Failing to Forecast Low Inflation and Phillips Curve Instability: A Euro‐Area Perspective

• Date: 01 March 2015
• DOI: http://doi.org/10.1111/1468-2362.12062
• Author: Fabrizio Venditti,Marianna Riggi
• Published date: 01 March 2015

Failing to Forecast Low Inflation

and Phillips Curve Instability:

A Euro-Area Perspective

Marianna Riggi and Fabrizio Venditti

Banca d’Italia, Rome, Italy.

Abstract

Professional forecasters failed to anticipate the sharp fall in inflation in

theeuroareain2013–2014. We investigate whether this forecasting

failure can be partly attributed to a break in the elasticity of inflation to

the output gap. Using structural break tests and time-varying parame-

ter models, we find that this elasticity has indeed increased substantial-

ly in the past 2 years. We offer two (observationally equivalent)

interpretations of this result. The first is that the increase in the

cyclicality of inflation has stemmed from lower nominal rigidities or

weaker strategic complementarities in price setting. A second possibili-

ty is that current output gap estimates are understating the amount of

spare capacity in the economy. We estimate that, in order to reconcile

the observed fall in inflation with the historical correlat ion between

consumer prices and the business cycle, the output gap should be wider

by around one third.

*We wish to thank two anonymous referees and the editor for useful comments. Th e views expressed

in this paper are those of the authors and do n ot necessarily reflect tho se of the Banca d’Ita lia.

International Finance 18:1, 2015: pp. 47–67

DOI: 10.1111/1468-2362.12062

© 2015 John Wiley & Sons Ltd

The ECB never expe cts inflation to d eviate from the ta rget of just unde r 2 per cent. Yet

each month inflation undershoots, and the ECB is apparently taken by surprise .

Wol f ga n g M €

unchau, Financial Times, 17 August 201 4

1

I. Introduction

Debate over the Phillips curve has gained momentum since the 2008 financial crisis.

In the course of the recession that followed th at crisis, a puzzle emerge d: inflation in

advanced countries did n ot fall as much as a tradit ional Phillips cu rve and past

experiences would predict , given the severity and the len gth of the recession (se e

Williams 2010 and Ball and Mazumder 2011). Instead, the decline in euro-area

inflation between 2013 and 2014 points in the opposite direction. Following the

sovereign debt crisis, t he euro area fell into a severe recession, w hich generated

sizeable output losse s, especially in Greece, Spa in, Portugal, Italy and Ireland. Th e

recession was followed by a sharp fall in euro-area consu mer price inflation, w ith

core (net of food and energy) inflation dropping to historical ly low levels in mid-

2014. Two aspects of this disinflation stan d out. First, it has o ccurred across

countries, though it ha s been relatively more intense in the countr ies hit the hardest

by the sovereign debt crisis. Second, it was not anticipated by professional

forecasters. This i s particular ly surprising given t hat the fall i n economic act ivity

in most euro-area countries since 2011 has generated significant gaps between

actual and potent ial output in thes e economies.

Two plausible explanations, not mutually exclusive, ca n be put forward. One is

that forecasters underes timated the output gap over t his horizon. Thi s hypothesis

relates to the usual di fficulty of separ ating trends f rom cycles in real ti me, a task

made even more difficult by the severity of the shock to GDP. The issue of

quantifying structural an d cyclical factors be hind economic act ivity is cruci al for

the conduct of monetar y policy and is at the center of the pol icy debate, as testi fied

to by ECB President Mario Draghi ’s 2014 Jackson Hole speech.

2

A second possibilit y

is that forecasters’predictions were cond itioned on an accurate mea sure of the

output gap (where by accurate we mean the output g ap that would have been

available to them ex post)buttheresponseofinflation was stronger than estimated

with data up to 2012. This second hypothesis, which has so far found less echo in

the debate, is the focus of our p aper. Drawing from the econometric l iterature,

which has long identified structura l breaks as the main cause of foreca st failure, we

investigate t hrough str uctural bre ak tests a nd time- varyi ng param eter mode ls

1

M€

unchau, Wolfgang (2014), ‘Draghi i s running out of legal ways to fixtheeuro’,Financial Times,17

Augu st.

2

See http://www.ecb.europa.eu/press/key/date/2014/html/sp140822.en.html.

48 Marianna Riggi and Fabrizio Venditti

© 2015 John Wiley & Sons Ltd