Last week, data indicated that the US trade deficit surged to a record $131.4bn in January, as businesses scrambled to stockpile goods ahead of President Donald Trump’s Schrodinger’s tariffs.
American economic data has generally been disappointing, but the worsening news on imports has in particular stirred up some angst. Because of the mechanics of how gross domestic product is measured and calculated (imports are subtracted as the point is to avoid double-counting and to measure domestic output) the widening trade deficit has helped send the Atlanta Fed’s widely-followed “GDPNOw” real-time economic forecasting model into a tailspin.
The 2.8 per cent contraction the model spat out was subsequently revised down to -2.4 per cent, and then to -1.6 per cent on Friday, after the latest US jobs numbers. But the nasty GDPNow reading naturally triggered a lot of alarming headlines about how the US seemed to be careening towards a brief recession.
Here’s Thomas Ryan, economist at the consultancy Capital Economics:
The ballooning of the trade deficit to a record high of $131.4bn in January once again stemmed from a huge surge in imports as businesses rushed to fast-track orders before new country- and product-specific tariffs took effect.
. . . This huge drag from net trade is what has done most of the damage to our first-quarter GDP estimate, which now stands at -2.5% annualised, as there has not been an offsetting rise in inventory buildup in the data. The good news is that this should reverse in the second-quarter as imports normalise without a corresponding inventory decline, which is why we are forecasting a strong rebound in GDP growth.
The main culprit, however, was a truly massive surge in US gold imports, as traders have also sought to get ahead of potential tariffs. And this matters a lot when we think about the economic implications.
While the motivation is the same (avoiding tariffs), the economic impact of movements in gold and other goods is markedly different. The vast majority of imports are either consumed or used in the production of other stuff, while gold mostly tends to sit inert and useless in a vault.
The tl;dr is that while all the uncertainty will unquestionably exact an economic toll, the Atlanta Fed’s GDPNow model’s horror reading can probably be relatively safely ignored.
The impact of gold in the US trade balance isn’t easy to spot, as movements in gold bars are well hidden in US statistics. They are strangely incorporated in the category “Finished metal shapes”, which accounted for $20.5bn out of the $36.2bn increase in goods imports in January.
“Unprecedented” is an overused word, but you can see just how extreme the January data is here.
Imports of other goods increased too, but to a far smaller degree. Pharmaceutical imports jumped $5.2bn month-on-month, for example, but this was only a 1.5x increase from January last year. Imports of passenger cars climbed by $1bn, but remained lower than in January last year.
In other words, bullion was boss in the January trade numbers. As David Mericle, economist at Goldman Sachs said:
We noted that most of the widening in the trade deficit since November reflects higher gold imports, which are excluded from GDP because they are not consumed or used in production. The details of the trade balance report indeed indicated that elevated gold imports contributed to the bulk of the increase in imports in January.
If you’re still not convinced of the importance of gold, let’s look at US trade with Switzerland.
Switzerland is the world’s biggest bullion refining and transit hub, and home to the world’s largest over-the-counter gold trading hub (alongside the UK). And the US trade deficit with Switzerland exploded to $22bn in January — nearly the size of the US goods trade deficit with China.
US goods import data matches Swiss customs data, which showed gold exports from the country to the US rose to 192.9 tons in January, from 64.2 tons in December.
You can enter different countries in the field above to see similar trends elsewhere. For example, the US has mostly enjoyed a trade surplus with Australia over the past decade, but a surge in Australia’s gold exports helped push the trade balance into negative territory in January.
But Switzerland seems to have been the big one, further evidence of how gold skewed things in January.
Perhaps fearful of being seen to be wrongly predicting a “Trumpcession”, the Atlanta Fed on Friday published an explainer of its GDPNow model and the gold glitch:
Although GDPNow does distinguish gold from other imports, the Bureau of Economic Analysis does, in tallying up the total of the net exports, subaggregate within GDP. Removing gold from imports and exports leads to an increase in both GDPNow’s topline growth forecast and the contribution of net exports to that forecast, of about 2 percentage points.
The topline growth forecasts also increased today — standard model -2.4 percent to -1.6 percent, “gold adjusted” model -0.4 percent to 0.4 percent — as data from today’s labor market report came in stronger than the model was expecting based on the limited February data the model received prior to that release.
So, a “gold-adjusted” GDP forecast of 0.4 per cent growth. Which isn’t great, but is very different from the scary headline number the Atlanta Fed model is still showing.
Goldman Sachs’ own gold-adjusted GDP forecast for the first quarter has been a more optimistic 1.3 per cent, but on Friday it cut its 2025 growth forecast and upped its “recession probability” to 20 per cent.
Here are the main points from the investment bank’s latest economic update, in case you’re curious, with Alphaville’s emphasis below:
— Larger tariffs will give a larger boost to consumer prices. In the absence of tariffs, we would have expected year-on-year core PCE inflation to fall from 2.65% in January to 2.1% by December 2025. Under our previous tariff assumptions, we expected core PCE inflation to remain in the mid-2s for the rest of the year. Our new tariff assumptions imply that it will instead rise a bit and peak at about 3% year-on-year, and in the risk scenario it would peak at around 3.3%.
— Larger tariffs are also likely to hit GDP harder through their tax-like effect on disposable income and consumer spending and their effect on financial conditions and uncertainty for businesses. While our previous tariff assumptions implied a peak hit to year-on-year GDP growth of -0.3pp, our new assumptions imply a peak hit of -0.8pp. In the risk scenario, this would grow to -1.3pp.
— Taking on board this additional 0.5pp drag on growth from our new larger tariff assumptions, we have reduced our 2025 Q4/Q4 GDP growth forecast to 1.7%, from 2.2% previously. This implies that GDP growth will be slightly below potential rather than slightly above. We have bumped up our unemployment rate forecast by 0.1pp to 4.2% in response.
— We have also raised our 12-month recession probability slightly from 15% to 20%. We have raised it by only a limited amount at this point because we see policy changes as the key risk, and the White House has the option to pull back if the downside risks begin to look more serious. If policy headed in the direction of our risk scenario or if the White House remained committed to its policies even in the face of much worse data, recession risk would rise further.
We’ll find out more when the first proper official US GDP estimate for the first three months of the year is published on April 30.
Read the full article here
