Monetary policy and inflation–output variability in Sri Lanka: Lessons for developing economies

• Published date: 01 February 2022
• Author: Kesavarajah Mayandy,Paul Middleditch
• Date: 01 February 2022
• DOI: http://doi.org/10.1111/rode.12823

Rev Dev Econ. 2022;26:259–279. wileyonlinelibrary.com/journal/rode

|

259

© 2021 John Wiley & Sons Ltd

1

|

INTRODUCTION

While the debate over the trade- off between inflation and output has been widely discussed in both the

theoretical and empirical strands of the business cycle literature, in the context of developing econo-

mies this area is less populated and not without challenge (see Adam & Dercon, 2009). Which is the

more appropriate target for a recently formed central bank presiding over a developing economy? In

this study we draw on the recent experience of Sri Lanka, to suggest that early- stage monetary policy

should focus on inflation targeting as a priority over economic activity. Our findings are consistent

Received: 4 March 2021

|

Revised: 27 July 2021

|

Accepted: 2 August 2021

DOI: 10.1111/rode.12823

REGULAR ARTICLE

Monetary policy and inflation– output variability in

Sri Lanka: Lessons for developing economies

KesavarajahMayandy1

|

PaulMiddleditch2

1Economic Research Department, Central

Bank of Sri Lanka, Colombo, Sri Lanka

2Department of Economics, University of

Manchester, Manchester, UK

Correspondence

Paul Middleditch, Department of

Economics, University of Manchester,

Oxford Road, Manchester M13 9PL, UK.

Email: paul.middleditch@manchester.ac.uk

Abstract

This paper considers the impact of monetary policy on the

inflation– output variability trade- off for Sri Lanka and in

doing so, draws lessons for the formulation of early- stage

monetary policy in developing economies. We examine

how this trade- off has changed across different monetary

policy episodes, and investigate the persistence of eco-

nomic shocks on the variability of inflation and output.

After exploring the contribution of monetary policy to-

wards macroeconomic performance more generally, we use

the experience of Sri Lanka to suggest that more recently

formed central banks in developing economies should focus

on inflation variability, especially where the impact of de-

mand and supply shocks is less persistent.

KEYWORDS

monetary policy, development, inflation targeting

JEL CLASSIFICATION

E52; E58; O23

260

|

MAYANDY et Al.

with De Mendonça and e Souza (2012) who highlight the need for a distinction in approaches for

emerging versus more advanced economies.

In a structural vector autoregressive (SVAR) model, we make use of subsamples to control for

structural changes to the economy, considering the complications presented by supply and demand

shocks. The Sri Lankan economy has registered high levels of growth alongside low levels of inf lation

over the last three decades, and its growth and inflation have been highly volatile with a notable and

regular cyclical behavior. The variability of inflation has fallen noticeably while that of output has

increased significantly, providing us with a natural experiment to measure the connection between

monetary policy in a developing economy with macroeconomic outcomes.

Since Taylor (1979), commentators have argued that the trade- off between inflation and output

might be described more accurately in terms of variability, resulting in a Taylor cur ve more theoret-

ically consistent with mainstream macroeconomics. The variability version of this trade- off, used in

Taylor (1999), suggests that an attempt to stabilize inflation will unfortunately result in larger fluctu-

ations in output, and vice versa for output stabilization. Policy- makers do not face a trade- off between

the levels of inflation and output, but between the variability of the two variables instead, Taylor

(2013). For a newly established central bank, the choice of targets is complicated by the lack of an

established reputation. Should a central bank follow the example of the Federal Reserve, or another

central bank with a history of early success in the area of monetary policy? The historical estimation

of central bank preferences is an obvious source of valuable information, especially the profile of

those preferences over time.

Authors such as Fackler and McMillin (2011) and Olson et al. (2012) have investigated the Taylor

relation for the case of the USA, suggesting that the trade- off between the variability of inflation and

output has changed favorably over time, meaning that the Taylor curve has spent periods closer to the

origin, especially in the 1950s. More generally, Cecchetti and Ehrmann (2002) explore how policy

preferences of a range of central banks have changed over time. Between the 1980s and 1990s aver-

sion to inflation variability increased in all 23 countries, though this has come at the cost of increased

output variability. By contrast Levin et al. (2004) find that inflation- targeting countries do not seem

to exhibit higher levels of variability in output, in comparison to non- inflation- targeting countries.

A later study by Cecchetti et al. (2006) finds that improved efficiency of monetary policy is often

supported by structural changes in the economy and a reduction in the variability of supply shocks.

In the rarer context of developing economies, a study by Ndou et al. (2013) investigates how

demand and supply shocks have affected the var iability trade- off over time for South Africa. The

Taylor curve tends to shift inwards during episodes of str icter inflation targeting. Also with a regime

approach, Onyukwu et al. (2011) investigate how monetary policy affected the nature of the trade- off

for Nigeria between 1995 and 2007. The magnitude of the impact of monetary policy varies across

different policy regimes. The findings suggest that monetary policy has a robust impact on output

growth compared to price stability, depending on the type of monetary policy regime in place; the

volatility of output and inflation becomes more persistent during periods when indirect control of

monetary policy is in place.

Using monthly time series data and an SVAR model, we examine three different monetary policy

regimes. Our choice of samples considers the following background for the case of monetar y policy in

Sri Lanka. Since the economic and financial sector reforms undertaken from 1977, the Central Bank

of Sri Lanka (CBSL) has moved from direct to market- based policy instruments, formally adopting a

monetary targeting policy framework from the 1980s. Under this framework, t he CBSL has conducted

monetary policy with the aim of maintaining reserve money as an operating target and broad money

as an intermediate target for economic stability. Further, since January 2001 the CBSL has allowed

the exchange rate to float freely. This has reduced the role of the exchange rate as a stabilization tool,