How do central banks enforce inflation targets

Inflation, inflation measurement and central bank policy

Inflation is a construct. It is perceived differently by different actors. This happens partly because shopping carts differ, partly because expectations are formed differently. This article discusses the heterogeneity of inflation and its perception and what this means for the target value of central bank policy.

The ECB's inflation target

The European Central Bank (ECB) is aiming for an inflation target of close to but below 2%. This inflation target is not anchored in the European Treaties. There, in Art. 127 TFEU, only the primary ECB objective of price stability is set down. This means that an inflation of 0% would be in line with the mandate of the contracts. The inflation target of close to but below 2% is based on the interpretation of the Governing Council as stated in a 2003 resolution. An original decision of the Governing Council in 1998 spoke of an inflation target of below 2%.

Even if a 2% target is not unusual in the concert of the central banks within the OECD countries, this interpretation of the inflation target has important economic implications: While the inflation rate, which has been below 2% since 2014, is no justification for an aggressive one after the decision of 1998 Expansion of the central bank money supply, the 2003 decision leaves more room for justification. On the other hand, at the zero interest rate limit, which has already been reached for some time, there are doubts as to the extent to which monetary policy is able to stimulate overall economic demand and induce price effects.

Reasons for a moderate, positive inflation rate

The ECB justifies the 2% target with three arguments:

  • First, a target inflation rate of close to 2% could reduce the risk of deflation.
  • Second, a positive average inflation rate is helpful in order to avoid that in a heterogeneous euro area some countries and regions achieve negative inflation.
  • Third, measuring inflation could be accompanied by small but systematic measurement errors. If so, measured inflation of zero could mean deflation.

Beyond these official reasons for moderate inflation, which are also listed on the ECB website, numerous additional arguments are discussed in the literature (Diercks, 2019). Two arguments in particular seem to carry weight among economists:

  • At least since Tobin (1972), a widespread and shared argument has been that the efficiency of labor markets benefits from positive inflation because employees have a high aversion to lower nominal wages. If such a nominal wage rigidity applies, inflation is required in order to achieve a real wage reduction and thus the preservation of jobs as well as an organic economic change in certain sectors or parts of the economy that are affected by negative productivity developments.
  • Another widely held argument for moderate inflation is that interest rates close to the zero rate limit severely limit the ability of central banks to help stimulate the economy through further rate cuts. Such a situation has occurred in the euro area since 2014. Increased efforts were made to push the nominal interest rate below zero. However, further rate cuts are difficult because market participants can escape into cash, which promises zero interest rates. At least this is the case if one abstracts from cash holding costs, which can reduce the effective nominal interest rate. Against this background, a positive inflation target can be helpful in avoiding the risk of the central bank being unable to act by setting a binding lower limit for interest rates.

GDP deflator or consumer price index?

Just as the Treaty on the Functioning of the EU (TFEU) does not explicitly define which inflation rate can still be regarded as price stable, it just as little does it specify which definition of the rate of price increase should be used.

One option would be to use the domestic production (GDP) deflator. This is supported by the fact that it is more important to allow moderate inflation in domestic production than in consumption in order to alleviate the problems of inflexible wages on the labor market. If the nominal wage is downwardly inflexible, then, from the company's point of view, a negative productivity shock can be better processed if the prices of their own products rise over time. Profit-oriented companies do not relate the nominal wage to the general consumer price level when making employment decisions, but to the price level of their own production output.

The European Central Bank uses the harmonized consumer price index (HICP) for its policy decisions. For this purpose, the changes in costs for purchasing typical national consumption bundles are aggregated to the level of the euro zone. At first glance, certain aspects of the risk of deflation speak in favor of the consumer price index. One problem with deflation could be that consumers postpone purchases of consumer goods in anticipation of further falling prices, which would result in a downturn in self-reinforcing tendencies. This danger is more likely to be averted if there is a steady rise in the consumer price level during normal times

Deflation also harbors dangers in combination with nominally rigid wages. In this case, falling producer prices lead to an indirect increase in real wages. A steady rise in the GDP deflator could therefore also be a more suitable guideline with regard to the risk of deflation than a steady rise in the consumer price index.

In addition, the GDP deflator also includes price increases for construction services and can therefore in principle also reflect cost increases for owner-occupied residential property, provided these are not only due to higher property prices. Even in an economy like the German one, which tends to specialize in the production of capital goods, the sole focus on the consumer price index can be misleading.

In the long term, the differences between the consumer price index and the GDP deflator balance each other out again and again. For a limited time, however, there are clear differences. Figure 1 shows an example of the differences for the euro area and Germany. In the so-called noughties, in which Germany exhibited a certain wage restraint and increasing competitiveness, the GDP deflator was mostly below the consumer price index. After the financial crisis, the eurozone has a less fluctuating inflation rate if one looks at the GDP deflator. In 2015 in particular, there was such a clear difference between the low inflation in consumer prices and the higher rate of increase in the GDP deflator that some economists criticized the ECB for fighting deflation which, when looking at the GDP deflator, does not exist (Alcidi, Busse and Gros, 2016). Interestingly, the original work on the Taylor rule of central bank policy also used the GDP deflator as the starting point, not the consumer price index (Taylor, 1993).

illustration 1
Consumer price index (HICP) and GDP deflator

Source: World Bank, own calculations.

The heterogeneity of consumer price increases

The price development of a consumer basket depends on its composition, and poorer households have different consumption habits than richer households. In developing countries in particular, there is a great deal of research that shows that inflation can differ significantly for different income groups (Arndt, Jones and Salvucci, 2015; Beck, 2015). But also for many EU countries, the data for the years 2001 to 2015 show that the household-specific inflation rates differ significantly. Households with low total expenditures had significantly higher inflation rates than higher-income households. On average over 25 EU countries, the annual inflation rate for the bottom decile of the respective country was around 0.8 percentage points higher than for the top decile (Gürer and Weichenrieder, 2020). In the individual countries, this is mostly due to the higher rates of increase in the price of rents, energy and food, which have a higher share of household expenditure in poorer households than in richer ones. Figure 2 illustrates the various expenditure shares, separated according to the deciles of total expenditure. Deciles with higher numbers generate higher absolute consumer spending. The figure on the left shows purchases with higher average price increases, the figure on the right with below-average price increases. Richer deciles of households tend to consume goods and services with lower price increases, poorer deciles tend to consume goods and services with above-average price increases. Inflation therefore favored richer households for the period under review. The experiences of the past few years do not necessarily have to be continued. But they make it clear that there is no such thing as a single rate of increase in consumer spending.

Figure 2
Distribution-specific inflation in Europe
Share of expenditure on goods with different rates of price increase

Note: Figure is based on the unweighted averages from 25 EU countries in the period 2001 to 2015.

Source: Gürer and Weichenrieder (2020).

To a certain extent, the diversity of inflation rates is also taken into account by the ECB. In addition to the HICP, a large number of communications also focus on core inflation, which does not include the usually greater fluctuations in food and energy prices. However, the development of food prices in particular can, under certain circumstances, have a particularly strong influence on the perceived inflation of consumers. Goods that are bought very often are likely to have a greater impact on perceived inflation (Hintze, 2005). Against this background, there is also the concept of an inflation rate that only includes goods that are frequently bought and not paid for through standing orders - Frequent Out of Pocket Purchases (FROOP), the prices of which determine FROOP inflation. Indeed, FROOP inflation in the euro area is consistently higher than the official inflation rate (Arioli et al., 2017).

Survey results also show that perceived inflation in the euro area is also significantly and consistently above the official HICP rate of increase. For women, who are still likely to buy everyday items more frequently, the perceived inflation is a little further above the official rate of price increase. However, FROOP inflation can only explain a small part of the inflation that is perceived as being too high (Arioli et al., 2017). If the individually felt inflation is reflected in inflation expectations, real effects are also possible. Higher inflation expectations seem to be reflected in lower savings. 2

Asset prices and owner-occupied residential property

The question of the extent to which different inflation rates should be taken into account for different income groups is hardly discussed for monetary policy. This is understandable insofar as the ECB does not have adequate instruments to influence the sectoral structure of inflation, nor does it have a special mandate. In the debate about the correct measure of inflation, it is more likely to be discussed to what extent asset prices should be included in the measurement of inflation. Because the HICP covers the cost of consumer goods, it is logical to neglect assets. Without prejudice to this, an inclusion of the price increases of assets would better indicate expansions of the money supply, which fizzle out without real effects. In addition, it is controversial whether the prices of owner-occupied residential property should not be included in the consumer price. This is not currently the case when calculating the HICP in the euro area.

In countries that take account of owner-occupied residential property, three different concepts are used (Eurostat, 2017, Section 1.3). With the acquisition cost concept, home ownership is taken into account in the calculation of the CPI in a similar way to other durable consumer goods (cars, washing machines): The expenses for new buildings or apartments that are not only sold between consumers are included at the time of purchase, a periodization of the expenses is omitted. Australia, New Zealand and Finland use this concept to measure their national inflation rate. The USA, Japan, Denmark, Norway, Switzerland, the Czech Republic, Mexico and South Africa assume the rent of an equivalent property for owner-occupied residential property (rental equivalence approach) .3 In the usage cost concept, which is used in Canada, Iceland and Sweden, only the ancillary housing costs are included (Maintenance costs, insurance, etc.) included in the rate of price increase.

Because owner-occupied residential property can be understood economically as an insurance instrument against price increases in the rental market, it can be argued that the rental equivalence approach overshoots the target. This speaks more in favor of the acquisition cost concept or an orientation towards the GDP deflator.

Heterogeneous inflation and its policy implications

If, with regard to structural change and nominally rigid wages, one advocates moderate inflation, around 2%, then, given the above discussion, a closer look at the GDP deflator as a background for monetary policy decisions appears sensible. With regard to the description of the welfare perspective, preference should be given to the consumer price index.

Inflation has temporarily different levels for different income groups. As much as it seems appropriate to honestly communicate these differences, it makes little sense to combine the consideration of alternative inflation measures with an expansion of the central bank's political mandate. As in environmental policy, national governments and the legislature have better instruments and better political legitimacy to carry out accompanying redistribution measures.

  • 1 But here, too, the GDP deflator may play a helpful indicator role because it is important for economic development whether purchases of domestic goods are postponed or “only” purchases of imports.
  • 2 For the Netherlands, see Vellekoop and Wiederholt (2019).
  • 3 On country practices, see Hill, Steurer and Waltl (2019).


Alcidi, C., M. Busse and D. Gros (2016), Is there a need for additional monetary stimulus? Insights from the original Taylor Rule, CEPS Policy Brief, No. 342, April.

Arioli, R., C. Bates, H. Dieden, I. Duca, R. Friz, C. Gayer, G. Kenny, A. Meyler and I. Pavlova (2017), EU consumers' quantitative inflation perceptions and expectations: An evaluation, ECB Occasional Paper, No. 186.

Arndt, C., S. Jones and V. Salvucci (2015), When do relative prices matter for measuring income inequality? The case of food prices in Mozambique, Journal of Economic Inequality, 13(3), 449-464.

Beck, U. (2015), Keep it real: Measuring real inequality using survey data from developing countries, WIDER Working Paper, No. 2015/133.

Diercks, A. M. (2019), The reader’s guide to optimal monetary policy, Federal Reserve Board, Working paper.

Eurostat (2017), Technical manual on owner-occupied housing and house price indices, (23 October 2020 ).

Gürer, E. and A. J. Weichenrieder (2020), Pro-rich inflation in Europe: Implications for the measurement of inequality, German Economic Review, 21, 107-138.

Hill, R. J., M. Steurer and S. R. Waltl (2019), Owner-occupied housing, inflation, and monetary policy, Graz Economics Papers, 2019-05.

Hinze, J. (2005), Perceived Inflation, Economic service, 85(12), 800-801.

Taylor, J. (1993), Discretion versus Policy Rules in Practice, Carnegie-Rochester Conference Series on Public Policy, 39, 195-214.

Tobin J. (1972), Inflation and unemployment, American Economic Review,62, 1-18.

Vellekoop, N. and M. Wiederholt (2019), Inflation expectations and choices of households, SAFE working paper, No. 250.