GRIN - Income inequality in China. Development and underlying drivers (2023)

table of contents

List of figures

table list

list of abbreviations

Section 1 Introduction

SECTION 2: Theoretical basis
2.1 Definitions
2.1.1 Measuring income inequality
2.1.2 Limitations of the Gini coefficient
2.1.3 Measurement of economic growth
2.2 Data
2.3 Theory of economic growth and income distribution

SECTION 3: Literature review on economic growth and income inequality
3.1 Income inequality: a necessary evil for economic growth?
3.2 The impact of income inequality on economic growth
3.3 Long-term implications of income inequality on economic growth
3.4 Relationship between income inequality and economic growth in China

SECTION 4: Income inequality in China
4.1 Extent of income inequality in China
4.2 Income inequality and insufficient consumption
4.2 Overview of Chinese government policy measures

SECTION 5: Drivers of Income Inequality in China
5.1 Economic Opening Process
5.2 The urban-rural income gap
5.3 Educational inequality
5.4 Regional inequality
5.5 Wealth inequality
5.6 Demographic change
5.7 Chinese tax system

SECTION 6: Policy Recommendations
6.1 Tax Reforms
6.2 Tax Policies
6.3 Dibon

SECTION 7: Conclusion

REFERENCES

ANNEXES

List of figures

Figure 1: Lorenz curve

Figure 2: Consumer spending as a function of disposable income

Figure 3: Income inequality in China, 1978 to 2017

Figure 4: Share of national income before taxes by income group, 1978 to 2015

Figure 5: Gini for selected developing countries, 2011

Figure 6: Lorenz curve for China, 2013 and 2017

Figure 7: Urbanization rate in China, 1978 to 2018

Figure 8: Urban-rural disposable income ratio, 1978 to 2017

Figure 9: Breakdown of urban and rural family income, 2011

Figure 10: Urban-rural disposable income ratio by region in China, 2013 to 2017

Figure 11: Household income structure in East, Central and West China, 2011

Figure 12: Wealth shares of different income percentage groups, 1994 to 2015

Figure 13: Sources of tax revenue for Germany, OECD and China in 2016

Figure 14: Gini coefficients for selected developing countries, 2016

table list

Table 1: Household savings rate, 2011 (in RMB)

Table 2: Proportion of Chinese households that saved, 2011

Table 3: Proportion of Chinese households saving by income group, 2011

Table 4: Urban households that saved by asset group and income, 2011

Table 5: Breakdown of rural and urban household income, 2011 (in RMB)

list of abbreviations

Illustration not included in this extract

Section 1 Introduction

Income inequality is a worrying issue that has persisted in economic debates for a long time in history, with different researchers from various fields exploring this topic with different interpretations and conceptualizations, from ancient Greek philosophers to modern philosophers, economists and politicians. inequality was extensively explored. According to the International Monetary Fund, income inequality has become one of the most devastating problems facing countries today, as the gap between rich and poor is experienced at significantly high rates in both developing and advanced economies (Dabla-Norris, Kochhar, Suphaphiphat, Ricka and Tsounta, 2015). This does not exclude developed countries, as the World Economic Forum recently observed a global trend of income inequality, as income distribution has worsened in 17 of the 22 OECD member countries since 1980 (Schwab, 2018).

Since adopting an open policy and implementing sweeping economic reforms in 1978, China has experienced rapid economic growth. The economic growth rate has increased uninterruptedly and the country has competed significantly with other countries on the international stage, consolidating itself as one of the emerging economies that have shown the greatest growth in recent decades. The country's economic growth between 2000 and 2010 averaged 9.15%, a figure that made China the second largest economy in the world (NBS, 2012). With this growth, people across the country are expected to enjoy the fruits. However, as in other economies, China's economic growth has been accompanied by large increases in income inequality. The country's income inequality in the 1980s was among the most equitable, a trend that has reversed over the past three decades, and income inequality has grown to become one of the highest in the world (Xie & Zhou, 2014). Yang (1999) even coined China as the country that experienced the fastest increase in income equality for which comparable data are available since 1978.

While some countries like the United States are highly developed in many ways, they also suffer from high levels of income inequality. For example, Germany was also heavily criticized by the IMF in its latest State of the Country report for its higher income and, in particular, its wealth inequality. The latter is now considered one of the fastest growing in the developed world (Dao, Perry, Klemm, & Hebous, 2019). As we are looking at rich countries, the question arises as to whether income inequality is in fact the reason for economic growth or a necessary result of it.

In this context, some might argue that a highly egalitarian society would lack the necessary incentives to stimulate economic growth; Remember: why should anyone invest in education and training if the benefits in terms of salary are low? Furthermore, rising income inequality may be consistent with welfare arguments as long as all income groups experience even small positive income growth. This has been the case in China since 1978, where all income groups experienced large increases in income, although larger for higher income groups, which eventually led to an increase in income inequality.(NBS, 2018)🇧🇷 As a result, absolute poverty, as measured by the poverty headcount rate, has been almost completely eradicated (Sicular, Li, Yue, and Sato, 2017). Consequently, measured by increased income and consumption, this could be described as Pareto efficient. This is certainly an area of ​​tension between academics with those who advocate income inequality who argue, among others, that income disparities are essential drivers of economic growth.

However, following the above argument, even if the lowest income groups recorded positive income growth, it would still mean that they became relatively poorer compared to the rest of society, so it is likely that they recorded participation losses. and opportunities. This circumstance may have several implications and indirect effects, such as in terms of health or crime, which may call into question the consistency of Pareto, as we will see in this thesis. Most importantly, high rates of inflation, often accompanied by high economic growth, can quickly wipe out these income gains if real wage increases are not accounted for, although inflation is a national average parameter, which does not necessarily capture price increases. for similar low-income families.

As Sampson (2016) argues, inequality within a society can be accompanied by a lack of upward mobility, stagnant wages and a hollowing out of the middle class, resulting in poorly distributed concentration of quality education, violence and poverty, all of which have a national impact. impact on general well-being and the economy. Furthermore, there are ethical and philosophical reasons for resentment of inequality per se, as there should be no differential treatment in access to economic resources. However, some might argue that people are responsible for the results of their actions and choices they've made throughout their lives. In some circumstances this may be correct, although in most cases the unfair treatment starts from the day they are born. Two children can have two completely different opportunities in life just based on the family they were raised in. Ravallion (2016) refers to this as the level of inequality of opportunity, which he considers to be the most crucial factor in assessing the negative impact of income inequality. on economic growth. Therefore, calls for equitable income distribution also raise questions about the extent to which income distributions can be described as equitable (Li and Gibson, 2013). Therefore, there is a contentious conceptual issue as to whether there is a trade-off between equality and efficiency (Wang, Wan, and Yang, 2014). Therefore, this article will shed light on the various characteristics and sources of income inequality in China and demonstrate their interrelationships in economic growth.

The Gini coefficient is used to determine the level of income inequality. In China, the Gini coefficient went from just 0.27 in 1980 to its peak of 0.49 in 2009, while a modest decline or stabilization can be observed, respectively, until the most recent estimate in 2017 (Chen, Pu and Hou , 2018; NBS, 2018). As much as there is controversy over these estimates by the official Chinese statistics agency and the fact that various sources lead to different results, there is no doubt that income inequality in the country has reached large proportions and has become a cause for concern. Inequality was widely noted among Chinese citizens, with most of them acknowledging that it has affected their lives. According to a 2012 national survey, the Chinese identified economic inequality as the most critical social problem they faced compared to other social problems such as unemployment and corruption (Wu, et al., 2013).

Historically, income inequality in China has been mainly driven by two main factors, including the access and extent of public resources and the emergence of market forces after major opening-up reforms (Li, Sato, and Sicular, 2013). From a conceptual point of view, both factors have the capacity to offset or reinforce the impact of the other. In the course, several dimensions of the drivers of inequality were developed, which manifest themselves in extensive urban-rural inequality, inequality between regions, inequality in education, inequality in wealth or demographic changes, among others.

This article aims to establish the current degree of income inequality in China, while identifying various forces and factors behind these changes since the economic opening process in 1978/79. This article also aims to explore the relationship between income inequality in China and the country's economic growth, as well as build on these findings by proposing appropriate policy recommendations.

Demonstrating the importance of studying the growing income inequality in China, the author argues that the growth of the Chinese economy may be affected by further increases in income inequality in the future. In this case, the author will base his argument on the view of Murphy, Shleifer and Vishny (1989) who explore the implications of greater income inequality on economic growth from the perspective of domestic demand. In this sense, the document will seek to demonstrate that, considering the combination between the perspective of domestic demand and China's unique economic structure, a further increase in income inequality could reduce the country's economic growth.

This thesis begins by providing an understanding of the definitions used in this article, presenting the data sources used, their underlying limitations and providing a theoretical basis on what the theory of economic growth suggests in the context of income inequality. To assess the relationship between income inequality and economic growth, this study uses a literature review approach starting in Section 3 to explore the mechanisms behind income distribution and economic growth first in a broad context and then specifically in the case of China . The literature review conclusions will be complemented by Chinese national data on consumption and saving patterns in Section 4.2. Subsequently, sections 4 and 5 will provide an important perspective on the development of income inequality since the economic opening reforms were adopted in 1978, thereby identifying the trends and sources that have relentlessly driven income disparities across society. Finally, Section 6 will provide policy recommendations based on the results set out above.

This thesis addresses the following research questions:

1. What is the relationship between economic growth and income inequality in general and in relation to China? What are the likely implications of higher levels of income inequality on economic growth? (Section 3 and 4.2)
2. How has income inequality evolved over time? How have Chinese lawmakers addressed this issue? (Section 4)
3. What are the drivers and reasons for changes in income inequality in China over the past two decades to the present? (Section 5)
4. What policy recommendations can be made to address this issue? (Section 6)

SECTION 2: Theoretical basis

2.1 Definitions

2.1.1 Measuring income inequality

Max Lorenz (1905) developed a theory that explains income distribution. The theory compared income distribution among households in a population over a period of time. According to Lorenz, perfect equality refers to a case where all families within society receive the same income.,while a perfect inequality represents a case where a single entity receives all income while the rest of the population receives nothing. Since both perfect equality and perfect inequality cannot be updated, the Lorenz curve presents a structure based on which the distribution of income within a given population can be understood.(Lorenzo, 1905)🇧🇷 Therefore, the curve is assumed to always lie below the line of perfect equality.(Clark, 1992).

The Gini coefficient was developed based on the Lorenz curve. The Gini refers to the ratio of the area between the line of perfect equality and the Lorenz curve. Therefore, as illustrated in Figure 1, from the areas indicated as B and the area C, the Gini can be calculated by the ratio of B to (B + C), which must be a value between 0 and 1. A value less than The Gini coefficient is indicative of a relatively equal distribution of income, while a higher value would suggest a comparatively unequal distribution.

The Gini measure will henceforth be used to indicate the extent of net disposable income inequality within certain groups of households, along with personal or size distribution of income. In addition to the Gini and Lorenz curve, there are other concepts to measure income inequality, such as the Theil index or the decile proportions, which will not be specifically addressed in this thesis (World Bank, n.d.).

Figure 1: Lorenz curve

Illustration not included in this extract

Source: authors' description

There are two main concepts that describe income inequality: through the distribution of personal income or size and the functional distribution of income. While the second focuses on the functional sources of income of an economy as a whole, the first groups individuals or families according to their income share within society observed in percentiles (Piketty, 2014). Both concepts will be of great importance for the next sections.

Most developing countries have Gini coefficients ranging from 0.35 to 0.50, while most developed countries range from 0.22 to 0.32, showing that developing countries tend to have a more unequal income distribution than their developed counterparts. Why this happens will be elaborated throughout this thesis.

Definition of Income

Income that can be understood as the potential dominance over consumption. It is used as the primary variable when measuring inequality rather than real consumption. In this context, disposable income, i.e. including after-tax deductions and transfers, is applied to measure income inequality, unless otherwise stated, rather than market income, which can be defined as pre-tax income and any redistributions. In some cases, disposable income can include or exclude rent or cash income, making comparability difficult, and it can also be used by governments to mislead the public about progress by changing definitions accordingly.

However, it should be taken into account that income is unable to accurately reflect the access of the respective individuals to an acceptable standard of living, which can be affected by various other factors such as insecurity, health, legal rights or education. (Lipton and Ravallion, 1995). This assumption will be important for the impact of income inequality on economic growth, as we shall see.

2.1.2 Limitations of the Gini coefficient

The Gini index as a measure to compare income inequality across countries carries several caveats. First, income can be calculated based on household income level or individual level, so income can also be defined differently. Considering ordinary disposable income, for example, does not capture the potential for low-income households in education, health care, or housing subsidies, or in agriculture-focused subsistence economies, income can be received by non-monetary means that also distort the overall picture. Therefore, differences in the methodology for calculating the Gini may limit cross-country comparisons on this measure (Chitiga, Owusu-Sekyere, and Tsoanamatsie, 2014). It is worth noting that this limitation is clearly reflected in the large variations that exist in income inequality estimates for China.¹(World Bank, 2019; Solt, 2019).

Second, informal sector income is often ignored when comparing income levels of different income percentiles. Particularly for developing countries, this gray sector constitutes a sizable, if not predominant, part of total economic output, further limiting the validity of the Gini measure as used today (Osberg, 2016). This aspect is obviously of great interest in relation to China as well and often leads to upward corrections in official data.

Furthermore, the Gini as a relative measure cannot reflect absolute improvements in income, in the sense that even for countries where all income groups have experienced an increase in income and thus reduced extreme poverty, the Gini may still rise. Consequently, this would be a violation of the Pareto principle of improvement. Likewise, in a scenario where all income groups in a given society experience income declines, the Gini coefficient could indicate an improvement in the level of income inequality (Osberg, 2016).

Finally, in addition, demographic differences are not specifically captured by the Gini, so countries with a higher proportion of students or retirees whose main source of income is pensions²they are likely to have higher Gini coefficients (Chitiga, Owusu-Sekyere, & Tsoanamatsie, 2014). The former is of particular relevance when considering China's income inequality, given the consequences of low fertility rates on China's demography.

However, the Gini coefficient together with the Lorenz Curve are the most commonly accepted tools that represent income inequality among academics in the field (World Bank, n.d.). Therefore, these are used for this study taking into account the limitations mentioned above.

2.1.3 Measurement of economic growth

The most common means of accounting for economic growth is the actual change in total production of goods and services within a country called GDP. Additional measures include per capita income and per capita consumption (Cho, Kim, and Rhee, 2014). The rate of economic growth can be influenced by microeconomic factors, such as human and natural resources, technological and capital resources, as well as by macroeconomic factors, such as the level of inequality, education, employment and population growth. the state of health, as well as the institutional framework of an economy (Barro, 2000). This study used gross domestic product (GDP) per capita as a measure of economic growth, to investigate the relationship between economic growth and income inequality with other macroeconomic factors such as education, employment and health spending as control variables. As GDP depends on pure and unbounded variables inside and outside a given economy, measurement errors are very likely to occur, for example through omitted variable bias.

While one measures economic growth, many actually claim to measure social progress or quality of life. This is an interesting field of study, although not the object of this thesis. For the sake of simplicity, the author assumes the positive correlation between economic growth and positive social progress. In addition to the widespread doubt whether GDP reflects development, there are several others. For example, the fact of not capturing environmental damage and resource depletion, among others, exemplifies the lack of sustainability that the production measure cannot express (Stiglitz, Sen and Fitoussi, 2010). Also, the fact that the value of unpaid work such as home or home renovations is not accounted for in GDP, which is considered a substantial part of the economy.

2.2 Data

Most data on income inequality was obtained from data collected by Chen, Pu and Hou (2018) -also in the SWIID database-, the NBS Statistical Yearbooks (2011-2018), the World Income Inequality Database (WID ), as well as the China Household Finance Survey with survey data from 2010 and 2011. In particular, additional data were obtained from the World Bank's Gini index database and OECD statistics. Smaller data was used in this regard from the CIA World Book and other authors.

Note that as the China Household Finance Survey (CHFS) is considered to be a detailed and unique dataset collected in China, the author has decided to exploit the results of this survey, despite its relatively old study period, as one of the important sources investigating the China. current inequality - with more information given below. A popular alternative would be the survey results provided by the China Household Income Project (CHIP), which are freely available after registration is confirmed.

Raw data for graphs and illustrated figures will be consistently provided in the Appendix, i.e. Appendices A to L.

China Household Finance Survey (CHFS) with 2011 survey data compiled by Gan et al. (2014)

The CHFS was first launched in 2010 by the Chengdu-based South Western University of Finance and Economics and is considered unique in its breadth in providing detailed information on Chinese household income, expenditure, assets, debt and wealth levels. (Gustafsson, Li and Sato, 2014). Another advantage of this data source is seen in the supplemental quarterly interviews on income, employment and expenditure attributed every two years from the main survey that provide regular updates (Gustafsson, Li, and Sato, 2014). The survey covered 29,500 people from 8,400 households.

As CHFS was first launched in 2010 and did not regularly provide open data access for its findings, comparability over time is limited. The survey data from 2013 turned out not to be as broad compared to previous publications, so this thesis mainly used the sample year 2011. Also, data from the most recent survey conducted in 2017 is not available. Free for external researchers. from SWUFE University. Given the perhaps outdated data, as well as possible underlying survey biases, this is a wake-up call when looking at the current dynamics of income inequality in China.

National Bureau of Statistics of China (NBS) - Statistical Yearbooks of China

The NBS began publishing adequate estimates of the Gini coefficients in 2002 through its China Statistical Yearbooks in per capita terms. These estimates are based on surveys of urban and rural residents. Data are supplemented with personal income records to correct for bias (Jain-Chandra, et al., 2018).

Weaknesses include the fact that multiple changes in its definition of income and population mean that the datasets are not consistently comparable over time (Chen, Pu, & Hou, 2018). Importantly, income did not include remuneration and benefits in kind, such as housing, medical care, pension and unemployment, or even bonus payments from the employer (Gustafsson, Li and Sato, 2014). According to Chen, Pu and Hou (2010), this would lead to an underestimation of the urban-rural income gap, since, they claim, urban residents receive much more public benefits than rural families. Furthermore, urban and rural households were previously defined based on their family register hukou classification (note that a proper definition of hukou will be given later). The authors emphasize that this particularly overlooks rural migrants working in cities, leading to various statistical distortions. However, there have been significant statistical changes in China since 2013 that have corrected many, though not all, warnings. As a side effect, this made comparisons over time (even) less accurate (Gustafsson, Li, & Sato, 2014).

Despite their limitations, data from the NBS and its China Statistical Yearbooks can be considered as the most important data source for calculating the Chinese Gini, especially for years for which other household survey data are not available. national level (Chen, Dai, Pu, Hou & Feng, 2010).

Chen, Pu e Hou (2018)

Five Chinese professors and researchers point to several caveats in their literature review of official income data widely used to calculate the Gini. They identify the data provided by the NBS as underestimating, as intragroup income inequality is not reflected in the pooled data samples they commonly claim (Chen, Pu, and Hou, 2018). Therefore, the urban-rural income ratio must be considered very low. No raw data provided prior to 2006, except data for limited income groups. Hence his motivation to present adjusted Gini numbers for the years that have elapsed since the economic liberalization reforms initiated in 1978.

Combining survey data from the NBS Department of Rural and Social Economy and the Urban Socioeconomic Survey, the first 7,500 different Gini coefficients were calculated and combined to be able to estimate Gini coefficients in per capita terms. 🇧🇷 🇧🇷 These estimates are listed in SWIID.

General limitations of data used

General reservations from data sources include the very universal problem of income underreporting. This is especially true for high-income families who generally tend to have better access to the informal sector or bribes.

Furthermore, there is a wide range of different estimates in the case of China (eg World Bank, CHFS, NBS, OECD, Chen, Pu and Hou, Ravallion report different estimates for a given year). This will certainly increase the chance of misinterpretation of numbers and will likely limit the comparability of data coming from different sources.

2.3 Theory of economic growth and income distribution

According to the classic economic growth models of Harrod, Domar and Solow, for GDP to increase, investment must exceed the amount needed to compensate for depreciated capital. Therefore, the level of saving and investment plays a crucial role in understanding economic growth (Gallo, 2002). An increase in the saving rate, holding all other variables constant, would lead to economic growth. This is one of the conclusions that can be drawn based on the initial Harrod-Domar model (Sampson, 2016). However, apparently these additional variables are changing and not constant. This is the case, for example, with the rate of capital production or population growth. Here, it is assumed that the variables are exogenous to economic growth. Although it is not so simplistic, as not only do growth processes influence the evolution of all these variables, but also the central saving rate can be affected by per capita income, cultural factors and, in fact, how income is distributed among different strata. of the population (Gallo, 2002). If the benefits of growth are unequally distributed among the highest shares of upper-tail income groups, the overall saving rate is expected to increase, given the higher expected propensity to save of higher-income groups. However, if most of the income growth is attributed to the group of individuals with a lower propensity to save and, therefore, a higher rate of consumption, the general saving rate is expected to fall as a consequence of growth (Rooster, 2002) . 🇧🇷 🇧🇷 This could harm subsequent growth, as some scholars argue in the context of economic growth and income distribution.

In stark contrast to the Harrod-Domar model, which assumes constant returns, the Solow model predicts that the saving rate has no long-term effect on the growth rate. As capital accumulation is not capable of stabilizing per capita growth, the effect is only short-term. As the final sustained source of economic growth, scholars later incorporated the exogenous technology variable, where technological progress increases labor productivity (Ranis, 2004).

As the technological factor plausibly seems quite endogenous, the so-called New Theories of Growth presented the concept of human capital that is integrated into the Solow Model with savings divided into human capital and physical capital. This enables constant returns on capital, even if physical capital shows diminishing returns. Similar to the initial Harrod-Domar model, it brings the savings rate back into the picture to explain long-term growth (Gallo, 2002). The most recent literature, from the 1990s, puts a relationship into play through political and economic factors. Economic factors predict a relationship between income distribution and the level of domestic demand, as well as the existence of imperfect credit markets. Likewise, Galor and Zeira (1993) establish that initial income inequality can lead to different investment options in human capital in the presence of imperfect loan markets. This, in turn, affects the level of investment and GDP growth. Countries with high income inequality will have fewer households with the financial means to invest in human capital compared to their more egalitarian counterparts (Galor and Zeira, 1993). Alesina and Perotti (1993), on the other hand, emphasize the importance of political factors based on their findings that income inequality can fuel sociopolitical unrest in the country, which in turn negatively affects investment. Therefore, they also establish the link with GDP growth by observing the level of investment as an instrumental variable (Alesina & Perotti, 1993).

However, to predict the influence of saving and investment, the variables cannot be understood as simple aggregate levels. It seems plausible to assume that certain sectors of the economy require more investment than others, thus contributing more to the growth process than others. Models based on the dual economy consider the manufacturing sector at the forefront of generating growth, considering inequality as a prerequisite for economic growth. These models are being presented by some scholars when they defend greater income inequality for developing countries, the Lewis model being perhaps one of the most influential. In it, Lewis (1954) builds a theory where a traditional sector (ie agricultural in rural areas) and a modern one (ie industrial based in urban areas) coexist. Initially, there is an abundance of labor in the agricultural sector and therefore low real wages. Therefore, industrialization can easily be fueled by an unlimited supply of cheap labor. The economy then finally reaches the so-called Lewis tipping point, once the surplus of rural labor is exhausted (Lewis, 1954). As a consequence, real wages in the unskilled industrial sector show strong gains, which in turn allows for further growth until a labor surplus is again achieved. The underlying basic assumption is that an economy's abundant labor supply up to the Lewis tipping point and recovers thereafter, provided balanced growth policies are adopted by the state. Consequently, the Lewis Model assumes, due to the existence of two very different sectors in the economy, initial unequal distribution of income that worsens the stronger the modern or industrial sector grows until the Lewis tipping point is reached and wages of skilled and unskilled workers. reduction in labor due to the initial labor shortage incurred, which in turn allows for major improvements in terms of income inequality.

While the Lewis model has been praised for its role in understanding economic transformation processes in developing countries, it has also received a number of criticisms, including, among others, the fact that an economy eventually suffers from food shortages, resulting in economic turmoil or that work cannot be infinitely abundant (Piazza, 2014). On the other hand, Ranis (2004) puts in perspective that an open market condition would make food imports more than plausible. Ranis (2004) concludes that although the Lewis model has lost great importance for economists when researching issues in developed countries, it is still widely used in parts of the developing world to explain economic growth and labor dynamics, which we will see throughout this article. this thesis in the case of China.

However, these classical views of economic development contrast with the alternative neoclassical models that evolved in the 1960s (Ge & Yang, 2010). To take as an example the work of Schultz (1964), who considers that wages are determined by market forces and not fixed from above. Furthermore, what changed Lewis's model is that the industrial sector can only attract labor at the expense of losing agricultural production. Furthermore, the increase in real wages is a constant process and not sudden, although there is no turning point, unlike Lewis (Ge & Yang, 2010; Schultz, 1964). Consequently, neoclassical theory is based on elasticities of substitution and production functions, which is why it emphasizes the importance of the market in resource allocation and therefore sees regional inequality as a temporary phenomenon (Wei, 2015).

In contrast, Keynesian economics takes marginal propensities to save as crucial. In Nicholas Kaldor's model, the economy is composed of two classes: workers and capitalists, where workers are less likely and capitalists are more likely to save (Galbraith, 2001). Thus assuming that investment in total income is exogenous in nature and does not depend on changes in the saving rate. Under conditions of full employment, the only possible distribution of income between earnings and wages can be achieved when saving equals investment (Gallo, 2002). Therefore, an increase in investment requires a corresponding increase in saving, whereby capitalists, as the most substantial savers, demand greater profits. This, in turn, results in a higher price level (Galbraith, 2001). Thus, Kaldor's model suggests a positive correlation between income inequality and economic growth. In the following decades, the model faced several criticisms about its restriction to only two classes or because the investment was exogenous.

The aforementioned theoretical concepts aimed to provide answers about the functional distribution of income versus the personal distribution of income. Although neoclassical theory provides some insight into the distribution of personal income through differences in factor endowments, it does not seem to be sufficient to explain the rampant Gini coefficients in developing countries like China. Furthermore, it does not explain how the differences in endowments were created in the first place. Theories about the distribution of personal income can be classified into theories that assume that income inequality is primarily the result of voluntary choice (e.g., Milton Friedman's theory of individual choice) and those in which institutions and heredity are fundamental (for example, the work of Acemoglu and Robinson on institutional persistence in Latin America), while representatives of the other believe that income is genetically determined or that the results of inequality are attributed to chance and luck (Wei, 2015).

The next section will be devoted entirely to the relationship between economic growth and income inequality, first in general and then in the context of China.

SECTION 3: Literature review on economic growth and income inequality

3.1 Income inequality: a necessary evil for economic growth?

Official statistics show that there has been a global increase in inequality in terms of income distribution over the past two decades (Schwab, 2018). Academic discussions, political debates and the popular press commonly react to this trend, seeing this increase in inequality as a problem that must be addressed through redistributive policies. However, there are some scholars who disagree with this notion, arguing that inequality is not necessarily a problem and therefore does not need targeted policies to address it. In this sense, these scholars point out that there is a greater need to develop policies that address poverty (Lyubimov, 2017; Saez & Zucman, 2016). Based on this premise, some researchers argue that some of the changes in society result in an increase in the income of higher-income individuals, without affecting income at the lower end of the income distribution. Such a change is in accordance with the Pareto principle, as it improves the income of high-income people, while the income levels of low-income people remain unchanged or also increase, respectively, although at a slower pace than the groups of low-income people. high income. (Jenkins, 2017).

Piketty is one of the authors who have presented significant arguments against income equality, postulating that income inequality will continuously increase within a given population with economic development and that this is not necessarily a bad sign (Lyubimov, 2017). The central aspect of Piketty's argument is capital. In his argument, Piketty postulated that capital is more unevenly distributed compared to labor income, which significantly affects household income, contributing enormously to income inequality (Piketty, 2014). Piketty noted that capital is made up of various assets, including intellectual property, financial capital, equipment, real estate, and land, among others (Lyubimov, 2017). As such, since capital is primarily owned by a small percentage of individuals, these individuals are able to accumulate more wealth within an economy as a result of the relatively higher income earned from capital. The fact that this capital is in the hands of a small fraction of the population contributes significantly to these people making large profits in terms of capital income (Wade, 2014). Once these people profit from their capital, they can invest more in assets and increase their income (Jones, 2015). For example, people with high equity yields can hire those who have a better understanding of asset management, allowing them to increase their return on investment. In contrast, households with few assets at their disposal do not have as many opportunities and are therefore forced to rely entirely on traditional financial services and labor income, receiving lower returns (Lyubimov, 2017).

The capital that has accumulated is inherited, so it falls within the small percentage of the population (Wade, 2014). In addition, the richest group of the population can be largely represented by professionals, including managers who receive income from work. Hence, Piketty sees rising income inequality as an inevitable and crucial element of capitalism, whose reductions in the past, especially during the World War period, have been associated with shocks.

Many scholars have pondered how the evidently higher Gini for most developing countries, especially emerging economies, compared to developed countries can be explained. Kuznets' model seemed to provide a plausible theoretical framework for this; others have expanded it, such as Fields (2019) with his theory on the different paths of expansions in the modern sector. The American economist Simon Kuznets was one of the first to develop a theory about the development of income inequality. Kuznets (1955) provided an explanation for this with his inverted-U hypothesis, which predicts that economic growth decreases income equality early in the economic development process (Jauch & Watzka, 2016). Once a certain stage is reached, a decline in income inequality is expected, along with a slowdown in economic growth. The modern sector, where productivity and labor wages are high, is the engine of economic growth which, however, is incapable of offering job opportunities to everyone. Therefore, it always leaves behind people who are still part of the traditional sector, leading to greater income inequality (Jauch & Watzka, 2016). However, levels of technology and human capital eventually rise in the wake of persistent economic growth, which in turn creates more job opportunities and a large middle class. This middle class, in turn, ends up demanding redistribution through taxes and public spending, thus reducing income inequality. This described inverted 'U' pattern predicts the evolution of income inequality along with economic growth according to the stage in which the respective country is located (Jauch & Watzka, 2016).

The Kuznets model and other models based on it offer several implications for developing or transition countries. First, income inequality is temporary and a prerequisite as the economy transforms to the point where inequality falls again. Second, emerging economies can overcome income inequality by following a path supported by today's fairly egalitarian developed countries. Although the Kuznets model and its inverted-U hypothesis can be empirically confirmed for a large set of countries, such as Thornton (2001), the Kuznets hypothesis has also received much criticism, including Thomas Piketty's argument that inequality is progressive and that, to be under control, it is necessary to develop an internationally coordinated policy (Lyubimov, 2017).

Another point of view that may help explain why some countries tend to stagnate at a high level of inequality without experiencing significant growth, as is the case in many Latin American countries, is what Ravallion (2016) calls inequality of opportunities. Depending on the degree of income inequality that prevents a certain portion of the population from having access to credit or education, determines how inequality translates into reduced growth. Thus, the income inequality experienced manifests itself in the exclusion of specific groups from society, whether based only on material, racial or gender wealth, etc., more potential for growth and, consequently, economic growth is lost.

Whether or not the Kuznets model applies, the fact remains that today's most developed economies with large redistribution schemes have low Gini levels (note that this excludes Anglo-Saxon countries). It is also true that these countries once had a worse income distribution. Therefore, what can be summarized is that high rates of economic growth for a country in transition are accompanied by increases in income inequality. However, the exact mechanisms behind this have yet to be agreed upon. Furthermore, there is widespread agreement that income inequality is a necessary consequence of a market economy, but how much inequality can or should be accepted is still up for debate. More on this discussion will be presented in the following subsections.

3.2 The impact of income inequality on economic growth

From a practical perspective, the relationship between the degree to which income is evenly distributed within a population and economic growth is not linear. More importantly, numerous unobserved parameters certainly lead to omitted variable biases for studies conducted in this field. Therefore, several empirical studies have produced quite different results. Previously, income inequality was believed to be positively related to economic growth, given that people are given the right incentives to spur that growth. In the 1950s and 1960s, researchers argued that high-income people's propensity to save more meant that income inequality drove higher levels of investment, which positively affected economic growth (Benhabib, 2003). In a review of 45 countries between 1966 and 1995, Forbes established that there is a positive relationship between the level of income inequality within a country and its growth in terms of the economy (Forbes, 2000).

However, in the 1990s, a myriad of empirical studies provided a different position. In a study carried out by Knell (1999) to establish the relationship between income distribution through the Gini coefficient and the per capita GDP growth rate, the researcher pointed out that, between 1960 and 1985, the annual growth rate decreased between 0.3 and 0.6 percent. percent as a result of a 10 percentage point increase in the Gini coefficient. In another study, Herzer and Vollmer (2012) examined the impact of income inequality on per capita GDP in 46 countries between 1970 and 1995, and the researchers concluded that an increase in income inequality negatively affected GDP growth. Significantly, researchers established that this relationship was consistent across developed and developing countries, as well as democratic and non-democratic countries (Herzer & Vollmer, 2012). However, when carrying out their literature review, the authors found that some studies point to a positive relationship between economic growth and income inequality (Herzer & Vollmer, 2012).

In a study by Barro (2000), the author investigated the relationship between income distribution and economic growth between 1965 and 1995 in 84 countries. In his findings, the author concluded that no significant relationship could be observed between them. However, with the categorization of countries into rich and poor, there were changes in the results. In this case, poor countries, including those with real GDP per capita below US$ 2,070 (in 1985 dollars), demonstrated a negative relationship between income inequality and economic growth (Barro, 2000). On the other hand, the group formed by rich countries showed a positive relationship between the variables. Ostry, Berg and Tsangarides (2014) conducted a review of previous studies on the relationship between income inequality and economic growth. The researchers found that most studies establish a negative relationship between variables, and economic growth is held back by rising income inequality (Ostry, Berg, and Tsangarides, 2014). Findings from their own empirical study also demonstrated a negative impact of income inequality on economic growth. The authors were quick to point out that several studies were found to take the opposite position (Ostry, Berg, & Tsangarides, 2014). Furthermore, an OECD study by Cingano (2014), an IMF study by Berg and Ostry (2013) and a Standard & Poor's study (2014) have demonstrated the negative implications of rising income inequality. Furthermore, United Nations reports and studies, including the 2013 United Nations Development Program report on inequality, regularly show the growth-hungry implications of income inequality (UNDP, 2013a).

The research does not provide adequate evidence on whether there is a shift in the direction of the implications of income inequality on economic growth between moderating growth and promoting growth, depending on its magnitude. Based on a study by Cornia et al., the authors argued that increasing income inequality resulted in economic growth up to a Gini coefficient value of 0.3, while a dampening effect of income inequality resulted in income growth in the economic growth was carried out with a Gini coefficient value greater than 0.45 (Cornia, Odusola, Bhorat and Conceição, 2017). Earlier, in Cornia and Court's 2001 policy brief, it was revealed that a growth-promoting effect would occur with a Gini coefficient between 0.25 and 0.4 (Cornia and Court, 2001). In contrast, Cho, Kim, and Rhee (2014), authors of the Korea Institute for International Economic Policy, conducted a study to explore the relationship between income inequality and economic growth between 1980 and 2007, and their findings revealed that the buffer effect of inequality of income can be performed with a Gini coefficient value lower than 0.245 (Cho, Kim and Rhee, 2014).

Furthermore, there are different implications and indirect effects of income inequality with an indirect impact on the economic growth rate. These can include negative health status effects due to competition and overwork (Rowlingson, 2011). The negative impact of income inequality on health is reiterated in the study by Pickett and Wilkinson (2015), in which the authors established a causal relationship between income inequality and the health and well-being of the population (Pickett & Wilkinson, 2015 ). Likewise, Reiss (2013) finds that income inequality may be related to mental illness. It can be argued that these negative health effects can lead to lower productivity and therefore lower economic growth.

Other important links to economic growth have been made with regard to the promotion of crime in the economy through a lower level of foreign direct investment, tourism or due to generally higher transaction costs of running a business. According to neoclassical theory, it may be rational for relatively poor individuals in a given country to commit a crime, since its benefits and costs differ from the rest of society, such as greater relative benefits (Danziger & Wheeler, 1975). For example, Rufrancos, Power, Pickett and Wilkinson (2013) confirm the neoclassical theory and establish a positive relationship between income inequality and crime. The authors reviewed 17 studies examining the association between the two variables using time series data. Their findings suggested that there was an increase in property crimes as a result of rising income inequality, particularly a high sensitivity of violent crime rates, including robbery and homicide, in reference to income inequality (Rufrancos, Power, Pickett, & Wilkinson, 2013). 🇧🇷 Other implications of income inequality may include the tendency to have less political stability, which has implications for the level of FDI or tourism, as argued by Alesina and Perotti (1993).

Taken together, these results show that starting with a relatively high Gini coefficient is likely to have a dampening effect on economic growth (Standard & Poor's, 2014). However, it should be noted that said average value is based on the analysis of data from several countries that, among others, achieved different levels of growth in terms of their economies, had different levels of income or had completely different causes of growth. 🇧🇷 Evidence suggests that the state of economic development of each economy may be critical in determining the extent to which there is a shift from growth-promoting to growth-depressing effects of income inequality.

3.3 Long-term implications of income inequality on economic growth

In the discussion of the previous section, this article performed a comparative static analysis, comparing different levels of income inequality and GDP growth in several countries. However, this section examines the long-term implications of rising income inequality for economic growth. It also explores the implications they can have on the demand and supply sides of GDP, including their resulting interactions.

On the supply side of GDP, greater income inequality can lead to a weakening of the economy's productive potential, especially in relation to human capital (Cingano, 2014). In cases where individuals within an economy feel that an additional contribution to the economy is unprofitable as a result of most of the national income being distributed among a small portion of the general population, they are likely to lose interest in making investments. in human capital including their own education. The improvement of human capital, in terms of quality, is essential for any economy to achieve economic growth (Galor & Zeira, 1993). Increasing income inequality becomes a serious challenge in cases where people are very unhappy with the fact that they ended up leaving their country (Bernstein, 2013). Empirical evidence shows that cross-border mobility is more common among well-educated young people, leaving affected countries at risk of brain drain and capacity for economic growth. A country whose human capital weakens as a result of individuals' reaction to rising income inequality experiences a reduction in its ability to experience low economic growth in the long run (Bernstein, 2013). Increased income inequality can also result in the impairment of human capital within an economy, preventing low-income people from accessing the health care system. From a general perspective, reduced investment in health and education results in reduced human capital formation, which ultimately results in stunted economic growth (Daguet, Colombier and Baur, 2015).

In addition, supply-side weaknesses can result from using income concentration to accumulate economic power that could be applied to exert political influence (Daguet, Colombier, and Baur, 2015). As such, one would expect high-income individuals within an economy to propose tax cuts or subsidies. The reduction in state revenue would subsequently require reductions in terms of state spending and targeted investments in education and infrastructure. With public services cut, the state's productive apparatus weakens, hampering economic growth (Bernstein, 2013).

Finally, high market income inequality can lead to a case where extensive income redistribution by the state is needed (Fenig, Mileva, and Petersen, 2013). For this to be achieved, higher levels of government revenue are needed. Thus, high levels of income security in these situations require either an expansion of the public debt or an increase in taxes. Tax increases prevent taxpayers from receiving performance incentives, resulting in capital flight and brain drain. As a consequence, the affected economy experiences a decrease in investment volumes, which in turn leads to a decrease in the economy's capital stock and economic growth (Fenig, Mileva, and Petersen, 2013). A similar result occurs in cases where the increase in state debt leads to an increase in interest rates, which suppresses private sector investment (Fenig, Mileva and Petersen, 2013).

On the demand side of GDP, rising income inequality has a dampening effect on demand for services and goods. In cases where high income inequality results in a small portion of the population making up high-income households, there is an increase in savings for the entire population, which results in a reduction in demand, due to the decrease that occurs in the rate of consumption of individuals with higher levels of disposable income. Taking Germany as an example, given the high quality of available statistical data, such a relationship can be seen in Figure 2 below. The income bracket with net monthly family income below 1,300 euros even shows a negative average savings rate, indicating the use of loans to prolong their consumption level above their income and higher savings rates the higher the income.

[...]

¹For example, for 2015: World Bank estimates 0.386, while Chinese statistics agency suggests 0.462, SWIID up to 0.469.

²For this reason, some studies only include families or people in the 18-64 age group, such as the OECD.