VoxEU Column International trade

China’s import demand for agricultural products: The impact of the Phase One trade agreement

In December 2019, the US and China reached a Phase One agreement, which mandates China to purchase additional imports from the US worth $200 billion in 2020 and 2021. This column shows that the most efficient way for China to increase imports from the US is to mimic the effects of an import subsidy. For agricultural products, this subsidy would need to be as high as 42% for 2020 and 59% for 2021 in order to meet the target. Such subsidies would divert agricultural imports away from other countries, especially decreasing Chinas imports from Australia and Canada.

In December 2019, the US and China reached a Phase One agreement to end the trade war that had started in early 2018. In this agreement, China committed to purchasing additional US imports worth $200 million in 2020 and 2021, covering products from the manufacturing, agricultural, energy, and services sectors (Bown 2020). While the Covid-19 pandemic in 2020 may render this agreement moot or at least delay its implementation, it is still worth asking whether it is actually achievable. As for example for agricultural goods, the agreement states that China would purchase imports worth $12.5 billion more in 2020 and $19.5 billion more in 2021 relative to the import level of 2017. While China is among the world’s largest importers of agricultural products, these numbers are significantly higher than what China had previously imported from the US. Overall, China’s agricultural imports increased more than 12 times between 1997 and 2015 (Figure 1). In 2017, China imported $132 billion of agricultural commodities from the rest of the world, of which $24.1 billion was sourced from the US. Hence, the Phase One agreement results in an import target of $36.6 billion in 2020 and $43.6 billion in 2021, as shown by the horizontal lines in Figure 1. Since these targets exceed the highest prior values of Chinese agricultural imports from the US, it is worth asking how and if they can be achieved.

Figure 1 China agricultural imports ($ billion) from US and ROW, 1997-2019

Notes: Data are obtained from China Customs Trade Data, and USDA for 2018 and 2019.


To address this question, in a recent paper we estimate a non-homothetic demand system for agricultural imports for China (Feenstra and Hong (2020). We adopt a framework developed by Fajgelbaum and Khandelwal (2016) that combines an almost ideal demand system (AIDS) defined over products and exporting countries with a gravity equation in trade, so as to identify the demand of each importer for the products of each exporter. To estimate the AIDS-gravity equation we use data from 30 Chinese provinces for 58 agricultural commodities from 78 major trading partners, between 1997-2017. We are able to distinguish the elasticity of demand by commodity and country of origin with respect to price and income. Based on these key estimated parameters, we forecast China’s future import demand for US agricultural products making some assumptions about rising spending on imports. We then ask how China can achieve the 2020 and 2021 import targets.

The effects of an import subsidy on US imports

We show that the most efficient way for China to import more from the US is to mimic the effects of an import subsidy on US imports. That is, state agencies could instruct importers of US agricultural goods to act as if there was a subsidy on those goods. We compute the necessary subsidies for three different import growth scenarios. If China’s agricultural imports grew at the same annual rate as during 2007-2017 -- about 10% per year -- then the effective subsidies would need to be 12% and 23% to achieve the import targets for 2020 and 2021, respectively. If import growth was only one-half of that amount, then the subsidies would need to be 18% and 41%. Whereas if there was no growth from 2017 onwards, as a result of the Covid-19 pandemic for example, then the subsidies would need to be as high as 42% and 59% (table 1). These subsidy rates for US products are admittedly very high, and it would be challenging to achieve such an increase in US imports through state command.

Table 1 China’s forecasted import demand from the US

Effective subsidy increases Chinese imports from the US, but reduces imports from other countries

While such an effective subsidy would substantially boost China’s purchases from the US, it would also shift trade away from the rest of the world. Under the zero-growth scenario, where China’s import demand for each agricultural product is the same as in 2017, the subsidies would lead to increases in US imports of 51.9% and 80.7% in 2020 and 2012 (Table 1). We assume that this increase would be fully offset by a decrease in imports from the rest of the world.

This would especially affect Australia and Canada which are both among the top 10 suppliers of goods to China, after the US (based on their exports to China in 2017). But also Brazil, Indonesia, Malaysia, Thailand, and Vietnam, as well as Argentina, France, Germany, Netherlands, and New Zealand (figure 2) would be affected. Australia would suffer the largest drop in exports to China as its exports would fall by $1 billion (or 10.8%) from a baseline of $9.18 billion in 2020, and by $1.67 billion (or 18.1%) in 2021. Brazil, the top supplier after the US, would sell $0.6 billion less to China under a 42% subsidy in 2020 and nearly $1 billion less under a 59% subsidy in 2021.

Figure 2 Impact of the subsidies on other countries under the zero-growth scenario (US$ billions)

We show that this diversion occurs for three reasons: (i) a conventional substitution effect within products, which depends on the number of competing countries selling each product in each province; (ii) an income effect that arises due to the effective subsidy, which can offset the substitution effect in part or in whole; (iii) a further substitution effect that can occur across products, particularly as some expenditure shares in the AIDS system reach zero so that there is a renormalization of all other demand shares.

Subsidies would lead to a rich pattern of trade diversion across countries

The export declines by country come from the impact of the subsidies on specific products. For necessity goods (which have a negative income effect), the increase in China’s imports from the US is less than the decrease in imports from the rest of the world, such that total imports fall. For example, China reduces its imports of coarse grains from the rest of the world by more than it increases its imports from the US (Figure 3). Since Australia is the major exporter of coarse grains to China, it is hit the hardest by the effective subsidy.

Figure 3 Impact of the subsidies on imports of coarse grains under the zero-growth scenario (US$ millions)

The results differ for soybeans, which are a luxury good with an income elasticity greater than unity. In this case, the income effect is so strong that China’s imports from some countries are complementary to those from the US, and therefore rise rather than fall with the Phase One targets. Since Brazil is a principal supplier of soybeans to China, besides the US, the effective subsidies will only affect Brazil modestly.

In addition, the substitution effect might lead to cross-product substitution away from other products. Increased soybean imports from the US would lower the price of soybean oil for consumers in China, leading to increased demand. That leads to cross-product substitution away from other products—in particular, away from canola oil that is processed from rapeseed. Hence, the effective subsidy on US soybean imports results in a decrease of rapeseed imports.  Since Canada’s top agricultural export to China is rapeseed, the cross-product substitution drives the decline in China’s imports from Canada.

Policy implications

The Phase One agreement commits China to a ‘managed trade’ approach, mandating it to buy an additional $200 billion of US goods and services in 2020 and 2021, relative to its baseline purchases in 2017. We show that the most efficient way for China achieve these ambitious targets is to mimic the effect of an import subsidy on US products. The magnitude of these effective subsidies needed to fulfil the Phase One agreement are admittedly very high and it can be expected that the increased imports from the US would result in trade diversion away from the rest of the world. The size and sign of these diversions differ starkly across source countries and especially countries which would see a decline in their exports to China due to an effective subsidy for US goods would be justified in raising a complaint to the WTO as such a manoeuvre would violate international trade norms of equal treatment of trading partners.

Authors’ note: This research was supported by Cooperative Agreement 58-3000-7-0087 between the U.S. Department of Agriculture’s Economic Research Service and the University of California at Davis. The findings and conclusions in this article are those of the authors and should not be construed to represent any official USDA or U.S. Government determination or policy, nor do they necessarily reflect the views of the National Bureau of Economic Research.


Bown, C (2020), “Unappreciated Hazards of the US-China Phase One Deal”, Trade and Investment Policy Watch, Peterson Institute for International Economics, 21 January. Available at:

Fajgelbaum, P D and A K Khandelwal (2016), “Measuring the Unequal Gains from Trade”, The Quarterly Journal of Economics, 131 (3): 1113-1180.

Feenstra, R C and C Hong (2020), “China's Import Demand for Agricultural Products: The Impact of the Phase One Trade Agreement”, NBER Working Paper No. w27383

945 Reads