Open Door System and Endogenous Growth in Indian Economy: An Empirical Analysis on the Role of Human Capital and R&D in Explaining Industrial Productivity

Abstract

The present study investigates the impact of human capital, knowledge capital which is a function of human capital, and real exchange rate scenario in explaining long-run industrial total factor productivity (TFP) from 1980 to 2015 on the theoretical basis of the open endogenous growth model. The variables employed in the contemporary study include manufacturing value added (MNVA) as industrial output measure, gross fixed capital formation (GFCF) as a measure of capital and labour input which is measured using employment data. Gross enrolment ratio (GER) is taken as a measure for human capital formation, expenditure on research and development (R&D) as a proxy for knowledge capital, and real exchange rate indicates global economic shocks. The study involves estimating TFP for Industrial Sector during the post-liberalization period by employing Cobb-Douglas production function. The ARDL bounds test technique for cointegration revealed long-run relation among the varying factors studied. The Toda-Yamamoto causality test concluded bi-directional causality running between, R&D expenditure and Industrial TFP which sends a strong signal to the policymakers for a well-framed long-term integrated approach for human & knowledge capital formation which will act as a strong impetus for manufacturing firms to come up in terms of augmenting production and productivity and expanding foreign market horizon.

JEL Classification: D24, E2, J24

Keywords

Endogenous growth Model total factor productivity (TFP)human capital and technology

Introduction

The developed countries like the USA, Britain, Japan as well as the sovereign units, which later accomplished this credential like Hong Kong, Singapore and Taiwan, are now in a towering position because of large-scale industrialization. Industrial development is a source of technological progress and sustained growth in these countries. The industrial sector has thus been viewed as instrumental in a fundamental transformation in an economy, and its progress is a critical concern because of its capability to engross people and its lead role in economic expansion. Its ability to stimulate other segments positively and fetch structural and technological conversion is well distinguished (Kaldor, 1966; Lewis, 1954). Hence, the long supporting industrial progression is an imperative facet to be considered.

If we take the growth and development of the industrial sector or aggregate economy, the motivating forces behind long-run economic growth fundamentally differ from short-run growth in the sense that the short-run economic growth is primarily determined by the amassing of factors such as labour and capital. In the long run, it is the productivity growth that determines sustained pathway. Human capital is vital to sustained productivity growth as far as a testament from the historical and empirical analysis goes. The technological progress and human capital together determine the sustained pathway of industrial growth. The technological progress is principally dependent on the portion of human capital involved in research and development (R&D) activities.

In the vibrant economic spectrum such as that of the industrial sector, there is a solid interweaving between human assets and R&D. A country with the ideal quantity of human wealth and R&D development will boost the progress and development of an economy by fetching significant expansion in industrialization technology sophistication as well as labour productivity. The role of the elements such as human wealth and R&D in augmenting growth is well explained in endogenous growth paradigms.

Endogenous Growth Model

The technical terminology of human capital has become a quite accustomed notion after the erudite research carried out by the Nobel Laureate G .S. Becker (1964). According to his view, human capitals are skill, knowledge, honesty, experience, etc., exemplified in a person, which upsurge earning, production, health position, etc. In influential research, the notion was acknowledged by Bontis (1998), Davenport (1999) and Pennings et al., (1998). Endogenous growth paradigm elucidates that human investment is the key vehicle to accomplish the increasing returns and it is the principal cause for departure between developed and developing regions if we assess them using the parameters of growth (Lucas, 1988; Romer, 1986). The research conclusion of inspiring works by Romer (1990), Rebelo (1991) and Stokey (1991) are the testaments for the same.

The contribution of human wealth in liaison with technological advancement has been central in social science research for a few decades; for example, scholars like Lucas (1990) and Romer (1990) contributed significantly to the writings of human wealth-led and technology-led progression through their seminal works. The new growth theory promulgated by Lucas (1990) branded human capital as a leading explanatory variable in affecting economic growth, while the research input of Romer (1990) articulated that economic growth is profoundly conditional upon R&D and its spillovers. The title role of R&D and human wealth in inducing technological improvement and thereby productivity has been well recognized by them by instigating a path-breaking advancement in the theoretical paradigms of growth by largely drifting away from the neoclassical assumptions of perfect competitions.

The role of human capital and R&D in industrial development is less tested in the industrial sector in an open economy context. Such examinations in the developing countries are very rare, and this calls for a scientific study relying on the theoretical bulwark of the endogenous growth models. Hence, in this context, an empirical study vividly discussing the role of human capital and R&D in an open economic paradigm in the premises of the Indian economy is initiated in this article. The unique contribution of this research is that it examines the relevance of endogenous growth paradigm in the context of Indian manufacturing sector in productivity augmenting through the factors such as human and knowledge capitals. Thus, it intends to analyse the long-term interlink among human capital, R&D and total factor productivity (TFP) of the industrial sector where TFP of the industrial sector is treated as an endogenous variable. The causal relationships among the variables are also explored and various aspects of policymaking are examined, which can be initiated in this regard.

Previous Studies

Human Capital and Manufacturing Output

Human capital is one of the stirring powerhouses which enable a country/industry to produce outside the production boundary. Snell and Dean (1992) enlightened academic literature by arguing that human resources add worth to an industrial body by boosting productivity, which can be accomplished by upgraded skill and knowledge addition, whereas Pennings et al. (1998) articulated that augmented level of human capital in an industrial body enables to stream better and high-quality goods and services at home and foreign market. Temple and Voth (1998) articulate a prototype in which sufficient correlation exists between the growth of the manufacturing sector and apparatus investment and the association is aided by human capital. In scientific research, using erudite models to establish the link between human capital and returns to scale, Hitt et al. (2001) established that there is a progressive curvilinear nexus between the above-mentioned two variables.

In the premises of Indian manufacturing industries, Panagariya (2006) argues there is capital bias in the major industries in the registered arena like petrochemicals. He views these industries as highly skill-intensive and believes that India is not utilizing the comparative advantage in the labour. In their influential work, Kang and Snell (2007) depicted that human wealth has a value-creating capability as its influence is noticeable in whole productivity in industrial production exercise. Panagariya (2008) observed the need for labour market reforms to bring unregistered manufacturing activities into the registered sector. Further, the significant role played by the human capital in easing the grander enactment of the software manufacturing sector in Egypt was recognized by Seleim et al (2007). Anwar (2008) worked out an error correction technique during 1980–2005 and appealed that besides foreign investment, human capital is also revealed to be a major driver in Singapore’s manufacturing growth.

Based on the error correction tool from 1978 to 2008, Olayemi (2012) claimed that government disbursement on learning has a favourable tie with the index of industrial production, while a negative association is proven with that of healthiness factor, an indicator of human wealth and capital creation in gross levels. Relying on time series data from 1980–2010, Adejumo et al. (2013) dissected the context of the Nigerian economy and stated that human capital has a large impact on industry value-added.

Using the cointegration technique of autoregressive distributed lagged model (ARDL) bounds for a period of 35 years, Sulaiman et al. (2015) resolved that both technology and human wealth are key variables in explaining Nigeria’s growth. Voigtlander and Squicciarini (2015) inspected the backing factors of the mid-eighteenth-century industrial uprising based on a study of 85 cities in France where the subscription of the encyclopaedia is found. They observed that the presence of upper tail knowledge is instrumental in raising productivity and innovation in the industrial sector rather than human capital indicated by literacy rate. Khan (2018) conducted a character study using a great inventors sample set of 434 men and a woman in Britain and stated that rather than specialized training, the accumulation of expensive human wealth and the role of elites, it is the inducement for ingenuity, plasticity and aptitude to make additive adjustments that can transform the prevailing technology into innovated product/mechanism under the prevailing domestic conditions. Based on a panel study of 13 economies and using data from 304 manufacturing firms and a hierarchical linear regression model, Ma et al. (2019) established that job-related training sessions and employee partaking improve innovation activity at the firm level.

Research and Development and Manufacturing Growth

Raut (1995) took a sample of Indian private manufacturing firms of light petrochemical and heavy industries over the period 1976–1986. While running a Cobb–Douglas production function by applying simple OLS, he confronted the case of R&D spillover as a relevant factor explaining manufacturing growth except in petrochemicals. By extending Romer’s model of endogenous growth, including energy intermediate product relationships, Zon and Yetkiner (2003) concluded in their study that the need for energy efficiency growth under mounting actual fuel prices warrants a blend of R&D and energy price policy. In their study based on 190 Indian manufacturing firms in Bengaluru and Hyderabad, Kale and Rath (2018) found that exports and R&D expenditure have a statistically significant influence on innovation in the Indian manufacturing sector. In recent research, Sikdar and Mukhopadhyay (2018) computed elasticity of industry-level TFP with that of R&D component in intermediaries and concluded that the R&D component inherent in intermediaries plays an important role in Indian industries.

Exchange Rate and Manufacturing Output

Real exchange rate incorporates the impact of foreign demand and supply externalities inflicting repercussions on the importation of raw materials and export of finished goods by amending external price value. Branson and Love (1988) deliberated this in the background of the USA, while Ekholm et al. (2012) detected the exchange rate pass-through effect on manufacturing products by the means of three major mediums.

Theoretical Frameworks

According to Solow (1956), the growth of output per capita will convert to zero due to shrinking returns to factor and constant returns to scale as a whole. If it does not occur and it remains positive in the long term, it may be demarcated as technological progress, as mentioned by him, and it remains exogenous as he did not model it. The growth in the long duration is influenced not by factor amassing but by the increment in factor productivity. The endogenous theoretical paradigms by Romer (1990), Lucas (1988) and Aghion and Howitt (1990) modelled hitherto exogenous technological growth by dismantling it into human capital, expansion in knowledge, external technology spillover, etc. The extant work comprises a long pathway analysis of industrial productivity movement and the internal abode of productivity upsurge such as human wealth and expansion in knowledge.

Objectives

To quantify TFP for Indian industries for the time span from 1980 to 2015;

To examine the part of human wealth, expansion in knowledge and real exchange rate of TFP for Indian industries and contemplate on the endogenous growth scenario in India; and

To understand the direction of causality between TFP and influencing variables.

Data and Estimation Procedure

Total factor productivity has been assessed using the Cobb–Douglas production function:

Q = A K^{α} L^{β}

From the capital (K), labour (L), energy (E), materials (M), and services (S) [KLEMS] database, the capital share and labour share of value-added from the year 1980 to 1981 have been obtained where the shares of capital input (α), labour input (β) together constitute value one, that is, constant returns to scale. The exponent of labour is estimated to be 0.45 and that of capital as 0.55.

\begin{matrix} Q = A K^{α} L^{β} \\ A (i . e ., T F P o r S o l o w Re s i d u a l) = Q / (K^{α} L^{β}) \\ L n Q = L n 4 + α L n K + β L n L \\ L n A = (L n Q - (α L n K + β L n L) \end{matrix}

Where Log A shows TFP or Solow residual which measures technical progress.

Data and Sources

Gross manufacturing value added at a constant price is taken as Q, and K is measured using gross domestic fixed capital formation (GDFC) for the industry. The base year is 2011–2012 and all the variables are taken at constant prices. Gross labour employment in the industrial sector is taken as the labour input. The data for measuring TFP are sourced from the database of RBI. Since the data are available in varying base years, the statistical method of splicing is used. All the variables are expressed in crores.

Gross enrolment ratio (GER) in primary education is taken as the measure of human capital. Knowledge capital is assessed using national R&D expenditure. The real exchange rate is taken as the interaction variable of supply and demand for goods and services reflected in the international prices of Indian currency. All variables are expressed annually. R&D is expressed in crores with the base period of 2011–2012. The data, except that of R&D expenditure, are sourced from the World Bank website, while R&D data are taken from The National Science and Technology Management Information System (NSTMIS). This study is undertaken for the period from 1980 to 2015.

Before estimating using a regression model, it is essential to have an inspection of the behaviour of the data. Therefore, Augmented Dickey–Fuller (ADF) (1979) ad Phillips–Perron (1988) tests were employed. As all variables are I(0), we continued our analysis with ARDL bounds experiment devised by Pesaran et al. (2001). The tool can be useful when variables are a mixture of I(0) and I(1) and also if all are I(1) processes. The prerequisite is that the variables must not be I(2); secondly, it also separately enables us to measure the short-term and long-term connections. Besides that, this regression procedure tackles the matter of endogeneity. As there is a multicollinearity issue between R&D and GER, there are two models devised.

The models in the functional form are represented as:

Model 1

L N T F P_{t} = α + \sum_{i = 1}^{n_{1}} γ L N T F P_{t - i} + \sum_{i = 0}^{n_{2}} β_{1} L N L G E R_{t - i} + \sum_{i = 0}^{n_{4}} β_{3} L N E X_{t - i} + ε_{t}

(1)

Model 2

L N T F P_{t} = α + \sum_{i = 1}^{n_{1}} γ L N T F P_{t - i} + \sum_{i = 0}^{n_{2}} β_{2} L N R D_{t - i} + \sum_{i = 0}^{n_{4}} β_{3} L N T O P_{t - i} + ε_{t}

(2)

We can apprehend the short-term undercurrents by translating Equation (1) into an error correction design (ECM) as follows:

Model 1

Δ L N T F P_{t} = α + \sum_{i = 1}^{n_{1}} γ Δ L N T F P_{t - i} + \sum_{i = 0}^{n_{2}} β_{2} Δ L N G E R_{t - i} + δ E C_{t - 1} + ε_{t}

(3)

Model 2

Δ L N T F P_{t} = α + \sum_{i = 1}^{n_{1}} γ Δ L N T F P_{t - i} + \sum_{i = 0}^{n_{2}} β_{2} Δ L N R D_{t - i} + δ E C_{t - 1} + ε_{t}

(4)

where LNTFP, LNGER, LNRD and LNEX, respectively, indicate the log forms industrial TFP, GER primary, national R&D expenditure and real exchange rate. δ reflects the speed of adjustment, EC_t–1 characterizes instability and Δ entitles the first difference. The error correction constant signs the rapidity of readjustment to the long-term stable function after short-term setbacks lead to instability, and in this procedure, the long-term causation is depicted by the term factor δ of the error correction, which is negative and statistically significant.

Empirical Results

The prima facie analysis of variables on the basis of graphical plots (not shown here) shows that all variables are trended. This indicates that they have a long-run behaviour, and this long-run behaviour in the data has to be captured using appropriate tools. TFP has a small downward sloping trend which indicates a fall in TFP over time. All other variables show a general upward trend.

Descriptive Statistics and Correlation Matrix

The mean values of the variables closely share a similarity in magnitude with that of the median for all the variables indicating favourable signs for the occurrence of the normal distribution for all the variables (Table 1). Moreover, the position of median and mode is in the middle of maximum and minimum values indicating that the variables are free from major skewness issues, reflecting the fact that we can obtain information and relationships which can be generalized.

Table 1.

Summary Statistics

	LTFP	LGER	LRD	LEX
Mean	5.55	4.57	9.37	3.38
Median	5.54	4.54	9.34	3.71
Max	6.37	4.7	11.46	4.16
Min	4.69	4.43	7.47	2.55
SD	0.52	0.09	1.24	0.61
Skewing effect	−0.01	0.25	0.2	−0.72
Kurtosis	1.9	1.55	1.72	2.04
Observations	35	35	35	35

Source: Authors’ own estimation.

The correlation matrix (Table 2) indicates TFP is highly correlated with the independent variables, while R&D and GER give primary evidence for the existence of probable multicollinear relation among the variables. Further sophisticated analysis using the variance inflation factor (VIF) technique reveals the presence of multicollinearity among the variables, which would lead to an erroneous standard error, creating issues with the significance of coefficients.

Table 2.

Pairwise Correlation Analysis

	LTFP	LGER	LRD	LEX
LTFP	1.00
LGER	−0.85	1.00
LRD	−0.78	0.93	1.00
LEX	−0.77	0.81	0.85	1.00

Source: Authors’ own estimation.

Unit Root Test Result

In order to add empirical evidence to the evidently long-run behaviour in the variables, the unit root tests such as Phillips and Perron (1988) and ADF tests (1979) are run. The traditional tests indicate that every variable becomes stationary only after taking the first difference (Table 3). As these tests are unable to detect and process the issue of systemic breaks in the data series, it is probable that these techniques may give biased findings.

Table 3.

Unit Root Test Result

Variables	ADF		PP
Variables	Level	1st Diff	Level	1st Diff
LTFP	−1.70	−4.96*	1.75	−4.96*
LGER	−1.21	−4.25*	−1.14	−4.27*
LRD	−0.85	−4.08*	0.76	−4.08*
LEXE	−2.42	−4.04*	0.2.08	−4.12*

Source: Authors’ own estimation.

Notes: * Indicates statistical significance at 1% level.

In light of the unit root outcomes, we established the variables which are an amalgam of series of I(0)) and I(1) that advocate us to take up the cointegration technique of ARDL bounds propounded by Pesaran et al. (2001) for further analysis. Apart from divulging estimates of short-term and long-term estimates of coefficients, it also overcomes the issue of endogeneity by accommodating suitable lag structure of the variable. Narayan and Narayan (2005) advocates this model for small samples as well, that is, ranging from 30 to 80. The information criterion of Akaike is used to select the optimal leg length of the regression equation. The estimated F statistics, as shown in Table 5, is recognized to be greater than the 95% upper limit of the critical level, confirming the existence of long-run links among the variables.

Short-Run and Long-Run Elasticities

As there involves a multicollinearity issue with regard to human capital and R&D, we cannot estimate both of them in a single equation model. Hence, in this context, we have devised two models involving industrial TFP as the dependent variable for both the models. One model carries human capital as the major determining variable, while in the next model, R&D is the major determining variable. The other regressor variable is the real exchange rate for both the models.

Model 1: The Model with Human Capital

The bounds test analysis (Table 4) confirms that F-statistics lie above the higher-level boundary for both the estimations, and the necessary condition of passing bounds test which is a pre-requirement to establish cointegration among the variables has been satisfied.

Table 4.

Bounds Test Result of F-stats

Model 1	K	Model 2	K
7.09*	3	6.12*	3

Source: Authors’ own estimation.

Note: * Denotes rejecting of null hypothesis at 5% level.

Short-Run Elasticities

In the instance of model 1, the real exchange rate is found to have 1-year lagged positive impact on TFP at a 10% significance level, indicating positive exchange rate shocks in terms of favourable import and export prices, which is passed to the current value of TFP. GER has instantiations of negative impact on TFP, that is, any negative shock to human capital in the current year instantaneously reflects industrial productivity. A positive shock in these directions would have a 2-year lagged positive impact on industrial TFP which is significant. Thus, an increase in GER has pros and cons.

In paradigm 2, the variables such as R&D and real exchange rate have statistically significant outcomes on TFP, with R&D having a negative impact, while the real exchange rate has a positive impact. It is obvious that the short-run results are reflecting some disequilibrium outcomes and the relationships are not solid. Hence, it is expected that the relationships become well defined only in the long run. The short run disequilibrium is getting corrected into long run equilibrium at the rate of 69 and 38% respectively per annum and the values are significant and negative pinpointing cointegrating relation exists (see Table 5).

Table 5.

Short-run Result

	Variable	Coefficient	Std. Error	t-Statistic	Prob
Model 1	D(LEX)	−0.18^*	0.48	−0.39	0.7
	D(LEX(−1))	1.19^*	0.65	1.83	0.08
	D(LEX(−2))	−0.59	0.46	−1.3	0.21
	D(LGER)	−5.94^***	1.57	−3.77	0.0
	D(LGER(−1))	0.19	2.31	0.08	0.94
	D(LGER(−2))	4.18^**	1.91	2.19	0.04
	D(@TREND())	0.12^***	0.03	4.17	0.0
	CointEq(−1)	−0.69^***	0.17	−4.09	0.0
	R ²		0.83
	Adj R²		0.69
	F-stats (prob)		5.68(0.00)
	DW-stats		2.56
Model 2	D(LRD)	−1.1^*	0.6	−1.85	0.08
	D(LEX)	1.44^***	0.5	2.87	0.01
	D(@TREND())	−0.22^***	0.06	−3.66	0.0
	CointEq(−1)	−0.38^***	0.11	−3.57	0.0
	R ²		0.51
	Adj R²		0.4
	F-stat (Prob)		4.76(0.00)
	DW-stats		2.43

Source: Authors’ own estimation.

Notes: *, ** and *** Indicate statistical significance at 10%, 5% and 1% level, respectively.

The Long-Run Relationship Under This Method

Coming into the long-run elasticity (Table 6) measures, the variables in paradigm 1 such as GER and real exchange rate have a negative influence on industrial TFP in the long run. The rate of change in the GER is not sufficient to make a positive impact on TFP of the manufacturing sector. This may emanate from the phenomena of disguised unemployment which is in the arena of the agriculture sector. Also, the industrial sector is in a nascent stage and not able to lift labour productivity as most of the labourers are in unregistered manufacturing activities. Also, the quality of schooling is poor as expressed by Panagariya (2006).

In model 1, the real exchange rate bears a negative relationship with that of manufacturing TFP, which reflects the fact that India is largely remaining as an importer and as an exporter, and its position is precarious. Hence, it is largely subjected to real exchange rate depreciation. But at the same time, the industrial sector is not able to benefit from importing technology as the technological spread is meagre in the industrial sector. We do not have a global market for branded manufacturing products, and our production standards usually conform to domestic demand requirements and remain unexposed to foreign exposure of technological standards, which is negatively reflected in its relationship with TFP. The trend coefficient is significant, indicating a particular relationship that is maintained throughout.

The long-run coefficients, as indicated in Table 6 for model 2, convey that R&D expenditure bears a positively significant coefficient with TFP. R&D indicates domestic production of knowledge and expansion of ideas, which is an outcome of human capital invested in the R&D sector. The positive influence of the real exchange rate in stimulating industrial TFP is captured in the model. There can be both positive and negative influences. The asymmetric ARDL method captures both the positive and negative impacts of a particular variable with the variable under focus. But we have used the normal ARDL model.

Table 6.

Long-run Elasticities

	Dependent Variable LTFP
	Lag Selection Based on Akaike information criteria
	ARDL lag (1,1,1)
Model 1	Variable	Coefficient	Std. Error	t-Statistic	Prob.
	LEX	−1.4^*	0.3	−5.5	0.0
	LGER	−14.9^*	2.2	−6.8	0.0
	C	75.4^*	10	7.5	0.0
	@TREND	0.2^*	0	5.6	0.0
	Dependent Variable LTFP
	Lag Selection Based on Akaike information criteria
	ARDL lag (1,1,1)
Model 2	Variable	Coefficient	Std. Error	t-Statistic	Prob.
	LRD	3.9^*	1.4	2.8	0.0
	LEX	1.7^*	0.7	2.3	0.0
	C	−25.8^*	11.5	−2.2	0.0
	@TREND	−0.6^*	0.2	−3	0.0

Source: Authors’ own estimation.

Notes: * Denotes statistically significant at 1% level.

Diagnostic Results

The diagnostic tests (Table 7) confirmed that the models are correctly specified; there is no serial correlation among the variables; error terms are normally distributed; there is no heteroscedasticity in the model; and the functional form of the model is normal. In a nutshell, the diagnostic test results pinpoint that our models are robust.

Table 7.

Diagnostic Check Result

Country	Ramsey RESET Test	LM Test	Jarque–Bera Test	ARCH Test
Model 1	0.73 (0.46)	1.84 (0.18)	1.25 (0.53)	0.34 (0.56)
Model 2	0.32 (0.74)	1.75 (0.14)	0.75 (0.68)	1.84 (0.18)

Source: Authors’ own estimation.

Note: Value in parenthesis is the p-value.

Stability of the Model

The paradigm stability in the ARDL estimation frame is scrutinized by applying the cumulative sum of recursive (CUSUM) and the CUSUM of squares (CUSUMSQ). This may be due to misspecification which results in biased coefficients ultimately affecting the explanatory power. Pesaran and Pesaran (1997) suggested CUSUM AND CUSUMSQ tests which assure the existence of stability of the model by checking the constancy of parameters. The lines of CUSUM and CUSUMSQ (see Figure 1) suggest that the tests’ statistics conform to the space between critical bounds of 5% significance level testifying the stability of the parameters in both estimates.

Figure 1.

Source: The authors.

Toda Yamamoto/Modified WALD Test of Causality

The Toda and Yamamoto (1995) approach to Granger causality test has been employed to understand the direction of causality. It involves a modified WALD test (MWALD) specified by Toda and Yamamoto (1995). The traditional granger causality test requires variables to be stationary in order to reflect the true results of causality. The Toda Yamamoto Granger causality test rules out this disadvantage as it is upgraded with an inbuilt ability to consider both I(0) and I(1)variable in equal footing. The results of the test indicate that there is a unidirectional causality escalating from TFP to GER and exchange rate. A feedback causal relationship has been established between TFP and R&D expenditure. Both variables strengthen each other which implies the quantum of research the expenditure influences the volume of research activities thereby affecting productivity. An increase in industrial TFP has a reverse influence on R&D implying an increase in industrial productivity that helps in allocating funds for R&D activity. TFP has a positive influence on the real exchange rate which is so obvious as an increase in industrial coverage has some favourable effects on the international value of the currency (see Table 8).

Table 8.

Causality Tests Result

Null Hypothesis: No Causality	Direction	Chi-squared	Prob.
	LGER to LTFP	10.17	0.11
	LTFP to LGER	11.57*	0.07
	LEX to LTFP	2.21	0.13
Model 1	LTFP to LEX	18.7***	0.00
	LEX to LGER	1.68	0.94
	LGER to LEX	14.99**	0.02
	LRD to LTFP	3.71**	0.05
	LTFP to LRD	3.92**	0.04
Model 2	LEX to LTFP	2.21	0.13
	LTFP to LEX	2.93*	0.08
	LEX to LRD	0.97	0.32
	LRD to LEX	13.2***	0.00

Source: Authors’ own estimation.

Notes: *, ** and *** Indicate statistical significance at 10%, 5% and 1% level, respectively.

Conclusion

The study involved estimating TFP of MNVA in the Indian economy from 1980 to 2015 followed by finding out the relationship between the TFP in the manufacturing sector with that of human and knowledge capitals and exchange rate. The unit root tests such as ADF (1979) and Phillips–Perron (1988) revealed that the variables are stationary only in the first difference form. We have performed two estimations as the variables of human and knowledge capital involve a multicollinearity problem. Industrial TFP estimated by the author is taken as the dependent variable in both the models, and in one model human capital measured by GER is primarily taken as the major independent variable, whereas in the other model, knowledge capital, measured by R&D, is the independent variable. The ARDL bounds test approach revealed cointegration among the variables in both the models. The negative long-run relationship between human capital and real exchange rate with that of manufacturing is established implying that the quality of human capital is very low and there is a need to bring in qualitative changes in the educational system of the country, to diversify the educational arena and to augment the skill-intensive education. In addition, job training mechanisms should be strengthened.

Also, in their paper, Joumard et al. (2015) observed the manufacturing sector in India is very feeble in its contribution in terms of income, export and employment, and manufacturing productivity is further low in terms of international comparisons. They also pinpointed that the kind of jobs percolated from it is feeble and mostly created in the informal sector. The major constraints as observed in their paper indicate stringent labour norms, infrastructural backwardness, low-quality jobs and working space. This should be read along with Panagariya (2006) who advocated labour market reforms as well as liberalization measures in labour-intensive sectors, which involves bringing it into the registered manufacturing arena through labour market reforms.

R&D-led growth hypothesis seemed to be relevant in the manufacturing sector which indicates R&D-led endogenous growth model is in vogue in explaining manufacturing productivity growth. This implies pursuing R&D activities in a much-augmented fashion gives enough room for private R&D activities. In a nutshell, the study involves advocating a well-framed long-term integrated policy for human and knowledge capital formation, which will act as a strong impetus for manufacturing firms to come up in terms of production and productivity and expansion of foreign market horizon.

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

References

Adejumo

A. V.

, Olomola

P. A.

, & Adejumo

O. O.

(2013). The role of human capital in industrial development: The Nigerian case (1980–2010). Modern Economy, 4(10), 639–651.

Aghion

, & Howitt

(1990). A model of growth through creative destruction (No. w3223). National Bureau of Economic Research.

Anwar

(2008). Foreign investment, human capital, and manufacturing sector growth in Singapore. Journal of Policy Modeling, 30(3), 447–453.

Becker

G. S.

(1964). Human capital: A theoretical and empirical analysis, with special reference to education. The University of Chicago Press.

Bontis

(1998). Intellectual capital: An exploratory study that develops measures and models. Management Decision, 36(2), 63–76.

Branson

W. H.

, & Love

(1988). US manufacturing and the real exchange rate. In Marston

R. C.

(Ed.), Misalignment of exchange rates: Effects on trade and industry (pp. 241–276). University of Chicago Press.

Davenport

(1999). The higher arithmetic: An introduction to the theory of numbers. Cambridge University Press.

Dickey

D. A.

, & Fuller

W. A.

(1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366a), 427–431.

Ekholm

, Moxnes

, & Ulltveit-Moe

K. H.

(2012). Manufacturing restructuring and the role of real exchange rate shocks. Journal of International Economics, 86(1), 101–117.

10.

Hitt

M. A.

, Bierman

, Shimizu

, & Kochhar

(2001). Direct and moderating effects of human capital on strategy and performance in professional service firms: A resource-based perspective. Academy of Management Journal, 44(1), 13–28.

11.

Joumard

, Sila

, & Morgavi

(2015). Challenges and opportunities of India’s manufacturing sector (Working Paper). OECD.

12.

Kaldor

(1966). Marginal productivity and the macro-economic theories of distribution: Comment on Samuelson and Modigliani. The Review of Economic Studies, 33(4), 309–319.

13.

Kale

, & Rath

B. N.

(2018). Does innovation matter for total factor productivity growth in India? Evidence from ARDL bound testing approach. International Journal of Emerging Markets, 13(5), 1311–1329.

14.

Kang

S. C.

, & Snell

S. A.

(2007). Intellectual capital architectures and ambidextrous learning: A framework for human resource management. Journal of Management Studies, 46(1), 65–92.

15.

Khan

B. Z.

(2018). Human capital, knowledge and economic development: Evidence from the British Industrial Revolution, 1750–1930. Cliometrica, 12(2), 313–341.

16.

Lewis

W. A.

(1954). Economic development with unlimited supplies of labour. The Manchester School, 22(2), 139–191.

17.

Lucas Jr

R. E.

(1988). On the mechanics of economic development. Journal of Monetary Economics, 22(1), 3–42.

18.

Lucas

R. E.

(1990). Why doesn’t capital flow from rich to poor countries? American Economic Review, 80(2), 92–96.

19.

, Zhai

, Zhong

, & Zhang

Z. X.

(2019). Deploying human capital for innovation: A study of multi-country manufacturing firms. International Journal of Production Economics, 208, 241–253.

20.

Narayan

, & Narayan

P. K.

(2005). An empirical analysis of Fiji’s import demand function. Journal of Economic Studies, 32(2), 158–168.

21.

Olayemi

S. O. O.

(2012). Human capital investment and industrial productivity in Nigeria. International Journal of Humanities and Social Science, 2(16), 298–307.

22.

Panagariya

(2006). Transforming India. In Conference on India: An Emerging Giant, Columbia University.

23.

Panagariya

(2008). India: The emerging giant. Oxford University Press.

24.

Pennings

J. M.

, Lee

, & Witteloostuijn

A. V.

(1998). Human capital, social capital, and firm dissolution. Academy of Management Journal, 41(4), 425–440.

25.

Pesaran

M. H.

, & Pesaran

(1997). Working with Microfit 4.0: Interactive econometric analysis. Oxford University Press.

26.

Pesaran

M. H.

, Shin

, & Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics, 16(3), 289–326.

27.

Phillips

P. C.

, & Perron

(1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335–346.

28.

Raut

L. K.

(1995). R & D spillover and productivity growth: Evidence from Indian private firms. Journal of Development Economics, 48(1), 1–23.

29.

Rebelo

(1991). Long-run policy analysis and long-run growth. Journal of Political Economy, 99(3), 500–521.

30.

Romer

P. M.

(1986). Increasing returns and long-run growth. Journal of Political Economy, 94(5), 1002–1037.

31.

Romer

P. M.

(1990). Endogenous technological change. Journal of Political Economy, 98(5, Part 2), 71–102.

32.

Seleim

, Ashour

, & Bontis

(2007). Human capital and organizational performance: A study of Egyptian software companies. Management Decision, 45(4), 789–801.

33.

Sikdar

, & Mukhopadhyay

(2018). Assessment of R&D and its impact on Indian manufacturing industries. International Journal of Computational Economics and Econometrics, 8(2), 207–228.

34.

Snell

S. A.

, & Dean Jr

J. W.

(1992). Integrated manufacturing and human resource management: A human capital perspective. Academy of Management Journal, 35(3), 467–504.

35.

Solow

R. M.

(1956). A contribution to the theory of economic growth. The Quarterly Journal of Economics, 70(1), 65–94.

36.

Squicciarini

M. P.

, & Voigtländer

(2015). Human capital and industrialization: Evidence from the age of enlightenment. The Quarterly Journal of Economics, 130(4), 1825–1883.

37.

Stokey

N. L.

(1991). Human capital, product quality, and growth. The Quarterly Journal of Economics, 106(2), 587–616.

38.

Sulaiman

, Bala

, Tijani

B. A.

, Waziri

S. I.

, & Maji

I. K.

(2015). Human capital, technology, and economic growth: Evidence from Nigeria. Sage Open, 5(4), 2158244015615166.

39.

Temple

, & Voth

H. J.

(1998). Human capital, equipment investment, and industrialization. European Economic Review, 42(7), 1343–1362.

40.

Toda

H. Y.

, & Yamamoto

(1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 66(1–2), 225–250.

41.

Van Zon

, & Yetkiner

I. H.

(2003). An endogenous growth model with embodied energy-saving technical change. Resource and Energy Economics, 25(1), 81–103.