully at the details, the monetary policy strategy of the Swiss National Bank is, in many respects, quite close to that of the European Central Bank. Let me elaborate on these points in turn.

As stressed by SNB staff members [13], the SNB

“[…] retained important ingredients of monetary targeting in its new monetary policy concept. At various occasions, the SNB stressed that it continues to monitor two sets of indicators providing leading information on future price developments […]. The first set of indicators is useful for forecasting short-run price developments, i.e over a horizon of one and a half to two years. It includes various indicators of the cyclical state of the economy, notably the output gap and supply and demand conditions in the labour market, as well as the real exchange rate of the Swiss franc. The second set of indicators comprises the monetary aggregates, which provide useful leading information on long-run price developments.”

The elements of continuity with the pre-2000 strategy emerge quite clearly, in particular in the form of the important role assigned to a deep analysis of monetary aggregates. The justification for closely monitoring monetary aggregates, their components, and their counterparts, is the same as that given by the European Central Bank, namely the information content they comprehend as regards inflation in the long run, which has been documented for both Switzerland and the euro area. Precisely because of this long-run link with inflation, and of the previously mentioned lead of trend money growth over trend inflation, a strong upward trend in money growth points to the existence of upward risks to the profile for trend inflation. When the ECB started raising rates in December 2005, for example, monetary analysis, and a close monitoring of trends in a number of money supply indicators, played a crucial role, by highlighting risks to the inflation profile which were not, at the time, immediately apparent just from the economic analysis. As the “Financial Times” [14] recently pointed out, the decision to raise rates, at that time “controversial”’, has been, 18 months later, “vindicated” by subsequent events.

The SNB’s differentiation between the two groups of indicator variables – on the one hand, short to medium-run indicators, such as a range of statistics including, prominently, inflation itself; on the other hand, monetary indicators, mostly relevant at a longer horizon – is clearly reminiscent of the ECB’s “two-pillar” strategy. And research undertaken at the SNB [15] has indeed shown such a dichotomy to provide a good characterisation of short-term and medium to long-run inflation dynamics for both Switzerland and the euro area.

A further important element common to both the SNB and the ECB is that neither central bank adjusts its policy by mechanically reacting to deviations of the inflation forecast from their respective definitions of price stability. Rather, both central banks focus on low-frequency – that is, trend – movements in the variables of interest, and both are willing to see through the shocks that buffet the economy occasionally from a medium-run perspective if overall macroeconomic circumstances require it.

There are, of course, also differences between the monetary policy strategies of the ECB and the SNB. For example, as I have already mentioned, at the SNB monetary information is taken into account in the construction of the inflation forecast, while at the ECB the results from the monetary analysis are used by the Governing Council to cross-check – on the basis of an expanding range of monetary tools – the signals coming from the economic analysis. In a sense, the SNB delegates the “cross-checking” activity to the projection process, whereas the ECB assigns that responsibility to the Governing Council, once the projection process has run its course independently. At that point, we cross-check the outputs coming from models and formal frameworks, often from within a variety of non-nested alternatives. We adopt something similar to what my friend Alan Greenspan would call a Bayesian approach to policy and Alan Blinder would describe as a rough approximation of optimal information processing: “Use a wide variety of models, and don’t ever trust any one of them too much.” [16]

I view these differences as minor. Our two central banks conduct similar exercises in robust policy-making within a shared framework of common principles: a clear definition of price stability, a medium-term policy orientation, and an important role assigned to monetary indicators for the assessment of longer-term risks to price stability.

Conclusions

Permettez-moi en conclusion, cher Jean-Pierre, cher Président, de souligner trois points.

D’abord, il est particulièrement émouvant pour l’ancien gouverneur de la Banque de France que je suis, d’être à Zurich. Nous avions célébré le 200ième anniversaire de la Banque de France, il y a 7 ans, en rendant un vibrant hommage à un citoyen suisse. La Banque de France a été créée par Napoléon en 1800 sur une proposition du financier Jean-Frédéric Perregaux, qui en a été ensuite le responsable. Monsieur Perregaux était l’un de ces financiers suisses, à la fois très imaginatifs, très sages et très prudents, dont la réputation est légendaire à travers les siècles.

Anschließend möchte ich hervorheben, wie vorbildlich die Beziehungen sind, die alle Zentralbanken der Welt mit der Schweiz sowie der Schweizerischen Nationalbank unterhalten. Die Offenheit der SNB und ihre aktive Beteiligung an geldpolitischen Fragestellungen und der geldpolitischen Forschung auf europäischer und internationaler Ebene werden sehr geschätzt. Jean-Pierre Roth, dem wir unsere regelmäßigen Zusammenkünfte in Basel zu verdanken haben, leistet als Präsident des Verwaltungsrats der Bank für Internationalen Zahlungsausgleich hervorragende Arbeit. Ihm möchte ich heute meine besondere Anerkennung aussprechen.

Finally, let me particularly emphasise that price stability is a necessary condition for sustainable growth and sustainable job creation. I made the point recently several times that since the creation of the Euro on the 1st of January 1999, during the first eight years, more than 12 million jobs were created in the euro area. I underlined it could be compared with less than 3 million during the eight years before the Euro. I also stressed that the current level of unemployment, at 7.1%, is the lowest for 25 years. Can I also stress here in Switzerland that the sound and wise monetary policy of the Swiss Central Bank and the level of price stability which has been attained throughout the last hundred years are one of the major causes of the full employment success of your country.

I thank you for your attention.

References

Assenmacher-Wesche, Katrin, and Gerlach, Stefan (2006a), ‘Interpreting Euro Area Inflation at High and Low Frequencies’, BIS Working Paper No 195, February 2006

Assenmacher-Wesche, Katrin, and Gerlach, Stefan (2006b), ‘Money Growth, Output Gaps and Inflation at Low and High Frequency: Spectral Estimates for Switzerland’, Swiss National Bank Working Paper 2006/5

Benati, Luca (2007), ‘Long-Run Evidence on Money Growth and Inflation’, European Central Bank, mimeo

Blinder, Alan (1998), “Central Banking in Theory and Practice”, Cambridge, Mass. The MIT Press

DeLong, Bradford J. (2000), ‘America’s Historical Experience with Low Inflation’, Journal of Money, Credit, and Banking, Vol 32(4), pp. 979-993

Friedman, Milton, and Schwartz, Anna Jacobson (1963), ‘A Monetary History of the United States, 1867-1960’, Princeton University Press for the National Bureau of Economic Research

Goodhart, Charles (2006), ‘Otmar Issing and the Monetary Pillar’, remarks delivered at ‘Monetary Policy: A Journey from Theory to Practice - An ECB Colloquium Held in Honour of Otmar Issing’, Frankfurt am Main, 16 - 17 March 2006

Hildebrand, Philippe (2006), ‘Comments on Fischer et al.’s paper’, delivered at the fourth ECB Central Banking Conference, ‘The Role of Money: Money and Monetary Policy in the Twenty-First Century’, Frankfurt am Main, 9 - 10 November 2006

Kohn, Donald (2005), “Central Bank Communication”, remarks by Governor Donald L. Kohn at the Annual Meeting of the American Economic Association, Philadelphia, Pennsylvania, January 9, 2005

Lucas, Robert E., Jr. (2006), ‘Remarks’ delivered at ‘Monetary Policy: A Journey from Theory to Practice - An ECB Colloquium Held in Honour of Otmar Issing’, Frankfurt am Main, 16 - 17 March 2006

Jordan, Thomas J., Peytrignet, Michel, and Rich, George (2000), ‘The Role of M3 in the Policy Analysis of the Swiss National Bank’, paper presented at the Central Bank Workshop on Monetary Analysis organised by the European Central Bank, Frankfurt am Main, 20 - 21 November 2000

Nelson, Edward (2007), ‘Ireland and Switzerland: The Jagged Edges of the Great Inflation’, Federal Reserve Bank of St. Louis Working Paper

Reynard, Samuel (2006), ‘Money and the Great Disinflation’, Swiss National Bank Working Paper 2006/7

Sargent, Thomas J., and Velde, Francois R. (1995), “Macroeconomic Features of the French Revolution”, Journal of Political Economy, Vol 103(3), pp. 474-518.



[1] See DeLong (2000).

[2] See Nelson (2007).

[3] See Friedman and Schwartz (1963).

[4] See Lucas (2006).

[5] See Goodhart (2006).

[6] Exemplified in the work of Michael Woodford.

[7] See Benati (2007).

[8] See Hildebrand (2006).

[9] For an analysis of the revolutionary experience with the ‘assignats’, see Sargent and Velde (1995, Section IV, ‘Rise and Fall of the Assignat’).

[10] See Kohn (2005).

[11] See Reynard (2006).

[12] See Jordan, Peyrignet and Rich (2000).

[13] See Jordan, Peyrignet and Rich (2000).

[14] See “Under fire, on target”, by Ralph Atkins and Lionel Barber, “Financial Times”, 18 May 2007, page 9.

[15] See Assenmacher-Wesche and Gerlach (2006a, 2006b).

[16] See Blinder (1998, page 12).

Media contacts