Concluding Comments: The SNB's Monetary Policy Framework Ten Years On

Kohli, Ulrich Article Concluding Comments: The SNB's Monetary Policy Framework Ten Years On Swiss Journal of Economics and Statistics Provided in Cooperation with: Swiss Society of Economics and Statistics, Zurich Suggested Citation: Kohli, Ulrich (2010) : Concluding Comments: The SNB's Monetary Policy Framework Ten Years On, Swiss Journal of Economics and Statistics, ISSN 2235-6282, Springer, Heidelberg, Vol. 146, Iss. 1, pp. 425-430, https://doi.org/10.1007/BF03399321 This Version is available at: https://hdl.handle.net/10419/185955 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/ © Swiss Society of Economics and Statistics 2010, Vol. 146 (1) 425–430 a Visiting Professor, University of Geneva; [email protected]. At the time of the conference, Ulrich Kohli was Chief Economist and Alternate Member of the Governing Board of the Swiss National Bank. The SNB’s Monetary Policy Framework Ten Years On: Concluding Comments Ulrich Kohlia It is obviously not a simple task to summarize, in real time so to speak, the findings of two days of intense discussion. I believe that we have learnt a great deal about the Swiss National Bank’s monetary policy strategy and related issues, and I would like to take this opportunity to thank all the participants for their contributions. Of course, I am especially grateful to Stefan Gerlach and Peter Kugler who first came up with the idea of this conference, to Marcel Savioz who coordinated it from the inside, and, last but not least, to Martina Oswald who was responsible for most of the organization. Let me try to briefly recap some of the main conclusions regarding our monetary policy strategy and touch on some still open issues. All in all, I think that we can conclude that the new monetary policy strategy has been very successful. It has served us – and Switzerland – well over the past ten years. It has proven flexible enough to deal with several major crises without having to compromise or depart from it. The performance in terms of price stability over the past ten years has been remarkable, and the volatility of inflation and of real growth has been lower than over previous decades. This is certainly the message conveyed by the papers of Jordan, Peytrignet and Rossi, and of Genberg and Gerlach. It might even be that the new monetary policy concept has contributed to lowering the natural rate of unemployment, as argued by Kugler and Sheldon. The SNB has also convincingly demonstrated its ability to control the Libor, which is an essential part of its strategy (Nautz, Offermanns and Abbassi; Moser). Is it inflation targeting? Many different opinions were expressed on this question. Some (Kugler and Sheldon; Posen) say it is, others (Jordan et al.) maintain it is not, and some are uncertain (Genberg: “it walks like a duckling, it quacks like a duckling, but it may be a swan”). Although inflation targeting can mean different things to different people, the SNB – just like the European Central 426 Ulrich Kohli 1 See Kohli (2008b). Bank (ECB) – does not view itself as an inflation targeter. Of course, if inflation targeting means that one believes that the main goal of monetary policy is to achieve and maintain a low rate of inflation, then we are all inflation targeters. Indeed, in this sense, the SNB has almost always been an inflation targeter, for most of its 102 year history. There are some important differences between the SNB’s strategy and the framework of self-declared inflation targeters such as the Reserve Bank of New Zealand, the Bank of England, or the Bank of Canada, though. Jordan et al. as well as Genberg and Gerlach have listed some of these, including the fact that the SNB does not target a numerical value for inflation, and that the SNB has never made any commitment as to the time it would take to return to price stability should some deviation occur. As pointed out by these authors, and also by Goodfriend, such flexibility is only possible if the central bank is highly credible. I also would like to add that the term “inflation targeting” is rather unfortunate, for it can convey the wrong impression to the public that the aim of monetary policy is to produce inflation, rather than price stability! The name of the game should be achieving and maintaining price stability.1 For the same reason, I find the title “inflation report” rather awkward and defeatist. I much prefer the name that we gave to our own document, namely “monetary policy report”. More importantly, inflation targeting usually means that the central bank is given a target by the government or the parliament – which is not the case in Switzerland since the SNB’s legal mandate of price stability is set once and for all –, but it also suggests that the target can be modified from time to time to suit the circumstances. While it might make perfect sense to map the course towards price stability when inflation is initially high, a moving target becomes counter-productive once that the country has graduated from the price stability class, for the framework then contains an element of discretion – admittedly exercised by an external body rather than by the central bank, but discretion nonetheless – which might raise time inconsistency issues. These difficulties are absent from our concept as it now stands. Another topic that has been addressed by a number of authors, and that keeps creeping back concerns the definition of price stability and the optimal rate of inflation. Let me stress that these are two separate issues. There is a large literature on the optimal rate of inflation, starting with Friedman’s contribution, and the recommendations are quite diverse, ranging from a modest negative rate to low positive ones. The question of the definition of price stability is quite a different one. Since the SNB is required by law to maintain price stability, the question The SNB’s Monetary Policy Framework Ten Years On: Concluding Comments 427 of the optimal rate of inflation is not on the table. What is, on the other hand, is how price stability is defined. The definition that the SNB has adopted is identical to the one of the ECB, namely an annual rate of inflation of less than 2% over the medium term. Given measurement errors, unrecorded quality changes, and the well-known substitution bias of the Laspeyres functional form, one can argue that an inflation rate of less than 2% is compatible with price stability. Nonetheless, one should not underestimate the destructive power of even low inflation rates. As pointed out by Jordan et al., a 2% yearly inflation rate means that the price level doubles in 35 years; it is multiplied by 7 over the course of a century. With a 3% inflation rate, the price level doubles in less than 24 years, and the price level is multiplied by nearly 20 within a century! In other words, over the course of a human life, the monetary unit would lose nearly 95% of its value, so that the residual worth of one dollar or one euro would be little more than five cents, a far cry from price stability. In any case, a change in the definition of price stability could only be justified if new objective, scientific findings regarding the measurement of the price level and of its rate of change became available, and if these truly warranted a redefinition of price stability. The definition cannot be changed simply as a matter of convenience or for operational purposes. Doing so would gravely endanger the credibility of the SNB. In my opinion, the SNB thus has no degree of freedom in this area. Given the smallness and the openness of the Swiss economy, its inflation rate is quite volatile. For this reason it is not possible to aim at an inflation rate close to, but just below the upper limit of the inflation range compatible with price stability. For instance, if the SNB targeted an inflation rate of 1.9%, one can be almost certain that the resulting inflation rate would be in excess of 2% nearly half the time, which would undermine its credibility. The experience of the ECB, with an average inflation rate of 2.1% since 2004, (i.e. outside the range it associates with price stability) is quite telling in this respect. Another important issue is what kind of interest-rate hypothesis the SNB should make when producing its inflation forecast, and whether the interest rate path should be published. Jordan et al. gave some compelling reasons why an inflation forecast explicitly based on a constant interest rate makes sense. The public understands its meaning. This is like chartering a course when sailing. If a vessel is heading for a lighthouse, this does not mean that its skipper intends to hit it. On the contrary, it suggests that a change of course will have to intervene. It follows that the SNB forecast – it should perhaps be called a projection – is not directly comparable to conventional forecasts. As mentioned by Savioz, the SNB consensus forecast is not meant to be the best possible predictor of future inflation, but it is a good predictor of future interest rate changes! Jordan et al. 428 Ulrich Kohli 2 See Kohli (2008a, 2008b). 3 See Taylor (2008). For more on this topic, also see Cùrdia and Woodford (2008). gave the pros and cons of using a market rate as a basis for the inflation forecast. A third option would be to use a model-based interest rate, one that perhaps even reflects the policy makers’ preferences. The difficulties I see there are at least threefold: (i) if there are more than one policy maker they would have to agree on the objective function; (ii) if one uses more than one model, it is not clear that the consensus inflation forecast would be coherent with a consensus forecast of the interest rate path; (iii) to the extent that the interest rate expected by the policy makers differs from market expectations, observed asset prices, which tend to enter the models as well, may be inconsistent with the projected interest rate path. Having said this, I am happy to concede that these questions deserve to be further investigated. The financial crisis has also brought forward some further important issues that had received little attention in the past. One such example is the difference between the ECB procedure and the SNB strategy (Jordan et al; Nautz et al; Martin). Whereas the ECB essentially fixes the overnight risk-free rate, the SNB, by targeting the 3-month Libor, is led to compensate any variations in the risk premium by adjusting the repo rate. This is what I have dubbed elsewhere the automatic monetary stabilizer mechanism.2 John Taylor has also referred to the Swiss procedure as being very helpful in the current circumstances, since there is no need then to make any ad hoc adjustments to the Taylor rule.3 Yet another question that requires more thinking concerns the use of the Libor as an operating target. Indeed, as asked by Jordan et al., is it the appropriate operating target? Given that there were hardly any transactions in the unsecured interbank market during the crisis one may wonder what the meaning of the Libor really is. Does the SNB target something that does not exist? Maybe, but I am not sure that this is really a problem. Price stability, after all, is defined in terms of something that does not exist either, namely the consumer price index (CPI). The CPI is not a market price, but rather an index, an indicator. Many market interest rates, including some of the most meaningful ones, are indexed on the Libor, so this suggests that the Libor is indeed important. Maybe, though, we should look at things the other way around. Rather than thinking of the mortgage rate – or any other market rate – as Libor plus, say, 75 points, it might be better to consider the Libor as the mortgage rate minus 75 points! In that sense, the Libor would be an indicator of general market conditions, an index so to speak. In this context, I find it surprising that the mark-ups – or agios – charged on mortgages relative to The SNB’s Monetary Policy Framework Ten Years On: Concluding Comments 429 4 See Assenmacher and Gerlach (2008), for instance. 5 See Angeloni and Faia (2009). Libor do not seem to have changed much during the crisis. Indeed, I would have expected these mark-ups to decline as the risk premium increased since mortgages are secured loans. Consequently, if we think of the Libor as an index of general market conditions, rather than an actual transaction price on unsecured loans, it may not fully incorporate the risk premium. In that case, the spread between the Libor and the overnight indexed swap (OIS) would be a poor measure of the risk premium. I gladly concede that these thoughts are rather speculative at this stage, which suggests that more research is needed. One further matter brought up by Posen is the question of price level targeting. Should bygones be bygones? There is a lot to be said in favor of price level targeting, possibly with a drift. In particular, if it is credible, it can add an element of stability in case of a prolonged deflation. Moreover, it seems indisputable that the mandate of price stability, taken literally, is best fulfilled if the central bank offsets past deviations and thus does indeed keep the price level stable! The Bank of Canada has initiated a research program that encompasses this question, which deserves to be pursued by others as well. One major issue that still needs considerable research concerns the place of financial stability considerations in monetary policy decisions, and the whole leaning against the wind argument (Rich; Posen). One can argue that the SNB monetary policy strategy, by relying on a three-year consensus inflation projection, already takes account of developments in asset markets, particularly so since several of its forecasting models attach special attention to the behavior of monetary aggregates and credit. Moreover, recent economic research demonstrates that in order to have a significant impact on asset prices, monetary policy would have to be extremely aggressive, thereby endangering economic activity.4 This would amount to killing the patient while overcoming the disease! There is also evidence that regulation, rather than interest rate policy, is best suited to tame financial imbalances.5 Finally, one must not forget that policy makers are not necessarily better than market participants at identifying developing asset market bubbles, and that asset prices do not necessarily all move in the same direction, which begs the question as to which one or which ones the central bank should react to. I, for one, would argue that if there is one asset price that the central bank should target, then it should be the price of the asset closest to it, i.e. money. That is, target the real price of money, namely 1/P, which simply amounts to aiming at price stability! Having said this, it is clear that the SNB is forward looking, and 430 Ulrich Kohli that it may take interest rate steps before price pressures are discernible: this is why it uses a three-year inflation projection as its guide. During the past ten years, the Swiss economy has gone through nearly two full cycles. As argued earlier, the SNB’s monetary policy strategy has served it well. Of course, the current conditions are quite extraordinary. Is the hardest part still to come? The exit from an exceptional situation must still be mastered. Pretty soon, the SNB strategy will become a teenager. A parent would see this coming phase with a certain amount of apprehension and anxiety. However, I believe that the SNB has a very steady hand, and that it can therefore view this next chapter with confidence and serenity. References Assenmacher-Wesche, Katrin, and Stefan Gerlach (2008), “Ensuring Financial Stability: Financial Structure and the Impact of Monetary Policy on Asset Prices”, IEW Working Paper 361, Institute for Empirical Research in Economics, University of Zurich. Angeloni, Ignazio, and Esther Faia (2009), “A Tale of Two Policies: Prudential Regulation and Monetary Policy with Fragile Banks”, Kiel Working Papers 1569, Kiel Institute for the World Economy. Kohli, Ulrich (2008a), “Comments“, presented at the press conference on the occasion of the IMF Article IV Consulation of Switzerland, Berne, March 17. Kohli, Ulrich (2008b), “Is Inflation Targeting Passé?“, presented at the Conference in honor of Alexandre Swoboda, Graduate Institute of International Studies, Geneva, May 30. Cùrdia, Vasco, and Michael Woodford (2008), “Credit Frictions and Optimal Monetary Policy”, presented at the 7th BIS Annual Conference, Lucerne, Switzerland, June 26–27. Taylor, John (2008), “Monetary Policy and the State of the Economy”, testimony before the Committee of Financial Services. U.S. House of Representatives, February 26.