Crossing the Growth Threshold: Service-Based Economy, Knowledge Process and Reshaping of Efficiency Model
Abstract: The transition from middle-income to high-income stage is fraught with risks of growth divergence. Economic transition is clouded by the following possibilities: (1) falling share of industrial sector through industrial depression and weakening growth momentum caused by the large urbanization costs; (2) the subordination of service sector as a result of nearly irreversibly industrial professional, which falters the process of service sector transition and upgrading; (3) inefficient knowledge production allocation and human capital upgrade due to the absence of incentivized compensation of knowledge consumption. We suggest that a country should reshape its efficiency model by upgrading knowledge factor and human capital as the pre-requisite. Given the dilemmas of transition, China should take the factorization trend of service sector and reshape efficiency model through institutional reform, ensuring that service sector will develop in tandem with industrial sector.
Keywords: Growth threshold, service-based economy, knowledge process, reshaping of efficiency model
JEL Classification: C13, E20, O11
For those late-moving countries after massive industrialization, urbanization and service-based economy is a critical stage of divergence that determines their growth sustainability and success of catch-up. The growth of transition economies, like China, could be interrupted by the demanding threshold of knowledge accumulation. Noncontinuity of growth implies the failure of direct usage of the intuition from industrialization on the service-based economy. The economic transition is undermined by the following risks: (1) at the macro level, service-led growth is vulnerable to the failure of industry/service coordination, which
is embodied in rising urbanization and falling share of industrial sector engineering industrial depression, or stagnant technology and efficiency improvement; (2) at the level of industries, despite the rising share of service sector, knowledge process1 as the core of service production and factorization faltered to preventing the completion of service sector transition and upgrade; ( 3) at the level of factor supply, the path of knowledge production and allocation and human capital upgrade is impeded due to the absence of efficiency compensation of human capital and knowledge consumption as the cornerstone for crossing the growth threshold.
Urbanization and service- based economy have increased the divergence or efficiency differences among the worldwide economies. Empirical analysis of this paper provides three scenarios: (1) high-efficiency model characterized by “high productivity, consumption and capital intensification capabilities” of OECD countries; (2) the “stop-go” unstable and inefficient model of Latin American countries dominated by traditional services and low- level consumption structure; (3) the successful transition of Japan and South Korea which crossed the growth threshold thanks to human capital accumulation in advance of 15 to 20 years.
As shown by international experience, ( 1) service-based economy as a brand-new efficiency model other than industrialization, in which the upgrade of quality based on knowledge and human capital (skilled workers with higher education) matters rather than the rising shares of service sector per se; (2) what matters is not whether investment continues to drive economic growth but whether developing countries possess capital intensification capabilities (i.e. capabilities to upgrade consumption structure), together with consumption capabilities; ( 3) for a superlarge economy like China, the transition period coincides with deepening industrialization in the absence of endogenous efficiency mechanism, the quantitative expansion of service sector cannot be over emphasized. Hence, China’s transition stage should also be marked as a buffer stage of structural upgrade to prevent de-industrialization of Latin America type and prevent the risk of repercussions on growth.
To illustrate the above views, the rest of this paper is arranged as follows. Section 2 describes the stylized facts of non- continuity of growth and growth divergence. Section 3 presents our views on the non-continuity of growth and the uncertainties in crossing the growth threshold. Section 4 describes propositions of service sector factorization trend. The last section is our conclusion.
2. Typical Facts of Growth Noncontinuity and Growth Divergence
In this section, the use of international comparative information and the observation of typical facts of growth non-continuity are based on the two overlapped contexts of economic evolution, i.e. transition from industrialization to urbanization and the leap from middle- income stage to high-income stage. Evolution from lowefficiency model to high-income efficiency model is a common experience for both developed countries and late-moving countries. Such change is formally described in economic theory as transition, cumulative causation and adjustment. While linking technology progress and increasing remuneration to long-term growth, Kaldor (1970, 1972, 1985) developed a systematic theory of cumulative causation followed by more formal statements in subsequent literature (e.g. Dixon & Thirlwall, 1975, Setterfield, 1997).
Based on database PWT8.1, we employ the following methodologies to observe the following facts of economic leap: ( 1) depict the relative labor productivity , relative capital formation per worker (for capital intensification) and relative consumption per worker (or aggregate consumer power supported by each and every worker) of sample countries during 1950- 2011 with the United States as benchmark for comparison; ( 2) employ labor productivity q and aggregate output Y to estimate Verdoorn coefficient - or
the ability to capture returns to scale; and based on GDP accounting by expenditure approach, employ the labor productivity q , aggregate capital formation i and aggregate household consumption level c of various countries to estimate Verdoorn coefficients ai and ac with the following estimation equations:
Catch-up trajectory of labor productivity in sample countries during 1950-2011 is shown in Figure 1:
2.1 Continuous Efficiency Improvement Is Key to Crossing the Growth Threshold
First, we explain the specific features of highefficiency model and low-efficiency model based on the catch-up trajectory of labor productivity shown in Figure 1. Our overall assessment is as follows: (1) during sample period of more than half a century, sample countries from East Asia, Latin America and Europe generally exhibit an S- shape catch- up trajectory; ( 2) countries that have completed catch- up, including European countries and East Asia’s Japan and South Korea, exhibit a significant S-shape catch-up trajectory; (3) the adjustment and repercussions of postwar Latin American countries lasting for over half a century warrant a great deal of attention; (4) East Asian countries including China, Thailand and the Philippines remain in the accelerated catchup process and lag behind developed countries and even Latin American countries in terms of labor productivity. Specifically, various efficiency models are characterized as follows:
Recovery and catch-up under high-efficiency model for France, Germany and Italy: for traditional European powers rebuilt on the ruins of World War II, their systems and institutions had been shaped by modern capitalism from the very start and characterized by high productivity and consumption. Therefore, what they did was to recover economic vitality rather than engage in any substantive
transition of efficiency model. Attempt of Japan and South Korea to build a high-efficiency model for catch-up: after Japan’s structural economic deceleration in the 1980s, the problems of original industrialization model were not perceived until in the 1990s, sparking a policy debate on whether Japan should copy the US model or partially converge on and accommodate its national uniqueness ( Dirks et al., 1999). Long- term adjustment and attempt of Latin America to build a high-efficiency model became thwarted: restrained by initial conditions and path dependence, Latin America is a typical example of failed attempt to build a high-efficiency model, forcing the economy into long-term adjustment and repercussions.
Compared with the above scenarios, emerging industrial countries such as China, Thailand, Indonesia and the Philippines are still in the catch-up stage under low-efficiency model. What warrants our attention is that the two rapidly industrializing countries of China and Thailand appear to face the urgency to adjust and reshape
efficiency model prior to reaching Latin America’s level of productivity.
Second, based on the relative consumption per worker and relative capital formation per worker in Table 1, we offer explanations on some statistical facts of economic catch- up to further clarify the connotations of efficiency models. Even if we put aside European countries with superior initial conditions for catch-up and focus on the comparison with Japan, South Korea and other less efficient countries, the facts that have been revealed are still shocking:
Fact 1: Capital intensification is the first to complete the catch-up process. When capital per worker ( ) reaches the US level, consumption per worker in the catch-up country is roughly equivalent to 40- 50% of the US level, which marks the completion of catch- up process and establishment of high- efficiency and highconsumption model. For instance, Japan’s average
was 0.87 during 1970-1980 and South Korea’s was 0.89 during 1991-1997. It took about 30 years
for both countries to reach the investment level of the United States starting from their respective economic acceleration. However, two things warrant our special attention: ( 1) rapid capital intensification prevented the potential risk of repercussions that may occur should the transition toward high- efficiency model take a long time to complete; ( 2) capital intensification was accompanied by significant increases of per capita consumption, giving rise to the benign cycle of capital intensification, rising consumption and labor productivity improvement.
Fact 2: Long- term economic adjustment fundamentally results from the manipulation of domestic vested interests, which prevents the development of high- efficiency model despite economic liberalization. As revealed by the comparative economic performance of Latin American countries during 1950-1980 and 19802011 in Table 1, shifting back to the exportoriented efficiency model after the eruption of debt crisis did not bring about any significant improvement of investment, consumption and labor productivity. In fact, the situation even deteriorated in some countries.
Fact 3: Capital intensification and investment/ consumption efficiency are equally important. It is necessary to turn the advanced economic model of Goodwin et al. (1997) into the efficiency triangle of “labor productivity, consumption and capital intensification.” As shown by the long- term adjustment of Latin American countries, the high-efficiency model cannot be accomplished in the absence of consumption or capital intensification capabilities as the two angles of the efficiency triangle.
2.2 Ability to Capture Increasing Return and the Efficiency Compensation of Consumption Are the Foundation for Crossing the Growth Threshold
The fact of increasing return to scale is implied in the S-shape path, which is evident in the long-term catch-up process. By illustrating the ability to capture increasing return to scale, this section continues to enrich the content of various efficiency models as shown in Figure 2:
Continuously increasing returns to scale are a common phenomenon: (1) the path of economic catch-up for various countries in sample period is
depicted by the scatter diagram Figure
2 of q and total output Y. In the sample period of more than half a century, for the economic recovery of traditional developed countries such as France and Germany and the creation of highefficiency model by Japan and South Korea, the continuous increase of labor productivity ( ) that accompanied rising total output ( ) and the nearly linear statistical relationship between them clearly indicate the vitality of a successful catch- up economy. For developing countries of East Asia, the linear relationship between labor productivity and total output growth in the course of their longterm growth also comply with the economic laws of increasing return to scale in the low-efficiency model. For Latin American countries such as Colombia, Mexico and Venezuela, there exists a significant non-linear relationship between labor productivity improvement ( ) and change in total output ( ), while the increasing return to scale resulting from economic expansion is not a longterm phenomenon as is the case with developed countries and East Asian countries. ( 2) Ability of total output Y to capture increasing return to scale. Steadiness in the ability to capture return to scale for developed countries: during the two time frames of 1950-1973 and 1973-2011 divided by 1973, aY for developed countries is as follows: US: 0.62, 0.61; France: 0.90, 0.84 ; Germany: 0.85, 0.83; Italy: 1.01, 0.82; Japan: 0.90, 0.90. The low-efficiency model of developing countries in East Asia also has strong abilities to capture return to scale: e.g. 1.01 for China during 19782011, 0.90 for Thailand during 1970-2011, 1.03 for the Philippines during 1970-2011 and 0.57 for Indonesia during 1970-2011.
Efficiency compensation of consumption: evidence for the existence of consumption/ investment dual efficiency model. Contribution of total investment expansion and total household consumption expansion to labor productivity improvement in Table 2 exhibits some major characteristics of high- efficiency model. ( 1) Consumption/investment dual efficiency model provides further evidence to the efficiency triangle. Judging by the factors of labor productivity improvement in developed countries, increasing investment scale and household consumption have significant abilities to capture increasing return to scale in the long- term and given a higher consumption elasticity of labor productivity compared to investment elasticity, the high-efficiency model of advanced economies is characterized by greater vitality of consumption. It needs to be noted that such a conclusion is based on the perspective of growth rate and the deeper logic is that under the high-efficiency model of high consumption and capital intensification capabilities, consumption contributes more to efficiency than does investment. In other words, human capital related to consumption contributes more to the efficiency triangle than does the expansion of material capital. ( 2) Efficiency compensation of consumption. As can be found from the comparison of the two sample periods of developed countries before and after 1973, the expansion of consumption contributed the most to labor productivity and, after developed countries entered into structural deceleration and maturity of urbanization, consumption played a more significant role in efficiency compensation and contribution. Thus, we further arrive at the following facts:
Fact 4: Stable high- efficiency triangle is based upon consumption/investment integration and the efficiency compensation of consumption is particularly significant during low-growth period.
China’s growth stage and commonalities of similar growth stage: evidence for the existence of ( capital- driven) single efficiency model is significant. In countries like China, high-efficiency model is ultimately dominated by strong capital intensification capabilities ( high investment per worker means that capital contributes less to efficiency than does consumption). In these countries, therefore, the significant effect of investment on increasing return to scale is different in comparison with developed countries before 1973. Non-continuity of growth from the standpoint of long- term adjustment: evidence for the existence of ( consumption- driven) single efficiency model. Volatility exists in the labor productivity of Latin American countries: in the long- term, labor productivity is on the decline or improves slowly and the effect of total investment of labor productivity is insignificant;
economic growth relies on the unstable support of consumption.
Fact 5: The nature of growth non-continuity is the disconnect between investment and consumption that undermines the basis of efficiency triangle; single investment efficiency model or single consumption efficiency model will both cause preventable efficiency leakage, forcing the economy into long-term adjustment fraught with uncertainties and frustrating the continuous upgrade toward high– efficiency model. Thus, low- efficiency model itself has the problem of instability.
3. Non-continuity of Growth and Crossing of Growth Threshold: Three Uncertainties
Due to stumbles in the adjustment of the old efficiency model, the disconnect between production and consumption may solidify lowefficiency model in the long-term growth process of developing countries and thus cause the noncontinuity of growth. In this paper, the noncontinuity of growth is defined as the long-term adjustment in the transition from low-efficiency model to high-efficiency model with the objective to cross the threshold and achieve continuous efficiency improvement through the accumulation of new factors. Such non-continuity is specifically embodied in the three uncertainties that constitute the middle-income threshold.
3.1 Uncertainty 1: Rupture between Industrialization and Urbanization Leading to Growth Stagnation
Traditional development theory is concerned with the escape of poverty trap and the achievement of modern growth path. Judging by the common experience of late-moving countries, the boundary between mass industrialization and urbanization in the course of long- term growth is explicitly embodied in the concentration of population in cities and rising share of service sector. In the stage of industrialization, economic growth is driven by resource concentration and utilization, while centralization (urban growth poles), economies of scale and standardization lie at the heart of efficiency improvement. After an economy enters into the stage of urbanization, the foundation for centralized resource utilization is lost: the growing diversity of demand, rising share of service sector and sophistication of technical innovation all require significant improvement of decentralized market decision-making, knowledge innovation and human capital accumulation as key drivers of economic growth. However, the problem is that if the efficiency model where industrialization is fueled by capital accumulation, as emphasized by traditional development theory, is driven by exogenous technology progress and primary labor factor, this growth pattern will trigger growth non- continuity and the rupture of efficiency improvement path as late– moving countries transition from industrialization to urbanization.
As revealed by empirical comparison, latemoving countries are trapped in a system of division of labor dominated by multinational firms, which prevents them from achieving technology progress and innovation; on the other hand, these countries also tend to follow an approach of homogeneous technology progress through “learning by doing” ( import of manufacturing equipment) (Research Group, 2007, 2009). However, the trek from technology progress of “learning by doing” to indigenous innovation is fraught with uncertainties, including uncertainties in human capital accumulation, market demand, capital market incentives, IPR protection and change of culture and corporate profitability model. Due to the lack of indigenous learning and knowledge process in the course of industrialization, the explicit growth path characterized by industry- led efficiency improvement gave way to growth divergence and the failure of coordination between industries and services in the new stage of growth transition and urbanization. Specifically, the falling share of industrial sector is accompanied by the falling speed of industrial growth compounded by rapid factory relocations and the rupture of industrial technology and efficiency upgrade.
3.2 Uncertainty 2: Spread of Inefficient Services Leading to Population Drift and Baumol’s Cost Disease
Uncertainty of transition is embodied at the level of industries, i.e. the subordination of service sector as a result of industrial division of labor cannot be fundamentally reversed and the factorization of service sector with knowledge process as the core is faltering, preventing service sector transition and upgrade and thus aggravating population drift and Baumol’s cost disease. Therefore, the process of an increasingly servicebased economy is fraught with uncertainties and divergence of structural transition: while it may propel the transition and upgrade of economic structure, enhance efficiency and economic stability and provide better social welfare, it may also lead to falling efficiency, economic instability and stagnation.
For successful transition, the following growth promotion mechanisms are implied in the growing share of service sector: (1) the promotion of coordination, which is a major aspect of a service-based economy (Leal, 2015). For instance, the ICT application of developed countries has dominated industrialization and industrial coordination has enhanced efficiency (Xie et al., 2016); (2) consumption growth and upgrade fuel demand for products and services with greater technical sophistication and induce the share of skills-intensive services to rise (Buera & Kaboski, 2012); (3) the price of skills-intensive services is consistent with the remuneration premium of highly skilled personnel, which incentivize human capital accumulation and knowledge production. Zhang and Guo ( 2016) demonstrated how human capital enhanced labor reproduction (with consumption as the starting point of knowledge process). However, in the process of transition and even the long-term adjustment of service-based economy, what late-moving countries experienced is the aggravation of Baumol’s cost disease under service-led growth and the lock-up of inefficient growth model. Under the effects of Engel’s law and consumer preference, the less efficient service sector of developing countries still enjoys sustained growth and substitutes more efficient industrial sectors through the “flying geese paradigm” of domestic industries, compromising the potentials of overall economic efficiency improvement. Given the extensive regulation of service sector, the growing share of service sector is inevitably accompanied by high cost and may even cause the economy to regress into a rent extraction model.
Given that the role of service sector as a vehicle of knowledge production cannot be brought into full play, the process of urbanization and service- based economy is still dictated by low-quality “population drift” - given the large number of rural population employed in the urban informal sector (such as peddlers), labor-intensive service sector will exist as a pool of low-quality labor. Population drift of Latin America is shown in Table 3:
3.3 Uncertainty 3: Lack of Efficient Compensation Mechanism for Consumption
As the economy becomes increasingly service- based, the contribution of service upgrade and growth to the improvement of overall economic efficiency (for the achievement of high- efficiency model) hinges upon a fundamental embedded mechanism: the efficiency compensation mechanism of consumption. Experience shows that consumption and economic structure become service-based at the same time but the divergence of high-efficiency and lowefficiency models in this process derives from differences in the efficiency compensation of consumption. The mechanism of efficiency
compensation induces the upgrade of human capital and service sector through consumption upgrade and propels the formation of knowledge process and transition toward high- efficiency model (see Figure 3 in Section 5). On the contrary, the stagnation of consumption upgrade will cause growth to stall.
During China’s rapid growth in the period of 1992- 2011, skyrocketing investments contributed almost 70% of efficiency growth. Disconnect between production and consumption under investment-driven growth will affect the cultivation of long- term growth potentials and improvement of low- efficiency model. While outsize investment excluded the opportunities to upgrade consumption and capture return to scale, distribution in favor of capital has repressed the propensity of consumption, as reflected in the imbalance between supply structure and consumption structure. Single- minded focus on the efficiency model of investment will disable long- term capital intensification as a result of emphasis on short- term investment. This single efficiency model has significant efficiency loopholes, which include: (1) in order to maintain short- term growth rate, economic authorities resorted to traditional low-efficiency options such as infrastructure construction and real estate, forcing the economy into a vicious cycle of growth, inefficiency, reinvestment and inefficiency; ( 2) low- level household consumption demand has restricted the boundaries of economies of scale and capital intensification; ( 3) certain groups who benefit from the old model ( especially the middle class in big cities) have a real demand for a diverse range of high-quality and sophisticated consumer goods, which cannot be satisfied by domestic industrial structure. As a result, such consumer power is driven to overseas outlets, causing a spillover effect on the efficiency improvement of overseas industries.
4. Abatement of Uncertainties in Efficiency Leap through Efficiency Improvement and Knowledge Process
In light of international growth experience and China’s economic reality, the following directions of adjustment should be explored to reduce the uncertainty in crossing the growth threshold. First, is it feasible to leap into urbanization and service-based economy while endogenous momentum is lacking? In other words, does China need a buffer period for industrial intensification? Second, what is the direction for the adjustment of China’s service sector? Third, why is consumption upgrade important?
4.1 Deepening, Coordination and Buffer of Industrialization
The form of industrial reorganization and intensification is subject to the domestic and international environments of economic development. Fundamentally speaking, a buffer period of industrial intensification is required for the transition from industrialization to urbanization. Urbanization should not be advanced as the single priority and still less should raising land price become a core strategy of urbanization. Artificial “de-industrialization” in such form leaves insufficient space and time for the upgrade of industrial technology and efficiency. For an industrial country like China that relies on primary factors as key drivers of economy, abandoning the process of industrial intensification prematurely will present barriers to continuous efficiency improvement. If industrialization loses steam, producer services will suffer as well and service sector will also lose the opportunities of efficiency improvement. As a result, the service sector will not be able to provide positive feedback and inducement to industrial intensification. In this sense, manufacturing intensification serves as a pillar of efficiency improvement.
Based on the above analysis, this paper argues that in the transition stage, a buffer zone should be provided for the improvement of China’s industrial structure, which comprises the following elements: (1) the regional potentials of China as a super- large economy should be utilized to optimize the flying geese pattern and structure of industries across regions. More
developed provinces and cities should focus on upgrading the structure of service sector and promote service-based economy; the structural upgrade of service sector should serve as the origin of human capital accumulation and knowledge production to promote the relocation and coordination of industries across regions in the flying geese pattern; ( 2) Coordination between industry and service sector: the key of such coordination is that the falling share of industrial sector should be based on the premise of improving industrial efficiency; the increasing share of service sector should not restrain the improvement of overall economic efficiency. Given China’s current status of transition, the development of service sector must be supported by structural upgrade and efficiency compensation. Otherwise, the urbanization risks of Latin America type may occur. ( 3) Gradually reshape the path of technology/ efficiency upgrade. Another important aspect of China’s manufacturing upgrade is to create an institutional mechanism of innovation and efficiency improvement to lift firms out from the dilemma of “learning by doing” characterized by equipment importation and low- price competition and help them transition toward more heterogeneous indigenous innovation.
4.2 Knowledge Process, Efficiency Improvement and Service Sector Upgrade
Inspired by Machlup ( 2007), we refer to knowledge production and allocation and the cycle and improvement of economic efficiency on such a basis as the “knowledge process”. High- efficiency model i n the advanced economic stage is based on the foundation of advanced service sector structure. Following this understanding, as advanced economies become increasingly service-based, they rely on service sector as the pre-requisite of economic growth and the knowledge sector spearheads the high- efficiency model. Empirical evidence for such understanding is as follows: (1) for a service-driven economy with a falling share of industrial sector, the efficiency and improvement potentials of service sector determine overall economic e ff i c iency and improvemen t potentials. Knowledge production and allocation in the sectors of higher education, R&D and IT services determine the efficiency improvement of other service and economic sectors. (2) Rising share and structural upgrade of service sector can be regarded as the substitution of traditional commodities ( goods and services) by the knowledge process, in which the allocation of knowledge to industrial sector will improve the performance of manufactured goods and replace traditional manual service sectors; meanwhile, knowledge and technology- intensive services become more diversified, with certain traditional consumption services replaced by knowledge consumption services. ( 3) Improvement of service sector tradability: supported by the network- based development of knowledge and information, service sector becomes more tradable, compensating for the falling share of industrial trade. It warrants our particular attention that, given the stronger monopolistic nature of knowledge, once knowledge- based trade in services is established, it is not only more profitable than traditional trade but may gain monopolistic and competitive advantages that are unpredictable in the short term as well.
4.3 Efficiency Compensation of Consumption and Growth Sustainability
Structural upgrade and particularly the effect of knowledge process, rather than size and proportion, are vital to the growth of consumption and services. As mentioned before, there are two possible paths for the transition toward advanced urbanization: first, under the inertia of quantitative industrial expansion, the development of service sector is dominated by the reproduction of low- skilled labor; the second path is the growth of service sector supported by knowledge process. As the economy becomes increasingly service- based, the expansion of service sector in size and proportion is inevitable. However, the impetus of such expansion should be the structural upgrade of service sector and the linkage between consumption upgrade and service sector growth. After the satisfaction of demand for essential goods, consumer choice is increasingly
determined by the psychological demand for diversity and novelty. In particular, the rapid development of service sector and contemporary knowledge and information sparked consumer fads. Knowledge process is thus related to consumption efficiency in time and space and finds expression in the factorization trend of knowledge- intensive services. Take, for instance, the joint cognition and knowledge sharing through face-to-face communication for instance:
Scenario 1: Capitalization of time through consumption: consumption in such sectors as education, leisure and entertainment no longer takes place instantly as is believed in traditional theory. Instead, consumption related to knowledge products should be regarded as a process: a new characteristic of modern producer services. In the process of face- toface communication, knowledge producers create and disseminate knowledge, while consumers receive knowledge. In the scenario of market exchange, consumers pay the cost according to the novelty of information streams ( transmission of information in time). In this process, consumers voluntarily pay for different knowledge streams based on the level of their psychological satisfaction and are willing to pay a premium for high-level knowledge.
Scenario 2: Consumption capitalizes space: the roundaboutness of consumption, i. e. the biggest difference of networking between the age of service- based economy and industrial age is the significant role of knowledge and information network. In addition to the capitalization of time, consumption has also capitalized space in advanced economies during urbanization. This is helped by the increased tradability of knowledge-intensive services and closer distance of face-to-face communication. Capitalization of space not only facilitates the dissemination of knowledge streams and novelties and increases knowledge productivity and output but is conducive to consumer market segmentation, making it possible to extract customized services from an ocean of redundant information. Similarly, dedicated knowledge service and premiums are also extracted, shedding light on the direction of structural upgrade and efficiency improvement of modern services.
5. Conclusion: Improving China’s Economic Efficiency through Reform
Evolution from industrialization t o urbanization involves the adjustment of growth pattern and the fostering of innovation. In the transition to urbanization and service- based economy, the key to crossing the threshold is knowledge creation through human capital accumulation. It is not only the foundation of a stable efficiency triangle, but also a brandnew model of service sector different from any other growth stages. Service-based transition of economic structure is driven by three factors: consumer preference, relative price, human capital accumulation. The last one is spurred by knowledge consumption and dynamic efficiency compensation. Among late- moving countries, a great deal of knowledge consumption and services related to human capital improvement falls into the category of public goods, thus is often strictly regulated. In this circumstance, the consumption of services driven by preference led to an undersupply in these sectors and requires supply to be increased through a price hike. This is equivalent to the imposition of “monopolistic rent” of knowledge service sector upon consumers, depriving consumer surplus, and service demand shifts overseas and the domestic system is left underdeveloped. Hence, market-oriented reform is inevitable in this stage. Structural upgrade of service sector is fundamental to efficiency improvement and prevents the regression from leading to monopolistic rent extraction model ( Research Group, 2014). The following questions need to be highlighted:
( 1) How to approach ser v ice sector development? If service sector development continues to follow the pattern of mass industrialization without even adjusting the pattern of resource allocation, China’s economy is likely to fall into long- term adjustment
and repercussions of Latin American type. Fortunately, China’s urbanization has not gone that far and some potential systemic problems are still avoidable. We contend that service sector should develop toward factorization as part of the following cycle starting with knowledge process development:
In a nutshell, the transition and upgrade of service sector and human capital accumulation are essential preparations for crossing the threshold.
(2) How to define the role of government? Efficiency compensation of consumption in the era of urbanization requires labor reproduction based on human capital enhancement. Due to their extremely high cost of production, externalities and dedicated use, knowledge and human capital ( particularly high- end) for higher education and R&D cannot be developed without the support of public sector. In order to achieve the factorization of service sector and growth leap, the government must improve the fairness of income distribution and attach importance to inputs for the development of knowledge process.
Specifically, the government should transform its functions and improve systems and rules against the backdrop of an increasingly service-based economy. With respect to income distribution, the comparative experience of Latin America, Japan and South Korea shows that a buffer period (about 20 years) is needed before the achievement of service- based economy. During this period, Japan and South Korea swiftly accumulated human capital to broaden the space for knowledge creation and reshape efficiency model for urbanization.
Inputs of knowledge process development include: (a) training system for skilled workers; ( b) incentives for skilled workers; ( c) highend talent system; and ( d) fundamental R& D support system. Without doubt, the government must adjust public spending to foster economic potentials. In a word, the role of government is essential in developing knowledge process and crossing the growth threshold under the servicebased economy.
( 3) How to incentivize innovations? In the serviced- based economy, China cannot cross the growth threshold and catch up with advanced economies without pursuing
innovation. Innovation is no longer a narrow concept of inventions and creations, but a more comprehensive and systematic concept involving the improvement of systems and rules, knowledge production, as well as allocation network and the integration of consumption and production. The reason is that the service-based economy requires more complex coordination of economic systems, channeling innovations and human capital incentives. Efficiency improvement and sustainable growth must derive from the production of “non-competitive” elements of growth, together with institutional rules, sharing of ideas and knowledge, education and information networks. The cycle of these elements lays the groundwork for a country to inspire innovation and cross the growth threshold.
 Buera, F. J., and J. P. Kaboski. 2012. “The Rise of the Service Economy.” American Economic Review, 102(6), 2540-2569.  Bulmer-Thomas, Victor. 2000. The Economic History of Latin America since Independence ( Chinese version). Beijing: China Economic Publishing House.
 Dirks, D., Jean-Francois Huchet, T. Ribault. 1999. Japanese Management in the Low Growth Era. Berlin: Springer Verlag, 8-9.
 Dixon, R., and Thirlwall, A. P. 1975. “A Model of Regional Growth- Rate Differences on Kaldorian Lines.” Oxford Economic Papers, 27(2), 201-214.
 ECLA, 1951. Economic Survey of Latin America 1949. New York: United Nations.
 Economic Growth and Macroeconomic Stability Subject Team. 2007. “Labor Supply Effect and Transition of China’s Economic Growth Trajectory.” Economic Research, no.12.  Economic Growth and Macroeconomic Stability Subject Team. 2009. “Urbanization, Industrial Efficiency and Economic Growth.” Economic Research, no.10.  Economic Growth Frontier Subject Team. 2012. “Trajectory, Efficiency and Potential Growth Level of China’s Economy.” Economic Research, no.11.
 ——. 2013. “Structural Characteristics, Risk and Efficiency Enhancing Path of China’s Economic Transition.” Economic Research, no.10.
 ——. 2014. “Low Efficiency Shock of China’s Economic Growth and Deceleration Governance .” Economic Research, no.12.
 ——. 2015. “New Factor Supply Theory, System and Policy Choice which Overcome Economic Deceleration.” Economic Research, no.11.
 Goodwin, N. R., F. Ackerman, and D. Kiron. 1997. The
Consumer Society. Washington, D.C.: Island Press.
 Herrendorf, B., R. Rogerson, and Á. Valentinyi. 2014. “Growth and Structural Transformation.” Handbook of Economic Growth, 6 (2): 855-941.
 Kaldor, N. 1970. “The Case for Regional Policies.” Scottish
Journal of Political Economy, 17(3): 337-348.
 ——. 1972. “The Irrelevance of Equilibrium Economics.” The
Economic Journal, 82(328): 1237-1255.
 ——. 1985. Economics without Equilibrium. UK: University
College of Cardiff Press.
 Leal, J. 2015. “Which Sectors Make Poor Countries so Unproductive? A Perspective from Inter-sectoral Linkages.”
Banco de Mexico, February 15.
 Machlup, Fritz. 2007. The Production and Distribution of Knowledge in the United States (Chinese version). Beijing: China Renmin University Press.
 Ozawa, T. 2005. Institutions, Industrial Upgrading, and Economic Performance in Japan: The “Flying- Geese” Paradigm of Catch-up Growth. UK: Edward Elgar Publishing.  Setterfield M. 1997. “History versus Equilibrium and the Theory of Economic Growth.” Cambridge Journal of Economics, 21(3): 365-378.
 Xie, Kang, Jinghua Xiao , and Cheng Fang. 2016 . “Coordinating Cost and Economic Growth: from the Perspective of Industrialization and Information Convergence.” Economic Perspectives, no.5.
 Yuan, Fuhua. 2012. “Structural Acceleration and Deceleration in the Process of Long-term Growth.” Economic Research, no.3.
 Zhang, Ping, Guanqing Guo. 2016. “Socialist Labor Reproduction, Labor Value Creation and Sharing: Theory, Evidence and Policy.” Economic Research, no.8.
Figure 3: Efficiency Compensation of Consumption