Water conservancy and farmers' income: A quasi-natural experiment based on the Three Gorges project

Water conservancy and farmers' income: A quasi-natural experiment based on the Three Gorges project | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Water conservancy and farmers' income: A quasi-natural experiment based on the Three Gorges project Bin Xu, Ke Wei, Zhiyang Shen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8882775/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Globally, large-scale water conservancy projects are widely used as a policy instrument to enhance climate resilience, stabilize agricultural production, and promote inclusive growth, yet credible evidence on their income and distributional effects in developing economies remains limited. This paper studies whether large-scale water infrastructure can contribute to these outcomes by examining the Three Gorges Project in China. We use a balanced county-level panel of 351 counties in the middle and lower Yangtze River basin from 2003 to 2020 and estimate a difference in differences design that exploits variation in exposure to the project’s flood control benefits over time. We find that the flood control function of the Three Gorges Project increases rural residents’ per capita disposable income, and the result is robust to a range of alternative specifications and placebo tests. Mechanism evidence suggests three channels: higher agricultural productivity, stronger household entrepreneurship and nonfarm business activity, and greater local investment. The income effect is larger in Central China and in counties with higher flood risk. We also find that the project reduces the urban rural income gap, consistent with water infrastructure supporting shared prosperity. Water Conservancy Project Three Gorges Project Farmers' income Common prosperity Figures Figure 1 Figure 2 Figure 3 1. Introduction At the global level, increasing climate variability and the rising incidence of floods have renewed interest in large-scale water infrastructure as a key policy tool for safeguarding agricultural production and rural livelihoods, particularly in developing and emerging economies. International organizations and governments have emphasized that effective flood control and water management are essential not only for disaster risk reduction, but also for promoting inclusive growth and preventing climate-induced poverty traps in rural areas. Recent empirical evidence on climate shocks shows that environmental shocks not only disrupt agricultural production but also alter farmers’ economic behavior and welfare dynamics. For example, Liebenehm et al. (2023) find that rainfall shocks increase individual risk aversion among rural households in Southeast Asia, particularly for net food buyers, implying that climate variability can shape farmers’ economic decisions and potentially perpetuate poverty in the absence of adequate credit and insurance. To develop rural reforms and advance comprehensive rural revitalization, the 2025 No.1 Central Document of Chinese government highlights the need to strengthen capacity for agricultural disaster prevention and mitigation and calls for a modern system for flood control and disaster reduction. Farmers’ income has long been at the center of the “three rural areas” agenda. A large body of research studies the drivers of income growth, with recent work emphasizing digital empowerment (Bowen & Morris, 2019 ; Mushi et al., 2022 ; Guo et al., 2023 ), policy support (Bosheng & Xiaoyang, 2021 ; Junqian & Xingmin, 2023 ), and farmers’ own initiatives (Yang et al., 2023 ). By contrast, relatively little attention has been given to hydraulic engineering and flood control infrastructure, despite their direct link to agricultural risk and rural livelihoods. This gap matters because extreme hydrological events are becoming more frequent and more damaging. Since the Industrial Revolution, technological progress and economic growth have been accompanied by large greenhouse gas emissions. The interaction of global warming and human activities has altered the water cycle across space and time, raising the likelihood of extreme events and affecting regional water security and national development strategies over the medium and long term. These risks are closely tied to the food system and to rural welfare. Floods threaten farmers’ lives and health and destroy productive assets. They reduce income, worsen quality of life, and can undermine mental health, sometimes pushing households into poverty or back into poverty. Koks et al. ( 2015 ) emphasize that flood risk outcomes depend not only on physical hazards and exposure, but also on social vulnerability, which is particularly relevant for farming households with limited adaptive capacity. When shocks are large and insurance and credit are limited, many households have little capacity to absorb losses. Historical experience warns us that the rise and fall of water infrastructure extends far beyond the projects themselves, profoundly impacting regional economic vitality and social stability. Cao & Chen ( 2022 ) revealed in their seminal study of the Grand Canal that the abandonment of trade routes can trigger social conflicts due to lost opportunities. Conversely, this demonstrates that constructing and maintaining inclusive water projects may serve as a fundamental stabilizer for safeguarding livelihoods and promoting shared development. Understanding how to respond to extreme hydrological events, raise farmers’ income, and reducing the risk of returning to poverty is therefore important for the goal of common prosperity. Water conservancy is often described as the lifeblood of agriculture and a foundation of the national economy. As Sun Yat-sen wrote in 1918, constructing slices and weirs can regulate water flow for navigation while also harnessing hydraulic power. The Three Gorges Project, the world’s largest hydro-junction, is a core infrastructure investment in the Yangtze River Basin. Its main benefits fall into three areas, flood control and disaster mitigation, power generation, and navigation, with flood control widely viewed as the most important. Batista and Firme ( 2025 ) examine the Mariana dam disaster in Brazil and find substantial and spatially uneven economic losses across neighborhoods, with pronounced impacts on agriculture and local economic activity. Their results underscore how water-related infrastructure events can directly affect rural incomes through production and environmental channels. Since 2007, the project’s flood control function has mitigated flood risks in the middle and lower Yangtze River regions, improved safety in vulnerable areas, and eased flood pressure in the lower reaches of the Jingjiang River. It has also been linked to addressing challenges that historically followed major floods, including environmental damage and post-flood epidemics. More broadly, effective disaster reduction can limit economic losses, stabilize production, and support wealth accumulation by reducing disruptions to social reproduction. So far, most research on water conservancy projects has emphasized ecological impacts, while much less attention has been paid to farmers’ income outcomes. In particular, few studies have systematically examined the link between water conservancy project construction and farmers’ income (Shi et al., 2002 ). Following Haishan & Yang ( 2022 ), this paper treats 2007, when the flood control function of the Three Gorges Project officially came into play, as the policy implementation node and applies a DID design to estimate the income effect of flood risk reduction. Using balanced panel data for 351 counties in the middle and lower Yangtze River basin over 2003 to 2020, we find that the flood control function of the Three Gorges Project significantly increases farmers’ per capita disposable income in flood-prone counties by about 0.054 percentage points. The estimate is stable across a set of sensitivity and robustness checks. Mechanism results suggest three pathways: improved agricultural productivity, stronger entrepreneurship, and greater investment inflows. We further document heterogeneous effects across county types. This paper contributes to literature in three ways. First, by leveraging the Three Gorges Project as a quasi-natural experiment and a balanced county panel, we provide causal evidence on how a major water conservancy project affects farmers’ income, and we unpack micro-level channels related to agricultural capacity, entrepreneurial activity, and investment attraction. Second, we quantitatively connect the project’s flood control function to both income growth and the urban-rural income gap, helping fill a gap at the intersection of water conservancy and rural economics. Third, the findings shed light on how large-scale water infrastructure can support regional coordination and shared prosperity, and they offer empirical support for policies that aim to translate water security and disaster mitigation into sustained rural income gains. 2. Theoretical analysis and literature review 2.1 Background Floods are among the most frequent, geographically extensive, and economically damaging natural hazards in China. These disasters were concentrated in the middle and lower Yangtze River basin, as well as parts of North China and Northeast China. The Yangtze River basin alone accounts for more than one fifth of China’s land area, and its combination of complex terrain and high precipitation has long made it a flood-prone region. Figure 1 shows that flood-vulnerable counties in the middle and lower reaches are largely distributed along the main river corridor, excluding municipalities. Figure 1 about here The Three Gorges Project was proposed as early as 1918, but for decades the technical and fiscal conditions for construction were not in place. With the economic expansion following reform and opening up, these constraints eased and project preparation accelerated. In July 1993, the State Council’s Three Gorges Construction Committee approved the preliminary design report and initiated technical design work for key components. Construction began in 1994 in Sandouping, Yichang, Hubei Province, and the river was successfully diverted in 1997. The Flood Control Law promulgated the same year strengthened the legal basis for flood control planning, project construction, and operational management, thereby supporting the institutional environment in which the project could function as part of the national flood control system. In July 2007, the project began to deliver flood control benefits in practice, with an initial reported peak clipping of 3,000 cubic meters per second. As a central component of Yangtze River governance, the project is widely viewed as a major investment in flood risk reduction. Yet the relationship between flood shocks and rural poverty suggests that disaster mitigation alone is not the endpoint. The broader policy question is whether large-scale flood control can also support sustained income growth and common prosperity by reducing risk and enabling rural development. The core of the “three rural areas” agenda ultimately centers on farmers’ livelihoods. A durable solution requires continued growth in farmers’ income and a narrowing of income gaps between urban and rural areas and across regions. Rural poverty remains persistent and, for some households, recurrent, reflecting the interaction of multiple constraints such as adverse natural conditions and a weak local economic base. Natural disasters, especially floods and droughts, can reinforce these constraints by destroying assets, interrupting production, and destabilizing earnings. When livelihoods are fragile and risk-coping capacity is limited, shocks can raise the likelihood of falling into poverty again, including in some eastern areas and major grain-producing regions. Even after the achievement of the goal of building a moderately prosperous society in all respects, consolidating poverty alleviation gains, raising farmers’ incomes, especially for low-income households, and narrowing the urban-rural gap are likely to remain central tasks of agricultural and rural development for the foreseeable future. 2.2 The impact of water conservancy projects on farmers’ income Water conservancy projects can shape farmers’ income through macro or micro channels. Kates et al. ( 2001 ) argue that disasters originate from the interaction between society and nature and may disrupt regional sustainable development. Dell et al. ( 2014 ) review the growing climate-economy literature and show that variations in temperature, precipitation, and extreme weather events have significant effects on agricultural productivity, income, and long-run economic growth, particularly in developing regions where farmers are highly dependent on water availability. Flood shocks can generate sudden shortages of resources, damage infrastructure, induce population displacement, and reduce ecosystem services. These effects may also trigger secondary disasters and broader social problems, which can weaken long-run development plans and sustainability goals. From a macro perspective, floods can affect economic performance through changes in post-disaster investment returns, human capital accumulation, and the pace and direction of technological progress. In flood-prone regions, floods impose direct losses on property, agriculture, industry, and transport, and they also create latent risks that depress investment and amplify uncertainty. Leiter et al. ( 2009 ) use a DID design to study European firms and find that exposure to flood risk reduces investor willingness in affected areas. Flood risk can also force local governments to devote substantial fiscal resources to flood control institutions and infrastructure, which can crowd out other productive spending. Even when regions have geographic advantages and strong market access, persistent flood risk can restrain growth by raising operating costs and discouraging long-horizon investment. At the household level, floods affect income and welfare through two main pathways. First, they cause direct losses of life and property, destroy housing, and damage productive assets. Second, they shift risk expectations and change consumption and saving behavior. When households suffer direct losses, they often cut non-essential consumption while increasing spending on relief and reconstruction, which reshapes both the consumption bundle and savings (Yan, 2025 ). In anticipation of future shocks, households may also raise precautionary savings and reduce discretionary consumption, consistent with precautionary savings theory. These behavioral responses matter for income dynamics because they influence labor supply choices, investment in farm inputs, and the capacity to finance recovery. At the meso level of the rural economy, flood impacts are especially visible in agricultural production and the destruction of basic means of production and livelihood. Floods inundate farmland and damage crops, reducing yields and lowering farm income. At the same time, damage to houses, farm tools, and other productive equipment increases the vulnerability of rural households. Kocornik et al. (2020) show that economic activity often remains concentrated in flood-prone areas, suggesting that farmers and rural communities may continue to rely on water-intensive locations despite persistent risks. Housing and farm implements are core assets for farm families and are often the most immediate victims of flooding. When these assets are destroyed, agricultural production may not return to normal quickly, reducing output in the current season and potentially affecting the next production cycle (Alemayehu, et al., 2025 ). Interrupted agricultural reproduction not only lowers income but can also raise production costs through replanting, repairs, and the need to replace damaged inputs, which further tightens household budgets. In this way, natural disasters can increase rural poverty incidence and deepen poverty persistence, limiting rural development and complicating the objective of common prosperity. Yuan et al. ( 2025 ) indicate that repeated exposure to shocks discourages farmers from adopting green production technologies due to heightened risk aversion, which in turn may influence income dynamics and technological diffusion in rural areas Disaster economics further suggests that disasters are fundamentally economic events, in the sense that they threaten welfare, destroy wealth, and weaken the foundations of socioeconomic sustainability. From this perspective, water conservancy projects can reduce flood damages and, at the same time, support rural development by improving the local production environment (Liu et al., 2025 ). On one hand, flood control infrastructure stabilizes the natural environment in flood-prone areas and protects lives and property. Lower risk can free fiscal capacity for productive public spending, which may support a reinforcing process of risk reduction and capital accumulation. Water conservancy projects also tend to be bundled with complementary infrastructure such as roads and bridges, improving market access and the local investment environment. Better connectivity can attract additional public and private investment, allowing resilience improvements to translate into economic gains. On the other hand, water conservancy projects can reduce flood-related crop losses and strengthen drought resistance through reservoirs, water regulation and storage systems, and ponds. By storing water during wet periods and providing irrigation during dry periods, these projects improve the timing and allocation of water resources, stabilize irrigation, protect yields, and reduce volatility in farmers’ operating income. The public goods nature of infrastructure implies potential spatial spillovers, while improvements in water management can promote adaptation to climate risk. Together, these mechanisms can raise incomes, reduce vulnerability, and help prevent poverty recurrence in flood-prone rural areas. The mechanism section develops these channels in detail. Based on this framework, we propose the following hypotheses: Hypothesis 1 The Three Gorges Project significantly increases farmers’ disposable income. Hypothesis 2 The Three Gorges Project drives income growth through three channels: enhancing agricultural productivity, stimulating entrepreneurship, and attracting investments. 3. Research design 3.1 Measurement model construction Since the Three Gorges Project can be regarded as a quasi-natural experiment, this paper uses the difference-in-differences (DID) to estimate the impact of water conservancy projects on farmers' income growth, treating counties in flood-prone areas as the treatment group and other counties as the control group, while establishing a baseline regression model in the following form: $$\:{Y}_{it}={\alpha\:}_{0}+{\beta\:}_{0}{did}_{it}+\gamma\:{X}_{it}+{\mu\:}_{i}+{\lambda\:}_{t}+{\epsilon\:}_{it}$$ 1 In formula (1), \(\:{Y}_{it}\) is the explained variable, representing the logarithmic value of the per capita disposable income of farmers in the county in year t. \(\:{\text{d}\text{i}\text{d}}_{\text{i}\text{t}}\) represents the interactive term, specifically, \(\:{\text{d}\text{i}\text{d}}_{\text{i}\text{t}}={\text{T}\text{r}\text{e}\text{a}\text{t}}_{\text{i}}\times\:{\text{P}\text{o}\text{s}\text{t}}_{\text{t}}\) , is a dummy variable equal to 1 if county i is located in a flood-prone area in the middle and lower reaches of the Yangtze River, and 0 otherwise. \(\:{\text{P}\text{o}\text{s}\text{t}}_{\text{t}}\) is a time dummy indicating the period in which the flood control function of the Three Gorges Project is in effect: \(\:{\text{P}\text{o}\text{s}\text{t}}_{\text{t}}\) =0 before the function takes effect, and \(\:{\text{P}\text{o}\text{s}\text{t}}_{\text{t}}\) =1 during and after its implementation. \(\:{\text{X}}_{\text{i}\text{t}}\) denotes a vector of control variables that may affect farmers’ income at the county level. \(\:{{\mu\:}}_{\text{i}}\) represents county fixed effects, which control for time-invariant unobserved heterogeneity across counties, while \(\:{{\lambda\:}}_{\text{t}}\) represents year fixed effects, capturing common shocks and macroeconomic trends over time. \(\:{{\epsilon\:}}_{\text{i}\text{t}}\) is the idiosyncratic error term. The coefficient of interest, \(\:{{\beta\:}}_{0}\) identifies the causal effect of the flood control function of the Three Gorges Project on farmers’ income growth. 3.2 Measurement and explanation of variables This paper uses farmers' disposable income from 2003 to 2020 to measure income growth. Based on whether counties are located in flood-prone areas in the middle and lower reaches of the Yangtze River, the sample is divided into a treatment group and a control group, with 133 counties classified as the treatment group. With reference to Haishan ( 2022 ), the year 2007 is taken as the “policy implementation node”, marking the point when the Three Gorges Project began to exert its flood control function. (1) Explained variables. Consistent with the existing literature, the dependent variable is measured as the logarithm of farmers’ per capita disposable income in each county. (2) Core explanatory variable. The key explanatory variable is the interaction term \(\:{\text{d}\text{i}\text{d}}_{\text{i}\text{t}}\) , which captures counties located in flood-prone areas during the period when the flood control function of the Three Gorges Project is in effect. (3) Control variables. The control variables include government size (Gov), measured as the ratio of local government general budget expenditure to GDP; financial development (Fin), measured by the ratio of outstanding loans from financial institutions to GDP at year-end; human capital (Hc), measured as the logarithm of the number of students per 10,000 residents; and regional economic development (Gdp), proxied by the annual average nighttime light intensity. 3.3 Data sources and descriptive statistics In this paper, the middle and lower reaches of the Yangtze river five provinces (excluding municipality directly under the central government) 351 county-level administrative region (region, counties and county-level cities) annual panel data as the research object, the average value data from global night light data ( https://ngdc.noaa.gov/eog/download.html ); For flood- prone areas, refer to Atlas of Major Natural Disasters and Society of China (2004) and Name Code of Flood Storage Areas of China (2001); Other data are from China County Statistical Yearbook. The descriptive statistics of the main variables are shown in Table 1 . Table 1 Variable definitions and descriptive statistics Variable type Variables Variable definition Sample size Mean Standard deviation Minimum Maximum Explained variable eq Logarithm of per capita disposable income of rural residents 5580 8.76 0.73 7.06 10.56 Core explanatory variable did Whether it is located in a flood-prone area the interaction term of the grouping dummy variable and the policy time dummy variable 5580 0.17 0.38 0.00 1.00 Control variable Is Secondary industry /GDP 5580 0.43 0.12 0.11 0.89 Gov Fiscal expenditure /GDP 5580 0.18 0.10 0.00 0.99 Fin Financial institution loans at year-end /GDP 5580 0.52 0.24 0.04 2.39 Hc Logarithm of number of students in school per 10,000 people 5580 6.21 0.30 3.47 7.06 Gdp Lighting 5580 3.79 7.08 0.02 57.16 4. Empirical analysis 4.1 Baseline regression Table 2 reports the baseline estimates of the effect of the Three Gorges Project’s flood control function on farmers’ per capita disposable income in flood-prone counties in the middle and lower reaches of the Yangtze River, providing evidence in support of Hypothesis 1 . Column (1) presents estimates from a specification without fixed effects. Column (2) adds year fixed effects, while Column (3) further includes county fixed effects and the full set of control variables. Across all three specifications, the estimated coefficient on the Three Gorges Project policy variable is positive and statistically significant. This pattern indicates that, after 2007, counties exposed to the project’s flood control benefits experienced higher farmers’ disposable income relative to the comparison group. In the preferred specification in column (3), which controls observed covariates as well as county and time fixed effects, the coefficient on the policy variable is 0.054 and is significant at the 1 percent level. The estimate implies that the flood control function of the Three Gorges Project increased farmers’ disposable income in flood-prone counties by about 0.054 percentage points. Taken together, these results suggest that flood control infrastructure can raise rural incomes in exposed areas, consistent with Hypothesis 1 . Table 2 Benchmark regression results (1) (2) (3) eq eq eq did 0.308 *** 0.036 *** 0.054 *** (0.015) (0.007) (0.007) Is 1.300 *** 0.766 *** 0.302 *** (0.066) (0.039) (0.026) Gov 0.567 *** -1.928 *** 0.122 *** (0.097) (0.074) (0.040) Fin 0.428 *** 0.029 0.000 (0.047) (0.021) (0.011) Hc -0.903 *** -0.026 ** 0.078 *** (0.027) (0.013) (0.008) Gdp 0.033 *** 0.016 *** -0.007 *** (0.001) (0.001) (0.001) _cons 13.317 *** 8.869 *** 8.139 *** (0.180) (0.082) (0.050) N 5580 5580 5580 \(\:{R}^{2}\) 0.490 0.894 0.986 Individual fixation No No Yes Time fixation effect No Yes Yes 4.2 Parallel trend test We conduct an event-study test of the parallel trend’s assumption using 2003 as the base year. Figure 2 plots the estimated coefficients from Eq. (2). The coefficients for the pre-2007 years are small and statistically indistinguishable from zero, suggesting no differential pre-trends between the treated and control counties before the flood control function of the Three Gorges Project began to operate in 2007. In contrast, the post-2007 coefficients are positive and statistically significant, and their magnitude increases over time, indicating that the income effect grows as exposure accumulates. Taken together, these results support the parallel trends assumption and justify using 2007 as the policy timing when the project’s flood control function started to take effect. 4.3 Robustness test 4.3.1 Placebo Test To assess the reliability of the baseline estimates and address the concern that omitted factors could lead to spurious rejection of the null, we conduct a placebo test based on random reassignment. Specifically, we randomly draw a set of treated counties and randomly assign a policy timing, then re-estimate the baseline specification. We repeat this procedure 500 times to obtain the distribution of placebo coefficients. Figure 3 summarizes the results. The placebo estimates are tightly centered around zero, and virtually all simulated coefficients are smaller in magnitude than the baseline estimate of 0.054. This pattern suggests that the baseline effect is unlikely to be driven by chance assignment or unobserved shocks unrelated to the Three Gorges Project, and it supports the interpretation that the estimated impact reflects the project’s flood control benefits. 4.3.2 PSM-DID Because treatment status is defined by whether a county is flood-prone rather than randomly assigned, treated and control counties may differ along other observable dimensions that also affect farmers’ income. To address this concern, we use PSM to construct a comparison group that is more similar to the treated counties in terms of the control variables. We implement 1:1 nearest-neighbor matching and then re-estimate the baseline DID specification on the matched sample. Column (1) of Table 3 reports the results. The coefficient on did remains positive and statistically significant, consistent with the baseline estimates. This finding suggests that differences in observable characteristics and sample selection based on flood exposure are unlikely to drive our main result. Table 3 Robustness test (1) (2) (3) (4) (5) eq eq eq eq eq did 0.055 *** 0.051 *** 0.018 *** 0.054 *** 0.054*** (0.007) (0.007) (0.006) (0.014) (0.009) _cons 8.216 *** 8.152 *** 8.257 *** 8.139 *** - (0.052) (0.049) (0.051) (0.107) - N 5290 5580 5580 5580 5580 \(\:{R}^{2}\) 0.986 0.986 0.989 0.986 0.106 Control variables Yes Yes Yes Yes Yes Individual fixation effect Yes Yes Yes Yes Yes Time fixation effect Yes Yes Yes Yes Yes Province - Time fixed effect No No Yes No No 4.3.3 Adding Control Variables Regional income growth may depend on initial development conditions because regional development can feature increasing returns and cumulative causation. To account for this possibility, we augment the baseline specification by interacting the county’s average nighttime light intensity in 2003 with a linear time trend. This term allows counties with different initial development levels to follow different underlying growth paths. Column (2) of Table 3 reports the results. The interaction term is statistically significant, indicating that initial economic conditions are indeed correlated with subsequent income growth. Importantly, the estimated effect of the flood control function of the Three Gorges Project remains positive and statistically significant after this adjustment. This implies that the baseline result is not driven by differential growth related to initial development levels, and that the project’s flood control benefits still contribute to higher farmers’ income. 4.3.4 Controlling for the Influence of Macro Factors In the baseline specification, we include county fixed effects and time fixed effects to absorb time-invariant county characteristics and common shocks. To further address the concern that provinces may face different macroeconomic conditions over time, we add province-time fixed effects, which flexibly control for all province-level shocks in each year. Column (3) of Table 3 reports the results. The estimated coefficient on did remains positive and statistically significant, indicating that our main finding is not driven by province-specific time-varying factors. 4.3.5 Adjust the standard error We adjust standard errors in two ways to account for potential correlation in the error term. First, we use two-way clustering at the county level and the province-year level. Column (4) of Table 3 reports the corresponding estimates. Second, we report spatial HAC standard errors that allow for spatial correlation within a 50 km radius and first-order serial correlation over time. Column (5) of Table 3 presents these results. Under both approaches, the estimated coefficient on did remains statistically significant, indicating that our main results are robust to alternative assumptions about heteroskedasticity, serial correlation, and spatial dependence. 4.3.6 Excluding the Influence of Other Policies Because counties in our sample may have been exposed to other major policies during the same period, policy overlap could bias the baseline estimates. We therefore control a set of contemporaneous policies that are likely to affect farmers’ income, and we examine whether the estimated impact of the Three Gorges Project changes once these policies are taken into account. Specifically, we sequentially add indicators for high-speed rail expansion, the rollout of e-commerce into rural areas, national poverty alleviation programs, targeted poverty alleviation policies, and incentive policies for major grain-producing counties. These policies can influence rural incomes through several channels. High-speed rail expansion may raise income by easing labor mobility, supporting county-level growth, and generating spatial spillovers. Rural e-commerce programs can reduce information frictions and expand market access, allowing farmers to adjust production and sales decisions, and potentially strengthening local agglomeration effects as market scale increases. Poverty alleviation policies, including the designation of poor national-level counties and targeted programs, can increase fiscal resources through transfer payments, subsidized credit, and related support, which may improve infrastructure and encourage the development of specialized local industries. The incentive policy for major grain-producing counties can shift local government priorities toward agriculture, improve production conditions, and attract investment into the agricultural sector. To isolate the independent effect of the Three Gorges Project on farmers’ income growth in the middle and lower Yangtze River basin, we include dummy variables for these policies in the DID specification. Table 4 reports the estimates. Across all specifications, the coefficient on did remains positive and statistically significant, and its magnitude is close to the benchmark estimate. These results indicate that the main finding is not explained by concurrent policy interventions, and the estimated income effect of the Three Gorges Project’s flood control function is robust to controlling for policy overlap. Table 4 excludes other policies during the same period (1) (2) (3) (4) (5) (6) eq eq eq eq eq eq did 0.054 *** 0.054 *** 0.054 *** 0.055 *** 0.054 *** 0.056 *** (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) High-speed rail opened 0.003 0.006 (0.006) (0.006) E-commerce into rural areas 0.027 *** -0.002 (0.006) (0.005) National-level poverty- stricken county -0.085 *** -0.146 *** (0.008) (0.009) Targeted poverty reduction 0.088 *** 0.127 *** (0.006) (0.007) Large grain-producing -0.012 -0.002 (0.011) (0.011) _cons 8.140 *** 8.145 *** 8.197 *** 8.192 *** 8.141 *** 8.313 *** (0.050) (0.050) (0.049) (0.049) (0.050) (0.046) N 5593 5593 5593 5593 5593 5593 \(\:{R}^{2}\) 0.986 0.986 0.986 0.987 0.986 0.987 Control variables Yes Yes Yes Yes Yes Yes Individual fixation effect Yes Yes Yes Yes Yes Yes Time fixation effect Yes Yes Yes Yes Yes Yes 4.3.7 Endogeneity Analysis The baseline results show that the flood control function of the Three Gorges Project significantly raises farmers’ income in flood-prone counties in the middle and lower Yangtze River basin. Although we mitigate omitted-variable concerns by including county fixed effects, year fixed effects, and province-time fixed effects, Fig. 1 suggests that flood-prone counties are concentrated along the Yangtze River. These locations are likely to share geographic and economic features, such as lower elevation, flatter terrain, proximity to the river channel, better transport access, and shorter distance to major markets. If these advantages are correlated with both treatment status and income growth, they could confound the DID estimates. In addition, because treatment assignment is based on flood exposure rather than random allocation, sample selection may remain a concern. To further address these endogeneity issues, we implement an IV approach. We use the linear distance from each county to the Three Gorges Project as an instrument. Because distance is time-invariant, we follow Nunn and Qian ( 2014 ) and interact distance with a time variable to generate identifying variation in a panel setting. The relevance condition is plausible. As Fig. 1 indicates, counties closer to the Three Gorges Project are more likely to be in flood-prone areas and to have larger exposed areas. Moreover, the project is located near the transition between China’s second and third topographic steps, where the river gradient is large and flows are rapid, making flood control benefits more likely to propagate to downstream locations. These features imply that proximity increases the probability and intensity of being affected by the project’s flood control function. The exclusion restriction is supported by the fact that distance to the Three Gorges Project is a predetermined geographic measure. Conditional on fixed effects and the included controls, it is unlikely to affect farmers’ income through channels other than exposure to the project’s flood control benefits. Table 5 reports the IV estimates. The first-stage results show a strong relationship between the instrument and treatment exposure. Standard diagnostics, including under-identification and weak-instrument tests, support the validity of the instrument. In the second stage, the estimated effect remains positive and statistically significant at the 1 percent level. Overall, the IV results reinforce the baseline conclusion that the flood control function of the Three Gorges Project increases farmers’ income. Table 5 Endogeneity analysis (1) (2) Phase I Results Phase 2 Results IV 0.056 *** (0.011) did 1.474 *** (0.281) First stage F number 27.58 Underrecognition test 29.08 [0.000] Weak instrumental variable test 27.58 {16.38} N 5513 Control variables Yes Individual fixation effect Yes Time fixation effect Yes Note :[] inside is the p-value of the LM statistic; {} inside is the critical value at the 10% level of the Stock-Yogo weak recognition test 5. Further analysis 5.1 Mechanism analysis We next examine the mechanisms through which the flood control function of water conservancy projects increases farmers’ income. We focus on three channels. First, flood control can strengthen agricultural production capacity. By reducing flood-related losses and stabilizing the production environment, water conservancy projects make agricultural production more predictable, which can facilitate mechanization, improve efficiency, and support larger-scale operations. In addition, by improving the timing and allocation of water resources, these projects can sustain irrigation during droughts, improve crop growing conditions, and raise output in the primary sector, thereby increasing farm operating income (Fabri et al., 2024 ). Second, flood control can encourage entrepreneurship. Lower disaster risk reduces income volatility and improves expectations about future returns, which can increase farmers’ willingness and ability to start businesses. Entrepreneurship can expand market access, raise the value added of agricultural products, and create local jobs, which supports poverty reduction and narrows the urban-rural income gap. Third, flood control can attract investment. By lowering disaster risk and improving complementary infrastructure such as roads and bridges, water conservancy projects can improve the local investment and financing environment and increase investors’ confidence in long-horizon projects (Luo & Fan, 2012). Risk reduction can also lower insurance costs for agriculture and infrastructure, which improves expected returns. In addition, water security can enable more diverse land uses, including tourism and ecological agriculture, which can increase the resilience of the local economy. Guided by these mechanisms, we test whether the income effect of the Three Gorges Project operates through agricultural production capacity, entrepreneurship, and investment. Agricultural production capacity is measured along two dimensions, the development level of agriculture and the scale of agricultural production. Using county-level data, we proxy agricultural development with per capita value added in the primary sector and agricultural scale with per capita grain sown area. We measure entrepreneurship using the relative level of entrepreneurial activity per 10,000 people, and we measure investment with the share of fixed asset investment in GDP. To avoid the limitations of conventional stepwise mediation tests, we follow Jiang ( 2022 ) and estimate the channel regressions directly. Table 6 reports the results. Table 6 Mechanism analysis of flood control (1) (2) (3) (4) Level of agricultural development Scale of agricultural production Promoting entrepreneurship Attracting investment did 0.040 *** 0.077 *** 0.091 ** 0.084 *** (0.011) (0.015) (0.038) (0.012) _cons 8.630 *** 1.126 *** 2.612 *** 0.373 *** (0.069) (0.110) (0.222) (0.097) N 5725 3860 5569 4762 \(\:{R}^{2}\) 0.955 0.796 0.731 0.799 Control variables Yes Yes Yes Yes Individual fixation effect Yes Yes Yes Yes Time fixation effect Yes Yes Yes Yes Column (1) of Table 6 shows that the flood control function of the Three Gorges Project significantly raises per capita value added in the primary sector at the 1 percent level, consistent with improved agricultural development. This result accords with the idea that risk reduction stabilizes production and helps protect grain output and farm operating income. Column (2) indicates that flood control also expands the scale of agricultural production, suggesting greater specialization and higher production efficiency. Column (3) examines entrepreneurship. The estimated effect is positive and highly significant, and it is the largest among the three channels. This pattern is consistent with farmers being more willing to start businesses when disaster risk falls and expected returns become more stable. Entrepreneurship can both expand market opportunities and increase the value added of agricultural products. Finally, column (4) shows that flood-prone counties became more attractive to capital after 2007, as measured by the share of fixed asset investment in GDP. This finding suggests that, once flood risk declines, the location advantages of counties along the Yangtze River can translate more effectively into investment inflows and more diversified land use. Overall, the results in Table 6 support the view that the Three Gorges Project increases farmers’ income through higher agricultural production capacity, stronger entrepreneurship, and greater investment. 5.2 Heterogeneity analysis 5.2.1 Eastern and central provinces Regional coordination is a core component of common prosperity, so we examine whether the income effect of the Three Gorges Project differs across regions. We split the county sample by location into eastern and central regions and re-estimate the baseline specification within each subsample. The results indicate meaningful heterogeneity. The estimated coefficient for the eastern region is negative, while the effect is larger and more positive in the central region. This pattern suggests that the flood control benefits of the project translate into stronger income gains in the central region than in the eastern region. A plausible interpretation is that baseline development conditions shape the marginal returns to flood risk reduction. Counties in the eastern region typically have higher economic concentration, stronger institutions, and better infrastructure, and farmers’ disposable income levels are already relatively high. As a result, the scope for additional income gains may be limited. In contrast, many counties in the central region start from lower income levels and face tighter infrastructure and risk-management constraints, leaving more room for improvement once flood risk declines. Consistent with this idea, the estimated effect of did tends to weaken as the initial income level rises, and the stability checks that add controls yield patterns that align with this interpretation. Overall, the heterogeneity results suggest that the project’s flood control function has a larger income effect in counties with slower initial development. In this sense, the Three Gorges Project appears to generate relatively greater gains in less developed areas, which is consistent with a reduction in regional disparities and with the goal of regional coordinated development under common prosperity. 5.2.2 Hubei Province and others Hubei Province is plausibly the largest beneficiary of the Three Gorges Project’s flood control function. Flood risk in the Jingjiang reach is widely viewed as especially severe because the river channel is highly curved and flood flows can be constrained, increasing the likelihood of dike failure. The north bank also borders the Jianghan Plain, which is low-lying and therefore highly exposed to inundation risk. Because the Three Gorges Project is located in Yichang, Hubei Province, it directly regulates upstream inflows into this reach and can reduce downstream flood pressure. Motivated by this geography, we split the sample into Hubei and all other provinces. Column (2) of Table 7 reports the results. The estimated coefficient on did is larger in Hubei, indicating that the income effect of flood control is stronger in Hubei Province than elsewhere. We then test whether the effect declines with geographic distance from the project site. Specifically, we augment Eq. ( 1 ) by interacting did with a measure of distance to the Three Gorges Project. Column (3) of Table 7 shows that the interaction term is negative and statistically significant, implying that the estimated income gains are concentrated closer to the project and diminish as distance increases. Together, these findings support the interpretation that Hubei is most exposed to the project’s flood control benefits and that the effect attenuates with distance. Table 7 Heterogeneity analysis (1) (2) (3) (4) eq eq eq eq did 0.059 *** 0.039 *** 0.189 *** 0.038 *** (0.007) (0.009) (0.050) (0.008) did* Eastern -0.056 *** (0.012) did* Hubei Province 0.041 *** (0.011) did* Distance from the Three Gorges -0.022 *** (0.009) did* Probability of flooding 0.345 *** (0.066) _cons 8.144 *** 8.118 *** 8.120 *** 8.130 *** (0.050) (0.051) (0.051) (0.051) N 5580 5580 5513 5245 \(\:{R}^{2}\) 0.986 0.986 0.986 0.986 Control variables Yes Yes Yes Yes Individual fixation effect Yes Yes Yes Yes Time fixation effect Yes Yes Yes Yes 5.3 Flood probability Because our analysis focuses on the flood control function of the Three Gorges Project, we also examine whether the income effect varies with local flood risk. We split counties by flood probability and re-estimate the baseline specification. The results show that the effect of did is larger in counties with a higher probability of flooding. Column (4) of Table 7 reports these estimates. This heterogeneity is consistent with a simple marginal-benefit logic. Counties facing higher flood risk tend to experience larger expected losses from floods, including greater threats to grain production and more frequent damage to housing, farm equipment, and other essential assets. These shocks can reduce income and increase the risk of poverty. By lowering disaster risk, the flood control function of the Three Gorges Project delivers larger gains precisely in the counties that are more exposed and more vulnerable, leading to a stronger observed increase in farmers’ disposable income. 6. Further analysis Finally, we examine whether the Three Gorges Project affects the urban rural income gap. Common prosperity requires not only aggregate growth but also a more inclusive distribution of income. In this sense, narrowing the income gap is a key component of common prosperity. We measure the urban rural income gap using the ratio of rural per capita disposable income to urban per capita disposable income, and we estimate the same DID framework using this outcome. To assess identification, we also conduct placebo and parallel trend tests analogous to those reported in the main analysis. Table 8 presents the results. Table 8 Further analysis (1) (2) (3) eq2 eq2 eq2 did 0.033 *** 0.015 *** 0.027 *** (0.004) (0.003) (0.008) Is 0.132 *** 0.094 *** 0.018 (0.018) (0.017) (0.018) Gov -0.340 *** -0.465 *** -0.006 (0.021) (0.024) (0.026) Fin 0.012 -0.076 *** 0.006 (0.010) (0.007) (0.007) Hc -0.036 *** 0.015 *** 0.059 *** (0.006) (0.006) (0.006) Gdp 0.001 *** 0.000 ** -0.002 *** (0.000) (0.000) (0.000) _cons 0.677 *** 0.456 *** 0.092 ** (0.038) (0.034) (0.037) N 3318 3318 3317 \(\:{R}^{2}\) 0.264 0.461 0.869 Individual fixation No No Yes Time-fixed effect No Yes Yes Table 8 about here After adding the full set of controls and urban fixed effects, the estimated coefficient on the Three Gorges Project policy is 0.027 and is statistically significant at the 1 percent level. This estimate implies that the project’s flood control function increases the ratio of rural per capita disposable income to urban per capita disposable income in flood-prone counties in the middle and lower Yangtze River basin by about 0.027 percentage points. Two considerations may help explain this result. First, rural areas typically start from a lower level of infrastructure, so the marginal gains from flood control and related water management improvements can be larger. Urban areas generally already have more complete systems for flood protection, water supply, power, transport, and emergency response, which can limit the incremental benefits from an additional large project. In contrast, many rural counties rely more directly on natural conditions for agricultural production, have weaker disaster resilience, and face higher costs of transport and logistics. These features make rural livelihoods more exposed to flood shocks. When a water conservancy project reduces flood risk and improves water availability, rural production and income may therefore respond more strongly. Second, the construction and operation of large water conservancy projects are often accompanied by policy support that is tilted toward rural and less developed areas. Examples include irrigation-related support, preferential electricity pricing, and complementary public investments. Such measures can reinforce the direct risk-reduction effects of the project by lowering production costs and easing constraints on rural development, which can contribute to a relative improvement in rural incomes compared with urban incomes. 6. Conclusions This paper uses a DID tool to evaluate the impact of the Three Gorges Project’s flood control function on farmers’ income. The results indicate that large-scale flood control infrastructure can contribute to income growth and to the goal of common prosperity. First, the flood control function of the Three Gorges Project significantly increases farmers’ per capita disposable income in flood-prone counties in the middle and lower Yangtze River basin, and the finding is stable across a wide set of robustness checks. In addition, the project not only raises farmers’ income but also increases the ratio of rural to urban per capita disposable income, implying a gradual narrowing of the urban-rural income gap. Second, mechanism evidence suggests that the income gains operate through three channels: stronger agricultural production capacity, greater entrepreneurship, and higher investment. Third, heterogeneity analyses show that the effects are larger in the central region, in Hubei Province, and in counties with higher flood probability. Taken together, these results suggest that the Three Gorges Project reduces flood risk while also creating a more stable environment for agricultural production and local economic activity. More broadly, the project appears to generate benefits beyond disaster prevention by supporting more balanced regional development and narrowing income disparities. The analysis offers evidence that can inform the evaluation of water conservancy projects and the design of future investments. Based on these findings, we draw two policy implications. First, the layout of water conservancy investments should be better aligned with local risk exposure and development needs. In high-risk flood-prone areas, including Hubei Province and counties along the middle and lower Yangtze River, policy should prioritize complementary small and medium-sized facilities that strengthen local irrigation and flood control networks. In less developed regions, water conservancy projects can be bundled with transport, electricity, and related infrastructure under the rural revitalization agenda to improve overall connectivity and amplify returns. Where conditions permit, integrating water conservancy investments with ecological restoration and local services can broaden benefits, for example by supporting ecological agriculture and rural tourism. Second, policy should strengthen the channels through which risk reduction translates into sustained income growth, especially entrepreneurship and investment. For entrepreneurship, local governments can improve access to credit and training for farmers in flood-prone areas, reduce administrative barriers for new businesses, and support organizational forms such as cooperatives and family farms that can raise productivity and market access. For investment, policies that attract private capital into water-related and agriculture-adjacent sectors, such as cold-chain logistics and agricultural technology, can help convert improved water security into durable local growth. Clear rules, transparent project selection, and well-designed PPP arrangements, together with targeted support measures such as fiscal incentives, can increase investment willingness and improve the long-run income effects of water conservancy projects. From an international perspective, the findings of this paper are consistent with evidence from other countries that large-scale flood control and water management infrastructure can generate substantial economic returns beyond risk mitigation. Studies on major river basin projects such as the Mississippi River flood control system in the United States, the Delta Works in the Netherlands, and multipurpose dams in countries like India and Brazil show that improved flood protection and water regulation can stabilize agricultural output, encourage private investment, and support structural transformation in rural areas. In developing economies, flood control projects along the Mekong and Ganges-Brahmaputra river systems have been found to reduce income volatility for farmers and facilitate the expansion of non-agricultural activities by lowering disaster-related uncertainty. However, international experience also suggests that the distributional effects of such projects depend critically on complementary institutions and policies, including land tenure security, access to finance, and local governance capacity. In line with this global evidence, the Three Gorges Project offers additional insights from China, showing that large-scale water conservancy investments, when implemented within a coordinated development framework, can simultaneously enhance disaster resilience and foster more inclusive income growth. This cross-country perspective underscores the broader relevance of China’s experience for other flood-prone developing regions facing similar trade-offs between risk reduction and long-term development. Declarations Conflicts of interests: None Note The table summarizes all the variables in the baseline regression and robustness check. Funding: None. Author Contribution Bin Xu: Conceptualization; Methodology; Formal analysis; Writing – original draft; Visualization.Ke Wei: Data curation; Investigation; Writing – original draft; Writing – review & editing; Validation.Zhiyang Shen: Supervision; Project administration; Writing – original draft; Writing – review & editing. Data availability: Data will be made available on the request. References Alemayehu YA, Ali AS, Mersha GT, Tesfahun T, Mengiste BM (2025) Extreme weather impacts on the socio-economic conditions of rural communities in Ethiopia: practical implications and recommendations for resilience and sustainability. Agricultural Food Econ 13(1):65. .https://doi.org/10.1186/s40100-025-00411-4 Batista ICA, Firme VDAC (2025) Economic impacts of the Brazil’s Mariana dam disaster on different levels of neighbourhood. Spat Economic Anal 1–23. https://doi.org/10.1080/17421772.2025.2502355 Bosheng ZHANG, Xiaoyang CAO (2021) Impact of targeted poverty alleviation policy on the urban-rural income gap in national poverty-stricken counties. Resour Sci 43(8):1549–1561 (in Chinese). https://doi.org/10.18402/resci.2021.08.04 Bowen R, Morris W (2019) The digital divide: Implications for agribusiness and entrepreneurship. Lessons from Wales. J Rural Stud 72:75–84. https://doi.org/10.1016/j.jrurstud.2019.10.031 Cai X, Yao L (2024) Assessing the option water-saving willingness of irrigation areas considering farmers’ risk tolerance. Risk Anal 44(1):70–86. https://doi.org/10.1111/risa.14140 Cao Y, Chen S (2022) Rebel on the canal: Disrupted trade access and social conflict in China, 1650–1911. Am Econ Rev 112(5):1555–1590. https://doi.org/10.1257/aer.20201283 Dell M, Jones BF, Olken BA (2014) What do we learn from the weather? The new climate-economy literature. J Econ Lit 52(3):740–798. https://doi.org/10.1257/jel.52.3.740 Fabri C, Vermeulen S, Van Passel S, Schaub S (2024) Crop diversification and the effect of weather shocks on Italian farmers' income and income risk. J Agric Econ 75(3):955–980. https://doi.org/10.1111/1477-9552.12610 Guo X, Wang L, Meng X, Dong X, Gu L (2023) The impact of digital inclusive finance on farmers' income level: evidence from China's major grain production regions. Finance Res Lett 58:104531. https://doi.org/10.1016/j.frl.2023.104531 Haishan C, Yang W (2022) Flood Control in Super Large Water Conservancy Facilities and Regional Economic Development: A Case Study of Three Gorges Project. World Econ 05137–161. https://doi.org/10.19985/j.cnki.cassjwe.2022.05.007 (in Chinese) Jiang T (2022) Mediating effects and moderating effects in causal inference. China Industrial Econ 5:100–120 (in Chinese). https://doi.org/10.19581/j.cnki.ciejournal.2022.05.005 Junqian W, Xingmin Z (2023) Can Grain Production Incentives Promote Farmers’ Incomes and County Economic Development? A Quasi-natural Experiment Based on the Incentive Policy for Large Grain-producing Counties. J Finance Econ 49(01):124–138 (in Chinese). https://doi.org/10.16538/j.cnki.jfe.20221022.402 Pagan-Cajigas Z, Guikema S, Otaduy‐Ramirez R, Woolley V, Smith K, Hu T, Chen T (2025) Water outage predictions for natural hazards using synthetic water distribution systems. Risk Anal. https://doi.org/10.1111/risa.70004 Kates RW, Clark WC, Corell R, Hall JM, Jaeger CC, Lowe I, Svedin U (2001) Sustainability science. Science 292(5517):641–642. https://doi.org/10.1126/science.1059386 Kocornik-Mina A, McDermott TK, Michaels G, Rauch F (2020) Flooded cities. Am Economic Journal: Appl Econ 12(2):35–66. https://doi.org/10.1257/app.20170066 Koks EE, Jongman B, Husby TG, Botzen WJ (2015) Combining hazard, exposure and social vulnerability to provide lessons for flood risk management. Environ Sci Policy 47:42–52. https://doi.org/10.1016/j.envsci.2014.10.013 Leiter AM, Oberhofer H, Raschky PA (2009) Creative Disasters? Flooding Effects on Capital, Labour and Productivity Within European Firms. Environ Resource Econ 43(3):333–350. https://doi.org/10.1007/s10640-009-9273-9 Liebenehm S, Schumacher I, Strobl E (2024) Rainfall shocks and risk aversion: Evidence from Southeast Asia. Am J Agric Econ 106(1):145–176. https://doi.org/10.1111/ajae.12403 Liu W, Qi P, Xu J, Feldman M, Xu D (2025) Does livelihood risk matter in disaster preparedness? Insights from flood risk areas of rural China. Clim Risk Manage 48:100705. https://doi.org/10.1016/j.crm.2025.100705 Luo Y, Liming, Fan (2012) Spatial Characteristics of the Income Effects of Rural Infrastructure in China: Evidence from Spatial Dependence and Spatial Heterogeneity. Manage World 571–87 (in Chinese). https://doi.org/10.19744/j.cnki.11-1235/f.2012.05.007 Mushi GE, Di Marzo Serugendo G, Burgi P-Y (2022) Digital Technology and Services for Sustainable Agriculture in Tanzania: A Literature Review. Sustainability 14(4). https://doi.org/10.3390/su14042415 Nunn N, Qian N (2014) US food aid and civil conflict. Am Econ Rev 104(6):1630–1666. https://doi.org/10.1257/aer.104.6.1630 Shi G, Pang J, Wang Y (2002) An Analysis of the Correlation between Water Conservancy Project Construction and Farmers’ Income. Chin Rural Econ 434–39 (in Chinese). https://doi.org/10.20077/j.cnki.11-1262/f.2002.04.006 Yan J (2025) Abnormal Weather and Household Consumption. Rev Dev Econ. https://doi.org/10.1111/rode.70047 Yang CL, Yan JD, He XD, Tian SQ (2023) What determines the survival of farmer entrepreneurship: Micro-evidence from China. Int Rev Econ Finance 86:334–348 Yuan B, Sun W, Huang J (2025) Once Bitten, Twice Shy: The Impact of Natural Disasters on the Adoption of Green Production Technologies by Farmers Based on the Risk Aversion Perspective. Twice Shy: The Impact of Natural Disasters on the Adoption of Green Production Technologies by Farmers Based on the Risk Aversion Perspective. https://doi.org/10.1002/ldr.70388 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 09 May, 2026 Reviewers agreed at journal 06 May, 2026 Reviewers invited by journal 04 Mar, 2026 Editor assigned by journal 15 Feb, 2026 Submission checks completed at journal 15 Feb, 2026 First submitted to journal 14 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8882775","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":600699900,"identity":"cbfd7eb3-dccd-4b40-abc8-c2d53b632b0c","order_by":0,"name":"Bin Xu","email":"","orcid":"","institution":"Jiangxi University of Finance and Economics","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Xu","suffix":""},{"id":600699901,"identity":"f2866e49-b89b-46ed-9090-28efbe38f77e","order_by":1,"name":"Ke Wei","email":"","orcid":"","institution":"Jiangxi University of Finance and Economics","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Wei","suffix":""},{"id":600699902,"identity":"b21733eb-d65f-40a1-87cf-e2b4415977cd","order_by":2,"name":"Zhiyang Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvUlEQVRIiWNgGAWjYHACNiC24YGwbYjXkgbVkka8lsMMxGuRb2B/9uDjjvMy/PyHD79gSLhHWAtjA4+54cwzt3kkZ6SlWTAkFBPWwszAwybN23abx+AGj5kB448EwlrYGNifSf9tO8djf/78NwOGBCK0AEPKTJqx7QCPAUMO8wOitEgw85hJ9rYl80jcSDMD6iBCi3x7+zOJn2129vz9hx9/+ECMFqD/4YBNghgNqLo/kKpjFIyCUTAKRgYAAGMvL+gfRBAGAAAAAElFTkSuQmCC","orcid":"","institution":"Univ. Lille, CNRS, UMR 9221 - LEM - Lille Economie Management","correspondingAuthor":true,"prefix":"","firstName":"Zhiyang","middleName":"","lastName":"Shen","suffix":""}],"badges":[],"createdAt":"2026-02-14 22:38:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8882775/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8882775/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104214674,"identity":"f297d243-a6a6-4e03-b2ef-18beaedef362","added_by":"auto","created_at":"2026-03-09 08:43:03","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":88088,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlood-prone areas in the middle and lower reaches of the Yangtze River\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8882775/v1/fa1e931854be99677d12cfe7.jpeg"},{"id":104214678,"identity":"6b0baa54-7ed2-41b0-a28f-ef59a071fd83","added_by":"auto","created_at":"2026-03-09 08:43:05","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":55607,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eParallel trend test\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8882775/v1/66d9c2c8acc1173bc288e1f2.png"},{"id":104214666,"identity":"fc26eb16-5d81-4669-99fb-fda5d22381c7","added_by":"auto","created_at":"2026-03-09 08:42:59","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":33229,"visible":true,"origin":"","legend":"\u003cp\u003ePlacebo test\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8882775/v1/3e5401f4801119a6bdcb93dc.jpeg"},{"id":104405227,"identity":"72eda003-9704-4051-a9aa-b8bb9796473f","added_by":"auto","created_at":"2026-03-11 12:22:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1509671,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8882775/v1/a59d1d4a-360a-46b5-b4a1-cc7a0759df86.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Water conservancy and farmers' income: A quasi-natural experiment based on the Three Gorges project","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAt the global level, increasing climate variability and the rising incidence of floods have renewed interest in large-scale water infrastructure as a key policy tool for safeguarding agricultural production and rural livelihoods, particularly in developing and emerging economies. International organizations and governments have emphasized that effective flood control and water management are essential not only for disaster risk reduction, but also for promoting inclusive growth and preventing climate-induced poverty traps in rural areas. Recent empirical evidence on climate shocks shows that environmental shocks not only disrupt agricultural production but also alter farmers\u0026rsquo; economic behavior and welfare dynamics. For example, Liebenehm et al. (2023) find that rainfall shocks increase individual risk aversion among rural households in Southeast Asia, particularly for net food buyers, implying that climate variability can shape farmers\u0026rsquo; economic decisions and potentially perpetuate poverty in the absence of adequate credit and insurance.\u003c/p\u003e \u003cp\u003eTo develop rural reforms and advance comprehensive rural revitalization, the 2025 No.1 Central Document of Chinese government highlights the need to strengthen capacity for agricultural disaster prevention and mitigation and calls for a modern system for flood control and disaster reduction. Farmers\u0026rsquo; income has long been at the center of the \u0026ldquo;three rural areas\u0026rdquo; agenda. A large body of research studies the drivers of income growth, with recent work emphasizing digital empowerment (Bowen \u0026amp; Morris, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Mushi et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Guo et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), policy support (Bosheng \u0026amp; Xiaoyang, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Junqian \u0026amp; Xingmin, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), and farmers\u0026rsquo; own initiatives (Yang et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). By contrast, relatively little attention has been given to hydraulic engineering and flood control infrastructure, despite their direct link to agricultural risk and rural livelihoods.\u003c/p\u003e \u003cp\u003eThis gap matters because extreme hydrological events are becoming more frequent and more damaging. Since the Industrial Revolution, technological progress and economic growth have been accompanied by large greenhouse gas emissions. The interaction of global warming and human activities has altered the water cycle across space and time, raising the likelihood of extreme events and affecting regional water security and national development strategies over the medium and long term. These risks are closely tied to the food system and to rural welfare. Floods threaten farmers\u0026rsquo; lives and health and destroy productive assets. They reduce income, worsen quality of life, and can undermine mental health, sometimes pushing households into poverty or back into poverty. Koks et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) emphasize that flood risk outcomes depend not only on physical hazards and exposure, but also on social vulnerability, which is particularly relevant for farming households with limited adaptive capacity. When shocks are large and insurance and credit are limited, many households have little capacity to absorb losses. Historical experience warns us that the rise and fall of water infrastructure extends far beyond the projects themselves, profoundly impacting regional economic vitality and social stability. Cao \u0026amp; Chen (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) revealed in their seminal study of the Grand Canal that the abandonment of trade routes can trigger social conflicts due to lost opportunities. Conversely, this demonstrates that constructing and maintaining inclusive water projects may serve as a fundamental stabilizer for safeguarding livelihoods and promoting shared development. Understanding how to respond to extreme hydrological events, raise farmers\u0026rsquo; income, and reducing the risk of returning to poverty is therefore important for the goal of common prosperity.\u003c/p\u003e \u003cp\u003eWater conservancy is often described as the lifeblood of agriculture and a foundation of the national economy. As Sun Yat-sen wrote in 1918, constructing slices and weirs can regulate water flow for navigation while also harnessing hydraulic power. The Three Gorges Project, the world\u0026rsquo;s largest hydro-junction, is a core infrastructure investment in the Yangtze River Basin. Its main benefits fall into three areas, flood control and disaster mitigation, power generation, and navigation, with flood control widely viewed as the most important. Batista and Firme (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) examine the Mariana dam disaster in Brazil and find substantial and spatially uneven economic losses across neighborhoods, with pronounced impacts on agriculture and local economic activity. Their results underscore how water-related infrastructure events can directly affect rural incomes through production and environmental channels. Since 2007, the project\u0026rsquo;s flood control function has mitigated flood risks in the middle and lower Yangtze River regions, improved safety in vulnerable areas, and eased flood pressure in the lower reaches of the Jingjiang River. It has also been linked to addressing challenges that historically followed major floods, including environmental damage and post-flood epidemics. More broadly, effective disaster reduction can limit economic losses, stabilize production, and support wealth accumulation by reducing disruptions to social reproduction.\u003c/p\u003e \u003cp\u003eSo far, most research on water conservancy projects has emphasized ecological impacts, while much less attention has been paid to farmers\u0026rsquo; income outcomes. In particular, few studies have systematically examined the link between water conservancy project construction and farmers\u0026rsquo; income (Shi et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Following Haishan \u0026amp; Yang (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), this paper treats 2007, when the flood control function of the Three Gorges Project officially came into play, as the policy implementation node and applies a DID design to estimate the income effect of flood risk reduction. Using balanced panel data for 351 counties in the middle and lower Yangtze River basin over 2003 to 2020, we find that the flood control function of the Three Gorges Project significantly increases farmers\u0026rsquo; per capita disposable income in flood-prone counties by about 0.054 percentage points. The estimate is stable across a set of sensitivity and robustness checks. Mechanism results suggest three pathways: improved agricultural productivity, stronger entrepreneurship, and greater investment inflows. We further document heterogeneous effects across county types.\u003c/p\u003e \u003cp\u003eThis paper contributes to literature in three ways. First, by leveraging the Three Gorges Project as a quasi-natural experiment and a balanced county panel, we provide causal evidence on how a major water conservancy project affects farmers\u0026rsquo; income, and we unpack micro-level channels related to agricultural capacity, entrepreneurial activity, and investment attraction. Second, we quantitatively connect the project\u0026rsquo;s flood control function to both income growth and the urban-rural income gap, helping fill a gap at the intersection of water conservancy and rural economics. Third, the findings shed light on how large-scale water infrastructure can support regional coordination and shared prosperity, and they offer empirical support for policies that aim to translate water security and disaster mitigation into sustained rural income gains.\u003c/p\u003e"},{"header":"2. Theoretical analysis and literature review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Background\u003c/h2\u003e \u003cp\u003eFloods are among the most frequent, geographically extensive, and economically damaging natural hazards in China. These disasters were concentrated in the middle and lower Yangtze River basin, as well as parts of North China and Northeast China. The Yangtze River basin alone accounts for more than one fifth of China\u0026rsquo;s land area, and its combination of complex terrain and high precipitation has long made it a flood-prone region. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows that flood-vulnerable counties in the middle and lower reaches are largely distributed along the main river corridor, excluding municipalities.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e \u003cb\u003eabout here\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe Three Gorges Project was proposed as early as 1918, but for decades the technical and fiscal conditions for construction were not in place. With the economic expansion following reform and opening up, these constraints eased and project preparation accelerated. In July 1993, the State Council\u0026rsquo;s Three Gorges Construction Committee approved the preliminary design report and initiated technical design work for key components. Construction began in 1994 in Sandouping, Yichang, Hubei Province, and the river was successfully diverted in 1997. The Flood Control Law promulgated the same year strengthened the legal basis for flood control planning, project construction, and operational management, thereby supporting the institutional environment in which the project could function as part of the national flood control system. In July 2007, the project began to deliver flood control benefits in practice, with an initial reported peak clipping of 3,000 cubic meters per second. As a central component of Yangtze River governance, the project is widely viewed as a major investment in flood risk reduction. Yet the relationship between flood shocks and rural poverty suggests that disaster mitigation alone is not the endpoint. The broader policy question is whether large-scale flood control can also support sustained income growth and common prosperity by reducing risk and enabling rural development.\u003c/p\u003e \u003cp\u003eThe core of the \u0026ldquo;three rural areas\u0026rdquo; agenda ultimately centers on farmers\u0026rsquo; livelihoods. A durable solution requires continued growth in farmers\u0026rsquo; income and a narrowing of income gaps between urban and rural areas and across regions. Rural poverty remains persistent and, for some households, recurrent, reflecting the interaction of multiple constraints such as adverse natural conditions and a weak local economic base. Natural disasters, especially floods and droughts, can reinforce these constraints by destroying assets, interrupting production, and destabilizing earnings. When livelihoods are fragile and risk-coping capacity is limited, shocks can raise the likelihood of falling into poverty again, including in some eastern areas and major grain-producing regions. Even after the achievement of the goal of building a moderately prosperous society in all respects, consolidating poverty alleviation gains, raising farmers\u0026rsquo; incomes, especially for low-income households, and narrowing the urban-rural gap are likely to remain central tasks of agricultural and rural development for the foreseeable future.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 The impact of water conservancy projects on farmers\u0026rsquo; income\u003c/h2\u003e \u003cp\u003eWater conservancy projects can shape farmers\u0026rsquo; income through macro or micro channels. Kates et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2001\u003c/span\u003e) argue that disasters originate from the interaction between society and nature and may disrupt regional sustainable development. Dell et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) review the growing climate-economy literature and show that variations in temperature, precipitation, and extreme weather events have significant effects on agricultural productivity, income, and long-run economic growth, particularly in developing regions where farmers are highly dependent on water availability. Flood shocks can generate sudden shortages of resources, damage infrastructure, induce population displacement, and reduce ecosystem services. These effects may also trigger secondary disasters and broader social problems, which can weaken long-run development plans and sustainability goals.\u003c/p\u003e \u003cp\u003eFrom a macro perspective, floods can affect economic performance through changes in post-disaster investment returns, human capital accumulation, and the pace and direction of technological progress. In flood-prone regions, floods impose direct losses on property, agriculture, industry, and transport, and they also create latent risks that depress investment and amplify uncertainty. Leiter et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) use a DID design to study European firms and find that exposure to flood risk reduces investor willingness in affected areas. Flood risk can also force local governments to devote substantial fiscal resources to flood control institutions and infrastructure, which can crowd out other productive spending. Even when regions have geographic advantages and strong market access, persistent flood risk can restrain growth by raising operating costs and discouraging long-horizon investment.\u003c/p\u003e \u003cp\u003eAt the household level, floods affect income and welfare through two main pathways. First, they cause direct losses of life and property, destroy housing, and damage productive assets. Second, they shift risk expectations and change consumption and saving behavior. When households suffer direct losses, they often cut non-essential consumption while increasing spending on relief and reconstruction, which reshapes both the consumption bundle and savings (Yan, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In anticipation of future shocks, households may also raise precautionary savings and reduce discretionary consumption, consistent with precautionary savings theory. These behavioral responses matter for income dynamics because they influence labor supply choices, investment in farm inputs, and the capacity to finance recovery.\u003c/p\u003e \u003cp\u003eAt the meso level of the rural economy, flood impacts are especially visible in agricultural production and the destruction of basic means of production and livelihood. Floods inundate farmland and damage crops, reducing yields and lowering farm income. At the same time, damage to houses, farm tools, and other productive equipment increases the vulnerability of rural households. Kocornik et al. (2020) show that economic activity often remains concentrated in flood-prone areas, suggesting that farmers and rural communities may continue to rely on water-intensive locations despite persistent risks. Housing and farm implements are core assets for farm families and are often the most immediate victims of flooding. When these assets are destroyed, agricultural production may not return to normal quickly, reducing output in the current season and potentially affecting the next production cycle (Alemayehu, et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Interrupted agricultural reproduction not only lowers income but can also raise production costs through replanting, repairs, and the need to replace damaged inputs, which further tightens household budgets. In this way, natural disasters can increase rural poverty incidence and deepen poverty persistence, limiting rural development and complicating the objective of common prosperity. Yuan et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) indicate that repeated exposure to shocks discourages farmers from adopting green production technologies due to heightened risk aversion, which in turn may influence income dynamics and technological diffusion in rural areas\u003c/p\u003e \u003cp\u003eDisaster economics further suggests that disasters are fundamentally economic events, in the sense that they threaten welfare, destroy wealth, and weaken the foundations of socioeconomic sustainability. From this perspective, water conservancy projects can reduce flood damages and, at the same time, support rural development by improving the local production environment (Liu et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). On one hand, flood control infrastructure stabilizes the natural environment in flood-prone areas and protects lives and property. Lower risk can free fiscal capacity for productive public spending, which may support a reinforcing process of risk reduction and capital accumulation. Water conservancy projects also tend to be bundled with complementary infrastructure such as roads and bridges, improving market access and the local investment environment. Better connectivity can attract additional public and private investment, allowing resilience improvements to translate into economic gains. On the other hand, water conservancy projects can reduce flood-related crop losses and strengthen drought resistance through reservoirs, water regulation and storage systems, and ponds. By storing water during wet periods and providing irrigation during dry periods, these projects improve the timing and allocation of water resources, stabilize irrigation, protect yields, and reduce volatility in farmers\u0026rsquo; operating income. The public goods nature of infrastructure implies potential spatial spillovers, while improvements in water management can promote adaptation to climate risk. Together, these mechanisms can raise incomes, reduce vulnerability, and help prevent poverty recurrence in flood-prone rural areas. The mechanism section develops these channels in detail. Based on this framework, we propose the following hypotheses:\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 1\u003c/strong\u003e \u003cp\u003eThe Three Gorges Project significantly increases farmers\u0026rsquo; disposable income.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eHypothesis 2\u003c/strong\u003e \u003cp\u003eThe Three Gorges Project drives income growth through three channels: enhancing agricultural productivity, stimulating entrepreneurship, and attracting investments.\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Research design","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Measurement model construction\u003c/h2\u003e \u003cp\u003eSince the Three Gorges Project can be regarded as a quasi-natural experiment, this paper uses the difference-in-differences (DID) to estimate the impact of water conservancy projects on farmers' income growth, treating counties in flood-prone areas as the treatment group and other counties as the control group, while establishing a baseline regression model in the following form:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{Y}_{it}={\\alpha\\:}_{0}+{\\beta\\:}_{0}{did}_{it}+\\gamma\\:{X}_{it}+{\\mu\\:}_{i}+{\\lambda\\:}_{t}+{\\epsilon\\:}_{it}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eIn formula (1),\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{Y}_{it}\\)\u003c/span\u003e\u003c/span\u003e is the explained variable, representing the logarithmic value of the per capita disposable income of farmers in the county in year t.\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{d}\\text{i}\\text{d}}_{\\text{i}\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e represents the interactive term, specifically, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{d}\\text{i}\\text{d}}_{\\text{i}\\text{t}}={\\text{T}\\text{r}\\text{e}\\text{a}\\text{t}}_{\\text{i}}\\times\\:{\\text{P}\\text{o}\\text{s}\\text{t}}_{\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e, is a dummy variable equal to 1 if county i is located in a flood-prone area in the middle and lower reaches of the Yangtze River, and 0 otherwise. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{P}\\text{o}\\text{s}\\text{t}}_{\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e is a time dummy indicating the period in which the flood control function of the Three Gorges Project is in effect: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{P}\\text{o}\\text{s}\\text{t}}_{\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e=0 before the function takes effect, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{P}\\text{o}\\text{s}\\text{t}}_{\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e=1 during and after its implementation.\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{X}}_{\\text{i}\\text{t}}\\)\u003c/span\u003e \u003c/span\u003e denotes a vector of control variables that may affect farmers\u0026rsquo; income at the county level. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\mu\\:}}_{\\text{i}}\\)\u003c/span\u003e\u003c/span\u003e represents county fixed effects, which control for time-invariant unobserved heterogeneity across counties, while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\lambda\\:}}_{\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e represents year fixed effects, capturing common shocks and macroeconomic trends over time. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\epsilon\\:}}_{\\text{i}\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e is the idiosyncratic error term. The coefficient of interest, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{{\\beta\\:}}_{0}\\)\u003c/span\u003e\u003c/span\u003e identifies the causal effect of the flood control function of the Three Gorges Project on farmers\u0026rsquo; income growth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Measurement and explanation of variables\u003c/h2\u003e \u003cp\u003eThis paper uses farmers' disposable income from 2003 to 2020 to measure income growth. Based on whether counties are located in flood-prone areas in the middle and lower reaches of the Yangtze River, the sample is divided into a treatment group and a control group, with 133 counties classified as the treatment group. With reference to Haishan (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), the year 2007 is taken as the \u0026ldquo;policy implementation node\u0026rdquo;, marking the point when the Three Gorges Project began to exert its flood control function.\u003c/p\u003e \u003cp\u003e(1) Explained variables. Consistent with the existing literature, the dependent variable is measured as the logarithm of farmers\u0026rsquo; per capita disposable income in each county.\u003c/p\u003e \u003cp\u003e(2) Core explanatory variable. The key explanatory variable is the interaction term \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{d}\\text{i}\\text{d}}_{\\text{i}\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e, which captures counties located in flood-prone areas during the period when the flood control function of the Three Gorges Project is in effect.\u003c/p\u003e \u003cp\u003e(3) Control variables. The control variables include government size (Gov), measured as the ratio of local government general budget expenditure to GDP; financial development (Fin), measured by the ratio of outstanding loans from financial institutions to GDP at year-end; human capital (Hc), measured as the logarithm of the number of students per 10,000 residents; and regional economic development (Gdp), proxied by the annual average nighttime light intensity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Data sources and descriptive statistics\u003c/h2\u003e \u003cp\u003eIn this paper, the middle and lower reaches of the Yangtze river five provinces (excluding municipality directly under the central government) 351 county-level administrative region (region, counties and county-level cities) annual panel data as the research object, the average value data from global night light data (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://ngdc.noaa.gov/eog/download.html\u003c/span\u003e\u003cspan address=\"https://ngdc.noaa.gov/eog/download.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e); For flood- prone areas, refer to Atlas of Major Natural Disasters and Society of China (2004) and Name Code of Flood Storage Areas of China (2001); Other data are from China County Statistical Yearbook. The descriptive statistics of the main variables are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eVariable definitions and descriptive statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003cp\u003etype\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eVariable definition\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSample\u003c/p\u003e \u003cp\u003esize\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eStandard\u003c/p\u003e \u003cp\u003edeviation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMinimum\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eMaximum\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExplained\u003c/p\u003e \u003cp\u003evariable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eeq\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLogarithm of per capita disposable income of\u003c/p\u003e \u003cp\u003erural residents\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e7.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e10.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCore explanatory\u003c/p\u003e \u003cp\u003evariable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003edid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWhether it is located in a flood-prone area the interaction term of the grouping dummy variable\u003c/p\u003e \u003cp\u003eand the policy time dummy variable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003cp\u003evariable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSecondary industry /GDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFiscal expenditure /GDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFinancial institution loans at year-end /GDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e2.39\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLogarithm of number of students in school per\u003c/p\u003e \u003cp\u003e10,000 people\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e7.06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGdp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLighting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e7.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e57.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Empirical analysis","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n\u003ch2\u003e4.1 Baseline regression\u003c/h2\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e reports the baseline estimates of the effect of the Three Gorges Project\u0026rsquo;s flood control function on farmers\u0026rsquo; per capita disposable income in flood-prone counties in the middle and lower reaches of the Yangtze River, providing evidence in support of Hypothesis \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. Column (1) presents estimates from a specification without fixed effects. Column (2) adds year fixed effects, while Column (3) further includes county fixed effects and the full set of control variables. Across all three specifications, the estimated coefficient on the Three Gorges Project policy variable is positive and statistically significant. This pattern indicates that, after 2007, counties exposed to the project\u0026rsquo;s flood control benefits experienced higher farmers\u0026rsquo; disposable income relative to the comparison group. In the preferred specification in column (3), which controls observed covariates as well as county and time fixed effects, the coefficient on the policy variable is 0.054 and is significant at the 1 percent level. The estimate implies that the flood control function of the Three Gorges Project increased farmers\u0026rsquo; disposable income in flood-prone counties by about 0.054 percentage points. Taken together, these results suggest that flood control infrastructure can raise rural incomes in exposed areas, consistent with Hypothesis \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003ctable id=\"Tab2\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eBenchmark regression results\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(3)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.308\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.036\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.015)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eIs\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1.300\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.766\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.302\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.066)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.039)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.026)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eGov\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.567\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-1.928\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.122\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.097)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.074)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.040)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eFin\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.428\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.029\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.000\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.047)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.021)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eHc\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.903\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.026\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.078\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.027)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.013)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.008)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eGdp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.033\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.016\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.007\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.001)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.001)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.001)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e_cons\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e13.317\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.869\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.139\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.180)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.082)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.490\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.894\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n\u003ch2\u003e4.2 Parallel trend test\u003c/h2\u003e\n\u003cp\u003eWe conduct an event-study test of the parallel trend\u0026rsquo;s assumption using 2003 as the base year. Figure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e plots the estimated coefficients from Eq.\u0026nbsp;(2). The coefficients for the pre-2007 years are small and statistically indistinguishable from zero, suggesting no differential pre-trends between the treated and control counties before the flood control function of the Three Gorges Project began to operate in 2007. In contrast, the post-2007 coefficients are positive and statistically significant, and their magnitude increases over time, indicating that the income effect grows as exposure accumulates. Taken together, these results support the parallel trends assumption and justify using 2007 as the policy timing when the project\u0026rsquo;s flood control function started to take effect.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n\u003ch2\u003e4.3 Robustness test\u003c/h2\u003e\n\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.1 Placebo Test\u003c/h2\u003e\n\u003cp\u003eTo assess the reliability of the baseline estimates and address the concern that omitted factors could lead to spurious rejection of the null, we conduct a placebo test based on random reassignment. Specifically, we randomly draw a set of treated counties and randomly assign a policy timing, then re-estimate the baseline specification. We repeat this procedure 500 times to obtain the distribution of placebo coefficients. Figure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e summarizes the results. The placebo estimates are tightly centered around zero, and virtually all simulated coefficients are smaller in magnitude than the baseline estimate of 0.054. This pattern suggests that the baseline effect is unlikely to be driven by chance assignment or unobserved shocks unrelated to the Three Gorges Project, and it supports the interpretation that the estimated impact reflects the project\u0026rsquo;s flood control benefits.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.2 PSM-DID\u003c/h2\u003e\n\u003cp\u003eBecause treatment status is defined by whether a county is flood-prone rather than randomly assigned, treated and control counties may differ along other observable dimensions that also affect farmers\u0026rsquo; income. To address this concern, we use PSM to construct a comparison group that is more similar to the treated counties in terms of the control variables. We implement 1:1 nearest-neighbor matching and then re-estimate the baseline DID specification on the matched sample. Column (1) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e reports the results. The coefficient on did remains positive and statistically significant, consistent with the baseline estimates. This finding suggests that differences in observable characteristics and sample selection based on flood exposure are unlikely to drive our main result.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab3\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eRobustness test\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(3)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(4)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(5)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.055\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.051\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.018\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054***\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.006)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.014)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.009)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e_cons\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.216\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.152\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.257\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.139\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.052)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.049)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.051)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.107)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5290\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.989\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.106\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eProvince - Time fixed effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNo\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.3 Adding Control Variables\u003c/h2\u003e\n\u003cp\u003eRegional income growth may depend on initial development conditions because regional development can feature increasing returns and cumulative causation. To account for this possibility, we augment the baseline specification by interacting the county\u0026rsquo;s average nighttime light intensity in 2003 with a linear time trend. This term allows counties with different initial development levels to follow different underlying growth paths. Column (2) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e reports the results. The interaction term is statistically significant, indicating that initial economic conditions are indeed correlated with subsequent income growth. Importantly, the estimated effect of the flood control function of the Three Gorges Project remains positive and statistically significant after this adjustment. This implies that the baseline result is not driven by differential growth related to initial development levels, and that the project\u0026rsquo;s flood control benefits still contribute to higher farmers\u0026rsquo; income.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.4 Controlling for the Influence of Macro Factors\u003c/h2\u003e\n\u003cp\u003eIn the baseline specification, we include county fixed effects and time fixed effects to absorb time-invariant county characteristics and common shocks. To further address the concern that provinces may face different macroeconomic conditions over time, we add province-time fixed effects, which flexibly control for all province-level shocks in each year. Column (3) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e reports the results. The estimated coefficient on did remains positive and statistically significant, indicating that our main finding is not driven by province-specific time-varying factors.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.5 Adjust the standard error\u003c/h2\u003e\n\u003cp\u003eWe adjust standard errors in two ways to account for potential correlation in the error term. First, we use two-way clustering at the county level and the province-year level. Column (4) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e reports the corresponding estimates. Second, we report spatial HAC standard errors that allow for spatial correlation within a 50 km radius and first-order serial correlation over time. Column (5) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e presents these results. Under both approaches, the estimated coefficient on did remains statistically significant, indicating that our main results are robust to alternative assumptions about heteroskedasticity, serial correlation, and spatial dependence.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.6 Excluding the Influence of Other Policies\u003c/h2\u003e\n\u003cp\u003eBecause counties in our sample may have been exposed to other major policies during the same period, policy overlap could bias the baseline estimates. We therefore control a set of contemporaneous policies that are likely to affect farmers\u0026rsquo; income, and we examine whether the estimated impact of the Three Gorges Project changes once these policies are taken into account. Specifically, we sequentially add indicators for high-speed rail expansion, the rollout of e-commerce into rural areas, national poverty alleviation programs, targeted poverty alleviation policies, and incentive policies for major grain-producing counties. These policies can influence rural incomes through several channels. High-speed rail expansion may raise income by easing labor mobility, supporting county-level growth, and generating spatial spillovers. Rural e-commerce programs can reduce information frictions and expand market access, allowing farmers to adjust production and sales decisions, and potentially strengthening local agglomeration effects as market scale increases. Poverty alleviation policies, including the designation of poor national-level counties and targeted programs, can increase fiscal resources through transfer payments, subsidized credit, and related support, which may improve infrastructure and encourage the development of specialized local industries. The incentive policy for major grain-producing counties can shift local government priorities toward agriculture, improve production conditions, and attract investment into the agricultural sector. To isolate the independent effect of the Three Gorges Project on farmers\u0026rsquo; income growth in the middle and lower Yangtze River basin, we include dummy variables for these policies in the DID specification. Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e reports the estimates. Across all specifications, the coefficient on did remains positive and statistically significant, and its magnitude is close to the benchmark estimate. These results indicate that the main finding is not explained by concurrent policy interventions, and the estimated income effect of the Three Gorges Project\u0026rsquo;s flood control function is robust to controlling for policy overlap.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab4\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eexcludes other policies during the same period\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(3)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(4)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(5)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(6)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.055\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.054\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.056\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eHigh-speed rail opened\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.003\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.006\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.006)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.006)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eE-commerce into rural areas\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.027\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.002\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.006)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.005)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eNational-level poverty-\u003c/p\u003e\n\u003cp\u003estricken county\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.085\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.146\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.008)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.009)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eTargeted poverty reduction\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.088\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.127\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.006)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eLarge grain-producing\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.012\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.002\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e_cons\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.140\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.145\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.197\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.192\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.141\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.313\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.049)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.049)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.046)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5593\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.987\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.987\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section3\"\u003e\n\u003ch2\u003e4.3.7 Endogeneity Analysis\u003c/h2\u003e\n\u003cp\u003eThe baseline results show that the flood control function of the Three Gorges Project significantly raises farmers\u0026rsquo; income in flood-prone counties in the middle and lower Yangtze River basin. Although we mitigate omitted-variable concerns by including county fixed effects, year fixed effects, and province-time fixed effects, Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e suggests that flood-prone counties are concentrated along the Yangtze River. These locations are likely to share geographic and economic features, such as lower elevation, flatter terrain, proximity to the river channel, better transport access, and shorter distance to major markets. If these advantages are correlated with both treatment status and income growth, they could confound the DID estimates. In addition, because treatment assignment is based on flood exposure rather than random allocation, sample selection may remain a concern. To further address these endogeneity issues, we implement an IV approach.\u003c/p\u003e\n\u003cp\u003eWe use the linear distance from each county to the Three Gorges Project as an instrument. Because distance is time-invariant, we follow Nunn and Qian (\u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e) and interact distance with a time variable to generate identifying variation in a panel setting. The relevance condition is plausible. As Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e indicates, counties closer to the Three Gorges Project are more likely to be in flood-prone areas and to have larger exposed areas. Moreover, the project is located near the transition between China\u0026rsquo;s second and third topographic steps, where the river gradient is large and flows are rapid, making flood control benefits more likely to propagate to downstream locations. These features imply that proximity increases the probability and intensity of being affected by the project\u0026rsquo;s flood control function.\u003c/p\u003e\n\u003cp\u003eThe exclusion restriction is supported by the fact that distance to the Three Gorges Project is a predetermined geographic measure. Conditional on fixed effects and the included controls, it is unlikely to affect farmers\u0026rsquo; income through channels other than exposure to the project\u0026rsquo;s flood control benefits. Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e reports the IV estimates. The first-stage results show a strong relationship between the instrument and treatment exposure. Standard diagnostics, including under-identification and weak-instrument tests, support the validity of the instrument. In the second stage, the estimated effect remains positive and statistically significant at the 1 percent level. Overall, the IV results reinforce the baseline conclusion that the flood control function of the Three Gorges Project increases farmers\u0026rsquo; income.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab5\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eEndogeneity analysis\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePhase I Results\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePhase 2 Results\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIV\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.056\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1.474\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.281)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eFirst stage F number\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e27.58\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eUnderrecognition test\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e29.08\u003c/p\u003e\n\u003cp\u003e[0.000]\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eWeak instrumental variable test\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e27.58\u003c/p\u003e\n\u003cp\u003e{16.38}\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e5513\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003ctfoot\u003e\n\u003ctr\u003e\n\u003ctd colspan=\"3\"\u003eNote :[] inside is the p-value of the LM statistic; {} inside is the critical value at the 10% level of the Stock-Yogo weak recognition test\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tfoot\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"5. Further analysis","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n\u003ch2\u003e5.1 Mechanism analysis\u003c/h2\u003e\n\u003cp\u003eWe next examine the mechanisms through which the flood control function of water conservancy projects increases farmers\u0026rsquo; income. We focus on three channels. First, flood control can strengthen agricultural production capacity. By reducing flood-related losses and stabilizing the production environment, water conservancy projects make agricultural production more predictable, which can facilitate mechanization, improve efficiency, and support larger-scale operations. In addition, by improving the timing and allocation of water resources, these projects can sustain irrigation during droughts, improve crop growing conditions, and raise output in the primary sector, thereby increasing farm operating income (Fabri et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Second, flood control can encourage entrepreneurship. Lower disaster risk reduces income volatility and improves expectations about future returns, which can increase farmers\u0026rsquo; willingness and ability to start businesses. Entrepreneurship can expand market access, raise the value added of agricultural products, and create local jobs, which supports poverty reduction and narrows the urban-rural income gap. Third, flood control can attract investment. By lowering disaster risk and improving complementary infrastructure such as roads and bridges, water conservancy projects can improve the local investment and financing environment and increase investors\u0026rsquo; confidence in long-horizon projects (Luo \u0026amp; Fan, 2012). Risk reduction can also lower insurance costs for agriculture and infrastructure, which improves expected returns. In addition, water security can enable more diverse land uses, including tourism and ecological agriculture, which can increase the resilience of the local economy.\u003c/p\u003e\n\u003cp\u003eGuided by these mechanisms, we test whether the income effect of the Three Gorges Project operates through agricultural production capacity, entrepreneurship, and investment. Agricultural production capacity is measured along two dimensions, the development level of agriculture and the scale of agricultural production. Using county-level data, we proxy agricultural development with per capita value added in the primary sector and agricultural scale with per capita grain sown area. We measure entrepreneurship using the relative level of entrepreneurial activity per 10,000 people, and we measure investment with the share of fixed asset investment in GDP. To avoid the limitations of conventional stepwise mediation tests, we follow Jiang (\u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e) and estimate the channel regressions directly. Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e reports the results.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003ctable id=\"Tab6\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eMechanism analysis of flood control\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(3)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(4)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eLevel of agricultural\u003c/p\u003e\n\u003cp\u003edevelopment\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eScale of agricultural\u003c/p\u003e\n\u003cp\u003eproduction\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePromoting\u003c/p\u003e\n\u003cp\u003eentrepreneurship\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eAttracting investment\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.040\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.077\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.091\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.084\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.015)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.038)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.012)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e_cons\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.630\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1.126\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2.612\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.373\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.069)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.110)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.222)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.097)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5725\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3860\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5569\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4762\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.955\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.796\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.731\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.799\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation\u003c/p\u003e\n\u003cp\u003eeffect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eColumn (1) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e shows that the flood control function of the Three Gorges Project significantly raises per capita value added in the primary sector at the 1 percent level, consistent with improved agricultural development. This result accords with the idea that risk reduction stabilizes production and helps protect grain output and farm operating income. Column (2) indicates that flood control also expands the scale of agricultural production, suggesting greater specialization and higher production efficiency. Column (3) examines entrepreneurship. The estimated effect is positive and highly significant, and it is the largest among the three channels. This pattern is consistent with farmers being more willing to start businesses when disaster risk falls and expected returns become more stable. Entrepreneurship can both expand market opportunities and increase the value added of agricultural products. Finally, column (4) shows that flood-prone counties became more attractive to capital after 2007, as measured by the share of fixed asset investment in GDP. This finding suggests that, once flood risk declines, the location advantages of counties along the Yangtze River can translate more effectively into investment inflows and more diversified land use. Overall, the results in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e support the view that the Three Gorges Project increases farmers\u0026rsquo; income through higher agricultural production capacity, stronger entrepreneurship, and greater investment.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n\u003ch2\u003e5.2 Heterogeneity analysis\u003c/h2\u003e\n\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\n\u003ch2\u003e5.2.1 Eastern and central provinces\u003c/h2\u003e\n\u003cp\u003eRegional coordination is a core component of common prosperity, so we examine whether the income effect of the Three Gorges Project differs across regions. We split the county sample by location into eastern and central regions and re-estimate the baseline specification within each subsample. The results indicate meaningful heterogeneity. The estimated coefficient for the eastern region is negative, while the effect is larger and more positive in the central region. This pattern suggests that the flood control benefits of the project translate into stronger income gains in the central region than in the eastern region.\u003c/p\u003e\n\u003cp\u003eA plausible interpretation is that baseline development conditions shape the marginal returns to flood risk reduction. Counties in the eastern region typically have higher economic concentration, stronger institutions, and better infrastructure, and farmers\u0026rsquo; disposable income levels are already relatively high. As a result, the scope for additional income gains may be limited. In contrast, many counties in the central region start from lower income levels and face tighter infrastructure and risk-management constraints, leaving more room for improvement once flood risk declines. Consistent with this idea, the estimated effect of did tends to weaken as the initial income level rises, and the stability checks that add controls yield patterns that align with this interpretation.\u003c/p\u003e\n\u003cp\u003eOverall, the heterogeneity results suggest that the project\u0026rsquo;s flood control function has a larger income effect in counties with slower initial development. In this sense, the Three Gorges Project appears to generate relatively greater gains in less developed areas, which is consistent with a reduction in regional disparities and with the goal of regional coordinated development under common prosperity.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec24\" class=\"Section3\"\u003e\n\u003ch2\u003e5.2.2 Hubei Province and others\u003c/h2\u003e\n\u003cp\u003eHubei Province is plausibly the largest beneficiary of the Three Gorges Project\u0026rsquo;s flood control function. Flood risk in the Jingjiang reach is widely viewed as especially severe because the river channel is highly curved and flood flows can be constrained, increasing the likelihood of dike failure. The north bank also borders the Jianghan Plain, which is low-lying and therefore highly exposed to inundation risk. Because the Three Gorges Project is located in Yichang, Hubei Province, it directly regulates upstream inflows into this reach and can reduce downstream flood pressure. Motivated by this geography, we split the sample into Hubei and all other provinces.\u003c/p\u003e\n\u003cp\u003eColumn (2) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e reports the results. The estimated coefficient on did is larger in Hubei, indicating that the income effect of flood control is stronger in Hubei Province than elsewhere. We then test whether the effect declines with geographic distance from the project site. Specifically, we augment Eq.\u0026nbsp;(\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e) by interacting did with a measure of distance to the Three Gorges Project. Column (3) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e shows that the interaction term is negative and statistically significant, implying that the estimated income gains are concentrated closer to the project and diminish as distance increases. Together, these findings support the interpretation that Hubei is most exposed to the project\u0026rsquo;s flood control benefits and that the effect attenuates with distance.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab7\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eHeterogeneity analysis\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(1)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(2)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(3)\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e(4)\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eeq\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.059\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.039\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.189\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.038\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.007)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.009)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.008)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid* Eastern\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.056\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.012)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid* Hubei Province\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.041\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.011)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid* Distance from the Three Gorges\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.022\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.009)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003edid* Probability of flooding\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.345\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.066)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd rowspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e_cons\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.144\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.118\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.120\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.130\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.050)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.051)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.051)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(0.051)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eN\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5580\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5513\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5245\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.986\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl variables\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIndividual fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eTime fixation effect\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYes\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\n\u003ch2\u003e5.3 Flood probability\u003c/h2\u003e\n\u003cp\u003eBecause our analysis focuses on the flood control function of the Three Gorges Project, we also examine whether the income effect varies with local flood risk. We split counties by flood probability and re-estimate the baseline specification. The results show that the effect of did is larger in counties with a higher probability of flooding. Column (4) of Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e reports these estimates. This heterogeneity is consistent with a simple marginal-benefit logic. Counties facing higher flood risk tend to experience larger expected losses from floods, including greater threats to grain production and more frequent damage to housing, farm equipment, and other essential assets. These shocks can reduce income and increase the risk of poverty. By lowering disaster risk, the flood control function of the Three Gorges Project delivers larger gains precisely in the counties that are more exposed and more vulnerable, leading to a stronger observed increase in farmers\u0026rsquo; disposable income.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"6. Further analysis","content":"\u003cp\u003eFinally, we examine whether the Three Gorges Project affects the urban rural income gap. Common prosperity requires not only aggregate growth but also a more inclusive distribution of income. In this sense, narrowing the income gap is a key component of common prosperity. We measure the urban rural income gap using the ratio of rural per capita disposable income to urban per capita disposable income, and we estimate the same DID framework using this outcome. To assess identification, we also conduct placebo and parallel trend tests analogous to those reported in the main analysis. Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e presents the results.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFurther analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eeq2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eeq2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eeq2\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003edid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.033\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.015\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.027\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eIs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.132\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.094\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.018\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eGov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.340\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.465\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.024)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.026)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eFin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.076\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eHc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.036\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.015\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.059\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.006)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.006)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.006)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eGdp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.001\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.002\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e_cons\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.677\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.456\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.092\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.038)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.034)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.037)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3318\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3318\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3317\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.461\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.869\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndividual fixation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime-fixed effect\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e \u003cb\u003eabout here\u003c/b\u003e\u003c/p\u003e \u003cp\u003eAfter adding the full set of controls and urban fixed effects, the estimated coefficient on the Three Gorges Project policy is 0.027 and is statistically significant at the 1 percent level. This estimate implies that the project\u0026rsquo;s flood control function increases the ratio of rural per capita disposable income to urban per capita disposable income in flood-prone counties in the middle and lower Yangtze River basin by about 0.027 percentage points. Two considerations may help explain this result. First, rural areas typically start from a lower level of infrastructure, so the marginal gains from flood control and related water management improvements can be larger. Urban areas generally already have more complete systems for flood protection, water supply, power, transport, and emergency response, which can limit the incremental benefits from an additional large project. In contrast, many rural counties rely more directly on natural conditions for agricultural production, have weaker disaster resilience, and face higher costs of transport and logistics. These features make rural livelihoods more exposed to flood shocks. When a water conservancy project reduces flood risk and improves water availability, rural production and income may therefore respond more strongly. Second, the construction and operation of large water conservancy projects are often accompanied by policy support that is tilted toward rural and less developed areas. Examples include irrigation-related support, preferential electricity pricing, and complementary public investments. Such measures can reinforce the direct risk-reduction effects of the project by lowering production costs and easing constraints on rural development, which can contribute to a relative improvement in rural incomes compared with urban incomes.\u003c/p\u003e"},{"header":"6. Conclusions","content":"\u003cp\u003eThis paper uses a DID tool to evaluate the impact of the Three Gorges Project\u0026rsquo;s flood control function on farmers\u0026rsquo; income. The results indicate that large-scale flood control infrastructure can contribute to income growth and to the goal of common prosperity. First, the flood control function of the Three Gorges Project significantly increases farmers\u0026rsquo; per capita disposable income in flood-prone counties in the middle and lower Yangtze River basin, and the finding is stable across a wide set of robustness checks. In addition, the project not only raises farmers\u0026rsquo; income but also increases the ratio of rural to urban per capita disposable income, implying a gradual narrowing of the urban-rural income gap. Second, mechanism evidence suggests that the income gains operate through three channels: stronger agricultural production capacity, greater entrepreneurship, and higher investment. Third, heterogeneity analyses show that the effects are larger in the central region, in Hubei Province, and in counties with higher flood probability. Taken together, these results suggest that the Three Gorges Project reduces flood risk while also creating a more stable environment for agricultural production and local economic activity. More broadly, the project appears to generate benefits beyond disaster prevention by supporting more balanced regional development and narrowing income disparities. The analysis offers evidence that can inform the evaluation of water conservancy projects and the design of future investments.\u003c/p\u003e \u003cp\u003eBased on these findings, we draw two policy implications. First, the layout of water conservancy investments should be better aligned with local risk exposure and development needs. In high-risk flood-prone areas, including Hubei Province and counties along the middle and lower Yangtze River, policy should prioritize complementary small and medium-sized facilities that strengthen local irrigation and flood control networks. In less developed regions, water conservancy projects can be bundled with transport, electricity, and related infrastructure under the rural revitalization agenda to improve overall connectivity and amplify returns. Where conditions permit, integrating water conservancy investments with ecological restoration and local services can broaden benefits, for example by supporting ecological agriculture and rural tourism.\u003c/p\u003e \u003cp\u003eSecond, policy should strengthen the channels through which risk reduction translates into sustained income growth, especially entrepreneurship and investment. For entrepreneurship, local governments can improve access to credit and training for farmers in flood-prone areas, reduce administrative barriers for new businesses, and support organizational forms such as cooperatives and family farms that can raise productivity and market access. For investment, policies that attract private capital into water-related and agriculture-adjacent sectors, such as cold-chain logistics and agricultural technology, can help convert improved water security into durable local growth. Clear rules, transparent project selection, and well-designed PPP arrangements, together with targeted support measures such as fiscal incentives, can increase investment willingness and improve the long-run income effects of water conservancy projects.\u003c/p\u003e \u003cp\u003eFrom an international perspective, the findings of this paper are consistent with evidence from other countries that large-scale flood control and water management infrastructure can generate substantial economic returns beyond risk mitigation. Studies on major river basin projects such as the Mississippi River flood control system in the United States, the Delta Works in the Netherlands, and multipurpose dams in countries like India and Brazil show that improved flood protection and water regulation can stabilize agricultural output, encourage private investment, and support structural transformation in rural areas. In developing economies, flood control projects along the Mekong and Ganges-Brahmaputra river systems have been found to reduce income volatility for farmers and facilitate the expansion of non-agricultural activities by lowering disaster-related uncertainty. However, international experience also suggests that the distributional effects of such projects depend critically on complementary institutions and policies, including land tenure security, access to finance, and local governance capacity. In line with this global evidence, the Three Gorges Project offers additional insights from China, showing that large-scale water conservancy investments, when implemented within a coordinated development framework, can simultaneously enhance disaster resilience and foster more inclusive income growth. This cross-country perspective underscores the broader relevance of China\u0026rsquo;s experience for other flood-prone developing regions facing similar trade-offs between risk reduction and long-term development.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflicts of interests:\u003c/h2\u003e \u003cp\u003eNone\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eNote\u003c/h2\u003e \u003cp\u003eThe table summarizes all the variables in the baseline regression and robustness check.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003eNone.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eBin Xu: Conceptualization; Methodology; Formal analysis; Writing \u0026ndash; original draft; Visualization.Ke Wei: Data curation; Investigation; Writing \u0026ndash; original draft; Writing \u0026ndash; review \u0026amp; editing; Validation.Zhiyang Shen: Supervision; Project administration; Writing \u0026ndash; original draft; Writing \u0026ndash; review \u0026amp; editing.\u003c/p\u003e\u003ch2\u003eData availability:\u003c/h2\u003e \u003cp\u003eData will be made available on the request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlemayehu YA, Ali AS, Mersha GT, Tesfahun T, Mengiste BM (2025) Extreme weather impacts on the socio-economic conditions of rural communities in Ethiopia: practical implications and recommendations for resilience and sustainability. Agricultural Food Econ 13(1):65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1186/s40100-025-00411-4\u003c/span\u003e\u003cspan address=\".10.1186/s40100-025-00411-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBatista ICA, Firme VDAC (2025) Economic impacts of the Brazil\u0026rsquo;s Mariana dam disaster on different levels of neighbourhood. Spat Economic Anal 1\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17421772.2025.2502355\u003c/span\u003e\u003cspan address=\"10.1080/17421772.2025.2502355\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBosheng ZHANG, Xiaoyang CAO (2021) Impact of targeted poverty alleviation policy on the urban-rural income gap in national poverty-stricken counties. Resour Sci 43(8):1549\u0026ndash;1561 (in Chinese). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18402/resci.2021.08.04\u003c/span\u003e\u003cspan address=\"10.18402/resci.2021.08.04\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBowen R, Morris W (2019) The digital divide: Implications for agribusiness and entrepreneurship. Lessons from Wales. J Rural Stud 72:75\u0026ndash;84. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jrurstud.2019.10.031\u003c/span\u003e\u003cspan address=\"10.1016/j.jrurstud.2019.10.031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai X, Yao L (2024) Assessing the option water-saving willingness of irrigation areas considering farmers\u0026rsquo; risk tolerance. Risk Anal 44(1):70\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/risa.14140\u003c/span\u003e\u003cspan address=\"10.1111/risa.14140\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Y, Chen S (2022) Rebel on the canal: Disrupted trade access and social conflict in China, 1650\u0026ndash;1911. Am Econ Rev 112(5):1555\u0026ndash;1590. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/aer.20201283\u003c/span\u003e\u003cspan address=\"10.1257/aer.20201283\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDell M, Jones BF, Olken BA (2014) What do we learn from the weather? The new climate-economy literature. J Econ Lit 52(3):740\u0026ndash;798. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/jel.52.3.740\u003c/span\u003e\u003cspan address=\"10.1257/jel.52.3.740\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFabri C, Vermeulen S, Van Passel S, Schaub S (2024) Crop diversification and the effect of weather shocks on Italian farmers' income and income risk. J Agric Econ 75(3):955\u0026ndash;980. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1477-9552.12610\u003c/span\u003e\u003cspan address=\"10.1111/1477-9552.12610\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo X, Wang L, Meng X, Dong X, Gu L (2023) The impact of digital inclusive finance on farmers' income level: evidence from China's major grain production regions. Finance Res Lett 58:104531. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.frl.2023.104531\u003c/span\u003e\u003cspan address=\"10.1016/j.frl.2023.104531\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaishan C, Yang W (2022) Flood Control in Super Large Water Conservancy Facilities and Regional Economic Development: A Case Study of Three Gorges Project. World Econ 05137\u0026ndash;161. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.19985/j.cnki.cassjwe.2022.05.007\u003c/span\u003e\u003cspan address=\"10.19985/j.cnki.cassjwe.2022.05.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e(in Chinese)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang T (2022) Mediating effects and moderating effects in causal inference. China Industrial Econ 5:100\u0026ndash;120 (in Chinese). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.19581/j.cnki.ciejournal.2022.05.005\u003c/span\u003e\u003cspan address=\"10.19581/j.cnki.ciejournal.2022.05.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJunqian W, Xingmin Z (2023) Can Grain Production Incentives Promote Farmers\u0026rsquo; Incomes and County Economic Development? A Quasi-natural Experiment Based on the Incentive Policy for Large Grain-producing Counties. J Finance Econ 49(01):124\u0026ndash;138 (in Chinese). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.16538/j.cnki.jfe.20221022.402\u003c/span\u003e\u003cspan address=\"10.16538/j.cnki.jfe.20221022.402\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePagan-Cajigas Z, Guikema S, Otaduy‐Ramirez R, Woolley V, Smith K, Hu T, Chen T (2025) Water outage predictions for natural hazards using synthetic water distribution systems. Risk Anal. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/risa.70004\u003c/span\u003e\u003cspan address=\"10.1111/risa.70004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKates RW, Clark WC, Corell R, Hall JM, Jaeger CC, Lowe I, Svedin U (2001) Sustainability science. Science 292(5517):641\u0026ndash;642. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1059386\u003c/span\u003e\u003cspan address=\"10.1126/science.1059386\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKocornik-Mina A, McDermott TK, Michaels G, Rauch F (2020) Flooded cities. Am Economic Journal: Appl Econ 12(2):35\u0026ndash;66. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/app.20170066\u003c/span\u003e\u003cspan address=\"10.1257/app.20170066\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoks EE, Jongman B, Husby TG, Botzen WJ (2015) Combining hazard, exposure and social vulnerability to provide lessons for flood risk management. Environ Sci Policy 47:42\u0026ndash;52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envsci.2014.10.013\u003c/span\u003e\u003cspan address=\"10.1016/j.envsci.2014.10.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeiter AM, Oberhofer H, Raschky PA (2009) Creative Disasters? Flooding Effects on Capital, Labour and Productivity Within European Firms. Environ Resource Econ 43(3):333\u0026ndash;350. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10640-009-9273-9\u003c/span\u003e\u003cspan address=\"10.1007/s10640-009-9273-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiebenehm S, Schumacher I, Strobl E (2024) Rainfall shocks and risk aversion: Evidence from Southeast Asia. Am J Agric Econ 106(1):145\u0026ndash;176. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ajae.12403\u003c/span\u003e\u003cspan address=\"10.1111/ajae.12403\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu W, Qi P, Xu J, Feldman M, Xu D (2025) Does livelihood risk matter in disaster preparedness? Insights from flood risk areas of rural China. Clim Risk Manage 48:100705. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.crm.2025.100705\u003c/span\u003e\u003cspan address=\"10.1016/j.crm.2025.100705\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo Y, Liming, Fan (2012) Spatial Characteristics of the Income Effects of Rural Infrastructure in China: Evidence from Spatial Dependence and Spatial Heterogeneity. Manage World 571\u0026ndash;87 (in Chinese). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.19744/j.cnki.11-1235/f.2012.05.007\u003c/span\u003e\u003cspan address=\"10.19744/j.cnki.11-1235/f.2012.05.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMushi GE, Di Marzo Serugendo G, Burgi P-Y (2022) Digital Technology and Services for Sustainable Agriculture in Tanzania: A Literature Review. Sustainability 14(4). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su14042415\u003c/span\u003e\u003cspan address=\"10.3390/su14042415\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNunn N, Qian N (2014) US food aid and civil conflict. Am Econ Rev 104(6):1630\u0026ndash;1666. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/aer.104.6.1630\u003c/span\u003e\u003cspan address=\"10.1257/aer.104.6.1630\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi G, Pang J, Wang Y (2002) An Analysis of the Correlation between Water Conservancy Project Construction and Farmers\u0026rsquo; Income. Chin Rural Econ 434\u0026ndash;39 (in Chinese). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.20077/j.cnki.11-1262/f.2002.04.006\u003c/span\u003e\u003cspan address=\"10.20077/j.cnki.11-1262/f.2002.04.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan J (2025) Abnormal Weather and Household Consumption. Rev Dev Econ. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/rode.70047\u003c/span\u003e\u003cspan address=\"10.1111/rode.70047\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang CL, Yan JD, He XD, Tian SQ (2023) What determines the survival of farmer entrepreneurship: Micro-evidence from China. Int Rev Econ Finance 86:334\u0026ndash;348\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan B, Sun W, Huang J (2025) Once Bitten, Twice Shy: The Impact of Natural Disasters on the Adoption of Green Production Technologies by Farmers Based on the Risk Aversion Perspective. Twice Shy: The Impact of Natural Disasters on the Adoption of Green Production Technologies by Farmers Based on the Risk Aversion Perspective.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ldr.70388\u003c/span\u003e\u003cspan address=\"10.1002/ldr.70388\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"the-annals-of-regional-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"arsc","sideBox":"Learn more about [The Annals of Regional Science](https://link.springer.com/journal/168)","snPcode":"168","submissionUrl":"https://submission.springernature.com/new-submission/168/3","title":"The Annals of Regional Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Water Conservancy Project, Three Gorges Project, Farmers' income, Common prosperity","lastPublishedDoi":"10.21203/rs.3.rs-8882775/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8882775/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGlobally, large-scale water conservancy projects are widely used as a policy instrument to enhance climate resilience, stabilize agricultural production, and promote inclusive growth, yet credible evidence on their income and distributional effects in developing economies remains limited. This paper studies whether large-scale water infrastructure can contribute to these outcomes by examining the Three Gorges Project in China. We use a balanced county-level panel of 351 counties in the middle and lower Yangtze River basin from 2003 to 2020 and estimate a difference in differences design that exploits variation in exposure to the project\u0026rsquo;s flood control benefits over time. We find that the flood control function of the Three Gorges Project increases rural residents\u0026rsquo; per capita disposable income, and the result is robust to a range of alternative specifications and placebo tests. Mechanism evidence suggests three channels: higher agricultural productivity, stronger household entrepreneurship and nonfarm business activity, and greater local investment. The income effect is larger in Central China and in counties with higher flood risk. We also find that the project reduces the urban rural income gap, consistent with water infrastructure supporting shared prosperity.\u003c/p\u003e","manuscriptTitle":"Water conservancy and farmers' income: A quasi-natural experiment based on the Three Gorges project","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-09 08:42:15","doi":"10.21203/rs.3.rs-8882775/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-09T21:14:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"41191547197769696330855617979329439873","date":"2026-05-07T01:27:52+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-04T14:36:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-16T04:34:17+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-16T04:32:36+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Annals of Regional Science","date":"2026-02-14T22:30:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-annals-of-regional-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"arsc","sideBox":"Learn more about [The Annals of Regional Science](https://link.springer.com/journal/168)","snPcode":"168","submissionUrl":"https://submission.springernature.com/new-submission/168/3","title":"The Annals of Regional Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"500d0968-e641-4134-a035-74bda38a03db","owner":[],"postedDate":"March 9th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-09T21:14:00+00:00","index":18,"fulltext":""},{"type":"reviewerAgreed","content":"41191547197769696330855617979329439873","date":"2026-05-07T01:27:52+00:00","index":16,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T08:42:15+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-09 08:42:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8882775","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8882775","identity":"rs-8882775","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}