A service of the

Download article as PDF

Recent price developments in the euro area demonstrate that deflation is a serious issue of concern. Some euro area countries are already facing a deflation in the usual macroeconomic sense of a “general price decline”. This article discusses the economic consequences of deflation and suggests that the ECB’s expansionary monetary policy was inevitable.

The possible economic consequences of deflation are high-priority issues for most central bankers and economists, but they have never attracted much attention among the general public in Germany.1 The picture has changed in the last couple of years, as the sustained crisis in many euro area countries has raised concerns about the stability of the euro area as a whole and has stimulated political debates about the fiscal imbalances and redistribution issues among member states. The recent decision by the European Central Bank (ECB) to embark on a massive quantitative easing (QE) programme – for several reasons quite a few years behind the schedule of other large central banks – has led public commentators in Germany to claim that the German tradition of stability-oriented monetary policy has been forsaken by the ECB.2 This step, it is argued, will possibly result in a sustained softening of the euro, thereby destroying the wealth accumulated by prudent savers, among other adverse effects.3

The first purpose of this article is to discuss briefly the economic consequences of deflation. The second objective is to present publicly available empirical facts about recent price developments in the euro area, which demonstrate that deflation is a serious issue of concern. Specifically, we argue that Europe was on the brink of disaster and – given the imposed constraints of austerity obsession and fiscal policy hands-tying – the ECB’s monetary easing was inevitable.4

A refresher course in deflation economics

Deflation is defined as “a general decline in prices, with emphasis on the word ‘general’”.5 Macroeconomists usually are very clear about the distinction between relative price changes versus a general decline in the price level. However, sector-specific price declines might affect the skewness of all relative price changes,6 which has lasting impacts on the aggregate inflation rate as a consequence of nominal price rigidities. However, a process of self-enforcing deflation only becomes relevant in situations of protracted insufficient aggregate demand. As Ben Bernanke stated several years before the recent financial crisis and subsequent Great Recession:

Deflation is in almost all cases a side effect of a collapse of aggregate demand, a drop in spending so severe that producers must cut prices on an ongoing basis in order to find buyers. Likewise, the economic effects of a deflationary episode, for the most part, are similar to those of any other sharp decline in aggregate spending – namely, recession, rising unemployment, and financial stress.7

At least two channels through which deflation affects economic activity are regularly cited within the deflation literature: the debt-deflation theory arguments dating back to Keynes and Fisher and the aggregate demand/expectations nexus.8 In short, for debt-deflation theory proponents, deflation is seen as a redistributionary process between debtors and creditors as long as contracts are fixed in nominal terms. This creates a huge burden on creditors, possibly leading to a credit crunch. Pigou argued that such an adverse effect might well be overcompensated by an increasing real value of aggregate wealth, giving rise to a stimulating effect on consumption and aggregate demand as a whole.9 Such an effect could arise in “modern macroeconomic models” because of inter-temporal consumption behaviour (more wealth over the life cycle, ceteris paribus, implies less savings today and more consumption today, given that net wealth exists). This first channel is mainly the concern of financial intermediaries, central bankers and professional investors, but it seems to be of minimal interest to the general public, at least in Germany.

There is a second channel, which could be labelled the “aggregate demand/expectations nexus”, which asserts that deflation, as a process of general price declines driven by insufficient demand, may turn self-reinforcing if persistent expectations of future deflation become strong. In a world with forward-looking consumers and investors, it is reasonable to assume that consumption and investment expenditure is shifted into the future if further price declines are expected, thus reinforcing the deflationary spiral.

Under such circumstances, it is necessary for monetary policy to act fast and, more specifically, to act before expectations are anchored to future deflation. This makes credible, strongly committed and timely monetary policy actions crucial. As Ben Bernanke argued, the deflation-fighting argument is strengthened if it is assumed that deflation is “in almost all cases” a phenomenon of insufficient demand, which according to any macroeconomic school (at least in the short run) can and should be cured by monetary expansion.10 However, nominal interest rates can only be lowered to the zero bound, unless central banks adopt unconventional measures to pursue their goals. This is what happened with the ECB’s decision to “lift all boats” through unconventional monetary easing. Several observers – most prominently former ECB chief economist Ottmar Issing – argued recently that the ECB was (extremely) overreacting and that deflation was not a big issue.11 In the following, we present arguments demonstrating why this is not the case.

Figure 1
Headline inflation rates
in %

Note: Unweighted HICP. Last observation is December 2014.

Source: Eurostat; own calculations.

Arguments and assessments

We begin by looking at the development of general price changes. Figure 1 presents inflation rates for several European countries as well as for the euro area as a whole. Year-on-year inflation rates dropped to zero or even turned negative for most euro area member countries during the last several months. This is most pronounced for Greece and Spain. One could argue that this development is to a large extent driven by falling energy and food prices.12

Figure 2
Core inflation rates
in %

Note: Unweighted HICP. Last observation is December 2014.

Source: Eurostat; own calculations.

This is surely a reasonable argument, given the sustained decline in energy and food prices over the last two years. We therefore calculated price indices excluding food and energy prices (see Figure 2). Clearly, this measure of core inflation did not fall in such a dramatic fashion as the overall inflation rate. Nevertheless, the core inflation rates of Greece, Spain and Portugal have fallen close to or even below the zero bound. Furthermore, one can observe that core inflation rates in the remaining economies are also far below the ECB’s target of below, but close to, two per cent. Looking at the recent historical evidence, one could argue that the current development looks a bit like 2009 – when the ECB escaped a deflationary situation without relying on a QE programme of such a massive dimension. It is worthwhile, however, to analyse the data in more detail.

Figure 3
Fractions of price changes in selected ranges

Note: The length of the bars indicates the fraction of (the 93) product categories with annual inflation rates belonging to a given interval. The black line indicates the percentage of items with negative inflation rates. Last observation is December 2014.

Source: Eurostat; own calculations.

Therefore, we looked at the distribution of price changes across all main 93 product categories contained in consumer price indices calculations (Figures 3 and 4). At first glance, whether looking at the euro area as a whole or specifically Germany, the situation does not differ much from 2009. However, things are different in Italy, Spain, Portugal and Greece. The fraction of product categories characterised by falling prices has lately increased to 30 per cent in Italy and ranges between 50 per cent and 80 per cent in Spain, Portugal and Greece. The figures furthermore show a lasting tendency and an increase in the percentage of items with low or falling inflation rates in the last couple of months. The data indicate that in the southern euro area countries, prices are already falling on a wide scope, which is totally different from the situation in 2009. We also replicated the exercise for all product categories excluding energy and food. However, this does not change the qualitative argument. These indicators strengthen our argument that we face a general price decline rather than merely relative price declines in the southern euro area countries that have been induced by lower energy prices. This statement is supported by Figure 4, which displays box plots of annual price changes for the 93 product categories.13

Figure 4
Distribution of price changes
Inflation in %

Note: Distribution of annual inflation rates using all 93 unweighted items. Last observation is December 2014.

Source: Eurostat; own calculations.

The width of the box shows the interquartile range. The line inside the box indicates the median and the dot in the box represents the mean. If mean and median deviate, we usually have a skewed distribution. Figure 4 makes clear that, while inflation rates declined in almost all countries, this happened in a very pronounced manner in Greece, Spain and Portugal. In 2014 the distribution became more skewed in several countries, which indicates that relative price adjustments played a role. More importantly, however, the dispersion of price changes became smaller in several countries and especially for the euro area as a whole. Such a narrowing of the distribution usually occurs if the concept that “prices in general move with the same tendency” takes precedence over relative price changes, which usually are accompanied by more dispersed distribution patterns across product prices.14 For Spain, Italy and Greece, we can infer that we do not merely face the danger of possible future deflation; indeed, deflation in the usual macroeconomic sense is already present, as prices in general show a falling tendency broadly across all product prices.

Figure 5
Probability distribution of inflation forecasts

Note: Based on values reported in the first quarter.

Source: Eurostat, own calculations.

In order to visualise the current operation of the expectations channel, we study recent survey-based inflation expectations of firms and households. We can proxy firms’ expectations by using data from the ECB’s Survey of Professional Forecasters (SPF). Figure 5 shows average probability distributions of inflation expectations for the current and following years based on the ECB surveys, which are always conducted in the first quarter of the respective year. Unfortunately, quantitative consumer inflation expectation data are rare for the euro area. Instead, we use the balance statistics of the EU Commission household survey (see Figure 6).15

Figure 6
Consumer inflation expectations
balance statistics

Note: The balance statistic is a weighted average of the fractions of answers, excluding the non-response category. The balances show the difference between positive and negative answers, measured as percentage points of total answers. Balance values range from -100, when all respondents choose “decreasing prices” to +100, when all respondents choose “increasing prices”. Shaded areas indicate recent periods of falling inflation expectations.

Source: European Commission: Joint Harmonised EU Programme of Business and Consumer Surveys.

There are two aspects which should be mentioned. First, inflation expectations among both professional forecasters and households have declined tremendously over the last two years. Second, we can infer increasing uncertainty from the SPF data, for which information on the cross-sectional distribution is available. The data clearly show that the professionals gave increasing weight to the possibility of falling prices in the euro area as a whole. This gives rise to serious concerns about the stability of medium- to longer-term inflation expectations. Competent central banks should and do care about this issue.

Last but not least, there is a further important channel that is mostly ignored in the Ordnungspolitik discussion that is still ongoing in Germany: there has been a significant and lasting increase in policy-induced uncertainty in the euro area in recent years, implying that the assessment of policy-induced adjustment results, in terms of growth and inflation forecasts, is extremely difficult for market participants as well as for the general public.16

Figure 7 reveals that the levels of economic policy uncertainty in the US and the euro area were broadly similar before the Great Recession, but they have decoupled from each other during the last several years. Specifically, policy uncertainty is more volatile in Europe and is currently at least 50 per cent higher than in the US. There are different channels through which increased uncertainty adversely affects economic activity – ranging from investment to consumption to labour market channels17 – but a detailed discussion is beyond the scope of this paper. However, it should be clear that the danger of persistent stagnation and hence a lasting deflationary period is much stronger in Europe than in the US, as economic policy uncertainty in Europe is more pronounced.

Figure 7
Economic policy uncertainty

Note: A value of 100 is a normalised value of the average number of articles that contain terms related to economic policy uncertainty based on automated text-search results for a representative sample of newspapers in the respective countries.

Source: http://www.policyuncertainty.com.


There are several conclusions to be drawn from the discussion above. First and foremost, deflation is a serious issue, and the empirical facts support the interpretation that some Southern euro area countries – namely Greece, Spain and Portugal – are already facing a deflation in the usual macroeconomic sense of a “general price decline”. Second, the inflation expectations of experts and households have shifted significantly downwards over the last year. Third, economic policy uncertainty in the euro area increased after 2010 and has remained at historically high levels. All of these findings justify strong expansionary macroeconomic policy measures. The question remains: is monetary policy the right tool with which to address this issue? Several authors raise concerns that QE in the euro area might be ineffective and that the prospective fiscal burden in the long run might not justify drastic measures.18 This argument has several dimensions. First, the independence of central banking could be negatively affected; second, quantitative easing possibly comes with a “sin” of prospective budget financing; and third, the turn of the ECB towards massive monetary easing might be a reflection of construction failures in the economic policy constitution of the European Economic and Monetary Union. All of these arguments have validity to a certain extent.

First, the credibility of the central bank is surely affected. Compared to other central banks, the ECB has acted slowly in moving towards quantitative easing over the last couple of years. This in itself reflects the fact that it was extremely difficult to eliminate certain ideological firewalls, as there was always a fear of the original sin. QE, however, is the logical continuation of the “whatever it takes” stance the ECB has adopted under Mario Draghi’s leadership. The “whatever it takes” announcement became necessary as the very existence of the euro area as a common currency area was endangered by the European political blockade on fiscal expansion in the crisis-stricken countries. The ECB’s decision was never a pure “lender of last resort” decision but rather a very political one, in a broad sense. To put it differently, it was a decision to take responsibility for the survival of the currency union in periods of broad political irresponsibility. In that respect, the ECB has lost its political innocence, as its “whatever it takes” stance is undoubtedly a political statement.

Second, the argument that QE is irresponsible as it leads to heavy debt burdens for future generations is flawed. It remains implausible to hope that the broad institutional malfunctioning which makes the coordination of fiscal policy not only impossible but undesired comes at “no costs”. The ongoing political blockade of fiscal expansion and the bargaining over fiscal austerity require an aggressive monetary policy stance. In our view, the question is not whether there were better solutions but – due to the urgency of the problem of self-reinforcing deflationary spirals – whether there were any better solutions at hand.

Third, the ECB decision does not solve the political blockade in the euro area that prevents a rethinking of austerity. Instead of moving towards expansionary macroeconomic policy for the currency area as a whole, we face a situation where institutions and governments have become stuck in arguments of moral hazard, conditionality and austerity.

To sum up, we argue that a serious and dangerous deflationary situation is already at work in several southern euro area countries. Given the worldwide macroeconomic consensus that deflationary processes should be avoided, the choice for a QE programme was surely right. To blame the European Central Bank for overstepping its mandate and for saving the euro area from breaking up is like blaming a car driver for avoiding a crash by steering around it. It is possible that the developments in the next few months might lead to a “Grexit” or even the breakup of the euro area. Nonetheless, the ECB and its QE policy should not be blamed for such a result, as the Bank is doing all it can to prevent such a development.

  • 1 For long-term trends, see for example Google’s Ngram viewer using the search items “Inflation” and “Deflation” for books published in German.
  • 2 See A. Winkler: Auf dem Tiefpunkt: Die deutsche geldpolitische Debatte, Ökonomenstimme, 27 February 2015; and A. Winkler: Auf dem Prüfstand: die Argumentation gegen die EZB-Staatsanleihenkäufe, in: Wirtschaftsdienst, Vol. 95, No. 3, 2015, pp. 178-185.
  • 3 To some extent, the economic policy debate among German economists seems to be de-coupled from international strands of argumentation. See also J. Muellbauer: Combatting Eurozone deflation: QE for the people, VoxEU, 23 December 2014.
  • 4 See K. Bernoth, P. König, C. Raab, M. Fratzscher: Unbekanntes Terrain: Anleihekäufe der Europäischen Zentralbank, in: Wochenbericht des DIW Berlin, No. 13, 2015, pp. 307-316.
  • 5 B. Bernanke: Deflation: Making Sure “It” Doesn’t Happen Here, remarks by Governor Ben S. Bernanke before the National Economists Club, Washington DC, 21 November 2002.
  • 6 L. Ball, G.N. Mankiw: Relative-Price Changes as Aggregate Supply Shocks, in: Quarterly Journal of Economics, Vol. 110, No. 1, 1995, pp. 161-193.
  • 7 B. Bernanke, op. cit.
  • 8 J.M. Keynes: The consequences to the banks of the collapse of money values, in: J.M. Keynes: Essays in Persuasion, London 1931, Macmillan; I. Fisher: The Debt-Deflation Theory of Great Depressions, in: Econometrica, Vol. 1, No. 4, 1933, pp. 337-357.
  • 9 A.C. Pigou: The Classical Stationary State, in: Economic Journal, Vol. 53, No. 212, 1943, pp. 343-351.
  • 10 B. Bernanke, op. cit.
  • 11 O. Issing: Die Deflations-Diskussion grenzt an Hysterie, in: Neue Zürcher Zeitung, 20 January 2015; M. Hellwig: Jens Weidmanns gefährliche Argumente, Frankfurter Allgemeine Sonntagszeitung, 4 January 2015.
  • 12 O. Issing, op. cit.; K. Bernoth et al., op. cit.
  • 13 While the changes displayed in Figure 4 are unweighted, the picture does not change qualitatively if we weight the changes by their respective weights in usual HICP calculations.
  • 14 L. Ball, G.N. Mankiw, op. cit.
  • 15 Question 6 in the monthly Joint Harmonised EU Programme of Business and Consumer Surveys refers to price expectations (“prices in general”) for the next 12 months. The balance statistic is a weighted average of the percentage of answers indicating increasing, “staying about the same” and decreasing prices. The more answers shift towards the “decreasing prices” category, the more the balance statistic will fall. The survey data is available for many European countries.
  • 16 S.R. Baker, N. Bloom, S.J. Davis: Measuring Economic Policy Uncertainty, 19 May 2013, http://www.policyuncertainty.com/media/BakerBloomDavis.pdf.
  • 17 N. Bloom: Fluctuations in Uncertainty, in: Journal of Economic Perspectives, Vol. 28, No. 2, 2014, pp. 153-176.
  • 18 K. Bernoth et al., op. cit; and A. Winkler: Auf dem Prüfstand , op. cit.

Download as PDF

DOI: 10.1007/s10272-015-0538-y