Undershooting Ahead

We think the ECB will continue to provide reassurance this Thursday beyond “merely” delivering another widely expected 25-bps cut. Bringing the policy rate to 2% takes a particular significance since this level is often seen as the “neutral rate”. Yet, we think the ECB, without providing explicit forward guidance, will make it clear that it can “break” this level and descend into accommodative territory without a too painful internal discussion. The new macroeconomic projections should help. Indeed, relative to the March batch, the Eurosystem’s economists will take on board the appreciation in the euro exchange rate and the further decline in oil prices, which should raise the probability that inflation could undershoot the ECB’s target for a longer share of the forecasting horizon than in March. The picture for the real economy is unlikely to be rosy, even if the German fiscal push could lift the GDP projection somewhat towards the end of the projections.

Still, the pace and magnitude of such cuts will largely depend on the outcome of the EU-US trade talks – which is a strong reason why providing explicit forward guidance is difficult now. Uncertainty on this front has risen further after a decision by the US Court of International Trade invalidating a lot of the tariffs imposed by executive orders (those taken in retaliation against drug and people smuggling, and the “reciprocal tariffs). The Court’s decision has for now been stayed by the Federal Circuit Court, and the case may have to go all the way to the Supreme Court. We explore the case in some detail as it raises some fundamental issues, notably on the separation of power in the US, which has brought together a surprising alliance of Democrats, traditional Republicans and Libertarians. We note however that the US President would still have at his disposal a vast arsenal of tariffs to use, should the Supreme Court side with the CIT. A risk is that EU negotiators, instead of dealing with one, major tariff on most products, end up having to respond to a myriad of sector-specific tariffs, which could make the discussion even more complex.

Clear “dovish bias” – despite uncertainty – for the ECB

The ECB’s more straightforward dovish tilt was already quite apparent at the April meeting, and we do not think that the Governing Council needs to innovate much this week, on top of delivering a widely expected additional 25-bps cut, to continue to provide reassurance for the Euro area amid persistent uncertainty. Removing in April the reference to the degree of restriction of monetary policy in the prepared statement - which had turned the 2% level for the deposit rate, likely hit this week, into a “totemic threshold” announcing tough battles to get past it – was a key move in our view. Christine Lagarde justified it in the April Q&A by stating that “the neutral rate (…) is a concept that works for a shock-free world, but we are not in a shock-free world.” This opens the door to bringing the policy stance more squarely in accommodative territory “if need be” beyond this week’s move without too much pain. Public statements by prominent hawks, such as Pierre Wunsch, Governor of the National Bank of Belgium, who in an interview with the Financial Times on 17 May openly contemplated “bringing the policy rate slightly below 2%” strengthened that message. Merely continuing to state this Thursday that the European Central Bank will “take whatever measure is appropriate” to deliver on the inflation target would suffice to give a strong indication to the market that more accommodation is in the pipeline beyond this week. ​

Indeed, the possibility of an inflation undershoot – often summarily dismissed just a few months ago – must be taken seriously, and this is something which could be reflected in the ECB’s new set of forecasts. In the March 2025 vintage of the projections, convergence to 2% was slower than in the December batch for headline inflation, albeit with a lower “anchoring” for late 2026 and 2027 and even some undershooting at the end of 2026 (see Exhibit 1). Conversely, on core inflation – explicitly more relevant for the ECB now - convergence to 2% was expected to happen more quickly in 2025 (see Exhibit 2). A further downward revision – entailing some longer brush with “undershooting” – would be natural in the new version of the forecasts released this Thursday.

Indeed, the March projections were established while the euro exchange rate was still under pressure from the “US exceptionalism” theme. The assumption for the euro dollar over the forecasting period stood at 1.04, resulting in a 1.6% depreciation relative to the 2024 level. The June batch, conversely, will reflect the more recent “flight from the dollar” thesis. `Given the usual “cut-off” date for the collection of the market data used to build the assumptions, the average level seen in the first half of May – EURUSD 1.126 – should be close to the new FX assumption, producing an 8.2% gain relative to 2024. In trade-weighted terms (see Exhibit 3) the upgrade should also be visible, at +4%. That would be enough to lower inflation and GDP growth by at least 0.2% by the second year of the projections, using the ECB’s own elasticities.

Oil prices are pulling in the same direction. Oil prices were held at USD74.7 per barrel for 2025 in the March projections. The spot price stood at USD64.2 per barrel on Friday, and over the first half of May, the future contract for December 2025 stood on average at USD65. This will have a strong impact on headline inflation (around 0.3/0.4%), with some more limited effect on core prices dynamics (e.g. via production prices and second-round wage effects).

There is enough here to trigger some meaningful downward revision to the ECB’s already “dovish” March trajectory. Of course, towards the end of the forecasting horizon, the impact of the technical assumptions will fade, and the underlying cyclical story will take precedence. On the real economy side of the projections, a lot will hinge on the Eurosystem’s take on the German fiscal push. In our view, it should not start materialising in a visible manner in the Euro area’s aggregate demand dynamics before well into 2026, and German profligacy will be to some extent offset by fiscal restraint in most other member states. In addition, given the magnitude of the current external headwinds, it would be very “brave” for the central bank to produce a rosy macro story in this particular projection exercise.

True, the hawks could insist on the fact that weak business confidence in the Euro area (see Exhibit 5), now obvious in both the services and the manufacturing sector, is not translating into depressed selling price expectations (see Exhibit 6), at least in services. Indeed, according to the European Commission business survey in the services sector, expected selling prices are still standing around one standard deviation above their long-term average (we have used here 1999-2024), with little sense of improvement recently. We suspect however that this reflects more an extrapolation of recent cost pressure than the emergence of properly new challenges. Services are particularly sensitive to labour cost developments, and it has taken long for wages to moderate – as often in Europe given its institutional features. The deceleration has however finally started in earnest, as illustrated by the sharp slowdown in negotiated wages in Q1 2025 (see Exhibit 7). We have little doubt this will continue in the months and quarters ahead. The latest European Commission quarterly survey in the services sector signalled another decline in hiring difficulties, which are slowly converging towards their long-term average, while concerns over demand are building up, even if they are still manageable by historical standards (see Exhibit 8).

The ECB hawks will of course choose to focus on the fact that difficulties arising from access to funding are also starting to wane and never hit their long-term average, which would add to the sense that the level of restriction of monetary policy is already relatively low. This is no longer the issue though, as the removal of the comment on the level of restriction from the prepared statement suggested in April already: irrespective of whether the neutral rate has been hit, we believe that the majority of the Governing Council will consider that the ECB can take the risk of offsetting the external headwinds with more monetary “push” beyond this week, given that the risks of inflation undershooting are now more prevalent than upside risk.

The US arsenal is not depleted

The pace of the subsequent cuts though will depend on the level of uncertainty and the magnitude of the trade shock, and this means that the ECB cannot engage in full-bodied forward guidance. The Governing council meets without any clarity on the trade war, as the normal term for the negotiations with the US is now back to 9 July. In addition, while the EU’s trade commissioner has made it public that the schedule of his conversations with the US side was unchanged, the latest Court decisions in the US – and new announcements by Donald Trump on steel and aluminium tariffs – could change the features on any future agreement. If rather than having to deal with a “one size fits all” reciprocal tariff on the US market, the EU faces a myriad of sectorial tariffs, negotiations would take a different – and possibly even more complex – turn. Even more fundamentally, if ultimately the White House is limited in its capacity to make changes to tariffs via executive orders, this would be another source of complications in the talks.

We first need to dig into quite a bit of “legalese” to understand the – profound – issues at stake. The Court of International Trade (CIT) – which has jurisdiction over most trade-related matters across the US – ruled on 28 May that the International Emergency Economic Powers Act (IEEPA) does not authorize the President to set Trafficking tariffs (the ones used against Canada, China and Mexico, linked to Fentanyl smuggling and immigration) or Worldwide Retaliatory tariffs (the more commonly called “reciprocal tariffs”). The news had of course been (carefully) saluted by the market last Friday, but the government has appealed to the Federal Circuit Court which has granted a stay of execution on the CIT’s decision (which was to be entirely expected) while it is looking into the substance of the case. The issue may have to be ultimately settled by the Supreme Court.

Precedents are not necessarily that encouraging. Indeed, it is not the first time that the CIT has ruled against Donald Trump. In 2020 it struck a tariff on imports of steel from Turkey. Its reasoning at the time was both procedural – it found the White House had exceeded the time limit set to make a decision – and substantial – it found the White House had taken a too extensive interpretation of its powers. Yet, on appeal the Federal Circuit Court found in favour of the US government, and the Supreme Court declined to take up the case, which at the time was seen as a readiness at the highest level of the US Judiciary to grant the executive branch much leeway on trade policy.

Yet, the case is quite different this time. Indeed, to impose tariffs on Turkish steel, the first Trump administration had invoked Section 232 of the Trade Expansion Act of 1962 (TEA), which can be used only after a report by the Commerce Department finds that imports threaten national security. Section 232 explicitly states that the President can impose tariffs. However, in the case of trafficking and reciprocal tariffs, the White House invoked a different statute, the IEEPA. This implies declaring a national emergency defined as an “unusual and extraordinary foreign threat which has its source in whole or substantial part outside the US, to the national security, foreign policy or economy of the United States.” The IEEPA does not explicitly mention tariffs in the long list of powerful tools which the President can be used (the President can “investigate, block during the pendency of an investigation, regulate, direct and compel, nullify void, prevent or prohibit (…) importations”).

In a nutshell, the CIT considers that the issue at stake for the reciprocal tariff – a persistent trade deficit – does not meet the definition of a “national emergency”, and also accepted the plaintiffs’ view that the absence of explicit mention of tariffs in the list of tools matters: the plaintiffs argued (see link here, page 27) that “under the major questions doctrine, when Congress delegates powers of vast economic and political significance, it must speak clearly” and the CIT ruled that “an unlimited delegation of tariff authority would constitute an improper abdication of legislative power”. Ultimately, separation of powers is the fundamental issue there, and this explains why several prominent libertarian or “traditional Republican” lawyers supported the plaintiffs in the process, including former associates of current conservative members of the Supreme Court (such as John Danforth, former Missouri Senator who was Clarence Thomas’ mentor). The court’s reasoning on trafficking was different: the point there is that they could not find a logical and direct link between the imposition of tariffs and the end of drug or people trafficking. The Court’s language there was almost mischievous. To quote verbatim: “tax deals with a budget deficit by raising revenue. A dam deals with flooding by holding back a river. But there is no such association between the act of imposing a tariff and the unusual and extraordinary threats that the (…) order purports to combat.”

Now, even assuming that the Federal Circuit Court and then the Supreme Court side with the CIT, there would be many avenues open to the US administration to slap tariffs by executive order. Incidentally, the CIT has indicated one of them: indeed, the CIT ruling considers that if the President’s real issue is with balance of payments difficulties, Section 122 of the Trade Act of 1974 can help. Separately, if it is about sectorial issues, then section 301 of the same Act provides solutions. There are however limits to the scope of Section 122: the maximum tariff under this rule is 15%, and it cannot be levied for more than 5 months without Congress consent. Section 301 however would offer much more leeway. Rather than focusing on one sweeping order – e.g. the reciprocal tariffs – the White House could accumulate a series of sector-based tariffs, and there are already investigations (necessary to use Section 301) underway on pharmaceuticals, aircraft, or shipbuilding. Or he could “simply” proceed via the Section 232 of the TEA. Legal challenges could arise if such accumulation were construed as a way to circumvent limits to the other solutions, but the outcome would be uncertain.

The “sector by sector” approach has taken a new shine with Donald Trump’s announcement over the weekend that the tariff on steel and aluminium – taken this time, as for the “Turkish case” of 2020, under Section 232 of the TEA - would be lifted from 25% to 50%. Beyond the economic impact on foreign suppliers of the US – Canada protested vigorously - such decision in any puts in jeopardy any deal with the US. As of last Sunday, the UK government was trying to ascertain if the “special treatment” it had managed to snatch for its own steel industry on the US market was still valid, but crucially, the framework agreement Prime Minister Starmer negotiated with D. Trump as of now is not legally binding.

All in all, last week’s decision is probably more a political setback for Donald Trump than the “beginning of the end” of the trade war. On the margin, it may convince Europeans to take their time in offering concessions to DC while legal uncertainty dissipates, but it is far too early to sound the “all clear” on that front. We would highlight Donald Trump’s points on China “not delivering on their side of the agreement” over the weekend as illustrative of a key risk: that faced with legal and political setbacks, his administration responds by “doubling down” and trigger even more uncertainty and volatility.