The South African Financial Cycle and its Relation to Household Deleveraging

• Published date: 01 June 2020
• Date: 01 June 2020
• Author: Steven F. Koch,Adél Bosch
• DOI: http://doi.org/10.1111/saje.12245

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

doi: 10.1111/saje.12245

145

© 2020 Economic Society of South Africa

THE SOUTH AFRICAN FINANCIAL CYCLE AND ITS

RELATION TO HOUSEHOLD DELEVERAGING

ADÉL BOSCH*,†,‡ AND STEVEN F. KOCH‡

Abstract

This paper considers the extent to which South African households have deleveraged, since the

global financial crisis of 2007/2008. We extend the official South African Reserve Bank business

cycle methodology to date financial cycles, from which we identify the peaks and troughs of the

South African financial cycle going back to 1966. Our composite financial cycle index peaks in

April 1974, January 1984 and May 2007; it has bottomed out in July 1979 and February 1999.

Thus, we still await the trough. We further compare and contrast the deleveraging process in the

current downward phase to the experiences from previous financial cycles. We find that the average

period of the financial cycle in South Africa is much longer (approximately 17.3 years) than that of

the business cycle (approximately 5.8 years), and that deleveraging has not yet matched the degree

of deleveraging seen in previous downward phases. Our results suggest that further deleveraging is

necessary, before we can expect to turn the financial corner.

JEL Classification: E32, E44, E61, G01

Keywords: business cycles, macroeconomy, policy coordination, financial crises

1. INTRODUCTION

The 2007/2008 global financial crisis raised important questions about the behaviour of

financial cycles and the length of time it takes households to deleverage from the peak,

especially when the upward phase of a financial cycle is supported by an asset bubble.

When an asset price bubble bursts, asset prices fall sharply, making it harder for house-

holds to sell off assets (usually not by choice) or to use them as collateral. Interest rates

may fall rapidly in an attempt to support the economy, succeeding in lowering general

debt levels, but that may not fully describe household financial health. Furthermore, the

link between deleveraging and economic activity has not received much attention (see

Dynan, 2012), although Brown et al. (2011), and Bricker et al. (2011) examine balance

sheet adjustment while households deleverage. Thus, we add to a small, but growing

literature that considers the behaviour of the financial cycle, with a focus on households.

Typical post-war recessions, according to Borio (2014) were triggered by monetary

policy attempts to contain inflation. However, when a financial boom originates in an

* Corresponding author: Adel Bosch, Phd candidates, Department of Economics, University of

Pretoria, Hatfield, Pretoria, South Africa. E-mail: Adel.Bosch@resbank.co.za

† Economist, Statistics Department, South African Reserve Bank

‡ Department of Economics, University of Pretoria

The views expressed herein are those of the author and do not reflect the official views of the South

African Reserve Bank Group.

South African Journal

of Economics

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

© 2020 Economic Society of South Africa

environment of low and stable inflation, but then bursts, it becomes what Koo (2013)

refers to as a “balance sheet” or financial cycle recession, where companies seek to repay

excessive debts instead of seeking profit. Reinhart and Rogoff (2009) show that financial

cycle recessions are deeper and followed by weaker recoveries, because, as Borio et al.

(2014) suggests, policy makers in these situations have little room for policy manoeu-

vring. In 24 advanced economies, since the 1960s, Bech et al. (2014) find that monetary

policy is relatively ineffective in a balance sheet recession and its subsequent recovery.

Evidence also suggests that the amplitude, length and potential disruptive force of the

financial cycle are closely related to the financial, and possibly also, monetary regimes in

place (e.g. Lowe and Borio, 2002; Drehmann et al., 2012).

Since the global financial crisis, international literature has advanced our understand-

ing of the impact of asset prices and financial markets on the real economy and business

cycles. These financial drivers (excessive growth in house prices, credit extension and asset

prices – an asset bubble) manifest in what is commonly referred to as the financial cycle

(Adarov, 2018). Most empirical research on the financial cycle focuses on developed

economies or groups of emerging market economies (see Borio, 2014; Borio et al., 2017;

Claessens et al., 2011, 2012; Drehmann et al., 2012; Schularick and Taylor, 2012;

Aikman et al., 2015; Gonzalez et al., 2015),1 while empirical findings from South Africa

are focussed mainly on determining and characterising the financial cycle (Boshoff, 2005,

2010; Kabundi and Mbelu, 2017; Farrell and Kemp, 2018). What is missing for South

Africa, is an analysis of the aftermath of the global financial crisis, when policy making

and its timing is most crucial. A full analysis of credit supply conditions, however, falls

outside of the scope of this paper.

There are a growing number of studies measuring and describing financial cycles. They

generally find that such cycles display vastly different properties than business cycles,

which have, in general, received more attention.2 The financial cycle is usually measured

by three key financial variables, credit, equity and house prices (e.g. Kindleberger and

Aliber, 2005; Minsky, 1992; Claessens et al., 2011, 2012). Farrell and Kemp (2018)

consider similar information, when they determine the financial cycle for South Africa.

In this research, we confirm the main points from their analysis, applying different meth-

ods. We also extend their research by comparing the behaviour of debt, with a particular

focus on the deleveraging process, across the financial cycles that we were able to date.

Thus, we are able to provide a series of stylised facts about the behaviour of financial

variables during the financial cycle; such information should be of use to policy makers –

monetary, regulatory and fiscal – as it can be used to underpin the design of optimal

policy.

We uncover a clear debt build-up in the lead-up to the 2007/2008 global financial

crisis, preceding a slow and difficult deleveraging process that continues. We identify the

1 Borio et al. (2017) looks at the US, while Borio (2014) and Drehmann et al. (2012) look at 7

developed countries. Claessens et al. (2011) analyses 21 advanced economies and Claessens et al.

(2012) divide 44 countries into 21 advanced countries and 23 emerging market economies.

Schularick and Taylor (2012) and Aikman et al. (2015) look at 14 developed economies, while

Gonzalez et al. (2015) groups South Africa together with 27 other countries.

2 Burns and Mitchell (1946) originally proposed the business cycle definition that has supported

decades of research on the business cycle.