The Indian growth miracle and endogenous growth
Introduction
Although R&D has been assumed to be the key factor in endogenous growth models, so far there has been little empirical analysis that explores the role of R&D in explaining growth for a “miracle” economy. Consequently, whether there are scale-effects in the rate of innovation and hence whether the high growth rates in the “miracle” economies will continue remains largely unexplored. In fact, the implications of R&D-based theories for developing economies remain unclear. The main objectives of this paper are to use data for more than half a century to: 1) test which second-generation endogenous growth theory is most applicable in explaining growth in India; and 2) examine the importance of R&D, among other variables, in explaining growth rates in India. The growth theories tested in this paper are the following two second-generation endogenous growth models: 1) the semi-endogenous growth theories of Jones (1995a), Kortum (1997) and Segerstrom (1998); and 2) the Schumpeterian growth theories of Aghion and Howitt (1998), Peretto (1998),Young (1998) and Howitt (1999).
This study is, to the best of our knowledge, the first attempt to test the importance of R&D for growth and the validity of second-generation endogenous growth models for a developing country. An important question addressed in this paper is whether R&D activities play a crucial role in explaining growth for a developing country like India or whether R&D-driven growth is limited to highly developed countries. The analysis is limited to India because it is one of the very few developing countries for which R&D data over sufficiently long time periods are available to enable tests of the importance of R&D for growth in general, and specifically to discriminate between different second-generation growth models. The R&D data for India cover the period from 1950 to 2005, a time span that even exceeds that of the R&D data that are available for almost all OECD countries.1 Most R&D data for developing countries are available only from the 1980s, which is far too short a period to discriminate between growth models using aggregate data.
Furthermore, India is an ideal candidate for testing R&D driven growth given that it has experienced a significant increase in the growth rate of total factor productivity (henceforth, TFP) since the 1980s. This raises the question of whether R&D or economic reforms have been the key factors behind India's take-off in the late 1980s. The transition from low to high growth coincides with significant economic reforms in several key sectors of the Indian economy since the late 1980s. Hence, the literature often argues that the high growth rates have been driven by economic liberalization or an attitudinal change favoring the reforms (Panagariya, 2002, De Long, 2003, Rodrik and Subramanian, 2005).
Another possibility is that growth in India has been predominantly driven by transitional dynamics and factor accumulation over the past two decades. Young (1995), for example, finds that capital deepening, increasing educational attainment, and increasing labor force participation rates explain most of the high growth rates experienced among the Four Tigers in the post WWII period. However, factor accumulation is unlikely to have been entirely responsible for the high growth rates recently experienced in India for two reasons. First, the growth in India's savings rate has not coincided with the increasing productivity growth rates, as experienced by the Four Tigers. In fact, the growth in India's savings rate has fluctuated around a constant level over the period 1950–2005. If savings-induced capital deepening was the principal factor behind productivity growth in India, we would expect growth to have been distributed evenly over the period 1950–2005 and not concentrated in the latter part of the period. Second, the factor accumulation hypothesis predicts that the growth in the capital-output ratio precedes growth in labor productivity. However, Granger causality tests suggest that the capital deepening has been a result of productivity growth and not the other way around.2 Finally, Madsen and Ang (2009) show that more than half of per capita growth in India during the period 1950–2006 is explained by TFP and that most growth in the post-reform period has been driven by TFP growth.
To account for the influence on TFP growth of the liberalization attempts by the Indian Government, we include several control variables that capture the impact of the economic reforms, i.e., an index of financial liberalization, tariff rates, the fraction of firms that are foreign owned and patent rights protection. These variables will also shed some light on the influence of the economic reforms on TFP growth in India, an issue that remains largely unexamined. Both aggregate data (1950–2005) and firm-level data (590 firms over the period 1993–2005) are used in this paper to provide insight into the ability of the second-generation growth models in explaining the productivity growth experience for India.
The next section contains a brief anatomy of the most important reforms undertaken in India since independence with a special focus on R&D policies. The innovation-driven endogenous growth theories and their growth implications are presented and briefly discussed in Section 3. Data and graphical evidence are presented in Section 4. Time series tests using aggregate data are performed in Section 5. Section 6 provides estimates of TFP growth. In addition to discriminating between the second-generation endogenous growth models, the empirical analysis also examines the roles of international technology spillovers and distance to the frontier in driving productivity growth. Some robustness checks of the estimates are provided in Section 7. Analysis using firm-level data are undertaken in Section 8 to complement the results based on aggregate data. The last section concludes the paper.
Section snippets
From Hindu to miracle growth rates
Between 1950 and 1990, India's per capita income grew at an average annual rate of only about 2%. During this period, the Indian government implemented restrictive trade, financial and industrial policies. Shortly after independence the Indian state took control of major heavy industries and private firms were only allowed to enter a few consumer and intermediate industries while being subject to widespread industrial licensing requirements and price regulation (Aghion et al., 2008).
R&D-based endogenous growth models
Consider the following ideas production function, which can be used to discriminate between endogenous growth theories (see, e.g., Ha and Howitt, 2007, Madsen, 2008):where gA is TFP growth, A is the level of TFP, X represents research input (semi-endogenous growth theory) or the productivity-adjusted research input (Schumpeterian growth theory), Q is product variety, L is employment, X/Q is research intensity, λ is the R&D productivity parameter, σ
Data
The models are estimated using aggregate data and firm-level data. Annual data for the period 1950–2005 are used in the aggregate analysis and the firm-level analysis considers the post-liberalization period of 1993–2005. TFPt is computed as Yt/[Kαt·(HL)1 − αt] where Yt is real GDP, Kt is real capital stock, Lt is labor force and Ht is an index of human capital. (HL)t measures the quality adjusted workforce and α measures the capital elasticity. The assumption of constant returns to scale in
Unit root tests
Unit root tests are performed based on the procedure of Ng and Perron (2001) that takes into account the possible presence of a structural break using data over the period 1950–2005.7
TFP growth estimates
The TFP growth equation given by Eq. (5) is estimated in 5 and 10-year differences to filter out the influence of business cycles. The research intensity measures are taken to be the average over all years covered by the differences. Distance to the frontier is evaluated at the first year of the differences. The standard errors are derived based on the Newey–West procedure in order to provide heteroscedasticity and autocorrelation consistent estimates.
The estimation results are reported in
Robustness checks
The surge in India's growth in the post-reform period raises the question of the extent to which the reforms have contributed to the Indian growth miracle and whether the results in the previous section are robust to the inclusion of control variables, particularly the variables that capture the most important economic reforms. Eq. (5) is extended with several control variables in this section.
The following control variables are considered. First is the ratio of foreign direct investment (FDI)
Firm-level analysis
The results in the previous section may have suffered from a small sample problem. More importantly, the Schumpeterian results are derived under the assumption of free entry of firms. This assumption may not have been satisfied for India during the post-independence period given that the economy was subject to a set of industrial licensing requirements that restricted entry and expansion of both domestic and foreign firms, and this is dubbed “license raj” (Panagariya, 2002, Rodrik and
Conclusions and implications of the findings
The objectives of this paper are two-fold: first, to test which of the two second-generation endogenous growth models is consistent with the data for India; and second, to examine the extent to which the roles R&D activity, international R&D spillovers, distance to the technology frontier and variables representing the economic reforms since independence, explain growth in a “miracle” economy like India. The study is motivated by the significant increase in GDP growth observed in India during
Acknowledgments
Helpful comments and suggestions from workshop participants at Monash University, conference participants at the Harvard Project on Asian and International Relations at Harvard University, the 13th Australasian Macroeconomics Workshop at the University of Sydney and the 2008 Asia Pacific Week at the Australian National University and, particularly, two referees are gratefully acknowledged. James B. Ang and Jakob B. Madsen acknowledge support from ARC Discovery Grants from the Australian
References (63)
- et al.
Financial liberalization, financial sector development and growth: evidence from Malaysia
Journal of Development Economics
(2007) - et al.
Buyer investment, export variety, and intrafirm trade
European Economic Review
(2008) - et al.
Endogenous growth in a cross-section of countries
Journal of International Economics
(2000) - et al.
Testing endogenous growth in South Korea and Taiwan
Journal of Development Economics
(1999) - et al.
Product variety and technical change
Journal of Development Economics
(2009) - et al.
Determinants of patent rights: a cross-national study
Research Policy
(1997) - et al.
Testing for unit roots in heterogeneous panels
Journal of Econometrics
(2003) Technology spillover through trade and TFP convergence: 135 years of evidence for the OECD countries
Journal of International Economics
(2007)Industrial development, technological change, and long-run growth
Journal of Development Economics
(1999)Does learning to add up add up? The returns to schooling in aggregate data
India's growth prospects revisited
Economic and Political Weekly
A model of growth through creative destruction
Econometrica
Endogenous Growth Theory
The unequal effects of liberalization: evidence from dismantling the license Raj in India
American Economic Review
International Data on Educational Attainment: Updates and Implications
Oxford Economic Papers
The local power of some unit root tests for panel data
International R&D spillovers
European Economic Review
India since independence: an analytic growth narrative
New product varieties and the measurement of international prices
American Economic Review
Export variety and country productivity: estimating the monopolistic competition model with endogenous productivity
Journal of International Economics
Industrial R&D promotional programme
R&D and absorptive capacity: theory and empirical evidence
Scandinavian Journal of Economics
Mapping the two faces of R&D: productivity growth in a panel of OECD industries
Review of Economics and Statistics
Innovation and Growth in the Global Economy
Accounting for trends in productivity and R&D: a Schumpeterian critique of semi-endogenous growth theory
Journal of Money Credit and Banking
Why do some countries produce so much more output per worker than others?
Quarterly Journal of Economics
The performance of panel unit root and stationarity tests: results from a large scale simulation study
Econometric Reviews
Cited by (78)
Sustainable development in China? A nonparametric decomposition of economic growth
2023, China Economic ReviewThe determinants of AI innovation across European firms
2023, Research PolicyPoland's innovation performance: drivers, challenges, and policy implications
2023, Procedia Computer ScienceExternal technology dependence and manufacturing TFP: Evidence from China
2023, Research in International Business and FinanceCan buying weapons from your friends make you better off? Evidence from NATO
2023, Economic Modelling