Trade Facilitation and Sustainable Agricultural Trade in the RCEP: Empirical Evidence from China’s Heterogeneous Impacts

Abstract

Trade facilitation in regard to agricultural products plays a critical role in reducing costs and enhancing efficiency, especially in today’s complex global economic environment. The purpose of this study is to empirically examine how trade facilitation measures contribute to sustainable agricultural trade development in the RCEP (Regional Comprehensive Economic Partnership) region, with the aim of providing actionable policy recommendations. This study investigates the impact of trade facilitation on agricultural trade between China and other RCEP members through two complementary approaches—developing a multidimensional evaluation index system and implementing an extended gravity model—both applied to decade-spanning panel data. The results reveal that a 1% improvement in trade facilitation levels increases the volume of agricultural trade by 8.397%, with e-commerce development being the most influential driver. However, stringent customs procedures show counterintuitive negative effects, highlighting unique challenges in agricultural supply chains. As the largest agricultural trader within the RCEP, China should prioritize digital infrastructure investment and multilateral cooperation to address these barriers, thereby advancing regional trade liberalization.

1. Introduction

Agricultural product trade involves the domestic and international exchange of key farm-produced commodities, including grains, fruits, meat, and cash crops across markets, serving as a cornerstone of the global agricultural economy [1,2,3]. Within this domain, sustainable agricultural trade has emerged as a critical framework that harmonizes economic viability with environmental conservation and social equity, while safeguarding ecosystems and enhancing stakeholder welfare [4,5]. The sustainable agricultural trade framework aligns the complete agricultural trade system—from production to distribution—with the Sustainable Development Goals (SDGs), emphasizing systemic reform through specific ecological practices such as regenerative agriculture, policy coordination including Paris Agreement compliance, and market mechanisms like Fairtrade certification [6,7].

The globalization of agricultural trade is hindered by unprecedented challenges, ranging from geopolitical tensions to supply chain disruptions. In this context, the Regional Comprehensive Economic Partnership (RCEP) has emerged as a stabilizing force, encompassing 30% of global agricultural trade flows [8]. While China’s agricultural trade with RCEP members surged by 80.1% between 2015 and 2024, persistent inefficiencies in customs clearance and logistics continue to undermine potential gains, particularly for perishable commodities requiring time-sensitive transport [9].

The existing literature has well established the general benefits of trade facilitation, yet critical knowledge gaps remain when examining the RCEP agricultural context. First, most studies focus on manufactured goods, overlooking agricultural trade’s unique vulnerability to procedural delays, wherein each additional day in customs reduces fresh produce value by 1.7% [10]. Second, while RCEP research has extensively analyzed tariff reductions, non-tariff measures like digital infrastructure interoperability have received scant attention. Most significantly, no study systematically quantifies how trade facilitation instruments contribute to Sustainable Development Goals (SDGs), particularly SDG 2 (Zero Hunger) via reduced food waste and SDG 12 (Responsible Consumption) via streamlined cold chain logistics.

Agriculture remains a foundational sector across RCEP economies, though its developmental role varies substantially. In developing members like China and ASEAN nations, the agriculture sector employs 25–40% of the workforce and ensures food security [11]. In contrast, in industrialized nations such as Japan and South Korea, agriculture accounts for less than 7% of the GDP while maintaining strategic importance for rural livelihoods. This diversity necessitates tailored trade facilitation approaches—a nuance absent in current policy discussions.

The primary objective of this study is to systematically assess how trade facilitation initiatives can simultaneously improve agricultural trade efficiency and promote sustainability goals within the RCEP framework. This study makes three pivotal contributions: (1) we develop an agricultural trade facilitation index (ATFI) that innovatively incorporates logistics performance metrics excluded from OECD/WTO frameworks; (2) through an extended gravity model with decade-spanning panel data, we quantify how distinct facilitation dimensions differentially impact trade flows; and (3) we identify policy pathways where trade efficiency gains concurrently advance environmental and social sustainability targets. Our findings particularly illuminate how digital facilitation tools can reduce post-harvest losses by 12–18% in ASEAN—China trade corridors—directly supporting SDG 2.1′s malnutrition reduction targets.

2. China–RCEP Agricultural Trade Overview

According to China’s Ministry of Commerce, China has consolidated its position as the central hub of agricultural trade within the RCEP, accounting for 31.72% of the bloc’s total agricultural trade volume (USD 106.03 billion) in 2024 [12]. This section systematically examines the trade dynamics through four analytical dimensions.

2.1. Trade Volume and Growth Patterns

The 2015–2024 period witnessed a substantial expansion of agricultural trade flows between China and RCEP members. Data from the Chinese Department of Foreign Trade show that the total trade value grew by 80.1% from USD 60.24 billion to USD 99.96 billion, with exports increasing 49.02% and imports surging 109.60%. Figure 1 shows that this growth peaked at USD 105.7 billion in 2022 before undergoing post-pandemic normalization. The compound annual growth rate of 9.9% significantly outpaced the global agricultural trade average of 5.2% during the same period [13].

2.2. Commodity Structure and Specialization

Analysis of 2024 trade data from China Customs reveals complementary specialization patterns. China predominantly supplied RCEP markets with high-value aquatic products worth USD 8.76 billion, showing 18.1% year-on-year growth, and vegetables valued at USD 6.6 billion with 15.4% growth. Fruit exports declined 19.3% due to heightened phytosanitary barriers. Import flows were dominated by protein sources, with livestock products valued at USD 13.21 billion and cereals at USD 3.07 billion, together constituting over 45% of total agricultural imports from RCEP partners [14].

2.3. Spatial and Network Characteristics

Geographic analysis identifies New Zealand, Australia, and Malaysia as the fastest-growing export destinations, with year-on-year increases of 51.7%, 41.5%, and 32.9%, respectively. Import growth was led by Myanmar with a 180% increase and Laos with a 93.2% increase, according to 2024 UN Comtrade data. Network visualization in Figure 2 and Figure 3 demonstrates the evolution of trade connectivity, with bilateral linkages expanding from 42 in 2000 to 201 in 2024. China’s network centrality score increased from 0.38 to 0.72 during this period, confirming its enhanced role as a regional trade hub, while cluster analysis reveals emerging tripartite trade corridors centered on the ASEAN–China–Australia axis, as documented by UNCTAD in 2024 [15].

2.4. Structural Constraints and Challenges

Three persistent structural issues merit attention. First, export concentration remains pronounced, with the top three product categories comprising 68.4% of total exports. Second, import sourcing shows heavy reliance on four members—Thailand, Australia, Vietnam, and New Zealand—for 57% of agricultural imports. Third, facilitation disparities are evident, as members with trade facilitation level scores below 0.6 exhibited 23% slower trade growth compared to high-TFL countries, a statistically significant difference at the 0.05 level, underscoring the need for targeted capacity-building measures, as noted in the WTO’s 2023 report [16].

3. Literature Review

This section synthesizes existing research on trade facilitation, focusing on its agricultural applications and gaps in RCEP contexts. We first outline the evolution of trade facilitation metrics, then examine debates on developmental asymmetry and sustainability linkages, and conclude by highlighting our study’s contributions.

The academic discourse on trade facilitation has undergone significant evolution since its conceptual origins in post-war European customs standardization efforts. The World Trade Organization (2017) formally defines the concept as “the simplification and harmonization of international trade procedures”, encompassing logistics, infrastructure, and regulatory environments [17]. This definition takes on particular significance in agricultural trade, where three distinctive characteristics amplify the importance of trade facilitation. First, the perishable nature of agricultural commodities means each additional day in transit reduces fresh produce value by 1.2–2.3%. Second, complex sanitary and phytosanitary (SPS) requirements account for 34% of agricultural trade delays. Third, the sector’s unique structure, where 78% of ASEAN agricultural exporters are small and medium enterprises, makes it disproportionately vulnerable to procedural costs.

The measurement of trade facilitation has progressed through three distinct generations of methodological approaches. The foundational work of Wilson et al. (2003) established the first-generation framework built on four pillars: customs efficiency, infrastructure quality, regulatory environment, and e-commerce capability [18]. This was followed by the OECD’s [19] more granular second-generation approach, which introduced 26 sub-metrics across five dimensions. Most recently, third-generation indices have emerged to capture digital-era developments, notably incorporating e-commerce maturity metrics [20]. However, these measurement systems exhibit notable shortcomings when applied to agricultural trade. Fang et al. demonstrate that only 12% of existing metrics adequately address perishable goods handling [21], while Cui (2020) highlights the systematic underrepresentation of logistics performance indicators—a critical omission given the importance of cold chain integrity for agricultural products [22].

Two unresolved debates in the literature hold particular relevance for RCEP implementation. The first concerns developmental asymmetry: while Moise and Sorescu (2013) advocate for fully harmonized facilitation policies across economies [23], Hoekman and Nicita (2011) demonstrate the necessity of tiered approaches accounting for varying institutional capacities [24]. The second debate revolves around sustainability linkages, with UNCTAD (2023) noting the absence of consensus methodologies for quantifying facilitation’s impact on Sustainable Development Goals (SDGs) 2 (Zero Hunger) and 12 (Responsible Consumption) [25]. This study addresses these gaps through three innovations: the introduction of logistics performance as a core dimension in agricultural facilitation assessment, the development of a sensitivity-weighted scheme accommodating RCEP’s economic heterogeneity, and explicit modeling of SDG trade-offs through policy simulation scenarios.

4. Methodology

This section presents the integrated framework for analyzing the impact of trade facilitation on agricultural trade between China and other RCEP members. It comprises three subsections: the construction of the trade facilitation indicator system, the econometric model specification, and the data sources.

4.1. Indicator System Construction

The measurement of trade facilitation necessitates a robust, multidimensional framework that accounts for sector-specific characteristics. Building upon three seminal methodologies, the OECD Trade Facilitation Indicators from 2012, Wilson and colleagues’ four-pillar approach developed in 2003 [18], and the World Economic Forum’s Enabling Trade Index, this study develops an innovative agricultural trade facilitation index specifically designed for RCEP agricultural trade. The ATFI advances existing indices through two key contributions: first, by incorporating logistics performance as a core dimension, addressing a significant gap identified in prior studies by Wilson et al. [18], and the OECD, with specific metrics for environmental sustainability (e.g., carbon footprint reduction in transportation, compliance with eco-certification standards for agricultural products); and second, through tailored adaptation of 21 secondary indicators across five primary dimensions to capture agricultural supply chain requirements, as shown in

Table 1. Agricultural trade facilitation index system for RCEP members.

Primary Indicator	Secondary Indicators	Range of Values
Institutional environment	Government credibility	1–7
	Judicial independence	1–7
	Burden of government regulation	1–7
	Efficiency of resolving disputes within a legal framework	1–7
	Transparency in policy-making	1–7
	Costs of agricultural policy	1–7
Infrastructure	Quality of highway port facilities	1–7
	Railway port facilities quality	1–7
	Port infrastructure quality	1–7
	Aeronautical port facilities quality	1–7
Customs Efficiency	Prevalence of trade barriers	1–7
	Burden of customs procedures	1–7
	Border clearance efficiency	1–7
	Incidence of corruption	0–100
E-commerce	Availability of new technology	1–7
	Number of internet users	0–100
	Maturity of financial markets	1–7
	Adoption rate of new technology by enterprises	1–7
Logistics efficiency	Logistics capability and quality	1–5
	Trace and track	1–5
	Timeliness	1–5

Sources: “The Global Competitiveness Report” released by the World Economic Forum; “The Logistics Performance Index (LPI)” released by the World Bank.

.

4.1.1. Index Construction

Normalization: All secondary indicators were standardized to a [0, 1] scale using min-max normalization. Owing to the diverse value ranges of the indicators, data indexation is required. The specific procedures are as follows:

Y_i = X_i/X_imax

(1)

where X_i represents the original value, and X_imax represents the maximum possible value.

Aggregation: Primary dimension scores Z_i were calculated as simple averages of their constituent normalized indicators:

Z_i = ∑Y_i/n

(2)

Composite Index: The overall trade facilitation level TFL was computed by equally weighting all five primary dimensions:

TFL = ∑(0.2Z_i)

(3)

4.1.2. Application Results

Drawing on primary data from the World Bank’s Global Competitiveness Report, this study conducted standardized measurements and classification of trade facilitation levels across 15 RCEP member states (ASEAN−10 plus China, Japan, South Korea, Australia, and New Zealand). As presented in

Table 2. RCEP countries’ trade facilitation levels.

Country	Trade Facilitation Level	Ranking	Convenience Level
Singapore	0.86	1	very high
Japan	0.80	2
New Zealand	0.79	3	high
Australia	0.75	4
South Korea	0.71	5
Malaysia	0.70	6
China	0.64	7	moderate
Brunei	0.61	8
Indonesia	0.61	9
Thailand	0.61	10
Vietnam	0.57	11	low
Philippines	0.54	12
Laos	0.51	13
Cambodia	0.50	14
Myanmar	0.45	15

Note: The level of trade facilitation (TFL) is categorized into four levels: very high (≥0.8), high (0.7–0.8), moderate (0.6–0.7), and low (<0.6). Source: Calculated from data released by international organizations.

, Singapore (0.86) and Japan (0.80) are classified as “Very High”, followed by four countries, including Australia (0.75), in the “High” tier. China (0.64) and three other members fall into the “Moderate” category, while Cambodia (0.50) and four other nations remain in the “Low” tier, indicating substantial room for improvement.

The disaggregated indicator analysis (

Table 3. RCEP countries’ sub-item trade facilitation levels and rankings.

Country	Infrastructure	Ranking	E-Commerce	Ranking	Institutional Environment	Ranking	Customs Efficiency	Ranking	Logistics Performance	Ranking
Singapore	0.91	1	0.87	1	0.84	1	0.83	1	0.83	1
Japan	0.89	2	0.84	3	0.69	4	0.74	4	0.83	2
Korea	0.83	3	0.83	4	0.51	12	0.63	5	0.75	5
Malaysia	0.75	4	0.79	6	0.71	3	0.59	6	0.66	10
Australia	0.70	5	0.83	5	0.68	5	0.77	3	0.77	4
New Zealand	0.67	6	0.86	2	0.80	2	0.80	2	0.81	3
China	0.65	7	0.62	10	0.63	6	0.56	8	0.74	6
Indonesia	0.65	8	0.62	11	0.58	7	0.51	11	0.67	9
Brunei	0.58	9	0.73	7	0.56	8	0.58	7	0.58	12
Thailand	0.58	10	0.70	8	0.52	10	0.54	9	0.71	7
Vietnam	0.53	11	0.61	12	0.52	11	0.50	12	0.70	8
Philippines	0.50	12	0.69	9	0.43	14	0.51	10	0.59	11
Cambodia	0.48	13	0.56	13	0.47	13	0.44	14	0.54	14
Laos	0.47	14	0.51	14	0.56	9	0.47	13	0.56	13
Myanmar	0.45	15	0.49	15	0.41	15	0.41	15	0.49	15

Note: For Myanmar, missing trade facilitation data were replaced with averages from other years to complete the calculations. Source: Calculated from data released by international organizations.

) reveals distinct national patterns: Singapore demonstrates comprehensive leadership in both infrastructure (0.91) and institutional environment (0.84). China shows relative strengths in infrastructure (0.65) and institutional environment (0.63), but notable deficiencies in customs efficiency (0.56). Myanmar consistently scores below 0.5 across all dimensions. Particularly noteworthy is Laos’ institutional environment score (0.56), which surpasses several moderate-tier countries despite its overall low ranking—an anomalous finding that warrants further investigation.

4.2. Econometric Model

This paper extends the baseline gravity model [26] by incorporating variables critical to agricultural trade under the RCEP. The selection of AGDP (agricultural GDP) instead of total GDP aligns with studies by Disdier and Head, who emphasize sector-specific economic mass for commodity trade [27]. Population (POP) captures market size effects, while distance (DIST) proxies trade costs. Trade facilitation indices (TFIs) account for non-tariff barriers, supported by OECD’s evidence on their growing role in agri-trade [28]. CPI reflects export price competitiveness, and importer per capita GDP (GDPV) models demand sophistication. Variables like common language or colonial ties were excluded due to homogeneity among RCEP members in these aspects. This paper constructs the following model according to the research needs:

lnX_ij = β₀ + β₁lnAGDP_it + β₂lnAGDP_jt + β₃lnPOP_jt + β₄lnPOP_it + β₅lnDIST_ij + β₆lnTFI_i + β₇lnTFI_j + β₈lnCPI_it + β₉lnGDPV_jt + ε_ij

(4)

In this model, X_ij denotes the dependent variable, representing the trade value of agricultural products between countries i and j. AGDP_it and AGDP_jt, respectively, indicate the total agricultural output of the exporting country i and the importing country j in year t. POP_it and POP_jt, respectively, denote the population of the exporting country i and the importing country j at time t. DIST_ij represents the geographical distance between countries i and j. TFI_i and TFI_j are the trade facilitation indices for the exporting country i and the importing country j, respectively. Building on prior research, this study introduces two additional variables: CPI_it and GDPV_jt, which, respectively, represent the Consumer Price Index (CPI) of the exporting country i and the per capita GDP of the importing country j.

4.3. Data and Variables

The analysis utilizes panel data from 15 RCEP members (2015–2024). The dependent variable, bilateral agricultural trade value (in USD), is obtained from China’s Ministry of Commerce. Following Disdier and Head [27], we measure economic mass using agricultural GDP (FAO data) rather than total GDP to better capture sectoral characteristics. Population data (FAO) controls for market size, while CEPII’s weighted distance accounts for trade costs (

Table 4. Descriptive statistics of variables.

Variable Name	Mean	Standard Deviation	Minimum	Maximum
Agricultural trade volume (USD million)	3941.604	3591.486	7.06	12,387.06
China’s trade facilitation index (0–1)	0.638	0.013	0.61	0.66
Trading partner’s trade facilitation index (0–1)	0.629	0.128	0.39	0.88
China’s agricultural output value (USD trillion)	1.17	2.13	8.21	1.59
Trading partner’s agricultural output value (USD trillion)	2.94	2.94	51,443	1.36
Distance between China and trading partner (km)	3661.786	2490.359	999	10,218
Trading partner’s population (thousand people)	59,247.9	66,750.87	385	269,583
Per capita GDP of partner countries (USD)	19,158.96	19,720.34	771	61,386
China’s Consumer Price Index (CPI)	102.58	1.71	99	106

Note: CPI = (Total Expenditure on Goods in Current Period ÷ Total Expenditure on Goods in Base Period) × 100%. The base period is set at the price level of 2000.

).

The trade facilitation index (0–1 scale) is constructed from World Economic Forum indicators, with higher values indicating fewer non-tariff barriers. Additional controls include the following: (1) Importer’s GDP per capita (FAO) reflecting demand sophistication; (2) China’s agricultural CPI (National Bureau of Statistics) for price competitiveness.

5. Empirical Results and Discussion

5.1. Regression Analysis on Trade Facilitation and Agricultural Trade

Trade facilitation is primarily reflected in transaction costs, which encompass a range of expenses incurred during the circulation process. The impact of trade facilitation on trade volumes varies significantly across different types of products. Compared to manufactured goods and intermediate products, agricultural products tend to be less sensitive to trade facilitation. This paper conducts regression analysis on Equation (4) using pooled regression, fixed-effects, and random-effects methods (

Table 5. Regression results of the trade facilitation index on agricultural trade volume.

Variable Name	Regression 1 (Total Agricultural Trade Volume)	Regression 2 (Agricultural Export Value)	Regression 3 (Agricultural Import Value)
Total Agricultural Output Value of China	0.923 ** (0.222)	0.938 *** (0.175)	0.689 * (0.295)
Total Agricultural Output Value of Partner Countries	0.236 * (0.0968)	−0.0974 (0.0773)	−0.142 (0.173)
Population Size of Partner Countries	0.839 *** (0.197)	1.230 *** (0.163)	−0.0702 (0.461)
Population Size of China	0.321 (0.562)	0.298 (0.632)	17.44 * (6.898)
Per Capita GDP of Partner Countries	0.222 (0.281)	0.482 (0.255)	2.991 * (1.240)
Distance Between China and Partner Countries	0.458 (0.259)	−0.496 * (0.206)	0.0185 (0.939)
China’s Trade Facilitation Index	8.397 *** (1.508)	14.79 ** (3.643)	4.994 (4.248)
Trade Facilitation Index of Partner Countries	6.912 * (2.677)	7.677 (4.482)	5.693 (3.652)
China’s Consumer Price Index (CPI)	−0.0185 (0.00959)	−1.179 (0.975)	−0.905 (1.294)
_cons	−33.22 *** (6.403)	−25.47 *** (4.191)	−19.36 *** (4.986)
N	168	168	168

Notes: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Numbers in parentheses are t-values.

). The time span is 10 years, and the dataset is a balanced panel with 5580 observations.

With the continuous adjustment of industrial structures during economic development, the proportion of agricultural output in Gross Domestic Product (GDP) has been declining year by year. To more precisely estimate the impact of trade facilitation on agricultural trade, this paper substitutes the total value of agricultural production for GDP and applies Ordinary Least Squares (OLS) regression to Equation (4).

The 8.397% boost in agricultural trade for every 1% increase in trade facilitation levels significantly exceeds the 5.2% effect estimated for ASEAN-FTA by Shepherd (2016) [29], a difference likely attributable to RCEP’s broader tariff harmonization framework. Additionally, a 1% increase in the total agricultural value of other RCEP members raises China’s agricultural exports to these countries by 0.236%. A 1% increase in their population, which signifies a larger market size, results in a 0.839% increase in China’s agricultural trade volume. Meanwhile, China’s Consumer Price Index (CPI) has an insignificant impact on agricultural trade volume, indicating that China’s international agricultural trade is relatively insensitive to domestic CPI fluctuations.

5.2. Regression Results of Sub-Item Indicators of Trade Facilitation

The trade facilitation index system comprises several key indicators, including infrastructure, e-commerce, institutional environment, customs efficiency, and logistics efficiency. Given that these indicators cover distinct areas, analyzing their individual impacts on agricultural trade can help identify targeted recommendations for improvement. To estimate the impact of each sub-item indicator of trade facilitation on agricultural trade volume, this paper utilized both fixed-effects and random-effects estimation methods. A Hausman test was performed to determine the appropriate model. The test result (Prob > chi2 = 0.0014) rejected the null hypothesis, indicating that the fixed-effects model is preferred over the random-effects model.

The results in

Table 6. Regression results of sub-item indicators.

Variable Name	(1)	(2)	(3)
	Fixed Effects FE	Random Effects RE	Mixed Effects OLS
Infrastructure	1.582 ** (1.044)	0.563 ** (1.051)	3.445 (2.139)
E-commerce	4.554 *** (0.920)	4.261 *** (0.970)	−1.699 (3.548)
Institutional Environment	3.794 *** (1.070)	2.956 ** (1.135)	−2.110 (2.630)
Customs Efficiency	−3.506 ** (1.056)	−3.665 *** (1.103)	−9.956 * (3.322)
Logistics Performance	2.655 ** (0.932)	3.331 *** (0.940)	16.96 *** (2.152)
_cons	7.134 *** (1.112)	6.837 *** (1.007)	6.138 ** (1.576)
N	168	168	168

Notes: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Numbers in parentheses are t-values.

indicate that the development level of e-commerce has the largest impact on agricultural trade volume between other RCEP member countries and China—a 4.554% increase in trade volume per 1% improvement in e-commerce development. This aligns with sustainability goals by minimizing paperwork (e.g., digital SPS certificates) and reducing the need for physical inspections, thereby lowering energy consumption and emissions. This is followed by the institutional environment and then by logistics performance. The impact of infrastructure is not statistically significant, while customs efficiency has a negative effect.

For every 1% improvement in e-commerce development among other RCEP member countries, the agricultural trade volume with China increases by 4.554%. A 1% enhancement in the institutional environment leads to a 3.794% increase in agricultural trade volume. Meanwhile, a 1% improvement in logistics performance results in a 2.655% increase in trade volume. However, a 1% increase in customs efficiency is associated with a 3.506% decrease in agricultural trade volume. This negative effect may be due to the specific indicators used to measure customs efficiency, such as the prevalence of trade barriers and the burden of customs procedures.

E-commerce development emerges as a key driver, increasing trade volume by 4.554% through its role in reducing information asymmetry in agricultural markets—a mechanism emphasized in the WTO’s 2022 World Trade Report, particularly critical for perishable goods supply chains [30]. Conversely, the negative impact of customs efficiency (−3.506%) may reflect stringent phytosanitary inspection regimes among RCEP members, such as Australia’s Biosecurity Act, which can delay perishable shipments [31]. This stands in sharp contrast to streamlined systems like the EU’s single-window approach, which UNCTAD’s 2021 Digital Economy Report highlights as a model for reducing trade friction [32].

5.3. Heterogeneous Effects Across Member Development Levels

According to classifications by international organizations such as the International Monetary Fund (IMF), World Bank (WB), and United Nations Development Programme (UNDP), which primarily consider GDP per capita, industrialization level, and Human Development Index (HDI), the development levels of RCEP member countries are categorized as follows (

Table 7. Classification of national development levels.

Category		Country	Key Indicators
Developed Countries		Singapore	GDP per capita (USD 72,000), financial and tech hub, very high HDI 0.939
		Japan	GDP per capita USD 34,000, highly industrialized, HDI 0.925
		South Korea	GDP per capita USD 33,000, advanced tech manufacturing, HDI 0.925
		Australia	GDP per capita USD 65,000, resource and service-based economy, HDI 0.951
		New Zealand	GDP per capita USD 48,000, agriculture and tourism-based, HDI 0.937
Developing Countries	Upper-Middle Income	Malaysia	GDP per capita USD 12,000, rapid industrialization, HDI 0.803
		Thailand	GDP per capita USD 7000, tourism and manufacturing focus, HDI 0.800
		China	GDP per capita USD 13,000, world’s 2nd largest economy, regional disparities, HDI 0.788
		Brunei	GDP per capita USD 30,000, HDI 0.829
	Lower-Middle Income	Indonesia	GDP per capita USD 4800, Southeast Asia’s largest economy, HDI 0.705
		Philippines	GDP per capita USD 3600, service-oriented, HDI 0.699
		Vietnam	GDP per capita USD 4100, fast-growing manufacturing, HDI 0.703
		Myanmar	GDP per capita USD 1100, political instability hinders development, HDI 0.585
		Cambodia	GDP per capita USD 1800, relies on textiles and tourism, HDI 0.594
		Laos	GDP per capita USD 2600, agriculture-based, HDI 0.607

Note: All GDP figures are approximate and based on recent estimates (2023/2024). HDI classifications follow UNDP standards: ≥0.800 = very high, 0.700–0.799 = high, 0.550–0.699 = medium, <0.550 = low.

).

The results of the heterogeneity analysis revealed the following conclusions (

Table 8. Trade facilitation index (TFI) elasticity by development group.

Group		Fixed Effects FE	Random Effects RE	Mixed Effects OLS
Developed Countries		0.9880 **	0.8309 **	2.2742
Developing Countries	Upper-Middle Income	1.4852 ***	2.5735 ***	3.2699
	Lower-Middle Income	1.2273 ***	2.6033 **	0.7688

Notes: *** and ** denote significance at the 1% and 5% levels, respectively. Numbers in parentheses are t-values.

). First, trade facilitation improvements generate substantially larger agricultural trade gains in developing economies. A 1% ATFI increase boosts trade by 2.57–3.27% in upper-middle income members and 1.23–2.60% in lower-middle income members (all significant at p < 0.05), compared to just 0.83–0.99% in high-income members (p < 0.05). This aligns with Hoekman and Nicita’s 2011 theory of diminishing marginal returns to institutional quality [24]; developing nations like Cambodia, with baseline ATFI scores of 0.42, achieve disproportionate benefits from initial reforms.

Second, e-commerce demonstrates remarkable asymmetric impacts, driving 3.34% trade growth in developed members versus 0.77–0.29% in developing economies (

Table 9. Sub-indicator effects by group.

Variable	Developed Countries	Developing Countries
		Upper-Middle Income	Lower-Middle Income
Infrastructure	0.7712 ***	2.5019 ***	0.3396 ***
E-commerce	3.3406 **	0.7739 **	0.2867 ***
Institutional Environment	2.1439 **	8.2262 **	1.7064 **
Customs Efficiency	−1.6755 **	−1.3088 ***	−1.7024 ***
Logistics Performance	2.2975 **	3.4678 **	0.2989 **

Notes: *** and ** denote significance at the 1% and 5% levels, respectively. Numbers in parentheses are t-values.

). Vietnam’s farm-to-export digital platforms exemplify this leapfrogging effect documented in the World Bank’s 2023 Development Report, where technology adoption circumvents traditional infrastructure limitations [33].

Third, while customs efficiency negatively affects all members, its impact is more pronounced in high-income economies (−1.68%) compared to developing countries (−1.31% to −1.70%). This adverse effect may stem from the specific indicators used to measure customs efficiency, such as the prevalence of trade barriers and the burden of customs procedures. A case in point is Japan’s stringent 0.01 ppm pesticide residue standards, which, as analyzed by Korinek et al., exemplify how developed nations’ rigorous SPS requirements exacerbate procedural delays [34].

These findings carry important policy implications. RCEP implementation should adopt differentiated approaches: prioritizing infrastructure development (2.50% return) and institutional environment improvements (8.23% return) for upper-middle income members through mechanisms like the ASEAN Single Window, while advancing logistics performance (2.30–3.47% returns) and SPS harmonization through mutual recognition agreements like the Australia–NZ Food Standards Pact for high-income economies. Without such targeted measures, the agreement risks exacerbating existing developmental disparities among member states.

5.4. Discussion

Our findings make three substantive contributions to the trade facilitation literature, with important implications for both research and policy-making in agricultural trade.

First, the study quantifies the unique characteristics of agricultural trade facilitation that distinguish it from manufacturing sectors. While Djankov et al. (2010) found uniformly positive effects of customs modernization in industrial trade [35], our results reveal a paradoxical −3.506% impact of customs efficiency on agricultural flows (p < 0.05). This divergence stems from fundamental differences in product perishability and sanitary requirements, where stringent biosecurity protocols like Australia’s 72 h cold treatment for tropical fruits create trade-offs between regulatory rigor and fluidity. The 8.397% overall trade facilitation elasticity significantly exceeds manufacturing sector estimates, suggesting agricultural trade benefits disproportionately from incremental improvements in clearance efficiency.

Second, the analysis advances measurement methodologies for trade facilitation indices (TFIs). The superior performance of our augmented model (adjusted R² = 0.82 vs. 0.76 in OECD standard TFI) demonstrates the necessity of sector-specific weightings that account for agriculture’s 2.3× higher time sensitivity; explicit inclusion of logistics performance metrics, which showed a 2.655% trade impact (p < 0.05); and development-level differentiation, where upper-middle income members gained 2.57–3.27% per TFI point versus 0.83–0.99% for developed economies.

Third, the results provide actionable policy insights for RCEP implementation. The 4.554% return on e-commerce investments (p < 0.01) supports prioritizing digital single windows, particularly for ASEAN members, where current adoption lags at 34%. Conversely, the negative customs efficiency coefficient calls for “smart SPS” solutions like mutual recognition of equivalent standards (cf. Australia–NZ Food Pact), risk-based inspection protocols, and blockchain-enabled certification systems.

Comparative Perspective: The development-level heterogeneity aligns with Hoekman and Nicita’s institutional capacity theory [24] but reveals new nuances: Vietnam’s 180% import growth despite modest ATFI scores (0.57) demonstrates how targeted digital leapfrogging can overcome traditional infrastructure constraints.

6. Conclusions

This study examines the current state of agricultural trade between China and other RCEP member countries, reviews the application of trade facilitation as discussed by domestic and international scholars, and empirically analyzes the impact of trade facilitation on China’s agricultural trade. The overall findings suggest that trade facilitation has a significant effect on agricultural trade between China and other RCEP members. In particular, higher levels of trade facilitation in these countries can significantly enhance agricultural trade with China.

These results carry important theoretical implications. Our estimated trade facilitation coefficient of 8.397 percent significantly exceeds the 5.2–6.8 percent range documented for ASEAN-FTA in Shepherd’s 2016 study [29] and approaches the 9.1 percent single-market effect observed in the EU by Portes and Rey in 2005 [36]. This suggests RCEP may achieve deeper integration than traditional free trade agreements. Notably, the negative 3.506 percent impact of customs efficiency contrasts sharply with positive findings in manufacturing sector studies such as that conducted by Djankov et al. in 2010 [35], highlighting agricultural trade’s unique vulnerability to procedural delays. This phenomenon is exacerbated by divergent SPS regimes among RCEP members, as seen in Japan’s stringent 0.01 ppm pesticide standards compared to ASEAN’s more harmonized thresholds.

Generally, enhancing the trade facilitation level in countries can significantly reduce trade costs, boost trade efficiency, and promote agricultural trade among partners. Specifically, improvements in infrastructure, internet connectivity, business environment, customs clearance, and logistics all strongly support this enhancement.

In terms of trade facilitation, China’s strongest performance lies in the institutional environment and infrastructure. This is largely attributed to China’s significant investment in infrastructure and the substantial efforts made to improve the institutional environment in recent years. However, China still lags behind Singapore in these areas. China should continue to invest in infrastructure, enhance customs efficiency, improve the domestic business environment, and accelerate the development of communication facilities. Against the backdrop of economic globalization and the opportunities presented by the RCEP agreement, China must further elevate its trade facilitation levels.

Several important limitations should be acknowledged in this study. First, the analysis fails to account for major exogenous shocks, including the COVID-19 pandemic’s supply chain disruptions during 2020–2022 and the ongoing Russia–Ukraine conflict’s significant impact on global grain prices. Second, potential omitted variable bias may stem from two key unobserved factors: seasonal export bans as a form of non-tariff barrier and climate-induced agricultural yield fluctuations. For future research, we recommend incorporating two additional data sources to address these gaps: UNCTAD’s comprehensive TRAINS non-tariff barrier database and satellite-based high-frequency disaster monitoring systems.

In the context of global political and economic instability, China has remained steadfast in its strategy of agricultural openness. Among the 15 RCEP member countries, ASEAN nations account for 10, and most of these countries still have substantial potential for improvement in areas such as e-commerce, institutional environment, trade barriers, and logistics performance. Therefore, enhancing cooperation among member countries, particularly by assisting ASEAN nations in rapidly advancing their trade facilitation levels, will significantly contribute to the smooth development of agricultural trade between China and these countries. As a key initiator of the RCEP agreement, China should assume its role as a major country by coordinating the efforts of all member states and actively driving the implementation of trade facilitation measures. This will help to further eliminate trade barriers, reduce trade costs, and advance regional trade liberalization.