Bundesbank Puts the Average US Import Tariff at Almost 12 Percent

By The Edge Research Team

For a year and a half the tariff debate has run on assertion. Each side has had its own number for how high the wall now stands, its own guess at who pays for it, and its own view of whether the damage has already landed or is still in transit. What has been missing is an institution with a macroeconomic model, no commercial position in the outcome and an obligation to publish its workings.

The Deutsche Bundesbank supplied one. Its Monthly Report for July 2026, dated 28 July, carries a full-length article — released a day ahead, on 27 July — on United States tariffs and geoeconomic fragmentation, and it does something the running commentary has not: it states the level, quantifies the effect on prices, output and trade volumes, and shows how much worse a further escalation would be. The central finding is stark in its simplicity. The average United States import tariff rate has risen from 2.3 percent in 2024 to almost 12 percent at present — a level, the Bundesbank notes, last reached in the early 1940s.

That is the anchor figure, and it reframes everything downstream of it. A jump of roughly nine and a half percentage points in the average applied rate is not a trade dispute; it is a change in the price structure of the world’s largest import market. The Bundesbank’s estimates of what follows are notable less for their size than for their asymmetry — the price effect lands quickly and almost fully on the importing economy, while the output effect is spread thin across everyone else.

All figures in this article are taken directly from the Deutsche Bundesbank Monthly Report July 2026, Volume 78 Number 7. The Monthly Report itself is dated 28 July 2026; the tariff article carries a publication stamp of 27 July 2026, having been released a day ahead of the full report. The output and trade estimates are model simulations against the stated baseline, not forecasts. Where the report gives a range, the range is reproduced.

What the tariffs have already done to prices

The Bundesbank’s most immediate result concerns inflation in the United States itself. Tariffs increased the inflation rate, as measured by the private consumption deflator, by 0.6 to 0.7 percentage point in February 2026. That is a substantial contribution to a headline number, and it is worth being precise about which index it refers to: the private consumption deflator, not the consumer price index. The two behave differently, and conflating them is the most common error in tariff commentary.

More striking is the pass-through estimate. The Bundesbank finds that observed price movements amount to just over 80 percent of the price effect that would be expected under full tariff pass-through. This is not a statement that 80 percent of the tariff has been added to shelf prices. It is a statement that the price response so far has come in at just over four fifths of what a complete, textbook pass-through would have produced. The distinction matters: it means exporters have absorbed relatively little, and the burden has been carried mostly at the importing end.

The final line in that table belongs to the escalation scenario, not to the current path, and should be read only in that context. In the Bundesbank’s escalation simulation the euro-area price level would be 0.2 percent higher in 2027. Symmetric retaliation is a further branch on top of that, which the Bundesbank says would have amplified these effects again — so 0.2 percent is the escalation figure, not the post-retaliation one. On the current trajectory the euro area’s problem is one of demand, not of prices.

The output and trade effects are smaller than the rhetoric

Here the report cuts against the loudest claims on both sides. After three years, euro-area real gross domestic product is just under 0.3 percent below the baseline. That is a real cost — but it is a fraction of the figures that circulated when the tariff regime was first announced, and it is spread over three years. Global trade flows are dampened by just over 3 percent. United States real imports run around 9 percent below the level that would have been expected in 2027.

The pattern is consistent: the tariffs bite hardest on the volume of trade and on the price level in the tariffing economy, and comparatively lightly on real output elsewhere. Trade is redirected rather than extinguished. That is the mechanism behind one of the report’s sharper observations — Chinese exports to the United States fell around one third, a collapse in a bilateral corridor that did not produce anything like a proportionate fall in global trade, because the goods went somewhere else.

The Bundesbank is careful about how much weight that redirection can bear. On the European end of it, the affected products account for only around 10 percent of Chinese exports to the European Union, which means the macroeconomic impact of the diversion effects has, in the Bundesbank’s words, been very limited so far. Diversion is a real mechanism; it is not yet a large one.

The escalation branch roughly doubles the damage

The Bundesbank does not stop at the current regime. It models a further escalation, and the results scale in a way that should concentrate attention on what has not yet happened rather than on what has.

An effective rate above 20 percent would be roughly double the almost-12-percent level already reached — itself the highest since the early 1940s on the Bundesbank’s own historical comparison. The doubling of the damage is the key number: it says the cost function is not flat, and that the marginal escalation is more expensive than the average one already incurred.

What German firms report on the ground

The report pairs its model work with survey evidence from the Bundesbank’s own firm panel, and this is where the transmission becomes concrete. Around 11 percent of German firms exported to the United States in 2025 — a minority, and a useful corrective to the assumption that the German economy as a whole is directly exposed. But the effect on that minority has been sharp, and the uncertainty effect has been broader still.

The standard deviation of firms’ sales expectations widened by 3.4 percentage points to 8.8 percent. That is the statistical signature of firms no longer agreeing with one another about what happens next, and it is the channel through which tariff policy reaches investment decisions in companies that never ship a container to Baltimore. Just over half of the affected firms postponed investment. Around two thirds expected weaker demand, and only 16 percent reported no effects at all.

The distortion nobody is pricing

One finding in the report has nothing to do with tariffs and deserves separate attention. Citing the World Trade Organization, the Bundesbank records that artificial-intelligence-related goods accounted for around 42 percent of the 2025 increase in global goods trade, while making up around one sixth of trade overall. A single product category is therefore carrying a wildly disproportionate share of trade growth.

The implication runs in both directions. It means aggregate trade figures have been flattered by a concentrated technology cycle and are weaker underneath than they look. It also means that any interruption to that cycle would remove a support that the tariff analysis does not otherwise account for. For readers in the Gulf, where technology and data-centre investment has become a stated policy priority, this is the more consequential of the report’s two structural findings.

Why it matters: A central bank has now put a defensible number on the tariff regime, and the numbers do not support either of the prevailing narratives. The regime is historically large — almost 12 percent against 2.3 percent two years ago, a level not seen since the early 1940s — and the price effect is real and concentrated in the United States, at 0.6 to 0.7 percentage point on the private consumption deflator with just over 80 percent of full pass-through observed. But the output damage abroad is modest and slow: just under 0.3 percent of euro-area real GDP over three years. The risk is not in what has happened. It is in the escalation branch, where the damage roughly doubles and the effective rate passes 20 percent.

Outlook: Four things to watch. First, whether the 15 percent ceiling on the vast majority of European Union exports holds, since the escalation scenario is defined by its breach. Second, the pass-through ratio — if it moves above the just-over-80-percent mark, exporters have stopped absorbing and the price effect will extend. Third, the German firm panel’s investment-postponement share, currently just over half, which is the cleanest available read on whether uncertainty is easing. Fourth, the artificial-intelligence trade concentration at around 42 percent of the 2025 increase, because a slowdown there would remove the main support under global trade volumes.

Sources: Deutsche Bundesbank, Monthly Report July 2026, Volume 78 Number 7, published 28 July 2026.

Get Complimentary Insights

Follow us