[author]SHI Lei

[content]

The Influence Mechanism of Changes in Civil Litigation Rate of China----Testing and Amending the Curve Theory

*Author Shi Lei

Associate Researcher, Koguan School of Law, Shanghai Jiao Tong University

Abstract: The classic curve theory posits an inverted U-shaped relationship between the rate of civil litigation and economic development, yet it does not indicate whether this relationship is influenced by other factors. Due to the lack of large-scale regional data, existing researches hardly directly test and amend the theory within the Chinese context. This empirical research based on provincial panel data shows that the civil litigation rate in China is characterized by a “decline then rise”  temporal pattern and an “east-west divide”  spatial pattern. It demonstrates that there is an inverted U-shaped relationship between the civil litigation rate and economic development, situating presently in the curve’s initial phase where the economy maintains a positive, albeit diminishing, impetus over an extended period. More importantly, the degree of marketization and the level of education play a moderating role in attenuating and reinforcing the curve relationship, respectively. The discovery of this interaction mechanism in China advances theoretical development, leading to the establishment of a new curve theory based on multiple conditions. This contributes to thinking about how to systematically improve the judicial capacity of courts, for instance, by combining the enhancement of internal management efficiency with the construction of predictive mechanisms for region-specific litigation demands, as well as optimizing macro-level policies for social collaborative governance.

Introduction

The changing trends in China’s civil litigation rate are a crucial basis for deepening judicial reform. The civil litigation rate is generally defined as the ratio of civil litigation cases to the population within a specific time and region, providing a direct reflection of societal attitudes toward litigation and judicial demand. In recent years, the increasingly prominent “case-to-judge imbalance”  has posed severe challenges to the judicial capacity of Chinese courts, drawing widespread attention from both practitioners and academics. To address this issue, current reforms focus on enhancing judicial supply capacity from within the courts, such as through reforms in judicial quotas, streamlined case management, and the construction of smart courts, yielding positive outcomes. However, the external demand perspective should not be overlooked.

On one hand, understanding the patterns of judicial demand helps in adopting timely and appropriate responses, avoiding overreaction to short-term fluctuations or excessive efforts when demand is about to decline, which could lead to wasted resources and harm judicial fairness. On the other hand, identifying the regional characteristics of judicial demand facilitates the rational allocation of limited judicial resources, promoting regional co-development. Therefore, in the context of understanding judicial demand, the civil litigation rate becomes a critical object of study.

Domestic and international academic research has produced abundant findings on the evolution of civil litigation rates. International studies focus on the macro relationship between the civil litigation rate and economic development, proposing the classic curve theory while continuously identifying other factors influencing the civil litigation rate. However, these studies have not further explored the interrelationships and pathways among these factors, nor have they paid much attention to the experiences of developing countries. Domestic research primarily concentrates on depicting and explaining the overall changes in China’s civil litigation rate, analyzing the impacts of economic, legal, and social factors. However, due to a lack of large-scale regional data, it is difficult to fully identify the complex mechanisms of influence and effectively respond to and advance existing theories.

Thus, the unresolved questions are: What are the characteristics of changes in China’s civil litigation rate over the past two decades? What factors influence these changes? How do these factors interact to shape the evolution of the civil litigation rate?

To address these questions, this study uses provincial panel data from 1997 to 2020 to depict the changing trends and regional disparities in China’s civil litigation rate, focusing on its relationship with economic development. It empirically tests and revises the curve theory, thereby exploring in depth the mechanisms influencing changes in the civil litigation rate to provide policy recommendations for scientifically forecasting litigation demand and allocating judicial resources. This study highlights that the evolution of China’s civil litigation rate exhibits a “decline then rise”  temporal pattern and an “east-west divide”  spatial pattern. It shows an inverted U-shaped relationship with economic development while also being influenced by institutional environments and social structures. Importantly, some factors within the latter category, such as the degree of marketization and the level of education, play moderating roles in attenuating or reinforcing the curve relationship. Compared with the traditional curve theory based on single conditions, the new curve theory based on multi-layered conditions offers greater explanatory power.

1.1 The Traditional Construction of Curve Theory

For a long time, when exploring the reasons behind changes in civil litigation rates, researchers have primarily considered economic factors. From an intellectual genealogy perspective, if history is broadly interpreted, the study of civil litigation rate changes falls into the category of long-term “historical”  research. In this context, the paradigm shift initiated in the 1950s by James Willard Hurst and the “Wisconsin School”  in legal history and legal studies cannot be overlooked. Traditional legal doctrinalism viewed legal development as self-contained, free from external influences. In contrast, the Wisconsin School challenged this notion of legal endogeneity, asserting that legal development is exogenous and shaped by external social factors, particularly the demands of economic development. Social issues arising from economic progress often need to be transmitted through litigation to judicial decisions and further legislative practices. Litigation, as a medium of legal development, provides a foundational theme and basic framework for scholars studying civil litigation rates.

Under a functionalist perspective, the classic curve theory of civil litigation rates was constructed. This theory posits that the civil litigation rate follows a trajectory of first rising and then falling as economic development progresses, forming an inverted U-shaped curve. The inflection point of the curve occurs when economic development reaches a high or mature level. The growth rate of civil litigation gradually slows with economic progress and eventually shifts from positive to negative at the inflection point. The explanatory logic of this theory is built on three key points:

First, the normative effects hypothesis serves as the starting premise, suggesting that legal or regulatory uncertainty increases the willingness of social actors to seek judicial resolution for disputes. This premise assumes a stable perception of law among social actors, who view recourse to courts as the optimal solution to issues of normative ambiguity. Second, economic development acts as the triggering condition for normative uncertainty, particularly during states of economic transformation that induce significant social changes. Third, the inflection point on the curve arises when law responds to reduce normative uncertainty by addressing gaps in social norms and shaping rational social relations.

Specifically, American scholar Lawrence M. Friedman argued that during the transition from traditional to modern society, the rapid economic growth associated with marketization and industrialization leads to an expansion of market transactions and increased complexity in social relationships. This, in turn, intensifies social conflicts and gives rise to new types of disputes, prompting individuals to seek help from courts due to the inadequacy of existing norms and rules, thus increasing the civil litigation rate. However, when industrialization reaches a mature stage, improvements and rationalization of the legal system, along with the rising complexity and cost of legal rules and judicial procedures, suppress litigation willingness, causing the civil litigation rate to plateau or decline.

The process of modernization forms the practical basis for traditional curve theory, and empirical research has typically validated this theory from the perspective of industrialization. Foundational research was conducted by Spanish scholar José Juan Toharia, who used data on civil litigation rates and socio-economic indicators from Spanish courts between 1900 and 1970 to reveal the curve relationship between changes in civil litigation rates and economic development. From a temporal perspective, he noted that during Spain’s “economic miracle”  in the 1960s, the civil litigation rate shifted from being highly positively correlated with economic development to being weakly correlated. From a regional perspective, Toharia classified Spanish provinces into five categories based on their level of industrialization, finding that economically developed and highly industrialized provinces maintained consistently low litigation rates, while economically underdeveloped and less industrialized regions experienced significant increases in litigation rates alongside rapid economic growth. Similarly, American scholars Joel B. Grossman and Austin Sarat used data from U.S. federal district courts between 1902 and 1972 and socio-economic data from various states, discovering that in regions with the highest levels of industrialization, civil litigation rates showed a certain degree of decline. They concluded that “the influence of industrialization on litigation follows a curve pattern.”  Additionally, related studies based on long-term judicial data from other regions in the United States, as well as from the UK and Germany, also supported the curve relationship within the modernization narrative. However, the traditional curve theory focuses primarily on the macro-level impacts of economic transformations, and its narrow interpretation of economic development tends to obscure the micro-level effects of social wealth growth on litigation willingness.

The absence of micro-level logic regarding litigation in the traditional curve theory not only results in an incomplete economic explanation but also risks excluding other potential influencing factors. Consequently, subsequent research has shifted toward more diverse explanatory pathways, focusing primarily on institutional factors from a national perspective and social factors from an individual perspective. The specific factors and their roles will be detailed later, but it is important to note here that, after considering multiple factors, subsequent studies have demonstrated various forms of the relationship between economic development and changes in civil litigation rates. First, the relationship is nonlinear. American scholar Frank Munger, using trial court data from West Virginia, confirmed the inverted U-shaped curve relationship. However, other researchers, using long-term judicial data from Belgium, discovered irregular dynamic curves. Second, the relationship is positively correlated. This is a more common research finding, but evidence from a few developing countries may simply indicate that these nations have not yet reached the threshold required for the curve to emerge. Third, the relationship is negatively correlated. Research based on Japanese prefectural panel data from 1986 to 2002 showed that civil litigation rates increase during economic recessions, a finding also supported by recent data from Spain. Fourth, there is no correlation. Empirical studies from Italy have provided evidence supporting this perspective. These findings suggest that the relationship between the economy and litigation requires further clarification.

Finally, many studies acknowledge the complex relationship between economic factors and other explanatory variables, but few delve deeper into this complexity. An exception is the research by American scholar Tonja Jacobi. Using data from U.S. state courts between 1975 and 2000, she revealed that the more dispersed governmental power is, the higher the civil litigation rate, and that this relationship is moderated by regional economic levels. Specifically, the higher a state’s economic level, the weaker the positive correlation between power dispersion and litigation rates. Inspired by this, we have reason to hypothesize that the relationship between economic development and civil litigation rates may also be influenced by other factors. Exploring these interactive relationships could significantly advance the development of curve theory and deepen our understanding of the mechanisms driving changes in civil litigation rates.

1.2 The Curve Theory’s Explanation of China’s Civil Litigation Rate and Its Challenges

Research on China’s civil litigation rate, while adopting diverse explanatory approaches, has paid particular attention to the role of economic development, given the tremendous achievements in China’s modernization. According to the assumptions of traditional curve theory, civil litigation rates should increase rapidly during the early stages of modernization. If this is a universal principle, then China’s civil litigation rate should have exhibited a similar trend during the period from the beginning of reform and opening-up to the early 21st century. Existing studies generally agree that there is a positive correlation between China’s civil litigation rate and economic development, but they differ in assessing the explanatory power of economic factors. Ran Jingfu was among the first to analyze the relationship between China’s civil litigation rate and modernization during 1978-2000. He argued that economic development was highly positively correlated with the civil litigation rate. He also highlighted the complexity of economic influence through phenomena such as the temporary decline in civil litigation rates during 1988-1989 and 1999-2000, and the variation in litigation rates across different types of cases. Additionally, he emphasized the role of value orientation and policy choices. Tang Ming and Li Hao, using nationwide civil trial data from 1978-2003, observed a decline in the civil litigation rate from 1996 to 2003, which ran counter to economic growth trends. Based on this, they concluded that the curve’s inflection point had occurred and that the overall changes in China’s civil litigation rate aligned with the curve theory.

However, shifting the perspective from the national to the local level may weaken the explanatory power of economic factors. Regarding the decline in the number of first-instance economic cases in the late 1990s, He Xin compared the operational conditions of grassroots courts in Guangdong and Hunan. From the micro perspective of litigation demand, he argued that economic factors had limited explanatory power, while social changes and the dysfunction of courts better explained the decline. Notably, subsequent studies incorporating updated data and expanded sample sizes pointed out that civil litigation rates began to rise again after 2006. However, their regression analyses yielded divergent results, with some supporting a positive correlation between economic development and litigation rates and others supporting a negative correlation.

In summary, the continuous updating of data and shifts in perspective have reshaped our understanding of the relationship between changes in China’s civil litigation rate and economic development, sparking considerable debate. Nevertheless, existing studies still face significant limitations in two areas, leaving several key questions insufficiently examined and discussed.

First, how can the current changes in China’s civil litigation rate be more comprehensively described?

From a temporal perspective, the latest studies examined data only up to 2014 and predicted that civil litigation rates would continue to rise thereafter. Whether this prediction holds true remains to be confirmed by more recent data. From a spatial perspective, regional heterogeneity must be fully considered. China’s vast territory and unbalanced regional socioeconomic development inevitably result in varying judicial demands. Existing research, constrained by the difficulty of obtaining data, has primarily relied on national-level data from the China Law Yearbook to analyze overall changes in civil litigation rates. This leads to a lack of regional detail, making it difficult to identify unique patterns or specific trends across different areas.

Second, can curve theory adequately explain the Chinese experience?

Although previous research in Europe and the United States did not precisely define the standards for the curve’s inflection point, it generally suggested that the point occurs when economic development reaches an advanced level. According to the United Nations and other international organizations, Spain and the United States had already entered the ranks of advanced economies by 1970. The studies by Toharia and Grossman, which empirically validated curve theory and its inflection point, were conducted within this context of advanced development. China, as the largest developing country in the world, has experienced rapid economic growth since the reform and opening-up period. By 2020, China had achieved its goal of building a moderately prosperous society in all respects, with its GDP doubling compared to 2010. Particularly in developed eastern regions such as Shanghai, Beijing, Tianjin, Jiangsu, and Zhejiang, per capita GDP in 2020 approached or even exceeded the widely recognized $20,000 threshold for middle-income advanced economies. If we consider China’s development over the past decade to be more mature and modernized than before, the explanatory power of curve theory for China’s civil litigation rate requires reevaluation.

In fact, existing research on China’s civil litigation rate has not truly and effectively tested the curve theory. There are currently three approaches to testing the curve theory. The first is the overall temporal observation method. If a country or region experiences continuous economic growth while its civil litigation rate initially increases but later stabilizes or declines, a curve relationship is presumed to exist. This approach essentially provides a descriptive observation of correlation while ignoring all confounding factors, making it not a rigorous test in the strict sense. The second is the regional comparison method. Within the same country, if economic growth is positively correlated with changes in civil litigation rates in less developed regions, but negatively correlated or unrelated in more developed regions, the curve relationship is considered valid. Compared to the first method, this approach has the advantage of controlling for some national-level confounding factors. The third is the regression analysis method based on large samples, which is currently the most scientific way to test the curve theory. The technical approach involves setting control variables and demonstrating the existence of an inverted U-shaped curve relationship by showing that the first-order term of the independent variable is significantly positive and the second-order term is significantly negative. Compared to the first two methods, this approach incorporates the rich information provided by continuous variables, more comprehensively controls for other factors, significantly enhances the validity of conclusions, and uses mathematical models to intuitively reveal the true relationship between the explanatory and dependent variables. Previous studies in China have primarily adopted the first temporal observation method, which cannot provide reliable conclusions. Even the few studies attempting to use regression analysis have been constrained by the limited availability of national-level samples, making their results similarly unconvincing. Thus, the effective identification of factors influencing China’s civil litigation rate largely depends on access to large-scale regional data.

Finally, and more importantly, if other explanatory factors exist, how do they interact with economic factors to influence changes in China’s civil litigation rate? Since the first two questions have not been adequately addressed, this deeper theoretical issue has remained unexplored. As highlighted in the previous review, addressing the lack of large-scale regional data would not only enable a proper test of the curve theory and identification of additional influencing factors but also allow further exploration of the interactions between these factors. This would provide a substantive response to and advancement of the curve theory based on China’s unique experience.

1.3 Conditional Adjustments to the Curve Theory and Mechanism Hypotheses

This paper proposes that there may exist a “multi-layered conditional” curve theory capable of providing a deeper explanation of changes in civil litigation rates while revising and expanding upon the “single-condition” traditional curve theory. As previously mentioned, traditional curve theory posits a fixed curve relationship between civil litigation rates and economic development. Not only is economic development regarded as the decisive influencing factor on civil litigation rates, but the relationship between the two (such as its direction and intensity) is assumed to be unaffected by other factors. As a result, the validity and manifestation of the curve relationship rely solely on economic development as a single condition.

In contrast, the multi-layered conditional curve theory argues that the curve relationship between civil litigation rates and economic development is not fixed but rather influenced by institutional environments, social structures, and other unknown factors, which together produce diverse forms. The curve relationship no longer depends solely on economic development but varies across different layers of social attributes, resembling the way light refracts when passing through clouds. Specifically, the new curve theory recognizes that civil litigation rates and economic development exhibit a curve relationship. It also acknowledges that civil litigation rates are influenced by a variety of other factors, and among these, some factors interact with economic development, exerting a moderating effect on the curve relationship. In this sense, the multi-layered conditional hypothesis advances curve theory both in breadth and depth, reflecting the structural characteristics of the mechanisms influencing civil litigation rates.

Regarding the primary influencing factors and mechanisms, further theoretical elaboration is required. According to existing research, the choice of dispute resolution methods is fundamentally driven by two reasons: subjective recognition and rational calculation. Subjective recognition refers to individuals acknowledging the legitimacy of state judicial authority and believing that resorting to courts is superior to other methods, such as inaction or reconciliation, in resolving disputes, thereby forming an emotional and subconscious preference for litigation. In contrast, rational calculation refers to the process where individuals, based on their available resources--such as wealth, knowledge, and information--objectively and consciously weigh the pros and cons of litigation outcomes. In the decision-making process, the primacy of these two factors often shifts depending on the specific circumstances of a dispute. These two behavioral drivers, derived from general observations of human nature, possess universality and can serve as a logical foundation for explaining the mechanisms influencing civil litigation rates in China.

Figure 1 depicts the pathways of influence among the various factors within the revised curve theory, detailed as follows.

Figure 1 Diagram of Conditional Adjustments and Influence Mechanisms in Curve Theory

1.3.1 The Impact of Economic Development on Civil Litigation Rates

Economic development provides the most fundamental impetus for the emergence of litigation, but its influence is not constant. For example, during the processes of economic modernization and rural revitalization, the frequency and scope of rural land transfers have gradually expanded. However, neither the original Property Law, the Rural Land Contracting Law, nor their interpretative documents have clarified certain specific technical issues, leading to frequent farmland disputes. With the implementation of land certification in recent years, land transfers have become more standardized, and related litigation has gradually decreased. Other empirical studies indicate that economic development has led to an increase in labor disputes, but following the implementation of the Labor Contract Law (as amended) in 2008, related cases have decreased. Moreover, labor contract practices in the more developed eastern provinces and cities are more standardized than in other regions, resulting in fewer cases being filed. These situations align with traditional curve theory. Economic development generates normative uncertainty, expands the sources of disputes, and drives people to seek judicial relief to clarify the boundaries of rights and obligations. As standardization progresses, the courts shift their focus from resolving “legal issues” to “routine management,” leading to the stabilization or decline of civil litigation rates.

This channel of influence through normative uncertainty is primarily based on subjective recognition. As mentioned earlier, the implicit premise of the normative effects hypothesis is that social groups generally possess a high level of legal awareness, recognize the lack of norms in new issues, and trust that resorting to courts is reliable. In countries such as the United States and the United Kingdom, where traditions of “judge-made law” and “rule by judiciary” are longstanding, subjective recognition of litigation may not pose a problem, and normative uncertainty can smoothly drive an increase in civil litigation rates. However, in the context of China’s modern development, this channel of influence has not always been unimpeded. From the reform and opening-up period to the early 21st century, the people’s judiciary underwent a process of “restoration and emergence--partial disconnection--partial recovery,” during which people harbored concerns about resorting to courts. Studies have noted that the overall sense of judicial fairness in society remained relatively low between 2006 and 2013, only beginning to rise gradually after 2013. Furthermore, in the early stages of development, the relatively low degree of legal recognition among Chinese citizens was reflected in their emphasis on petitioning (xinfang), perceiving that appealing to higher levels of political authority was more effective than relying on the law. As hypothesized later in this paper, the realization of this channel of influence may rely more on the adjustment of subjective recognition through other factors.

The corresponding changes in litigation costs and benefits, as another channel of influence stemming from economic development, may have a more direct impact in China. Litigation costs are key factors in a rational person’s decision to pursue litigation. If litigation costs exceed expected benefits, disputing parties are likely to choose lower-cost alternatives such as reconciliation. In terms of increasing personal wealth, economic development enables individuals to bear higher litigation costs, thereby strengthening their willingness to litigate. However, as living standards continue to rise, the opportunity costs of litigation increase correspondingly, and people gradually lose the willingness to pursue litigation for small disputes, opting instead for reconciliation. At this stage, civil litigation rates may decline. For example, prior to the 2021 amendment to the Civil Procedure Law, small claims cases were generally rare across the country, far below the Supreme People’s Court’s anticipated goal of constituting 30% of all civil cases. Grassroots courts have reported that “some citizens feel that disputes involving less than ten thousand yuan are not worth pursuing in court.” Similarly, in economically developed Wenzhou, the average claim amount for private lending disputes increased from 389,800 yuan in 2008 to 797,600 yuan in 2016. This indirectly reflects the relatively high litigation costs for small disputes, which may discourage the choice of litigation. Thus, both channels of influence, regardless of their relative primacy, emerge in China. Theoretically, civil litigation rates and economic development can be assumed to exhibit a curve relationship.

1.3.2 Institutional Environment and Social Structure’s Impact on Civil Litigation Rates

The institutional environment influences civil litigation rates by adjusting the thresholds and functionality of dispute resolution methods. The institutional environment refers to a series of social behavioral norms and related domains that the state can establish or intervene in. It primarily includes several sub-factors. First, the legal profession as a factor: The technical complexity of the law often requires disputing parties to seek professional lawyers, who, motivated by profit, may persuade clients to litigate. The increasing proportion of civil cases handled by lawyers in China, along with their concentration in economically developed eastern regions, illustrates this rational tendency of the profession. Therefore, the availability of legal services strengthens litigation willingness. Second, judicial policy as a factor: Lowering litigation thresholds and simplifying procedures may reduce litigation costs, encouraging more lawsuits. Relevant policy reforms include the implementation of the Measures for the Payment of Litigation Fees in 2007, the reform of the case filing registration system in 2015, and recent reforms aimed at streamlining complex and simple cases. Third, trial capacity as a factor: Transparent litigation procedures and improved judicial competence enhance public trust in the judiciary and litigation willingness. The Third and Fourth Plenary Sessions of the 18th Central Committee of the Communist Party of China fully outlined plans to deepen judicial reform and comprehensively advance the rule of law, which have significantly bolstered court trial capacity. Judicial quality, efficiency, and credibility have been further strengthened, likely reinforcing the public’s willingness to litigate. Fourth, marketization as a factor: The role of market forces in resource allocation has been increasingly emphasized and elevated in China’s economic reforms. However, the impact of marketization on civil litigation rates has received little attention. At low levels of marketization, government officials’ rent-seeking behavior coexists with business operators’ pursuit of political connections. Insiders with “privileged access” may manipulate judicial processes, favoring litigation, while outsiders, trusting authority, may also prefer judicial remedies. As marketization progresses, government control over resource allocation diminishes, fair competition environments emerge, transaction mechanisms become transparent, free resource flows are established, and property rights are better protected. The decision-making costs associated with information asymmetry are significantly reduced, making parties more inclined to use efficient and flexible informal dispute resolution methods such as negotiation or mediation. Thus, theoretically, civil litigation rates are negatively correlated with marketization but positively correlated with the availability of legal services, litigation fee reforms, and legal reforms.

Social structure also influences litigation willingness. Social structure here refers to the organizational forms, class systems, and relational patterns that shape and influence individual social behavior, including factors such as educational level, income inequality, and urbanization. First, the role of education cannot be overlooked, as China has long incorporated moral and legal education into its national education system. It is generally believed that the higher a person’s educational level, the stronger their legal awareness, the greater their acceptance of judicial dispute resolution methods, and the better their ability to process and absorb complex information. This makes them more adept at understanding legal rules and procedures, thus achieving outcomes most favorable to themselves and strengthening their willingness to litigate. Second, against the backdrop of unbalanced urban and regional development, income disparities among residents may affect litigation behavior. Income inequality exacerbates survival pressures and societal dissatisfaction among low-income groups, intensifying social conflicts and interpersonal disputes. For low-income individuals, the opportunity costs of litigation are lower, making it a viable means of redistributing benefits, thereby increasing their willingness to bring disputes to court. Third, urbanization affects how interpersonal relationships are handled. As more labor migrates from rural to urban areas, the rise in urban population density leads to more frequent and complex interactions, resulting in the emergence of a “society of strangers.” Disputing parties often lack traditional relational ties or stable social interactions, reducing moral concerns and lowering the hidden costs of litigation. As a result, such individuals are more inclined to resolve issues through litigation. Thus, theoretically, civil litigation rates are positively correlated with educational level, income inequality, and urbanization.

1.3.3 The Moderating Role of Marketization and Education Factors on the Curve Relationship

In terms of interaction, the impact of economic development on civil litigation rates may be further moderated by factors such as marketization and education. In theoretical explanations, the moderating effects specifically target the two economic channels of normative uncertainty and litigation benefits.

Marketization factors weaken the impact of economic development. On one hand, the degree of marketization influences the legal problematization of emerging disputes. A higher degree of marketization signifies a more orderly market environment. Mature industry organizations, communication mechanisms, and private dispute resolution systems can absorb some of the new social issues arising from economic development before state intervention and foster new norms. From this perspective, the degree of marketization suppresses litigation willingness by strengthening people’s recognition of alternative dispute resolution methods. On the other hand, a high degree of marketization also signifies a robust credit system and cooperative environment. During economic development, even when market operators and participants use legal means more frequently to standardize transactions, they may choose to avoid legalizing certain disputes to maintain long-term cooperative relationships and market reputations. In such cases, the additional costs associated with litigation may be too high, thereby limiting the early-stage function of economic development in reducing relative litigation costs.

In contrast, education factors enhance the impact of economic development. On one hand, groups with higher levels of education have stronger legal awareness. Therefore, when economic development induces a large number of disputes, compared to groups with lower education levels, they are more capable of identifying new issues with unclear norms and are thus more likely to bring such issues to court. On the other hand, groups with higher education levels have stronger legal competence, enabling them to further reduce litigation costs in situations where wealth growth brings about relatively lower costs. This makes even small claims litigation yield worthwhile returns. Consequently, as economic development continues, this group’s willingness to litigate becomes even stronger.

In summary, the traditional curve theory based on single conditions has the potential to be expanded into a curve theory based on multi-layered conditions. Not only is there a curve relationship between civil litigation rates and economic development, but this relationship is also influenced by factors at the institutional and social structural levels. Empirical tests will follow to examine these aspects.

2. Data Structure and Research Methods

2.1 Data Sources and Variable Settings

The core data on litigation case volumes used in this study primarily come from provincial yearbooks (excluding Hong Kong, Macau, and Taiwan regions of China). Provincial yearbooks are generally compiled by the local chronicle compilation committees of provinces, municipalities, and autonomous regions, with data provided by statistical bureaus and other relevant government departments. Most provincial yearbooks can be traced back to the 1980s, but data from the early years are often incomplete. Considering the administrative change in 1996, when Chongqing became a directly governed municipality, the provincial yearbooks from 1997 to 2020 were selected as the data source for this study.

Provincial yearbooks are currently the most comprehensive and effective publicly available source for data on litigation case volumes. Although the work reports of provincial higher people’s courts are also important sources of judicial data, their accessibility is limited. Both online resources and published compilations have significant time gaps. In fact, a random comparison of provincial higher court work reports with the relevant content in yearbooks for certain years shows that the data recorded in the yearbooks are almost identical to those in the work reports for the same years, particularly over the last decade. In some cases, content from yearbooks is directly extracted from the work reports. Therefore, the two sources are complementary, and when yearbook data for certain years are unavailable, the corresponding work reports can be used as supplements.

Additionally, this study references and utilizes publicly available judicial data obtained from provincial statistical yearbooks, websites of provincial higher people’s courts, and local chronicles to cross-check, correct, and fill in significant errors and missing data in the provincial yearbooks. Furthermore, other original data related to variable settings are primarily sourced from the China Statistical Yearbook, Compilation of Statistics for Sixty Years of New China, China Lawyer Yearbook, and provincial statistical yearbooks. Details on data usage are provided in the source notes accompanying each figure.

Table 1 Variable Settings and Calculation Methods

Table 1 presents the specific variable settings and calculation methods. The dependent variable in this study is the civil litigation rate, defined as the number of first-instance civil case closures (including commercial cases) per 10,000 people. Ideally, the civil litigation rate would represent the ratio of civil litigation case volumes to the actual number of societal disputes. However, since the total number of disputes is difficult to ascertain in practice, population size--representing the potential number of disputants--can serve as a reasonable proxy. Therefore, the case-to-population ratio is a frequently used and relatively reasonable substitute variable.

Moreover, first-instance case closures, rather than case filings, were chosen to calculate the civil litigation rate based on three considerations. First, first-instance case closures reported in provincial yearbooks include carried-over cases, avoiding discrepancies in statistical definitions. Second, publicly available data on case closures are significantly more comprehensive than those on case filings, offering greater completeness. Third, under China’s strict judicial time-limit system, case closures and case filings exhibit nearly identical trends. As shown in Figure 2, the national average case closure-to-filing ratio has remained above 95% for an extended period and often exceeds 98%, with minimal variation. Thus, using first-instance case closures introduces negligible bias.

Figure 2: Ratio of Civil First-Instance Cases Concluded to Accepted (1997–2020)

For explanatory variables, this study designed three categories of indicators.

The first category comprises economic development variables, which are the core explanatory variables. Economic development levels are represented by real GDP per capita, based on 1995 constant prices. The second category measures the impact of institutional environments. The proportion of foreign direct investment (FDI) in GDP reflects the degree of marketization. The number of lawyers per 10,000 people represents the scale of legal professionals, measuring the accessibility of legal services and reflecting the degree of legal institutionalization. Additionally, following the approach of American scholars Ginsburg and Hoetker, dummy variables are set for the years 2007 and 2014 to capture the effects of the 2007 implementation of the Measures for the Payment of Litigation Fees and the post-2014 nationwide promotion of “comprehensive rule of law.” The third category involves variables related to social structure. Specifically, the average years of education for individuals aged six and above represents the educational level of residents. The Gini coefficient represents the income gap between urban and rural residents. The proportion of the urban permanent population to the total population represents the urbanization rate.

Finally, this study compiled balanced panel data for 30 provincial administrative regions in China from 1997 to 2020. Data for Hebei Province were excluded due to significant gaps in first-instance case closure data. For small amounts of missing data on case closures and the number of lawyers, linear interpolation was used based on patterns of data variation. After variable calculations, a dataset with a total of 720 samples was formed. Table 2 presents the descriptive statistics of the dataset.

Table 2 Descriptive Statistics of Variables

2.2 Statistical Model

This study employs the fixed effects model for statistical analysis using the plm package in the R programming language. There are two reasons for this choice. First, the Hausman Test results (p < 0.01) indicate that the fixed effects model is superior to the random effects model. Second, the fixed effects model controls for unobservable regional differences and the impacts of policy implementation at different times, thereby mitigating the estimation bias caused by omitted variables. Additionally, to avoid reverse causality and address potential endogeneity issues, variables such as real GDP per capita, the proportion of foreign investment, and the scale of the legal profession are lagged by one period before being incorporated into the model. Some variables are also log-transformed to reduce measurement bias and scale disparities. Finally, to test the curve relationship, the model includes the quadratic term of the logarithm of real GDP per capita. If the linear term is positive and the quadratic term is negative, this would mathematically demonstrate an inverted U-shaped curve relationship. The specific formula for the statistical model is expressed as follows:

In the formula, i（=1，2…30）represents the 30 provinces, and t（=1，2…24）represents the 24 years from 1997 to 2020. The dependent variable is the civil litigation rate, αis the intercept term......are the coefficients of the independent variables, indicating how many units the dependent variable changes when the independent variable changes by one unit, represents the province fixed effects,  represents the time fixed effects, and is the error term.

3. Temporal and Spatial Characteristics of Changes in China’s Civil Litigation Rate

To effectively and concisely reflect the trends, this section adopts Jacobi’s method, using median values to depict civil litigation rates at the national, major economic region, and provincial levels, thereby outlining the temporal and spatial characteristics of China’s civil litigation rate changes from 1997 to 2020.

First, China’s civil litigation rate shows an overall upward trend but with distinct phase differences. Figure 3 indicates that from 1997 to 2020, the national median civil litigation rate increased from 38.72 cases per 10,000 people to 97.34 cases per 10,000 people. Among these phases, 1997–2006 represents the early stage of a gradual decline in litigation rates. After experiencing slight growth in 1998 and 1999, the national civil litigation rate began to decline, reaching its lowest median value of 30.36 cases per 10,000 people in 2006, a 21.55% decrease compared to 38.70 cases per 10,000 people in 1999.

Starting in 2007, the national civil litigation rate entered a sustained growth phase, reaching its highest median value of 99.53 cases per 10,000 people in 2019, a 2.28-fold increase compared to 2006, before slightly declining in 2020. This “decline-then-rise” phased change does not entirely align with the steady and continuous growth trend of per capita real GDP. Judicial reforms may explain part of the reason.

Before 2007, high litigation costs suppressed demand, while litigation fee reforms afterward reduced costs and released demand. Furthermore, since the implementation of the comprehensive rule-of-law strategy in 2014, the deepening of judicial reforms has significantly improved court trial quality and efficiency. Concurrently, the reform of the case registration system fundamentally addressed the issue of “difficulty in filing cases,” thereby enhancing citizens’ trust in the judiciary and their willingness to litigate. Figure 4 clearly shows a “dual-peak” pattern in the annual growth rate median of the national civil litigation rate in 2008 and 2015, partially supporting the above inference, though further detailed arguments are still required.

Figure 3: Median Trends in China’s Civil Litigation Rate and Per Capita Real GDP (1997-2020)

Second, after dividing regions according to their traditional economic development levels, the changes in civil litigation rates in the eastern, central, and western regions generally follow the aforementioned temporal characteristics, though with notable differences. As shown in Figure 3, the median civil litigation rate in the eastern region consistently exceeds that of the central and western regions. From 1997 to 2020, the median rate in the eastern region rose from 52.59 cases per 10,000 people to 110.76 cases per 10,000 people, an increase of 110.61%. In comparison, the median rates in the central and western regions started at similar levels, rising from 40.03 cases per 10,000 people and 32.17 cases per 10,000 people to 83.84 cases per 10,000 people and 102.29 cases per 10,000 people, representing increases of 109.44% and 217.97%, respectively.

Additionally, after 2007, the eastern region’s civil litigation rate exhibited fluctuating growth, closely mirroring changes in per capita real GDP for most of the period. However, after 2018, its growth rate fell below that of GDP. By contrast, the civil litigation rates in the central and western regions experienced steady and accelerating growth after 2007, with a marked increase after 2014 that outpaced changes in per capita real GDP. Moreover, Figure 4 shows that since 2011, the annual growth rate median of civil litigation rates in the eastern region has been consistently lower than that in the central and western regions.

In summary, while the civil litigation rate in the eastern region has generally remained higher than in the central and western regions, the past decade has seen a slowdown in growth in the eastern region and an acceleration in the central and western regions, with the latter regions even showing signs of catching up with the former. Combined with corresponding economic conditions, this east-west divergence in civil litigation rate changes aligns with the assumptions of the curve theory.

Figure 4: Median Trends in the Annual Growth Rate of China’s Civil Litigation Rate (1998–2020)

Figure 5: Spatial Distribution of the Median Civil Litigation Rate in China (1997–2020)

Finally, from a cross-sectional comparison among provinces, the “east-west divide” in the spatial pattern of China’s civil litigation rate has remained relatively stable. Figure 5 displays the geographic distribution of the median civil litigation rates across provinces, municipalities, and autonomous regions from 1997 to 2020. Using the natural break classification method, the provinces were categorized into five regions, ranked from highest to lowest based on their medians.

Apart from Xinjiang and Ningxia, the high-value and second-high-value regions are primarily concentrated in six eastern provinces and municipalities: Beijing, Shanghai, Tianjin, Jiangsu, Zhejiang, and Liaoning. Among these, Beijing and Shanghai are the only high-value regions, with medians of 127.68 cases per 10,000 people and 101.93 cases per 10,000 people, respectively, far exceeding those of other provinces.

The mid-value region is more dispersed, covering seven provinces and municipalities in the eastern, central, and western regions. Together with the high-value and second-high-value regions, they form a “C-shaped” zone extending from the east to the north.

The second-lowest and lowest-value regions are predominantly located in the central and western provinces, including 13 provinces from these regions and 2 provinces from the east. Their medians are all below 40.06 cases per 10,000 people, over 1.5 times lower than the high-value regions. The lowest median was recorded in Hainan, an eastern province, at 24.18 cases per 10,000 people, followed by Tibet at 24.50 cases per 10,000 people.

Thus, while there are evident differences in civil litigation rates between eastern and western provinces overall, exceptions exist. For instance, certain western regions like Xinjiang and Ningxia exhibit relatively high civil litigation rates, whereas some eastern regions, such as Guangdong and Hainan, display relatively low rates. This suggests that economic development alone may not fully explain the variation in civil litigation rates in China; other regional social factors may also play an important role.

In summary, the changes in China’s civil litigation rate from 1997 to 2020 exhibit two distinct characteristics. First, the “decline-then-rise” temporal feature, with the civil litigation rate declining gradually between 1997 and 2006, followed by sustained growth from 2007 to 2020. Second, the “east-west divide” spatial feature, where the civil litigation rate in the eastern region is higher than in the central and western regions, but the growth rate in the east has gradually lagged behind the central and western regions since 2011. In terms of explanatory reasoning, the curve theory may hold, and economic development is likely a necessary but insufficient condition.

3. Regression Analysis of Influencing Factors

Regarding the influencing factors of China’s civil litigation rate, the fixed effects regression analysis results show that the civil litigation rate and economic development exhibit an inverted U-shaped curve relationship, currently situated in the “outward convex arc” segment of the curve’s early stage. Moreover, this curve relationship is moderated by marketization and education factors. Thus, the curve theory based on multi-layered conditions offers better explanatory power than the traditional curve theory.

3.1 The Curve Relationship Between Economic Development and Civil Litigation Rate

Table 3 reports the fixed effects regression analysis results for the influencing factors of China’s civil litigation rate. The first three columns sequentially display the results of incorporating economic, institutional, and social factors into the regression model step by step. The fourth column adopts the least squares dummy variable (LSDV) model, equivalent to the fixed effects model, to reveal the regression coefficients for the two reforms. Fixed effects for both years and provinces were controlled in all models.

From the specific results, the first-order and second-order terms of the real GDP per capita variable are statistically significant at the 1% level, with positive and negative coefficients, respectively, indicating a statistically significant inverted U-shaped curve relationship between the civil litigation rate and economic development. Additionally, both the stepwise regression models and the LSDV model yield similar regression coefficients and consistent signs of significance, suggesting that potential multicollinearity issues introduced by the second-order term do not affect the robustness of the curve results.

Of course, the mathematical inverted U-shaped curve relationship does not fully equate to its real-world manifestation. Figure 6 illustrates a curve diagram based on the regression coefficients obtained from the results. The dashed line represents the theoretical trend between the civil litigation rate and economic development, while the solid black line shows their actual relationship within the sample range.

As shown in the figure, the actual relationship between China’s civil litigation rate and economic development lies in the early segment of the inverted U-shaped curve, forming an “outward convex arc.” The inflection point is projected to occur when real GDP per capita reaches 568,070 yuan. This suggests that, under the assumption that other conditions remain constant, as China’s economy continues to develop, the growth rate of the civil litigation rate will gradually slow, but it is still far from reaching the inflection point.

Table 3: Fixed Effects Regression Analysis of Influencing Factors on China’s Civil Litigation Rate (1997–2020)

Figure 6: The Curve Relationship Between China’s Civil Litigation Rate and Economic Development

3.2 Other Factors Influencing Civil Litigation Rates

The regression results in columns (3) and (4) of Table 3 indicate that China’s civil litigation rate is also influenced by factors at the institutional environment and social structure levels. Specifically:

In terms of the institutional environment, all four factors have significant effects on civil litigation rates, aligning largely with theoretical hypotheses.

First, the coefficient for the proportion of foreign investment is significantly negative at the 1% level, indicating that for every 1% increase, the civil litigation rate decreases by 0.027%. This suggests that higher levels of marketization, characterized by reduced government intervention and improved market order, lead to lower litigation willingness among market participants. Second, the coefficient for lawyer scale is significantly positive. For every 1% increase in the number of lawyers per 10,000 people, the civil litigation rate increases by 0.159%. This indicates that as the availability of legal services improves, people are more willing to choose litigation to resolve disputes. Third, the 2007 litigation fee reform had a significant impact, but the regression coefficient is negative. This result is counterintuitive. It is generally believed that litigation fee reform significantly reduced litigation costs and should have encouraged more litigation. A possible explanation is that the reform’s incentive effects were not long-term. To investigate further, the study reclassified the time periods for regression analysis. It found that when focusing on the four years before and after 2007, the variable coefficient was significantly positive, but after five years, the effect became insignificant. This partially demonstrates the short-term incentive effects of litigation fee reform. Fourth, the coefficient for the 2014 legal reform dummy variable is significantly positive, with the civil litigation rate increasing by 0.225%. This suggests that the comprehensive promotion of the rule of law has been effective, enhancing citizens’ legal awareness and judicial trust, thereby strengthening their willingness to litigate.

In terms of social structure, the three factors show differing relationships with civil litigation rates. First, the coefficient for education level is significantly positive at the 1% level. For every one-year increase in average years of education, the civil litigation rate increases by 0.189%. This aligns with the hypothesis that higher education levels strengthen disputants’ willingness to litigate. Second, the coefficient for income disparity is not significant, suggesting that urban-rural income disparity is unrelated to litigation willingness in China. Third, the coefficient for urbanization rate is significantly negative in the long term, contrary to conventional expectations. One explanation is that during urbanization, city governance levels improve alongside, leading to better social management and reduced litigation rates. Another explanation is that China’s urbanization process differs from that of developed Western countries. It has not fully transitioned to a “society of strangers” but may still resemble the “grid-based society of acquaintances” described by Su Li. Although more people reside in cities, they still place high value on “face” and “relationships,” making them averse to litigation.

3.3 Moderating Effects of the Curve Relationship

Further analysis is required to provide a conclusive response regarding the moderating effects on the curve relationship. However, before addressing this, it is necessary to examine regional heterogeneity to highlight the significance of discovering interaction effects. Table 4 reports the fixed effects regression results of the interaction effects.

Columns (1) and (2) present the regional regression results for the eastern region and the central-western regions. The results show that the first-order and second-order terms of real GDP per capita are both significantly positive and negative at the 1% level, respectively, indicating that the curve relationship between civil litigation rates and economic development holds true for both the eastern and central-western regions. These results are robust.

In column (3), the interaction terms between regional dummy variables and the first-order and second-order terms of real GDP per capita are significantly negative and positive at the 1% level, respectively. This suggests that there are regional differences in the curve relationship: the curve in the central-western regions is flatter, with the inflection point occurring further to the right compared to the eastern region. In other words, the positive impact of economic development on civil litigation rates is more enduring in the central-western regions, though its initial impact is weaker than in the eastern region.

One possible explanation is that the eastern region, as the pioneer of reform, opening-up, and economic development, encountered many new issues earlier, thereby establishing normative “precedents” and contributing to the formulation of related laws. These exemplary experiences alleviated normative uncertainty during subsequent development in the central-western regions, thereby reducing the intensity of the initial impact of economic development. Furthermore, the central-western regions have long lagged behind the eastern region in terms of socioeconomic development, which results in lower litigation opportunity costs, imperfect market institutions, and lower education levels--factors that may encourage litigation and prolong the impact of economic development. In this sense, differences in spatial and social attributes across regions affect the curve relationship.

Table 4: Fixed Effects Regression Analysis of Interaction Effects on Influencing Factors (1997–2020)

The regression results in column (4) of Table 4 illustrate that marketization has a moderating effect on the curve relationship. The first-order and second-order terms of real GDP per capita are significantly positive and negative at the 1% level, respectively, while the interaction terms between the proportion of foreign investment and the first-order and second-order terms are positive and negative, respectively, and significant at the 5% level. This indicates that the inverted U-shaped curve relationship between civil litigation rates and economic development is moderated by the degree of marketization.

Figure 7 depicts this moderating effect based on the regression coefficients. As shown in the figure, holding other conditions constant, higher levels of marketization lower the curve between economic development and civil litigation rates, making it steeper and more pronounced, with the inflection point shifting to the left. This means the curve reaches its turning point more quickly, consistent with the theoretical hypothesis that marketization weakens the curve relationship.

However, this weakening effect applies to the overall trend. If examined in stages, a steeper curve implies that in the early stages of economic development, the positive impact of the economy on litigation is stronger--each unit of economic growth leads to a larger increase in civil litigation rates. Conversely, in the later stages of economic development, the positive impact diminishes more rapidly and eventually disappears.

The regression results in column (5) of Table 4 reveal that education has a moderating effect on the curve relationship, opposite to that of marketization. Education reinforces the curve relationship between civil litigation rates and economic development. This can be seen in the regression results: the first-order and second-order terms of real GDP per capita are significantly positive and negative at the 1% level, while their interaction terms with education level are significantly negative at the 1% level and significantly positive at the 5% level, respectively.

This indicates that, controlling for other variables, higher education levels flatten the inverted U-shaped curve, with the inflection point shifting progressively to the right. Moreover, when the average years of education exceed 7.25 years, the inverted U-shaped curve changes into a monotonically increasing curve. When the average years of education exceed 8.69 years, the curve transitions into a positive U-shaped curve.

Figure 7: The Moderating Effect of Marketization on the Curve Relationship

Figure 8 illustrates that, within the actual sample range, the shape of the curve relationship has significantly changed at different levels of education.

When education levels are low (average years of education below 7.25 years), the curve relationship gradually flattens as education levels increase, approaching a straight line. When education levels are high (average years of education above 7.25 years), the curve relationship progressively transitions into a positive U-shaped curve as education levels rise. The curve is effectively situated in the latter, upward-sloping segment, where civil litigation rates continuously increase and accelerate with economic development, showing no signs of slowing or declining. Therefore, the reinforcing effect of education on the curve relationship is demonstrated by the fact that higher education levels lead to a progressively stronger positive impact of economic development on civil litigation rates, rather than a weakening effect.

Figure 8: The Moderating Effect of Education Level on the Curve Relationship

3.4 Robustness Tests

This study applies modifications to models, variables, and samples to enhance the robustness of its findings.

First, to address potential endogeneity and autocorrelation issues, multiple models are alternately used to test the curve relationship.

In the first model, although the study already uses the lagged one-period real GDP per capita to address reverse causality, given the possibility of autocorrelation in real GDP per capita, the lagged variable’s effectiveness in dealing with endogeneity might be limited. Therefore, the study uses river density as an instrumental variable for a two-stage least squares (2SLS) regression. River density satisfies the relevance and exclusivity requirements of an instrumental variable. On the one hand, production and business activities require substantial water resources, and regions with high river density facilitate industrial clustering and expansion, thus promoting economic development. On the other hand, river density is a natural factor that does not directly influence litigation willingness and is unrelated to civil litigation rates. Since river density does not change over time, fixed effects are not controlled in this model.

In the second model, considering that panel data may exhibit autocorrelation, the study introduces first-order autocorrelation (fixed effects linear model with AR(1) disturbance) into the error term of the fixed effects model for regression analysis.

In the third model, as citizens’ litigation willingness may be influenced by past societal litigation conditions, leading to endogeneity due to omitted variables, the study introduces the first-order lag of the civil litigation rate as an explanatory variable. The dataset is treated as dynamic panel data, and higher-order lags of the civil litigation rate are used as instrumental variables, applying the difference generalized method of moments (DIF-GMM) for regression analysis.

In the fourth model, since DIF-GMM estimators are prone to biases caused by weak instrumental variables, the study further adopts the system generalized method of moments (SYS-GMM). Higher-order lags of the civil litigation rate and real GDP per capita are used as instrumental variables to more effectively address endogeneity issues. The SYS-GMM model satisfies the test conditions of significant first-order autocorrelation and insignificant second-order autocorrelation in the residuals.

Table 5 reports the regression results for the instrumental variable two-stage least squares method, the fixed effects model with first-order autocorrelation, the difference GMM, and the system GMM. Across all results, the core explanatory variables--the first-order and second-order terms of real GDP per capita--are consistently significantly positive and negative at the 1% level, respectively. These findings align with the results in Table 3, indicating that the inverted U-shaped curve relationship between civil litigation rates and economic development is robust.

Table 5: Multiple Model Tests of the Curve Relationship

Additionally, in testing the moderating effects, considering the potential correlation between economic development and the levels of marketization and education, this study follows common practices by lagging the moderating variables by two periods for fixed effects regression to eliminate endogeneity effects.

The coefficients of the interaction terms in Table 6 are significant and consistent in sign (positive or negative) with the earlier results, indicating that the conclusions about the moderating effects on the curve relationship are robust.

Table 6: Fixed Effects Regression Analysis with Moderating Variables Lagged by Two Periods

Secondly, this study replaces real GDP per capita with the value-added of the tertiary industry as the core explanatory variable for regression analysis. The value-added of the tertiary industry reflects the transition in economic development models and is more relevant to measuring the validity of the curve theory in terms of normative uncertainty.

Column (1) of Table 7 shows that the first-order and second-order terms of the value-added of the tertiary industry are significantly positive and negative at the 1% level, respectively, indicating the validity of the curve relationship. Moreover, in columns (2) and (3), the interaction terms between the first-order and second-order terms of the tertiary industry value-added and the proportion of foreign investment are significantly positive and negative, respectively. Similarly, the interaction terms with education level are significantly negative and positive, respectively. These findings further confirm that the degree of marketization weakens, while education level strengthens, the curve relationship.

Table 7: Fixed Effects Regression Analysis of the Curve Relationship with Variable Replacement and Sample Adjustments

Finally, to eliminate the interference of outliers, this study conducts regression analysis on a sample excluding data from Beijing. The results reported in columns (4), (5), and (6) of Table 7 are consistent with the corresponding results in Tables 3 and 4.

These findings further strengthen the robustness of the conclusions regarding the curve relationship and its moderating effects, providing stronger support for the curve theory based on multi-layered conditions.

5. Research Conclusions and Policy Implications

Overall, this study utilizes Chinese litigation data and quantitative methods to effectively revise the traditional curve theory based on single conditions. It innovatively proposes a new curve theory based on multi-layered conditions, particularly identifying the existence of interaction mechanisms. Specifically, the curve relationship between China’s civil litigation rate and economic development is influenced by factors at the institutional environment and social structure levels.

Empirical research shows that changes in China’s civil litigation rate exhibit significant temporal and spatial heterogeneity. Temporally, the civil litigation rate declined gradually from 1997 to 2006, then began to rise rapidly from 2007 onward, demonstrating a “decline-then-rise” temporal characteristic. Spatially, the civil litigation rate shows a “east-west divide” regional pattern, with rates higher in the eastern region than in the central and western regions. However, in recent years, the growth rate in the eastern region has been slower than in the central and western regions.

In terms of influencing mechanisms, economic development affects litigation willingness through two channels: normative uncertainty and litigation benefits. This is reflected in the inverted U-shaped curve relationship between the civil litigation rate and economic development. Within the range of China’s sample, the relationship is situated in the “outward convex arc” segment at the early stage of the curve, showing a trend where the growth in civil litigation rates is slowing and moving closer to the inflection point.

Additionally, factors such as education level, availability of legal services, litigation fee reforms, and trial capacity construction positively influence the civil litigation rate. Conversely, marketization and urbanization levels are negatively correlated with civil litigation rates. These institutional and social factors alter the civil litigation rate by affecting disputants’ subjective recognition and rational calculation.

More importantly, marketization and education, belonging to the institutional environment and social structure levels respectively, play moderating roles. Marketization weakens the curve relationship, while education strengthens it. The discovery of these interaction mechanisms validates the rationality of the new curve theory and, to some extent, reveals the structural foundation of the influencing mechanisms for civil litigation rates.

Although this study delves deeply into the mechanisms behind changes in civil litigation rates through the lens of economic factors, it does not overlook their complexity. Future research can further advance in the following directions:

First, from a spatial perspective, explain the regional differences in civil litigation rates, identify the dominant factors in each region, and categorize them to more precisely guide the allocation of judicial resources. Second, differentiate case types and examine the formation mechanisms of disputes in areas such as marriage and family, tort, contracts, and intellectual property. Third, explore the impact of cultural morality and ideology on civil litigation rates beyond economic, institutional, and social factors.

In addition to theoretical dialogue, this study is motivated by practical concerns about modernizing China’s court trial system and trial capacity. The “many cases, few judges” issue essentially reflects an imbalance in the supply and demand of court litigation services. Addressing this problem generally requires actions on both the supply and demand sides.

On the demand side, litigation procedure reforms, such as the 2021 amendment to the Civil Procedure Law, which expanded the scope of single-judge trials and increased the thresholds for small claims, can reduce the workload of judges. However, such reforms may incentivize more litigation and risk undermining the fundamental values of justice.

In contrast, reducing litigation volume through social governance approaches, such as litigation source governance and multi-tiered dispute resolution mechanisms, helps establish a systematic dispute resolution framework. This approach effectively resolves most disputes before litigation, curbs unreasonable or improper lawsuits, and ensures the orderly conduct of regular court trials.

On the supply side, the most straightforward strategy is to increase the number of judges to cope with the rising volume of litigation. However, resources are limited, and the number of judges cannot grow indefinitely. This strategy only addresses immediate needs. Comparatively, improving case-handling efficiency through management and organizational optimization is more economically rational and yields longer-term benefits. Therefore, enhancing social governance and increasing managerial efficiency are fundamental and value-oriented reform methods for improving the judicial capacity of courts.

Regarding the two fundamental reform approaches, the findings of this study can enrich related policy proposals: First, on the supply-side management for efficiency improvement, there has been considerable discussion about dynamically optimizing resource allocation within courts based on demand. However, further exploration is warranted, as concerns about “overcapacity” are currently unnecessary. Empirical results suggest that economic development in China continues to drive the growth of civil litigation rates, albeit with gradually weakening intensity. Nevertheless, this positive influence is expected to persist for a long time. According to statistical models, the inflection point of the curve will occur when real GDP per capita reaches over 560,000 yuan. As of 2020, Shanghai’s real GDP per capita--the highest in the country--was just over 120,000 yuan, still more than four times below the inflection point. Therefore, as China’s economy continues to grow, civil litigation rates are likely to keep increasing in the near future, requiring proactive measures to address this trend.

Second, recognizing the importance of dynamic regional adjustments is crucial. In China’s unitary system, courts do not independently obtain personnel, financial, or material resources but rely on allocation by higher authorities. If a single court or region’s courts receive insufficient resources, no amount of internal optimization or enhanced management can compensate for the lack of foundational support. Since 2016, the nationwide provincial-level unified management reform for court personnel, finances, and materials has enabled feasible resource allocation at the provincial level. Some local courts have also demonstrated through empirical research the necessity of differentiated approaches to address operational conditions and resource needs within provinces.

While resource allocation within provinces is vital, the judiciary falls under central jurisdiction. To promote regional equitable development, the central government should assume the responsibility of nationwide management and dynamic adjustment of judicial resources. Empirical research has already shown that the central and western inland regions generally lag behind the eastern coastal regions in development, with relatively scarce judicial resources. Yet, civil litigation rates in these regions have risen rapidly over the past decade. If this trend continues, the east-west judicial resource supply-demand gap may further widen.

To address resource misallocation, the regional dynamic allocation of judicial resources should be based on litigation demand forecasting and tailored to the influencing mechanisms of different regions. On one hand, constructing demand forecasting models requires incorporating more diverse and detailed socioeconomic factors, such as population, economy, institutions, and education. The selection of these variables should be gradually refined through empirical research and interdisciplinary collaboration. Furthermore, building a unified judicial big data system is indispensable for data collection and result presentation. Achieving interconnectivity among court information systems, real-time data monitoring, and integrating socioeconomic databases from various sectors into a centralized platform can help create accurate and responsive forecasting models.

On the other hand, litigation demand should be categorized into different regional types, and tailored forecasting models should be designed accordingly. While this study focuses on general and regular explanatory theories, the empirical results reveal spatial heterogeneity and interaction mechanisms, suggesting that forecasting models cannot adopt a “one-size-fits-all” approach. The types of variables and their weights in the models should differ across regions. Additionally, regional classifications should evolve over time. Only through such differentiation can judicial resource allocation become more efficient.

On the demand-side structural reform, focusing solely on courts and judges’ roles in social governance is insufficient. To highlight the state’s critical role and potential in guiding societal litigation behavior, this study specifically examines the impact of institutional and structural factors on civil litigation rates. The significant influence of these factors is supported by empirical results. Thus, the judiciary’s supply-demand imbalance should be addressed as a problem of collaborative social dispute governance, rather than one exclusively solved by courts.

From the empirical results, three macro-policy reform directions emerge: First, further enhance marketization, reduce improper government intervention, leverage regional industrial advantages, and establish a healthy business environment. Marketization weakens the curve relationship, suggesting that improved autonomous market order can effectively facilitate timely dispute resolution. Efforts should focus on refining industry-specific self-governance norms and establishing robust communication and coordination mechanisms among multiple stakeholders in market disputes. Second, strengthen legal education with an emphasis on fostering compliance awareness and a comprehensive understanding of rights and obligations. Education increases litigation willingness and reinforces the curve relationship, indicating that China’s educational system has effectively improved citizens’ legal competence and rights awareness. However, it also reveals a lack of obligation awareness among some individuals, who may exploit legal systems for improper gains, undermining social justice. Therefore, legal education should emphasize the balance between rights and responsibilities. Third, enhance lawyer professional protections and ethics, and establish mechanisms for lawyer involvement in pre-litigation dispute resolution. The positive correlation between the availability of legal services and civil litigation rates suggests that improved access to legal services enhances citizens’ ability to “access justice.” However, caution is needed to address the self-serving tendencies of the legal profession. A minority of lawyers may deliberately complicate disputes or solicit cases to occupy judicial resources and harm litigants’ interests for personal gain. Specific measures, such as government subsidies or relaxed fee standards for mediation services, can incentivize lawyers to actively intervene and resolve disputes before litigation.

By implementing these measures, the state can reduce social disputes, alleviate courts’ case-handling pressure, and ultimately achieve a balanced supply-demand dynamic in judicial resources, even as citizens’ living standards continue to improve.