US global value chain integration: a major impediment for Trump’s protectionist trade agenda

Co-authored by Moniek Tulen during her internship at RaboResearch

Many campaign promises by President Trump revolve around the implementation of a protectionist trade agenda. In order to bring back jobs to the US, Trump promised his voters to tear down NAFTA and to impose tariffs on Chinese and Mexican import, and German cars. So far, a large-scale package of protectionist measures has not been implemented (see Erken, Marey and Wijffelaars, 2017). NAFTA has not been abolished, but will be renegotiated. And the only trade barrier that Trump has imposed is a tariff on Canadian softwood lumber. One explanation for the lack of vigorous protectionist trade policy could be reluctance to shake up cross-border trade. In case of heavily-integrated international supply chains, slapping on a straightforward blanket levy would not only be detrimental for Chinese, German or Mexican exporters, but would also harm US manufacturers of intermediate goods and services.

In this Special we consider global value chain integration of the US with four countries: China, Mexico, Germany and the Netherlands. First, we will provide some background information on global value chain integration in the literature. Second, we will examine the five industries that are most important in terms of US trade. Third, we will assess actual US global value chain integration by looking at the share of value added of American intermediates that are used in final goods destined for the US market, and, vice versa, the dependence of US firms on foreign intermediates.

In his inauguration speech on the 20^th of January, Donald Trump was fairly straightforward about his stance towards free trade: “We must protect our borders from the ravages of other countries making our product, stealing our companies and destroying our jobs. Protection will lead to great prosperity and strength.” Trump’s protectionist promises have spurred a stream of studies on US trade, protectionism and the potential impact on the US economy and its main trading partners. In this section, we position the recent US trade topic within the broader discussion revolving the complexities of trade, which has been ongoing for decades. However, let’s first have a look at some recent new insights directly related to the impact of possible US protectionism.

Based on the Brookings export monitor, Parilla and Muro (2017) show that many local economies in the US are dependent on exports. In various clusters reflecting local industrial strength, such as aerospace in Seattle, automotive in Michigan and electronics in Portland, exports account for at least 15% of GDP and has created thousands of jobs. Any disruption of trade patterns, either be it a renegotiation of NAFTA or retaliation by foreign countries in response to US trade barriers, will hit these local communities hard. The Economist (2017) also illustrates the importance of trade for local US communities. They calculated that, without NAFTA, especially Texan farmers and producers of automotive parts in Michigan would face large additional costs.

Although these studies underline the importance of trade for local communities, it does not give a clear picture of the complexity of trade relations, which is far more integrated than gross trade data reveals. Many US firms are not only involved in final goods exports to other countries, but are also highly dependent on imports of intermediate goods produced abroad. Intermediates have a share of 43% in total US import, with intermediates from NAFTA countries even having a share of 50% (see Parilla, 2017). But is also works the other way around: 60% of US gross exports consists of intermediate goods and services, and only 40% encompasses final products. Both the OECD (2013a) and Baldwin and Lopez-Gonzalez (2015) show that global value chains integration (GVCI) has become more and more important over the last decades. This process has enabled firms to outsource labour-intensive tasks in the value chain to low-costs countries, whereas knowledge-intensive activities are located at the home base (OECD, 2013b).

So, what benefits do firms exactly have from slicing up the value chain globally? In general, there is much empirical evidence supporting the positive effect of openness to foreign trade and foreign direct investment (FDI) on economic performance. First of all, openness and FDI are conduits to internalise knowledge developed overseas (see, e.g., Romer, 1989, Cohen and Levinthal, 1990). The empirical evidence is quite strong: foreign R&D capital generates larger positive effects on domestic productivity, the more a country is open to foreign trade (Coe and Helpman, 1995; Erken, Donselaar and Thurik, 2016) or involved in foreign direct investment activity (Branstetter, 2006). From a firm-level perspective, Keller and Yeapl (2009) provide empirical evidence that FDI spillovers account for about 14% of productivity growth in US firms between 1987 and 1996. Second, there appears to be direct positive productivity effects generated by foreign trade (Bassanini, Scarpetta and Hemmings, 2001; Edwards, 1998). A larger degree of openness increases market opportunities abroad, provides specialisation opportunities and increases foreign competition, which forces domestic firms to reduce X-inefficiencies[1] and invest in Research & Development (R&D) and innovation. For example, Haskel, Pereira and Slaughter (2007) show for the UK that a 10% point increase in foreign presence in an industry raises total factor productivity of that industry’s domestic plant by 0,5%.

In general, value chain integration contributes positively to firm performance (see Chang et al. (2016) and Ataseven and Nair (2017) for two recent meta-analyses). But how these integration processes play out on a global scale is more difficult to assess. In a hallmark review of trade theory, Helpman (2006) concludes that traditional trade models fail to describe the complexity of trade nowadays, such as the international fragmentation of production. Theoretical refinements have been made to fill this void, but empirical testing is difficult due to limited availability of data sets. Helpman states: “Hypotheses that require detailed firm-level data about trade in different types of products, such as intermediate inputs versus final goods […] cannot be examined.”

However, on a higher level of aggregation, empirical work is growing, which in general acknowledges the increasing importance of input-output linkages for economic activity (e.g. Frohm and Gunella, 2017; Acemoglu, Akcigit and Kerr, 2016). The complex relations between the different fragments within global value chains imply that any disruption in these supply chains generates amplified effects throughout industries and economies on a higher scale. Frohm and Gunella estimate that 1% change in economic activity in an industry’s global value chain translates into 0.3% points impact on activity of the sector. Moreover, these interlinkages might have prevented policymakers around the globe to resort to protectionist measures. Gawande, Hoekman and Cui (2015) conclude that the increasing fragmentation of production across global value chains – i.e. vertical specialisation – may have prevented countries to raise trade barriers and induce protectionist measures in the follow-up of the global financial crisis of 2008. This might explain why we haven’t seen any protectionist trade measures by Trump yet. Biondi (2017) recently examined value chain integration between the Mexican automotive industry and the US economy, as this industry has been scapegoated by Trump on several occasions. His conclusion is that 17% of the export value of the Mexican automotive sector is sourced from the US.

In our empirical analysis, we will not only view global value chain integration of the Mexican industries and the US, but additionally will analyse US value chain integration with China, Germany and the Netherlands as well.

[1] X-inefficiencies has been defined by Leibenstein (1966) as the difference between a firm’s potential and actual utilisation of resources. Leibenstein concludes that a majority of X-inefficiencies are the result of poor motivation of business management, which is most likely linked to the market structure and the extent to which businesses face competition in their markets.

In order to examine US global trade integration, we use data from the 2015 release of the OECD-WTO Trade in Value Added (TiVA) database. This database offers country-level trade flows in 2011 for a detailed breakdown of 34 industries (two-digit ISIC). Next, we will use two datasets within TiVA:

As data on foreign value added in gross exports are not directly available on bilateral levels, we adopt a so-called proportionality assumption, which means that the value structure of exports to a bilateral partner is similar to the structure of total exports. The OECD also uses this assumption in order to calculate domestic value added on a bilateral level.

The United States runs a trade deficit with the world of roughly USD 800bn. The largest share of this deficit is due to trade between the US and China (see Figure 1). In 2015, China exported roughly USD 500bn worth of goods and services to the US, whereas it only imported USD 150bn from the US.

The US also runs trade deficits with Mexico (USD 60bn), Japan (USD 70bn)and Germany (USD 80bn), albeit much smaller than its deficit with China. In contrast, the US has a trade surplus with the Netherlands of USD 20bn.

China industry breakdown

It is possible to break down the US trade deficit with its main trading partners. Roughly 75% of the deficit with China can be attributed to trade in only three industries: electronic products (i.e. computers and mobile phones), wholesale and retail trade, and textiles and footwear. This is not surprising, as more than 90% of all imports of tablets and laptop (Capital Economics, 2016) and more than more than 60% of all mobile phones imported in the US are assembled in China (Figure 3). In addition, textiles and footwear are important US imports from China as well. China covers again 60% of all imported footwear in the US (Figure 3).

Although some studies report higher shares of Chinese manufacturing in total American imports of textiles/shoes and electronic products, keep in mind that China is increasingly losing out on its comparative low-cost advantage due to rising prosperity and wages. A consequence of these rising costs is that many foreign firms relocate Chinese production lines to other low-cost Asian countries or have begun planning the relocation of these activities (see BMI, 2016). Beginning of this year, Seagate, a hard-disk producer, closed its plant in Suzhou, China. And Taiwan’s Foxconn, known for producing Apple’s iPhones, is planning to set up 12 new locations in India rather than expand production lines in China.

Mexico industry breakdown

An industry breakdown of US-Mexican trade shows that roughly four sectors are responsible for the US trade deficit: motor vehicles, mining, electronic products and electrical machinery (Figure 4). What is even more interesting is that the US trade deficit of these four sectors together amounts to USD 90bn, thereby exceeding the total US trade deficit with Mexico (of USD 60bn). This is due to the fact that Mexico runs large trade deficits with the US in other sectors, such as fuels.

From a trade perspective, the motor vehicle industry is especially important for Mexico. The OECD STAN database shows that Mexico’s auto industry has a share of 3% within total Mexican value added, a share of 4.1% in Mexico’s gross operating surplus and provides direct jobs to 433 thousand people. Moreover, Biondi (2017) concludes that 70% of vehicle exports go to the US and Mexico is now the world’s 4th largest automobile exporter, behind Germany, Japan and South Korea.

Germany industry breakdown

Germany shows a trade surplus in each of its top-10 most important trading sectors with the US, with the most prominent sectors being motor vehicles (USD 14.6bn), machinery and equipment (USD 8.6bn), and chemicals and chemical products (USD 7.2bn). During the G7 summit in May 2017, Trump’s criticized German-American trade ties and especially mentioned the German automotive sector: “Germans are bad, very bad. See the millions of cars they are selling to the U.S. Terrible. We will stop this.” In an interview with German tabloid Bild Zeitung in January, Trump had already stated that he would impose a 35% tariff on every German car exported to the US.

The Netherlands industry breakdown

In contrast to China, Mexico and Germany, the US has a trade surplus of approximately USD 20bn with the Netherlands. Most of this surplus can be attributed to high exports of American services industries to the Netherlands, such as R&D and other business activities, and renting of machinery and equipment. Reversely, exports of the Dutch financial intermediation sector to the US are relatively strong. The trade deficit of the Dutch economy with the US is quite extraordinary from an international context, as well as the fact that the Netherlands generally runs trade surpluses with the rest of the world. It is important to state, however, that the Netherlands often functions as a hub for re-exports to the European hinterland, e.g. Germany or France. Statistics Netherlands (2017) shows that the Dutch deficit with the US turns into a slight surplus if the data is adjusted for re-export.

In this section, we present our empirical findings of US global value chain integration with China, Mexico, Germany and the Netherlands. The extent of global value chain integration is assessed from two different angels. First, we examine the share of value added of American intermediates used in final goods destined for the US market. Second, we look at the dependence of US firms on foreign intermediate goods and services.

American value added in foreign final goods exports to the US

On the macro level, we find that 4.1% (roughly USD 42bn) of the total value of final goods and services shipped to the US consists of American produced intermediates. On a lower level of aggregation, however, the degree of integration is very unevenly distributed across industries and trading partners. Table 1 shows the share of American value added in Mexican, Chinese, German and Dutch final goods exports to the US.

Integration of the value chain is especially prominent in three Mexican industries: motor vehicles (18.1%), electronic equipment (17.2%) and electrical machinery (16.7%). Our calculation for the Mexican motor vehicle industry is surprisingly close to calculations by Biondi (2017), who concludes that 17% of the export value of the Mexican automotive sector is sourced from the US.

A breakdown of the Mexican motor vehicle industry shows that American value added (USD 5bn) in Mexican final auto exports to the US (USD 26bn) is distributed fairly evenly across the five largest American supplying sectors, with the wholesale and retail industry (USD 800mln) responsible for the largest piece of the pie (Figure 7). Outside Mexico, the most integrated sector of the three remaining countries that we have examined is the Chinese electronic equipment industry (Figure 8). Apart from this sector, no other sector in either China, Germany or the Netherlands has an American value added share exceeding 4%. One could argue that these industries are most susceptible to protectionist policies by the US. However, it is important to keep in mind that, reversely, US firms are also dependent on foreign intermediates in these industries.

Foreign intermediates in the American exports

Global value chain integration is not one-dimensional. It is also important to look at the dependence of US firms on foreign intermediate goods and services. Table 2 provides this overview.

The three American industries that are most heavily dependent on foreign intermediates are motor vehicles; coke, petroleum and fuels, and basic metals. For these industries roughly one-third of the value that is eventually exported consists of foreign value added. In the motor vehicles sector, this makes up for a total of USD 33bn, while it is even higher for exports of coke and fuel products (USD 38bn). China is the second-most important supplier of intermediates to the US motor vehicles industry (USD 4.4bn). More in general, Canada, Mexico, China, Germany and Japan are the most important suppliers of intermediate goods to American firms. So, even if the US does not make up for a major share in the foreign intermediates used in Chinese, German and Dutch final products, this conclusion does not hold the other way around. Of course, one could argue that if Trump decides to target China or Germany, US firms can substitute their German and Chinese inputs and look for other suppliers. We do not assess these dynamic and possibly mitigating effects. However, these adaptations are time consuming and involve transaction costs, as other suppliers often do not instantly meet proper requirements and know-how. Moreover, the substitutes are likely of less quality and/or more expensive, as otherwise they would have already been the preferred option.

Although the Netherlands does not appear in Table 2 due to relative small trade volumes, De Boeck (2017) shows that 70% of all Dutch exports to the US consists of intermediate goods and services. This is an indication that the global value chain between the US and the Netherlands is largely integrated as well.

In this Special we have looked at the US trade relations with three countries that have been in Trump’s crosshairs for some time: China, Mexico, Germany, and a fourth: the Netherlands. Merely examining gross export and import data misses out on the complexity that lies underneath these flows. In fact, in many sectors the value chain is sliced up globally, which means that US firms are highly dependent on imports of intermediate goods produced abroad and, vice versa, foreign firms exporting to the US use American manufactured intermediates in their final products as well. This so-called global value chains integration has become more and more important over the last decades. Consequently, disrupting these supply chains would not only hurt foreign exporters, but domestic firms and consumers as well. The larger the integration of US firms within global value chains, the more difficult it becomes to implement protectionist trade measures.

Our empirical analysis shows global value chain integration of US firms is most prominent in three Mexican industries: motor vehicles (18.1%), electronic equipment (17.2%) and electrical machinery (16.7%). For example, for the Mexican motor vehicles industry, this means that 18.1% of final exports to the US consists of American value added.Outside Mexico, the most integrated sector of the three remaining countries that we have examined is the Chinese electronic equipment industry (5.5%). Apart from this sector, no other sector in either China, Germany or the Netherlands has an American value added share exceeding 4%. One could argue that these industries are most susceptible to protectionist policies by the US. But even if the US is not heavily involved in providing intermediates for Chinese, German and Dutch final products, this conclusion does not hold the other way around. More in general, Canada, Mexico, China, Germany and Japan are the most important suppliers of intermediate goods to the US economy. Even if US firms decide to substitute intermediates from either of these countries, adaptations would take time and would result in higher transaction costs.

Gawande, Hoekman and Cui (2015) show that global integrated value chains may have prevented countries to raise trade barriers just after global economic meltdown in 2008. In a similar fashion, US involvement in the global value chain might explain why we have hardly seen any protectionist trade measures by the Trump team yet. And to make matters more complicated, global supply chain integration is just one of many reasons why Trump’s hands are tied as far as trade protectionism is concerned (see Erken, Marey and Wijffelaars, 2017). Consequently, we do not believe Trump will resort to full-fletched protectionism, which might not only a blessing for the world economy, but equally so for the US economy.

Acemoglu, D., U. Akcigit & W. Kerr (2016). Networks and the macroeconomy: An empirical exploration. NBER Macroeconomics Annual, 30(1), 273-335.

Ataseven, C., & A. Nair (2017). Assessment of Supply Chain Integration and Performance Relationships: A Meta-Analytic Investigation of the Literature. International Journal of Production Economics.

Baldwin, R., & J. Lopez‐Gonzalez (2015). Supply‐chain trade: A portrait of global patterns and several testable hypotheses. The World Economy, 38(11), 1682-1721.

Bassanini, A., Scarpetta, S., & P. Hemmings (2001). Economic growth: the role of policies and institutions. Panel data evidence from OECD countries. OECD, Paris.

BMI (2016), China - Offshoring trend to ramp up despite Beijing's best efforts, London.

Biondi, F. (2017), The Mexican automotive industry and Trump’s USA, Bruegel.

Branstetter, L. (2006). Is foreign direct investment a channel of knowledge spillovers? Evidence from Japan's FDI in the United States. Journal of International economics, 68(2), 325-344.

Capital Economics (2016), How much would a Trump tariff hurt China?, London.

CBS (2017), Trade surplus excluding re-exports €20 billion lower, Den Haag.

Chang, W., A.E. Ellinger, K.K. Kim & G.R. Franke (2016). Supply chain integration and firm financial performance: A meta-analysis of positional advantage mediation and moderating factors. European Management Journal, 34(3), 282-295.

Coe, D.T. & E. Helpman (1995), International R&D spillovers. European Economic Review, 39(5), 859-887.

Cohen, W.M., & D.A. Levinthal (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative science quarterly, 128-152.

Edwards, S. (1998). Openness, productivity and growth: what do we really know? The Economic Journal, 108(447), 383-398.

Erken, H.P.G., P. Donselaar, P., & A.R. Thurik (2016). Total factor productivity and the role of entrepreneurship. Journal of Technology Transfer, 1-29.

Erken, H.P.G., P. Marey and M. Wijffelaars (2017), Why Trump’s protectionist trade agenda fails, Rabobank, Utrecht.

Frohm, E., & V. Gunnella (2017). Sectoral interlinkages in global value chains: spillovers and network effects. ECB working paper series no. 2064.

Gawande, K., B. Hoekman & Y. Cui (2014). Global supply chains and trade policy responses to the 2008 crisis. World Bank Economic Review, 29(1), 102-128.

Haskel, J.E., S.C. Pereira & M.J. Slaughter (2007). Does inward foreign direct investment boost the productivity of domestic firms? Review of economics and statistics, 89(3), 482-496.

Helpman, E. (2006). Trade, FDI, and the Organization of Firms. Journal of economic literature, 44(3), 589-630.

Keller, W., & S.R. Yeaple (2009). Multinational enterprises, international trade, and productivity growth: firm-level evidence from the United States. Review of Economics and Statistics, 91(4), 821-831.

Kraemer, K., G. Linden & L. Dedrick, (2011). Capturing value in global networks: Apple’s iPad and iPhone. UC Irvine. UC Berkeley and Syracuse working paper.

Leibenstein, H. (1966). Allocative efficiency vs. “X-efficiency”. American Economic Review, 56(3), 392-415.

OECD (2013a), Global value chains (GVCs): United States, Paris.

OECD (2013b), Interconnected economies: benefiting from global value chains, Paris.

Parilla, J. and M. Muro (2017), US metros most exposed to a Trump trade shock, Brookings.

Parilla, J., How US states rely on the NAFTA supply chain, Brookings.

Romer, P.M. (1989). Increasing returns and new developments in the theory of growth, National Bureau of Economic Research, NBER working paper no. w3098, Cambridge (Mass).

The Economist (2017), Farmers and Texans would lose most from barriers to trade with Mexico.