Zoekopties
Home Media Explainers Onderzoek & publicaties Statistieken Monetair beleid De euro Betalingsverkeer & markten Werken bij de ECB
Suggesties
Sorteren op
Pauline Avril
Paul Bochmann
Senior Financial Stability Expert · Macro Prud Policy&Financial Stability, Systemic Risk&Financial Institutions
Mar Domenech Palacios
Graduate Programme Participant · Research, Monetary Policy Research
Stephan Fahr
Principal Financial Stability Expert · Macro Prud Policy&Financial Stability, Macroprudential Policy
Magdalena Grothe
Senior Lead Economist · International & European Relations, International Policy Analysis
Aoife Horan
Peter McQuade
Senior Economist · International & European Relations, International Policy Analysis
Cosimo Pancaro
Riccardo Pizzeghello
Financial Stability Analyst · Macro Prud Policy&Financial Stability, Systemic Risk&Financial Institutions
Martino Ricci
Senior Economist · International & European Relations, International Policy Analysis
Josep Maria Vendrell Simon
Niet beschikbaar in het Nederlands

Risks to euro area financial stability from trade tensions

Prepared by Pauline Avril, Paul Bochmann, Stephan Fahr, Aoife Horan, Cosimo Pancaro and Riccardo Pizzeghello

Trade tensions can be a threat to financial stability, with both the implementation of trade restrictions and trade policy uncertainty resulting in adverse consequences. In this special feature, we show that trade policy uncertainty can adversely affect the real economy as well as banks’ funding, asset quality, profitability and lending. Policy authorities need to identify risks stemming from trade tensions, monitor their transmission and evaluate their potential impact on financial stability. Sound capital and liquidity buffers are financial institutions’ first line of defence to absorb shocks stemming from trade disruptions. However, banks should also conduct regular assessments to identify and evaluate these specific risks. In addition, they should diversify portfolios to minimise their exposure to such risks. A box within the special feature analyses the risks of euro area equities repricing across sectors in response to developments in the United States, with a particular focus on news relating to trade policy.

Escalating global trade tensions have emerged as a significant concern for global growth and financial stability. Following a decade of robust economic growth, trade openness, measured by the global trade-to-GDP ratio, has largely stagnated since 2008 (Chart B.1, panel a), reflecting, among other things, growing scepticism towards globalisation.[1] In addition, adverse geopolitical developments unrelated to trade policy itself may aggravate trade-related tensions, altering the volume of global trade as well as the relative shares of imports and exports between trading partners.

Trade policy interventions have surged in recent years, and especially since 2021. An increasing number of trade measures have been implemented by G20 countries, with restrictive policies prevailing significantly over those aimed at liberalising trade (Chart B.1, panel b). The recent escalation of trade frictions between major economies – especially between the United States and its trading partners – has fuelled trade policy uncertainty and emerged as a critical concern for businesses and policymakers alike. This increased uncertainty has the potential to redirect trade flows, reconfigure value chains, deter investment and dampen economic growth.

Chart B.1

Global trade tensions have escalated in recent years

a) Euro area trade with the United States and global trade openness

b) Trade measures implemented by G20 countries

(2002-24; left-hand scale: percentage of GDP, right-hand scale: percentage of GDP)

(2009-24, number of trade measures)

Sources: World Bank, Global Trade Alert and ECB calculations.
Notes: Panel a: trade openness is calculated as exports plus imports divided by GDP. Panel b: data have been adjusted for reporting lags. The cut-off date for each year is 31 December. Trade measures are classified as harmful if they are likely or almost certain to involve discrimination against foreign commercial interests. Harmful measures include tariffs, quotas and other barriers aimed at protecting domestic industries from foreign competition.

This special feature analyses the implications of elevated trade policy uncertainty for euro area financial stability. The euro area is an open economy that is deeply integrated into international supply chains. This means there is a potential for trade policy uncertainty and for the implementation of trade restrictions to have significant adverse effects. Moreover, the United States is the most important trading partner outside the euro area for most member countries and aggregate euro area trade with the United States has increased strongly since 2010, notwithstanding the global stagnation in trade openness (Chart B.1, panel a). Against this background, this special feature provides a conceptual overview of the channels through which trade tensions can affect euro area financial stability. It then presents empirical evidence of the effects of trade policy uncertainty on euro area economic growth, systemic stress and financial vulnerabilities. Finally, it zooms in on the implications for euro area banks by analysing their exposure to sectors that are reliant on US trade.

1 Trade tensions can heighten financial stability risks

The rise in trade policy uncertainty could have an adverse impact on both the financial system and the real economy. Heightened uncertainty may lead to abrupt shifts in sentiment which would increase the volatility of asset and commodity prices and widen risk premia, tightening financial conditions. It may also increase exchange rate volatility, affecting portfolio investment decisions and capital flows. Reacting to both the direct impact and market risk losses, financial institutions may adjust their portfolio holdings, potentially through fire sales, thereby amplifying financial stress.[2] With regard to the real economy, higher uncertainty over future trade policy can delay and/or reduce investment as firms encounter difficulties when assessing future demand for their products and services. Moreover, consumer confidence can worsen, causing a decline in spending by households (Figure B.1).

In addition to trade policy uncertainty, the implementation of trade restrictions tends to reduce external demand, increase prices and raise production costs. Trade restrictions hinder technological advancement, stifling innovation, productivity and economic growth.[3] In addition, firms affected by trade barriers abroad may reroute trade to the domestic market or to countries with fewer restrictions, which would increase competitive pressures faced by domestic firms at home. Overall, trade policy uncertainty and the implementation of tariffs can lead to weaker economic growth, a higher likelihood of corporate distress and, hence, losses for financial institutions, lowering their resilience. Restrictions on trade, capital flows and international investments also constrain the ability of financial institutions to diversify their portfolios, leading to heightened concentration risk in their balance sheets.

Figure B.1

Trade tensions and their transmission to financial stability

Source: ECB.

The effects of trade tensions on financial stability, both via the real economy and via financial market channels, are mutually reinforcing. Given the linkages between the real economy and the financial sectors, feedback processes may reinforce the direct effects in each sector. The various factors together can lead to systemic risks if the transmission of losses is not mitigated in a timely manner.

2 Trade policy uncertainty and macro-financial stability

The methods used to measure trade tensions focus on trade policy uncertainty and trade restrictions to capture the different ways tensions are transmitted to financial stability risks. This analysis employs the trade policy uncertainty index developed by Caldara et al., which relies on newspaper coverage to count the number of occurrences of terms related to trade policy and uncertainty.[4] The indicator can be driven by market sentiment or changes in expectations relating to restrictive trade policies rather than policies that have actually been implemented (Chart B.2, panel a). To estimate the extent to which uncertainty represents a risk to the financial system, this special feature uses a quantile vector autoregressive model leveraging data for the period between the first quarter of 1990 and the third quarter of 2024. This method assesses the impact of rising trade policy uncertainty on the distributions of euro area GDP growth, systemic stress (as measured by the ECB’s composite indicator of systemic stress, CISS) and financial vulnerabilities (as measured by the ECB’s systemic risk indicator, SRI).[5]

Chart B.2

Rising trade tensions heighten risks to economic growth and financial vulnerabilities

a) Trade policy uncertainty index

b) Impact of trade policy uncertainty on real GDP and financial conditions

(Jan. 2008-Apr. 2025, index)

(x-axis: two, four and six quarters ahead, conditional on trade policy uncertainty shock; y-axis: percentage point impact on cumulated quarterly GDP growth, units of indicators)

Sources: Caldara et al.*, Eurostat and ECB calculations.
Notes: Panel b: the real GDP response is cumulated, thus showing the change in output level over two to six quarters ahead. CISS stands for composite indicator of systemic stress; SRI stands for systemic risk indicator. The size of the trade policy uncertainty shock is set to 1 standard deviation (34 basis points over Q2 1990-Q3 2024), while recent changes have been larger (128 basis points in Q4 2024 and 240 basis points in Q1 2025). See footnote 6 for details of the modelling approach.
*) Caldara, D., Iacoviello, M., Molligo, P., Prestpino, A. and Raffo, A. “The economic effects of trade policy uncertainty”, Journal of Monetary Economics, Vol. 109, 2020, pp. 38-59.

Trade policy uncertainty raises downside risks to economic growth over the medium term. Econometric analysis indicates that an increase in trade policy uncertainty raises the probability of adverse economic developments.[6] This is reflected by the downward shift of the distribution of future growth (the difference between the 10th and the 90th percentiles, the top and bottom of the bars in Chart B.2, panel b) in an asymmetric manner around the median response. Specifically, an increase of 1 standard deviation in trade policy uncertainty lowers the median real GDP forecast by 0.15 percentage points after four quarters and the lower tail (the 10th percentile) by 0.75 percentage points, representing major downside risks. By contrast, the shift in the upper tail (the 90th percentile) is contained, indicating limited upside risks.[7]

Rising trade policy uncertainty results in a limited increase in systemic risk. Not only does trade policy uncertainty dampen future economic activity, it also heightens the financial vulnerabilities captured by the SRI, although to a much more limited extent. Over time, the uncertainty raises leverage through an increase in the debt-to-GDP ratio, the main component of the SRI, but it has only limited repercussions for financial stress.[8]

Box A
How spillovers from US developments differ across euro area equity sectors

Prepared by Mar Domenech Palacios, Magdalena Grothe, Peter McQuade, Martino Ricci and Josep M. Vendrell Simón

This box analyses the risks of euro area equities repricing across sectors in response to developments in the United States, with a particular focus on news relating to trade policy. The EU is heavily exposed to trade with the United States, having one of highest levels of value-added content in US imports of any economy.[9] Since the 2024 US presidential election, protectionist policies have re-emerged as a key policy instrument. In this context, potential trade policy shocks have become a source of concern for international financial markets, increasing policy uncertainty and acting as an important driver of euro area equities. Such shocks can affect euro area equity prices unequally across sectors and translate into repricing risks of different magnitudes. This box highlights the sensitivity of euro area equity prices to US trade policy news across sectors.

Euro area equity prices are sensitive to broader US macroeconomic developments. Equity markets in the euro area, like many other markets, are influenced by both domestic and foreign factors. Among the foreign factors, spillovers from the United States are especially significant, given the size of the US economy and its pivotal role in the global economic and financial system. Euro area equity prices thus typically respond to US macroeconomic shocks, albeit with considerable variation across sectors. Moreover, linkages between euro area and US equity markets have expanded in recent years, with cross-border listings on US exchanges increasing and euro area non-banks becoming more exposed to US issuers.

Chart A

Repricing sensitivity of euro area and US equity markets to tariff-related news during the United States-China trade dispute of 2018-19 and in 2025

a) The impact of tariff policy announcements on euro area industrials equities

b) The impact of tariff policy announcements on US industrials equities

c) The impact of tariff policy announcements on firm-level euro area equities

(x-axis: days, y-axis: percentages)

(x-axis: days, y-axis: percentages)

(x-axis: percentages, y-axis: relative frequency)

Sources: Bloomberg Finance L.P., ECB, Haver Analytics, LSEG, MSCI and ECB calculations.
Notes: Panels a and b: average cumulative stock market returns (percentages) for euro area (panel a) and US (panel b) industrials, measured from the day before each tariff announcement during the period 2018-19 (in blue, with 95% confidence intervals in grey). The returns are estimated using an event study approach based on 11 tariff announcements, as set out in Amiti et al.* (see Table 1 of the paper for the list of events). The model includes dummy variables for the three days preceding and four days following each announcement. Regressions are conducted using daily stock return data spanning the periods 2017-19 and 2022-25. The yellow line shows the average cumulative stock market response, as a percentage, following US tariff announcements in 2025 − 1 February: tariffs announced on Canada, Mexico and China; 10 February: 25% tariffs imposed on steel and aluminium; 4 March: additional 10% levy announced on imports from China, Canada retaliates against the United States; 11 March: additional 25% tariff announced on steel and aluminium, followed by EU and Canada retaliation on March 12 and global trade tariffs on 2 April. The event on 8 April is not included as it was followed by announcements of the 90-day pause in tariffs within the same market opening day. Panel c: distribution of firm-level returns from the STOXX Europe 600 on the events indicated above: 11 tariff announcements in 2018-19 and six tariff announcements in 2025. All other days are days with no tariff announcements between 2018 and 2025.
*) Amiti, M., Gomez, M., Kong, S.H. and Weinstein, D., “Trade Protection, Stock-Market Returns, and Welfare”, Working Papers, No 28758, National Bureau of Economic Research, May 2021.

Adverse tariff announcements tend to lead to significant equity market repricing. Experience of the trade dispute between the United States and China in 2018 and 2019, during which a series of tariffs were introduced, shows that trade tensions tend to have a significant negative effect on equity valuations, both in the country that imposes tariffs and more broadly in other countries. In that period, equity prices declined by around 2% on average in both the United States and the euro area following the announcement of tariffs (Chart A, panels a and b, blue lines). The effect of the tariffs imposed between February and May 2025 was broadly similar for the euro area but much more pronounced for US markets (Chart A, panels a and b, yellow lines). The fall in equity indices in response to tariff announcements is typically very broad-based across firms. This is indicated by the distribution of firm-level returns, which is positioned clearly to the left compared with days when no trade announcements were made, both in 2018-19 and in response to the tariff announcements in 2025 (Chart A, panel c).[10] The response of the most exposed euro area firms to recent tariff announcements appears more pronounced than during the first Trump presidency, as the left tail of the distribution is more negative.

The resurgence of trade tensions and tariff-related news has put pressure on equity prices in some euro area industries. Euro area equities have outperformed US stocks since the inauguration of President Trump. All the same, the potential implementation of new tariffs weighed on the equities of firms with greater trade exposure in both the euro area and the United States on the days when tariffs were announced (Chart B, panel a).[11] While there has been a broader sell-off in US stocks in recent months until April, firms identified as US tariff losers have significantly underperformed.[12] Also in the subsequent equity market rally following news on pausing some of the tariffs, US tariff losers have not recovered all earlier losses. Zooming in on the high-frequency response of euro area equity prices across sectors to recent broad-based US tariff events suggests that several sectors are particularly exposed, with the automotive, consumer products, IT, industrials, materials and financial sectors experiencing sizeable declines (Chart B, panel b).[13]

Market repricing remains a prominent risk to euro area financial stability, and the potential for further US trade policy shocks to euro area markets remains high. By taking a more granular approach and examining equity prices across sectors, this box sheds light on the variation in repricing risks in the euro area. The results suggest that all euro area equity sectors are sensitive to adverse US trade policy shocks, with the automotive, consumer products, IT, industrials, materials and financial sectors considered by markets to be the most exposed. The effects do not appear to help most US companies either, as US stocks also fell after the announcements.[14] These risks are particularly relevant given the current state of economic and policy uncertainty, as shocks from US developments have become more frequent and severe. The effect of US tariffs on euro area equities has already been pronounced. If they encompass broader production and supply mechanisms – especially those integral to complex value chains – they could trigger more significant financial stability risks. Tariffs could disrupt both direct trade and production dynamics across several countries if they are extensive, lasting and aimed at sectors crucial to global production networks. Moreover, the interconnected nature of global supply chains means that the repercussions could extend far beyond the immediate targets, affecting other regions and key industries. Thus, further substantial tariff shocks could pose pronounced risks to both euro area and global financial stability.

Chart B

Varied repricing sensitivity to news on international trade policies

a) EU and US equity market developments and tariff-related news

b) Comparison of sectoral EU equity market sensitivity to tariff news on selected dates

(17 Jan.-13 May 2025, index: 17 Jan. 2025 = 100)

(percentage points)

Sources: Bloomberg Finance L.P. and ECB calculations.
Notes: The EU (US) Tariff Losers basket tracks the performance of EU (US) stocks negatively exposed to the imposition of import tariffs by the new US Administration. Stocks selected on the basis on input from UBS analysts. The EU Tariff Losers index consists of 33 stocks, primarily from the industrials (34%), consumer discretionary (31%) and consumer staples (15%) sectors. The US Tariff Losers index consists of 38 stocks, primarily from the consumer discretionary (71%) and industrials (16%) sectors. Panel a: the chart shows equity returns since one business day before the US presidential inauguration on 20 January 2025. Vertical lines for US trade threats denote the days when President Trump threatened to introduce tariffs, while vertical lines for US tariff announcements denote the days when concrete trade measures were announced. US tariff announcements refer to 1 February for tariffs against Mexico and Canada, 10 February for 25% tariffs on aluminium and steel, 4 March for an additional 10% levy on imports from China and 2 April for the announcement of global trade tariffs. The event on 8 April is not included as it was followed by announcements of the 90-day pause in tariffs within the same market opening day. US tariff threats refer to 22 January against China and the EU, 27 January against Colombia, 30 January against BRICS countries, 26 February against the EU and 13 March against EU wine and champagne. EU tariff retaliation refers to 12 March. Vertical lines show market close of business on the day before the event. Panel b: changes are calculated as the percentage difference between the price 30 minutes after the market opened on 3 April 2025 (3 February 2025 for the yellow bars) and the last closing price before President Trump’s announcements on 2 April 2025 (tariff announcement against Mexico and China on 1 February 2025 for the yellow bars).
The latest observations are for 13 May 2025 for panel a) and 09:30 on 3 April 2025 for panel b).

3 Trade policy uncertainty and banking system stability

Trade policy uncertainty can have adverse repercussions for euro area banks through financial markets and balance sheet effects. Based on bank-level data for the period between the first quarter of 2015 and the third quarter of 2024 and using panel local projections, this section assesses the impact of an increase in trade policy uncertainty on a range of key banking yardsticks.[15] More specifically, the analysis examines the effects of a surge in trade policy uncertainty on bank stock prices, credit default swap (CDS) spreads, bond spreads, the cost of risk, the return on assets and lending.[16]

An increase in trade policy uncertainty has significant adverse effects on banks’ stock prices and on some market-based metrics of bank risk. A 1 standard deviation increase in trade policy uncertainty leads to a decline in euro area bank stock prices of 1.9% on impact and 10.4% after six months.[17] It also has a statistically significant impact on bank CDS spreads, which increase by 12 basis points after six months (Chart B.3, panel a). Furthermore, bank bond spreads increase by 7 basis points after six months. These patterns are consistent with the notion that greater uncertainty fuels risk aversion on the part of investors, who demand more compensation for bearing the perceived higher risk associated with banks. The relative persistence of the market reactions suggests that investors price in the medium-term implications of trade policy uncertainty as growth weakens and, accordingly, bank asset quality and resilience deteriorate. These lasting effects of trade policy uncertainty on market valuations and banks’ funding costs may have broader implications for the ability of banks to raise capital and extend credit.

Trade policy uncertainty also leads to higher provisioning, lower profitability and a reduction in lending to the real economy. A 1 standard deviation increase in trade policy uncertainty raises banks’ cost of risk, measured as loan impairment divided by total loans, by 8 basis points after six months and 11 basis points after one year, while banks’ return on assets declines by 10 basis points after six months and 8 basis points after one year. The contraction in total bank lending to the real economy amounts to 0.6% after six months and increases to 1.9% after one year (Chart B.3, panel b).

The gradual build-up of these effects over time reflects the persistent impact of heightened uncertainty on the real economy. Trade policy uncertainty slows economic growth, weakening the ability of borrowers to repay loans and thus leading to an increase in non-performing loans and higher provisioning. Banks may also respond to trade policy uncertainty by reducing lending. Both credit supply and credit demand may be dampened by adverse macroeconomic developments caused by heightened uncertainty. On the credit supply side, greater uncertainty may make banks less willing to lend to non-financial corporations (NFCs) and households. On the credit demand side, households and NFCs may be reluctant to take out loans or invest in times of elevated uncertainty. Curtailing lending may then exacerbate the economic slowdown, further reducing borrowers’ ability to repay their loans. Higher funding costs, weaker lending and a deterioration in asset quality caused by trade tensions could undermine bank profitability. A higher cost of funding and lower loan volumes could squeeze banks’ net interest margins. Loan losses and valuation losses on asset holdings could also weigh on profitability, representing risks to bank capital accumulation and solvency. These dynamics suggest that the impact of trade policy uncertainty on the banking sector extends beyond short-term fluctuations, with potential medium-term consequences for credit supply and economic growth.[18]

Chart B.3

An increase in trade policy uncertainty has adverse effects on banks

a) Impact of a trade policy uncertainty shock on euro area bank market-based variables

b) Impact of a trade policy uncertainty shock on euro area bank balance sheet and profit and loss variables

(left-hand scale: percentages, right-hand scale: basis points)

(left-hand scale: basis points, right-hand scale: percentages)

Sources: ECB (supervisory data), Eurostat, Bloomberg Finance L.P. and Caldara et al.*.
Notes: Impulse responses across different time horizons to a 1 standard deviation trade policy uncertainty shock (Caldara et al.*), based on panel local projections exploiting quarterly data from Q1 2015 to Q3 2024. The dependent variables are logarithm of stock prices, CDS spread, bond z-spread, cost of risk, return on assets and logarithm of loans. The estimations are based on a sample of 34 banks for stocks, 40 banks for CDS spreads, 42 banks for z-spreads, 78 banks for cost of risk and return on assets, and 69 banks for logarithm of loans. The models include four lags of the dependent variable, macro-financial controls (VIX, three-month EURIBOR and EURO STOXX), country-specific macroeconomic controls (GDP growth, inflation and one-year sovereign bond yields), bank-level controls (total assets, cash/assets ratio, cost of risk, return on assets, cost/income ratio and Tier 1 capital ratio) and bank fixed effects. All values are statistically significant at levels of at least 10%, except where shaded.
*) Caldara et al., op. cit.

Higher solvency ratios mitigate some of the negative effects stemming from trade policy uncertainty, while some of the adverse repercussions are more severe for banks lending to sectors that rely on extra-EU trade. Banks with stronger capital positions are better able to absorb the adverse effects of increased trade policy uncertainty. Neither their bond spreads nor their lending is significantly affected by a trade policy uncertainty shock (Chart B.4, panel a). By contrast, banks with a larger share of lending to sectors that rely on extra-EU trade are more vulnerable to trade policy uncertainty, as they can be more materially affected by disruptions in trade flows.[19] Indeed, a higher exposure to these sectors amplifies the effect of a trade policy uncertainty shock, leading to a sharper increase in the cost of risk (i.e. to a materially greater deterioration in their asset quality) and a more pronounced decline in lending (Chart B.4, panels b and c).

Chart B.4

High levels of capitalisation cushion some negative effects of a trade policy uncertainty shock while exposure to more vulnerable sectors increases them

a) Impact depending on capitalisation

b) Impact depending on foreign input reliance

c) Impact depending on foreign market reliance

(left-hand scale: basis points,
right-hand scale: percentages)

(left-hand scale: basis points,
right-hand scale: percentages)

(left-hand scale: basis points,
right-hand scale: percentages)

Sources: ECB (supervisory data), OECD, Eurostat, Bloomberg Finance L.P. and Caldara et al.*.
Notes: Impulse responses across different time horizons to a 1 standard deviation trade policy uncertainty shock (Caldara et al.*), based on panel local projections exploiting quarterly data from Q1 2015 to Q3 2024. With regard to controls, the model specifications follow those in Chart B.3. The key difference is the inclusion of an interaction term, based on the distribution of the interaction variable. A bank is classified as highly capitalised or highly exposed in terms of foreign input reliance (FIR) or foreign market reliance (FMR) if it falls within the top quartile of the distribution. The foreign reliance indicators can be interpreted as the share of total domestic output exposed to foreign disruptions in global value chains: upstream disruptions for FIR and downstream disruptions for FMR. To calculate extra-EU foreign reliance at the bank level, the FIR/FMR of all non-EU trade counterparties are first aggregated at the NACE-2 borrower country level. The bank exposure to extra-EU trade reliance is then calculated as the sum of Σ(foreign market reliance score_i,j*NFC exposure share_i,j) where i =NACE-2 level and j=country of the borrower. Coefficients are reported according to the value of the corresponding dummy variable. Shaded areas indicate non-significance at the 10% confidence level.
*) Caldara et al., op. cit.

4 Euro area banking system exposure to sectors reliant on US trade

Sectors which are heavily reliant on US trade, and therefore most exposed to tariffs imposed by the US Administration, could face profitability and debt servicing capacity challenges. Using the foreign trade reliance indicators described above, dependence on US trade can be assessed at a sectoral level. Sectors that trade in goods associated with mining and manufacturing, as well as services such as the professional activities and technology sectors, are highly exposed to trade with the United States through both export and import dependence (Chart B.5, panel a). In particular, the profitability of sectors trading in goods subject to US tariffs could be affected by falling sales and margins as well as by rising input costs if tariffs are reciprocated. The broader economic dampening impact outlined above could exacerbate these profitability effects via lower investment and lower consumption. Additionally, affected sectors may face increased competition in domestic markets, should global trade flows be redirected in response to reduced margins on US sales.

Chart B.5

The profitability and debt servicing capacity of sectors that are more reliant on US markets and where tariffs have been applied could be negatively affected

a) Indicators of US trade reliance across sectors

b) Credit to sectors highly exposed to US trade in goods

c) Larger banks tend to have higher exposures to US markets

(2019, risk score)

(Sep. 2024, share of NFC exposures to US-exposed goods trading sectors, percentages)

(Sep. 2024, weighted bank portfolios by exposure to US exports (top graph) and imports (bottom graph), by bank size)

Sources: OECD, Eurostat and ECB (AnaCredit, SHS).
Notes: Bank data as at September 2024. Panel a: US trade reliance at NACE-1 level aggregated from country and NACE-2-level scores, weighted by net turnover of the sector. Sector classification into goods or services is determined according to whether the value of extra-EU goods or services trade is higher. The FMR and FIR indicators are obtained from the OECD and are based on the methodology developed by Baldwin and Freeman*. They identify the sectors across EU countries which are potentially most exposed to rising trade tensions with the United States. Note this methodology does not account for trade elasticities and the substitutability of goods which could also affect sectoral revenue impacts from tariffs. Panel b: high-risk sectors are those sectors above the 75th percentile scores for each indicator. The share of NFC exposures to each sector takes both loans and debt securities into consideration. “Other” consists of euro area countries that cannot be shown individually for reasons of AnaCredit confidentiality. Panel c: the exposure of bank portfolios to US trade via their lending activities is calculated as (US foreign reliance scorei,j*share NFC exposuresi,j) where i =NACE-2 level and j = country of the borrower and the foreign reliance score is either FMR (top) or FIR (bottom). Bank size is categorised by total assets size (large: >€20 billion; medium: €20 billion to €5 billion; small: <€5 billion).
*) Baldwin, R. and Freeman, R., “Risks and Global Supply Chains: What We Know and What We Need to Know”, Annual Review of Economics, Vol. 14, No 1, 2022, pp. 153-180.

Euro area banks’ exposures to sectors that are reliant on the United States as a trading partner may face rising credit risk as trade restrictions are increased. The euro area banking system’s exposure to sectors which could be impacted by US tariffs can be calculated using granular data on banks’ loans and debt security holdings.[20] Total exposures to EU sectors for which a significant share of output is reliant on either exports to or imports from US goods markets represent 9.6% of all NFC exposures (1.5% of total assets).[21] However, the share of lending at risk varies materially across euro area banking sectors and individual institutions (Chart B.5, panel b). For example, large banks allocate a greater portion of their portfolios to sectors which are reliant on trade with the United States (Chart B.5, panel c). This pattern does not hold across all countries, however, as exposures to sectors reliant on US trade in Germany and Spain tend to be larger for smaller banks. Overall, banks could face a rise in default rates and associated provisioning requirements, should the debt servicing capacity of these highly exposed sectors be negatively affected by tariff-induced profitability shocks.

Credit risk may also rise in other areas of banks’ portfolios, given the fact that there are various channels through which rising trade tensions may affect the real economy. While the direct impact of higher US tariffs on credit risk in the euro area banking system will depend on the specific measures implemented and the sectors affected, the indirect effects of materially rising trade tensions are in any case likely to be significant. The lower economic growth, build-up of financial vulnerabilities and falling investment volumes caused by rising trade tensions could contribute to a broader decline in asset quality beyond the immediate targets of the tariffs. Greater risk aversion and declining bank profitability could cause a contraction in credit. Retail portfolios could also be affected by falling consumption and higher uncertainty, especially if unemployment increased in sectors such as manufacturing and trade.[22]

5 Conclusions

The rise in trade tensions observed at the global level may not only adversely affect the real economy, it may also have an impact on financial stability. There may be repercussions associated with rising downside risks to growth over the medium term and a build-up of financial vulnerabilities. The adverse consequences for banks would include less favourable funding conditions. These effects would be accompanied by a higher cost of risk and a reduction in profitability and lending.

Policy authorities need to identify the risks stemming from trade tensions, monitor them and evaluate their potential impact on financial stability. This will enable them to identify vulnerabilities more easily, gain deeper insight into how trade tensions can affect the financial system and proactively develop potential policy responses. This proactive approach would ensure that policy reaction is rapid and coordinated, thereby bolstering the overall resilience of the financial system.

Financial institutions should also take a number of proactive steps to cope with risks stemming from trade tensions. While sound capital and liquidity buffers are the first line of defence to absorb shocks stemming from trade disruptions, financial institutions should conduct regular assessments to identify and evaluate the specific risks associated with trade tensions. They should diversify portfolios to minimise exposure to these risks and perform stress tests and scenario analyses to understand how trade tensions could impact their financial positions and operations. The results of these analyses could then be used to develop contingency plans which would make it possible to respond swiftly and effectively, should the risk materialise.

  1. For an in-depth analysis of the economic implications of trade fragmentation, see Attinasi, M.G., et al., “Navigating a fragmenting global trading system: insights for central banks”, Occasional Paper Series, No 365, ECB, 2024.

  2. See the special feature entitled “Turbulent times: geopolitical risk and its impact on euro area financial stability”, Financial Stability Review, ECB, May 2024.

  3. See Attinasi, M.G., et al., op. cit.

  4. Caldara et al. used newspaper coverage of trade-related economic uncertainty to capture shifts in policy expectations relating to trade. See Caldara, D., Iacoviello, M., Molligo, P., Prestipino, A. and Raffo, A., “The economic effects of trade policy uncertainty”, Journal of Monetary Economics, Vol. 109, 2020, pp. 38-59.

  5. For the SRI, see Lang, J.H., Izzo, C., Fahr, S. and Ruzicka, J., “Anticipating the bust: a new cyclical systemic risk indicator to assess the likelihood and severity of financial crises”, Occasional Paper Series, No 219, ECB, 2019. For the CISS, see Holló, D., Kremer, M. and Lo Duca, M., “CISS – A composite indicator of systemic stress in the financial system”, Working Paper Series, No 1426, ECB, 2012, and related data.

  6. The econometric model employed is a structural quantile vector autoregressive (QVAR) following the work done by Bochman, P., Dieckelmann, D., Fahr, S. and Ruzicka, J., “Financial stability considerations in the conduct of monetary policy”, Working Paper Series, No 2870, ECB, 2024, and Chavleishvili, M., Kremer, M. and Lund-Thomsen, F., “Quantifying financial stability trade-offs for monetary policy: a quantile VAR approach”, Working Paper Series, No 2833, ECB, 2024. The identification of shocks is based on recursive short-run restrictions akin to a Cholesky decomposition where variables are ordered as follows: trade policy uncertainty (TPU), GDP, SRI and CISS. The analysis uses the TPU as provided by the authors, i.e. without controlling for the implementation of tariffs or increased volatility. The impact of 1 standard deviation assumes that the TPU is exogenous with no feedback from the other variables in the QVAR. The model is estimated using quarterly data from the first quarter of 1990 to the third quarter of 2024. GDP enters as a quarterly (log) growth rate while the SRI, TPU and CISS enter in levels. The estimate based on the QVAR, a small-scale empirical model designed to analyse tail risks, may overestimate the effects, especially compared with larger-scale models where the trade policy uncertainty dimension can be more specifically identified.

  7. The real economy risks surrounding trade policy uncertainty were also discussed in the ECB’s March 2025 projections. See Box 2 of the ECB staff macroeconomic projections for the euro area, March 2025, published on the ECB’s website on 6 March 2025.

  8. Additional analysis using the economic policy uncertainty index developed by Baker, S., Bloom, N. and Davis, S., “Measuring Economic Policy Uncertainty”, The Quarterly Journal of Economics, Vol. 131, No 4, November 2016, pp. 1593-1636, reveals stronger effects, especially for financial variables, as it raises financial stress (CISS) in the short run. It also has a dampening effect on financial variables such as credit, house prices and stock market prices and therefore a negative effect on the SRI.

  9. Almost 12% of intermediate components that are processed and assembled into finished goods, and subsequently imported into the United States, can be traced back to production in the EU.

  10. The economic literature also suggests that announcements of the implementation of tariffs have greater and more lasting effects on US equity markets than threats about such policies. See, for example, Amiti, M., Gomez, M., Kong, S.H. and Weinstein, D., “Trade Protection, Stock-Market Returns, and Welfare”, Working Papers, No 28758, National Bureau of Economic Research, 2021.

  11. For an analysis of how increased trade tensions transmitted through social media communication affect financial markets, see Ferrari Minesso, M., Kurcz, F. and Sole Pagliari, M., “Do words hurt more than actions? The impact of trade tensions on financial markets”, Journal of Applied Econometrics, Vol. 37, Issue 6, 2022.

  12. The EU (US) Tariff Losers basket tracks the performance of EU (US) stocks negatively exposed to the imposition of import tariffs by the new US Administration. Stocks selected on the basis of input from UBS analysts. The EU Tariff Losers index consists of 33 stocks, primarily from the industrials (34%), consumer discretionary (31%) and consumer staples (15%) sectors. The US Tariff Losers index consists of 38 stocks, primarily from the consumer discretionary (71%) and industrials (16%) sectors.

  13. Chart B, panel b) focuses on the distribution of equity market responses to broad-based tariff announcements that affect a wide range of sectors. Targeted sectors have been particularly negatively affected when sector-specific tariffs have been announced. This was the case for the 10 February US announcement of tariffs on steel and aluminium imports and the 13 March threat to impose 200% tariffs on US alcoholic beverage imports from the EU. In the latter case, shares in specific European alcoholic beverage exporters fell in the 30 minutes after the threat, in some cases by around 3%, while the decline in the STOXX Europe 600 index was more muted, at around 0.3%.

  14. Similarly, in the United States automotive and IT sector stocks fell the most in the 30 minutes after the US tariff announcements.

  15. Jordà, Ò., “Estimation and Inference of Impulse Responses by Local Projections”, American Economic ReviewVol. 95, No 1, 2005, pp. 161-182.

  16. The estimates based on a panel local projection model, a small-scale empirical model, may overestimate the effects of trade policy uncertainty, especially compared with larger-scale models where the trade policy uncertainty dimension can be more specifically identified. However, it should be noted that the results of this analysis hold qualitatively, both when excluding the COVID-19 period and when accounting for it by including a time dummy for the relevant quarters.

  17. The standard deviation of the trade policy uncertainty index over the sample period considered in the analysis in this section is 53.51 index points.

  18. Bank-level results appear more pronounced compared with the macro-financial results reported in Section 2. However, this is consistent with the fact that the composite financial stability indicators employed in Section 2 also contain several other indicators beyond those investigated in the bank-level analysis. These can act as confounding factors. For example, as regards the SRI, reduced bank lending and a decline in stock market prices would lower it (ceteris paribus), but there would also be an effect on GDP (in the denominator), so the impact is not clear ex-ante.

  19. Exposure to extra-EU trade is measured using the OECD’s foreign input reliance (FIR) and foreign market reliance (FMR) indicators. These are based on the methodology developed by Baldwin, R. and Freeman, R., “Risks and Global Supply Chains: What We Know and What We Need to Know”, Annual Review of Economics, Vol. 14, No 1, 2022, pp. 153-180. Data were accessed on the OECD’s website on 20 February 2025. These indicators are provided at a country NACE-2 level and show reliance on foreign exports and imports respectively, via both direct and indirect exposure, in global value chains. Aggregate reliance on extra-EU trade at a sectoral level is calculated as the sum of reliance on all non-EU counterparties. The exposure of bank portfolios to extra-EU trade reliance via their lending activities is then calculated by weighting loan shares for each sector-country pair by the respective FMR and FIR scores.

  20. As the US Administration has focused primarily on tariffs on goods, only EU sectors which trade goods are considered (Chart B.5, panel a).

  21. Other segments of banks’ portfolios may also be exposed to increased credit risk from rising trade tensions with the United States. This includes lending to non-EU firms which trade heavily in US markets or lending to US sectors facing “reciprocal” tariffs.

  22. Future analyses should aim to quantitatively measure the contribution of both the direct and the indirect channels in driving the impact on banks stemming from rising trade tensions.