Capital flows to Asia and Latin America: Does institutional quality matter?

• DOI: http://doi.org/10.1111/twec.12783
• Author: Vandana Arya,Tony Cavoli,Rajabrata Banerjee
• Date: 01 July 2019
• Published date: 01 July 2019

ORIGINAL ARTICLE

Capital flows to Asia and Latin America: Does

institutional quality matter?

Vandana Arya

|

Rajabrata Banerjee

|

Tony Cavoli

School of Commerce, University of South Australia, Adelaide, South Australia

KEYWORDS

capital flows, emerging economies, Institutional quality, productivity

1

|

INTRODUCTION

By the end of the twentieth century, many Asian economies, particularly the East Asian econo-

mies, experienced rapid economic growth. In contrast, the Latin American countries witnessed a

complete reversal of fortune (IMF, 2006).

1

For example, the average per capita GDP of the Asian

economies in 1950 was only 11% of the USA in 1950. However, this share jumped to 27% in

2001 and further increased to 39% in 2015. In comparison, the average per capita GDP of the

Latin American economies was 30% of the USA in 1950, which dropped to 23% in 2001 and

moderately increased to 28% in 2015.

2

This raises concerns about why the Latin American econo-

mies lagged behind the Asian economies at the start of the twenty‐first century. After comparing

the growth differences between these two regions, the factors that have been recently highlighted

are liberalisation of trade and capital accounts, country‐specific institutional quality (IQ), external

debt, education and export‐oriented foreign direct investment (FDI; Sachs & Williamson, 1985;

Solimano, 2006).

With the liberalisation of trade and capital accounts in 1980s and 1990s, there was a substantial

rise in all types of capital flows in the emerging market economies (EMEs), such as FDI, portfolio

flows and other flows (mainly, bank flows). However, very few studies account for the relative

importance of each of these capital flows to measure the growth differences between Asia and Latin

America. Moreover, while there is some evidence that the marginal effect of capital flows is condi-

tional on the quality of institutions of the host countries,

3

there is no estimate of the threshold level

of institutional quality at which the marginal effect is positive. Thus, the main purpose of this paper

was to examine the role of institutional quality in determining the marginal effect of capital flows

on GDP per capita growth and total factor productivity (TFP) growth of Asia and Latin America in

the period 1990–2013. We use dynamic panel regression techniques to estimate the threshold level

of institutional quality at which capital flows exert a positive effect on economic growth.

1

http://www.imf.org/extesach,rnal/pubs/ft/fandd/2006/06/elson.htm

2

The figures are based on the Conference Board 2015 database, http://www.conference-board.org/data/economydatabase/

3

See Bekaert, Harvey, and Lundblad (2005), Chanda (2005), Alfaro, Kalemli‐Ozcan, and Volosovych (2008), Klein (2005),

Klein and Olivei (2008).

Received: 16 April 2018

|

Revised: 10 December 2018

|

Accepted: 7 January 2019

DOI: 10.1111/twec.12783

World Econ. 2019;42:2039–2069. wileyonlinelibrary.com/journal/twec © 2019 John Wiley & Sons Ltd

|

2039

The paper has three important contributions. First, it is pertinent to inquire into factors that

determine the long‐run growth trajectory of the emerging market economies (EMEs). Conse-

quently, we adopt a growth accounting framework and measure how each factor, such as labour,

capital and TFP, has contributed towards GDP growth in the period 1990–2013. The findings

would help us to understand the nexus between the level of institutional quality, per capita GDP

and technological progress (accounted by TFP) in Asia and Latin America, respectively. Second,

by studying the interaction effect of capital flows and institutional quality we determine the thresh-

old level of institutional quality at which capital flows exert a positive effect on economic growth.

We also calculate the marginal effect of capital flows for low and high levels of institutional qual-

ity. This study, to the best of our knowledge, is the first to test the interaction effect of institutional

quality and capital flows on GDP and total factor productivity growths of EMEs. Capital flows are

diverse in nature and have different stabilising properties. Investigating their effect separa tely on

economic growth allows for a systematic comparison between foreign direct investment (FDI),

non‐FDI inflows and total flows. Finally, we look at the regional differences between Asia and

Latin America to calculate the marginal effects of capital flows on economic growth.

Our findings from the growth accounting exercise suggest that in the period 1990–2013, the

share of TFP is the key contributor of growth in Asia during high growth periods, whereas in Latin

America the share of labour and capital are more prominent than TFP. Moreover, in Asia, contribu-

tions from TFP are higher for economies with better institutional quality. Consequently, our empiri-

cal results show that after controlling for endogeneity the effect of total gross capital inflows on

GDP per capita growth is positive (0.69) when the threshold level of institutional quality reaches

29th percentile in our total sample. Similarly, the positive effect of FDI on GDP per capita growth

is higher (6.66 as compared to 4.69) above the institutional threshold level of the 33rd percentile.

Moreover, the conditional threshold level of institutional quality in Asia is smaller than that for

Latin America, suggesting that the overall institutional quality is better in Asia and that robust posi-

tive effects of capital flows on GDP per capita growth can be achieved with lower threshold levels

of institutional quality. Similarly, when we examine the effects of capital flows on TFP growth, we

find the effect of capital flows is contingent on the threshold level of institutional quality. However,

for Asia, the effect of capital flows at the very high levels of institutional quality (91st percentile) is

almost negligible. The finding suggests the possible presence of diminishing returns to institutional

quality that in economies with very high level of institutional quality, the positive effect of capital

flows on productivity growth is no longer conditional on institutional quality.

The rest of the paper is organised as follows. Section 2 provides a brief overview of the related

literature and develops the hypotheses. Section 3 provides a description of the data and measure-

ment issues. Section 4 presents the growth anatomy of Asia and Latin America in the period

1990–2013 by adopting a growth accounting framework. Section 5 discusses the empirical method-

ology. Section 6 presents the baseline results for the full sample and provides some robustness and

sensitivity tests. Section 7 presents the results for Asia and Latin America, separately. Section 8

concludes this paper.

2

|

PRIOR LITERATURE AND HYPOTHESIS DEVELOPMENT

There is a long‐standing literature that looks at the effects of capital flows, mainly FDI, and insti-

tutional quality on economic growth. The literature on endogenous growth theories has emphasised

the role of technological progress and capital deepening as the main drivers of long‐run economic

growth (Aghion & Howitt, 1997; Basu & Weil, 1998; Grossman & Helpman, 1993; Peretto,

2040

|

ARYA ET AL.