Distortions, Growth Catch-up, and Sustainable Growth
Abstract: A distortion is a departure from the allocation of economic resources from the state in which each agent maximizes his/her own welfare. Distortions can be divided into endogenous distortion (i.e. market imperfections) and policy-imposed distortion. The relationship between distortion and development is complex, thus favorable distortion would only be possible under certain conditions, where, as argued in this paper, four crucial mechanisms may play roles — advantage of backwardness, second-best principle, coordination failure and political economy perspective. Empirically, both international experience and evidence from China suggest that distortions have a positive effect on total factor productivity (TFP) in the early stages of development, but with increasing income levels this role gradually diminishes. Especially in the phases of middle and high income, the negative effects of distortions are significant and become an important factor leading to the middle-income trap. Therefore, reducing and correcting distortions is the key to achieving sustainable growth. Regarding China, it is necessary to eliminate the distortions in a clear way and let the market play the decisive role in resource allocation. Otherwise, in the name of “growth catch-up,” the policy-imposed distortion will occur frequently, and the direction of market-oriented reform will become blurred and swing. Mitigating unfavorable distortions is largely a process of exploring the favorable borderline of government and market, which constitutes a major challenge for all economies.
Keywords: distortions, growth catch-up, government–market relationship, sustainable
JEL Classification Codes: D23; H1; O1; O2 DOI: 1 0.19602/j .chinaeconomist.2018.09.02
The question of whether rapid economic growth is necessarily followed by slowdown and stagnation has been at the center of recent academic debate (Eichengreen et al. 2011; Pritchett and Summers 2014). It can be further divided into some sub-questions, including what are the driving forces of rapid growth, what are the changing roles played by market reforms, factor accumulations (especially investment and labor force) and government behavior in the course of development, and what are the key factors to achieve sustainable development. Undoubtedly, these questions sorely need answering in the case
of present- day China, which is challenged by economic slowdown and socioeconomic structural transformations. This paper attempts to shed some light on this issue by focusing on “distortion,” a factor of both theoretical and practical importance.
From a theoretical perspective, a “distortion” refers to a departure from the perfect competitive equilibrium with no externalities in which resources have been optimally allocated so that each economic agent maximizes his/her own welfare. It is usually caused by market imperfection. In the above sense, distortion is pervasive in economies of all income levels, but their characteristics, extents, types, forms and impacts on socioeconomic development differ substantially across countries.
In particular, judged from their causes, distortions can be divided into two categories. The first is “endogenous distortion,” caused by market imperfection and underdevelopment. The second is “policyimposed distortion,” caused by government policies and interventions. The latter is often founded upon or even on the pretext of the former (Bhagwati 1969).
There are, in theory, two extreme types of distortions. The “night-watchman state” where economy functions under the laissez-faire principle with minimal distortions, and “paternalist state” whose system of welfare covers from “cradle to grave,” which are commonly considered as two extreme types. Most countries, if not all, lie in between them.
Judged from the forms, there are more indirect and highly institutionalized distortions, with the aim of adjusting market failure and of providing public services of positive externality, such as elementary education, environmental protection and basic research. Some are more direct and of administrative color, with governments directly involved in economic constructions and industrial development, usually through state-owned enterprises and selective industrial policies. It should be stressed that direct distortions are often associated with strategies of catching-up. The latter could, however, backfire in practice, especially when national conditions and factors related to the development phase are disregarded.
Judged by impact on economic performance, distortions differ in many aspects across different countries. Generally speaking, at the early stage of development, facing domestic institutional imperfection and heated competition from abroad, intensive distortions, especially selective industrial policies and development strategies, can be used for mobilizing resources, accumulating capital and protecting “infant industries.” Nevertheless, as an economy becomes more mature in terms of both industrial structure and market institutions, distortions’ adverse effects on the economy begin to manifest themselves. In particular, through rent seeking by the government sector, distortions may change the incentives and behavior of private economic agents, thereby hindering innovation and long-term growth. In summary, the question of what the “necessary” distortions are in the context of growth catch-up
1 remains up for discussion. To a large extent, it pertains to the choice of China’s development path in the future. Without a solid theoretical understanding of distortion, there is a risk that reforms will falter or be blindly pursued, thus trapping the country at the middle-income level.
This paper attempts to provide a comprehensive perspective on the relationship between distortions and economic performance, and to discuss the policy implications of distortions for China’s reforms and sustainable development. The rest of the paper is organized as follows. Section 2 reviews the recent literature on distortions. Section 3 discusses the favorable distortions and their applicable conditions. Section 4 provides evidence from international experiences for the impact of distortions on economic performance. Section 5 turns to China’s case, with the emphasis on the marketization process of Chinese
regions. The last section concludes the paper and provides policy discussions.
2. Literature Review
Regarding academic literature, distortion as a research topic was firstly proposed by Bhagwati (1969)，in which he generalized the theory of distortions (mostly in trade) and welfare and proposed a conceptual distinction between “endogenous” distortions and “policy-imposed” distortions.
Since then, distortions have received attention, but mainly on the experiences of developing countries, including taxation, state-owned enterprises, administrative monopoly, trade and industrial policies, financial repression, exchange rate management and government regulations. In particular, economics of development and institutional economics have provided intensive theoretical analysis and policy discussions on economic take-off, and development strategies in developing countries, such as McKinnon (1973)，Sah and Stiglitz (1987), Grossman and Helpman (1994), Parker (1995), Shleifer and Vishny (1998), Hall and Jones (1999), Qian (2000), Gordon and Li (2009) and Acemoglu and Robinson (2012). Relatively speaking, thanks to matured market system, endogenous distortions are less pervasive in advanced economies, where distortions are mainly to do with taxation and government discretionary stabilizing policies as documented in Judd（1985），Chamley（1986），Lucas（1990）and Easterly（1993）.
More interestingly, as China reaches the middle-income level, its economic structures, comparative advantages, sources and path of long-term growth have all been subject to substantial changes. Thus, the relationship between government and market in general, and the mixed impact of distortions on the economy in particular, have been at the center of recent academic discussions, including Parker (1995), Young (2000), Lin (2003), Zhu et al. (2005), Zhang (2008), Zhang (2005), Hsieh & Klenow (2009), Zhang and Cheng (2010), Li and Lin (2011), Song et al. (2011), Yan and Hu (2013) and Qian (2016). The topics of interest include factor price distortions, taxation distortions, financial repression, fragmentation of domestic markets and distortions regarding openness. Recently, Yang (2016) further stressed that mitigating the distortions on factor allocation is the key to the structural reforms in today’s China.
It should be emphasized that although some studies do not directly address the “distortions,” they focus on some questions of high relevance, especially the aforementioned “policy-imposed” distortions. Indeed, the very nature of the distortion should be understood in the context of the government–market relationship and the role of government in development.
From this angle, early literature on structuralist economics of development focuses on market imperfection in developing countries, and claims that the hand of government can help repair the deficiencies of the market. In other words, because of the presence of “endogenous distortions” (including market rigidity, hysteresis, shortage over-supply, inelasticity of supply and demand, and undershooting), government needs to proactively intervene in the functioning of markets through distorting policies and measures, including forced savings, “big-push,” industry selection and import substitution. Despite their positive effects in some areas, the focus on the government interventions brings about various problems, and thus it is replaced by Neoclassical economics of development, which emphasizes the forces of markets and the importance of price mechanism. In fact, the story is more complex in practice than in theory. Although some East Asian economies (such as Japan and South Korea) avoided the “middle income trap” thanks to, to a large degree, the efforts of government, some followers, such as Malaysia and Indonesia, seem to have become trapped at the middle-income level in the aftermath of the Asian Financial Crisis in the late 1990s. Later on, intensive discussions and hot debates emerged around the so-called “Washington Consensus” versus the “Beijing Consensus.” The underlying question of those concerns is once again that of the relationship between government and market. Of course, the topic became even more relevant in the wake of the 2008 crisis and a growing body of literature has emerged since then.
In his theoretical framework labeled New Structural Economics, Lin（2014）stressed that an economy should specialize according to its comparative advantages and this path of development cannot be reached through a laissez-faire market mechanism and thus, the efforts of government are needed. Since comparative advantages are mainly based on factor endowment, including capital, labor force and natural resources, the changes in the latter can also cause the changes in comparative advantages. Indeed, to exploit the comparative advantages, both markets and government are indispensable. In this regard, two cases can be considered: (1) If the government gives priority to the development of the industry of no comparative and competitive advantages, the enterprises in this industry will need protection ( i. e. a distortion) because of their lack of viability; ( 2) If the industry has comparative advantages, the enterprises involved will have high international competitive ability thanks to their viability, thereby improving their capital accumulations. As Lin argued, government should proactively improve the industrial upgrading and technical innovations through “growth identification” and “facilitation.” In essence, the New Structural Economics developed by Lin attempts to strike a balance between government and market. It seems that the former has been overemphasized by early structural development economics and the latter by the Washington Consensus.
In criticizing the traditional theory of comparative advantages, Stiglitz (2017) emphasized the importance of learning. He argued that in the traditional theoretical framework, comparative advantage relies on the assumption that knowledge is publicly available and its focus is on factor endowment (such as the capital–labor ratio). Nevertheless, since capital is mobile across countries, capital endowment seems to be unhelpful for understanding static comparative advantage. Instead, comparative advantage is mainly determined by immobile factors, such as knowledge, labor and institutions. The most important of these is learning ability at the level of society. Given the fact that the market seems to be inefficient in producing and spreading knowledge, governments, especially those in developing countries, need to play a proactive role in improving learning ability.
From a broader perspective, Bardhan ( 2016) addressed the roles of government and their complexity. He argues that because of the comprehensiveness of the goals of development and the multidimensionality of government functions, along with coordination failure and the difficulty in collective action, the role of government needs redefining. From both historical and logical perspectives, Bardhan related stages of development and government’s roles. Jia (2011) and Wen (2016) also stressed the following fact: Government interventions played an important and even indispensable role in the success of the early industrialized countries.
Fukuyama（2012，2015）stressed the importance of “state capacity,” and thus provided an argument in support of strong government. He claimed that the prosperity of a country needs state capacity, accountable government (democracy) and rule of law. According to him, the United States 1956) implies that if it is in, is suffering from weak state capacity, but enjoys accountable government and rule of law; in contrast, China has strong state capacity, but with weak or immature accountable government and rule of law. Besley and Persson（2009）also documented that experiences in advanced economies show that taxation and contract enforcement, as two important elements of state capacity, constitute necessary preconditions for prosperity. Nevertheless, it should be noted that there is no standard definition for the term “state capacity.” Loosely speaking, in our view, it includes market support (property rights delimitation and protection, contract enforcement, market rules), resources allocation (such as taxation) and adjusting social relationships.
Acemoglu et al. (2015) also argued that the state capacity is a crucial explanatory factor for both the Asian Miracle and the mediocre growth performances of many African and South American countries. This claim is further supported by cross-country data (Gennaioli and Rainer 2007) and subnational data
Michalopoulos and Papaioannou, 2013；Bandyopadhyay and Green, 2012). It is noteworthy that the importance of state capacity is by no means unconditional. It should be constrained by accountable government and rule of law. Without those two constraints, a strong government might turn into its
opposite, the Leviathan, or grabbing-hand government. Thus, a strong government should also be a limited government.
Historically, there has been a rise and fall in the government’s role in economic development. In particular, during economic booms, it is not uncommon that their role is either ignored or considered as an impediment to prosperity. In contrast, during crisis their role is often overemphasized. The 2008 crisis is not an exception. Indeed, in the wake of this event, there was a growing body of literature focusing on the (mainly positive) role of government. In this regard, four points should be made: (1) Government, as a last resort, is able to tackle a crisis and mitigate various shocks, through stabilizing policies such as debt-transmission from the private sector to the public sector, especially in advanced economies; ( 2) Government helps with innovation. For example, Mazzucato ( 2013) showed that government in advanced countries actually plays a crucial role in innovation through industrial policies; (3) Government, as a venture capitalist, can also fill the gap between public and private investment; (4) Government can not only make up for market deficiency, but proactively create the market as well. This
2 argument is in sharp contrast with some in Neoclassical economics, and thus has led to some criticism Mingardi 2015).
It should be stressed that government interventions and distortions are not interchangeable. In other words, not all government intervention is seen as distortion. More rigorously, distortions can be defined as government intervention that leads to the deviation from optimal allocation of resources. In this spirit, the “market-augmenting government” of Olson (2000) and the “market-enhancing government” of Aoki (1997) should not be considered as distortions. According to Olson, the success of economic development should be based on two conditions: One is the clear delimitation of rights, another is the protection of the economy from the extortion of public power. The “market-augmenting” government is indeed defined by these conditions, which, in much the same spirit of “state capacity,” imply that the government should play a role in market creation and function. When discussing government in Eastern Asian economies, Aoki and others proposed the concept of “market-enhancing.” He argued that because government failure is not less pervasive than market failure in economic activities, various industrial and social organizations, such as associations of enterprises, trade unions, financial intermediaries, and Chambers of Commerce, can contribute to addressing these two kinds of failure. Thus, government should also foster the development of these organizations and build institutional frameworks to coordinate with them. To judge whether a government intervention is a kind of distortion, therefore, the key criterion is whether this intervention can reinforce the market mechanism.
In sum, like government, distortions play mixed roles in development, which depend on a number of socioeconomic characteristics and institutional factors. Given such complexity, it is very important to investigate, from both theoretical and empirical perspectives, the interaction between distortions and development.
3. How Are Favorable Distortions Possible?
This paper mainly addresses distortions in developing countries, which usually experience two kinds
3 of mutually reinforcing distortions—“endogenous distortions” and “policy-imposed distortions.” In the paper, we pay more attention to the latter, which is believed to be able to contribute to the economic takeoff and catch up.
In light of neoclassical economics, catch up of developing countries is a natural process thanks to
“convergence.” In practice, however, convergence is often conditional and, indeed, the gap between developed and developing countries is in many cases even getting bigger. In some East Asian economies including Japan, South Korea, Singapore and China’s Taiwan, government interventions are sometimes considered to be a contributing factor to their impressive growth performance (World Bank 1993). In addition to these East Asian economies, strong government interventions (such as tariff barriers, exploitation of colonies and overseas markets, and infant industry protection) can be found in many
Western countries during their early industrialization process. Although these government interventions are often distorting in the light of neoclassical economics, their positive impact on economic development may generally surpass their negative impact, and thus they can be considered as “favorable” distortions.
Hence, how are the favorable distortions possible? What are their preconditions? What are the logic and mechanism underlying them? In what follows, we provide four arguments on those questions— advantage of backwardness, second- best principle, coordination failure and political economy perspective.
3.1 Advantage of Backwardness
Gerschenkron (1951) firstly dealt with “economic backwardness” and pointed out that a country, such as the Soviet Union backward relative to Britain when it embarked on industrialization, did not go through the same stages. His theory predicts that the more “economically backward” a country is, the more we will see, for example, more rapid rates of industrial growth and a more active role for the government and large banks in supplying capital and entrepreneurship.
A developing economy can also be at an advantage because of its relative backwardness— in that it can borrow technologies, business models and marketing procedures from more advanced economies; and in that imitation may be easier and faster than original innovation on which the leading economies have to rely. However, when the advantage of backwardness gradually narrows and the uncertainty of frontier innovation increases, the government will find it more difficult to collect enough information and make correct decisions. The “government-picking-the-winner” model turns out to be a failure. That is why, as economies move to higher incomes and rely more on innovation as the engine of growth, the government intervention as a major form of distortions will amplify its adverse effects on the economy. Therefore mitigating distortions and letting the market play a decisive role in resource allocation are the key to sustainable development and therefore also the key to avoiding middle-income trap.
3.2 Second-Best Principle
Another factor of interest is the “second-best principle.” The second-best principle (Lipsey and Lancaster 1956) implies that if it is not feasible to remove a particular market distortion, introducing a second (or more) market distortion(s) may partially counteract the first, and lead to a more efficient outcome. In this regard, the argument put forward by Qian (2016) is of direct relevance to the topic of this paper. As he pointed out, some favorable distortions are possible given the “second-best principle.” Contrary to the “first-best principle,” claiming that a distortion must lead to a decline in efficiency, the “second-best principle” means that in the presence of multiple distortions, the reduction of one distortion
does not necessarily increase efficiency, and the addition of another distortion does not necessarily decrease efficiency.
Arguably, “rent seeking” behavior, which is often caused by government regulations, industrial policies and development strategies, may improve the marketization process in the presence of institutional barriers. Nevertheless, in the early stage of development and industrialization, both entrepreneurship and capital are scarce resources. In this context, investments, learning and innovations may be encouraged by rent seeking behavior—as long as the latter is consistent with the principle of market-oriented reform. Since both government regulations and institutional barriers on the one hand, and rent seeking behavior on the other, can be seen as distortions, thus in this sense, a distortion can be used to tackle another. In recent years, this logic has been discussed in the literature on government behavio（rBardhan, 2016）.
Obviously, China’s reform proceeds in the presence of many distortions, especially the inherent “original sin” distortions (such as underdevelopment of market system and inadequate protection of property rights), and it needs to be “adjusted” by distortion of other kinds, such as active government intervention. For instance, when property rights are secured, the efficiency of private enterprise is generally high, and other forms of ownership of enterprises, due mainly to the principal–agent problem, will cause distortions. But if property rights are insecure because of imperfect rule of law, then privately owned firms will need to pay to secure protection for their property. Therefore, in this context the reform is likely to choose one distortion to deal with another, such as the use of local government power to protect property rights against a higher level of government, which may improve efficiency. In this sense, some distortions may be reasonable and necessary. But despite their favorable effects under some circumstances, distortions as transitional institutional arrangements in line with the second-best principle may incur high costs, and even become significant barriers to further deepening reforms as an economy enters a higher stage of maturity.
3.3. Coordination Failure
Generally speaking, different types of governance should be consistent with different goals of the organization. At different stages of development, an economy may suffer from various kinds of coordination failure, and in this context, government (or the state), market, community and so forth all can play a role as “coordinator” ( either complementary or contradictory) in dealing with this challenge. Clearly, the applicability of the coordination mechanism or arrangement highly depends on the socioeconomic environments and thus, to avoid a simplistic and ahistorical perspective, one should emphasize that there is no universal and optimal model for this.
In theory, the market may be an outstanding coordinator of non-cooperative interactions, inefficient governance and performance incentives. However, if the residual claim and control are misallocated, and the long-run investment decisions have important strategic complementarity, the market as a coordinator ceases to be efficient. In particular, the poor tend to be more affected by imperfections and inadequacy of credit and insurance markets, which discourages productive investments, innovations and human resource development. In this context, non-cooperative interactions tend to be inefficient and therefore coordination and selective incentives by the government are needed to stimulate cooperative actions among market agents.
Market failure in coordination provides the basis for government intervention. In spite of different focus, this argument is in the same spirit as the aforementioned “second best principle.” In some areas and socioeconomic environments (such as underdeveloped markets), the hand of government is indispensable to deal with market failure. Therefore, in addition to the role of night watchman and
protector of property rights, government may play a multiple role as coordinator, guide and stimulator, especially in the context of structural changes and an uncertain development path. As a matter of fact, development and transition countries appear to encounter more challenges from coordination failure and thus government should be associated with proper capacity in mobilizing resources, making decisions, regulating markets, enforcing laws, and collecting and treating information. Ideally, government should remain neutral regarding specific interests, and be subjected to institutional balance and checks to avoid abuse of government power. But although government interventions aim to tackle market failure, they may also be subjected to malfunction and imperfection of their own. For example, although industrial policies regained their popularity in the wake of the 2008 financial crisis, the adverse effects of the government-picking-the-winner model (such as rent seeking behavior) have become more evident. It implies that when intervening in the market, government should tackle both coordination failure and its own deficiency. Obviously, in practice it is extremely hard to do so, especially to succeed in striking a balance between market mechanism and policy distortions. That just shows that favorable distortions, which are not always easy to see, are indeed only possible under certain strict conditions.
3.4 Perspective of Political Economy
Distortion is mostly defined from an economics perspective and thus examined with reference to market equilibrium. If taking the somewhat broader perspective of political economy, which may take account of aspects beyond economics such as national security, geopolitical or ideological factors and so forth, some distortions could be taken as normal or favorable. For instance, the success of the “two bombs, one satellite” program in China before market reforms and opening up is highly praised in the Beijing Consensus, but apparently conflicts with the narrowly defined principle of comparative advantage or cost–benefit analysis. Another example is to sacrifice the interests of farmers to give priority to the development of heavy industry (also from the experience of China). These examples
7 indicate that whether a policy is favorable or not depends on which benchmark or position is chosen.
4. International Evidence 4.1. Descriptive Analysis
Our analysis of international experiences is based on a sample of 50 countries of different income levels. It includes major economies in various continents and at different stages of development. Notably, given the fact that the aggregation of the 50 countries under consideration accounts for 80% and 90% of world population and GDP, respectively, this sample is highly representative. Moreover, this representative sample will also help to tackle the potential bias due to extremely small economies.
Regarding the measure of economic distortions, we rely on the Index of Economic Freedom (IEF for short; overall index), which is jointly published by the Heritage Foundation and the Wall Street Journal. As a commonly used indicator of economic freedom, or, accordingly, an opposite indicator of distortions, the overall IEF is composed of 10 sub-indexes, including “property rights,” “freedom from corruption,” “fiscal freedom,” “government spending,” “business freedom,” “labor freedom,” “monetary freedom,” “trade freedom,” “investment freedom” and “financial freedom.” All the sub-indexes are
8 valued from 0 (least free) to 100 (most free). Thanks to its broad coverage of a number of dimensions of market functioning and economic activities, it is believed that the IEF can serve as a good indicator for the multi-faceted distortions we discussed above.
our conjecture that market distortions will contribute to TFP growth in the early stages of development, and, as the economy expands this positive impact of distortions, will fade away and even change its sign.
5. Evidence from China
This section turns to the experiences of China since the reform and opening up in 1978. Specifically, with the help of panel analysis, we explore the relationship between China’s policy distortions, measured by the Marketization Index, and the province-level (or equivalent region) TFP growth.
5.1. Distortions Measured by Marketization Index
As a commonly used indicator of market development or distortions, the provincial Marketization Index (MI) is compiled by the National Economic Research Institute (NERI). It is a composite measure of China’s marketization process in five dimensions: government–market relations, development of non-state economy, development of product market, development of factors market, and development of intermediary and legal environment. The MI and all of its components are measured on a scale from 0–10. A province scores a higher MI when it stays in a leading position compared with other provinces in its progress towards market economy and suffered fewer distortions compared with lower MI provinces. Here we use some statistical charts to explain the basic facts and characteristics of China’s marketization process in 1997–2014.
First, China’s marketization in general moves forward, while its speed varies between different periods. China’s MI continued to rise during 1997–2014 (the only exception is 2010, slightly down by 1.36%), indicating that China’s market–oriented reform has been effectively pursued. Overall, the growth rate of MI varies widely between different periods. (1) From 1999–2004, the MI increased rapidly, probably due to China’s accession to the World Trade Organization (WTO). (2) From 2005– 2010, the MI continued to rise, but the growth rate gradually slowed. (3) From 2011–2014, growth of the MI picked up and the average annual grow rate was about 5%.
As illustrated in Figure 4, the variation of the MI between provinces expanded after 2008, and its distribution is roughly consistent with the level of regional economic development; the eastern region as a whole had the highest MI, followed by the central region, and then the western region, which had the lowest score.
Furthermore, each MI dimension follows a different trend. Most dimensions of the MI showed an upward trend in most periods (Figures 6). The dimension with a clear downward trend is Government– Market Relations. In 2006–2013, the Government–Market relations dimension fell from 7.18 to 5.70, and rose slightly to 6.02 in 2014. The Government–Market relations dimension consists of three indicators: the proportion of resources allocated by the market, the government intervention in the enterprise and the government size. The three indicators fell by 1.33, 0.77 and 1.53 points, respectively, indicating that since the financial crisis of 2008, the Chinese government has strengthened its power in the resources allocation and expanded its size.
5.2. Provincial TFP Growth in China
As suggested in some research, compared with the developed countries, the contribution of China’s TFP to economic growth has yet to be further improved. For instance, Brandt and Zhu (2010) decomposed China’s economic growth since the reform and opening-up policy in 1978, and found the contribution of TFP to economic growth in China was 62% in 1978–1988, 48% in 1988–1998 and 47%
in 1998–2008, with a gradually declining tendency. In this section, we count TFP for provinces from 1978–2015 by using the Solow growth calculation method. Since the Solow model is the most widely used method for calculating TFP, our improvement is mainly about the estimation of production factor inputs. Based on TFP growth, we can identify the main driving force of China’s economic growth and judge whether the Chinese economy has entered the efficiency-driven stage.
5.2.1 Equation of TFP calculation
The production function is characterized as a Cobb–Douglas production function:
A is TFP, α is the capital output elasticity, (1–α) is the human capital output elasticity. Deriving the derivative of time t, equation (1) changes to
(4) Furthermore, we use the Hodrick– Prescott ( HP) filtering method to remove the stochastic perturbation factor of , and finally estimate the growth rate of the TFP .
5.2.2 Input accounting
(1) Physical capital
The widely used method for estimating the physical capital stock is the perpetual inventory method initiated by Goldsmith (1951): . To use this method, four parameters need to be determined: the capital stock in the base period K, investment in every period I, capital price index i and depreciation rate δ. Firstly, learning from Young (2000), we set 1952 as the base period, and multiply the investment in fixed assets in 1952 by 10 to obtain the capital stock for that year. Secondly, we use the capital formation in every year as the investment increment. Thirdly, for the capital price index, in 1952–1990, we use the retail price index as a proxy for it; and in 1991–2015, we use the price deflator of fixed asset investment to deflate the value of capital stock. Fourthly, the depreciation rate from 1952 to 1977 was calculated to be 5%, and it increased by 0.1% from 1978. This approach is adopted from Wang et al. (2009).
(2) Human capital
According to the Solow model, the human capital could be defined as the product of the number of workers (L) and the average number of years of education (H) of labor. The number of labor (L) could be obtained from the China Statistical Yearbook. The average education year (H) could be calculated based on the census data of 1982, 1990, 2000, 2005 and 2010.
(3) Capital output elasticity
Chen et al. (1988), Chow and Li (2002) and Lu and Cai (2016) used the aggregate production function to estimate the capital output elasticity. Bai and Zhang (2015) argued that the share of capital income from the aggregate production function is constant, but this is inconsistent with China’s reality. Therefore, we use the income approach to calculate the capital income share of each province as a capital output elasticity. The calculation formula is as follows:
(4) Statistics of TFP
Following the above method, we estimate China’s provincial TFP growth. The statistics for the variables are as follows:
Figure 7 presents the TFP in the three regions13 of China. In 1978–2015, TFP steadily increased in each region but at a different pace: TFP in the eastern region is significantly faster than in the central and western regions. The gap of TFP between central and western gradually increased.
TFP growth of each region is shown in Figure 8. In most years from 1978 to 2015, China’s eastern, central and western regions saw positive TFP growth, but the growth rate fluctuated greatly in different years. In the 1980s and 1990s, TFP growth showed a tendency to firstly increase and then decrease. From 2000, the TFP growth trend stabilized, and after 2010 it significantly declined.
5.3. Empirical Model
In this section, we establish a panel regression model to further investigate the long- term 14 quantitative contributions of marketization to total provincial TFP.
We construct the following empirical regression model.
(6) The subscripts i and t represent the provinces and years. The explanatory variable TFP of the
regression model is set as the TFP growth rate of the province i in year t. The core explanatory variable MIi, t is the Marketization Index ( MI) of province i in year t. The fixed effect eliminates all the variations among provinces that do not change over time, while the time-fixed effect removes all timelevel disturbances that do not vary with the province. represents the residual term of the regression model.
We focus on the estimated coefficients of the independent variable MIi, t. The economic connotation of the coefficient is interpreted as the average effect of the MI on the dependent variable, that is, the influence of the MI on the TFP growth rate.
Other control variables of the regression include: (1) GDP per capita, which is used to control the levels of economic development. (2) Government size, which is defined as government expenditure as a proportion of GDP. (3) Openness, which is defined as the export share of GDP. (4) Investment rate, which is defined as the share of capital formation in GDP.
Regarding sample grouping, most of the provinces in China were in the upper-middle-income group and the lower-middle-income group from 1997 to 2014.15 Therefore, the subsample regressions are only for the lower-middle-income and the upper-middle-income group. The remaining empirical strategies are consistent with the international evidence section.
5.4 Regression Results
Table 4 demonstrates the regression results of equation (6). Columns (1)–(4) of Table 4 show the regression results for fixed effect using the full sample, and columns (5)–(6) show the subsample regression results. In columns (1)–(3), the MI has no significant effect on TFP growth. In column (4), both of MI's primary and secondary terms are significant, which confirms the nonlinear relationship between distortion and economic growth: that is, the impact of MI on TFP growth first declines and then increases. The inflection point is 8.38, which corresponds to the market-wide level of the whole country in 2013–2014. Columns (5) and (6) show that the MI has a significant effect on the growth rate of TFP in the upper-middle-income group, but not in the lower-middle-income group. For the upper-middleincome group, MI increases by 1 point and the growth of TFP increases by 0.4% on average.
For other control variables, GDP per capita ( GDPPC) has a negative effect on TFP growth, indicating that provinces with a high development level might lose the late-mover advantage and have more difficulty in promoting TFP growth. Trade openness ( TradeOpen) has significantly positive coefficients, which is in line with the international evidence. The investment rate ( InvRatio) has negative effects on TFP growth in every column, and it has a greater negative effect on the lower-middle-income group. This indicates that there are large numbers of inefficient investments in China, and the growth driven by low-efficiency investments is not conducive to TFP growth, confirming the views of Woo (1998), and Bai and Zhang (2015).
The evidence from China is generally consistent with international experience. But it also has its own characteristics. Both the evidence from China and international experience confirm the non- linear relationship of distortion for TFP growth. With the development of the economy, the promotion of marketization on the growth of TFP will first decline and then rise, that is, the effect of distortion on the growth of TFP will rise first and then decline. In the international experience, however, the suppression effect of distortions occurs in the high-income group; while in China, the suppression effect of distortions occurs in the upper-income group. Since China has too few samples in high-income groups, we cannot obtain the regression results of high-income groups. However, combining international experience with the Chinese experience, we can see that as more provinces enter the upper-middle-income and high-income stage, the negative effects of distortions on TFP growth will
further manifest themselves.
5.5 Robustness Analysis
To check the robustness of the previous findings, we use the sub-indexes of MI as alternative measures for policy distortions. To investigate the non-linear relationship between distortions and TFP growth, we also divide the sample period into two sub-samples: lower-middle-income group and highermiddle-income group. Other control variables remain unchanged. The results are shown in Table 5.16
The Government–Market Relations (MI1) have a significant negative effect on TFP growth in the lower-middle-income group, and a significantly positive effect in the upper-middle-income group.
The Development of Non-State Economic (MI2) promoted TFP growth in the lower-middle-income group, but not in the upper-middle-income group. The reason may be that after undergoing a largescale reform of state-owned enterprises, the Chinese economy has formed a vertical structure, of which upstream industries (such as energy, finance and telecommunications) are still occupied by state-owned enterprises, and most downstream industries (such as consumer goods, manufacturing, and consumer services such as hotel and entertainment industries) have been dominated by private economy (Li et al. 2014). With China's accession to the WTO, both downstream industry and upstream industry have greatly improved. Therefore, in the upper-middle-income group, the state-owned economy and the nonstate-owned economy are equally important for improving TFP growth. The relative share of the nonstate-owned economy in the overall economy has no significant effect on TFP growth.
The Development of Factors Market (MI4) has no significant effect on TFP growth in the lowermiddle-income group, and a significant positive effect on TFP in the upper-middle-income group. The report of the 19th National Congress of the CPC pointed out that economic system reform must focus on improving the property rights system and the market-oriented allocation of factors. Further improving the allocation efficiency of factor market is an important guarantee for the sustainable development of the economy.
Finally, the Development of Intermediary and Legal Environment (MI5) has a significant positive impact on TFP growth in both the lower-middle-income group and the upper-middle-income group. The regression results for other variables are consistent with those in Table 4.
6. Concluding Remarks and Policy Implications
Our empirical results based on international experiences show that distortions can hinder the engine of long-term growth, measured by the TFP, in high-income economies. By contrast, distortions may help middle-income economies to achieve faster TFP growth under certain conditions.
Empirical study of China’s provinces generally shows that marketization significantly promoted the growth of the country’s TFP during the period 1997–2014. That is to say, distortions have significant negative effects on the country’s TFP growth during this period. However, in the early stage of development (1978–1991), distortions played a significantly positive role in the TFP growth.
Overall, despite some slight differences, the main findings from international and Chinese experience seem basically consistent, so that we can conclude that in the early stage of development, appropriate distortions can help break through the poverty trap. But as the economy climbs to a higher income level, the adverse effects of distortions become more and more significant and eventually hinder sustainable growth.
Our findings have important implications for China’s policy making. First, the transition from middle-income to high-income economy is also a process of mitigating market distortions. Although the latter might have contributed to the country’s economic takeoff and growth catchup in the past, serious problems due to distorting market systems have been created and become substantial impediments to reform and development in the future. In view of this, to avoid the so-called middle-income trap and to achieve sustainable development, the key is to promote market-oriented reforms and, in particular,
17 reduce inappropriate government interventions and institutional distortions.
Second, it by no means implies that the economy can function well without government. From a
theoretical standpoint, government services and distortions differ in many important ways. In particular, an efficient government with appropriate “state capacity” is indispensable in a modern market-based economy. As a matter of fact, it is not uncommon that in advanced economies with mature market systems, government also plays a proactive role in many socio-economic dimensions, especially in encouraging innovation. In other words, the distortions per se can be viewed as the situation in which government does not assume its role properly. What should be done to let the government play a “better” role in a market economy context as stated in the Third Plenary Session of the 18th CPC Central Committee? In our view, instead of “replacing” market mechanism, government should play a proactive role in the “market-reinforcing” process. In other words, various government interventions should only
be judged by their contributions to market-oriented reforms, especially to helping the market play a decisive role in resource allocation.
Third, with clear objectives of economic reforms, “favorable distortions” should be subjected to strict constraints. Although “favorable distortion” is indeed often taken as a supportive argument for keeping distortion and the important role of government in the economy, the relationship between distortion and economic performance is complex, especially nonlinear. In particular, favorable distortion substantially depends on various socioeconomic conditions related to stage of development. If these conditions are not met, distortions will hamper development. For instance, the advantage of backwardness, second-best principle, and coordination failure all have their theoretical preconditions, and, generally speaking, are appropriate for developing countries which have immature market systems, serious structural problems and challenges regarding “takeoff” and “transition”. Since those countries including China enter into a higher level of development with an improved institutional environment, the so-called favorable distortions will become less plausible. For this reason, China’s policy makers should take a clear stand to reduce distortions, and let the market play a decisive role in resource allocation. Otherwise, in the name of “growth catch-up,” the policy-imposed distortion will occur frequently and the direction of market-oriented reform will become blurred and swing. Mitigating unfavorable distortions is largely a process of exploring the favorable borderline between government and market, which constitutes a major challenge for all economies.
TFP = total factor productivity.Notes: 1. Standard errors in parentheses, with ***, **, * denoting 1%, 5%, and 10% level of significance, respectively. 2. All regressions above include a constant.Source: The authors. Table 5: TFP Growth and the State-Owned Economy (1997–2014)