Persson: Sweden, EMU and the krona
Thank you for the invitation to come here. The theme is an important one, even if we do not yet know when a referendum will be held and whether a decision on EMU membership will be taken in Sweden.
There is now just over one month left until 1 January 2002, when 307 million inhabitants of 12 European countries will have the same new banknotes and coins. 15 billion banknotes and 50 billion coins will come into circulation. In the course of a couple of months the national banknotes and coins in the eurozone will cease to be legal tender.
The distribution and exchange will involve enormous challenges in terms of practical implementation, logistics and security. Yet I should imagine that many people will soon see it as a positive experience to be able to use the same means of payment throughout almost the whole of western Europe. Seeing the banknotes and shopping in euro prices across the continent brings home the existence of the monetary union in a more tangible way for the citizens of Europe.
For us Swedes the euro will enforce our sense of being outside the euro. In addition, we have behind us a period during which the krona has weakened without this being apparent entirely motivated by fundamental factors. In my opinion, these are problems that are worth a closer examination. It is on these issues that I intend to speak today.
1. The role of the Riksbank
The role of the Riksbank in this context is to implement the decision taken by the government and parliament. The Sveriges Riksbank Act, which has governed our operations since 1999, states that "the Riksbank shall decide on the application of the foreign exchange rate system decided upon by the government." The General Council of the Riksbank last expressed an opinion on the issue in 1997 when asked to respond to the government's investigative report. In its response to the submission the General Council emphasised that the economic advantages of a membership probably outweighed the disadvantages and that Sweden therefore should join the EMU from the start. The current Executive Board has not had reason to take a stance on the issue. Over the past few years we have regularly made technical preparations to enable a possible future membership to be implemented at no great cost and in a problem-free manner if and when a decision is finally taken.
2. Economic consequences of a Swedish EMU membership
The most obvious advantage with the disappearance of the currency borders is that which is normally known as transaction gains. This is the removal of the cost and inconvenience involved in exchanging currency when one visits or trades with different countries in the eurozone. Of course the foreign exchange costs in the eurozone were limited right from the start of the EMU in 1999, but they will now disappear completely. For the economy as a whole, however, this exchange of cash has involved a relatively small cost, which is nevertheless symbolically important.
A somewhat larger cost has arisen from the handling charges the companies have been forced to bear to hedge against undesirable fluctuations in exchange rates. This cost disappeared as early as 1999 within the eurozone, but it is not particularly large in relation to our GDP - perhaps a few tenths of a percentage point.
A single currency will promote competition within the EU and benefit the consumers, who will experience lower prices and thereby have more money in their pockets. When everyone uses the same unit of currency, it is easier to compare prices between countries. It will also be an advantage for the national economy as a whole, since competition facilitates the work on holding down inflation. Some differences in prices are unavoidable and arise from transport and information costs, but not all of the differences can be attributed to these costs. Despite the fact that we now have a common market with free trade for goods and services throughout the entire EU, there are still large differences in prices between different countries within the union. The average price difference is 16 per cent, which means that the difference is approximately 40 per cent larger than is the case between various states and regions in the USA. The price of a normal Volvo, for instance, can differ by 15 per cent or more from one country to another in Europe.
Some of this difference is created by difference in cultures and languages, but new studies indicate that the actual psychological barrier involved in differing units of currency may bear a large part of the responsibility. The US economist Charles Engel, for instance, has calculated in extensive studies of price differences between the USA and Canada that a currency border creates price differences corresponding to a geographical distance of almost 8,000 km, even if the countries on either side of the currency border (as in the case of Canada and the USA) are entirely open to one another and share the same language and culture.
The largest negative effects of having different currencies probably have to do with trade and resource allocation. In the same way that we believe inflation results in large national economical costs by creating uncertainty about future expenses and income, a fluctuating exchange rate does too. In addition, the allocation of resources in society is distorted by an exchange rate that is not fully motivated.
The EMU has particular significance for trade. Many economists have long attempted to estimate this gain by looking at the extent to which fixed and relatively floating exchange rates affect trade and they have found a very small difference. However, a single currency is not the same as a fixed exchange rate. The practical and psychological differences are probably considerable. Professor Andrew K. Rose at Berkeley in the USA has, together with other economists, instead compared trade within monetary unions with the trade between countries without a monetary union, where he has made adjustments for such facts as monetary unions often share the same language, culture and regulatory framework.
When Rose recently presented such a study for Sweden, his opinion was that the EMU could over the course of a few decades increase trade and integration by as much as 30-50 per cent. Rose himself admits that the figures are uncertain and other economists believe in much smaller trade gains. However, even the more cautious calculations indicate that the monetary union could provide considerable gains. For instance, Professor Torsten Persson estimates, using an alternative calculation of Rose's material, that the gains of a monetary union are only about one-fifth of what Rose claims. This would still involve a clear trade gain if the reasoning is applied to Sweden. Hopefully, the new debate stimulated by Rose's articles will lead to new insights in this very important field.
What about the disadvantages? After all, giving up one's own currency also involves giving up a certain amount of one's autonomy.
The arguments against a possible Swedish membership of the EMU have often touched on the opportunity to pursue an independent stabilisation policy. The starting point for this is that the Swedish economy could for one reason or another find itself out of synch with economic developments in the eurozone, what is known as an asymmetrical shock. Having an independent monetary policy and currency would then give us the opportunity to stabilise our economy ourselves - act as a form of shock-absorber, which we will lose if we have a joint monetary policy governed by the prospects for inflation and the business climate throughout the eurozone.
This was also the primary reason why the Calmfors Commission's report on the consequences of Swedish membership in the EMU (1) came to the conclusion that Sweden should wait until its labour market and wage formation showed sufficient flexibility to deal with a serious disturbance. However, it should be noted that the Commission was envisaging a "small" eurozone with around 6 countries taking part, whereas now we are talking about 12 countries, representing 40-50 per cent of Sweden's trade. Moreover, in practice the Swedish business climate has rarely come seriously out of synch with the eurozone. When Sweden has done so, it has mainly been the result of mistakes in Sweden's own economic policy and not due to any unforeseeable external disturbance that has afflicted the Swedish economy. Also, in this respect Sweden has changed in recent years: the introduction of inflation targeting and expenditure ceilings should reduce the risks of new economic policy mistakes.
In addition, there are strong indications that the correlation between Swedish and eurozone growth will increase if we join the EMU, as trade and integration with the euro countries will thereby be strengthened. We can see already how the long-term Swedish market rates have covaried strongly with those in the eurozone in recent years, even during periods with differences in business climate between Sweden and the eurozone. As far as a Swedish membership of the EMU would reduce these differences in business climate, it will lead to a long-term interest rate development that is more adapted to the business climate.
If we still believe that Sweden could become radically out of synch with the rest of Europe, there is of course an alternative to using an independent exchange rate and monetary policy as shock absorbers. The more robust budget situation Sweden enjoys today, for example, should provide a slightly greater scope than during the early 1990s for using fiscal policy for the purpose of stabilisation. One can also imagine building up buffer funds to pay for reduced payroll tax in crisis situations (known as internal devaluation) of the type introduced in Finland (2). Another variant of an asymmetrical disturbance is that asset prices - the property market and stock exchange - due to a too strong economic development are forced up for a short while. However, in situations such as this, the required tightening can usually be managed with the help of fiscal policy.
Naturally, stabilisation policy via the budget has its limitations since it only works with delay and is dependent on many other long-term considerations. In addition, the Stability and Growth Pact, which Sweden is included in whether or not a member of the EMU, provides a limitation on increased government expenses.
The most important "shock absorber" even if we do not join the EMU, but especially if we do, is that wages can adapt to changes in the competition situation, i.e. that there is a high degree of flexibility and pliability in wage formation. It would thus be possible to avoid an increase in unemployment during an adaptation period. Naturally, it can be more difficult, politically, to achieve this than to adapt an exchange rate. A single monetary policy therefore requires greater discipline from labour market parties since the connection between wage costs/inflation and monetary policy will be less obvious than is the case today.
The krona
The reasoning I have pursued so far has assumed that the exchange rate actually functions as a shock absorber: that the floating exchange rate adapts to competitiveness and the business climate, i.e. that it strengthens against other countries when Sweden's economy shows better growth and weakens when the Swedish economy shows weaker growth than other countries. The exchange rate would thus counteract imbalances that arise. If Sweden alone were to suffer a negative disturbance that required prices and wages in Sweden to develop differently than those in the euro countries, a floating exchange rate would facilitate the adaptation.
Here the development of the krona in recent years gives some cause for reflection. Does the krona really function as a shock absorber in practice? Or do fluctuations in the exchange rate risk being affected by so many other factors that the role as shock absorber is counteracted? It is difficult to see the most recent severe weakening of the krona as an adaptation to the fact that Sweden has come out of synch with the surrounding world. So what actually steers the krona rate?
Since the 1970s the krona has weakened by almost 90 per cent against Sweden's most important competitor countries. Approximately half of this weakening can be explained by the fact that inflation in Sweden has been higher than in other countries. However, now inflation in Sweden is developing roughly in line with international inflationary trends. The weakening of the krona in recent years therefore cannot be explained by differences in the inflation rate.
If we instead look at the exchange rate adjusted for inflation differences, the "real" exchange rate, we find some traditional explanations for much of the weakening that has taken place from the 1970s until the mid-1990s. Our terms of trade were weakened, productivity developed more weakly in Sweden than in other countries and during certain periods investors were concerned over a rapidly growing central government debt.
However, developments during the latter part of the 1990s were rather different. The growth in productivity and GDP has been better and the inflation rate as low as the average among our main trading partners, while the central government debt has declined, which should have favoured a stronger krona. Nevertheless, the krona has depreciated.
A small part of the most recent depreciation can be explained by real economy factors: the terms of trade have deteriorated slightly due to falling prices on telecom products; exports have declined more heavily in Sweden than in many other countries following on from the economic slowdown in the USA; and a falling order intake for exports has created a declining demand for the krona.
However, there are also other explanations that are not directly concerned with real economy factors, but have instead been steered by financial flows, i.e. that the krona has been affected by a net outflow of capital from Sweden. The development of the krona and the stock market has shown a strong correlation since the beginning of 2000. This is probably connected with foreign investors selling Swedish shares in line with falls on the stock market, which has affected Sweden more than other countries, as a large part of the stock market value is owned by foreign investors. Moreover, Swedish investors have bought a large amount of foreign shares, which is probably connected with the new pension system and the freer investment regulations for the AP pension funds. It has been possible for the AP pension funds to increase their foreign exchange exposure, i.e. the percentage of foreign assets, from 10 to 15 per cent with effect from this year. The funds can increase their exposure by five percentage points a year to a maximum of 40 per cent in 2006.
On top of this comes a factor that is difficult to explain on the basis of classical exchange rate theories. The krona is affected to a large extent by general unease on the global financial markets. During times of financial unrest the krona weakens when investors leave the Swedish market for ones considered a safer investment. This was particularly noticeable during the Mexico crisis of 1995 and the Asia crisis of 1997-98. The recent unrest on the financial markets in connection with uncertainty over the global business climate has once again had a negative effect on the krona. Following the terrorist attacks in the USA we saw a further depreciation of the krona (and for some other minor currencies), as well as a weak development in the stock market.
Finally, the generally stronger dollar must be mentioned as one of the explanations for the weak krona. Some of the dollar's strength can be attributed to higher productivity in the USA, but most economists still believe the dollar to have been overvalued on the basis of traditional economic theory.
The krona has thereby been steered to some extent by factors that are not connected with the economic situation and developments in the real economy. This means that the weak krona does not function fully as a shock absorber against the outside world. Sweden has an undervalued currency despite our economy developing as well as other countries' economies. The monetary policy problem is that a weak krona risks creating inflation via import prices, which in turn can force a tighter monetary policy stance than would otherwise be necessary.
There are international examples of how the exchange rate, instead of acting as a shock absorber, has rather worsened the effects of an economic slowdown. Japan and Switzerland have experienced prolonged crisis periods during the 1990s, while their currencies have remained very strong and have thus caused problems for the export industry.
One effect of the weakening of the krona that we have now is that the export industry and the competitive import industry should become more competitive in the short term. However, there are some structural problems here: an unmotivated weak exchange rate risks strengthening certain parts of the export industry that do not constitute expansive future sectors. Sweden's future hardly lies in competing with low costs and prices, but rather with quality, cutting edge technology and know-how. When the krona eventually appreciates to a level more motivated on the basis of fundamental factors, Sweden will have an industrial structure poorly equipped to meet the requirements of the future.
Conclusions
The Riksbank assumes that the krona will appreciate. It is possible that we have already seen the beginning of this development very recently. When the upturn does take place, it may prove even stronger than the Riksbank estimated in its latest Inflation Report. This does not prevent the fact that the recent valuation of the krona has brought new perspectives to the question of Swedish membership in the third phase of the EMU, i.e. participation in the eurozone.
The volatility of the krona creates difficulties for export and import companies in making decisions regarding trade and investment. However well monetary policy functions, it still cannot supply the efficiency gains at the microeconomy level that ensue from a single currency. The effects on growth through both price competition over the borders and increased trade could be greater than we have previously often assumed.
An independent monetary policy will of course remain the main argument against EMU membership. The question is how valuable this is. In part, I do not consider the risk very large of Sweden alone getting caught up in a crisis where these very instruments will be the best for stabilising our economy. In part, we have seen that the exchange rate can be governed for fairly long periods of time by factors on the financial markets that are not primarily related to the real economy. Instead of acting as a shock absorber for business climate developments in Sweden that are out of synch with other countries, there is a risk that the krona, by weakening more than motivated by fundamental factors, could instead upset Sweden's possibilities for stabilising its development.
The economic gains of remaining outside the eurozone thus appear uncertain, while recent research indicates that Swedish debate has probably undervalued the integration gains from EMU.
When you hold your first euro banknote next year and take a closer look at it, you will be able to see that Sweden is actually present, on the European map that adorns one side of every banknote. May I be permitted the cautious interpretation of this as a sign of things to come?
Notes:
1) Report on the EMU investigation (SOU 1996:158).
2) A more detailed discussion of the stabilisation policy tools within the EMU can be found in the report "Stability and stabilisation policy in the EMU" (SOU 2001:62), which can be downloaded from http://finans.regeringen.se/propositionermm/sou/index.htm.
