Is the unemployment–inflation trade‐off still alive in the Euro Area and its member countries? It seems so

• DOI: http://doi.org/10.1111/twec.13004
• Published date: 01 September 2020
• Date: 01 September 2020
• Author: Antonio Ribba

World Econ. 2020;43:2393–2410. wileyonlinelibrary.com/journal/twec

|

2393

© 2020 John Wiley & Sons Ltd

Received: 3 December 2019

|

Revised: 29 April 2020

|

Accepted: 6 July 2020

DOI: 10.1111/twec.13004

ORIGINAL ARTICLE

Is the unemployment–inflation trade-off still alive in

the Euro Area and its member countries? It seems so

AntonioRibba

University of Modena and Reggio Emilia, Modena, Italy

KEYWORDS

Euro Area, inflation, structural VARs, unemployment

1

|

INTRODUCTION

In this paper, we wish to investigate the existence of a negative relation, that is, a possible trade-off,

between unemployment and inflation in the Euro Area and (some of) its member countries in the last

20years. Moreover, we also aim to provide some measures of the size of the trade-off.

Since the work by Samuelson and Solow (1960) on the US Phillips curve, as they labelled the

inverse relation between inflation and unemployment discovered by Phillips (1958) for the UK, the

trade-off between these two variables—at least the short-run trade-off—has been a building block of

traditional and less traditional macroeconomic models. The importance of the Phillips curve in the

macroeconomic research over the last 60years is likely to be related to its ability to characterise the in-

teraction between the nominal and the real side of the economic system in a single and simple relation.

Nevertheless, there is no shortage of periodic announcements of the disappearance of this relation.

In a research concerning the US economy, Blanchard (2016) finds that the Phillips relation still

holds in the United States. Another interesting result of this study regards the conclusion that in recent

decades, the relation between unemployment and inflation could be better described in terms of the

old Phillips curve prevailing in the 60s, that is, an inverse relation between the two variables specified

in levels, rather than in terms of the accelerationist model dominating the subsequent periods, in which

following the lead of Friedman (1968) the unemployment rate was related to the changes in inflation.

Blanchard's finding on the reduced importance of the accelerationist model in more recent decades

has been recently confirmed by Gali and Gambetti (2018). In their study, the authors provide estimates

of a wage Phillips curve for the US economy.

In another recent study on the US Phillips curve, Ball and Mazumder (2019a) explore the role of

anchored inflation expectations in explaining the relation between inflation and unemployment in the

more recent decades. Even their results are consistent with Blanchard's conclusion that a relation be-

tween the two variables in levels seems to better characterise the Phillips curve in the last twenty years.

As emphasised by modern research on this subject (see, among others, Mankiw, 2001 and

Ribba,2007), the trade-off is essentially a proposition concerning the response of inflation and unem-

ployment to shocks on the demand side of the economy, as those associated with waves of pessimism/

2394

|

RIBBA

optimism affecting economic agents, with changes in government spending or with unexpected

changes in monetary policy choices undertaken by central banks. Instead, in the presence of shocks

affecting the supply side, as oil shocks or technology shocks, the two variables will move in the same

direction (see e.g., Ball & Mankiw,2002). Indeed, as far as oil shocks are concerned, a very recent

research by Raduzzi and Ribba (2020) has shown that oil price movements have played a notable role

in driving fluctuations of prices and output both at the Euro-area aggregate level and in the small

Euro-area economies over the Economic and Monetary Union (EMU) period.

Thus, and in general, there is no reason to expect that in every historical phase and in every coun-

try, inflation and unemployment will exhibit a negative relationship, since the detection of a positive

or a negative pattern will rest on the specific nature of dominant exogenous shocks hitting the eco-

nomic system.

In this paper, we use the structural VAR methodology to study the trade-off, since it appears to

be a particularly suitable approach for the topic under investigation, owing to its potential ability to

separate demand shocks from supply shocks. We consider the Euro Area and the group of Euro-area

countries that have been part of the EMU since 1999. Moreover, we also include in our investigation

Greece, that joined the Euro Area in 2002. Thus, we select those countries that exhibit an historical

path of adoption of the common currency of around twenty years and we, therefore, exclude those

countries that joined the Euro Area more recently and for which, as a consequence, we do not dispose

of enough data in order to draw sound conclusions.

A bivariate structural VAR including inflation and unemployment was estimated by King and

Watson (1994) to study the trade-off in the post-war period in the United States. Benati (2015), by

using the structural VAR methodology, explores the long-run trade-off between unemployment and

the inflation rate in the United States and other industrialised economies in the post-war period.

However, as far as the Euro Area is concerned, he considers the sample period 1970–98, that is, the

period preceding the start of the currency union.

Battharai (2016) uses panel VAR techniques and presents some interesting results for 35 OECD

countries. In particular, he finds that in the majority of the countries included in the investigation, and

more precisely 28 out of 35, the Phillips relation is empirically significant. However, although Euro-

area countries are also considered in the investigation, the sample data cover the period 1990–2014,

and hence a period over which two different monetary policy and exchange rate regimes have charac-

terised these countries, with national central banks operating in Euro-area countries until 1999.

Abbritti and Weber (2018) have recently studied the relationship between labour market institu-

tions and the behaviour of inflation and unemployment in response to exogenous shocks. They use a

panel VAR model and find a notable effect exerted by a set of labour market institutions on business

cycle fluctuations of Euro-area countries.

Instead, as far as the EMU period is concerned, Bobeica and Sokol (2019) study the evolution

of underlying inflation, that is, HICP inflation excluding energy and food, by using a Phillips curve

framework. They concentrate on the period 2008–18, from the Great Recession (or the Great Financial

Crisis, the expression used by the authors) onwards. Their main conclusion is that while a Phillips

relation can explain the behaviour of inflation over the period 2013–17, it is difficult to interpret the

weakness of underlying inflation in terms of a Phillips curve in more recent quarters.

In related research, Kortela, Oinonen, and Vilmi (2018) find that the Phillips curve is still a useful

tool to interpret the behaviour of inflation in the Euro Area. In particular, the authors conclude that

the behaviour of inflation in recent years can be mainly explained by the strong decrease in oil prices

and by the significant economic slack.

The study conducted by Boeckx, Dossche, and Peersman (2017) aims to investigate the effects pro-

duced in Euro-area countries by the quantitative easing adopted by the ECB in response to the severe