Peak trade, a pause, or just back to normal: What’s up with world trade?

19 Oct 2016


  • Mark Thirlwell
  • Global Value Chains
  • International Trade
  • World Economy

Health warning: Apologies readers, this is a very long post! Although a fair bit of that is charts...

In my recent post on the outlook for global growth and world trade, I noted that the IMF devotes a whole chapter (pdf) in this month’s World Economic Outlook (WEO) to investigating some of the possible causes of the current slowdown in world trade growth.

That slowdown has been visible in a series of downgrades to official trade forecasts, as I’ve highlighted in both the cited post and in previous ones. It’s also quite apparent in the trajectory of the ratio of global trade to GDP – often used as a summary globalisation statistic. As the chart below shows, this ratio had been on a rising trajectory until it was interrupted by the global financial crisis. Post-crisis, there was a quick recovery in the level of the ratio, but still no sign as yet of any resumption of the previous upward trajectory.

Ratio of World trade to World GDP

This kind of pattern has prompted a range of different diagnoses about the state of international trade[1]. One hypothesis is that we have now reached ‘peak trade’; that is, that there are no further gains in globalisation to be had from this point onwards as the benefits from trade liberalisation, technology and the expansion of global value chains (GVCs) have pretty much all been exhausted. A closely- related hypothesis is that the previous period of rapid growth in world trade was actually quite abnormal, and so now we are returning to a more ‘normal’ relationship between trade and overall economic growth. On the other hand, another view would suggest instead that rather than a new normal for trade growth, instead we’re stuck in a kind of extended pause, due mainly to the current slow pace of growth in the global economy, and that once this period is over, trade growth will return to something closer to its previous pace.

So, just what does the IMF have to say about this debate? The analysis in the new WEO starts off by pointing out that there has been a sharp decline in the rate of growth of both the volume and the value of world trade in goods and services in recent years. Indeed, the Fund emphasises that while the pace of volume growth has slowed significantly since the end of 2011, the value of world trade in US dollar terms has actually collapsed since the second half of 2014, with the value of goods and services trade falling by more than 10 per cent in 2015.

World import growth since 2013: Values, Prices and Volumes

However, there’s no great mystery as to what was behind that deep fall in trade values: basically, there was a sharp fall in import prices due mostly to a sizeable drop in commodity prices (especially the oil price) and the impact of a marked appreciation in the US dollar[2]. What’s rather more puzzling is the recent trend in trade volumes, which the Fund points out have only grown by around three per cent a year since 2012 - a rate which is less than half the average rate of growth experienced during the previous three decades.

World Real GDP and trade growth, 1960 to 2015

At the same time, there also seems to have been a change in the relationship between trade volumes and GDP growth. From the mid-1980s to the period before the global financial crisis, trade volumes grew at around double the pace of GDP growth. In contrast, over the most recent period, trade growth has shown signs of struggling to keep pace with real output growth.

Ratio of growth in World Imports to growth in World GDP

This broad aggregate story also conceals some interesting details: as I’ve noted before, the downturn in growth has been much stronger for goods trade than it has been for trade in services, for example. And within goods trade, although the slowdown has occurred across all sub-components, the severity of the decline in growth has varied quite significantly by product. For example, the size of decline in trade in nondurable consumption goods has been smaller than the fall in growth rates for capital goods and primary intermediate goods.

Real import growth by product group

Another difference is that although both developed economies and emerging markets have seen their import growth slow, the downturn in developed economies came earlier, associated with the Eurozone debt crisis, and has since shown a modest recovery. Meanwhile, for emerging economies the trade slowdown was initially much less severe but has since intensified over the past two years, driven by a decline in the pace of Chinese import growth and by much weaker conditions in many commodity exporting developing countries.

Real import growth by market group, and Contribution to real import growth

To identify the key factors behind these changes, the IMF’s economists estimated the historical relationship between import volumes of goods and services and growth in demand (broken down into investment, private consumption, government expenditure and exports each weighted by their import content) and shifts in relative prices. This estimated relationship was then used to predict import growth and then actual and predicted import growth were compared to see how well the current slowdown in trade can be explained by these factors. On this basis, the Fund finds that for world trade, about three-quarters of the decline in import growth can be explained by the combined effect of slower overall economic growth, a shift in economic activity away from import-intensive investment to less import-intensive consumption, and the fact that the growth slowdown has been synchronised across a range of markets.

In other words, the key to understanding most (but not all) of the slowdown in world trade is pretty straightforward: slower economic growth in general and weaker investment in particular takes most of the blame.

What about the other (unexplained) 25 per cent of the slowdown in world trade? Here the Fund canvasses a range of possibilities, with a focus on trade costs (including tariffs, transport and logistics costs), trade policies and the role of GVCs. And as a bonus, a review of the same data also tells some interesting stories about several key features of the contemporary international trade landscape.

First, and despite frequent talk of a swing to protectionism, formal tariff barriers to trade actually remain quite low overall, if appreciably higher on average for emerging economies than for developed markets. On the other hand, progress in lowering tariffs further has now stalled. So the data show that for both developed and emerging economies, import tariffs fell quite substantially between 1986 and the conclusion of the Uruguay Round of trade negotiations in 1995, and then continued to decline thereafter, albeit at a more modest pace, with the trend continuing until about 2008. Since then, tariff declines have been negligible. Still, they haven’t gone back up, either.

Tariff rate in advanced economies, and Tariff rate in emerging economies

Second, there are signs that protectionism in the form of non-tariff barriers has increased, with the World Bank’s data on temporary trade barriers data and the Centre of Economic Policy Research’s Global Trade Alert measures both showing an increase in trade restricting measures.

Temporary trade barriers, and Trade measures recorded by GTA

A third factor influencing the global picture on trade costs is the spread of free trade agreements (FTAs). The growth in numbers of FTAs was particularly strong over the decade of the 1990s, when new FTAs were signed at an average rate of nearly 30 a year, but since 2011 the rate has dropped down to about ten a year. Still, although there are fewer new agreements, those that are signed tend to be larger, covering both larger countries and broader areas of the economy. The combined impact of those two trends means that the overall coverage of FTAs measured as a share of global GDP has continued to rise, but at a slower rate than before.

FTAs by year signed, and Coverage of FTAs

Next, the logistical and transport costs association with trade (in both dollar terms and in terms of the time taken to trade) have fallen significantly in emerging economies since 2006 up until quite recently and have remained relatively flat for advanced economies for most of the same period[3].

Cost to import, and Time to import

One last feature of the trade environment that is often linked to the current trade slowdown is the role played by global value chains (GVCs)[4]. Here the basic argument is that the rapid expansion of GVCs in the 1990s and early 2000s that helped supercharge trade flows back then is now mostly over[5]. It’s true that measures of GVC participation do fit with that pattern, with participation for both developed and emerging economies rising until the financial crisis and then levelling off in the years since[6].

GVC participation - Global, and GVC participation by market

So, where does this (lengthy) review leave us? The key result is that the main factor behind the slowdown in world trade appears to be the overall weakness in the global economy, which in turn implies that a big turnaround in trade growth is likely to require a recovery in the pace of global activity, including in particular in investment rates. At the same time, the lack of additional trade momentum coming from past drivers of rapid trade expansion (lower trade costs, lower trade policy barriers, expanding GVCs) suggests that even a return to a more healthy global GDP growth rate might not be sufficient to fully restore world trade growth to its past dynamism, at least not without new measures designed to reinvigorate cross-border commerce.

[1]  For another earlier and very detailed look at some possible causes of the global trade slowdown, see the various chapters in this book released by the CEPR last year, which covers a range of explanations taking in changing trade patterns, the role of GVCs and protectionist trade policies. A more recent investigation of these themes is available in this ECB occasional paper (pdf) published just last month, which reaches similar conclusions to the IMF analysis discussed in this post: the authors find that global trade growth has mainly been dampened by two factors: (1) First and most important are compositional factors including changes in the geography of economic activity (due to a rise in the relative weight of emerging economies in the world economy, with those countries tending to have a lower trade intensity than developed economies) and shifts in the composition of demand (away from import-intensive, investment-led growth); and (2) Second and less important but still significant, structural shifts including slower growth in GVCs, rising non-tariff barriers and the exhaustion of gains from previous positive developments such as increasing WTO membership and lower transportation costs.
[2]   Most commodities are priced in US dollars and as the dollar appreciates, commodity prices tend to fall since the same quantity of resources can now be bought with fewer greenbacks.
[3]  The Fund notes that an exception to this story was the cost of air freight, which rose quite steadily between 2002 and 2012 before falling back again due to the later decline in oil prices.
[4]  I took a look at some measures of Australia’s integration into GVCs earlier this year.
[5]  That could reflect a natural maturity of the process, or it could be down to the rise of non-tariff barriers and political and business pressures towards greater ‘localisation’.
[6]  Participation in GVCs is measured here as the sum of (1) the domestic content in a country’s exports that is then reused in the exports of its trading partners and (2) its own exports’ foreign value added as a share of gross exports.