RaboResearch - Economic Research

Why Trump’s protectionist trade agenda fails


  • So far, Trump has failed to implement much of his protectionist trade agenda
  • Although we do expect targeted protectionist measures going forward, we do not expect a full-fletched package because of five reasons:
    • The US has to respect WTO rules, or face retaliation
    • The value chain is sliced up globally, which limits possibilities for protectionist policies without hurting domestic firms
    • Retaliation by trading partners would seriously harm US economic growth
    • US companies benefit much more from offshore production than gross trade data show
    • Automation is more to blame for US job losses than trade. In order to revitalise American manufacturing, Trump should focus on fostering domestic innovation and human capital
  • If Trump decides to impose protectionist measures anyway, this would have negative consequences for the US and world economy. We have used an macro econometric model NiGEM to calculate the impact of an additional US tariff of 20% on foreign import. This would result in a GDP loss of -0.7 percentage points in 2017 and -1.4 percentage points in 2018.
  • If countries decide to retaliate, the damage would be even more serious. US GDP would face a loss of -1.3 percentage points in 2017 and -3.1 percentage points in 2018. Moreover, global trade would decline by 2% in 2017 and 5% in 2018.

Trump’s protectionist agenda: a short recap

Figure 1: Trade deficit of the US with China is USD 350bn in 2016
Figure 1: Trade deficit of the US with China is USD 350bn in 2016Source: Rabobank based on UNCTAD data

Many campaign promises by President Trump revolve around the implementation of a protectionist trade agenda. In his view, the large trade deficit that the US runs with Mexico and China is an indication of dishonest trade practices by these countries against the US (Figure 1). In order to bring back jobs to the US, Trump promised his voters to tear down NAFTA. He called this trade agreement between the US, Canada and Mexico the single-worst trade deal. Moreover, China was framed as being a currency manipulator. According to Trump, China artificially pushes the Renminbi lower in order to sell abundant cheap goods and services on foreign markets. On the G7 summit in May, Trump said Germans are “bad, very bad”, because of the large amount of German car sales in the US, which the US president vowed he would stop. In order to turn the tide on trade, Trump threatened to impose a 45% tariff on all Chinese imports of goods and services, a 35% tariff on Mexican imports and a 35% tariff on the import of German cars. 

Box: Trade and angry white males

Trump’s main message to frame liberal trade as the culprit for US job losses resonated well especially with angry low- and medium educated Caucasian Americans in rural areas (Figure 2). This group saw earnings and employment deteriorate over the last couple of years.

Figure 2: Caucasian Americans voted Trump
Figure 2: Caucasian Americans voted TrumpSource: Edison Research for the National Selection Pool
Figure 3: Medium skilled have a real wage gap of 7%
Figure 3: Medium skilled have a real wage gap of 7%Source: BLS, Macrobond and Rabobank

Especially the low and medium educated show a cumulative real wage gap compared to the beginning of the century, which peaked at more than 8 percent in 2013 (Figure 3). The low educated seem to be catching-up for lost income. Yet, the medium skilled still faced a real wage gap of almost 7 percent in 2016 compared to levels seen in 2000. Also, the quality of jobs for medium-skilled workers has been deteriorating. The mid-segment has been pushed towards the low-end distribution of the labour market. Figure 4 illustrates this development: think of an assistant-accountant or mechanic in the car industry who has lost his/her job and can only find a new job flipping hamburgers in a fast-food restaurant and cleaning offices in the evening hours.

Figuur 4: Erosion of job quality
Figuur 4: Erosion of job qualitySource: BLS and Rabobank

Recently, many authors have been writing about problems of the low and medium-educated working class (see, e.g., Vance, 2016), but the increasingly tough position of the middle class has already been addressed more than a decade ago by Autor, Katz and Kearney (2006). These economists showed an increasingly polarising US labour market, with employment rising in high-wage and low-wage jobs at the expense of the middle-skilled jobs. This phenomenon is called job polarisation and has been subject to different empirical studies for both the US (Autor and Dorn, 2013), and Europe (Goos, Manning and Salomons, 2009). Job polarisation has been more prominent in the US than in most European countries. Especially the institutional context could explain why. While the US is characterised by a very flexible labour market and low union bargaining power, European countries have higher levels of social security.

Limited progress

New guidelines on trade have been released on 9 May by US Secretary Ross. The US will pursue a more aggressive stance on trade remedies and will be reconsidering trade agreements with the intention to boost American exports. In general, Trump has hardly implemented any large protectionist trade measures, even though he does not need approval from Congress to change trade policies (Marey et al., 2017). So far, Trump has only signed one executive order terminating negotiations on the Trans-Pacific Partnership (TPP) and has been focusing on targeted bilateral measures. He imposed a duty on Canadian softwood lumber and maintained current tariffs and anti-dumping measures on Mexican sugar. It is highly unlikely that these incremental trade measures will trigger a trade war. Moreover, on many fronts Trump seems to be backtracking on his initial promises. On 12 April, for instance, Trump stated that China is not a currency manipulator after all and on 27 April, the White House announced that it would renegotiate NAFTA, rather than terminate the agreement. 

Five reasons why Trump won’t resort to full-blown protectionism

There are several reasons why the Trump team might be following an incremental bilateral trade strategy rather than unleashing a full-fletched package of protectionist measures. Below, we will discuss five reasons.

Reason 1: WTO rules and political reasons

First of all, there are the obvious political reasons to refrain from full-blown protectionism. The White House needs Chinese cooperation to put maximum pressure on Pyongyang in order to eliminate the North-Korean nuclear threat. Another important political issue is that the US is a member of the World Trade Organisation (WTO). The WTO uses a non-discriminatory clause, known as the most favoured nation (MFN) principle. This principle means that countries cannot randomly discriminate trade policy between trading partners that are also members of the WTO. For example, the US cannot slap on a 45 percent levy on Chinese imports and at the same time continue to charge EU countries at an average tariff of just 3 percent. Hence, the most favoured trade policy regime of the US, e.g. lower customs duties on certain products, should apply to all WTO members. In his 2017 Trade Agenda sent to Congress on 1 March, Trump indicated that he is willing to violate WTO policy, if he believes WTO policies support unfair trade practices. If Trump decides to walk away from WTO agreements, this could certainly trigger retaliation penalties against the US by other countries. The negative effect would be not only detrimental to the world economy, but to the US even more so. We will elaborate on these effects more extensively later in this Special.

Reason 2: Excessive US consumption

The most important reason behind the US trade deficit has not been dishonest trade practices by foreign countries, such as China and Mexico, but has been credit-driven consumption growth by Americans households. Higher consumption has increased demand for Chinese goods. Subsequently, China used the incoming trade dollars to buy American assets, especially US Treasuries. These capital inflows kept US interest rates low and induced American households to prop up spending even more. Simultaneously, saving rates have decreased to ultimately low levels of 5 to 6 percent, down from 12 to 14 percent in the 1970s. Currently, US foreign borrowings range between USD 50 to 100bn each quarter, needed to finance the American trade deficit (Figure 5).

China and the US will have to deal with these macroeconomic imbalances, as they are unsustainable in the long run. In the short run, however, the dependency on foreign capital inflows makes the US economy vulnerable to say the least. The net international investment position of the US dropped from -10 percent of GDP at the beginning of the century to -40 percent in 2015 (Figure 6). If China decides to repatriate capital from the US in response to (hypothetical) trade barriers imposed by Trump, this would result in a jump in US interest rates. This would spur a chain reaction of falling US asset prices, a fast depreciating USD and a surge in household savings. Such a shock would have serious implications for US private consumption, investment and, ultimately, economic growth.

Figure 5: Low domestic savings and increased foreign borrowings have weakened US international investment position
Figure 5: Low domestic savings and increased foreign borrowings have weakened US international investment positionSource: Macrobond, Rabobank. Explanation: we use an HP filter on the series net lending/borrowing
Figure 6: US international investment position has deteriorated to -40% in 2015
Figure 6: US international investment position has deteriorated to -40% in 2015Source: EIU, Rabobank

Reason 3: Global supply-chain integration

In many sectors, the value chain is sliced up globally. This 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. Gross export and import data do not show these interlinkages. Ultimately, disrupting US globally-integrated supply chains – by for instance tariffs on imports – would not only hurt foreign exporters, but US firms and US consumers as well. The magnitude of these distorting effects depend on the level of global supply-chain integration, which is very heterogeneous across industries and trading partners.

Erken and Tulen (2017) show that US value added in Mexican motor vehicle exports to the US (Figure 7), electronic equipment, and electric machinery is 18.1%, 17.2 % and 16.7%, respectively. In contrast, US value added in Chinese and German gross exports is much lower. US intermediate manufacturers are responsible for only 5.5 percent of total value added of shipped electronic equipment as final goods from China to the US.

Figure 7: Breakdown of Mexican final motor vehicle exports to the US
Figure 7: Breakdown of Mexican final motor vehicle exports to the USSource: Erken and Tulen, 2017. Explanation: VA = value added.

Reversely, US firms use foreign intermediates in their production process as well (Table 1). Canada, Mexico, China, Germany and Japan are the most important suppliers of intermediate goods to US firms. Disruption of these supply chains could be very costly for US manufacturers. Of course, US firms can decide to substitute intermediates from either of these countries. That said, adaptations tend to be time consuming and will involve transaction costs, as other suppliers often do not instantly meet suitable 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.

Table 1: US firms are very dependent on foreign intermediates
Table 1: US firms are very dependent on foreign intermediatesSource: Erken and Tulen, 2017

Reason 4: Trade flows do not reflect revenue flows

US companies benefit much more from offshore production than gross trade data is showing. These profits in turn return to the US and are used to pay for high-wage jobs, such as marketing and product development. The share of the total revenue captured by the American mother company is significantly larger than the share earned by the assembling country abroad.

In the case of for instance Apple, the largest share of iPhones are manufactured in China. This means that the export value of every iPhone shipped from China to the US contributes directly to the US-China trade deficit. A study from the Asian Development Bank (2010) shows that in 2009, iPhone shipments from China to the US added USD 1.9bn to the US trade deficit. But this does not mean that China captures all the revenues of the total export value of these phones. Research by Kraemer, Linden and Dedrick (2011) on revenue streams of the iPhone 4 and iPad illustrates that only a small proportion of revenues from retail sales end up in China for the assembly of these electronic products (Figure 8 and 9).

Figure 8: Chinese labour captures 1.8% revenue share for manufacturing the iPhone 4
Figure 8: Chinese labour captures 1.8% revenue share for manufacturing the iPhone 4Source: Kraemer et al., 2011. Explanation : the iPhone show no post Retail, as Apple is directly paid by a cellular company, which handles distribution and sales
Figure 9: Chinese labour captures a 1.6% revenue share for manufacturing the iPad
Figure 9: Chinese labour captures a 1.6% revenue share for manufacturing the iPadSource: Kraemer et al., 2011

In fact, for each sold iPhone 4, Chinese labour only captures a meagre 1.8 percent ($10) of the revenue ($550), whereas the bulk of the profit share goes to Apple (58 percent), i.e. the US. Ultimately, these profits are also used to nourish high-wage activities in California, such as marketing and product development. Of course, one could argue that Apple could move the assembly of iPhones back to the US, but the US does not have a so-called comparative advantage in low-skilled jobs and low-paid manufacturing anymore. Consequently, relocation of these activities would weigh heavily on Apple’s profit margins, as the average hourly wage costs in the US ($36) outnumbers the Chinese costs ($4) by almost a factor 10 (Figure 10).

Figure 10: Manufacturing hourly labour compensation in US is ten times higher than in China
Figure 10: Manufacturing hourly labour compensation in US is ten times higher than in ChinaSource: Conference Board

Even if large US corporates would decide to relocate overseas manufacturing activities to the US, it is questionable if this would translate into one-on-one jobs creation in the US. An important motivation for firms to reshore production to the home base is automation, which has been eroding the labour cost advantage of low-wage emerging economies, such as China. However, automation also implies that the capital-labour share of these production activities will shift, which weighs on the employment gains of reshoring. Indeed, the OECD (2016) states that empirical work looking into reshoring to OECD economies is particularly characterised by additional capital investment, rather than additional jobs creation. And this brings us to the next reason: the role of automation. 

Reason 5: Automation is more to blame for US job losses than trade

Trump has been targeting trade as the culprit behind American job losses. Partly, this claim is backed by the empirical literature. For instance, Acemoglu et al. (2015) show that Chinese import competition resulted in American job losses between 2 and 2.5 million over more than a decade (1999-2011). Furthermore, Ebenstein et al. (2014) argue that trade pushed American workers from high-wage manufacturing jobs to the lower-paid segments of the labour market, resulting in real wage losses of 12 to 17 percent. Oldenski (2014), however, argues that the wage effect of offshoring is distributed unevenly, with gains for the most highly-skilled workers and relative losses for middle-skilled workers. This is in line with the general observation of a polarised American labour market (see Box 1).

But Trump has been ignoring one big elephant in the room: automation also has been an important reason why American jobs have been shredded (Autor et al., 2015). Declining costs of ICT and fast-increasing processing power have increased options to automate cognitive-routine tasks. Think about office and administrative supporting jobs. Empirical research shows that automation might be far more important in explaining polarisation of the US labour market than trade. Hicks and Devaraj (2015) show that 88 percent of US job losses between 2000 and 2010 is the result of productivity growth due to e.g. automation, rather than trade or a change in domestic demand.

The bottom-line is: Trump can impose trade barriers, but he cannot reverse technological progress. The fact that trade barriers are unlikely to restore jobs lost in the past decades should refrain Trump from turning to large-scale protectionist measures. 

The economic impact of protectionist trade policy on the US and the world

We have outlined five reasons why we believe Trump won’t will not resort to full-fletched protectionism. So, what if Trump does agree with the conclusions regarding the negative impact of protectionism on corporates, but is more concerned about the negative impact of globalisation on certain groups of workers? What if he is more concerned about being able to keep some of his campaign promises and this makes trade a clear target given the difficulties he is facing in terms of domestic fiscal reform? In short: what is the economic impact on the US and world economy if Trump decides to implement protectionist trade measures anyway.

To answer this question, we have used the macro-econometric trade model NiGEM to run two scenarios. In the first scenario, we assume Trump decides to impose an additional uniform import tariff of 20 percent on all products and services imported in the US, in the third quarter of 2017. In the second scenario, we rerun the first scenario, but assume that foreign countries will retaliate by imposing an additional tariff 20% on goods and services exported from the US.

Scenario 1: Protectionist measures by the US

Table 2: Scenario 1: additional uniform import tariff of 20% on all exports to the US
Table 2: Scenario 1: additional uniform import tariff of 20% on all exports to the USSource: Rabobank

The results of scenario 1 are illustrated in Table 2. The table shows the effects in percentage points vis-à-vis our baseline. In the US, for example, the economy would grow by 1.4 percentage points less in 2018 than our baseline. We currently expect the US economy to grow by 2.4 percent in 2018. Hence, in case of a uniform import tariff the US economy would grow by 1.0%. In the rest of the world we see that countries with a large internal market, like China and Germany, are relatively shielded from external trade shocks. Small open countries like the Netherlands and countries that have the strongest trade ties with the US, like Canada and Mexico, are hit relatively hard. Canada suffers more than Mexico, as the Canadian economy depends more on exports than the Mexican economy.

What causes growth to slow?

So why do trade barriers have a negative effect on GDP? Let’s start with the US. The 20 percent additional tariff leads to higher import prices in the US and accordingly to higher domestic inflation. Not only the price of imported consumer goods rises, but also the price of intermediate inputs. This will result in higher export prices, as production becomes more costly. Higher import and export prices lead to lower private consumption, lower investment, and lower export and import growth.

On top of that, higher inflation feeds into higher interest rates. Higher inflation will force the Fed to hike its monetary policy rate to stem inflation. Of course, the Fed would take into account the fact that it is dealing with cost-push inflation and not demand-pull inflation. As such, in order to prevent significant damage to the economy, the Fed will calibrate its policy rate on nominal GDP developments (nominal GDP targeting) instead of the usual mix of nominal GDP and inflation (Taylor rule). Obviously, inflation developments will still influence Fed policy, as nominal GDP growth also depends on inflation. In any case, higher interest rates discourage private consumption borrowings and investment. At the same time, higher interest rates increase demand for dollar assets. Consequently, the US dollar appreciates, slightly lowering the negative impact of the import tariffs. That said, the USD appreciation props up prices of US export goods and services, lowering global demand for US products.

The upshot is that the US economy suffers. The same holds for economic growth in the rest of the world. As imports in the US are more expensive, export growth in other countries slows. Lower export growth also feeds into, amongst other things, lower investment, employment and consumption growth. As lower growth in the rest of the world also lowers import demand in the rest of the world, the US economy takes another hit in the second round.

Scenario 2: Retaliation

In scenario 1, we have only imposed an additional import tariff in the US. But protectionist measures rarely stay unanswered and countries tend to retaliate. For instance, in response to the softwood lumber levy, Canadian officials have announced that the government will look at options to cease shipment of thermal coal from ports in British Columbia. These are small-scale threats, but a large-scale retaliation by China to US-imposed trade barriers could have a devastating effect on the US economy.

Table 3: Scenario 2: the 20% import tariff by the US on all foreign import is retaliated by trading partners by an additional 20% tariff on all US products shipped abroad
Table 3: Scenario 2: the 20% import tariff by the US on all foreign import is retaliated by trading partners by an additional 20% tariff on all US products shipped abroadSource: Rabobank

To estimate the effect in case of retaliation, we ran a second scenario using NiGEM. The US again imposes an additional uniform import tariff of 20 percent in the third quarter of 2017. However, the rest of the world retaliates by imposing an additional import tariff of 20 percent on products and services shipped from the US to these countries, in the fourth quarter of 2017. Table 3 shows the results. In our second scenario, the impact on US GDP could be as large as 3 percentage points, which means that the US economy would end up in a recession.

The US is hurt via the same channels as the other countries in the previous scenario, and the other countries via the same routes as the US in scenario 1. The main difference is that the US faces a higher import tariff by all of its trading partners, while the other countries are only confronted with an additional tariff in the US. Accordingly, the negative impact on economic growth in the US is larger than in most other countries. Canada again is the exception, due to, as explained, it’s relatively large dependence on exports to the US.

Conclusion and discussion

Figure 11: German car sector outperforms US in terms of R&D and productivity
Figure 11: German car sector outperforms US in terms of R&D and productivitySource: Rabobank based on EUKLEMS database, OECD STAN database and OECD ANBERD database. Explanation: R&D intensity measures business R&D expenditure as a % of value added and encompasses the motor vehicles industry (ISIC 29). TFP encompasses the total transport equipment sector (ISIC 29-30). TFP measures productivity growth that cannot be attributed to either additional capital or labour inputs.

In this Special, we have discussed several reasons why we believe Trump is unlikely to unleash a full-fletched protectionist trade agenda. The most important reasons are political ties, membership of the WTO, the possibility and the cost of retaliation, the existence of global value chain integration and revenue streams, and the fact that automation rather than trade has caused most job losses in the US. That said, all of these reasons are based on economic rationale. If Trump will continue to face setbacks with his domestic policy agenda, the pressure to take steps implementing his protectionist agenda might rise. Yet, while we believe such steps could boost his popularity in the short run, it is likely to cost him in the longer run, due to the negative economic impact on GDP, retailers and households.

In order to foster job growth, Trump should instead focus on fostering domestic innovation and human capital. That way he could boost US total factor productivity growth and revitalise competitiveness of US manufacturing. In this sense, German foreign minister Gabriel Sigmar had a case in point, stating that Trump should not focus on raising import barriers on German cars sold in the US, but rather should pay attention to building better American cars (Figure 11).


Acemoglu, D., D.H. Autor, D. Dorn, D., G.H. Hanson and B. Price (2015). Import competition and the great US employment sag of the 2000s. Journal of Labor Economics, 34(S1), S141-S198.

Autor, D.H. and D. Dorn (2013). The growth of low-skill service jobs and the polarization of the US labor market. American Economic Review, 103(5), blz. 1553–97.

Autor, D.H., D. Dorn and G.H. Hanson (2015). Untangling trade and technology: Evidence from local labour markets. The Economic Journal, 125(584), 621-646.

Autor, D.H., L.F. Katz and M.S. Kearney (2006). The Polarization of the U.S. Labor Market. American Economic Review Papers and Proceedings, 189-194.

Asian Development Bank, How the iPhone widens the United States trade deficit with the People's Republic of China.

Center for Automotive Research (2016). The growing role of Mexico in the North American automotive industry – Trends, drivers and forecasts.

Ebenstein, A., A. Harrison, M. McMillan and S. Phillips (2014). Estimating the impact of trade and offshoring on American workers using the current population surveys. Review of Economics and Statistics, 96(4), 581-595.

Erken and Tulen (2017). US global value chain integration: a major impediment for Trump’s protectionist trade agenda. Rabobank, Utrecht

Goos, M., Manning, A., and Salomons, A. (2009). Job polarization in Europe. American Economic Review, 99(2), 58-63.

Hicks, M.J. and S. Devaraj (2015). The myth and reality of manufacturing in America, Ball State University.

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

Marey, P., A. Dumitru, H.P.G. Erken, M. Wijffelaars and E. Barendregt (2017). De impact van Trump op de economie. Rabobank, Utrecht.

OECD (2016). Reshoring: myth or reality? OECD Science, Technology and Industry Policy Papers no. 27, Paris.

Oldenski, L. (2014). Offshoring and the polarization of the US labor market. Industrial & Labor Relations Review, 67(3), 734-761.

Vance, J.D. (2016). Hillbilly Elegy. HarperCollins Publishers.

Hugo Erken
RaboResearch Netherlands, Economics and Sustainability Rabobank KEO
+31 6 2223 1650
Philip Marey
RaboResearch Global Economics & Markets Rabobank KEO
+31 30 71 21437
Maartje Wijffelaars
RaboResearch Global Economics & Markets Rabobank KEO
+31 6 2257 0569

naar boven