The EU aims to finance common priorities in defence, energy transition, and digital infrastructure with supranational EU bonds. This column argues that the current exclusion of these supranational bonds from main sovereign indices limits their investor base and translates into worse behaviour during crises. Coordination problems among countries and the reliance on perceived ECB support create further structural challenges for EU bonds. A more coherent approach would involve a European Debt Agency that consolidates the existing issuance of multiple EU institutions and would lead to lower borrowing costs on these EU bonds.
Europe is embarking on one of its most ambitious projects since the creation of the euro. Through new programmes in defence, energy transition, digital infrastructure, and strategic technologies, the EU aims to finance common priorities while creating something the euro area has always lacked: a true European safe asset. This safe asset would strengthen the transmission of monetary policy, support risk sharing across countries, and speed up progress toward a capital markets union (Draghi 2024).
Already during the pandemic, Support to mitigate Unemployment Risks in an Emergency (SURE) and NextGenerationEU were meant to provide cheap financing to countries such as Italy and Spain and to test Europe’s ability to issue safe debt collectively. Thanks to unprecedented urgency and political focus, the European Commission started issuing so-called ‘EU bonds’ backed by the budget of the Union without having to resort to off-balance sheet vehicles such as those proposed by Brunnermeier et al. (2017). Building on this, commentators from across the national and political spectrum are now calling for an expanded use of this instrument (e.g. Garicano 2022, Blanchard and Ubide 2025).
The scale of what is already underway is remarkable. By the end of 2026, the European Commission is expected to have roughly €1 trillion of bonds outstanding. This will make it a larger borrower than Belgium and the Netherlands combined. In 2025, it was already the world’s largest net issuer of debt in euros, ahead of France and Italy. On paper, this should be a major European success. These bonds are widely understood as joint and several liabilities of the member states. They carry a AAA rating, something that US Treasuries have lost, and their credit default swap premia are low and stable. The EU appears to have created an ideal instrument for financing continental priorities and possibly a first step toward fiscal union.
Not issuing as a formal government has a large cost
In practice, this has not happened and EU bonds do not behave like safe assets. More surprisingly, they fail for reasons that have nothing to do with fiscal risk, like their detractors have tried to argue. Figure 1 shows the five-year spread of EU bonds over German Bunds. The spread has averaged an astonishing 50 basis points since 2022, when it widened sharply during the ECB tightening cycle. It is consistently higher than the spread paid by the Netherlands, another AAA borrower, and it has even exceeded the spread of Spain, a country with a much lower rating and one of the main intended beneficiaries of EU borrowing during the pandemic.
Figure 1 Five-year yield spread with German government bonds
A popular explanation is that markets were initially unfamiliar with these instruments and that the premium fell as the Commission built a track record. In Bonfanti and Marcucci (2025), I show that this explanation is incorrect. To illustrate this, I compare the Commission’s yields with those of three institutions that belong to the institutional architecture of the EU. These are the European Stability Mechanism (ESM), the European Financial Stability Facility (EFSF), and the European Investment Bank (EIB). I also include Kreditanstalt für Wiederaufbau (KfW), which is a German policy bank that has no institutional connection to the EU. All four institutions are AAA rated and all of them are ultimately backed by tax revenues.
Despite large differences in mandates, governance, and legal structures, none of these issuers is formally classified as a sovereign because they lack direct legal sovereignty. The left panel of Figure 2 shows that, as a result, they all pay almost exactly the same yield across all maturities, which is much higher than the rate paid by Germany. Even KfW, which is fully guaranteed by Germany and is backed by the same revenue stream as the German state, pays a sizeable premium. The right panel further shows that the decline in yields in recent years is common to all supranationals. The Commission did not gain credibility: every institution in this group followed the exact same pattern.
Figure 2 Yield to maturity in secondary market for selected AAA-rated issuers
Supranational bonds are arbitrarily excluded from indices
The reason is structural. Supranational bonds are excluded from the main sovereign indices that serve as benchmarks for large institutional investors. In a market where many investors must stay close to their benchmarks, this exclusion sharply reduces the set of potential buyers. Using novel data on the exact composition of fixed-income indices, I show that if indexed investors matched their benchmarks exactly, the demand for EU bonds would be over 80% lower than for comparable sovereign bonds. This is purely driven by the legal classification of supranationals as ‘quasi-governments’ and is not related to the characteristics of the bonds.
This matters because a smaller investor base translates into worse behaviour during crises. When market stress hits, some investors need to sell assets to raise cash. With fewer potential buyers available to purchase supranational bonds, their prices fall further. Even investors that are not tied to an index require a premium to hold EU bonds because they anticipate that this smaller investor base makes these securities less valuable precisely when liquidity is most needed.
Investors with liquidity needs avoid EU bonds
To test this explanation, I look at the actual investor base across different AAA-rated issuers. Mutual funds and foreign investors (mostly central banks) avoid EU bonds and are willing to pay a large premium to hold sovereign bonds because they need assets with a high price in bad times. By contrast, commercial banks and insurance companies, which finance themselves with stable deposits or long-dated liabilities, buy EU bonds enthusiastically because they are not forced to sell in bad times.
Their arbitrage does not eliminate these differences because foreigners already own almost the entirety of the supply of sovereign bonds. The scarcity is extreme: the net stock of AAA rated bonds in euros is about €3 trillion, which is smaller than the foreign exchange reserves of the People’s Bank of China alone.
National governments will resist changes to the status quo
If this segmentation is driven by index exclusion, why not simply include EU bonds in sovereign indices? The Commission has tried, but index providers have declined to do so. My research highlights two key reasons. The first is that the market is stuck in an equilibrium where supranationals behave poorly in crises, which makes them unattractive index constituents. They would behave like safe assets if everyone treated them as such, but no index provider wants to be the first to make the switch.
The second reason is a fundamental political economy problem. Reclassifying EU bonds as sovereigns would reduce the index weight of existing national bonds and raise their funding costs. Governments with strained finances therefore have strong incentives to resist any change and can exert considerable pressure on index providers. This creates a coordination failure that markets cannot solve alone.
Conditional quantitative easing and the ECB as an unwilling kingmaker
These slow-moving structural features cannot explain the dramatic swings in the EU spread, which at one point even quadrupled within a year. The explanation lies in the interaction between this market segmentation and monetary policy expectations. The crucial variable is not the policy rate or the volume of current asset purchases. What matters is the perceived willingness of the ECB to intervene forcefully during a severe liquidity crisis. I refer to this as conditional quantitative easing, or a measure of the ‘whatever it takes’ commitment.
Because EU bonds are weaker in stressed markets, they rely more on the expectation of central bank support. When the ECB looks unwilling to intervene, measured by rising implied volatility in bond futures as in Haddad et al. (2025), EU bonds become less attractive because investors know they will fall harder without that backstop. When investors believe the ECB will intervene freely, EU bonds don’t look too different from sovereigns. It is only when investors doubt that support that the spread widens quickly.
Figure 3 shows a very close correlation between this measure of conditional support and the EU spread. This relationship is driven entirely by expectations: the ECB stopped quantitative easing (QE) in 2019, but the spread remained low because markets expected strong support in a crisis and a few months later this expectation proved correct. The implication is clear: the spread will widen again whenever the ECB is perceived as unwilling to stabilise stressed markets due to its monetary policy objective.
Figure 3 Five-year spread with Germany and implied volatility from Bund futures
Policy implications: Marginal improvements will not work
What should the Commission do? It has already tried a long list of policies. It adopted a single funding strategy and created a repo facility, the ECB improved the treatment of Commission bonds in its collateral framework, and a futures market has been launched. None of these measures made the spread paid by the Commission deviate from other supranationals. These measures failed because they did not address the exclusion of EU bonds from the asset class of sovereign bonds.
More recent proposals such as Hildebrand et al. (2025) aim to achieve sovereign status by introducing a predictable issuance calendar, clear revenue streams, larger issuance sizes, and ‘credible commitments’. These characteristics already describe institutions such as KfW and the EIB, yet their bonds do not behave like safe assets. The solution to this problem will not come from incremental changes.
My counterfactual computations show that even very large increases in EU issuance would not close the spread as long as these bonds keep their supranational classification. Instead, if investors began viewing them as sovereign bonds, the spread would shrink to negligible levels even with current quantities and for the same perceived default risk. The cost for member states would be small since their borrowing costs would rise only marginally thanks to the insatiable demand for safe assets coming from foreign central banks. This would also change the attractiveness of the euro as a reserve currency relative to the dollar.
Time to get out of the shadows and be bold
I have a bold proposal to move to a better equilibrium. The outstanding debt of EU institutions, including the Commission, the European Investment Bank, the European Stability Mechanism, and the European Financial Stability Facility, is already close to €2 trillion. Yet most Europeans have never heard of it, and this is a feature, not a bug. EU leaders have long recognised the need for joint borrowing but had to hide it within separate legal entities to avoid political backlash. This approach made joint borrowing possible, but at a large financial cost. Investors never learned to treat these bonds as sovereign and Europe has been paying the price ever since.
A more coherent approach would involve a European Debt Agency that consolidates the existing issuance of EU institutions, as I argued in Bonfanti (2023). This would not require any new net borrowing. It would simply organise better what already exists. The resulting market would be large enough that index providers and global investors could not ignore it. Yet implementing such a reform would require politically costly treaty changes.
In August 2025, Vladimir Putin remarked that “It was clear that the EU […] didn’t have much sovereignty. Now it has become obvious that it has none at all.” The statement is provocative, but it captures a reality that Europe can no longer afford to ignore – fragmented fiscal sovereignty weakens the Union’s capacity to act. This is not just an inefficiency. It is a strategic constraint that carries real geopolitical costs. My research shows that the gains from overcoming this fragmentation are large and the path forward is clear. With fiscal pressures mounting and external threats rising, Europe must finally decide whether it is willing to build the sovereignty it claims to have.
Source : VOXeu
