Leaving China: Which countries might benefit from a relocation of production?
- Uncertainty due to the US-China trade war will probably accelerate a shift of foreign production out of China
- Our new ‘Where Will They Go index’ shows that Thailand, Malaysia, Vietnam, Taiwan and India are likely to benefit from these production relocation activities
- There will probably not be one big winner, as Trump will likely target the countries that benefit most from the shift
- Instead, the benefits will probably be spread across Southeast Asia, in specific industries, such as semi-conductors in Malaysia or the automotive sector in Thailand
- For China, the shift of foreign activities will hurt the economy in the short term, as jobs will be shed. In the medium to longer term, reduced foreign direct investments will likely lead to slower technological progress and, consequently, lower economic growth
- Dutch companies in China will not likely leave en masse since their activities seem to mostly (68%) service the domestic Chinese market
- Nevertheless, a substantial part of these Dutch companies use China as a manufacturing hub to service American consumers (24%). These firms might consider moving part of their production to Thailand, Malaysia, Vietnam, Taiwan or India
Trade war uncertainty is taking its toll
US-China trade tensions have accelerated after the US administration recently decided to add a 10% tariff on the remaining USD300bn of Chinese import goods China responded by allowing the Chinese yuan to depreciate against the US dollar (Giesbergen et. al, 2019). Financial markets are realizing that any ‘trade talks’ between the US and China will break down sooner or later, as we have repeatedly argued before (here and here). In the meantime, the damage to producer- and investor confidence has long been done and an increasing number of international firms have started rerouting their (intermediate) imports from China to Southeast Asia (most notably Vietnam).
In the longer term though, a possibly more important effect could be the relocation of production from China to other Southeast Asian countries. There is ample (albeit anecdotal) evidence that international firms in China have started to move, or are thinking about moving, production elsewhere, for example, Harley Davidson, Nintendo and Apple (via suppliers such Foxconn and Goertek). Vietnam is often mentioned as the main beneficiary of these shifts in firm activity from China. However, Vietnam is too small to absorb the entire chunk of relocated production capacity from China and a natural question is: where will these companies go?
In this Special, we try to answer this question by looking at countries in Asia that have export baskets similar to China’s, low wages and an attractive investment climate. We have used these metrics to develop what we call the Where Will They Go index (WWGT). We find that alongside Vietnam, Thailand, Malaysia, Taiwan and India will likely absorb production capacity moving away from China. Mexico will also likely absorb a part, but these activities will mostly encompass companies that almost exclusively service the US market. Other companies will likely try to stay close to China to be able to keep servicing Chinese and other Asian customers. We are sceptical that the overall effect will be positive for those countries that absorb production capacity relocated from China.
Our expectation is that no single country will be able to absorb the entire chunk of relocated activities. For one thing, if any single country (even partially) replaces Chinese production, President Trump will likely slap tariffs on these countries as well, limiting the upside. The reason is that Trump views large US trade deficits with countries as an indication of “unfair” trade and he would rather see the US companies involved move production back home and “bring jobs back to America”. This is the reason Trump has recently targeted Vietnam and India.
Secondly, a surge in investment and demand for labor in potential relocation countries could easily wipe out any current cost differential with China. This is likely to happen to Vietnam in the coming few years. Moreover, many of the potential relocation countries will also suffer from lost exports to China, which taken together with an overall decline in sentiment, global growth, world trade and the disruption of global supply chains could very well overshadow the effect of increased investment.
Are international firms actually moving out of China?
There is much anecdotal evidence that international firms have indeed been moving out of China or are planning on doing so (Table 1: The Economist, Bloomberg [here, here, here], Nikkei Asia Review, Wall Street Journal and the American Chamber of Commerce [2018, 2019] and Figure 1). However, part of this trend was already going on prior to the US-China trade tensions and has been driven mostly by a rapid increase in manufacturing wages in China in the past years. The growth of inward foreign direct investment (FDI) in China has been declining consistently in the past 10 years (Figure 2), while manufacturing wages in China have almost tripled over the same period (Figure 3). In fact, rising wages were mentioned as the main challenge of doing business in Asia by a majority (66%) of Japanese companies responding to a December 2018 survey by the Japanese External Trade Organization (JETRO). So moving production capacity from China is a trend that will be accelerated by US-China trade tensions instead of being solely caused by it.
In addition, a large part of international firms will continue their activities in China because their main motivation for operating in China is not cheap production costs, but access to China’s huge consumer market. Firms that use China as a manufacturing hub serving different markets are more likely to relocate these activities.
As yet, hard data does not show a decrease in US FDI to China. In fact, data from the US Bureau of Economic Analysis shows that between 2018 and 2017, the US stock of FDI in China grew by 8% (from USD 108bn to USD 117bn), while the US FDI stock grew only marginally in other Asian countries, such as Taiwan (3%) and India (3%) over the same period. However, it is important to point out that 2018 only covers the beginning of US-China trade tensions, while large shifts in FDI flows only become visible with a certain time lag.
To get a more forward-looking view on investment sentiment towards China, one could look at non-resident portfolio flows to China (the buying and selling of Chinese stocks and bonds by people that do not live in China). Such portfolio flows could give a hint of future FDI flows to China. However, in recent years, the historical correlation between FDI flows and non-resident portfolio flows to China (0.64) has broken down (Figure 4). The reason is that portfolio flows are also driven by short-term factors such as investor sentiment and ‘search for yield’. As such, portfolio flows to China in 2018 actually peaked while FDI flows leveled off. All in all, we believe the pain for China will come with a lag, but also with a vengeance.
How do we gauge where companies will go?
We take a forward-looking approach to gauge which countries could benefit from international companies moving their production from China. Such companies will likely move to countries with:
(i) export baskets that are similar to China’s;
(ii) manufacturing wages which are similar or lower than in China;
(iii) an attractive long-term investment climate in terms of Ease of Doing Business;
(iv) sound institutional quality.
Of course, factors such as transportation costs matter too, which is one reason Vietnam is so popular (it shares a border with China in the north). However, transportation costs are a smaller part of transaction costs than they traditionally were and to keep things simple in this analysis, we chose to ignore transportation costs within Southeast Asia. If we were to take the entire Emerging Market spectrum into account, Mexico, for example, could be quite interesting for some companies. Our analysis also disregards factors such as currency volatility and market size (GDP). The latter would mostly be important for companies looking to tap a country’s consumer market rather than looking for lower production costs. If we were to take market size into account, India’s score, for example, would be higher.
For factor (i) we use the Export Similarity Index of Finger and Kreinin (1979). This index measures the overlap of two countries’ export baskets and thus their export competitiveness. The index ranges from 0% to 100%, where a higher percentage indicates a higher export similarity (see the Appendix for more technical details). A score of 100% here means that a country has exactly the same export basket as the country it is being compared to. Figure 5 shows that within (Southeast) Asia the export baskets of Vietnam, Thailand, South Korea, Taiwan and Japan are the most similar to China, while those of Mongolia, Bangladesh, Myanmar and Pakistan are the least similar. Intuitively, this makes sense because countries such as Pakistan and Bangladesh have specialized in producing textiles, while China specializes more in, for example, the assembly of electronic products. On the other, exports in the latter countries could nevertheless get a boost if parts of China’s exports are replaced by these countries due to the sheer size of China’s export basket.
For factor (ii) we use manufacturing wages. Both academic (Yin et. al., 2014) and survey-based evidence (the 2018 JETRO survey) suggest that manufacturing wages are an important factor for firms when considering moving production to other countries.
For factors iii and iv, we use the World Bank’s Ease of Doing Business Index and the World Bank Governance Indicators, which are both widely used measures to gauge the investment climate and institutional quality of a country in areas such as corruption, political stability and the protection of property rights. Both have been found to be good indicators of FDI attractiveness (e.g. Ali et al., 2010 and World Bank, 2013).
Where Will They Go index: WWTG
Putting our four factors together, we arrive at what we call the ‘Where Will They Go index: WWTG’. The index corroborates why Vietnam has been such a popular destination for moving production from China. Vietnam’s export basket looks like China’s (so the country has the expertise to produce similar products to China), while manufacturing wages are 64% lower than in China (Figure 6). Vietnam does not score well on institutional quality (which, for example, reflects its lack of a democratic system), although it is politically stable and has a decent investment climate (Figure 7). Finally, Vietnam has signed about 17 Free Trade Agreements with several country blocks.
Thailand, Malaysia, Taiwan and India are also likely contenders for production relocation from China (Table 2). Thailand tops the list as its export basket is similar to China’s, wages are 25% lower than in China and the country has investor-friendly policies. Thailand’s main caveat though is political stability, even after recent elections (which leads to policy uncertainty). Again, it should be noted that we only focus on Southeast Asia here due to its proximity to China, and we exclude factors such as market size. Outside of Asia, Mexico is likely to be a contender for FDI originally meant for Chinese as well (due to its similarity to China’s export basket. India could also be substantially higher on the list if its market size was taken into account.
Who will gain the most?
We believe that no single country will reap substantial benefits from an acceleration in production capacity relocation from China because of increased trade tensions.
For one thing, President Trump will likely step in by slapping tariffs on the countries most likely to gain from these developments, as the trade surplus that these countries will run up with the US will probably jump in a short period of time. For the Trump administration these deficits are an indication that countries resort to dishonest trade practices, an accusation Trump has leveled at China on numerous occasions. Moreover, Trump might take additional protectionist measures in order to persuade firms moving out of China to reshore (part of) these activities to US soil. Trump has already threatened Vietnam with tariffs and called it the “single worst abuser of everybody”. The same applies to India. Trump has stripped India of its status as a beneficiary developing country, which effectively means that about USD 6bn worth of exports from India to the US will be subject to tariffs.
International firms will likely anticipate this ‘Trump tax’ and diversify their production locations across multiple countries. This would be similar to the ‘China plus one’ production strategy some international firms are currently pursuing, consciously placing some of their production capacity outside China for diversification of such risks.
Secondly, surging investment in Southeast Asian countries will put pressure on property prices and wages in these countries and will reduce their cost differential with China. For Vietnam, some of these pressures are already being reported in the media. Thirdly, for many Southeast Asian countries China is also one of their biggest export markets. China represents more than 10% of total exports for Vietnam, Thailand and Taiwan for example. The negative effect of a US-China trade war on Chinese GDP growth, sentiment, world trade and the disruption of global supply chains is likely to outweigh the benefits of increased investments (which will take time to fully materialize).
Ultimately, we think that the benefits from relocation of production from China to other Asian countries are likely to be spread out. Specific sectors within countries could gain, for example the semiconductor sector in Malaysia and the automotive sector in Thailand (which might be why Harley Davidson has relocated its production to Thailand). Countries such as Bangladesh, Pakistan and Sri Lanka might also gain, albeit mostly from increased export demand for textiles and tea instead of more FDI.
What will this mean for China?
In the short term, the movement of production capacity from China will likely increase unemployment as jobs move away. As an anecdotal example, US-China trade tensions have reportedly already cost about two million industrial jobs in China.
In the medium to longer term, there could be a large negative effect on productivity (Erken et al., 2018). Ironically, if China wants to move up the value chain and produce more high-tech goods (as it has repeatedly indicated), it needs FDI and more open trade. These are precisely the two factors that it will miss out on as a result of a US-China trade war.
There is a large body of economic literature that shows that FDI and trade have large foreign knowledge spillovers that influence potential growth. For example, international knowledge developed abroad is more effectively absorbed into domestic productivity if a country is more open to either foreign trade (Coe and Helpman, 1995) or foreign direct investments (Branstetter, 2006). Moreover, openness to foreign trade fosters market competition, which stimulates firms to reduce inefficiencies and increase efforts to innovate (for more on this, see our Special The US-China trade War in the rerun). All in all, a reduction in productivity growth will hurt China’s potential growth rate, which increasingly relies on technological innovation, rather than, for example, labor force growth. Finally, global value chains going through China will be disrupted, we will publish an in-depth analysis on this topic soon.
What are the implications for the Netherlands?
Our analysis has some important implications for Dutch companies in China. Firstly, Dutch companies with substantial production facilities in China that service the US market will face the same problems as their US counterparts. These Dutch companies might also (partially) move production capacity to countries ranking highly on our WWTG index, depending on their type of product.
Secondly, in a world that is becoming increasingly protectionist, it might make sense for Dutch multinationals to source goods from more than one region and supplier, i.e. diversify across suppliers and countries. This is similar to the ‘China plus one’ strategy employed by some international firms in China.
Thirdly, it is unlikely that Dutch companies in China will move away en masse. In 2016, the Dutch Ministry of Foreign Affairs conducted a survey on Dutch companies operating in China and found that 68% of the respondents are mainly in China to service the Chinese market. These companies will probably continue to operate in China. However, in the same survey, 24% of the respondents mentioned that they also service American consumers from China. These companies have a greater incentive to move part of their production out of China.
 As an example, Bangladesh’s export of men’s suits and jackets (SITC code 6203) comprises about 15% of its total exports while this is only 0.5% for China. Nevertheless, China’s export of these products is almost twice as big as Bangladesh (USD 12.4bn vs USD 6.6bn). So even if Bangladesh only captures a relatively small part of China’s export in these products, Bangladesh’s export of men’s suits and jackets could increase substantially.
 We use real, annual and comparable (including bonuses and social security etc.) manufacturing wages, based on the 2018 JETRO Survey on Business Conditions of Japanese Companies in Asia and Oceania. Although this survey focuses on Japanese companies in Asia, it seems reasonable to assume that wages paid by Japanese companies to Chinese workers are approximately the same as for other international companies.
 The World Bank Governance Indicators are a widely used set of indicators of institutional quality, which is loosely defined as a measure of the quality of law, individual rights protection and government of a country. See Kaufmann et al.(2010) for a more detailed discussion on these indicators.
 To calculate the WWTG index, we normalize the four variables into z-scores (see Table 1 in the Appendix) and take a weighted average, where we assign a weight of 0.3 to wages and export similarity and 0.2 to institutional quality and ease of doing business. We assign a higher weight to wages and export similarity because we think these factors will matter more for international companies thinking about moving production abroad.
 Mexico’s score on Export Similarity with China, is 0.43, which puts Mexico on the third sport on this metric, just after Thailand. However, Mexico does not score very well on institutional quality (-0.5) and its real annual manufacturing wages (based on OECD data) are higher than the Southeast Asia average (USD 16K vs USD 10K). Overall, Mexico would rank as 7th on our WWTG index (right after Singapore). However, Mexico’s proximity to the US would give it an advantage over Southeast Asia for firms that almost exclusively service the US market.
 We calculate the index based on a 4 digit SITC classification of export products.
Ali, F.A., N. Fiess and R. MacDonald (2010). Do institutions matter for foreign direct investment?. Open economies review, 21(2), 201-219.
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.
Coe, D.T. and E. Helpman (1995). International R&D spillovers. European Economic Review, 39(5), 859–887.
Erken, H.P.G., Giesbergen, B.C.J., and de Vreede, I. (2018). Re-assessing the US-China trade war, Rabobank.
Finger, J. and M.E. Kreinin (1979). A measure of export similarity and its possible uses. The Economic Journal, 89 (356), 905-912.
Giesbergen, B.J.C., de Groot, E. and Every, M. (2019). US-China trade war update: No turning back, Rabobank.
Kaufmann, D., A. Kraay and M. Mastruzzi (2010). The worldwide governance indicators: methodology and analytical issues. World Bank Policy Research Working Paper No. 5430. Available at SSRN: https://ssrn.com/abstract=1682130
World Bank (2013), Doing business 2013: Does doing business matter for foreign direct investment?. Washington DC; World Bank, available from download here.
Yin, F., Y. Mingque and L. Xu (2014). Location Determinants of Foreign Direct Investment in Services: Evidence from Chinese Provincial-level data. LSE Asia Research Centre Working Paper 64, available for download here.
How is Export Similarity calculated?
Export Similarity (ES) measures the overall overlap in the share of exported products between two countries (Finger and Kreinin, 1979). It is widely used as a measure to gauge how comparable the export baskets of two countries are (and their substitutability for FDI and sourcing). ES is calculated as follows: