Eurozone inflation differentials likely to stick around

In April, eurozone inflation inched up to 7% after it had fallen to 6.9% in March. While down from its peak, this is still far above the ECB’s target level of 2%. Furthermore, the generally more sticky core inflation showed little signs of easing yet. In April, it stood at 5.6%, close to March’s all-time high of 5.7%. Apart from high average inflation, the large inflation variation between eurozone member states could also become a headache for the ECB. In any case, while eurozone figures feed into our view that the ECB will keep its policy rate higher for longer, major inflation variation makes the one-size policy less likely to fit all than ‘usual’.

In this paper, we examine whether inflation differentials are likely to remain higher for a longer period of time, and if so, which countries are likely to be at the top and bottom of the list. To do so, we must first assess the current key drivers of inflation differentials. Second, we delve deeper into the determinants of more sticky core inflation and its variation across the block. The latter provides useful insight into the expected inflation differentials further ahead.

The disparity in inflation between eurozone member states is enormous. Energy inflation still shows the largest variation, but the role of core differentials is not only important but it is also growing.

Despite recent tightening, inflation differentials within the eurozone are still very high (see Figure 2). In April, the gap between the highest and lowest inflation rate across the block was 12.3 percent. The Baltic states and Slovakia currently have the highest inflation rate, while Luxembourg and Belgium have the lowest (see Figure 3). The differences in inflation between member states can be attributed to a variety of factors. In short, while energy is still the category experiencing the widest inflation differentials, in essence it appears that price developments differ widely among all the main categories of goods and services, i.e. energy, food, and core (see Figure 4).

This also results from an analysis in which we decompose the difference in headline inflation between a member state and the eurozone into a price and a composition effect. The former captures whether prices of different goods and services have increased faster in country x than in country y. The latter captures the effect of the composition of the consumption basket on inflation. In other words, if energy, food and core inflation would be at the euro area average in all countries, would that lead to different headline inflation rates because people in country x spend a larger portion on e.g. energy than people in country y?

From Figure 5 it is clear that the majority of inflation variation can be explained from the price effect. Last summer, the composition effect played a larger role, as especially countries with a relatively energy intensive consumption basket, such as the Baltics, saw above average inflation.

Energy prices still account for a sizeable share of the divergence in inflation rates, partly because of base effects. We expect that it will take some time for these differences to largely subside.

Some countries, such as the Baltics (see Figure 7), are still experiencing double-digit energy inflation, while others, such as Belgium and the Netherlands, are experiencing double-digit contractions. There are numerous reasons for these differences, which can be broadly classified as i) varying increases in energy price indices since the beginning of 2021 (see Figure 8) and (ii) timing, i.e. base effect.

Retail energy prices have diverged significantly over the last two years, owing to factors such as varying previous energy dependence on Russia and at which price alternatives could be found, heterogeneous retail energy markets across the block, non-harmonized calculation of energy price indices, and the characteristics and timing of national support packages targeting prices (see box and Appendix). Base effects are also important. Because inflation is calculated on a year-on-year basis, it is important to consider what happened to prices not only in the current month, but also in the same month a year ago. Since retail energy prices started their ascend at different points in time and at different speeds across the block, base effects, which reduce inflation – as price indices flatten or even start to fall, kick in at different times.

Regardless of the country, base effects are likely to cause substantial swings in energy, and hence headline inflation, over the course of this and the following year. We do expect the differences in energy inflation between member states to narrow as wholesale energy prices stabilize and government support measures fade.[1] Indeed, we have already seen that energy and thus headline inflation have fallen faster in most of the countries that faced the highest energy inflation last summer. At the same time, government intervention to reduce or limit prices has become less visible in inflation figures in many countries, as a lot of those measures were implemented more than a year ago.

With the role of energy inflation as driver of inflation differentials declining, the role of core inflation will grow. In the medium term, developments in core inflation are likely to become the primary driver of inflation and differentials within the monetary union.

[1] Clearly the process could be slowed by a renewed increase in wholesale energy prices and a change in government support policies.

Based on a number of indicators, we expect that the currently high core inflation will be more (less) persistent in the countries that currently have the highest (lowest) core inflation. We therefore expect inflation differentials to remain substantial.

Core inflation is generally more stable and less volatile than the volatile energy and food components, and it has a larger weight in the consumption basket. Because core inflation is so important to the inflation outlook, it's worth delving deeper into the drivers of core inflation. There are a few indicators that we pay close attention to in order to predict if current differences will last. In order to get a feeling for the relative stickiness of inflation, we have built a heatmap using Z-scores.

Persistence and current core inflation

The first indicator is the persistence of core inflation. In some countries core inflation is more sticky than in others. This is due to a variety of factors, ranging from automatic wage indexation, cultural effects, pricing power of domestic companies and so on. In order to estimate the persistence we have estimated a simple autoregressive model in style of this ECB paper. We scale the persistence factor by the current core inflation, since a high persistence does not have to be a problem when core inflation is actually low (in fact it could be beneficial).

Labour market

Because the eurozone lacks optimal labour mobility, there are significant differences between national labour markets. The unemployment rate in the Netherlands and Germany, for example, is significantly lower than in the rest of Europe. Generally, a tighter labour market also triggers higher wage growth (see Figure 10, wage growth is proxied by a change in labour costs). In turn, cross country analysis shows that within the euro area core inflation is higher in those countries where labour costs increased the most (see Figure 11).[2] We have included the difference between the unemployment rate and the non-accelerating-wage-rate unemployment rate to capture labour market tightness (NAWRU). This variable accounts for labour market differences. The unemployment rate in Italy, for example, remains high, but relative to the NAWRU, the labour market in Italy has not been as tight since 2008.

[2] Admittedly, the causality unlikely runs in one direction, as higher core inflation could also induce higher wage growth. Yet we would expect a larger time lag than the other way around – unless inflation expectations are spot on.

Economic growth

Economic growth is another driver of core inflation because it is easier to raise prices for products and wages for workers when the economy is growing. To make a fair comparison, growth forecasts must be adjusted for potential growth, similar to the adjustment we make for the labour market. In the Baltic countries, for example, structural growth exceeds the eurozone average. Hence, we look at the output gap, rather than the effective growth forecast. The Netherlands truly stands out in this case, as it is expected to grow significantly faster than its potential growth rate.

Energy passthrough

In our previous report (Why inflation is diverging in the eurozone) we argued that energy played a significant role in explaining inflation differences. Energy prices have risen so dramatically that even goods and services that are not typically associated with energy costs see price increases as companies try to pass on their higher costs – rather than drastically cut margins. The extent to which this occurs is determined by the initial price shock (which was much higher in the Baltic countries than in France, for example!), as well as less tangible factors such as pricing power. To assess the effect on non-energy prices, we ran a series of regressions on energy prices. We did this for all 19 member states as well as the eurozone average. It turns out that Slovakia, Lithuania and Latvia are the most vulnerable.

Heatmap

Figure 14 shows a heatmap with all of those elements combined. We expect the differences in core inflation to persist next year, based on these figures. The countries with core inflation that is currently well above the eurozone average (Slovakia, Estonia, and Lithuania) are the most likely to have high core inflation next year. Latvia, despite having core inflation of 10.0 percent, is likely to see core inflation fall as economic growth is expected to be well below potential. Countries that are likely to have lower core inflation, on the other hand, already have core inflation that is lower than the eurozone average. Luxembourg and Spain are two examples.

Despite some convergence since the summer of 2022, inflation differentials between eurozone member states remain substantial. Moving forward, further convergence is to be expected as energy price swings weaken, energy price passthrough fades, and the effect of government support measures falls away. The larger-than-usual disparities in core inflation, however, will likely take longer to resolve. In some countries there still is a substantial passthrough effect of higher energy prices, whilst at the same time, there is a large variety in the outlook for the economy and labour market. We expect that the currently relatively high core inflation in Slovakia, Lithuania and Estonia will prove especially sticky.

Because of the large variety in inflation rates across the block, a one-size-fits-all monetary policy is even less likely to fit everyone than in 'normal' times. So in due time, the ECB could have more to worry about than the historically high inflation itself.