## Jean-Pierre Roth: Autonomy on the edge of a large monetary zone - the Swiss experience
Address by Mr Jean-Pierre Roth, Chairman of the Governing Board of the Swiss National Bank, at the Leonard Davis Institute for International Relations, Hebrew University, Jerusalem, 28 May 2006.
This paper was prepared with the support of Ms Petra Gerlach-Kristen, Economic Research Section, Swiss National Bank, Zurich.
* * *
This is my second visit to Israel and I am delighted to be back. I am especially happy to speak at this institute which, according to the will of its founder, Leonard Davis, should promote a better understanding of Israel external policy. The first time I was in Jerusalem was in November 2004 when the Bank of Israel celebrated its 50th anniversary.
The Swiss National Bank is itself approaching an important anniversary. Next year, it will have been 100 years since we opened our doors for business. Since 1907, our institution has seen large changes in the monetary environment. When the SNB began operating, the Swiss franc's parity was defined in terms of gold and silver. In the early 1920s, Switzerland adopted a gold standard and fixed exchange rates remained the dominating feature of its monetary regime until the collapse of the Bretton Woods system. Since then, we have learned how to conduct our monetary policy independently and how to live with flexible exchange rates. Our policy was guided by monetary targets until the end of the 1990s, and only recently, like most central banks, have we shifted our focus to price stability more directly.
I would like to share with you some of the lessons we have learned from that period with flexible exchange rates and, in particular, review our recent experience of living on the edge of the euro area.
## Living on the edge of the euro area
As you can imagine, the introduction of the euro in January 1999 constituted a sort of revolution in our monetary environment. For the first time in our history, we are now surrounded by a single currency area, which also happens to be the destination of 60% of our exports. In fact, our situation, in terms of economic power, can be compared to that of the State of Massachusetts in the United States. Can you imagine the State of Massachusetts having its own currency? The natural question is: How do we come to the conclusion that monetary independence is still desirable?
Indeed, given our geographical position, we, Swiss, frequently have to explain why we decided not to adopt the euro or not to peg our currency to the euro, as Denmark has done. The first question I would like to address today is therefore: What are the advantages for Switzerland of retaining its national currency, as opposed to adopting the euro? I will then look at the conditions that are necessary for successful monetary autonomy. Finally, I will turn to the experience of what is now seven years of coexistence with the euro area.
While my focus is naturally on Switzerland, I would like to emphasise that we are not the only ones living on the edge of the euro area. Norway, Sweden and the UK conduct independent monetary policies; most Central and Eastern European countries are eager to join the euro area, but have not been admitted yet, and, of course, Israel is not a member state either.
In fact, Israel and Switzerland share a large number of similarities in their relations to Europe.
Let me start with the obvious. We are both small open economies. Our populations are roughly the same, with around seven million people. Our economies are service-oriented, with about two-thirds of GDP originating in that sector. Moreover, Israel and Switzerland are both export-oriented, with Israeli exports amounting to 44% of GDP and Swiss exports accounting for 49%. One-quarter of our countries' exports are services, three-quarters are goods, most of which are manufactured goods, especially capital goods. Granted, the greatest part of Israeli exports, roughly 40%, goes to the US, while the largest share of Swiss exports, around 60%, goes to the EU. Due to our geographical position and a long history of cross-border activity, it is understandable why we are more Europedependent than Israel. Nevertheless, Europe is a critical trading partner for your country, too, absorbing almost a third of Israeli exports.
BIS Review 46/2006
Consequently, economic fluctuations in the euro area are important for both Israel and Switzerland. But in the Swiss case - and this is a kind of paradox for a country that does not belong to the EU - the relations are especially close: Switzerland shows a higher degree of cyclical integration with the euro area than the individual euro members themselves! Indeed, over the last five years, Swiss GDP growth has shown a correlation of 86% with economic growth in the euro area, while within the euro area, growth rates have had a correlation of only 65%. Israel's growth has a correlation with euro area growth of 62%.
The deep integration of Switzerland in the European markets is not a surprise, since we are located at the heart of the Union, next to its largest members, but also because we have signed several bilateral agreements with the EU giving us free access to the European goods and labour markets. Israel's situation is quite different in that respect, but it remains a fact that, in both countries, fluctuations on the real side are closely linked to those in the euro area.
Regarding the level of growth, Israel, as a relatively young country with a growing population, has outperformed both the euro area and Switzerland, with an average growth over the last five years of 2% versus 0.8% and 0.6% respectively. Nevertheless, Israeli GDP per capita is still only roughly half of that in Switzerland.
On the nominal side, our two countries' links with the euro area differ markedly. Swiss and euro area inflation have a correlation of 39%, which, although low, corresponds to the average correlation of inflation rates within the euro area itself. Israel's inflation shows a weaker link here, displaying a correlation with euro area inflation of only 19%. Comparing again the levels instead of the correlations, Switzerland has a long tradition of monetary stability with an inflation rate that has been clearly lower than those in Israel and the euro area. Over the last five years, the average inflation rates were 0.8% in Switzerland, 2% in Israel and 2.2% in the euro area. While for Switzerland, low and stable inflation is not particularly noteworthy, it is important to remember that price stability in Israel is a recent achievement and that inflation remains relatively volatile here.
Overall, this quick look at the data suggests that Israel and Switzerland have different records in terms of stability, that both countries are highly integrated into the euro area, but that, for obvious reasons, Switzerland seems to be more "European" than Israel. If Switzerland is indeed as integrated as any regular euro area member, what advantages do we see in maintaining our monetary independence, while our neighbours have abandoned their national currencies in favour of the euro?
## Reasons for monetary independence
As you well know, the large majority of European countries have come to the conclusion that adopting the euro is beneficial. For them, one of the major advantages of joining a monetary union is stronger economic growth thanks to lower transaction costs and reduced exchange rate risk. Moreover, for countries that have historically experienced high and volatile inflation and thus high interest rates, adopting the euro brings a fall in nominal and real interest rates, thereby reducing government and private debt burdens.
Given these positive effects, why has Switzerland not embraced the euro or, at least, pegged the franc to the euro?
To my mind, there are four main reasons for this.
- The first is that retaining our own currency, within the framework of flexible exchange rates, gives us the freedom to choose the level of domestic inflation.
- Secondly, monetary independence allows a better reaction to country-specific shocks.
- Thirdly, monetary integration would lead to higher interest rates in our capital market.
- And finally, any tentative move to peg the franc to the euro, as long as Switzerland is not at least in the process of becoming an EU member, would destabilise rather than stabilise our currency.
Let me review these points in detail.
BIS Review 46/2006
## Choice of monetary policy goal
The European Central Bank's main task is to guarantee price stability in the euro area. The ECB achieves this by, in its own wording, "aim[ing] at maintaining [consumer price] inflation below, but close to, 2% over the medium term". The National Bank Act defines the SNB's main task, too, as the achievement of price stability. Our monetary policy concept, which we adopted in 2000, defines price stability, and I quote again, as "a rise in the national consumer price index of less than 2% per annum".
Let me draw your attention to three subtleties here. First, our mandate is to ensure price stability, but we do tolerate some inflation! For both the ECB and the SNB, prices should slightly increase or potentially stagnate, but not decrease over time. Rising prices may not correspond to your idea of price stability, and you may think that even low inflation should be avoided since inflation causes inefficiency. The reason why central bankers think that price stability is achieved even when there is in fact some inflation, is that the better quality of new products and the improvement of existing ones can justify non-inflationary price adjustments.
A second detail is that we do not name one figure as the desirable inflation rate. Rather, we say that we think of price stability as inflation somewhere between 0 and 2%. The reason for this is that uncertainties about the extent of product innovations make it difficult to define a certain inflation rate as optimal. Moreover, given the importance of trade in our GDP, our inflation rate may react rapidly to changes in foreign prices, which should not necessarily trigger a monetary policy reaction. A target band gives us more flexibility in the implementation of monetary policy. No doubt, however, that a point target can also be attractive, as it is clearer and constitutes a stronger anchor for expectations. Given the SNB's excellent track record with regard to price stability - twelve years of inflation below 2% - we can benefit from firmly anchored expectations without having to accept the unnecessary constraints that a point target would imply.
The third subtlety is that, while the SNB defines price stability as inflation of less than 2%, the ECB, which has more or less the same definition, aims for a rate close to 2%, and the Bank of Israel has an inflation target of 1 to 3%. Why is it that the three central banks have different views as to what exactly constitutes price stability?
There are several considerations here, but the most important is the historical perspective. For central banks that have faced relatively high and volatile inflation levels in the past, a comparatively high inflation target is already ambitious. Only once expectations have been anchored, can further steps towards price stability be envisaged. This is true for Israel, where keeping inflation between 2% and 3% is a huge step forward as compared with historical records. For the ECB, which still has to cope with adjustment difficulties (such as higher inflation in countries that are catching up, like Spain or Greece), an inflation rate "close to 2%" seems reasonable.
Monetary independence thus allows a country to choose the level of inflation it finds appropriate.
Monetary independence also helps reduce the variability of inflation and economic conditions, and I will turn to this issue next.
## Reaction to country-specific shocks
The second reason why I think that Switzerland should not adopt the euro is that monetary independence gives us the necessary leeway to react to country-specific shocks. Although I mentioned earlier that the Swiss economy is highly integrated into the euro area, we are not an average European country. Over the last 20 years, Switzerland has become a platform for global activity: while we count 4.2 million jobs within our country, Swiss firms employ another 1.8 million people abroad. Moreover, we are an international financial centre: about 15% of Swiss GDP originates from the financial sector. As a consequence, Switzerland's business cycle is closely related to the development of the world economy and to the situation in financial markets.
Mood swings in international financial markets and in investment decisions impact on the Swiss economy through different channels. On the one hand, they affect our banking and insurance industry as well as our exports. On the other hand, they may trigger a flight into the Swiss franc. The resulting exchange rate volatility is a major risk factor for our trading sector.
Given our openness to capital flows and the importance of our production of capital goods, shocks to the global economy and international financial markets have a larger and faster effect on economic conditions and inflation in Switzerland than in the euro area. Consequently, the optimal path of
BIS Review 46/2006
monetary policy may differ. Monetary independence gives us the capacity to react to asymmetric shocks and to limit their negative impact on the domestic economy.
To sum up my arguments thus far, monetary independence allows Switzerland, first, to live with a comparatively low rate of inflation and, second, to reduce the impact of Switzerlandspecific shocks, which limits the volatility of Swiss inflation and growth. The third reason for monetary independence, which I would now like to turn to, is the protection of the Swiss interest rate advantage.
## Continuity of the interest rate advantage
One major attraction of adopting the euro for countries with a record of high inflation was that national interest rates would drop in the run-up to the monetary union and converge to the German level, which was the lowest among the euro's founding members. Swiss interest rates, by contrast, have historically been lower, even lower than those in Germany.
The reason why Switzerland enjoys the lowest interest rate levels in Europe has nothing to do with the appeal of our bank secrecy, as many believe. Switzerland is not a net importer of capital, but rather a major capital exporter. Traditionally, the Swiss economy displays a large excess of savings over investments. In 2005, our current account surplus represented 14% of GDP and Switzerland exported more capital than the whole euro area.
Another natural reason for the presence of low interest rates is our history of low and stable inflation rates, at least in comparison with other countries. As a result, international investors view Switzerland as a kind of safe haven, which gives rise to an interest rate advantage in nominal as well as in real terms.
If Switzerland were to give up its monetary independence by pegging the franc to the euro or by joining the euro area, it would lose its privileged status and, consequently, Swiss interest rates would increase to European levels. Since higher interest rates mean lower asset prices and lower levels of investment, and also higher mortgage rates, the Swiss are understandably not keen on giving up their national currency. Since any major decision in Switzerland has to be approved in a public referendum, the prospect of euro membership in the near future is thus highly unlikely.
The final point I would like to make is that going for a currency peg instead of adopting the euro outright is not an option for Switzerland either. It could even be a dangerous solution.
## Currency peg not an option
It is no secret that the majority of Swiss are against EU membership, at least at the moment. But you may ask: Even if Switzerland isn't an EU member, couldn't it peg the Swiss franc unilaterally to the single currency? Wouldn't this help eliminate the exchange rate risk, without giving up monetary flexibility entirely?
There is no such thing as a free lunch in monetary affairs! It is not at all certain that a peg would engender the foreign exchange stability that exporters dream of. We have learned from our experience with flexible exchange rates that any explicit foreign exchange target stimulates the appetite of speculators instead of calming the atmosphere. And this would be especially true if our pegging decision were to be taken unilaterally, without the prospect of a full and permanent integration into the euro area. Today, with global markets and global speculation, it would be an illusion to believe that a central bank can defend a specific exchange rate target unless there is strong convergence of economic fundamentals with the economy whose currency it is pegging the domestic currency to.
Furthermore, a nominal peg against the euro would do little to stabilise the real value of the Swiss franc - the only thing that matters for the international competitiveness of the Swiss economy - or to stabilise the franc vis-à-vis third currencies, especially the US dollar.
Given our exposure to global economic and financial shocks, and their arguably smaller importance for the euro area, pegging the Swiss franc to the euro is not a realistic option. Flexible exchange rates give us the room of manoeuvre we need.
## Conditions for monetary independence
Having reviewed the advantages of retaining our national currency and flexible exchange rates, I would now like to turn to the conditions necessary for successful monetary independence. I see three crucial factors here: credibility, transparency and efficiency.
## Credibility
The first is the need for a highly credible central bank. If the public believes that the central bank will not deliver on its promise to keep inflation low and stable, price stability is difficult to preserve. In this case, joining a monetary union whose policymakers are seen as more credible may be the easiest way to anchor inflation expectations.
Luckily, the SNB enjoys high credibility at home and abroad. Average inflation over the last 20 years amounted to 1.8%, and the Bank proved on many occasions that it was ready to act in order to defend price stability even in difficult situations. Moreover, the SNB was granted formal independence in 2000 through a constitutional amendment. Such independence is commonly seen as a means of enhancing the credibility of a central bank, since it limits the influence of the government, which may have an interest in keeping interest rates low, and thus stoking inflation. In my view, however, the formal independence we got in 2000 is unlikely to have changed the public's perception of the SNB much, as the Bank enjoyed de facto independence already before that date. Of course, our credibility rests on a commitment to price stability that is not just our own, but one that is also shared by the government and the population at large.
An important determinant of a central bank's credibility is its operational independence. Successful monetary policy is based on thorough economic analysis and professional implementation. To achieve quality and continuity in this area, a central bank must be able to attract and retain staff versed in monetary economics, market operations and many other fields. Only if a central bank's finances are insulated from daily politics, can it compete freely with the private sector in the access to human resources.
## Transparency
A second condition for successful monetary independence is a well-communicated monetary policy strategy. Policymakers need to announce clearly what they are trying to achieve in both the long and the short term, and with what means. Transparency of this kind makes it easier for market participants to judge the likely course of future monetary action. As a consequence, monetary policy changes rarely cause major fluctuations in financial markets, and the central bank can set interest rates to achieve the goal of price stability without having to fear financial distress.
The SNB attaches great weight to communication. Our monetary policy goal is clearly spelt out in the National Bank Act and in our monetary policy concept. Our interest rate decisions are communicated together with an explanation that includes our inflation forecast, which is based on the assumption that interest rates remain constant. This allows the market to understand how we judge future developments in the absence of monetary policy changes and thus makes it possible to infer the likely direction of future interest rate adjustments. In between monetary policy decisions, my two colleagues in the Governing Board and I give frequent speeches - as today - to clarify our view on things. Overall, I believe that our communication with the public is multifaceted and highly efficient.
## Efficient foreign exchange market
The third condition I view as crucial for successful monetary independence is an efficient foreign exchange market, as changes in monetary policy often translate into movements in the exchange rate. The amplitude of these movements can depend critically on the depth and efficiency of the foreign exchange market. An interest rate increase generally leads to an appreciation of the domestic currency, and this appreciation can be particularly large in a narrow market, where there may be few agents willing to sell. Hence, changes in monetary policy can cause large - or even excessive exchange rate volatility if the foreign exchange market is not well-developed. Deep markets, by contrast, allow the exchange rate to stick more closely to the fundamental factors of the economy. This in turn facilitates the conduct of monetary policy.
How deep is the Swiss foreign exchange market? A survey by the Bank for International Settlements shows that, in 2004, 6% of all currency transactions worldwide involved the Swiss franc. Considering
that Switzerland contributes less than 1% to world GDP, this implies that the Swiss franc market is very deep indeed. Hence, when we change interest rates, the exchange rate does not respond in an overly volatile fashion.
I would like to point here out that the SNB has not always been keen on free financial markets. For many years, we were afraid of a possible "internationalisation" of our currency. Until the early 1980s, we had a general feeling that our monetary policy was more likely to be successful if the Swiss franc were only traded domestically. We were overestimating our capacity to control markets with administrative measures. With time and experience, we came to realise that, in the long run, an internationalisation of financial markets was inevitable and that it was overall beneficial, since it would help reduce exchange rate volatility. Today, we are convinced that central banks should accept the markets' strength rather than spend time and energy trying to curb it.
In sum, Switzerland, with an independent central bank and efficient financial markets, meets the conditions for successful monetary autonomy. Let us now consider whether the SNB has made good use of it.
## Evidence of the SNB's monetary independence
Whether the SNB's monetary independence has been effective can be judged in two ways. First, by studying how monetary policy has reacted to shocks that were specific to Switzerland, and second, by assessing differences in monetary policy between Switzerland and the euro area in times of calm. Let me first discuss our reactions to asymmetric shocks.
## Monetary policy under special circumstances
Since the introduction of the euro, there have been three major shocks that I think affected Switzerland more than the euro area. These were the stock market crisis of 2000, the September 11 events, and the war in Iraq.
Do you remember how the NASDAQ Composite Index reached its maximum on 10 March 2000 and within a year, lost half its value? As a consequence, global demand for capital fell sharply. The weakness in financial markets and the cooling in global economic sentiment were a real shock to the Swiss economy in general, because of its specialisation in the production of capital goods, and to its banking sector in particular. Within twelve months, our exports and investments contracted sharply. Facing a slowdown in economic activity and less pressure on the price front, we cut interest rates for the first time in March 2001, a few weeks after the Federal Reserve and just ahead of the ECB.
September 11 and the war in Iraq were, from a Swiss central banker's perspective, new asymmetric monetary shocks. Both events caused a flight into the Swiss franc, which gained ground against the dollar and the euro. With a further deterioration of economic conditions and reduced inflation pressure, we cut rates twice by 50 basis points: the first time, in line with the Federal Reserve and the ECB, on September 17, and the second time, a week later. The ECB, by contrast, held rates stable until November 2001.
A similar situation could be observed in the run-up to the war in Iraq. In 2002, the SNB cut interest rates on two occasions, while the ECB did not move until December. In March 2003, we even pushed interest rates down to 0.25% in an attempt to stop the Swiss franc from appreciating and deflationary tendencies from developing. The ECB lowered its rate by only 25 basis points, down to 2.5%. Consequently, the interest rate differential between the euro and the franc rose from 1.25 percentage points to a maximum of 2.25 percentage points between December 2000 and spring 2003.
These three episodes suggest that the SNB made use of its monetary independence when faced with country-specific shocks. Shadowing the ECB's policy would not have been appropriate given the circumstances.
## Monetary policy in the absence of asymmetric shocks
Let me now turn to the question of whether the SNB's monetary policy differed in any important way from the ECB's during times when there were no major asymmetric shocks. Insofar as the euro area and Switzerland were exposed to the same shocks during such times, we would expect to see
identical policy reactions. The comparison of the course of interest rates in the euro area and Switzerland yields three interesting results.
First, interest rates in Switzerland lie below those in the euro area. This reflects the Swiss interest rate advantage. Since January 1999 - the date of the introduction of the single currency in Europe - the average 3-month interest rate on euro deposits has been 3%, and the average 3-month Swiss franc rate, 1.4%. Assuming a difference in the expected levels of inflation between the euro area and Switzerland of 1 percentage point, the Swiss real interest rate advantage thus accounts for about 60 basis points.
A second finding is that euro area and Swiss interest rates basically move in tandem, as one would expect considering we face the same economic conditions. The high correlation of interest rates does, however, not imply causality. Rather, our monetary policies are similar because our economies are highly integrated.
The third and final result is that the SNB tends to adjust policy before the ECB does. It is possible that shocks impact faster on our small open economy than on the euro area, making an earlier reaction of Swiss policy desirable. I also think, and I will return to this in a minute, that the SNB's small size helps us agree faster on interest rate adjustments.
## Conclusions
What do I conclude from our experience since the introduction of the euro? To start with, I disagree with the notion that small open economies on the edge of a large monetary zone are bound to do worse on their own and had best join the union.
On the contrary, it would be easy to demonstrate that living on the edge of the euro area is far preferable to being surrounded by an array of small European currencies facing crisis periodically, as was the case before 1999. Remember the instability of the European monetary scene in the 1990s and the volatility of hard currencies. Today, the euro is providing market stability, which is also beneficial to third currencies like the Swiss franc.
In monetary affairs, size does not really matter, but the rules of the game are more challenging for small countries than for large economies. Small countries have no choice but to respect market realities. They have to accept exchange rate flexibility even if, in the short run, some exchange rate development might interfere with the interests of their export sector. If they were to complement inflation targets with specific exchange rate objectives, they would create market confusion and destabilise expectations. The implementation of a clear and coherent monetary strategy is of the essence.
In my view, smallness in combination with monetary independence and flexible exchange rates is a strength rather than a weakness. For us, being able to set our own monetary policy has meant that we could react to country-specific shocks. Had we joined the euro area, we would not have had this flexibility.
Furthermore, monetary independence is attractive for small open economies, because policymakers can concentrate on national data when setting interest rates. In a monetary union, data from different countries have to be combined. Also, the decision process in the union is more complicated, since there are diverging interests to be reconciled. As a consequence, monetary policy is more inertial than in a small open economy with a currency of its own.
Looking at our development over the last ten years, I do believe that price stability, due to monetary independence and flexible exchange rates, has been beneficial to the Swiss economy. It has maintained a low interest rate environment which has facilitated the private sector's adjustment to the rapid changes of the world economy.
## And what about Israel?
I think that the arguments in favour of Switzerland's monetary independence apply even more to this country. While we at the SNB concentrate primarily on domestic developments and trends in the euro area, the Bank of Israel also has to take into account economic conditions in the US. Thus, there are three major sources of shocks that Israeli monetary policy has to respond to in order to achieve the goal of price stability.
BIS Review 46/2006
Despite difficult circumstances, Israel was able to stabilise its monetary framework in the 1990s. The Bank of Israel now operates with an inflation target and will soon benefit from a new law that will modernise its decision-making process and strengthen its autonomy.
All the necessary ingredients are now in place for the implementation of a successful independent policy in this country, as was once also the case in Switzerland. May I wish Stanley Fisher and his colleagues all the best in this endeavour.