Economie globale
The global economy has entered a new macroeconomic regime.
The first half of 2026 has been marked by significant shocks, yet global growth has proved more resilient than expected. Financial conditions have stabilized rapidly, macroeconomic imbalances have remained contained, and several major economies have continued to expand at a relatively solid pace.
But this resilience should not be mistaken for a return to the relatively benign environment of the past decade.
The underlying drivers of growth have changed fundamentally. Real interest rates are now structurally positive, the cost of capital has increased, and economies face simultaneously rising investment needs in defense, energy transition, digital infrastructure and industrial sovereignty. At the same time, the fragmentation of the global economy is increasingly producing divergent economic trajectories rather than a common shock.
In this environment, the key question is no longer simply whether the global economy can absorb another shock. It is whether the conditions underpinning its current resilience can continue to hold.
Our latest Perspectives conjoncturelles trimestrielles identifies five conditions that will determine the durability of the current cycle: the normalization of the oil market, the continuation of the AI investment cycle, the anchoring of inflation expectations, fiscal sustainability and financial-market resilience.
The global economy is proving more adaptable to shocks than in previous cycles, but the mechanisms behind this resilience are different.
The return of positive real interest rates has fundamentally altered the allocation of capital. Governments and companies are no longer operating in an environment of abundant and inexpensive financing. At the same time, investment requirements have increased substantially, particularly in defense, energy, digital infrastructure and industrial capacity.
This creates new trade-offs. The central issue is increasingly one of capital allocation rather than simply access to financing.
Technological developments are also contributing to a widening divergence between economies. The United States is benefiting from an investment cycle centered on artificial intelligence, while several Asian economies are attracting substantial semiconductor investment. Europe, by contrast, remains more dependent on a recovery in domestic demand and on the delayed transmission of public investment plans.
Central banks remain important in stabilizing the cycle, but they are no longer its primary driver. Fiscal policy, geopolitics, industrial strategies and technological investment are becoming increasingly important determinants of economic performance.
This is why the resilience observed during the first half of 2026 should be regarded as conditional rather than structural.
The resumption of hostilities involving Iran in July provides a useful illustration of this new environment.
The conflict has not fundamentally altered our baseline scenario, which assumes alternating periods of military tension and negotiation rather than uncontrolled escalation. However, it has highlighted an important change: the global economy has fewer buffers available to absorb another major energy shock.
During the first phase of the conflict, strategic reserves, commercial inventories and spare production capacity helped limit the impact on oil prices. These buffers have since become considerably thinner. A prolonged disruption to oil flows would therefore have a faster impact on energy prices, inflation expectations and financial conditions than it did earlier in the year.
Our baseline remains based on a gradual normalization of the oil market and the emergence of a structural supply surplus, allowing Brent prices to converge toward the $60–70 per barrel range in 2027. But the key change is the increased sensitivity of the global economy to an additional supply shock.
The second condition is the continuation of the AI investment cycle. AI has evolved beyond a sector-specific theme and has become an important driver of investment, trade and financial valuations. Yet this dynamic depends increasingly on expectations of future productivity and profitability. A downward revision to those expectations could affect not only technology valuations but also corporate investment and global demand.
Third, inflation expectations need to remain anchored. The energy shock has temporarily interrupted the disinflationary process, but medium-term expectations remain relatively stable. This credibility allows central banks to adopt a gradual approach. A more persistent energy shock, however, could generate broader cost pass-through and prolong restrictive monetary conditions.
Fourth, fiscal sustainability is becoming increasingly important. Governments must finance defense, energy transition, reindustrialization and aging populations at a significantly higher cost of capital than during the previous decade. A deterioration in sovereign risk could amplify pressure on long-term interest rates and refinancing costs.
Finally, financial markets remain resilient, but that resilience reflects strong confidence in the baseline scenario. Tight credit spreads, limited equity risk premia and the concentration of equity indices in a small number of technology companies all create potential vulnerabilities if expectations change.
The common thread is therefore interdependence. A shock to one condition can quickly affect the others.
The United States remains the strongest-performing major developed economy in our scenario, with growth projected at 2.1% in both 2026 and 2027. The important distinction is that the U.S. cycle appears to be slowing in pace, rather than fundamentally changing in nature.
Financial conditions are central to this assessment.
Credit spreads remain historically tight, equity markets are strong, household financial wealth is high and corporate lending is gradually recovering. Consequently, the Fed Funds rate alone no longer provides a complete picture of the effective monetary stance.
The U.S. economy is increasingly being supported by business investment, particularly in digital infrastructure, data centers, semiconductors and computing capacity. At this stage, AI is primarily generating a capex and aggregate-demand shock, rather than a fully visible productivity shock in national statistics. Investment is already supporting economic activity, corporate earnings and financial markets, even though the productivity gains expected from AI have yet to be fully demonstrated.
This creates the principal vulnerability of the U.S. outlook: concentration.
A growing share of investment, profits and stock-market performance is linked to a limited number of technology companies. The risk is not necessarily a traditional speculative bubble. Much of the investment is real. The vulnerability instead comes from the expectations embedded in current valuations.
If the economic return on AI investment were to disappoint, the adjustment could begin in financial markets before transmitting to corporate investment and the wider economy. The principal risk is therefore not an abrupt recession, but a gradual questioning of the financial mechanisms currently supporting the investment cycle.
Inflation is another important feature of the U.S. outlook. We expect average CPI inflation of 3.4% in 2026, easing only gradually to around 2.6% in 2027. The issue is no longer simply the speed of disinflation, but the level at which inflation ultimately stabilizes.
Our scenario therefore assumes that monetary tightening can increasingly operate through financial conditions and the structure of the Federal Reserve’s balance sheet rather than through further increases in the policy rate. We maintain the Fed Funds rate at 3.75%, while the 10-year Treasury yield gradually moves toward 4.6–4.7%.
The eurozone presents almost the mirror image of the United States.
We have revised our 2026 growth forecast down to 0.4%, while maintaining a more positive outlook for 2027, with growth expected to reach 1.1%. This is primarily a question of timing rather than a fundamental deterioration in the medium-term outlook.
The mechanisms required for a recovery are already emerging. Real wages are rising again, inflation has moderated, monetary easing has improved financing conditions and credit demand is gradually recovering.
Yet these improvements have not translated into sufficiently strong private demand.
Households remain cautious, while companies are waiting for a more sustained improvement in order books before committing to a new investment cycle. The eurozone therefore remains in a low-growth regime, with confidence, industrial production and financial conditions still acting as important constraints.
The main source of medium-term optimism is Germany’s investment plans, alongside European initiatives in defense, energy transition and AI.
The challenge is not the scale of these measures, but their speed of transmission to the real economy. The implementation of projects, the ramp-up of industrial capacity and the response of private investment will take time.
We therefore expect the synchronization between recovering domestic demand and a stronger investment cycle to occur in 2027 rather than 2026.
Inflation should average 2.5% in 2026, before returning to 2.0% in 2027. Against this backdrop, we expect the ECB deposit rate to remain at 2.25% through the end of 2027. The increase in long-term Bund yields should instead reflect higher term premia, greater sovereign issuance and Europe’s investment needs rather than a renewed tightening cycle.
The United Kingdom faces a more difficult combination of constraints.
We expect growth of just 0.8% in both 2026 and 2027, with no clear domestic engine capable of generating a sustained acceleration. Consumption remains constrained by housing costs and accumulated losses in purchasing power, while private investment remains subdued.
At the same time, inflation is proving more persistent than in the eurozone. We expect average inflation of 3.3% in 2026, before gradually declining to an average of 2.8% in 2027.
This limits the Bank of England’s room for maneuver. We expect Bank Rate to remain at 3.75% throughout 2026, with gradual cuts beginning in 2027. Yet monetary easing should not be confused with a normalization of financial conditions.
Long-term gilt yields are expected to remain around 5.3–5.4% in 2027, reflecting a higher term premium, substantial government financing requirements and concerns surrounding fiscal credibility.
The result is an unusual configuration in which the central bank can reduce short-term rates while sovereign financing conditions remain tight.
This makes fiscal credibility an increasingly important economic variable. The future government’s ability to demonstrate that higher public investment will genuinely improve potential growth will be crucial for keeping the term premium under control.
Japan is following a different path altogether.
The Bank of Japan is progressively exiting a monetary framework built around negative rates, large-scale JGB purchases and yield-curve control. We expect growth of 0.5% in 2026 et 0.8% in 2027, while the policy rate gradually rises toward 1.50% in 2027.
The domestic backdrop is becoming more supportive. Strong wage increases are improving household purchasing power and lending greater credibility to a wage-price cycle compatible with the Bank of Japan’s inflation objective.
But Japan remains highly dependent on imported energy and raw materials. This means that a renewed energy shock could still weigh significantly on real incomes and consumption.
The key challenge for the BoJ is therefore to normalize short-term rates without destabilizing the JGB market.
We expect 10-year JGB yields to rise toward 3.5% by the end of 2027. This normalization is necessary, but an excessively rapid increase could affect bank portfolios, public debt servicing costs and institutional asset allocation. Conversely, normalization that is too slow could weaken the yen and amplify imported inflation.
The implications extend well beyond Japan.
As domestic Japanese yields become more attractive, Japanese insurers, pension funds and banks may have less incentive to increase their exposure to foreign bonds once currency-hedging costs are taken into account. The issue is not an immediate repatriation of capital, but a potential structural decline in Japan’s marginal demand for U.S. Treasuries, gilts and other developed-market bonds.
The yen therefore becomes an important test of the consistency of the BoJ’s strategy. Too much weakness would suggest that normalization is insufficient; excessive appreciation could tighten domestic financial conditions too rapidly.
The global economy is not entering a synchronized downturn. Nor, however, is it returning to the conditions that prevailed before the pandemic and the subsequent inflationary shocks.
The current cycle is more political, more fragmented and more dependent on investment and capital allocation.
The United States is benefiting from a powerful technology investment cycle, but its growing concentration creates financial vulnerabilities. Europe has the ingredients for a stronger recovery, but the transmission of fiscal and monetary support to private demand is taking longer than expected. The United Kingdom faces a particularly tight interaction between inflation, fiscal constraints and sovereign financing conditions. Japan is normalizing monetary policy after decades of exceptional accommodation, with consequences that increasingly extend to global bond markets.
Across all four economies, the same broader principle applies: policy space is becoming more conditional.
The resilience of the global economy ultimately depends on several delicate balances holding simultaneously: oil markets must continue to normalize; AI investment must deliver sufficient returns to sustain current expectations; inflation expectations must remain anchored; governments must preserve fiscal credibility; and financial markets must continue to absorb a higher cost of capital.
None of these conditions is guaranteed.
This does not mean that the baseline scenario is about to break down. Rather, it means that the distribution of risks has changed. The probability of continued resilience remains high, but the economic cost of a deterioration in one of its key conditions is becoming increasingly significant.
For investors, corporates and policymakers, the implication is clear: the central challenge is no longer simply forecasting the next shock, but understanding the transmission mechanisms that determine whether the global economy can absorb it.
That is the defining feature of the new macroeconomic regime: resilience remains possible, but it is increasingly conditional.
