Firms’ access to external finance differs widely across Europe. This column uses data from the ECB’s Survey on the Access to Finance of Enterprises to map these differences and to study what they mean for monetary policy. The authors find that countries in which firms face tighter financial frictions respond less to the ECB’s monetary policy changes, and that they face a worse trade-off between output and inflation. A model of the monetary union suggests that financial integration in the spirit of the Savings and Investments Union would make monetary policy more effective, but potentially at a cost to financial stability.
Europe’s credit and banking markets are still far from being fully integrated: whether a firm can obtain a loan, and on what terms, depends to a large extent on the country in which it operates. The Savings and Investments Union – the successor to the Capital Markets and Banking Union agenda – is meant to change this, so that firms across Europe face a level playing field in their access to external finance (European Commission 2025). Much of the debate is about what this would do for investment and growth (Caivano et al. 2025) and how the reform should be designed (Angeloni and Cavallini 2026). Less attention has been paid to what it would mean for monetary policy.
The ECB’s single monetary policy does not affect all member states alike. Countries with more fragile banking sectors benefit the least from unconventional monetary policy measures (Burriel and Galesi 2018), and transmission is stronger where more households live hand-to-mouth (Almgren et al. 2022). Firms matter, too: those with high default risk respond less to monetary policy (Ottonello and Winberry 2020), and in the euro area the response of investment varies with firm age and industry (Durante et al. 2022).
In our recent paper (Haas and Jamilov 2026), we study how differences in firms’ financial frictions across European countries shape the transmission of monetary policy. We establish three results:
- Firms’ access to external finance differs substantially across Europe, and persistently so.
- High-friction countries respond less to monetary policy and face a worse output-inflation trade-off.
- Financial integration would make monetary policy more effective, but potentially at a cost to financial stability.
The geography of financial frictions in Europe
To study the geography of financial frictions, we use granular data from the ECB’s Survey on the Access to Finance of Enterprises (SAFE). The survey, introduced in 2009, is conducted twice a year (quarterly since 2024), with around 11,000 firms surveyed in each round. We have access to a granular micro dataset that contains all surveys conducted between 2014 and 2023 and identifies firms at the regional level. We count a firm as facing a financial friction if it reports an obstacle in obtaining a bank loan: its application was rejected, it received only part of the amount requested, it declined the loan because the cost was too high, or it did not apply because it expected to be rejected. Our measure is the share of such firms in a region or country.
Figure 1 shows a clear core-periphery pattern: firms in north-western and central Europe face significantly fewer obstacles than firms in the South and East. At the country level, averaged over 2014–2023, the share of firms facing financial frictions ranges from below 4% in Luxembourg to 27% in Greece – a nearly sevenfold gap.
In the paper, we document two further facts. First, this heterogeneity is mostly due to country-specific factors: firm size, age, and sector play only a minor role. Second, it is persistent, with only gradual convergence over time. This points to structural, country-specific causes. Indeed, we show that financial frictions are strongly related to bank loan recovery rates (European Banking Authority 2025) and to a broad range of institutional-quality indicators.
Figure 1 The geography of financial frictions in Europe


Note: Share of firms that face financial frictions, in percent, region-specific averages. Regions are sorted into five groups (left) and into core and periphery (right).
Source: ECB SAFE and authors’ calculations.
High-friction countries respond less to monetary policy
Does this geography matter for monetary policy? In the cross-section, it is not immediately clear whether countries where firms face tighter frictions are more or less responsive: the financial accelerator (Bernanke et al. 1999) points to amplification, whereas Ottonello and Winberry (2020) find that US firms with high default risk respond less.
We find that high-friction countries respond less. Using data for 22 countries over 2000–2020 – the 19 pre-pandemic euro area members plus Denmark, Bulgaria, and Croatia, whose currencies were pegged to the euro – we estimate how the responses of output and inflation to identified ECB monetary policy shocks vary with the level of financial frictions. Figure 2 shows the result for a monetary expansion. The blue line is the difference in the response that comes with a one-standard-deviation increase in financial frictions. It is negative and persistent: after two years, output has risen by about 14 basis points less. Inflation also rises by less, with a gap of up to around 6 basis points.
Could financial frictions be picking up something else? High-friction countries also tend to have lower productivity, more hand-to-mouth households, and more concentrated banking sectors. The other lines in Figure 2 control for each in turn: the output differential is essentially unchanged, and the inflation differential becomes larger. We find the same pattern for unemployment and investment.
Figure 2 Heterogeneous responses to a monetary expansion
a) Output


b) Inflation


Note: Difference in the response of output (top) and inflation (bottom) to a one-standard-deviation expansionary monetary policy shock that is associated with a one-standard-deviation increase in financial frictions, in basis points. The other lines control, one at a time, for country-level total factor productivity (TFP), the share of hand-to-mouth households (HtM), and banking-sector concentration (HHI). Shaded areas are 90% and 68% confidence bands around the baseline.
Why financial frictions dampen monetary transmission
To understand this result, we build a New Keynesian model of a monetary union with 22 countries that differ in the bank loan recovery rate – how much lenders recover when a firm defaults – as well as in productivity and size. We map our measure of financial frictions into recovery rates using data from the European Banking Authority (2025).
Where recovery rates are low, the cost of external finance rises steeply as firms borrow more. When the central bank cuts interest rates, firms in these countries therefore expand borrowing and investment by less. The model matches the data well: the differences in the responses of output, investment, and inflation between high- and low-friction countries are quantitatively similar to those we estimate.
A worse output-inflation trade-off in high-friction countries
We highlight two results from the model. First, high-friction countries face a worse output-inflation trade-off. We summarise this trade-off with the Phillips multiplier of Barnichon and Mesters (2021): the cumulative response of inflation relative to that of output after a monetary policy shock, or how much inflation a country incurs for each unit of output gained. A higher multiplier means a worse trade-off.
Figure 3 plots the multiplier for each country in our model against its loan recovery rate. The relationship is negative at every horizon, and the dispersion across countries widens over time. After three years, the same output gain costs roughly a fifth more inflation in Greece, the country with the lowest recovery rate, than in Luxembourg (a multiplier of 0.25 against 0.21). Where the credit channel is weak, a monetary expansion does relatively little to stimulate real activity, but inflation responds all the same. The geography of financial frictions thus translates into a geography of output-inflation trade-offs.
Figure 3 The Phillips multiplier


Note: The Phillips multiplier is the ratio of the cumulative inflation response to the cumulative output response following a monetary policy shock. The figure plots country-level multipliers at horizons of 2, 6, and 12 quarters against the loan recovery rate, with fitted regression lines. Model simulations.
Financial integration: Monetary policy effectiveness versus financial stability
Second, financial integration makes monetary policy more effective, but at a potential cost to financial stability. We study a convergence experiment in the spirit of the Savings and Investments Union: the loan recovery rate is made common across all countries and, as a benchmark, raised to its upper bound, so that loan enforcement no longer differs across countries. In the paper, we show that the long-run gains are sizeable – output is roughly 8% higher than in the baseline – and concentrated in the countries where frictions were tightest.
Figure 4 shows how this changes monetary policy. In the integrated economy, the same interest rate cut has a larger effect on output and inflation: monetary policy becomes more effective. But credit and leverage respond more strongly, too. Financial integration is therefore a double-edged sword: it strengthens monetary policy, but it may weaken financial stability – a trade-off between macroeconomic and financial stabilisation that echoes Coimbra and Rey (2024).
Figure 4 Financial convergence and the effects of monetary policy


Note: Panel (a) shows loan recovery rates in the baseline economy and in the convergence counterfactual with a common loan recovery rate at its upper bound. Panels (b) to (f) show cumulative aggregate responses in the two economies to a cut in the policy rate of one percentage point (annualised). Model simulations.
Inequality, spillovers, and robustness
The paper contains further results. A monetary expansion raises cross-country inequality in output, wages, and firm leverage. Country-specific shocks spill over to the rest of the union through the common monetary policy response, the more so the larger the country. And the geography of monetary transmission is preserved in richer versions of the model that add firm heterogeneity within countries and endogenous default, an additional liquidity friction, and capital mobility across countries. The same holds when we vary the length of firms’ life cycle.
Conclusion
Where a firm is located in Europe still shapes its access to external finance, and this matters for monetary policy. In line with evidence for US firms (Ottonello and Winberry 2020), we find – in the data and in the model – that low-friction countries respond most strongly to a common interest rate cut. Financial convergence in the spirit of the Savings and Investments Union would make monetary policy more effective, but potentially at the cost of higher leverage and financial instability.
Source : VOXeu








































































