Skip to main content

[Speech]Panel : "Globalization at an Inflection Point": Remarks At the Jackson Hole Economic Policy Symposium Hosted by the Federal Reserve Bank of Kansas City
(August 26, 2023)


UEDA Kazuo
Governor of the Bank of Japan
August 28, 2023

I would like to thank the organizers for inviting me to this symposium. It is my great pleasure to participate on this panel. The topic is a very broad and difficult one: addressing structural changes in the global economy caused by the possible trend of de-globalization and discussing their macroeconomic implications. Being no expert in trade theory, I would like to confine myself to sharing with you the picture we are seeing from Japan on these key questions.

To provide a preview of my main points, trade and foreign direct investment (FDI) patterns in Asia are changing partially in response to rising geopolitical tensions. As far as Japan-related trade and FDI flows are concerned, there has been some diversification of production from China into the rest of Asia, to some extent into North America, and also back to Japan. Some of this has been going on for a while, and thus can be more appropriately viewed as continued attempts at globalization, while some more recent diversification seems to reflect a response to geopolitical risks. The net effect of the latter flows on Japan and the world economy remains very much uncertain, but skewed to the downside. The uncertainty poses a difficult challenge for monetary policy making.

I. Japanese Experience in the 1990s

Let me begin with the Japanese experience some three or four decades ago with respect to trade policy and its effect on the pattern of trade involving Japan, the U.S., and Asia, although the nature of trade conflict back then was very different and less threatening than what is happening now. As a result of sharp increases in Japanese exports to the U.S. in the 1970s and 1980s, the U.S.-Japan trade relationship deteriorated significantly--culminating in the adoption of trade restrictive measures such as tariffs, voluntary export restraints, etc. The trade tensions resulted in a permanent change in the structure of foreign trade involving the U.S. and Asia. By sometime in the 1990s, as shown in Chart 1, a significant portion of Japanese exports to the U.S. had taken a detour by way of the rest of Asia. I hasten to add that a more important cause of the change in the trade structure was the rise in Japanese wages relative to the rest of Asia. But I suspect that it is possible to carry out a statistical analysis that finds the U.S.-Japan trade relationship had some role. Needless to say, I also would add that a changing pattern of regional business investment--increases in inward FDI into the rest of Asia and stagnant domestic investment in Japan--was a major driver of such a relocation of production.

Another consequence of the trade friction was the tendency for Japanese firms to produce in the U.S. by setting up plants there (horizontal FDI). As pointed out by Alfaro and Chor (2023), total sales of goods of Japanese origin in the U.S. are much larger than U.S. imports from Japan.

II. Trade and FDI in Asia

Coming back to the main theme of the panel--possible fragmentation of the world economy due to geopolitics--I think that the basic reference is the IMF's World Economic Outlook released this April (IMF (2023)), especially the analysis in Chapter 4. One of the most important conclusions of the analysis is that the emergence of permanent barriers to FDI between the U.S. and China blocs, according to their baseline hypothetical scenario, would cause a large decline in world GDP. The negative effect on GDP would be much more serious in Southeast Asia, as a result of the area's geoeconomic proximity to China and its heavy reliance on inward FDI.1

How does such a picture fit with recent trade and FDI flows around Asia? Chart 3 shows movements in Japanese exports. The left half shows that the share of Japan's exports to China rose during the pandemic but has declined somewhat since then, while those to the U.S. and the rest of Asia have held up. For individual countries, there is a noteworthy trend, shown in the right half of the same chart, that exports to Vietnam have risen consistently and those to India have risen more sharply for a few years.2

Turning to Japan's outward FDI, Chart 4 shows that FDI to China has been stagnant for some time, FDI to North America has rebounded since 2019, and FDI to the rest of Asia has been steady, with some strength in FDI to Vietnam and India.

Chart 5 shows FDI plans for 2023 and beyond among large Japanese firms based on a survey carried out by the Development Bank of Japan (DBJ). North America stands out as the most important destination for Japan's FDI, followed by China and the rest of Asia, where four countries--Vietnam, Thailand, India, and Indonesia--appear to be more important than others.

These charts as well as some anecdotal evidence that we obtain from our contacts may be summarized as follows. There is some diversification of production bases from China into ASEAN, India, and North America. Flows to ASEAN and India are motivated not only by geopolitical considerations but also by demand increases in the host countries. Flows into the U.S. are also demand-driven but may be affected by U.S. industrial policies such as the IRA and the CHIPS and Science Act as well.3

Regarding reshoring back to Japan, Chart 6 shows that an increasing number of Japanese firms have plans to expand domestic production capacity, but not entirely at the expense of foreign production capacity. At the sectoral level, firms in the auto, general machinery, and chemical industries still plan to increase capacity in Asia, while there is a clearer tendency in semiconductors and related industries to increase domestic capacity with support measures by the government.

  1. The data on inward FDI into China, based on the Balance of Payments statistics, indicate sharp declines recently (Chart 2). This does not quite match trends in Japanese/U.S. data and needs to be analyzed further.
  2. July trade data were published on August 17. They show strong Japanese nominal export growth to North America and Europe and declines to Asia, except to India and Indonesia.
  3. South Korean and Taiwanese exports of high-tech goods are usually a good indicator of global trends in the sector. Since around 2022, exports of high-tech goods from these areas to China have been declining, while those to the U.S. remain steady. It is, however, still difficult to determine whether this is a result of geopolitical factors or just a reflection of the macroeconomic strength of the U.S. and Chinese economies.

III. Course of Fragmentation and Globalization

Such a picture seems to be somewhat at odds with the IMF simulation analysis I described earlier. If emerging Asia were to be adversely affected by fragmentation, businesses would try to move out of the area. Asian countries, even including China, however, continue to be hosts to Japanese FDIs. In fact, one of the charts presented in IMF (2023) on global FDI flows in strategic sectors (their Figure 4.4.) also shows that, although flows into China have been declining since 2019, flows into Asia excluding China have been fairly resilient.

Interpretation of the resilience of Asian production sites is not straightforward. Perhaps many firms assume that fragmentation risks would be contained to a small number of specific sectors. Or they may just keep doing the same as before until the geopolitical picture becomes clearer. Yet another possibility is that, as the title of the panel suggests, we are slowly approaching an inflection point beyond which many things will change drastically. It appears fair to say that, at least in the semiconductor sector, friend-shoring and reshoring activities are taking place on a non-negligible scale.

Meanwhile, regional economic integration in developing Asia has become even deeper. Chart 7 shows that intraregional shares in developing Asia have risen not just for total trade and FDI, but also for intermediate goods and ICT goods trade--evidence of deeper vertical integration within the region. Such an increase in regional economic dependence, of course, is partly a reflection of China's attempts to de-route production by way of the rest of Asia, as pointed out by Alfaro and Chor (2023), akin to what happened to Japan and Asia 40 years ago. As was the case with Japan then, rising wages in China must be a major factor behind such relocation, in addition to geopolitical forces. Anyway, globalization forces are still alive in the region. Should the region's economic integration go into reverse, the world would lose not just some of the gains from free trade, but also the efficiency gains from Marshallian externalities arising from an agglomeration of manufacturing plants in the region.

Another non-linearity-related story regarding the region is about the medium of exchange. China's attempt to relocate production or cultivate trade relationships has extended well beyond Asia to now encompass regions such as South America and Africa. Along with such efforts on the trade front, China has strategically encouraged the use of the renminbi in trade finance. The currency's role remains rather small compared with that of the U.S. dollar as the world's vehicle currency, but it has been growing in some areas according to SWIFT data, for example (Chart 8). The choice of the medium of exchange is essentially a multiple-equilibrium story. Thus, even a temporary change in the structure of trade flows could result in a persistent change in the choice of currency supporting trade flows.

IV. Implications for the Outlook for Japan's Economy and Monetary Policy

Let me now turn to the question of how all this affects Japan's economic outlook. The Japanese economy started the year with an expansion led by consumption and investment--3.7% (SAAR) growth in real GDP in Q1. The strength of the economy was to a certain extent a response to the relaxation of pandemic-related restrictions, including a resurgence in inbound tourism. Growth in Q2 was also high, at 6%, but this was largely due to declines in imports while strength in tourism continued. Private consumption declined in Q2, partly due to bad weather, but we think that domestic demand is still on a healthy trend--although this is something that needs to be checked with Q3 data. Business fixed investment is supported by record-high profits as well as structural factors, such as labor shortages, digitalization, climate changes, and the tendency toward expanding domestic, relative to foreign, capacity.

On the inflation front, the rise in import prices in 2021-2022 has spilled over to domestic prices. The CPI inflation rate (all items less fresh food) was 3.1% in July, but it is expected to decline toward the end of this year. We think that underlying inflation is still below our target of 2%. This is why we are sticking with our current monetary easing framework.

As pointed out earlier, the tendency toward reshoring of manufacturing activities has been a positive for the economy. New investment projects in the semiconductor industry are providing stimulus to the local areas in terms of rising sales in related industries and employment.

Offsetting this is the slowdown in some parts of the world. In particular, the pace of economic activity in China has been a disappointment. Monthly data for July--such as for retail sales, fixed asset investment, and industrial production--were on the weak side. The underlying problem appears to be the adjustment in the property sector and its spillover to the rest of the economy. It is very difficult at this point to detect the contribution of the geopolitical factors to the slowdown in the economy. For the Japanese economy, some offset is provided by the relative strength in the U.S.

Longer-run effects of geopolitical factors on the Japanese economy are unsurprisingly very uncertain. In addition to the factors described earlier, the tit-for-tat war, mainly in the semiconductor sector, between major advanced economies and China is a risk. The enthusiasm about reshoring in key industries that is partly motivated by government subsidies is expected to lead to industrial clusters and accumulation of human capital, thereby raising potential growth. However, there may not be adequate infrastructure to support growth of the projects that are starting. Japan may lose out in the global race to attract top firms. The widespread use of industrial policies globally could just lead to inefficient factories.

Central banks will have a hard time factoring in these forces when making policy decisions. As described, the economic outlook is clouded by a number of effects that geopolitics/de-globalization could generate, many of which will affect the supply side of the economies, as well as the demand side. It will take time to determine how long-lasting these effects will be. As production location shifts over time, researchers will find it difficult to obtain stable statistical results involving regional variables.

Such an environment shares some similarities with the one central banks faced over the last few years. There were a series of pandemic-related supply shocks, the durability of which was very uncertain in real time. Some affected the aggregate demand side of the economies as well. Hopefully, we will learn to cope with such an environment appropriately.


  • Alfaro, Laura, and Davin Chor. 2023. "Global Supply Chains: The Looming 'Great Reallocation,'" paper presented in this symposium.
  • International Monetary Fund. 2023. World Economic Outlook, April.