China’s S&T Reform and TFP over Past Four Decades of Reform and Opening-Up
After 40 years of relentless efforts, China’s reform of the S&T system has achieved great progress and has invigorated the vibrancy of innovation. As noted in the Report of the 19th CPC National Congress, China should continue reforming its S&T system, create a market-based innovation system with companies as the backbone and production-education-research integration, support the innovation of SMEs and promote S&T commercialization. S&T innovation has yet to play a greater role in supporting economic development. After taking stock of the achievements and challenges of China’s S&T reform, this paper puts forward policy recommendations on deepening S&T reform and implementing an innovation-driven development strategy.
Based on China’s aggregate data of economic development during 1979-2015, this paper employs growth accounting method to estimate the TFP index denoted by the Solow residual. Our research found that since reform and opening up in 1978, China’s TFP experienced ups and downs and TFP improvement spurred rapid economic development, while capital input remains the primary source of China’s economic growth. During 20082015, China’s TFP growth slowed and high capital input coexisted with low productivity and high economic growth.
scientific and technology system, total factor productivity, capital driven JEL classification code: O14, O32, D24
DOI: 1 0.19602/j .chinaeconomist.2018.01.04
1. China’s S&T Reform over the Past 40 Years of Reform and Opening-Up
After its founding in 1949, the People’s Republic of China developed a state-controlled science and technology (S&T) system under the planned economy. With highly centralized administration and an inflexible personnel system, research institutions were detached from the production system (Li, 2000). It took decades for China’s leadership to realize the inefficiencies of the planned economy. Since reform and opening-up in 1978, the Communist Party of China (CPC) Central Committee has shifted its priority to the “four modernizations” (i.e. modernization of industry, agriculture, national defense, and science and technology) (Kou, 2008). With the creation of a market-based system, Chi-
na’s innovation and S&T systems transitioned from the planned system to a market-based one.
1.1 China’s S&T Reform in Retrospect
Since reform and opening-up in 1978, China has been striving to explore an optimal path for S&T development and reforming its S&T system at both the macro and micro levels with the following objectives: to integrate S&T research with socio-economic development, to optimize S&T resources allocation and develop enterprises into the backbone of innovation, to promote indigenous innovation and innovation-driven development, to create a national innovation system, and build an innovationbased country.
In this paper, we divide China’s S&T reform since 1978 into five stages marked by the following milestones: (1) the National Science Conference in March 1978 and the enactment of the Outline of National S&T Development Planning 1978-1985 (Draft); (2) the Decisions of the CPC Central Committee on S&T reform enacted by the State Council in 1985; (3) the National Science Conference convened by the CPC Central Committee in May 1995, which unveiled the “Strategy of Invigorating China Through Science and Education” and released the Decisions on Accelerating the Progress of Science and Technology; (4) the release of the Outline of the National Mid- and Long-Term Planning for the Development of Science and Technology (2006-2020) and the strategy to develop an “innovation-based country” by the State Council; (5) the innovation-driven development strategy adopted at the 18th CPC National Congress in 2012. In these stages, China’s S&T reform exhibited different characteristics.
1.1.1 Reconstruction of the S&T system (1978-1984)
The Third Plenary Session of the 11th CPC Central Committee in 1978 unveiled a new chapter of China’s S&T development as the government’s priority shifted to economic development and the existing S&T system became increasingly obsolete (Liao, Dai, 2009). The National Science Conference in 1978 identified science and technology as a productive force, intellectuals as part of the working class and the modernization of science and technology as the key to the “four modernizations”. In 1980, the CPC Central Committee adopted a strategic guideline entitled “We Must Rely on Science and Technology for Economic Development and Science and Technology Must Serve Economic Development”. In this stage, the objective of reform was to reconstruct China’s S&T system and explore an appropriate path of reform in line with a market-based economy.
1.1.2 Adjustment and innovation of the S&T system (1985-1994)
The Decisions of the CPC Central Committee on S&T Reform enacted on March 5, 1985 marked the beginning of China’s S& T reform, which aimed to “swiftly and extensively apply the results of science and technology to production and give full play to the role of scientists and researchers in productivity and socio-economic development” and issued clear guidelines on reforms of S&T management and financing systems. In this stage, China reformed its scientific research institutions on a full scale to promote S&T research in sync with economic development (Liao, Dai, 2009).
1.1.3 Implementation of the “Strategy of Invigorating China through Science and Education” (19952005)
In 1995, the “Strategy of Invigorating China through Science and Education” was put forward at the National Science and Technology Conference - a key guideline in this stage. In 1998, the State Council decided to reform the administrative system of 242 scientific research institutes under the State Economic and Trade Commission (SETC) and released the Decisions on Enhancing Technology Innovation, Developing High Technology and Achieving Industrialization and the Outline of Report on the (Knowledge Innovation Project) Pilot Program of the Chinese Academy of Sciences. These
policy initiatives aimed to develop a national innovation system with enterprises as the backbone (Chao, Fang, Xie, 2013). The National Technology Innovation Conference held in 1999 was followed by a host of policies to promote corporate innovation. In this stage, the reform focused on developing a national innovation system and S&T commercialization.
1.1.4 Developing an innovation-based country (2006-2012)
In 2006, the State Council released the “Outline of National Mid- and Long- Term S& T Development Planning (2006-2020)”, which called for creating an innovation system with enterprises as the backbone, promoting production- education- research integration and turning China into an innovation-based country by 2020.
In January 2006, the CPC Central Committee and the State Council adopted the Decisions on Implementing the Outline of Science and Technology Development Planning to Enhance Indigenous Innovation; in February 2006, the State Council released the Supporting Policies for Implementing the Outline of National Mid- and Long-Term Science and Technology Development Planning (20062020). In March 2006, the CPC Central Committee and the State Council adopted the Decisions on Implementing the Outline of Science and Technology Development Planning to Enhance Indigenous Innovation.
In 2007, the 31st Session of the Standing Committee of the 10th National People’s Congress adopted the amended Scientific and Technological Progress Law. In this stage, the Chinese government attached great importance to the role of intellectual property rights in promoting socio-economic development. In 2008, the State Council adopted the Outline of the National Intellectual Property Strategy.
In 2009, the State Council released the Opinions on Further Promoting the Development of Smalland Medium-Sized Enterprises.
1.1.5 Developing an innovation-based country (2012-Present)
In November 2012, the Report of the 18th CPC National Congress adopted the strategy of innovation-driven development and called for “reforming the S&T system, integrating S&T research with economic development, fostering a national innovation system, creating a market-based system of innovation with enterprises as the backbone, and promoting production-education-research integration.”
In November 2013, the Third Plenary Session of the 18th CPC Central Committee adopted the
Decisions of the CPC Central Committee on Comprehensively Deepening Reforms (“Decisions”). Section (13) of the Decisions specifically elaborated on questions concerning S&T reform and its objectives, i.e. “to eliminate institutional barriers, improve indigenous innovation, follow innovation-driven socio-economic development, improve the national innovation system and build an innovation-based country”.
In March 2015, the CPC Central Committee and the State Council released the Opinions on Deepening Institutional Reforms and Expediting the Implementation of an Innovation- Driven Development Strategy, which identified 30 reform initiatives to create a level playing field, increase financial support, develop a market-oriented mechanism, commercialize S&T, improve the R&D system, expedite talent cultivation and flow, propel open innovation and enhance coordination. These reform initiatives aimed to promote the reasonable allocation of innovation resources, increase the dynamism of innovation and promote “mass entrepreneurship and innovation”.
On June 4, 2015, the 93rd Executive Meeting of the State Council approved the Opinions on Policy Measures to Promote Mass Entrepreneurship and Innovation, which called for financial capital to support entrepreneurship and innovation as drivers of industry development and job creation.
On October 29, 2015, the Fifth Plenary Session of the 18th CPC Central Committee adopted the
Suggestions of the CPC Central Committee on Formulating the 13th Five-Year Plan for Social and Economic Development, which called for “innovative, green, balanced, open and shared” development
and “putting innovation at the center of national development and fostering a pro-innovation social environment”.
On May 19, 2016, the CPC Central Committee and the State Council released the Outline of National Innovation-Driven Development Strategy, which called for “promoting S&T innovation, inspiring innovation through institutional reforms, and creating an innovation-based country”. It also laid out priorities for innovative industrial technology, original innovation, regional innovation layout, civil-military integration, innovation entities, major S&T programs, talent development, as well as innovation and entrepreneurship.
In November 2016, the General Office of the CPC Central Committee and the General Office of the State Council released the Opinions on Implementing Policies to Increase the Value of Knowledge to encourage R&D personnel to pursue innovation and entrepreneurship and commercialize S&T.
From the Decisions of the Third Plenary Session of the 18th CPC Central Committee to the Outline of Innovation-Driven Strategy, China not only identified the objectives for reforming the S&T system but carried out top-level designs for the reform as well. Priorities for reform of the innovation system include S&T administration, allocation of S&T resources and assessment of innovation results, production-education-research cooperation and R&D commercialization, talent attraction and cultivation, as well as innovation incentives and risk scattering.
1.2 Effects of Institutional Reforms
Over the past three decades, China’s reforms of the S&T system have achieved great progress. Specifically, China improved its S&T system, fostered innovation entities including research institutes,
universities, enterprises and technology intermediaries and enhanced the role of enterprises as the backbone of innovation (Yun, 2009); transformed the S&T operation mechanism, introduced competition into S&T resources allocation; reformed scientific research institutes of various types; enacted the Scientific and Technological Progress Law, Patent Law and Law on Promoting the Transformation of Scientific and Technological Results and initially developed a system of S&T policies and regulations; and enhanced S&T innovation capacity (Chen, 2013).
Since the 18th CPC National Congress, the State Council and its various departments released a host of policy documents on reform of the innovation system involving various aspects of macro-regulation and management, incentives and risk dispersion, production-education-research cooperation and S&T commercialization. Reforms have yielded policy dividends and unleashed great momentum of innovation and creativity.
1.2.1 Pro-innovation reforms and policies
The reform initiatives are intended to adopt innovative macro-regulation, promote production-education-research integration, develop mechanisms for S&T commercialization, increase fiscal support to innovation, foster innovation-intensive industries including the Internet, e-commerce, cloud computing and big data, and create a pro-entrepreneurship and pro- innovation environment.
1.2.2 S&T reforms boosted innovation, entrepreneurship and market vibrancy
S&T reforms have increased market vibrancy and unleashed great momentum of innovation and entrepreneurship in cities, municipalities and provinces such as Beijing, Tianjin, Shanghai, Zhejiang, Jiangsu and Shenzhen. Entrepreneurship and innovation have played a positive role in fostering new industries and business models and propelling economic development. In 2016, 16.51 million new businesses were registered in China, including 4.46 million businesses in the tertiary sector. In 2016, a wide range of sectors demonstrated a strong momentum, including IT, software, energy efficiency and environmental protection, new energy, high-end manufacturing, new materials, biomedicine, culture and creativity, financial services, professional technology services and R&D services. Since the Third Plenary Session of the 18th CPC Central Committee, private capital has played a pivotal role in innovation. According to Wind’s database, the number of disclosed venture investment events increased from 1,225 to 2,897 during 2013-2015, up 53.8% per year; the size of investment from disclosed events increased from 63.1 billion yuan to 408.56 billion yuan, up 154.4% per year.
1.2.3 S&T reforms propelled innovation
In 2016, China’s national innovation capacity ranked 15th in the world, up from 19th in the previous year. In 2016, innovation contributed 60% to scientific and technological progress. In 2016, total R&D spending as a share in GDP reached 2.5%. The output of S&T is remarkable. By the end of 2016, China had accepted a total of 3.465 million patent applications, ranking sixth in the world for six consecutive years, and licensed 1.34 million invention patents, ranking first in the world. China ranked second in the world in terms of the number of international papers published by scientists and researchers and third in terms of the number of paper citations. S&T research is brought closer with the economy. In 2016, the contract transaction volume of China’s technology market reached 1.14 trillion yuan. China now ranks first in the world in terms of S&T human resources and R&D personnel, accounting for 29.2% of the world total. China has achieved major technological breakthroughs with growing S&T competitiveness and international influence. China has developed a host of S&T research facilities and innovation platforms and a fairly complete mechanism for the sharing of public S&T resources. Reforms of the S&T system have facilitated implementation of an innovation-driven development strategy and yielded great results including the Tiangong space station, Jiaolong submersible, the five-hundred-meter Aperture Spherical Telescope (FAST), Wukong dark matter particle
explorer satellite, Mozi quantum communication satellite and jumbo jets.
1.3 Problems Existing in Reforms of the S&T System
We must also be mindful of the incompleteness of the S&T system, inefficiencies in the use of S&T resources and limited S&T support to economic development. Reforms of the S&T system still have a long way to go.
(1) Allocation of S&T resources still has great potentials to improve. S&T resources are scattered and not utilized in an efficient manner. The allocation of S&T funds is inefficient and wasteful. By putting a premium on short-term results, the current assessment mechanism discourages theoretical research and knowledge creation. Administrative interventions in research projects also affect the creativity of R&D personnel. Leading researchers have limited autonomy in the allocation of human and financial resources and the decision-making of technical pat hways.
(2) Innovation has yet to be led by companies. The government assumes the roles and functions of introducing S&T legislations, improving public services, creating a level playing field and enhancing supervision. Universities and scientific research institutes are the suppliers of innovation results and companies are responsible for applying those results. The research activities of key universities and institutes are not completely technology-oriented. S&T commercialization must be undertaken by the consortiums of universities, research institutes and ent erprises, corporatized research institutes and intermediaries. By allocating resources in favor of heavy and chemical industries, real estate and infrastructure, the government offers insufficient guidance and support to corporate innovation, failing to incentivize innovation. Market distortions also prevent resources and factors from supporting innovation. Companies tend to copy foreign technologies rather than pursue risky innovations.
(3) Production-education-research cooperation has yet to take hold. Universities and research institutes, which account for 70% of China’s scientific research resources, are disconnected from the industry, inhibiting R&D commercialization. This disconnect partially derives from the incomplete reforms of scientific research institutions. Despite government enthusiasm, enterprises and research institutes lack the motivations for innovation.
(4) Evaluation and incentive systems are obsolete. The evaluation mechanism that focuses on short-term results is unfavorable to theoretical research and public knowledge creation (Song, 2008). Due to the overemphasis on the publication of papers including SCI papers, researchers spend most of their time publishing papers instead of pursuing real research and are not motivated to commercialize R&D. The lack of operational agreements on the sharing of intellectual property rights and profits from commercialization has also discouraged researchers. In addition, researchers face barriers to move between public and private sectors.
1.4 Suggestions on Further Reforming the S&T System
Since the 18th CPC National Congress, the CPC Central Committee, ministries and local governments introduced a host of systems and measures on S&T reform and innovation-driven development.
(1) Promoting market-based allocation of S&T resources and developing an innovation system where companies hold sway. The Report of the 19th CPC National Congress calls for deepening S&T reform, creating a market-oriented innovation system with companies as the backbone and production- education- research integration, enhancing support to the innovation of SMEs and propelling S& T commercialization. The government should refrain from intervening in market operation at the micro level and withdraw from sectors free from market failure to give play to the fundamental role of market in the allocation of S&T resources. Research groups should be created and disbanded based on market mechanisms. Stakeholders must identify potential opportunities for cooperation
第四，科技评价制度、激励制度等不能适应科技发展新形势的要求。科技评价导向不够合理，片面追求短期效果的考核评价机制不利于科学研究中基础理论的探索和公共知识的创造（宋海龙，2008）。唯S C I论，“以论文论英雄”情况严重。大量科研人员为获得晋升、项目经费，将大量精力用在发表论文上，难以潜心做科研。激励优秀人才、鼓励创新创业的机制不完善，科技人员的积极性和创造性还没有得到充分发挥。具体表现在，一是从评价体系上看，目前大学和科研院所以立项、发表论文、获奖和职称评定为主要导向，科研人员缺乏科研成果转化的积极性。二是从利益分配机制上看，由于对知识产权、成果转化收益等合作成果分享缺乏明确可操作的约定，导致科研人员缺乏积极性。三是人才流动机制不健全。科研人员跨体制、跨部门双向流动机制的缺失，影响了科技人力资源在全社会的有效配置。
based on common interest and maximize return and efficiency based on market rules and competition. The government must carry out market-based reforms of factors to incentivize corporate R&D spending, protect intellectual property rights to ensure reason able return to innovation, share S&T resources among companies of different ownerships, implement the pre-tax deduction policy for R&D expenses, encourage companies to establish their own R&D institutions, offer incentives to innovative companies, and encourage companies to lead major national S&T programs.
( 2) Promoting market- based innovation. China’s industrial development requires more high-quality innovation that can be commercialized. In strategic emerging industries, in particular, universities and research institutes face the challenge of speeding up innovation. Therefore, universities and research institutes must extensively work with companies to meet corporate demand for innovation. We must also support the development of generic and public-interest technologies through service procurement and support innovation resources to be competitively allocated to companies to optimize the allocation of capital, technology and other factor s.
( 3) Incentivizing researchers to pursue innovation and entrepreneurship. We must develop a cooperation mechanism for R&D commercialization among companies, universities and research institutes that share risks and benefits. We must also strive to improve the institutional environment and provide more public technology service systems for S& T commercialization. Reform of the income distribution system must encourage individuals to innovate and commercialize science and technology. In addition, we must improve the venture investment mechanism and adopt innovative business models to propel the industrialization of science and technology.
We must promptly adjust industry regulation rules according to changing business models arising from new technologies, enhance the creation, protection and use of intellectual property rights and improve legislation to incentivize and protect proprietors, and integrate human sources of research institutions with the existing factors of SMEs.
( 4) Further improving talent cultivation and fostering young talents. We need to create an internationally competitive talent management system to attract high-end talents; foster discipline leaders through major scientific research and development programs, explore a corporate chief scientist system, and broaden channels for recruiting overseas talents; develop scientific research teams headed by chief scientists and encourage collaborative innovation among researchers; reduce institutional barriers to the free flow of scientists; and incen tivize and foster young scientists.
(5) Fostering a pro-innovation cultural environment. We should promote a culture that encourages long-term research and tolerates mistakes to support fundamental research. We should also encourage entrepreneurship, innovation and scientific awareness among the public in order for innovation to take hold as a national spirit and social culture.
2. China’s TFP Changes since Reform and Opening-Up in 1978 2.1 TFP Definition and Calculation
Economic growth derives from the growth of factor inputs and productivity. Productivity refers to how efficient intermediates are converted into output in the production process. Productivity can be divided into single factor productivity (SFP) and total factor productivity (TFP) by the scope of input factors. The Measuring Productivity - OECD Manual defines TFP as the contribution of all input factors to output growth.
Mainstream TFP measurement models include the Solow residual method and Stoch astic frontier analysis. Although the econometric model may relax the assumptions of a c omplete competition market and constant return to scale in growth accounting, it must provide priori hypothesis of parameters
and is restricted by the data quantity of sample observations, which may cause statistical problems like unstable parameter estimation. In estimating aggregate production function using an econometric model, it is appropriate to conduct the estimation based on a simultaneous equation that includes demand for production factors rather than a simple progression based on a single equation in order to address the problem of endogeneity (Li, 1992; Nadiri and Prucha, 2001). Since 1978, China has experienced a host of socio-economic volatilities and institutional reforms, as reflected in its socioeconomic indicators being much more volatile than those of advanced economies. Hence, the Solow residual econometric model and stochastic frontier approach (SFA) based on national aggregate data may lead to major deviations due to the poor robustness of parameters. Panel data of 30 provinces may also cause inaccuracies in production frontier based on the DEA method.
In comparison, growth accounting is more appropriate for the statistical analysis of productivity for countries or regions with significant economic development volatility. The Measuring Productivity - OECD Manual (2010) also recommends using growth accounting for TFP estimation, which is believed to be the most common method for estimating TFP (Ren, 2013). Hence, this study employs growth accounting to estimate China’s TFP index of 1979-2015 denoted by Solow residual and compare it with the result of estimation conducted by other scholars and research agencies.
Based on the assumptions of perfect competition, exogenous tech nological progress and constant return to the scale of input factors, Denison (1967) believed that the marginal output of input factors equals the return to factors, i.e. return to the scale of output is constant and the output elasticity coefficient equals the shares of the respective return to factor. Labor and capital income statistics can be used to determine the output elasticity coefficient of input factors and estimate TFP. Jorgenson and Griliches (1967) believed that Denison’s growth accounting method decomposed growth sources into labor, capital and technological progress, thus potentially overestimating TFP due to the omission of major variables. They argued that different production factors should be incorporated and precisely measured to reduce the overestimation of productivity caused by computational errors. On the basis of estimating labor, physical capital and TFP in relation to economic growth, this paper further incorporates the human capital variable that denotes workforce quality and the R&D variable that denotes intellectual capital to estimate TFP growth rate.
2.2 Estimation of the TFP Index
Based on the growth accounting method, this paper has estimated China’s TFP of 1979-2015, including (1) TFP index measured with the total workforce as labor input without considering the human capital factor ( tfp_wp); (2) TFP index estimated with human capital inventory as labor input ( tfp_hc); (3) TFP index taking into account the variations in R&D capital inventory that denotes intellectual capital ( tfp_rd). As Figure 1 shows, the three types of TFP share consistent trends with little difference between TFP containing human capital factor ( tfp_hc) and TFP containing R&D capital valuations ( tfp_rd), both of which are smaller than TFP without the human capital factor taken into account ( tfp_wp). The implication is that separating the human capital factor and TFP index from the residual value denoted by traditional TFP may address the overestimation of traditional TFP and help the “residual” part to represent technological progress and efficiency improvement of aggregate economic production.
As can be seen from Figure 1, China’s TFP index ( tfp_hc) had been in ups and downs with wild and frequent volatility during 1979-1993, a period of great economic and market transformations for China. The household contract responsibility system and the reform of SOEs in the wake of reform and opening up (1978-1984) vastly unleashed productivity and led to a surge in TFP. As reform deepened, institutional contradictions emerged and China’s TFP index plummeted, giving rise to great volatility in this stage. During 1994-2000, China’s TFP reduced gradually after a brief recovery. By 2000, the TFP index only reached 0.41%.
经济增长的来源分为两种，一是生产要素投入量的增长，二是生产率的提高。其中，生产率指生产过程中投入品转化为产出的效率，按衡量投入要素的范围大小可以分为“单要素生产率”（S i n g l e F a c t o r Productivity, SFP）和“全要素生产率（”Total Factor Productivity, TFP）。OECD在《生产率测算手册》中将全要素生产率定义为，测算所有投入要素对产出增长贡献的一种能力。
目前主流的T F P测算模型主要包括索洛余值法和生产前沿方法两大类。经济计量模型虽然可以放松增长核算方法中关于完全竞争市场和规模报酬不变等假设，但必须对估计的参数给出先验假设，并且受到样本观察值数据量的限制，容易出现参数估计不稳定等统计问题。另外，采用经济计量模型估计总量生产函数不宜仅采用单一方程进行简单回归，而应采用包括生产要素需求的联立方程体系进行估计，以处理相应的内生性问题（李子奈，1992；Nadiri 和 Prucha, 2001）。而且，自1978年以来，我国经历了一系列经济社会波动和经济体制机制改革。相对于西方发达国家而言，我国反映经济社会发展趋势的各类指标数据呈现较大幅度的波动。采用全国总量数据进行的索洛余值计量模型方法和随机前沿方法（S F A），可能会因为估计参数的稳健性低而出现较大偏误。而仅由30个省份构成的面板数据，也可能导致DEA方法构建生产前沿面的不准确。
相较而言，增长核算方法更适合经济发展趋势波动较大国家或地区的定期生产率统计研究。O E C D （2010）发布的《生产率测算手册》也因此推荐使用增长核算方法来估计全要素生产率，认为此方法是目前采用最广泛的测算全要素生产率的方法（任若恩等，2013）。因此，本研究采用经济增长核算方法，测算索洛余值代表的我国1979~2015年全要素生产率指数，并同目前学者和研究机构的测算结果进行比较，得出相关结论。
2.3 Comparative Analysis of Factor Contributions
2.3.1 TFP Contribution
With the exception of a few abnormal years, TFP contribution to growth also fluctuated during 1979-2000 and 2001-2015, which is consistent with the trend of the TFP index itself. Yet overall volatility was smooth. During 1979-2015, the average contribution of China’s TFP to economic growth stood at 21.27%. During 1979-2000, TFP contribution to economic growth experienced greater vol-
全社会就业人数作为劳动力投入，测算出的全要素生产率指数（t f p _ w p）；②考虑人力资本因素，以第二部分测算出的人力资本存量作为劳动力投入测算出的全要素生产率指数（t f p _ h c）；③考虑代表经济社会知识资本的R&D资本存量，测算包含R&D资本变动的全要素生产率指数（tfp_rd）。如图1所示，三类全要素生产率的变动趋势较为一致，其中考虑人力资本因素的全要素生产率（t f p _ h c）和包含R& D资本变动的全要素生产率
atility, averaging a low level of 14.56%. During 2000-2015, TFP contribution was more stable, averaging 31.13%，much higher than the average contribution in the previous round of volatility. This implies that technological progress and efficiency improvement became key drivers of China’s economic growth.
2.3.2 Comparison of production factors
Contribution of the input factor is the ratio of the product between the share of return to input factor and growth rate to economic growth rate. Take TFP ( tfp_hc) estimation for instance: the average contribution of physical capital to economic growth was the highest during 1979-2015 at 50.08%, followed by human capital contribution at 28.64% and TFP contribution at 21.27% (Figure 3).
During 1979-1981, physical capital contributed the most to economic growth, followed by human capital input and then TFP. During 1992-1986, TFP contribution increased rapidly amid a host of institutional reforms, next only to physical capital. During 2000-2007, China’s fixed capital investment increased rapidly with physical capital accumulation outpacing the economic growth rate and contribution to growth increasing over the years. TFP contribution also increased under the effect of innovation strategies and technology incentives. Human capital growth, however, kept falling with China’s ageing population and nine-year compulsory education becoming universalized. Since the eruption of the global financial crisis in 2008, China made tremendous fixed asset investments to shore up growth, causing the growth contribution of physical capital inventory to surge. During 2008-2015, physical capital contributed an average of 69.06% to growth, up 22 percentage points over 2000-2007. Meanwhile, China’s late-mover advantage for innovation diminished, as reflected in falling TFP contribution. During 2008-2015, average TFP contribution was 18 percentage points below the average level of 2000- 2007, reflecting the crowding- out effect of rapid but inefficient physical capital accumulation on TFP. During 2013-2015, China’s economy shifted gears from rapid growth to medium-high growth with the growth rate falling to 6.9% and the economy entering the new normal. During this period, physical capital contribution was as high as 70.98% and TFP contribution slightly increased to 21.43%, while the share of human capital contribution slid to 7.59%. The process of restructuring and reforms still has a long way to go .
从图1可以看出，1979~2015年，我国全要素生产率指数（t f p _ h c）总体呈现涨跌互现的波动情形，其中1979~1993年波动幅度较大且较为频繁。原因在于，1993年以前，我国正处于经济体制和市场条件发生剧烈变化的时期，改革初期（1978~1984年）的家庭联产承包责任制和国企放权让利等制度变革，使生产力得到极大解放，从而促进全要素生产率快速增长。随着改革的推进，制度上的深层次矛盾逐渐显现，我国全要素生产率指数快速下降，形成了这一阶段生产率增速大幅波动的趋势。1994~2000年，我国全要素生产率经过短期的恢复增长后逐年下降，到2000年，全要素生产率指数仅为0.41%。
与全要素生产率指数的变动趋势一致，除了个别异常年份外，全要素生产率对经济增长的贡献也大致经历了1979~2000年和2001~2015年两次先升后降的波动阶段，总体波动较为平稳。1979~2015年，我国全要素生产率对经济增长的平均贡献率为21.27%，其中1979~2000年，全要素生产率贡献率的波动幅度较大，平均贡献率为14.56%，对经济增长的贡献率相对较低；2001~2015年，全要素生产率贡献率波动幅度较小，平均贡献率为31.13%，远高于前一波动周期的平均贡献水平。说明以全要素生产率为代表的技术进步和效率提高已经成为驱动我国经济增长的重要因素，其对经济增长的贡献水平明显提升（见图2）。 2.各类生产要素贡献比较 投入要素的贡献，等于投入要素的报酬份额和增长率乘积与经济增长率的比值。以全要素生产率tf p _ h c
3. Concluding Remarks and Policy Recommendations
Based on growth accounting, this paper estimated China’s TFP changes and economic growth sources with the following conclusions: ( 1) Since reform and opening- up, China’s productivity growth has experienced ups and downs and served as a key driver of growth. During 2008-2015, as capital input continued to increase, the TFP index slowly decreased due to diminishing late-mover advantage. (2) China’s economic growth has been driven by capital input with physical capital contribution to economic growth remaining over 50% and steadily rising, up to over 70% in recent years. Yet in the face of growing energy and environmental constraints, it is rather challenge for the physical capital-driven growth to sustain in the long run. (3) Labor input contributed the least to China’s economic growth. Contribution of labor input to growth has been flat due to limited change of population and surplus labor that inhibited the improvement of marginal labor output. Falling labor contribution has to do with China’s diminishing demographic dividend and the slow response of labor market to industrial restructuring.
The above conclusions have the following policy implications fo r China:
Firstly, China’s economic development pattern should transition from the crude pattern to an intensive one. After the dawn of the 21st century, China’s productivity has been falling with economic growth heavily dependent on inputs, particularly after the eruption of the global financial crisis in 2008. This crude growth pattern is characterized by quantity and speed and will become unsustainable with China’s industrialization and urbanization slowing down and population increasingly ageing. Therefore, China should transition towards an intensive growth pattern driven by quality, efficiency and a combination of production factors. Growth should derive from technological progress and improving workforce quality.
Secondly, innovation holds the key to China’s growth sustainability. In the past, China’s economic growth was more dependent on factor accumulation than technological progress - a path that proves to be unsustainable. In the future, China’s economic growth should be supported by technological progress, particularly indigenous innovation, instead of factor accumulation.
Thirdly, economic reforms must be deepened to further increase China’s productivity. After the dawn of the 21st century, China’s economic reforms decelerated and productivity declined as reform dividends became exhausted. Falling productivity will inhibit the long- term growth rate, making
economic development unsustainable. To ensure long-term productivity growth, China must deepen reforms and promote fair competition.
Figure 1: China’s TFP Index during 1979-2015 (%)
Figure 2: TFP Contributions to China’s Economic Growth during 1979-2015 (%)
图1 我国1979~ 2015年的全要素生产率指数（%）
Figure 3. Factor Contributions to Economic Growth (%)