1. Introduction
The B&R Initiative, proposed by China in 2013, has drawn widespread political and academic attention. The inter-regional trade in goods and services along the B&R has positive effects on economic integration [
1], the globalization of medical and health industries [
2], inter-regional connectivity, industrial integration [
3] and the sustainable development of resources [
4,
5]. While people gain economic benefits from trade, they are increasingly aware that the associated environmental problems have brought serious threats to human existence. A large number of existing studies have focused on the impact of trade on carbon emissions [
6,
7,
8,
9], while less attention has been paid to the impact on water resources. Meeting the growing water demands of ecosystems and society is one of the major environmental challenges in the 21st century.
On the one hand, as the impact of climate change and environmental pollution intensifies, the availability of water resources decreases dramatically [
10,
11]. On the other hand, factors such as population growth and socio-economic development have led to increasing global water demand [
12,
13]. Under the action of dual pressures, the situation of water resources in the world has become more severe. According to the 2020 United Nations State of Food and Agriculture report, 3.2 billion people worldwide face water scarcity, and about 1.2 billion people live in agricultural areas with extreme water scarcity [
14]. Agriculture remains the largest user of water consumption globally, accounting for 87% of global water consumption [
15]. Grain production, however, is an important component of agricultural water consumption and plays an important role in economic development, water resources utilization, and social stability [
16]. Virtual water is not water in the real sense, but the water resources needed to produce products and services, that is, virtual water condensed in products and services. This concept was first proposed by Tony Allan, which provides a new perspective for the research of water resources security and management [
17,
18,
19,
20]. Compared with physical water resources, its easy transportation characteristics make virtual water trade a useful tool for mitigating water scarcity [
21,
22]. In this paper, virtual water refers to water resources that are embodied in grain in an “invisible” form. The virtual water strategy combines grain production and water use, aiming at a global reallocation of water resources by linking grain availability in water-rich areas with improved water scarcity in water-scarce areas through trade. In this work, the virtual water strategy is defined as the import of grain from water-scarce countries or regions to water-abundant countries or regions through trade in order to alleviate global grain supply and demand constraints and water scarcity.
In accounting for virtual water flows, there are mainly input–output models and linear optimization models. Some studies have focused on the use of input–output (IO) analysis to reveal virtual water in a country or region [
23,
24]. In the case of unbalanced regional production and consumption structures, it becomes crucial to assess the virtual water flows between regions. The multi-regional input–output (MRIO) model can systematically describe the input–output relationships between different sectors in different regions, and it has been widely used to calculate the virtual water flows between regions. For example, Zhang et al. [
25] used the 2012 Chinese MRIO table to account the virtual water volume traded between the Yellow River Delta and other provinces, and to assess the dependence of the Yellow River Delta on external water resources. The results illustrate that virtual water trade exacerbates water scarcity in this region, as virtual water exports are greater than imports. Zhang et al. constructed a virtual water trade network based on the MRIO model and accounted for virtual scarce water in sectoral exports of intermediate and final products to study the virtual water flow risks by sectors in northeast China [
26]. An et al. [
27] quantified the virtual water flow embodied in the inter-provincial grain trade in China, which was validated with the results calculated by MRIO. Several shortcomings have been revealed in the application of input–output models to address issues related to virtual water embodied in grain trade. First, the input–output model considers the economic sector level rather than individual product level [
28], making it difficult to accurately assess international grain trade based on these sectors. Secondly, the time resolution of the input–output table is poor, and there exists a time delay [
29]. Currently, the open-source data of the input–output table in the Eora database is only updated to 2016. The linear optimization model can compensate for the above shortcomings of the input–output model. Detailed supply and demand structures of grain products among countries can be obtained from the latest grain-related data. Linear programming is an important system optimization method in operations research, providing a scientific basis for making optimal decisions with limited human, material, financial and other resources in a rational way. From the perspective of virtual water, international grain trade shows an irrational structure of ”North-to-South Water Diversion” [
30,
31,
32]. The spatial dislocation of grain production and water poses a major challenge to sustainable development. The trade pattern of grain, coupled with the scarcity and endowment difference of water resources, has seriously threatened the development of agriculture. Therefore, reshaping the grain trade relations has positive significance for solving the water crisis [
33].
In addition, when weighing the rationality of the grain trade structure, it is mostly the transportation costs that are considered [
27,
34,
35]. Grains are used not only to meet ration consumption but also to meet the needs of industry and economic development. Opportunity cost refers to the loss of potential benefits resulting from choosing one better alternative and giving up the other when making a decision [
36]. According to the explanation of the opportunity cost (the maximum net income that may be obtained by making a choice but giving up another), it is necessary to explore the role of the opportunity cost of grain in order to stimulate the dynamism of grain in the industrial value chain and to enhance the added value of grain. In general, opportunity cost reflects the value of the best alternative and should be considered as part of any decision-making process. However, the hidden aspects of opportunity cost are difficult to capture and measure, and they are often ignored by people, so that optimal decisions cannot be made [
37,
38,
39,
40]. Therefore, when measuring the benefits of different decisions, decision makers should consider not only the resources sacrificed after making the choice but also the potential benefits lost in comparison with other options, that is, the opportunity cost. This study mainly focuses on the added value of industrial uses of grain (other service uses are not considered), using the added value generated per tonne of industrial grain as the opportunity cost, and a linear optimization model of grain trade in countries along the B&R is constructed. With the objective of minimizing the cost of grain trade, the optimal configuration of grain trade and the economic and water benefits in two situations with or without considering the opportunity cost are compared. This exploration helps policy makers understand the impact of grain trade structure on human society.
The rest of the paper is organized as follows:
Section 2 introduces the linear optimization model and method for quantifying benefits in two scenarios, and it presents the sample data selected for the study.
Section 3 obtains the optimal configuration of grain trade and compares the two optimization results to reveal the impact of opportunity cost on economic and water resource benefits. Eventually,
Section 4 draws conclusions on the optimal configuration and benefit analysis of grain trade in the B&R countries, and it puts forward more policy suggestions.
4. Conclusions and Policy Implications
The B&R Initiative has built a platform for China and related countries to smooth trade and promote the development of world economic and trade cooperation. With the steady increase in the scale of trade in recent years, the environmental concerns associated with the B&R initiative cannot be ignored. Water scarcity has become an important environmental issue facing the world today. Some studies have revealed at the regional level that the pattern of international grain virtual water flows exhibits a north–south pattern, which to some extent exacerbates the imbalance of regional resource allocation [
30,
31,
32]. In addition, this trade pattern is not conducive to balancing grain supply and demand across regions. For grain-exporting countries or regions where water resources are scarce, the lack of sustainability of agricultural production and water resources can exacerbate the pressure on local water resources, further threatening grain production. On the grain import side, if the embodied water resources in grain trade are not effectively utilized, it will result in an indirect waste of resources, which constrains sustainable economic and social development. In response to the current mismatch between the spatial distribution of grain production and water resources, this study designs the grain trade optimization model for countries along the Belt and Road to reconfigure the structure of grain supply and demand and achieve a rational allocation of resources, which is meaningful and challenging. In this study, the total grain trade cost is divided into two parts: transportation cost and opportunity cost. A linear optimization model is constructed to explore the optimal configuration of grain trade in B&R Initiative countries with the objectives of minimizing transportation cost and total trade cost, respectively. The results are as follows.
- (1)
Current situation of grain trade
Grain resources are highly concentrated. Grain-rich countries are mainly located in Eastern Europe and Southeast Asia. Ukraine and Russia are the largest grain-rich countries, with surpluses of about 40% of the total. China, Egypt, Iran and Turkey are the main demand countries in the grain trade, and their grain deficit accounts for about 75% of the total deficit. Comparing the relationship between grain balance, embodied virtual water balance, and water scarcity in various countries, it is found that on the demand side, countries with relatively abundant water resources such as Malaysia and Poland have introduced a large amount of virtual water through grain imports, which undoubtedly increases the pressure on countries upstream in the supply chain, especially water-scarce countries. On the supply side, Indonesia faces water shortages, but it transfers a large amount of virtual water through grain exports, which further exacerbates its own water scarcity. In general, there is a contradiction between the grain virtual water flow pattern and the actual water flow distribution in the countries along the B&R Initiative.
- (2)
Selection preference features
Both optimization results reveal that trading countries follow geographic proximity and long-tail characteristics in choosing partners. This is due to the fact that geographical proximity can reduce the transportation and communication costs of countries and avoid the increase of transaction cost caused by information asymmetry. Therefore, trading countries are more inclined to conduct trade with neighboring countries in order to obtain the advantages brought by geographical proximity. In addition, the grain trade of B&R Initiative countries has scale-free characteristics, indicating that trading countries have the characteristics of selection preference and tend to cooperate with countries with high trade volume to ensure the security of grain supply.
- (3)
Optimization of grain trade with consideration of transportation cost
The grain supply and demand structure obtained with the objective of minimizing transportation costs corresponds to a trade cost of about 466 billion US dollars, which is close to 1/4 of the total agricultural output. The corresponding loss of water resources is about cubic meters, reflecting that most of the virtual water embodied in grain trade flows from countries with low water productivity to those with high water productivity, resulting in an inefficient use of water resources.
- (4)
Optimization of grain trade with consideration of transportation cost and opportunity cost
Based on the optimization model with the objective of minimizing the transportation cost, the economic value of industrial grain is introduced as the opportunity cost, and then, the linear optimization model with the objective of minimizing the total trade cost (transportation cost and opportunity cost) is developed to reconstruct the grain supply and demand relationship. The results show that the introduction of opportunity cost constrains the grain trade flow from the outflow area with low industrial added value to the inflow area with high industrial added value, which has a positive economic impact. The economic value generated not only covers the transportation cost but also has an additional economic benefit of about 130 trillion US dollars. Furthermore, the analysis of grain virtual water trade shows that the optimization results also bring positive water resources benefits, saving about cubic meters of water resources.
Based on these findings, the following important insights can be drawn.
(1) Given the priority of some countries in grain trade, they have great potential to reshape the structure of grain supply and demand. In particular, China should take full advantage of the development opportunities brought by the BRI to strengthen extensive cooperation with neighboring grain-rich countries, such as Russia and India. Then, it radiates to the main exporting countries along the route, such as the Czech Republic, Romania, Kazakhstan and other countries. Finally, a mutually beneficial and win–win development pattern will be formed.
(2) The optimal grain trade configuration with consideration of opportunity cost yields significant economic and water resources benefits. Opportunity costs can be a means of improving trade structures and managing water resources. Noting this, the B&R economies should consider not only the natural endowments of the region but also the opportunity cost of grain products when reshaping the grain trade structure.
(3) Countries on the grain supply side can improve water productivity and vigorously develop grain production by improving crop varieties, enhancing soil fertility, and introducing water-saving technologies. Countries on the grain demand side should focus on developing their economies and improving the industrial and economic value of grain products.
(4) The optimized grain supply and demand structure with the introduction of opportunity cost has brought significant economic benefits to the B&R economies. However, the shortcoming is that some countries suffer economic losses due to the increase in trade costs compared to optimization results with consideration of transportation cost. For example, Lithuania exports grain to countries with lower industrial added value, while Belarus and Zimbabwe mainly import grain from countries with higher industrial added value, which leads to an increase in their total trade costs. Therefore, a regional cooperation organization should be established to explore the compensation mechanism for these countries and to conduct coupled management of physical–virtual water to ensure the sustainable and smooth flow of grains in B&R countries.
The findings have implications for grain security and water management, but there are still some shortcomings. Firstly, it is one-sided to evaluate the security of grain supply and demand only from the perspective of water resources, because the optimization of the grain trade structure also involves virtual land, virtual labor and other determinants. Benefiting from the results of the research at hand, further research will consider how to account for the virtual land embodied in grain trade and how to assess the ecological benefits resulting from the conversion of land resources saved due to the optimized structure of grain trade into ecological land. In addition, with the increased granularity of MRIO tables, it is necessary and meaningful to obtain the optimal configuration of grain trade in B&R countries based on the input–output approach and to compare and analyze the results with those in the study.