Today, all hopes of seeing sustained global economic growth in 2016 are riding on a continuing US recovery against the backdrop of a bleaker economic growth outlook for emerging market economies and a very slow recovery in the eurozone. However, despite its size and relatively low degree of openness to international trade, the US economy itself is not immune to contagion from the rest of the world's problems.
We will here focus on US exports that (unless there is a systemic financial crisis that spills over through finance and sentiment) will be among the main channels of contagion of external weakness to the US economy. US exporters are already subject to the dual pressures of a poor economic momentum in several parts of the world and the strong dollar, and these headwinds will continue to take their toll in 2016.
Exports: a relatively small, yet non negligible weight in the US economy
The weight of external demand as a component of GDP is significantly lower in the US than in other major developed economies (graph 1). This is true both when considering gross exports and when considering only domestic US “value added” integrated into products whose final destination is foreign countries (the latter calculation to avoid integrating the import content of exports and the proportion of US exports included in finished goods that are later re-imported into the US1).
Even so, the trend in exports has recently had a significant impact on US GDP (graph2). Rebounding exports were one of the main contributing factors to the US economic recovery in 2010-2011 after being a major drag during the 2009 recession.
Since 2011-2012, exports have had a smaller impact on the medium-term growth trend while weighing heavily on the volatility of quarterly GDP. The overall contribution of net trade to US economic growth has been negative over the last five years, but mostly because of imports rising more than exports, against a backdrop or recovering US internal demand. However, we should also keep in mind that large quarterly variations in exports were, recently, a major cause of unnecessary alarms on growth. Due to this volatility, it is more reasonable to focus on fluctuations in exports over a 12-month period.
Exports of goods now negative over 12 months
Although still positive, over one year, in Q2 2015, exports (goods & services in volume terms) had already sharply decelerated (up only +1.5% from +4.3% four quarters earlier, bringing their contribution to GDP growth down from +0.6pp to +0.2pp).
Export data already available for the beginning of Q3 show an increasing deterioration, with a -2.5% drop in volume (goods only, using a three-month moving average to smoothen volatility) over one year in August, their steepest fall since November 2009. A closer analysis of the breakdown and recent variation of US exports by product type and trading partner allows us to make a few observations and hypothesis.
Modest direct exposure to emerging countries (except Mexico), but indirect exposure via Canada should not be overlooked
Graph 4 shows that emerging countries, today the main sources of global economic worries, account (not counting Mexico) for only around 35% of total US exports. Approximately 10% go to China (including Hong Kong) and only slightly more to Latin America (“value added” data, despite some differences, shows similar orders of magnitude).
Mexico, the second-largest outlet for US exports, must be considered separately, as its economy fluctuates largely in line with the US economic cycle, although it also has a significant exposure to oil).
However, it is also necessary to factor in indirect exposure to emerging countries through Canada, the primary outlet for US exports and a major exporter of commodities (of which the recent plunge in prices stems, at least in part, from the slowdown in China).
So far, the fall in exports seems to bear the mark of the strong US dollar more than that of the global manufacturing cycle
Exports by product type show a clear predominance of capital goods and intermediate goods and a relatively small proportion of consumer goods (graph 5). Thus, US exports could be expected to fluctuate based on the global manufacturing cycle. However, an examination of recent variations in volume terms by product type (chart 6) shows that exports of industrial supplies have remained nearly stable over one year, while it is exports of capital goods and autos that dropped sharply, a hint that the strong USD may have played a major part (assuming exports of capital goods are more sensitive to forex variation than intermediate goods)3.
The fall in exports could also be more directly linked to the momentum of commodities prices than to the Chinese slowdown
Looking at recent variations in US exports by trading partner also gives a few indications, even though the lack of bilateral trade data in volume makes it hard to interpret these statistics. Official nominal figures are shown in the table below, together with “very rough” deflated figures (using the general US exports deflator), knowing that the latter can be very misleading due to the different product mix of exports to each region. The variation of total imports (in volume terms) for a number of large trading partners, shown on graph 9, also offers a basis for comparison.
What can be observed is a very sharp slide in the nominal value of exports to commodity-producing areas (Canada, Latin America, OPEC and Africa). The bulk of this nominal decline stems from oil prices themselves (the United States is a player in the processing chain for these products, at least to some regions) but, after closer examination of detailed US Census figures for August 2015, we also note some very sharp declines in exports of machinery and equipment for the oil and gas industry, most likely due to plunging volumes (accordingly, more than three-quarters of the $4.5 billion decline in capital goods exports between the first eight months of 2014 and the equivalent period of 2015 is attributable to the plunge in the drilling and excavating equipment sub-sectors).
Conversely, while down in nominal terms, exports to China and the rest of Asia ex-Japan (a region linked to China via multiple channels, with commodities not always the major ones) were probably still up, over one year, in August 2015 in volume terms (which seems to be confirmed by figures for total Chinese imports in volume, shown on graph 9).
Exports to Europe, for their part, seemed to be on a slightly declining trend in volume terms. In this case, the main culprit could be the strong dollar, as total imports from the Euro area have been rather on an upward path.
Finally, exports to Mexico were probably on the increase in volume terms, in line with the trend in total imports from Mexico (and with US internal demand, as they are often part of the value-added chain of products whose final destination is the US).
Taken together, the above mentioned figures thus suggest that the rise of the USD and the weakening of commodity-producing economies could have already taken a substantial part of their toll on US exports. Conversely, regarding trade to Asia, and given China’s recent additional loss of momentum, the worst is probably still to come.
What impacts are still to come from the strong USD?
It seems reasonable to assume (as shown by a number of models and research papers, see 3rd footnote and the box on recent estimates by New York Federal Reserve researchers) that the effect of a currency shock, while lasting several years, will usually (with notable recent counter-examples, such as in Japan) exert its maximum effect on a country’s export (and economic) growth rate after roughly one year (with some models pointing to two years).
With this timing in mind, it should be noted that, although the USD recently climbed an additional step in terms of effective exchange rate (following the renminbi devaluation in August 2015), roughly half of its 20% rise since June 2014 was already completed in December 2014 (essentially due to its sharp appreciation vs. the euro). Therefore, the stronger dollar could have had the time to generate a substantial part (perhaps slightly less than half) of its adverse effects on US exports by the end of 2015. It may thus be reasonable to expect a slightly (but not much) higher pressure in 2016, assuming the trade weighted USD remains near its current level. Moreover, regarding the more recent half of the USD appreciation, It should also be noted that the USD’s trade-weighted exchange rate, as calculated by the Federal Reserve, gives a 21% weight to the Chinese renminbi, an order of magnitude that makes sense for total trade (including US imports) but that is excessive if focusing only on US exports.
What impacts are still to come from the rest of the world’s momentum?
The negative effect, on US exports, of weak economic momentum in many parts of the world is likely to be felt with a shorter lag (less than a year) than the effect of currency variations.
More than a year after oil prices started to decline, while the velocity of this decline has slowed and the economies of oil-producing countries have been put to a hard test, the falling export volume to these regions is likely to continue but will not necessarily intensify.
In the direction of Europe, the portion that is yet to come of the adjustment in the volume of exports (linked to past fluctuations in the EUR/USD, assuming no major changes in this exchange rate next year) will be partly offset by the continuation of the region’s economic recovery (that we expect to be slightly stronger in 2016 than in 2015). As such, the most likely scenario is that the decline in US exports to Europe will continue, but without much worsening.
It is likely to be exports to Asia (China and the countries affected by the Chinese slowdown via channels other than commodities) that may hold the most unpleasant surprises in 2016 compared to 2015, due to the additional deterioration of the Chinese momentum in 2015. This will combine with the effect of the devaluation of the renminbi (and the acceleration of other Asian currencies’ declines). However, it is necessary to keep in mind that the weight of exports to these regions in US GDP is small (roughly a quarter of US exports and 2.5% of GDP) and that what is expected is that these countries’ economies will decelerate more, but still retain a positive GDP (and, probably, import) growth rate.
In conclusion, we can expect upheavals in exports, like the one we have just seen, to continue to affect the volatile US quarterly economic growth statistics from time to time (and this is likely to be the case when the first estimate of Q3 2015 US GDP growth is released, based on the latest available figures).
However, looking into 2016, aggregated GDP growth projections for the major trading partners of the United States (graph 10) do not signal a collapse in US exports. Exports will probably, at worst, contract slightly in volume, thus making a negative contribution to US GDP, but that will not be enough to disrupt the momentum of a recovery largely driven by domestic fundamentals. A (pessimistic) -2% decline in volume exports over one year would mean a negative -0.25pp contribution to GDP, or roughly -0.45pp decline in the contribution of exports to GDP compared to its level of Q2 2015: a material, but still manageable headwind.
Rising imports may also continue to play negatively (even more so due to the strong USD), but this is ambiguous as it is also the result of buoyant US domestic demand. Therefore, if a severe slowdown in Asia were to put an end to the US recovery, it will be either through a major financial shock activating spillover channels other than exports, or because it combines its effect with that of waning domestic demand drivers.
1. However, the most recent OECD data on “value added” trade is for the year 2011.
2. In Q1 2014 and Q1 2015, exports generated, each time, a large negative annualised contribution of -0.8pp (that combined with simultaneous rises in imports to heavily depress quarterly GDP) before these movements were partially corrected by opposite variations over the subsequent quarters. In Q1 2015, the total contribution of net trade to US quarterly GDP growth (-1.9pp, annualised) was the most negative since 1984.
3. On these topics, see, among other, Chapter 3 of the IMF’s October 2015 World Economic Outlook titled “Exchange rates and trade flows: Disconnected?”
US exports have recently contributed to false alerts on growth
US corporate profits declined -6.8% at annual rate in Q4 2015 vs Q3 of the same year, according to Bureau of Economy Analysis data (profits before taxes with inventory valuation adjustment).Most of this decline is due to the petroleum & coal sector, whose profit declined $124 bn, and tonet overseas profitsthat fell $6.4 bn.Domestic profits excluding petroleum and coal declined only -1.7% a.r, or $38.8bn.
Global Views Analyst
Bastien Drut & Tristan Perrier
Strategy and Economic Research at Amundi
Global Views Analyst