پویایی ساختار بازار و تنظیم اصلاحات آزادسازی اقتصادی در هند در سطح بنگاه به نفع بازار ، 1988-2006: روش انتقال زمان متفاوت پانل صاف رگرسیون (TV-PSTR)
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|13239||2011||14 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : International Review of Economics & Finance, Volume 20, Issue 4, October 2011, Pages 506–519
This paper for the first time employs the Time Varying Panel Smooth Transition Regression (TV-PSTR) approach to model the dynamic adjustments of firms and the evolution of industrial structure in the bigger setting of decades against the backdrop of India's dramatic liberalizing reform starting from 1991. Using Indian manufacturing firm data, it finds that the transition of market structure and productivity after liberalization did follow a smooth transition process. Instead of the previously assumed instantaneous ‘big-bang’ shift just after reforms, it actually took years for the Indian manufacturing industries start to react to the reforms, and the transitional impact of reforms took approximately four to eight years to complete. There is strong evidence of increased competition after the transition, with shrinked returns to scale (RTS) in most industries except for leather and chemical industries. The results on total factor productivity (TFP) are mixed: most import-competing industries, which suffer most from the shrinking of market size experienced no change or decreasing TFP growth; whereas the export-oriented industry, as the industry which benefits most from economy of scale, enjoyed a huge TFP growth following the reforms.
During the past several decades, lots of developing countries launched dramatic pro-market economic liberalization in an effort to attain higher growth. These liberalization efforts, broadly defined to include trade and entry liberalization, regulatory reform, ER regime reform and privatization, are believed to transform economies via more competition (domestic and foreign) and the removal of distortions in relative prices. Therefore it may bring the country welfare gains through several possible channels, which have been the questions receiving the extensive interests of country policy makers. First, it has been argued that, in imperfectly competitive markets, pro-market liberalization will bring welfare gains by reducing the dead weight losses created by domestic monopolies and oligopolies by increasing competition, and by reducing price-marginal cost markups.1 Secondly, there may be scale efficiency gain of trade by moving the firms down their average cost curves, thereby effectively raising firm size and scale efficiency. In addition, trade liberalization, as the major component of Indian pro-market liberalizing efforts, has been argued to have dynamic effects on firms' productivity growth through innovation. In contrast to the theoretical predictions on the effect of liberalization on competition and markups, theoretical predictions about firm activity (dynamic effects) from macro models of entry liberalization and deregulation are ambiguous (see endogenous growth theories by Grossman and Helpman, 1990, Grossman and Helpman, 1994 and Melitz and Costantini, 2008). Trade can potentially be growth generating as well as growth decelerating. Trade can enhance growth permanently by facilitating the international exchange of knowledge and technology. Trade can have growth decelerating effects if it, via market size effects, reduces domestic firms' incentives to innovate or diverts resources away from R&D.2 As outlined by Melitz and Costantini (2008), these dynamic effects are characterized by a heterogeneous firm-level adjustment process following the reforms. Firm-level productivity evolves stochastically, and innovation involves a trade-off between its cost and a return in terms of a better distribution of future productivity draws. Moreover, their model shows there will be an industrial evolution over its entire transition path to a new steady state. In particular, how the relative timing and magnitude of firm-level productivity improvements and export market entry decisions are also determined by non-technological factors such as the timing of trade liberalization announcements and the speed of liberalization. Although theory has provided us some implications on liberalization-growth nexus, few studies have systematically examined the growth performance of firms, or more specifically model the dynamic adjustments process of micro-economic industrial structure following the liberalizing reforms.3 In the context of developing countries, several studies have explored the relationship between firm productivity and trade reforms using firm level data. Tybout et al. (1991) find no evidence of increased productivity following liberalization in Chile. Harrison, 1994, Tybout and Daniel Westbrook, 1995, Pavcnik, 2002 and Fernandes, 2007 and Muendler (2004), on the other hand, do observe productivity increases following liberalization in, respectively, Cote d'Ivoire, Mexico, Chile, Colombia and Brazil. Krishna and Mitra (1998) find mixed results of change in productivity growth in four manufacturing industries with weak significance. However, they, as with almost all previous studies, assumed an instantaneous discrete shift in market efficiency parameters and productivity growth following the reforms, either by imposing a post-reform dummy or using one period lag output tariffs as a measure of trade reforms. However, linear estimation cannot account for nonlinear dynamics by imposing the unrealistic and restricting assumption of coefficient stability. Instead, it is more reasonable to model the transition after liberalization as a sequenced smooth process by a flexible functional form.4 There are two main reasons behind this argument: first, reforms taking time to gain credibility and market reactions; secondly, micro level restructure after unexpected macro level reforms usually take time due to various rigidities and adjustment cost of investment at micro level. It usually takes years before the dynamic effects of trade on productivity growth start to happen, and continue for years. This study takes a new approach to the question, Time Varying Panel Smooth Transition Regression (TV-PSTR) model.5 The starting point of this study is to recognize that the core mechanism that drives economic growth following liberalization is massive microeconomic restructuring and factor reallocation, which must be a slow process. Hence, it starts from the presumption that any changes in economic performance following reforms and liberalization may be more appropriately modeled as a steady transition rather than a discrete change. A standard explicit or implicit assumption underlying linear models is that there is a single structural break in the sample. In this study that assumption is replaced by a more general one stating that the parameters of the model may change continuously and smoothly as a function of time, which is more consistent with the firms' adjustment behaviors and economic evolution process. Moreover, this TV-PSTR model not only allows for discrete changes in parameters, but also allow for any form of nonlinear transition path. With discrete change (‘big bang’ shift) in parameters as a special case within this more general framework, the model doesn't lose its generality. Instead of using a priori information to fix the date of a transition, the speed and the timing of the transition are endogenously determined by the data. With the above merits, the TV-PSTR model has great advantage over conventional approach and doesn't put any prior estimation restrictions on the process of transition. Utilizing this TV-PSTR approach and the natural liberalizing experiment of India, I document some stylized facts about the evolution of India's industrial structure against the backdrop of India's unexpected dramatic liberalizing reform starting from 1991. More precisely, I estimate the transition of average industry level price-marginal cost markups, RTS and productivity growth in nine two-digit level manufacturing industries in India using firm level panel data. A full production function with substantial flexibility is employed, which allows for both non-perfect competitive market and non-constant return to scales. By relaxing the perfect competition and CRS assumptions,6 this study intends to capture the gains from market and scale efficiency. After we identify speed parameter and hence the dynamic production function, we will be able evaluate the average industrial total factor productivity (TFP) changes as the residuals in output growth after singling out estimated changes due to factor growth. Note that this study do not formally test whether liberalization results in growth. The results are, however, informative in two respects. Firstly, they post a clear picture of the adjustment and evolution process of Indian industries after pro-market liberalizing reforms. Secondly, they also point the way to improved econometric modeling of these processes. India is a suitable case study because of its long history of protecting its domestic manufacturing sector. Moreover, the extensive changes in the economic regime of India coming unexpectedly after several decades of restrictive external policies, provided a good controlled experiment. This unanticipation, together with the gradual nature of the reforms and the time reform policy takes to gain credibility, leads to slow adjustments of firms, which justify the use of the new approach. Last but not least, the Indian data set used in this paper contains detailed firm level data on a large sample of firms in a variety of industries, thereby facilitating analysis at a higher level of disaggregation than previous studies. The main findings of this paper are that the transition of market structure and productivity after liberalization do follow a smooth transition process. Instead of the previously assumed instantaneous ‘big-bang’ shift just after reforms, it actually took years for the Indian manufacturing industries start to react to the reforms, and the transitional impact of reforms took approximately four to eight years to complete. There is strong evidence of increased competition, which reduces the markup and make welfare gains possible from the reduction of dead weight losses. Except for the leather and chemical industries, RTS in most industries shrink after the transition. The effects of reforms on total factor productivity (TFP) are mixed: most import-competing industries, which suffer most from the shrinking of market size experienced no change or decreasing TFP growth; whereas the export-oriented industry, as the industry which benefits most from economy of scale, enjoyed a huge TFP growth in response to reforms. The main contribution of this paper is that for the first time it applies the Time Varying Panel Smooth Transition Regression (TV-PSTR) model to study the dynamics of market structure and productivity growth after liberalization in an emerging market. There is a growing body of empirical economic literature on the effects of reforms right after the reform year assuming there is a single structural break in the sample, but none of them characterize and model the adjustments and transition of the economy in the bigger setting of decades following the initial reforms. This study makes an attempt to fill this gap in the literature. This study proposes an innovative way of modeling slow structural changes as a smooth transition between states before and after any unexpected regime switching. In particular, instead of using a priori information to fix the date of transition, the speed and the timing of the transition are endogenously determined by the data. The remainder of this paper is organized as follows. Section 2 briefly discusses the complex economic liberalizing reforms in India. Section 3 outlines the Time Varying Panel Smooth Transition Regression (TV-PSTR) model and estimation methodology, discusses some econometric issues and describes the data. Section 4 presents and discusses the estimation results. Section 5 concludes.
نتیجه گیری انگلیسی
In this paper, I employ the TV-PSTR model to investigate the transition dynamics of price–marginal cost markups, returns to scales (RTS) and productivity growth in Indian manufacturing industries associated with India's unexpected dramatic pro-market economic liberalizing reforms. Using Indian manufacturing firm level data for the period from 1988 to 2006. I find that the transition after liberalization does follow a smooth process, instead of the previously assumed instantaneous ‘big-bang’ shift just after reforms. The dynamic adjustments of industrial structure do not occur immediately after the primary reforms, and do not occur over the same time period for all industries. Also the length of the transition process varies across industries. It actually took years for Indian firms in manufacturing industries to react to the reforms, and the transitional impact of reforms takes approximately four to eight years to complete. There is strong evidence of increases in competition, which pushes down the markup and makes it possible to obtain welfare gains from reduction of dead weight losses by increasing competition and lower markups. Except for the leather industries, RTS in most industries shrink after the transition. As predicted by the endogenous growth theory, the effects of reforms on TFP are mixed, depending on whether trade is increasing its market size or not, and hence possibly encouraging or discouraging R&D and innovation. After the liberalization reforms, generally, import-competing industries which suffer heavily from the shrinking market size, experienced no significant increase in TFP growth; whereas the export-oriented industry, which benefit most from economy of scale, enjoyed a huge TFP growth in response to reforms. In terms of policy implication, it is important to facilitate exports and exploit returns to scale, and innovation spurring policy environment to stimulate productivity growth and counteract the possible disincentive to innovate caused by shrinking market size. As pointed out by Panagariya (2004), a lack of acceleration of growth in the industrial sector is the most disappointing aspect of the 1990s experience, and also the key to explaining why India nevertheless continues to lag behind China. In 1980, the proportion of GDP originating in the industry was already 48.5% in China, in India it was only 24.2 (Table 7). Services, on the other hand, contributed only 21.4% to GDP in China but as much as 37.2% in India. In the following20 years, despite considerable growth, the share of industry did not rise in India. Instead, the entire decline in the share of agriculture was absorbed by services. Though a similar process was observed in China, the share of industry in GDP was already quite high there. To catch up, India are suggested to free the industry of continuing restraints: bring all tariffs down to 10% or less; abolish the small-scale labor-intensive industries reservation (draconian labor laws); institute an exit policy and bankruptcy laws; and privatize all public sector undertakings. In some ways, given the advantage India enjoys in the information technology sector over China, its overall prospects for growth are even better than those of China but only if the conventional industry is given a fair chance, but only when if it has a more strong formal sector.