The research team, led by Miao Qian, Xiuyu Huang, and Jiaomeng Li of Qingdao Agricultural University, constructed a multidimensional evaluation system to measure both cooperative development and collective economic development across each province. For cooperatives, the index captured five dimensions: standardized operation, operational performance, service scope, driving effect, and industrial upgrading, built from 18 underlying indicators such as average operating income per cooperative, farmer membership rates, and the share of cooperatives holding registered trademarks. The collective economy was measured across four dimensions covering scale and efficiency, profit distribution and accumulation, organizational management, and public service provision. Rather than relying on subjective weighting, the researchers used the entropy method, which assigns weights based on the information content of the data itself, reducing the bias that plagues expert-driven scoring systems.

The temporal picture that emerged is strikingly asymmetric. Between 2015 and 2023, the composite development score for farmers’ cooperatives actually declined slightly, falling from 3.499 to 3.324, a drop of roughly 5 percent. The rural collective economy, by contrast, grew steadily and substantially, its composite score more than doubling from 1.608 to 3.471. Within the cooperative sector, only industrial upgrading improved consistently, with its score rising from 0.695 to 0.992, driven by explosive growth in processing-oriented cooperatives, whose numbers expanded from 26,364 to 121,378, an average annual growth rate of about 21 percent. Service-oriented cooperatives grew from 108,700 to 162,560 over the same period. The data suggest a sector shifting from quantity to quality, even as its overall composite performance slipped.

To establish causality rather than mere correlation, the team employed a two-way fixed-effects panel regression that controls for both unchanging provincial characteristics and nation-wide temporal shocks. The baseline result showed that a one-unit increase in the level of cooperative development raises the level of collective economic development by an average of 0.351 units, a coefficient that remained highly significant at the 1 percent level across multiple model specifications. The researchers then confronted the classic endogeneity problem: prosperous villages might simply build better cooperatives, reversing the direction of causation. Using the one-period lag of cooperative development as an instrumental variable in a two-stage least squares estimation, they found the effect not only survived but grew slightly, to 0.436, with a first-stage F-statistic of 23.74 comfortably exceeding the Stock–Yogo threshold for instrument strength.

Robustness checks reinforced the conclusion. Replacing the entropy weighting with grey correlation analysis, an entirely different algorithm for constructing the composite index, produced a coefficient of 1.015 that remained significant at the 1 percent level, demonstrating that the result does not hinge on how the index was assembled. Adding further control variables for industrialization and trade openness left the core coefficient at 0.278, still significant. Interestingly, the degree of openness carried a significantly negative coefficient, suggesting that export-oriented economies can siphon resources away from local collective development, a nuance that underscores how global integration and village-level prosperity do not always pull in the same direction.

Perhaps the most consequential finding concerns how cooperatives actually deliver their benefit. When the researchers decomposed cooperative development into its five dimensions and tested each against the collective economy, only two channels proved significant: operational performance and industrial upgrading. Standardized operation, service scope, and the direct driving effect on farmers showed no significant impact. In other words, it is not the mere existence, formal compliance, or membership reach of cooperatives that powers collective growth, but their actual economic performance and their capacity to move up the value chain through branding, quality certification, and processing. This helps explain the well-documented phenomenon of hollow shell cooperatives in China, entities registered on paper that generate no substantive economic activity and, according to these results, no measurable collective benefit.

Geography matters enormously. When the sample was split into eastern, central, and western regions, the promotional effect was significant only in the east, with a coefficient of 0.318 at the 5 percent level, while the central and western coefficients of −0.078 and 0.085 failed every significance test. The authors attribute this to the east’s high degree of agricultural marketization, mature industrial chains, and standardized cooperative operations, which allow performance gains to transmit efficiently to village collectives. In the central and western provinces, cooperatives started later, remain smaller in scale, and lack the market infrastructure to convert organizational form into economic output. The lesson is uncomfortable for one-size-fits-all policymaking: the same institutional design produces radically different results depending on the surrounding economic ecosystem.

Two further nonlinear patterns sharpen the picture. First, the growth rate of the agricultural industry, measured as value added across farming, forestry, animal husbandry, and fisheries, acts as a positive moderator: as the moderator rises from its mean minus one standard deviation to its mean plus one standard deviation, the marginal effect of cooperatives on the collective economy climbs monotonically from 0.321 to 0.494. Vibrant agricultural sectors expand market demand, improve processing and logistics infrastructure, and lower the transaction costs of resource integration. Second, land transfer exhibits a single-threshold effect at a value of 0.3903. Below that threshold, the cooperative coefficient is 0.459; above it, the coefficient jumps to 0.626. Only once land consolidation reaches a critical scale can cooperatives fully exploit mechanization, centralized procurement, and branded marketing, unlocking economies of scale that fragmented smallholder plots cannot achieve.

The most surprising result may be spatial. Using a spatial Durbin model with an economic distance weight matrix, the researchers found that cooperative development in one province significantly boosts the collective economy of neighboring, economically linked provinces. The decomposition of effects is remarkable: the direct local effect is 0.383, but the indirect spillover effect is 1.198, more than three times larger, yielding a total effect of 1.581, all significant at the 1 percent level. Cross-regional diffusion of governance models, extension of market linkage networks, and the mobility of capital, talent, and agricultural services appear to dominate the story. Cooperative development, in this reading, is less a local intervention than a contagious one, with benefits propagating across provincial borders faster than they accumulate at home.

The policy implications are pointed. The authors argue that China should shift from subsidizing cooperative numbers to phasing out shell entities, strengthening annual reporting and dynamic exit mechanisms, and supporting chains that extend into processing, warehousing, logistics, and rural tourism. Eastern provinces should consolidate replicable models of deep cooperative-collective integration, while central and western regions need targeted fiscal support, talent deployment, and industrial chain repair before their cooperatives can deliver. Because spillovers are so large, removing barriers to cross-regional factor flows and building interprovincial cooperative federations could multiply returns. And because the land transfer threshold is real, orderly land consolidation emerges as a prerequisite rather than an afterthought. For a country betting heavily that organized smallholders can fund common prosperity, this study offers both a validation and a warning: the cooperative engine works, but only when it is fueled by genuine performance, adequate scale, and a living industrial environment.

Subject of Research: The impact of farmers' cooperatives on the development of China's rural collective economy

Article Title: The enabling effect of farmers’ cooperatives on the rural collective economy: evidence from China

Article References: Qian, M., Huang, X., & Li, J. (2026). The enabling effect of farmers’ cooperatives on the rural collective economy: evidence from China. Frontiers in Sustainable Food Systems, 10, Article 1971166. https://doi.org/10.3389/fsufs.2026.1971166

Image Credits: AI Generated

DOI: 10.3389/fsufs.2026.1971166

Keywords: farmers' cooperatives, rural collective economy, China, panel data, spatial spillover effects, land transfer, industrial upgrading, agricultural economics, rural revitalization, threshold effect, entropy method, two-way fixed-effects model