On secular stagnation and low interest rates: Demography matters

• Author: Stefano Neri,Marco Gross,Giuseppe Ferrero
• DOI: http://doi.org/10.1111/infi.12342
• Published date: 01 December 2019
• Date: 01 December 2019

DOI: 10.1111/infi.12342

ORIGINAL ARTICLE

On secular stagnation and low interest rates:

Demography matters

Giuseppe Ferrero

1

|

Marco Gross

2

|

Stefano Neri

1

1

Banca d’Italia, Economic Outlook and

Monetary Policy Directorate, Rome, Italy

2

Monetary and Capital Markets

Department, International Monetary

Fund, Washington, District of Columbia

Correspondence

Stefano Neri, Banca d’Italia, Economic

Outlook and Monetary Policy

Directorate, Rome, Italy.

Email: stefano.neri@bancaditalia.it

Abstract

Nominal and real interest rates in advanced economies have

been declining since the mid-1980s and reached historically

low levels after the outbreak of the global financial crisis.

Understanding why interest rates have fallen is of utmost

importance for monetary policy. This paper focuses on one

of the factors contained within the ‘secular stagnation’

view: adverse demographic developments. The empirical

analysis shows that these developments have exerted a

downward pressure on real interest rates in the euro area in

the last decade. Building on the projections of the

dependency ratios produced by the European Commission,

we show that the foreseen changes in the age composition of

the population may continue exerting downward pressure

on real interest rates.

1

|

INTRODUCTION

Nominal and real short- and long-term interest rates have been decreasing since the mid-eighties. These

rates reached historical low levels after the outbreak of the global financial crisis. Albeit falling

inflation expectations and risk premia have contributed to these developments, as central banks

increasingly focused on price stability, the fall in nominal rates has been associated primarily with a

decline in the real component. After the outbreak of the global financial crisis, monetary policies turned

very accommodative, also by means of unconventional measures, further lowering interest rates.

Understanding why interest rates have fallen is of utmost importance for monetary policy and

financial stability. If interest rates are low in ‘normal times’, monetary policy may be constrained by the

effective lower bound (ELB) to the policy rates when faced with a recession (Kiley & Roberts, 2017).

This is one of the key lessons of the global financial crisis: the probability of hitting the ELB is higher

than previously thought, as the real short-term rate consistent with output being at its potential,

unemployment at its natural level and inflation on target (the so-called ‘natural rate of interest’) has

International Finance. 2019;1–17. wileyonlinelibrary.com/journal/infi © 2019 John Wiley & Sons Ltd

|

1

d

r

B

o

a

a

l

w

o

p

b

r

‘

a

s

e

‘

e

p

f

d

c

i

f

p

o

e

b

i

E

(

p

e

t

H

r

r

r

b

t

l

2

F

© 2019 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/infi International Finance. 2019;22:262–278.

262

e

g

s

a

d

d

e

.

r

,

s

1

declined. Low interest rates pose risks to f inancial stability by reducing the prof itability and

resilience of financial insti tutions and encouraging exc essive risk-taking (Europ ean Systemic Risk

Board, 2016).

The objective of this paper is to provide evidence of the impact of demographic developments

on real interest rates and other macroeconomic variables in the euro area using a panel vector

autoregressive (VAR) model.

1

To the best of our knowledge, this is the first study on the euro

area.

We do not aim at quantitatively and jointly assessing the role of all the factors considered in the

literature.

2

We rather focus only on demographic developments, and in particular ageing, a key factor

within the secular stagnation view. Compared with technological innovation, another potential driver

of real interest rates, which is notoriously very difficult to predict, demographic developments are very

persistent and potentially more predictable. Demographic variables are also likely to be little affected

by business cycle fluctuations in macroeconomic variables.

Two explanations for the decline of interest rates have been put forward in the literature: one

relying on financial-cyclical factors (the ‘financial cycle’view), the other on structural factors (the

‘secular stagnation’view). Within the latter, some theories have focused on the structural deficiency of

aggregate demand, others on supply-side factors. We briefly review some of these theories in this

section. We refer the reader to Teulings and Baldwin (2014) and Ferrero and Neri (2017) for a more

extensive discussion.

3

Within the financial cycle view, Borio (2014) and Lo and Rogoff (2015) argue that during the

‘Great Moderation’, financial deregulation, excessively loose monetary policies, and overly optimistic

expectations about future asset returns led to a large increase in the supply of funds, to compressed risk

premia, and to lower interest rates. The correction in the financial cycle after the outbreak of the global

financial crisis, followed by a persistent contraction in aggregate demand and by an increase in the

demand for safe assets, led to a further decline in interest rates. Koo (2014) focuses on the

consequences of balance sheet recessions. The greater the damage to the balance sheet of financial

institutions, the longer it takes to clean them up and for the economy to recover. According to the

financial cycle view, when the deleveraging process ends and expansionary monetary policies are

phased out, interest rates will increase from their historically low levels.

Within the secular stagnati on camp, Summers (2014) and Eggertsson and Mehrotra (20 14) focus

on demand-side factors. Summ ers (2014) argues that the r eal interest rate consistent with full

employment has declined ov er the last decades in the adva nced economies. This declin e may have

been caused by slower popula tion growth, lower-priced capital goods, rising inequality, and th e

increased demand for safe a ssets. To the extent that the nominal short-term r ate cannot fall below its

ELB, the balancing of saving and investment may not be achievable . Eggertsson and Mehrotra

(2014) develop a simple overla pping generation model in wh ich the ‘equilibrium’real rate can be

permanently negative due to excess sav ings. With regard to supply-side fac tors, Gordon (2014)

examines the role of technolog ical innovation and argues tha t technological progress ha s slowed in

the last decades to its historic al average, after a period of sus tained growth between 1930 and 1 980.

However, there is no clear evid ence of the role of productivi ty growth in determining rea l interest

rates (Lunsford & West, 2017). Bl anchard, Furceri, and Pescat ori (2014) argue that the facto rs

responsible for the decline of r eal interest rates before the gl obal financial crisis are unli kely to be

reversed. Rather, they could be reinforced by some of its legacies, keeping the nat ural rate of interest

below pre-crisis levels.

Within the secular stagnation view, demographic developments play an important role. These

trends may affect both the demand and supply sides of the economy. Lower fertility and longer

longevity, which lead to a higher old-age dependency ratio, require more savings for old age, unless

2

|

FERRERO ET AL.FERRERO ET AL.

|

3

263

8.