Measuring the Financial Cycle in South Africa

• Author: Greg Farrell,Esti Kemp
• Published date: 01 June 2020
• Date: 01 June 2020
• DOI: http://doi.org/10.1111/saje.12246

South African Journal of Economics Vol. 88:2 June 2020

doi: 10.1111/saje.12246

123

© 2020 Economic Society of South Africa

MEASURING THE FINANCIAL CYCLE IN SOUTH AFRICA

GREG FARRELL† AND ESTI KEMP*,‡,§

Abstract

We measure the financial cycle of South Africa using three different methodologies. The financial

cycle is identified using credit, house prices and equity prices as indicators, and estimated using

traditional turning-point analysis, frequency-based filters and an unobserved components model-

based approach. We then consider the financial cycle’s main characteristics and examine its

relationships with the business cycle. We confirm the presence of a financial cycle in South Africa

that has a longer duration and a larger amplitude than the traditional business cycle. Developments

in measures of credit and house prices are important indicators of the financial cycle, although the

case for including equity prices in the measures is less certain. Periods where financial conditions

are stressed are associated with peaks in the financial cycle, suggesting that the estimated financial

cycle may have similar leading indicator properties to financial conditions or stress indices.

JEL Classification: E44, E61, G21

Keywords: Financial cycle, business cycle, spectral analysis, band-pass filters, turning points,

unobserved components, credit, house prices

1. INTRODUCTION

This paper sets out to measure the financial cycle in South Africa. Financial cycles provide

a broad perspective on the evolution of risks to financial stability, and therefore provide a

useful monitoring tool for policymakers who are required to set macroprudential policies.

A robust measure of the financial cycle is particularly important for South African poli-

cymakers at the present time, given the renewed emphasis on the financial stability regu-

latory and supervisory framework provided by the Financial Sector Regulation Act, which

was signed into law in September 2017. A study of South Africa’s financial cycle may also

be of wider interest given the country’s experience of not having suffered a systemic bank-

ing crisis in the period since 1970.1

1 According to Laeven and Valencia (2018), a standard reference for information on banking

crises. This has a number of interesting implications, including perhaps for studies of growth.

Rancière et al. (2008), e.g. note that countries that have experienced occasional financial crises

have, on average, grown faster than countries with stable financial conditions.

* Corresponding author: Esti Kemp, Pretoria, South Africa. E-mail: estivwdv@gmail.com

† School of Economics and Finance, University of the Witwatersrand

‡ South African Reserve Bank

§ School of Economics, University of Cape Town

South African Journal

of Economics

124 South African Journal of Economics Vol. 88:2 June 2020

© 2020 Economic Society of South Africa

Understanding financial cycles is viewed as critical for informing the use of countercy-

clical macroprudential policy.2 Che and Shinagawa (2014), e.g. find that attempting to

improve the financial soundness of banks during a downturn of the financial cycle could

amplify the cycle, and that policymakers should therefore be careful about the timing of

regulatory changes. However, despite their importance for policymakers, there is neither

consensus regarding the definition of financial cycles nor on the methodology that should

be employed to measure them. Furthermore, even though there is a large and growing

international literature,3 we are also not aware of published research that assesses the op-

tions available for measuring the South African financial cycle.

To fill this gap, we propose identifying the main characteristics of the financial cycle in

South Africa using three different approaches. This is motivated by the current lack of

consensus on the best method to use to measure financial cycles,4 and informed by the

most popular approaches in the literature. First, we apply traditional turning point anal-

ysis to identify the financial cycle by detecting peaks and troughs in the individual com-

ponent variables that make up the cycle. Second, we employ a frequency domain approach

that uses band-pass filters to isolate the cycles that correspond to medium-term frequency

intervals. Third, we use a multivariate model-based approach to extract cycles using un-

observed components time series models. We then provide a comparison of the results of

the three approaches, and compare the estimates of the financial cycle with those of the

business cycle to determine whether the cycles are distinct from one another. We begin,

however, by defining financial cycles and selecting a set of financial variables that can

potentially capture the main characteristics of the South African financial cycle.

2. DEFINITIONS AND DATA TRANSFORMATIONS

A working definition describes the financial cycle as reflecting self-reinforcing feedbacks

within the financial system and between the financial system and the real economy

(Borio, 2014b). Approaches to measuring financial cycles have therefore focused on the

co-movement of a broad set of financial variables (Bank for International Settlements,

2015). However, given macroprudential policy’s focus on systemic risks and the chal-

lenges of measuring risk perceptions, it is not clear which set of financial variables or

indicators best captures the financial cycle.

The indicators that have been found to give the most parsimonious description of the

financial cycle are credit and property prices (Drehmann et al., 2012; Borio, 2014b).

Credit aggregates (which can be used as a proxy for leverage) are often the sole focus

2 An important question for policymakers is whether macroprudential policy should be aimed at

controlling the ‘financial cycle’ or not. See Borio (2014a:32-33) and Constâncio (2014) for differ-

ing views on this issue.

3 Claessens and Kose (2017) provide a recent review of studies that examine the features of busi-

ness and financial cycles, and the linkages between them, for the credit, equity and housing

markets.

4 The Bank for International Settlements (2016:Box III.A) draw an analogy with the business

cycle here: it is often identified with movements in GDP, but no consensus exists on the best

method to use even after many years of research.