حسابداری خلاقانه و یا تخریب خلاقانه؟ رشد بهره وری در سطح شرکت ها در صنایع تولیدی چین
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|11766||2012||13 صفحه PDF||سفارش دهید||13360 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Development Economics, Volume 97, Issue 2, March 2012, Pages 339–351
We present the first comprehensive set of firm-level total factor productivity (TFP) estimates for China's manufacturing sector that spans China's entry into the WTO. For our preferred estimate, which adjusts for a number of potential sources of measurement error and bias, the weighted average annual productivity growth for incumbents is 2.85% for a gross output production function and 7.96% for a value added production function over the period 1998–2007. This is among the highest compared to other countries. Productivity growth at the industry level is even higher, reflecting the dynamic force of creative destruction. Over the entire period, net entry accounts for over two thirds of total TFP growth. In contrast to earlier studies looking at total non-agriculture including services, we find that TFP growth dominates input accumulation as a source of output growth.
China has enjoyed impressive labor productivity growth averaging nearly 8% for a period now spanning three decades. Considerable debate persists over the sources of this growth and the relative contributions of improvements in total factor productivity (TFP) versus the mobilization of additional resources, notably physical and human capital. Studies using aggregate data and combining agriculture and non-agriculture typically find TFP contributing approximately half of labor productivity growth (Bosworth and Collins, 2008 and Perkins and Rawski, 2008). In a widely cited study focusing solely on the non-agriculture sector covering the period up through 1998, Young (2003) paints a much less impressive picture of China's growth story. Correcting for potential biases in official deflators and the measurement of human capita, but otherwise using official data, Young reduces the estimate of productivity growth for the sector between 1978 and 1998 from a very respectable 3% to a more pedestrian 1.4%. Over this period, non-agriculture was the source of nearly 80% of GDP.1 The aggregate results hide important heterogeneity. TFP growth in industry, which represents forty percent of GDP and is the source of 90% of exports, is likely to be much higher than in the service sector, to which reform and market liberalization have only come with a long lag (Bosworth and Collins, 2008). Earlier empirical studies also identify a significant gap in productivity in industry between the rapidly expanding non-state sector and state-owned firms (Groves et al., 1994 and Jefferson and Rawski, 1994). Qualitatively, rising firm capabilities and productivity in industry have been linked to the expanding role of market forces, massive entry of new firms, and intense competition (Brandt et al., 2008). An analysis of Chinese manufacturing on par with that carried out for other countries has been handicapped by a lack of firm-level data sets. This constraint is gradually being relaxed, allowing more in-depth analysis at the micro level of key aspects of behavior in manufacturing that are missed at the macro level—see, for example, Bai et al., 2006, Dougherty et al., 2007 and Hsieh and Klenow, 2009, and Park et al. (2010). This paper builds on that work. Drawing on an unbalanced panel of firms between 1998 and 2007 that represents approximately 90% of gross output in manufacturing, we present the first comprehensive set of firm-level productivity estimates for Chinese manufacturing that spans China's entry into the World Trade Organization (WTO). The absolute size of China's manufacturing sector and its exports make this important in its own right. Over the period we examine, we find firm-level TFP growth of manufacturing firms averaging 2.85% for a gross output production function and 7.96% for a value added production function. Total TFP growth for the manufacturing sector was even higher due to massive entry of new firms with above average productivity levels and growth rates and the exodus of inefficient incumbents. When new firms replace exiting firms, the reallocation of input factors tends to enhance efficiency. Over the full sample period, our results identify net entry as the source of more than two thirds of total productivity growth, exceeding its contribution in U.S. manufacturing (Haltiwanger, 1997).2 In all, we find that TFP growth coming from improvements in continuing firms (the intensive margin of TFP growth) and through net entry (the extensive margin of TFP growth) was the source of over half of value added growth in manufacturing over the 1998–2007 period. TFP's contribution to labor productivity growth is even higher at two-thirds. The rest of the growth in value-added was the result of increases in total capital and labor use in manufacturing, much of which was associated with the entry of new firms. Our findings for the manufacturing sector are sharply at odds with the view of Young (2003) and others (Zheng et al., 2006) that productivity growth outside of agriculture has been mundane or ordinary. However, our results reveal that aggregate TFP growth in Chinese manufacturing remains constrained by limited efficiency-enhancing input reallocations between active firms, confirming results in Hsieh and Klenow (2009). These findings have important implications for government policy. First, the high firm-level TFP growth estimates imply that Chinese manufacturing output growth will not disappear any time soon as input accumulation diminishes. The labor force will peak in a few years (Perkins and Rawski, 2008), and rates of investment are expected to come down as China rebalances. TFP growth will also help firms in China weather rising labor and other input costs. Second, increasing competitive pressure and the adoption of new technology are often mentioned as drivers of TFP growth. Learning is not only important to the upgrading efforts and productivity growth among continuing firms, but is also equally important to the contribution of new entrants. For entrants, there are two dimensions to learning: first, identifying new opportunities making successful entry possible and second, improving productivity subsequent to entry. Policies that facilitate both kinds of learning are the key to sustained growth in the medium term. Third, as input growth slows and the technology gap with advanced countries narrows, further reforms to enhance efficient allocation of resources still provide important growth potential. A policy of liberalizing entry and facilitating exit has already played an important role in this regard. Removal of constraints that underpin productivity differences among existing firms, including those between the state and non-state sectors will have to be tackled next. Working with firm-level data for China has its difficulties. One of the additional contributions of this paper is to carefully describe and document these data. We make publically available online the complementary data we have constructed, including deflators, industry concordances, adjustment to capital stock series, etc. that are required to make full use of the data. Furthermore, in light of important concerns of Young and others, we examine the robustness of our results to a host of measurement issues. We show how alternative treatment of key variables often reduces productivity growth, but does not alter the basic picture.3 A particularly important aspect of the data work was the construction of linkages over time in firm-level observations when firm ID codes changed. This often occurs when active firms are restructured and it is important not to classify such instances as exit and subsequent entry. We find that one-sixth of the Chinese firms in our sample have at least one ID change. The ability to track firms as they are being restructured is an important precondition to being able to conclude that net entry has been the dominant force in productivity growth in Chinese manufacturing. The remainder of the paper is organized as follows. In the next section we describe our methodology for measuring productivity. Section 3 describes the data set and the construction of the key variables. An online Appendix provides more detailed documentation. In Section 4 we describe the Chinese results at the firm level, the performance of entrants and exiting firms, and the aggregate productivity growth experience. The latter allows us to “line up” our findings for industry with estimates from the literature for the entire economy. We also decompose the productivity residual to identify the types of heterogeneity most important to the aggregate evolution of productivity. Section 5 concludes.
نتیجه گیری انگلیسی
rawing on an unbalanced panel of firms that covers most of China's manufacturing sector, the purpose of this paper has been to examine the absolute size and the dynamics of productivity growth over a period that spans China's entry into the WTO. In our analysis, we have been especially attentive to a host of data and methodological issues. By all indications, productivity growth has been very rapid, a finding that appears to be robust to a host of measurement issues. This finding is in sharp contrast to alternative perspectives such as Young's (2003) that suggest modest productivity growth outside of agriculture. Improvements in productivity of “continuing firms”, either as a result of restructuring efforts, or investments in capability building are an important part of the picture. Equally, if not more important, are the gains to creative destruction, i.e. entry and exit, that China's decentralized reforms have increasingly allowed. For our preferred estimate, the weighted average annual productivity growth for incumbents is 2.85% for a gross output production function and 7.96% for a value added production function over the period 1998–2007. This is among the highest compared to other countries. Productivity growth at the industry level is even higher, reflecting the dynamic force of creative destruction. Over the entire period, net entry accounts for over two-thirds of total TFP growth, with growth of entrants after entry making an important contribution to this. In all, TFP growth dominates input accumulation as a source of output growth and is responsible for two thirds of labor productivity growth. These findings have important implications, not in the least for government policy. First, even with declining opportunities for growth through accumulation of additional inputs, the high TFP growth estimates suggest that we can expect robust growth in manufacturing in the near future. Moreover, TFP growth will help to maintain profitability in the sector in the face of rising labor and other input costs. Second, increasing competitive pressure and the absorption of foreign technology are often mentioned as drivers of TFP growth. Learning on the part of continuing firms as well as new entrants is critical to taking advantage of these sources of growth. For entrants, there are two dimensions to learning: first, identifying new opportunities that allow successful entry; and second, improvements in productivity subsequent to entry. Policies that facilitate learning among both kinds of firms are the key to sustained growth in the medium term. As Chinese firms narrow the technology gap with advanced countries, more of the learning will have to come from within firms. Third, despite the dynamism we document, our results also point to continued constraints on the growth of some of the most productive of firms. Problems in the allocation of credit and biases in favor of larger firms with state-sector connections are potential reasons for this. With growth prospects on the extensive margin limited, new reforms to enhance efficient resource allocation still provide important growth potential. A policy of liberalizing entry and facilitating exit has already played an important role in reallocating resources to new firms. Constraints that sustain the remaining productivity differences among existing firms, including between the state and non-state sectors, will have to be tackled next. More work is needed to identify the exact nature of the constraints impeding reallocation, and how they can be best removed.