Spillovers of the Conventional and Unconventional Monetary Policy from the US to South Africa

• DOI: http://doi.org/10.1111/saje.12262
• Author: Tumisang Loate,Alain Kabundi,Nicola Viegi
• Date: 01 December 2020
• Published date: 01 December 2020

South African Journal of Economics Vol. 88:4 December 2020

doi: 10.1111/saje.12262

435

© 2020 Economic Society of South Africa

SPILLOVERS OF THE CONVENTIONAL AND

UNCONVENTIONAL MONETARY POLICY FROM THE US TO

SOUTH AFRICA

ALAIN KABUNDI*,† , TUMISANG LOATE‡ AND NICOLA VIEGI§

Abstract

This paper assesses the effect of US monetary policy on South Africa during the period 1990–2018.

We separately analyse and compare the effect of conventional monetary policy, before the Global

Financial Crisis, and unconventional monetary policy, after the US monetary policy reached the

zero-lower bound. Our impulse response function results indicate that monetary policy in South

Africa responds mainly to local inflation, economic activity and financial conditions. While there

is strong correlation between the global and South African financial cycle, the financial cycle is

not transmitted to the real economy because of the sluggish response of industrial production and

domestic credit, especially after the global financial crisis. We see this as an indication of the effects

of structural issues to the real economy and constrained households’ balance sheet which has

prevented the local economy to take advantage of low local interest rates and the global economic

recovery after the crisis.

JEL Classification: E52, F36

Keywords: International spillovers, unconventional monetary policy, zero-lower bound, South Africa

1. INTRODUCTION

A country’s integration in the global financial system raises the question of what are the

exchange rate regime, the monetary policy framework and the financial stability reg-

ulation that guarantee the benefit of financial integration while allowing a significant

degree of independence in local policy making. The choice in South Africa, and in many

emerging countries, has been to rely on a combination of flexible exchange rate, inflation

targeting and free capital mobility. This is in accordance with the Mundellian trilemma

hypothesis (Obstfeld et al., 2005) by which an independent monetary policy is compat-

ible with free international capital mobility only at the cost of letting the exchange rate

be determined by market forces. This has been the consensus in international economics

for decades and it justifies the policy framework chosen in many emerging countries. The

financial crisis has raised questions over the validity of this consensus view. Rey (2016)

* Corresponding author: Department of Economics, University of Pretoria, Tukkiewerf Building,

Cnr Lynnwood and University Roads, Hatfield, Pretoria, South Africa. E-mail: tumisang.loate@

up.ac.za

† The World Bank Group

‡ Department of Economics, University of Pretoria

§ SARB Chair in Monetary Economics, Department of Economics, University of Pretoria

South African Journal

of Economics

436 South African Journal of Economics Vol. 88:4 December 2020

© 2020 Economic Society of South Africa

at the Jackson Hole meeting of 2013 famously raised the possibility that global financial

flows are transmitting the United States monetary policy to the rest of the world. Rey

(2015, 2016) argue that because the US dollar plays a central role in international trans-

actions, US monetary policy is transmitted to other countries through an international

credit and/or risk-taking channel. This transmission generates a strong co-movement in

risky assets across the globe thus generating a global financial cycle. The transmission of

the global financial cycle to local economies compels monetary policy in each country to

react to prevent local financial instability instead of focusing on its main macroeconomic

objective. Any country faces a dilemma: either allow free movement of capital, and lose

monetary independence, or introduce capital controls or macro-prudential tools to gain

renewed control on the instruments and goals of monetary policy. These results are strik-

ing and have strong implications for policy making in emerging countries.

This paper examines empirically the applicability of Rey (2016)’s findings to South Africa.

South Africa is an ideal candidate to test the dilemma hypothesis because it has the most de-

veloped and integrated financial market of any emerging country, with large equity and bond

markets, see Kavli and Viegi (2017). On the other hand its banking system is dominated by

a few local banks, financed by local sources, and thus less affected by resource allocation of

global banks Bruno and Shin (2013). First we analyse the co-movement between global and

local risky financial assets. We investigate this hypothesis by extracting the South African

financial cycle from a panel of financial variables, using the dynamic factor model. We then

analyse the correlation of this South African financial cycle with both the global cycle and

global financial risk. As suggested by Rey (2015, 2016) and Bruno and Shin (2015), the

presence of a high correlation between local and global financial cycles is the first indication

of a possible dependency of South Africa’s monetary policy to the Reserve policy.

The correlation of financial cycles is a necessary but not sufficient condition for de-

pendence. The other condition is that the financial cycle affect directly or indirectly the

setting of monetary policy in the periphery. To assess the channels of transmission of the

US monetary policy to South Africa, we use a medium Bayesian vector autoregressive

(BVAR) model with Minnesota priors, proposed by Bańbura et al. (2010) covering the

period January 1990 to February 2018. We identify US monetary policy shocks and we

then determine its’ spillovers to South Africa. We conduct the analysis by dividing the

sample into two sub-samples, before and after 2008, to account for the dramatic change

in US monetary policy after the global financial crisis, once the Fed funds rate reached

the zero lower bound and the Federal Reserve started its policy large-scale asset purchases

(LSAPs).

Our results indicate that the global financial cycle affects the valuation of risky assets,

as predicted by Rey (2015), but assets prices changes are not transmitted to the real econ-

omy because of a muted credit response. The exchange rate channel is the dominant one

and monetary policy responds mainly to local inflation and economic activity.

Before and after the financial crisis, an expansionary US monetary policy reduces

global risk, increases purchase of stocks by non-residents and leads to an appreciation of

the South African currency against the US dollar. While before the crisis a US monetary

policy expansion had a positive effect on South African real economy by expanding global

demand, after the crisis we see that a US monetary expansion had a contractionary effect

on industrial production and credit. These results highlight local structural issues in the

real sector and constrained households’ balance sheets post the global financial crisis.

437South African Journal of Economics Vol. 88:4 December 2020

© 2020 Economic Society of South Africa

The rest of the paper is organised as follows: after the literature review in the next

section, we present the analysis of the correlation of South African and global financial

cycles, which become highly correlated after the global financial crisis. In Section 4 in-

stead we briefly describe the methodology and the data used in the VAR analysis. The

results for the VAR analysis are then presented in Section 5. Finally Section 6 concludes.

2. GLOBAL TRANSMISSION OF US MONETARY POLICY

The importance of the financial channel in the global transmission of shocks has been

the subject of a large international finance literature, starting from the classical Mundell

(1960), on the assignment of instrument to target in open economy. The central ques-

tion of this literature has been the constraints imposed on national policy making by a

country’s integration into the international financial system. After the global financial

crisis the question has found new answers that have generated a wide debate. Rey (2015)

main argument is that global financial flows act as a transmission of US monetary policy

across the globe and that the exchange rate regime is largely irrelevant in insulating na-

tional economies from these shocks. This observation generalizes to the global economic

considerations that were previously largely limited to emerging countries policy experi-

ence by Calvo (1998), Calvo and Reinhart (2002). Similarly, Obstfeld (2019) elaborates

how the US dollar’s global role explains why the US unconventional monetary policy

had far-reaching effects to both emerging and developed markets than other developed

economies’ unconventional monetary policies. These roles include US dollar as a reserve,

trading “vehicle”, invoicing and funding currencies. According to the author, it is the lat-

ter two that matter more in the transmission of US monetary policy to other countries

via the international trade and global finance respectively.

Because the US dollar is the currency of global finance, a change in US monetary pol-

icy changes the cost of funding for global banks, Adrian and Shin (2009), the prices of

dollar assets in the US and abroad, and the allocation of capital and credit conditions

across the world, Giovanni et al. (2017). International capital flows are procyclical and

potentially destabilizing for the receiving country: a US monetary expansion increases

asset values, reduces the cost of funds and increases the risk appetite of global banks and

global financial intermediaries, Borio and Zhu (2012), Bruno and Shin (2013). This in-

creases international capital flows, increasing asset prices and credit provision across the

globe; the revaluation of local currencies relative to the US reinforces the mechanism by

making local assets more valuable in US dollar terms; this generates a local assets and

credit boom that local monetary policy has little instruments to fight, as argued by Rey

(2016). In this scenario, a flexible exchange rate is irrelevant, if not damaging, in insulat-

ing the country from global shocks. The only instrument available is some kind of capital

controls or macroprudential policy to slow down the procyclical effect of capital flows.1

The normative effect of the analysis is striking: either a country accepts that the benefits

of international capital flows come with potential destabilizing effects and a loss of mon-

etary sovereignty, or it has to impose capital control, or credit control or manage the

1 See Ostry et al. (2012) for a review of the instruments available to manage capital flows and

Chamon and Garcia (2016) for an analysis of capital control measures implemented in Brazil in

the years after the Global Financial Crisis.