Are different price developments in the euro area a cause for concern?

Speech by Dr. Willem F. Duisenberg President of the European Central Bank, at the Financial Services Industry Association, Dublin, 6 September 2000

Ladies and gentlemen, it is a pleasure for me to speak here in Dublin on the occasion of the annual dinner of the Financial Services Industry Association (FSIA). Public support for the euro in Ireland has been particularly strong, and this may be due to the prevailing expectation, stronger than elsewhere, of economic benefits resulting from the single currency. The introduction of the euro, however, also poses some new questions and challenges for the ECB and national economic policy-makers. Today I should like to take the opportunity to elaborate on one particular issue, that is the challenge posed by different inflation rates across countries or regions within the euro area. There is a general consensus that the euro area economies have experienced a great deal of convergence both in terms of economic growth and inflation during the past decade. In fact, the move to a single currency and the introduction of the euro are based on a successful convergence process towards a high degree of price stability in participating countries. However, this does not mean that countries have always had the same rates of inflation during the past decade. And, indeed, there is no reason to believe that they will - at all times - have the same rates of inflation within Monetary Union. I am well aware that divergences in inflation have increased lately in the euro area as a whole. For example, inflation in Ireland deviates significantly from area-wide inflation.

There are a number of reasons why differences in economic conditions have occurred in the past, and there is ample reason to expect that some degree of economic divergence will continue to emerge within the euro area. First, there may be reasons why some differences in inflation may occur over relatively short periods of time. Second, national or even regional rates of inflation may also differ within a single currency area over longer periods of time. Let me elaborate on each of these factors in turn.

Short-term differences in inflation may generally occur due to country-specific, statistical and "erratic" factors. First, government policies may lead to different price developments via tax changes, liberalisation measures and administrative price changes, as they are generally not of the same magnitude and timing across all euro area countries. The deregulation of national electricity and gas markets or measures such as the eco tax in Germany or the tobacco excise duty in Ireland may serve as examples of such policy measures. Secondly, since patterns of consumption are not the same in all participating countries, the weights used in the construction of the national components of the Harmonised Index of Consumer Prices (HICP) differ. For example, processed food constitutes around 13.5% of the HICP in the euro area as a whole, but 21% of the HICP in Ireland. This could, purely mechanically, generate measured inflation differentials across countries even if price changes for individual goods were equal in all these countries. Finally, different weather conditions or different numbers and timing of national holidays can create "noise" in measured inflation rates, which would be of only limited economic significance. Looking ahead, there is no reason to expect that such factors would disappear within the euro area.

In addition, a number of deeper economic factors may also contribute to short-term divergences of price developments. Trade patterns, production structures, tax systems, financial markets, and institutional structures differ significantly across euro area countries. Hence, common shocks may affect countries in a diverse way. The 1998/1999 emerging market crisis, the depreciation of the euro against the US dollar, and the recent surge in oil prices may all serve as examples of common shocks which have contributed to diverging developments within the euro area. Furthermore, second-round effects of common shocks may differ because of different institutional structures, such as the wage-bargaining system. Of course, the transmission of single monetary policy reactions to such shocks may also still differ because of financial structures. Likewise, differences in inflation may occur on account of the impact of asymmetric shocks, or differences in the response to such shocks. German unification was a clear example of such an asymmetric shock.

Looking ahead, some of these deeper economic factors may become less important as differences in trade patterns, tax systems, financial markets and institutional structures are likely to diminish. Others may become more important. In the face of increased openness and trade integration, a tendency towards country-specific and regional specialisation might emerge. This could lead to some inflation divergences between countries when shocks occur. However, to the extent that the linkages between industries increase across national boundaries as part of the process of growing integration in the euro area, the degree of synchronisation across national short-term economic developments may well increase.

Let me now turn to the factors which could lead to long-term differences in inflation. I should like first to address briefly the issue of price level convergence. Price level convergence could be expected to take place in the euro area for at least two reasons. First, the completion of the internal market and increased cross-border price transparency contribute to eroding the scope for the existence of substantial price differentials for products which are easily tradable across borders. To a large extent, this may have taken place already before the start of Economic and Monetary Union (EMU), but differences remain. One example of such a price convergence that has attracted public attention relates to car prices. Secondly, with regard to goods and services which are less easily tradable across national borders (such as housing and hairdressing), the long-term convergence of productivity and living standards across the euro area would create a tendency towards price level convergence. Countries with higher levels of economic development tend to have higher price levels for non-traded goods and services. While higher wages in the traded goods sector largely reflect higher levels of productivity, spillovers of wage increases to the non-traded goods and services sector often imply that wage levels in the economy as a whole will typically be higher in more developed economies. As the scope for increasing productivity growth is usually more limited in the non-traded goods sector, a general increase in wages as a result of productivity gains in the traded goods sector will raise the cost of producing non-traded goods and services. This leads to higher relative price increases for non-traded goods and services. In a dynamic context, as a less developed country is catching up with its neighbours, inflation in its non-traded goods sector could be rising more rapidly than in other countries, and as a result, inflation will be higher. This type of convergence normally takes place gradually and within certain limits among relatively homogeneous industrial countries, but could, of course, reach a much larger scale if differences in price levels among participating countries are more marked.

Are differences in inflation among euro area countries a cause for concern? The short answer is, "it depends". On the one hand, as I have just explained, differences in inflation rates among euro area countries had to be expected. They are in effect partly the result of the completion of the internal market and the creation of the single currency itself. On the other hand, if differences in inflation rates are quite large, they need a thorough examination of the underlying reasons as well as appropriate economic policy action at the national level.

The single monetary policy and the establishment of a more stable price environment in the euro area as a whole have been accompanied by a considerable degree of narrowing in inflation differentials. However, current inflation differentials for some euro area countries are quite large. An analysis of data for price developments in various US cities in the past suggests that inflation differences may occur within a monetary union, which can be at least as great as those currently prevailing between most euro area countries. A further feature of the US data is that the observed divergences in inflation rates among these cities, although persistent, are not permanent. There seems to be a tendency for price levels in individual cities to return to their initial levels relative to the national average. It can be assumed that similar adjustment processes would generally also operate in the euro area.

The cyclical component of consumer price developments may be affected by the orientation of monetary policy and changes therein, and could thereby exhibit more common patterns in the future. Typically, all kinds of specific shocks, such as oil price changes, seasonal patterns and weather conditions may then dominate the dispersion of price developments. To give you an example, the difference between the highest and the lowest annual national HICP inflation rate increased to 3.8 percentage points in July 2000, from an average of less than 2 percentage points recorded during 1998 and 1999. The biggest differentials have emerged in product groups such as food and energy. These have traditionally been most exposed to such oil price shocks and weather conditions. Non-energy industrial goods not only show the lowest, but also the least volatile differentials.

I have so far mentioned a number of reasons why divergences in price developments as such need not be particularly worrisome. There are, however, also circumstances under which inflation differentials are a cause for concern within a monetary union. Let us consider the case of a country where domestic demand has long been growing strongly, pushing real GDP at or above its potential. A further stimulus to domestic demand may then tend to generate upward pressure on domestic costs and prices. It will also lead to a higher demand for import goods. The external position deteriorates. The increase in inflationary pressures in such a case is not offset by a rise in productivity. Therefore the rise in prices relative to the external sector implies a loss of competitiveness for the economy. This in itself could exacerbate external disequilibria. Moreover, if the losses in competitiveness are not corrected, domestic and foreign markets for some products may be permanently lost, damaging longer-term growth prospects. Within a monetary union, the potential correction mechanism through the exchange rate is no longer available. Hence, the adjustment has to occur through domestic costs and prices. If rigidities in product, labour and capital markets hamper such adjustment, the correction may only take place in the form of weaker growth and a loss in employment growth.

Differences in price, and more generally, economic developments between euro area countries do not automatically warrant concern. Some differences in economic developments across euro area countries may even be seen as an advantage as they would average out and help to smooth cyclical developments at the area-wide level. Perfectly synchronous movements would increase the amplitude of cyclical movements in the euro area as a whole. However, a national policy response would be clearly warranted to deal with sizeable, growing and unsustainable divergences. The single monetary policy can only be geared towards the objective of price stability for the euro area as a whole. It is therefore not in a position to influence the dispersion of inflation rates across euro area countries. If action is required, it must be taken through national economic policies. They need to ensure the capability of individual economies to absorb and respond to shocks. The flexibility of national economic policies to deal with country-specific issues is required in all Member States of a currency union, but perhaps in particular in the smaller countries. Because of sheer size and weight, economic developments in the large Member States have less potential to deviate from area-wide averages than those in a small country. Moreover, national developments in large Member States may have greater spillover effects on other, smaller Member States' developments. Furthermore, small economies, such as Ireland, tend to have a higher degree of national openness, which implies that specific responses to divergent economic developments may be less effective and thus should be further-reaching than in the larger economies.

As regards structural policy, it is important to implement reforms in the product, labour and capital markets, which are aimed at increasing overall economic flexibility and which would increase the degree of flexibility of national economic policies in reacting to country-specific events and common shocks. This is important for all countries, independent of their cyclical position. Policies that support greater wage flexibility, wage dispersion in line with labour productivity, and greater mobility of labour across sectors and regions are likely to smooth the adjustment process, both in a booming economy and during a downturn. Also, measures to increase competition in the non-traded sector, particularly services, would help to smooth the adjustment process. All such policies are desirable in their own right, irrespective of membership of the euro area. However, the use of structural reforms to address excessive aggregate demand growth in the short term suffers from the substantial amount of time needed to prepare and implement such reforms. This should make it clear that national governments have no time to lose in taking measures to increase the flexibility of their national product, labour and capital markets. The current favourable economic environment provides an excellent opportunity to undertake the necessary structural reforms in euro area countries with vigour.

It is also necessary for fiscal policy to play its part. In general, unbalanced fiscal policies have a negative effect on economic efficiency and on price developments. They also increase strains on the single monetary policy, and may have negative spillover effects on other countries. The Stability and Growth Pact helps to guarantee sound public finances over the long term, which is an important contribution to macroeconomic stability and is also conducive to sustained economic growth and employment. If the public debt is reduced to a sustainable level and fiscal deficits are close to balance or in surplus, this creates room for manoeuvre to let automatic stabilisers play their role in the event of the economy slowing down. However, in addition, particular attention also needs to be given to fiscal policy in an environment of strong growth, budget surpluses and decreasing debt ratios. In such circumstances, fiscal policy should by all means refrain from giving a pro-cyclical impulse to the economy.

To summarise, the euro area has experienced a considerable degree of convergence in inflation rates during the past decade. National differences in output growth or inflation observed at the current juncture do not in general appear to be unusual either from a historical perspective or compared with developments in the United States. The more pronounced divergences in some countries strengthen the case for policy action at the national level. Looking ahead, there is reason to believe that divergent price developments may continue to emerge. Some sources of divergence may become less significant, while others may become more important. The ECB will continue to monitor divergent developments, but as our monetary policy can only be geared towards maintaining price stability in the euro area as a whole, it cannot play a role in tackling growth and inflation differentials between euro area countries. Therefore, national economic policies need to take action if divergences threaten to cause undue imbalances. In this context, structural policy, strongly backed by fiscal policy, which is not pro-cyclical and in line with the Stability and Growth Pact, as well as appropriate wage developments taking account of labour productivity growth and the level of unemployment, have to play a crucial role. Fortunately, professional forecasters not only expect continued and strong output growth to prevail in the euro area over the next few years, but the outlook also suggests increasing convergence both in terms of economic growth and consumer price inflation within the euro area.