INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 14

Modelling the Role of Absorptive Capacity in Foreign Direct

Investment - Economic Growth Nexus: A Focus on South Africa’s

Manufacturing Sector

Marius Ikpe

1, 2

Department of Economics, Enugu State University of Technology Enugu, Nigeria

DOI: https://doi.org/10.51583/IJLTEMAS.2025.1410000003

Received: 06 October 2025; Accepted: 12 October 2025; Published: 27 October 2025

Abstract: Overtime, scholars have come up with diverse opinion regarding the impact of foreign direct investment (FDI) on

economic growth in host communities. Similarly, empirical findings arising from different contexts and time have equally been

contradictory. Current effort to resolve these inconsistency narrows down to sectoral concentration of multinational firms and the

capacity of host communities to absorb foreign technology. In addition to these factors, this study deemed it necessary to

highlight the importance of government regulatory role. Analysis that defined this investigation was founded on the Solow’s

labour-augmented production function, while Autoregressive Distributed Lag (ARDL) bound test approach was employed in the

estimation that followed using data for the period 1991 - 2022. The analysis produced interesting findings: First, results show that

I. Introduction

Views are diverse and have remained unchanged regarding the growth-impact of foreign direct investment (FDI) on host

communities. In theory, there is a consensus of opinion aligning with the fact that FDI is an engine of growth; rather than

negative externalities, FDI provide positive external effects. However, overtime, empirical evidence regarding this relationship in

different contexts and time have remained contradictory. For instance, in some contexts and times, a good number of studies

provided evidence that supports the idea that FDI is a strong catalyst and engine of growth (Caves 1974; Kokko, 1994; Nair-

Reichert and Weinhold, 2001; Yao and Wei, 2007; Pegkas, 2015; Emako et al., 2022; Wondimu, 2023), some other studies came

up with findings that suggest a negative external effects instead (Haddad and Harrison, 1993; Aitken and Harrison, 1999;

Bornschier, Chase-Dunn, and Robinson, 1978; Fry, 1993; Ikpe and Nteegah 2014). There are also studies that found no evidence

that could link FDI to growth (Carbonell and Werner, 2018; Ikpe, 2019; Ozili, 2025). These evidence of contradiction in findings

notwithstanding, analysts still strongly hold the view that FDI plays key role in growth stimulation and development as evidence

are bound in many countries. Therefore, it is argued that, resolution lies in understanding the economic environment that could

serve as bases for harnessing the benefits of FDI as buttressed by Zhang (2001) which aligned perceived doubt on the positive

growth-impact of FDI to unfavourable country-specific conditions. One amongst notable country-specific conditions is explained

in terms of lack of absorptive capacity. This has been pointed out as reason why empirical evidence of positive effect of FDI on

growth is largest among developed countries (Carbonell and Werner, 2018). One other factor pointed out is the issue of sectoral

concentration of FDI within the industrial sectors of host communities. It is widely argued that spillover benefits of FDI differ

across sectors and countries, given that potential to absorb foreign technologies differ from sectors and economies (Alfaro, 2003;

Hirschman, 1958). Alfaro (2003) further emphasized that FDI in different sectors of the industrial sector (primary, manufacturing

and services) exert different effects on economic growth (see Emako et al, 2022). As a result, it is worthy of note that effect of

INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 15

sector (particularly the manufacturing sector) has remained unattractive to foreign multinational firms. The ranking of South

Africa in World Bank’s ease of doing business index has experienced a downward trend over the years; from 41 out of 185

countries in 2013, South Africa was subsequently ranked 72 in 2015, 74 in 2016 and 82 in 2017 out of 189 countries (World

Bank, 2018). Notable specific factors responsible for South Africa’s poor ranking includes lack of access to long term funds,

infrastructural deficit, insecurity, policy summersault, lack of business confidence, and lack of access to credits, high cost of

starting business, poor contract enforcement for businesses respectively. These factors do not guarantee security of investments,

and increases cost of investment. These has led to negative net-inflow of manufacturing multinational firms, thus leading to

declines in the contribution of manufacturing sector to GDP over the years. For instance, average growth rate of manufacturing

value added for South Africa declined from 5.93% over 1970 – 1975 to 0.33% over the period 1990 – 1998 (Bell and Madula

2001). Evidence points to the fact that this decline persists even in recent period; the contribution of manufacturing sector to GDP

is 15.2% in 2013 and 13% in 2017Q3 (Brand South Africa, 2014, 2018).

In an effort to resolve the observed inconsistency in previous findings in the context of South Africa, this study modelled for

absorptive capacity in the model of impact of FDI on the growth of the manufacturing sector. The study differs from previous

studies in the following ways: First, it modelled for human capital, openness to trade, and financial market development as

components of absorptive capacity (see Carbonell and Werner 2018). Carbonell and Werner argue that absorptive capacity is

required for FDI to enhance productivity and output growth. Previous studies did not consider this. Second, it is rightly argued

that, domestic credit for GDP transactions be controlled in the FDI-growth relationship, otherwise result from such estimation

could suffer from omitted variable bias (Carbonell and Werner, 2018). Previous South African studies ignored this assertion,

for a causal link between government regulatory role and economic performance. (see World Bank 2004). The study utilized

World Bank index of regulatory quality as proxy for capacity of government to formulate and implement sound policies and

regulations that could permit as well as promote private sector development.

The paper is divided into five sections; following this introduction is section two that examines the empirical evidence from

which literature gap was identified, section three shows the methodological procedures adopted in the analysis of data, four

presents result of the data analysis, while five concludes the study, highlighting policy implications of findings from the study.

II. Empirical literature

Divergent and inconsistent international evidence continues to support differing opinions about how foreign direct investment

(FDI) affects the growth and development of national economies. While some studies (Hansen and Rard, 2006; Lumbila, 2015;

Zakia and Ziad, 2007; Adigwe et al, 2015; Yao and Wei, 2007; Pegkas, 2015; Ali and Hussain, 2017) show a strong positive

growth impact in industrialized economies than in developing ones. Studies looked at the connection between FDI and several

characteristics that characterize an economy's absorptive capacity in order to provide explanation. Blomstrom et al. (1994) did not

identify any connection between education and FDI inflows in developing nations, while Li and Liu (2005) found evidence of a

strong positive correlation between human capital and the growth impact of FDI in both developed and developing nations. Some

nations have seen some good effects, while others have seen none at all (Castellani and Zanfei, 2003). In 123 countries, studies

such as Young and Lan (1997) were unable to find any evidence of FDI's beneficial effects on total factor productivity.

The aforementioned makes it abundantly evident that academics cannot agree on the expected impact of foreign direct investment

(FDI), either overall or at the sectoral level. Mixed results were also obtained when the individual effects of each of the many

elements that account for absorptive capacity were investigated. Because each country has unique characteristics, it is common to

have different conclusions for different countries. Even when the same model and data source are used, these factors greatly

INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 16

MANV

FDI

GfCAK

SSENROL

Douglas production function; this in this study is specified as Y = K

labour, α indicates constant returns to physical capital, and 1-α stands for constant returns to human capital. The output function is

thus derived from equation 1. In the Solow’s model, this is defined as a function of the initial output and the determinant of

ultimate steady state. In the empirical function, the study follows Carbonell and Werner (2018)’s reduced/parsimonious GETS

model for the Spanish economy with modification as adopted by Ikpe (2025),. Thus,

MANV = f(GfKAP, SSENROL, FDIGDP, CREDIT, TRDGDP, GVREG) (2)

where,

MANV is manufacturing value added, GfKAP represents fixed capital formation (proxy for stock of physical capital), SSENROL

approach to cointegration analysis can be applied to series that are integrated at I(0) or I(1) or [I(0) and I(1)]. However, it must be

ensured that none of the variables in the series are I(2) as that could make the computed F-Statistics required to determine

cointegration invalid. In order to identify the order of integration of series correctly, unit root analysis is implemented in this

study. With Perron (2006) suggesting that conventional unit root tests such as Augmented Dicky–Fuller (ADF; Dickey and Fuller

(1979, 1981), Philip Perron (PP; Phillips and Perron (1988), Kwiatkowski–Phillips–Schmidt–Shin (KPSS; Kwiatkowski et al.

(1992) and Ng and Perron (2001) provide biased results because of their low explanatory strength to identify unknown structural

breaks in the series, this study applies Zivot and Andrews (1992) unit root test, which takes into account unknown single

structural break in the series. It identifies the unknown structural breaks date by treating the break date as an endogenous variable.

The results in Table 2a and 2b show a structural break in all the data series. The stationary properties show mixed integration in

the data series of the two countries [I(0) and I(1)]. Interestingly, none of the variables is found to be integrated at second

INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 17

INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 18

six explanatory variables account for changes in the value added of the manufacturing sector. These include Foreign Direct

sector increases by 0.82% for every 1% increase in FDI. On the other hand, real sector credit (CREDIT) as a measure of

absorptive capacity significantly increases MANV by 0.10%. It indicates tendency of MANV to increase by 10% in response to

100% increase in real sector credit. Likewise, MANV significantly increases by 8.53% in response to 1% improvement in the

quality of government regulatory quality; this indicates tendency of MANV to increase by 853% should the economy level of

government regulatory quality increase by 100%. The economy’s levels of stock of physical capital (GfKAP), quality of human

capital (SSENROL), as well as openness of the economy (TRDGDP) are really not of importance in determination of levels of

MANV in South Africa. Among variables that define absorptive capacity, only real sector credit (CREDIT) significantly

-1.70%. The model has an ECM(-1) of -0.46, and it is negative, indicating adjustment to equilibrium at the rate of 46% in the case

of disequilibrium each period. See tables 5a and 5b for the long run, and short run estimates, as well as diagnostic tests. Evidence

INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,

MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)

ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025

www.ijltemas.in Page 19

insignificant effect on manufacturing growth performance at lower capacity, and positive or significant effect at higher capacity

of development. For instance, SSENROL as a measure of human capital development is at a very low level in South Africa,

hence findings from the study. The aggregate has a weak insignificant positive relation with MANV (0.0594) in the model.

Similarly, openness of the economy (TRDGDP) has a negative and insignificant relationship with MANV; This outcome is not

unconnected with the economy’s relative low level of development hence the negative effect.

Secondly, regulatory role of the state is a key factor that determines both the magnitude and nature of the individual private