Determinants and Transmission Channels of Financial Cycle Synchronization in EU Member States

Abstract

This paper investigates the determinants and transmission channels underlying the synchronization between financial and business cycles across European Union (EU) member states. For the empirical approach, we combine frequency-domain filtering techniques with spillover index analysis to track cross-country macro-financial interlinkages. We measure financial cycle correlations and spillovers in terms of common exposures to trade linkages, overlapping systemic risk episodes, and bilateral financial claims. An important finding is that financial and business cycles tend to move together, largely due to shared macro-financial conditions and systemic stress episodes. While the data reveal strong co-movement between these cycles, the analysis does not imply a specific direction of causality. In particular, it remains possible that shifts in financial conditions can amplify or even precede business-cycle fluctuations, as seen during major crises. The focus of this study is, therefore, on the interdependence and synchronization of these cycles rather than on causal sequencing. The analysis combines complementary filtering and variance-decomposition methods to quantify the interdependencies shaping EU financial stability, providing a basis for enhanced macroprudential policy coordination. The policy implications for macroprudential authorities entail taking into account cross-border effects and spillovers when implementing instruments for taming the financial cycle.

1. Introduction

Understanding how financial and business cycles interact has become central to the analysis of macro-financial stability. Seminal studies by Claessens et al. (2011) and Borio (2014) document that financial cycles, medium-term fluctuations in credit and asset prices, often amplify business-cycle dynamics, with important implications for policy coordination. Recent EU-focused work (Adarov, 2018; Schüler et al., 2020) highlights substantial cross-country heterogeneity in these linkages. Against this background, the present paper contributes new evidence on how cyclical synchronization and spillovers operate within the European Union.

The concept of cyclical fluctuations has a long intellectual history, with early analyses of trade and business cycles by authors such as Lowe, Keynes and Hayek. Classic definitions characterize the business cycle as recurrent expansions and contractions in aggregate economic activity (Burns & Mitchell, 1946). By contrast, the financial cycle refers to medium-term boom–bust fluctuations in credit and asset markets (Claessens et al., 2011; Borio, 2014). Although interconnected through leverage and risk-taking channels, financial and business cycles can differ in amplitude and duration, trends reinforced by late-20th-century financial liberalization and globalization (Krippner, 2005). Understanding how these two cycles interact is crucial for assessing macro-financial stability in highly integrated economies.

In view of developments during the last decades, EU economies have become closely linked, both in terms of trade relations and business-cycle dynamics. During this period, stimuli from the real economy were also transmitted to the financial sector, motivating the present analysis of how the correlation between economic cycles relates to the correlation between financial cycles across countries. A first premise of the empirical approach is that potential drivers of financial-cycle correlation include (i) business cycle synchronization, (ii) overlapping systemic stress episodes (proxied by the Country-Level Index of Financial Stress, CLIFS, as estimated by the European Central Bank), (iii) financial linkages measured by the value of bilateral cross-border claims (Bank for International Settlements data), and (iv) trade linkages via bilateral trade indicators. The results of our analysis illustrate that bilateral trade flows do not generate a statistically significant impact on financial-cycle correlation, while the strongest determinant of financial-cycle synchronization is business-cycle co-movement. In other words, when the business cycles in two countries become more synchronized, there is a strong prospect of synchronization in their financial cycles. Moreover, if the financial stress indices in two countries follow a similar pattern, then the dynamics of their financial cycles tend to evolve in tandem, reinforcing the link between financial cycles and systemic risk.

Beyond confirming existing macro-financial linkages, this study contributes by providing an EU-wide, harmonized assessment of both financial-cycle correlations and spillover effects within a single analytical framework. The focus is not to provide an exhaustive account of all transmission mechanisms, but rather to offer an integrated empirical perspective on how cross-border financial and business cycle interdependencies manifest within the European Union. The remainder of the paper is structured as follows: Section 2 reviews the relevant literature and identifies the gap addressed by our research. Section 3 describes the data and methodology, including the filtering and spillover techniques. Section 4 presents the empirical results. Section 5 discusses the findings in the context of related studies and draws policy implications. Finally, Section 6 concludes, highlighting the study’s contributions, practical significance, and limitations.

2. Literature Review

The financial, business, and trade cycle literature spans over a century of economic thought, including contributions by Schumpeter, Keynes, and Lucas, among others. Early work established that real-economy business cycles—fluctuations in output, employment, and prices—are recurring but imperfectly understood phenomena (Burns & Mitchell, 1946). The concept of a distinct “financial cycle” has gained prominence more recently, referring to longer-term credit and asset price waves that can fuel financial booms and busts (Claessens et al., 2011; Borio, 2014). Critically, the rise in global financial markets and the proliferation of non-bank credit (shadow banking) since the 1980s have altered cycle dynamics (Krippner, 2005; Borio, 2014). Financial deregulation and innovation mean that credit aggregates may only partially capture systemic leverage, posing challenges for measuring financial cycles. Moreover, financial and business cycles are inextricably linked with monetary conditions: both the historic work of Friedman and Schwartz (1963) and analyses of unconventional policies (e.g., Joyce et al., 2012) show that interest-rate and liquidity conditions can shape the amplitude of cyclical fluctuations.

In light of increasing economic interdependence, the idea of cross-country spillovers between financial cycles has gained interest from multiple research perspectives. Many studies examine “spillover effects” across various macro-financial domains using a wide array of variables. Another stream of research focuses on globalization and the emergence of global value chains, for example, trade in intermediate inputs, as a determinant of business-cycle synchronization. Financial integration is also an important driver of business-cycle co-movements: countries with restricted capital mobility tend to exhibit lower synchronization of economic activity than financially integrated countries. Despite growing interest in financial cycle synchronization, existing studies often focus on individual countries or global aggregates, offering limited insights into cross-country dynamics within Europe. Moreover, relatively little evidence is available on the direction and magnitude of financial spillovers across EU member states, leaving room to investigate the transmission mechanisms underlying these linkages.

Adarov (2018) provides one broad analysis of financial cycles, examining 34 advanced and developing economies over 1960–2015. Using both dynamic filters and state-space models, Adarov estimates country-level financial cycles in credit, housing, bond, and equity markets and constructs aggregate financial cycle indicators. The study finds that financial market activity exhibits a high degree of persistence and recurrence associated with the buildup and unwinding of imbalances. Financial cycles tend to fluctuate at much lower frequencies (9–15 years on average) than traditional business cycles, and the author documents notable intra-regional synchronization of financial cycles, including significant co-movement among European, American, and Asian financial cycles.

Another study on financial cycle synchronization appears in the Deutsche Bundesbank (2019) monthly report, which applies frequency-based (wavelet) analysis to selected euro-area countries. The variables examined, credit aggregates and house prices, serve as proxies for national financial cycles. The primary goal was to assess cross-country synchronization of financial cycles, particularly whether credit and residential real estate prices exhibit common cyclical patterns. The analysis finds that increases in lending, house prices, and real GDP tend to produce similar medium-term fluctuations across the countries studied. This implies that real (business) cycles and financial cycles should not be viewed as independent phenomena, but rather as interdependent. The Bundesbank’s empirical results show significant correlations between key financial and real variables both within and across countries, though no conclusions could be drawn regarding causality.

Borowski et al. (2020) investigate a specific structural channel, foreign bank ownership, and its effect on business-cycle synchronization between the euro area and other EU countries over 1998–2016. They find that economies with a higher share of foreign-owned bank assets are more synchronized with the euro-area cycle, mainly due to an alignment of investment cycles. The interpretation is that multinational banks (originating largely from the EU) treat host countries as extensions of their home market, standardizing lending behavior and transmitting shocks. Similarly, Beck (2021) studies how capital mobility relates to euro-area business-cycle synchronization, examining four channels: short-run business-cycle characteristics, exposure to common risks, foreign direct investment linkages, and contagion. The results show that greater capital mobility is associated with closer business-cycle synchronization in the long run within the EU. This aligns with the view that financial openness can lead to more correlated economic fluctuations.

From the spillovers perspective, various papers have analyzed the international propagation of macro-financial shocks. For example, Abosedra et al. (2020) study GDP growth volatility spillovers across 120 countries during 1960–2017, using a Diebold–Yilmaz variance-decomposition framework. They find that high-income countries are net transmitters of growth volatility, whereas low-income countries are net recipients, and that the latter take longer to absorb global shocks. Fang et al. (2021) investigate spillovers in stock, bond, and FX markets between China and G7 economies (2000–2018), finding that financial spillovers significantly drive asset price variation and that spillovers from G7 to China are larger than spillbacks from China to G7. Gomez-Pineda (2020) shows that global uncertainty and risk aversion are major volatility factors for both financial and business cycles worldwide. Škrinjarić and Orlović (2020) focus on economic policy uncertainty shocks, estimating spillover effects on stock market returns and volatility in several Central and Eastern European (CEE) markets. They note that uncertainty-driven spillovers are especially strong for certain CEE countries (e.g., Czechia, Lithuania, Slovenia, Poland), implying that developing markets’ asset prices react more sensitively to global uncertainty. Similarly, Kubinschi and Barnea (2019) examine industrial output spillovers in Europe and find that after 2009–2011, domestic industrial indices explained less of their own variance while cross-country spillovers gained importance, highlighting increasing interconnectedness. Gammadigbe (2022) looks at financial cycle synchronization in the West African Economic and Monetary Union, finding, contrary to mainstream literature, a lack of convergence in financial cycles; this suggests a one-size-fits-all macroprudential policy may be inappropriate even in a monetary union. Chari et al. (2022) show that a tighter ex ante macroprudential policy stance can amplify the impact of global risk shocks on portfolio flows in emerging markets, an important spillover at the “extremes” of global financial cycles. Their policy-shock approach (using high-frequency identification and a new macroprudential intensity index) finds that stricter macroprudential regulation reduces the volume and volatility of bank flows, but may push vulnerabilities into other parts of the financial system. Finally, Li et al. (2024) introduce an innovative approach to measuring and nowcasting financial cycles by incorporating textual sentiment data into a mixed-frequency dynamic factor model, improving predictive power. This array of literature underscores the multifaceted nature of business and financial cycle interactions, as well as the influence of global factors and policies.

For euro-area members, financial integration is also mirrored in the functioning of the TARGET2 system, which records settlement balances between national central banks within the Eurosystem. TARGET2 itself is not a policy instrument but an accounting mechanism that reflects cross-border liquidity movements arising from payment transactions. These balances highlight capital-flow dynamics within the monetary union but do not restrict convertibility or directly influence national monetary sovereignty—limitations that stem instead from the EU Treaties that transferred monetary powers to EU institutions. Understanding the evolution of TARGET2 positions helps contextualize intra-euro-area financial linkages and cyclical co-movements, particularly during stress periods.

Notably, historical analyses of financial crises (Reinhart & Rogoff, 2009) and of debt-induced recessions (Mian & Sufi, 2014) underscore the importance of monitoring financial cycles. Credit booms that coincide across countries can precede widespread crises (“this time is different” syndrome), reinforcing calls for coordinated macroprudential oversight. Altogether, the literature suggests that business and financial cycles are deeply intertwined through real and financial channels, and that both synchronization and spillover effects merit joint examination—especially in an economically integrated region like the EU. However, previous research has typically examined these mechanisms separately. Thus, a gap remains in our understanding of how business-cycle co-movements, financial stress overlaps, and cross-border linkages jointly determine the alignment of financial cycles in Europe. Building on the above insights, this study addresses the following research questions: To what extent do business-cycle co-movements and shared systemic stress episodes explain financial-cycle synchronization across EU countries? Which channels are cross-country financial cycle spillovers transmitted?

By empirically investigating these questions, our analysis aims to bridge the literature on cycle synchronization and that on spillovers, providing an integrated perspective on the determinants of financial cycle correlations in the EU.

3. Methodology

The main goal of the paper is to assess how various factors contribute to correlations between the financial cycles of selected EU countries. We focus on 20 EU member states (both euro-area and non-euro) based on data availability. The key variables included in the analysis are (i) the bilateral correlation of economic (business) cycles, (ii) the bilateral correlation of systemic financial stress indices, (iii) the value of bilateral financial claims between states, and (iv) a measure of bilateral trade links (exports).

For the empirical approach, we employ two complementary methods to extract financial cycle components: the Hodrick–Prescott (HP) filter and the Baxter–King (BP) band-pass filter. These filtering techniques are widely used to isolate cyclical fluctuations. The HP filter (using λ = 1600 for quarterly data) isolates relatively short-run fluctuations, whereas the BP filter (configured to pass 8- to 20-year cycles) captures the medium-term oscillations characteristic of financial cycles (Drehmann et al., 2012). Using both filters allows us to verify that our findings are not driven by the choice of cycle frequency. In our application, the BP filter eliminates both high- and low-frequency components of the time series, helping to isolate the financial-cycle component given uncertainty about the exact cycle length in the literature. We also use the traditional HP filter for business cycles (with the standard smoothing parameter λ = 1600), following the convention that the business cycle lasts roughly up to 8 years. In line with BIS and ESRB recommendations, we additionally compute the credit-to-GDP gap (deviation of the ratio of nominal credit to nominal GDP from trend) as an alternative financial cycle proxy for robustness, in the same way as (Alupoaiei et al., 2020).

The value of the smoothing parameter was established by Hodrick and Prescott (1981) for estimating the trend in the case of the business cycle, which lasts between 1 year and 8 years. From an economic perspective, the value used for smoothing parameter implies that the cyclical evolutions with a length of over 8 years are attributed to the trend component.

The HP filter yields a smooth trend in credit and asset price series, with the deviation capturing cyclical booms and busts. The Baxter–King filter, by focusing on a specified band of cycle lengths, provides a targeted view of medium-term financial fluctuations. Together, these methods ensure a robust identification of cycles across different frequency ranges.

After extracting financial and business cycle components for all countries, we compute correlation matrices for: financial cycle indicators, business cycle indicators, and the CLIFS financial stress indices. We then assess cross-country spillovers using the framework developed by Diebold and Yilmaz. In particular, we apply the Diebold and Yilmaz (2012) generalized forecast error variance decomposition approach, a spillover index (DYI), to quantify how shocks to one country’s cycle propagate to others. This involves estimating a vector autoregression (VAR) model for each set of variables (e.g., financial cycle indexes of all countries) and computing the fraction of each variable’s forecast error variance attributable to shocks from other countries. We use the Generalized Forecast Error Variance Decomposition (GFEVD) approach of Koop et al. (1996) and Pesaran and Shin (1998), which is invariant to variable ordering. In practical terms, the spillover index measures the share of cross-country variation due to spillovers as opposed to idiosyncratic (domestic) factors. We implement the Diebold–Yilmaz spillover analysis for financial cycles, business cycles, and financial stress indices, respectively.

It is worth noting some specifics of our regression framework. In the correlation analysis, we construct two sets of cross-sectional observations: (1) bilateral correlations (financial cycle correlations as the dependent variable) paired with corresponding bilateral business-cycle correlations, bilateral CLIFS correlations, and bilateral trade intensity measures; and (2) spillover indices (from the Diebold–Yilmaz analysis) as the dependent variable, paired with the analogous business-cycle spillover index, CLIFS spillover index, plus bilateral trade intensity and bilateral financial claims measures. These two cross-sectional regression models allow us to assess which factors significantly explain the strength of financial cycle linkages between country pairs. We emphasize that in the correlation case, the correlation matrices are symmetric (so we use the upper triangular elements only to avoid duplication), whereas in the spillover case, the directionality matters—we consider spillovers from i to j and from j to i separately, using appropriate explanatory variables for each direction. All regressions use robust standard errors to account for heteroskedasticity.

Before conducting the regressions, we examine the time-series properties of our data. Augmented Dickey–Fuller (ADF) and KPSS tests confirm that the filtered cycle components are stationary for all countries, which is consistent with the purpose of detrending. We acknowledge that testing for heteroskedasticity on filtered rather than raw data can limit the detection of shocks like the Global Financial Crisis; however, this approach ensures consistency across variables that have undergone the same filtering procedure. The robustness of the results was verified by comparing both HP- and BP-filtered series, with no material differences in statistical behavior. Finally, we note that our correlation and spillover measures capture associations rather than causation—they describe interconnectedness, not causal direction. We return to this point in the discussion and conclusion, suggesting avenues for future causal analysis.

The data employed in this study were gathered from multiple sources to ensure consistency across countries. Data were collected from the European Central Bank (ECB) Statistical Data Warehouse, the Bank for International Settlements (BIS) credit statistics, Eurostat, and national central banks. The Composite Indicator of Systemic Stress (CLIFS; Duprey et al., 2017) is used as a proxy for systemic financial stress, as it captures co-movements in credit spreads, volatility, and banking-sector risk premia. The sample period begins in 2004, thereby including the EU’s eastern enlargement wave, and ends in 2021 (incorporating the post-Global Financial Crisis and post-EU sovereign debt crisis years, as well as the initial pandemic shock). We chose 2004 as the start date to maximize data availability for newer EU members. The pandemic period of 2020–2021, while included in the baseline sample, was also analyzed in a sensitivity test by truncating those years. Euro-area countries (which share a common currency and the TARGET2 settlement system) are contrasted with non-euro EU members (which have independent monetary and exchange-rate policies) in our interpretation of results to account for structural differences in monetary regimes. The following countries were analyzed: Austria, Belgium, Bulgaria, Czech Republic, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Lithuania, Luxembourg, The Netherlands, Poland, Portugal, Romania, Slovenia, Spain, and Sweden. All series are quarterly except the CLIFS index (monthly, averaged to quarterly for analysis) and are expressed in percentage or index form as appropriate. The sample’s coverage across these 20 countries and 2004–2021 yields a balanced panel for the analyses.

4. Results

A first step in the analysis is to examine the pairwise correlations between the financial cycles of the EU member states in our sample. The results show that the highest financial-cycle correlations are generally observed among core euro-area countries. For instance, Germany’s financial cycle is very strongly correlated with Austria’s (the pair with the highest correlation), and high correlations are also evident between France and the The Netherlands, and between Finland and Austria, among others (see Table 1). By contrast, some country pairs exhibit negative financial-cycle correlations—for example, Hungary and Luxembourg show one of the most negative correlations, indicating that their financial systems often move out of phase. In general, a high positive correlation between two countries’ financial cycles indicates that their credit and asset markets tend to expand and contract together, reflecting similar leverage dynamics and asset price movements. A negative correlation suggests opposing cycle phases, which could occur if, say, one country’s credit boom coincides with another’s deleveraging phase. Overall, the highest financial-cycle correlations are observed among euro-area members (notably between large integrated economies like France, Germany, and The Netherlands), whereas correlations tend to be lower among Central and Eastern European (CEE) countries. This pattern indicates that financial-cycle synchronization remains stronger within the monetary union.

Table 1. Correlation matrix of financial cycles (HP-filtered).

	Austria	Belgium	Bulgaria	Czech Rep.	Finland	France	Germany	Greece	Hungary	Ireland	Italy	Lithuania	Luxembourg	The Netherlands	Poland	Portugal	Romania	Slovenia	Spain	Sweden
Austria		18%	17%	36%	82%	66%	85%	68%	21%	63%	21%	63%	20%	2%	46%	30%	73%	38%	56%	42%
Belgium	18%		32%	48%	6%	41%	25%	59%	−3%	21%	7%	42%	43%	49%	3%	36%	11%	38%	33%	8%
Bulgaria	17%	32%		53%	24%	20%	36%	42%	−29%	51%	50%	49%	23%	51%	5%	57%	29%	34%	39%	47%
Czech Rep.	36%	48%	53%		48%	50%	48%	59%	−12%	53%	38%	57%	44%	51%	33%	68%	47%	59%	69%	50%
Finland	82%	6%	24%	48%		55%	78%	64%	10%	59%	17%	69%	15%	16%	55%	49%	76%	34%	63%	57%
France	66%	41%	20%	50%	55%		49%	79%	27%	60%	29%	64%	20%	10%	46%	25%	82%	72%	50%	53%
Germany	85%	25%	36%	48%	78%	49%		57%	−11%	63%	24%	61%	36%	14%	41%	51%	57%	37%	65%	35%
Greece	68%	59%	42%	59%	64%	79%	57%		25%	59%	35%	82%	22%	39%	50%	46%	72%	54%	54%	62%
Hungary	21%	−3%	−29%	−12%	10%	27%	−11%	25%		1%	−12%	23%	−37%	−3%	1%	−19%	29%	−4%	−7%	5%
Ireland	63%	21%	51%	53%	59%	60%	63%	59%	1%		57%	55%	32%	24%	22%	49%	65%	63%	58%	64%
Italy	21%	7%	50%	38%	17%	29%	24%	35%	−12%	57%		16%	22%	25%	18%	22%	24%	41%	35%	63%
Lithuania	63%	42%	49%	57%	69%	64%	61%	82%	23%	55%	16%		11%	47%	37%	57%	74%	43%	46%	53%
Luxembourg	20%	43%	23%	44%	15%	20%	36%	22%	−37%	32%	22%	11%		40%	−10%	29%	−2%	51%	42%	16%
The Netherlands	2%	49%	51%	51%	16%	10%	14%	39%	−3%	24%	25%	47%	40%		−10%	49%	9%	27%	22%	37%
Poland	46%	3%	5%	33%	55%	46%	41%	50%	1%	22%	18%	37%	−10%	−10%		23%	52%	11%	42%	42%
Portugal	30%	36%	57%	68%	49%	25%	51%	46%	−19%	49%	22%	57%	29%	49%	23%		29%	38%	67%	43%
Romania	73%	11%	29%	47%	76%	82%	57%	72%	29%	65%	24%	74%	−2%	9%	52%	29%		51%	45%	57%
Slovenia	38%	38%	34%	59%	34%	72%	37%	54%	−4%	63%	41%	43%	51%	27%	11%	38%	51%		51%	53%
Spain	56%	33%	39%	69%	63%	50%	65%	54%	−7%	58%	35%	46%	42%	22%	42%	67%	45%	51%		47%
Sweden	42%	8%	47%	50%	57%	53%	35%	62%	5%	64%	63%	53%	16%	37%	42%	43%	57%	53%	47%

Note: color shading reflects correlation magnitude. Source: author’s calculations based on data provided by ECB.

Overall, financial cycles are most synchronized among countries that share the euro and exhibit deep financial integration, while non-euro and newer EU members display more diverse cyclical patterns. This pattern reflects the broader trajectory of economic and political integration within the Union since its founding in 1958, as the original members have progressively deepened both capital market and monetary interconnections. These findings also carry implications for the EU’s economic convergence agenda under the Maastricht Treaty: higher synchronization among the core economies suggests greater resilience within long-standing members, whereas weaker alignment in newer members may call for a more gradual, multi-speed path toward deeper monetary integration. Such differences could inform discussions on the prudence of future euro-area enlargements and the coordination of macro-financial policies across heterogeneous economies.

Turning to the business cycles, a similar correlation analysis is performed for real GDP cycles (business cycles) of the same countries. We find that many EU economies also exhibit high business-cycle correlations, reflecting the influence of EU-wide economic integration and common shocks. For example, the business cycles of Germany and Austria, or of Sweden and Poland, are strongly correlated (see Table 2 for the business-cycle correlation matrix). A high correlation between two countries’ business cycles implies that their economic activities (output, demand, etc.) tend to rise and fall together over time, often due to tight trade links, aligned policy frameworks, or shared exposure to global conditions. Overall, business-cycle correlations are also elevated within Europe, underlining that EU economies frequently experience synchronous booms and recessions. Even some non-euro countries (e.g., Sweden, Poland) show substantial business-cycle synchronization with euro-area neighbors, likely due to deep trade and investment ties.

Table 2. Correlation matrix of business cycles (output gaps, HP-filtered).

	Austria	Belgium	Bulgaria	Czech Rep.	Finland	France	Germany	Greece	Hungary	Ireland	Italy	Lithuania	Luxembourg	The Netherlands	Poland	Portugal	Romania	Slovenia	Spain	Sweden
Austria		92%	77%	85%	92%	86%	81%	92%	52%	88%	36%	93%	60%	65%	94%	85%	85%	68%	82%	87%
Belgium	92%		67%	80%	87%	90%	72%	96%	57%	88%	42%	96%	49%	73%	89%	83%	88%	59%	83%	83%
Bulgaria	77%	67%		89%	74%	79%	80%	63%	62%	78%	42%	69%	70%	57%	85%	73%	69%	82%	67%	87%
Czech Rep.	85%	80%	89%		86%	90%	83%	76%	64%	89%	55%	84%	64%	66%	93%	82%	83%	74%	84%	93%
Finland	92%	87%	74%	86%		79%	90%	86%	38%	88%	38%	90%	71%	71%	90%	74%	76%	61%	91%	87%
France	86%	90%	79%	90%	79%		67%	88%	71%	87%	45%	92%	39%	63%	91%	88%	95%	69%	75%	86%
Germany	81%	72%	80%	83%	90%	67%		68%	36%	74%	39%	76%	81%	72%	85%	66%	63%	63%	84%	87%
Greece	92%	96%	63%	76%	86%	88%	68%		51%	85%	33%	95%	44%	66%	86%	80%	86%	58%	76%	81%
Hungary	52%	57%	62%	64%	38%	71%	36%	51%		63%	43%	53%	32%	54%	66%	55%	65%	77%	37%	64%
Ireland	88%	88%	78%	89%	88%	87%	74%	85%	63%		46%	86%	64%	67%	89%	79%	82%	73%	83%	88%
Italy	36%	42%	42%	55%	38%	45%	39%	33%	43%	46%		43%	35%	49%	48%	37%	36%	30%	59%	48%
Lithuania	93%	96%	69%	84%	90%	92%	76%	95%	53%	86%	43%		45%	68%	92%	83%	91%	56%	85%	86%
Luxembourg	60%	49%	70%	64%	71%	39%	81%	44%	32%	64%	35%	45%		67%	63%	36%	28%	67%	62%	71%
The Netherlands	65%	73%	57%	66%	71%	63%	72%	66%	54%	67%	49%	68%	67%		68%	50%	53%	57%	73%	73%
Poland	94%	89%	85%	93%	90%	91%	85%	86%	66%	89%	48%	92%	63%	68%		85%	89%	76%	83%	94%
Portugal	85%	83%	73%	82%	74%	88%	66%	80%	55%	79%	37%	83%	36%	50%	85%		88%	60%	71%	78%
Romania	85%	88%	69%	83%	76%	95%	63%	86%	65%	82%	36%	91%	28%	53%	89%	88%		59%	71%	81%
Slovenia	68%	59%	82%	74%	61%	69%	63%	58%	77%	73%	30%	56%	67%	57%	76%	60%	59%		45%	79%
Spain	82%	83%	67%	84%	91%	75%	84%	76%	37%	83%	59%	85%	62%	73%	83%	71%	71%	45%		82%
Sweden	87%	83%	87%	93%	87%	86%	87%	81%	64%	88%	48%	86%	71%	73%	94%	78%	81%	79%	82%

Note: color shading reflects correlation magnitude. Source: author’s calculations based on data provided by Eurostat.

High business-cycle correlations often reflect structural factors such as strong trade and financial integration, as well as coordinated fiscal and monetary responses within the euro area. Taken together with the preceding results, this contrast highlights an important duality: while the EU’s Single Market has succeeded in synchronizing real economic activity across member states, financial cycle integration remains more uneven. This distinction suggests that the Single Market continues to function effectively even under episodes of global stress—such as the 2008–2009 and 2011 crises—whereas the monetary union’s cohesion depends more strongly on financial-sector depth and convergence. These patterns could help explain recurring bond market tensions (e.g., Italy 2011, France 2025) and provide guidance for managing future integration phases. This synchronization of economic activity provides context for interpreting financial-cycle co-movements, since synchronized business cycles can be a key driver of financial cycle alignment (as posited in RQ1).

Next, we consider cross-country spillover effects for financial cycles, as measured by the Diebold–Yilmaz spillover index. Table 3 presents the financial-cycle spillover matrix, where each entry (i, j) indicates the percentage of country i’s financial cycle variance attributable to shocks from country j (for i ≠ j). A larger off-diagonal value means country j’s financial cycle has a strong influence on country i’s financial cycle. Our findings reveal a core-periphery structure in spillovers: for many smaller or emerging EU economies, a significant portion of financial-cycle fluctuations can be traced to shocks emanating from major EU countries. For instance, Romania’s financial cycle is influenced to a large extent by financial-cycle developments in Austria, France, and Greece, where many parent banks of Romanian banks are headquartered. Similarly, Austria and Germany appear as dominant transmitters of financial-cycle shocks to several CEE countries. Notably, advanced euro-area economies (e.g., Germany, France) act as major sources of financial-cycle spillovers to smaller or emerging members, whereas the latter are more often on the receiving end of spillovers. This pattern reflects the transmission of credit conditions and banking-sector shocks through cross-border financial linkages (such as foreign bank ownership and lending).

Table 3. Financial cycle spillover index (% of variance in row country’s financial cycle coming from column country).

	Austria	Belgium	Bulgaria	Czech Rep.	Finland	France	Germany	Greece	Hungary	Ireland	Italy	Lithuania	Luxembourg	The Netherlands	Poland	Portugal	Romania	Slovenia	Spain	Sweden
Austria	12%	6%	0%	5%	9%	8%	10%	7%	4%	6%	0%	7%	7%	3%	4%	5%	6%	6%	4%	5%
Belgium	5%	16%	1%	5%	6%	6%	5%	7%	1%	3%	0%	6%	5%	7%	5%	2%	5%	4%	3%	5%
Bulgaria	0%	0%	42%	2%	0%	1%	0%	1%	3%	2%	6%	1%	1%	1%	0%	1%	1%	1%	1%	3%
Czech Rep.	6%	4%	4%	15%	3%	4%	3%	4%	4%	6%	6%	3%	6%	5%	3%	7%	4%	5%	6%	3%
Finland	7%	6%	1%	6%	12%	7%	8%	6%	1%	5%	0%	6%	6%	5%	7%	6%	8%	5%	6%	7%
France	8%	8%	2%	7%	9%	12%	8%	9%	1%	5%	1%	9%	5%	4%	7%	6%	10%	9%	7%	6%
Germany	9%	4%	0%	4%	8%	5%	18%	4%	0%	4%	3%	4%	10%	2%	6%	7%	4%	4%	7%	3%
Greece	9%	11%	6%	6%	9%	11%	8%	15%	3%	5%	2%	11%	4%	6%	11%	4%	11%	8%	5%	10%
Hungary	1%	1%	6%	0%	0%	1%	0%	0%	55%	0%	3%	1%	0%	1%	2%	1%	1%	1%	1%	0%
Ireland	4%	3%	1%	5%	3%	3%	3%	2%	5%	21%	2%	2%	1%	0%	2%	6%	3%	2%	5%	1%
Italy	0%	0%	7%	2%	0%	0%	1%	1%	4%	1%	55%	0%	2%	0%	0%	0%	0%	2%	1%	0%
Lithuania	7%	8%	4%	5%	8%	8%	7%	10%	4%	4%	0%	17%	3%	8%	6%	4%	10%	7%	4%	7%
Luxembourg	2%	2%	1%	3%	2%	2%	5%	1%	2%	2%	4%	1%	28%	6%	0%	1%	2%	4%	2%	1%
The Netherlands	3%	4%	0%	3%	4%	2%	2%	3%	2%	1%	1%	5%	6%	29%	1%	1%	4%	4%	1%	6%
Poland	3%	3%	2%	3%	3%	3%	3%	5%	2%	2%	1%	3%	0%	1%	24%	5%	4%	0%	3%	4%
Portugal	4%	3%	2%	5%	3%	3%	4%	2%	1%	6%	1%	2%	1%	2%	5%	21%	2%	3%	8%	2%
Romania	6%	7%	5%	7%	9%	9%	5%	8%	2%	6%	1%	9%	5%	6%	6%	4%	13%	6%	4%	6%
Slovenia	3%	2%	1%	6%	2%	4%	2%	2%	1%	3%	7%	2%	5%	5%	1%	4%	2%	17%	5%	5%
Spain	4%	5%	1%	5%	3%	4%	3%	3%	3%	8%	2%	1%	1%	1%	3%	10%	3%	4%	20%	1%
Sweden	4%	5%	11%	5%	6%	5%	3%	7%	2%	4%	4%	7%	2%	7%	5%	4%	6%	6%	4%	24%

Note: color shading reflects correlation magnitude. Source: author’s calculations based on data provided by Eurostat.

A similar spillover analysis for business cycles is summarized in Table 4, which displays the spillover matrix for real GDP cycles. The results indicate that some smaller economies’ business cycles are heavily influenced by larger neighbors. For instance, it emerges that Lithuania’s business cycle exerts a surprisingly strong influence on those of Poland and Finland—possibly due to regional supply-chain links or synchronized demand shocks in the Baltics and Northern Europe. In general, however, the business-cycle spillover indices tend to be more evenly distributed compared to financial-cycle spillovers, reflecting that real-economy shocks (like global demand or commodity price changes) are often widely shared across countries. The presence of cases where one country (even a smaller one) influences another’s business cycle more than the latter’s own internal factors might point to specialized regional relationships or sectoral interdependencies. Overall, cross-country business-cycle spillovers confirm that EU economies are tightly interconnected, though the pattern can vary by region (e.g., smaller open economies can transmit shocks to larger ones if sectoral links are significant).

Table 4. Business cycle spillover index (% of variance in row country’s GDP cycle due to column country).

	Austria	Belgium	Bulgaria	Czech Rep.	Finland	France	Germany	Greece	Hungary	Ireland	Italy	Lithuania	Luxembourg	The Netherlands	Poland	Portugal	Romania	Slovenia	Spain	Sweden
Austria	8%	6%	5%	5%	6%	6%	4%	7%	5%	6%	6%	7%	3%	3%	6%	7%	7%	4%	6%	5%
Belgium	7%	7%	6%	6%	6%	7%	5%	7%	6%	6%	5%	7%	4%	5%	6%	7%	7%	5%	6%	6%
Bulgaria	4%	4%	9%	4%	3%	4%	4%	4%	3%	4%	2%	4%	4%	3%	4%	4%	4%	4%	3%	4%
Czech Rep.	6%	6%	7%	9%	6%	7%	5%	6%	6%	7%	6%	6%	4%	3%	6%	6%	7%	6%	7%	7%
Finland	5%	6%	5%	5%	7%	5%	5%	5%	4%	5%	2%	6%	4%	4%	5%	4%	6%	5%	5%	5%
France	6%	7%	6%	6%	6%	7%	3%	7%	5%	7%	4%	7%	2%	3%	6%	6%	7%	5%	6%	6%
Germany	3%	2%	3%	3%	4%	2%	14%	2%	2%	2%	1%	3%	8%	5%	3%	2%	2%	4%	3%	3%
Greece	7%	8%	7%	7%	7%	8%	5%	8%	7%	7%	5%	8%	3%	5%	7%	7%	8%	6%	7%	6%
Hungary	4%	4%	3%	4%	4%	4%	4%	4%	13%	4%	3%	4%	6%	8%	5%	3%	3%	7%	4%	5%
Ireland	5%	5%	5%	5%	5%	5%	2%	5%	4%	7%	4%	5%	3%	2%	5%	5%	5%	5%	5%	5%
Italy	2%	2%	1%	3%	1%	2%	1%	2%	2%	2%	27%	2%	1%	0%	2%	3%	2%	1%	3%	2%
Lithuania	8%	8%	6%	7%	8%	8%	6%	8%	7%	7%	5%	8%	4%	6%	8%	7%	8%	6%	8%	7%
Luxembourg	1%	1%	2%	1%	1%	1%	5%	1%	2%	1%	0%	1%	17%	5%	1%	0%	0%	4%	1%	2%
The Netherlands	4%	4%	4%	4%	5%	3%	13%	4%	7%	3%	1%	4%	12%	27%	5%	3%	3%	6%	4%	6%
Poland	6%	5%	5%	6%	6%	6%	4%	6%	4%	6%	5%	6%	3%	2%	7%	6%	6%	5%	5%	5%
Portugal	4%	4%	2%	3%	3%	3%	1%	4%	2%	3%	7%	3%	1%	2%	3%	10%	4%	2%	3%	3%
Romania	6%	6%	5%	6%	5%	6%	3%	6%	3%	6%	4%	6%	1%	1%	5%	6%	8%	3%	5%	5%
Slovenia	3%	3%	4%	3%	3%	3%	4%	3%	6%	4%	2%	3%	6%	4%	3%	3%	2%	10%	3%	4%
Spain	5%	5%	6%	6%	6%	6%	6%	5%	5%	6%	7%	6%	4%	3%	6%	5%	5%	5%	10%	6%
Sweden	6%	7%	9%	7%	7%	7%	7%	6%	8%	7%	4%	6%	9%	7%	7%	6%	6%	8%	7%	10%

Note: color shading reflects correlation magnitude. Source: author’s calculations based on data provided by Eurostat.

Finally, we examine spillovers of systemic financial stress using the CLIFS indices. Table 5 presents the spillover matrix for CLIFS, indicating how financial stress shocks transmit among countries. We observe that certain countries experience nearly all of their financial stress domestically (high own variance shares), whereas others are more subject to external stress events. Notably, Sweden, Italy, Slovenia, and Ireland emerge as countries where domestic financial stress is largely driven by common international shocks (reflected by high incoming spillovers in Table 5). On the other hand, countries like Austria, Lithuania, Greece, and Germany have relatively low external contributions—meaning their CLIFS dynamics are more dominated by local conditions or idiosyncratic factors. In the case of Romania, we find that, beyond its own internal stress factors, the strongest external influences on Romanian financial stress come from a few key countries (e.g., France or Austria, in line with the earlier financial-cycle spillovers). Overall, the CLIFS spillover analysis suggests that systemic stress events (such as the 2008–2009 crisis or the euro debt crisis) propagate widely across EU countries, but the degree of propagation varies—core banking hubs often export stress, whereas smaller or deeply connected economies import stress. This highlights an important asymmetry: systemic financial stress tends to be a region-wide phenomenon, yet some core countries’ conditions weigh disproportionately on others’ financial stability.

Table 5. Spillover matrix for systemic stress (CLIFS) indices.

	Austria	Belgium	Bulgaria	Czech Rep.	Finland	France	Germany	Greece	Hungary	Ireland	Italy	Lithuania	Luxembourg	The Netherlands	Poland	Portugal	Romania	Slovenia	Spain	Sweden
Austria	10%	5%	2%	2%	4%	5%	5%	5%	7%	2%	4%	6%	3%	3%	7%	3%	4%	4%	1%	5%
Belgium	8%	13%	6%	7%	6%	7%	5%	7%	5%	8%	6%	6%	3%	9%	5%	7%	5%	7%	4%	7%
Bulgaria	1%	2%	16%	4%	4%	1%	2%	2%	1%	3%	1%	1%	3%	3%	1%	2%	0%	4%	5%	1%
Czech Rep.	4%	6%	8%	16%	8%	3%	5%	5%	1%	9%	7%	4%	9%	6%	4%	7%	5%	5%	12%	3%
Finland	7%	7%	10%	11%	16%	5%	5%	8%	3%	10%	6%	6%	9%	6%	6%	7%	3%	7%	13%	2%
France	6%	6%	5%	5%	5%	14%	7%	6%	6%	5%	6%	9%	3%	6%	6%	4%	7%	6%	4%	7%
Germany	4%	4%	3%	3%	3%	4%	13%	4%	3%	1%	4%	5%	4%	5%	6%	3%	5%	2%	2%	3%
Greece	9%	8%	7%	6%	8%	7%	6%	12%	5%	6%	7%	7%	6%	5%	8%	7%	7%	8%	6%	6%
Hungary	4%	3%	1%	1%	2%	5%	6%	2%	34%	1%	2%	4%	1%	3%	3%	2%	6%	6%	0%	6%
Ireland	4%	3%	5%	7%	7%	2%	1%	4%	1%	21%	1%	3%	5%	3%	2%	6%	2%	3%	9%	2%
Italy	6%	6%	8%	6%	6%	7%	6%	6%	3%	4%	22%	7%	4%	6%	7%	4%	6%	4%	5%	4%
Lithuania	7%	6%	6%	5%	6%	8%	6%	7%	5%	5%	6%	12%	4%	6%	8%	4%	6%	5%	4%	5%
Luxembourg	2%	2%	2%	2%	3%	1%	5%	3%	1%	1%	2%	2%	19%	1%	4%	3%	3%	2%	3%	1%
The Netherlands	3%	5%	3%	3%	2%	4%	5%	3%	5%	3%	3%	4%	3%	16%	2%	5%	3%	2%	2%	3%
Poland	7%	6%	5%	5%	6%	5%	7%	6%	4%	4%	5%	7%	5%	3%	15%	3%	5%	3%	3%	3%
Portugal	5%	5%	3%	4%	4%	3%	3%	5%	4%	9%	3%	3%	4%	4%	3%	20%	4%	3%	4%	5%
Romania	6%	6%	3%	5%	4%	9%	7%	7%	5%	4%	8%	8%	5%	6%	8%	4%	19%	4%	3%	7%
Slovenia	3%	3%	5%	4%	4%	2%	3%	3%	3%	2%	2%	2%	3%	2%	2%	1%	2%	21%	4%	1%
Spain	2%	2%	3%	3%	3%	1%	2%	2%	1%	1%	2%	1%	6%	3%	1%	5%	2%	4%	14%	3%
Sweden	3%	3%	1%	1%	0%	5%	2%	2%	5%	1%	2%	3%	1%	3%	3%	2%	5%	1%	1%	25%

Note: color shading reflects correlation magnitude. Source: author’s calculations based on data provided by Eurostat.

The results of the first cross-section regression model illustrate that two of the factors that significantly influence the correlation between the financial cycles of the Member States are the level of correlation between the economic cycles and the correlation between the indices of financial stress (Table 6). Thus, an increase in the link between two economies, materialized in a higher correlation of economic cycles, has an important impact on increasing the level of correlation of financial cycles. Moreover, if systemic risks and financial stress materialize synchronously in two Member States, this generates an increase in the link between the financial cycles of the two countries. However, it should be noted that the influence of the trade link between the two economies, materialized in the bilateral level of exports and imports, does not significantly influence the correlation of their financial cycles.

Table 6. Results of the regression model with cycles determined using the Band-Pass filter.

Financial Cycle Correlations
	Estimate	SE	t-Stat.	p-Value
Intercept	−0.715	0.1061	−6.7393	1.94 × 10−10
Trade links (Exp-Imp)	0.052234	0.19433	0.26879	0.78839
Systemic stress correlations (CLIFS)	0.95373	0.15431	6.1805	3.95 × 10−9
Business cycle correlations	0.58237	0.098801	5.8944	1.74 × 10−8
Number of observations: 190
Error degrees of freedom: 186
Root Mean Squared Error: 0.217
R-squared: 0.377
Adjusted R-Squared: 0.367
F-statistic vs. constant model: 37.5
p-value = 5.56 × 10−19

Source: author’s calculations based on data provided by BIS, Eurostat, ECB and IMF.

For the robustness of the results, the regression model was also analyzed under the conditions in which the financial cycles were calculated with a Hodrick–Prescott filter (Table 7). In this case, the variables that significantly influence the correlation of financial cycles are the correlations between the economic cycles of the countries, but also the correlations between their financial stress indices. However, different from the results of the first regression model, in this case, the impact of the two variables on the correlation of financial cycles is lower. Moreover, the impact of the variable on bilateral international trade does not have a significant impact on the correlation of financial cycles.

Table 7. Results of the regression model with cycles determined using the Hodrick–Prescott filter.

Financial Cycle Correlations
	Estimate	SE	t-Stat.	p-Value
Intercept	−0.51036	0.091937	−5.5512	9.67 × 10−8
Trade links (Exp-Imp)	−0.01934	0.1684	−0.11483	0.90871
Systemic stress correlations (CLIFS)	0.78985	0.13372	5.9068	1.63 × 10−8
Business cycle correlations	0.49518	0.085615	5.7838	3.04 × 10−8
Number of observations: 190
Error degrees of freedom: 186
Root Mean Squared Error: 0.188
R-squared: 0.362
Adjusted R-Squared: 0.352
F-statistic vs. constant model: 35.2
p-value = 4.67 × 10−18

Source: author’s calculations based on data provided by BIS, Eurostat, ECB and IMF.

In the case of models on spillover effects, all independent variables are statistically significant (Table 8). Thus, an increase in the value of claims between two states generates an increase in the spillover effect on their financial cycle. Another important element, which shows an even greater impact than bilateral claims, is the bilateral trade between countries. Thus, an intensification of the level of imports or exports between two countries generates an increase in the spillover effects on the financial cycle. A strong and significant effect on the financial cycle spillovers is also given by the spillover effects recorded at the level of the financial stress index. Thus, the more the systemic risks and financial stress within one state influence the dynamics of the financial stress index in another state, we find stronger spillovers for the financial cycles at the level of the two states.

Table 8. Results of the regression model on spillover effects.

Financial Cycle Spillovers
	Estimate	SE	t-Stat.	p-Value
Intercept	−0.01179	0.005797	−2.0342	0.043113
Financial Claims	0.045599	0.014627	3.1174	0.002065
Trade links (Exp)	0.056685	0.015291	3.7072	0.000264
Systemic stress spillovers (CLIFS)	0.26613	0.083494	3.1875	0.001641
Business cycle spillovers	0.65866	0.088738	7.4225	2.40 × 10−12
Number of observations: 228
Error degrees of freedom: 223
Root Mean Squared Error: 0.0249
R-squared: 0.322
Adjusted R-Squared: 0.31
F-statistic vs. constant model: 26.5
p-value = 5.73 × 10−18

Source: author’s calculations based on data provided by BIS, Eurostat, ECB and IMF.

However, the most important but also the most significant impact on the spillover effect of the financial cycle comes from the economic cycle spillovers. Thus, as the economic cycles in two states become tighter, a similar effect is observed regarding the dynamics of the financial cycles in these states.

In order to analyze the robustness of the results regarding the regression model on spillover effects, the analysis was performed taking into account only a part of the independent variables. The extended results of all 14 combinations of cross-section regression models are presented in Table 9. The variable on bilateral claims between states is statistically significant for all models and has a positive coefficient in all variants. Two other variables that are statistically significant for each of the models they were part of are the spillover effect on the financial stress index and the spillover effect on the business cycle. The coefficient of the bilateral trade variable illustrates that an intensification of the trade between two countries generates an increase in the spillover effect at the level of financial cycles, only for some of the combinations of models analyzed.

Table 9. Extended results of the regression models on spillover effects using all model combinations.

Financial Cycle Spillovers
Model Combination	Intercept	Financial Claims	Trade Links	Systemic Stress (CLIFS)	Business Cycle Spillovers
M1000	0.03719373 ***	0.065158 ***
M0100	0.040042969 ***		0.032677 *
M0010	0.020389364 ***			0.465769 ***
M0001	0.008593124 *				0.68782 ***
M1100	0.033242908 ***	0.06941 ***	0.040788 **
M1010	0.019454511 ***	0.047521 ***		0.39889 ***
M1001	0.006035901	0.051676 ***			0.648219 ***
M0110	0.018369888 ***		0.027722	0.457241 ***
M0011	−0.004925685			0.355744 ***	0.61403 ***
M1110	0.016848694 ***	0.051868 ***	0.034598 **	0.382129 ***
M1101	−0.003076275	0.056885 ***	0.062621 ***		0.711601 ***
M1011	−0.004900442	0.039356 ***		0.30383 ***	0.594639 ***
M0111	−0.011153638 *		0.051197 ***	0.329136 ***	0.674631 ***
M1111	−0.011792041 **	0.045599 ***	0.056685 ***	0.266135 ***	0.65866 ***

* 10% significance level; ** 5% significance level; *** 1% significance level. Source: author’s calculations based on data provided by BIS, Eurostat, ECB and IMF.

Table 10 extends the robustness exercise for the case of correlation-based models, where different combinations of models were analyzed. The only variable whose coefficients are statistically insignificant for all the models in which it was integrated is the one regarding bilateral trade. Conversely, both the correlation between financial stress indices and between economic cycles tend to have a significant impact on the correlation between financial cycles in all models they were included.

Table 10. Extended results of correlation-based regression models.

Financial Cycle Correlations
Model Combination	Intercept	Trade Links	Systemic Stress Correlation (CLIFS)	Business Cycle Correlations
M100	0.356659 ***	−0.0255
M010	−0.51262 ***		1.270818335 ***
M001	−0.20901 ***			0.790081 ***
M110	−0.50923 ***	−0.08232	1.272878261 ***
M101	−0.22179 ***	0.139062		0.796627 ***
M011	−0.71178 ***		0.95673209 ***	0.579255 ***
M111	−0.715 ***	0.052234	0.953733574 ***	0.582375 ***

*** 1% significance level. Source: author’s calculations based on data provided by BIS, Eurostat, ECB and IMF.

In summary, our results provide a detailed empirical characterization of both synchronization (via correlations) and spillovers (via variance decompositions) of financial cycles within the EU. Overall, the findings confirm that business and financial cycles are synchronized primarily through financial linkages rather than trade channels, supporting RQ1 and RQ2: countries with strongly overlapping business cycles and concurrent stress episodes exhibit higher financial-cycle correlation, and cross-border banking ties serve as a key transmission channel for financial cycle spillovers. At the same time, trade intensity appears to exert only a marginal influence on financial cycle co-movements in our sample, suggesting that real economy trade linkages are less important than financial factors in driving these dynamics. We also find evidence addressing RQ3: Euro-area membership tends to magnify both synchronization and spillovers of financial cycles. The common monetary policy and integrated financial markets in the currency union mean that euro-area countries share more in both the booms and busts of financial cycles, whereas non-euro countries retain a degree of insulation (at the cost of higher idiosyncratic volatility). These results set the stage for a deeper discussion of their implications in light of existing studies and for policy-making in the EU.

5. Discussion

The results reveal notable heterogeneity in the synchronization of financial cycles across EU member states. Core euro-area countries display the highest co-movement, reflecting both deeper financial integration and shared institutional frameworks, while newer or non-euro members exhibit more idiosyncratic patterns. This asymmetry highlights a two-speed dynamic of integration: the Single Market has succeeded in aligning real economic activity, but financial integration remains more fragmented.

These findings are broadly consistent with earlier evidence that financial cycles can display tighter synchronization than business cycles in highly integrated economies (Claessens et al., 2011; Borio, 2014), yet they extend the literature by offering a comprehensive EU-wide comparison. Our results also differ from those of Schüler et al. (2020), who focused primarily on G7 economies. By including all EU members, we capture heterogeneity arising from institutional development, euro adoption, and banking-sector depth.

Another important contribution lies in the analysis of trade linkages. The evidence of “trade decoupling” suggests that trade intensity alone cannot explain the pattern of financial co-movement across EU members. Even where trade integration is strong, financial cycles may diverge if cross-border credit flows and capital market depth differ substantially. This reinforces the idea that macro-financial synchronization requires not only common goods-market linkages but also institutional and regulatory convergence in financial systems.

From a policy perspective, these findings underscore the need for coordinated macroprudential frameworks and cross-border banking supervision. Financial stability risks and spillovers are likely to propagate asymmetrically across the Union. Hence, ensuring resilience in the euro area requires continued progress toward the Banking Union and Capital Markets Union, which can help dampen divergences and contain contagion during crisis episodes.

6. Conclusions

This paper has examined the co-movement of financial cycles across EU member states, highlighting their relationship with business-cycle synchronization and episodes of financial stress. The evidence shows that financial and real cycles tend to move together, particularly among euro-area countries, where shared monetary institutions and integrated financial systems amplify cyclical alignment. In contrast, newer and non-euro members display weaker synchronization, consistent with more limited financial integration and less developed financial intermediation.

These findings provide new insight into the architecture of European integration. They suggest that deeper institutional and financial convergence enhances resilience to common shocks, while persistent heterogeneity calls for a cautious, multi-speed approach to further euro-area enlargement. From a macroeconomic standpoint, the results reveal how structural differences in financial depth and policy coordination can influence the transmission of shocks, particularly during systemic crises such as 2008–2009 and the 2011 sovereign debt episode.

The analysis also speaks to the ongoing debate over economic convergence within the EU. While the Single Market continues to promote synchronization in real activity, the financial cycle remains more sensitive to national institutional settings and cross-border regulatory linkages. This duality points to the importance of advancing financial integration in tandem with fiscal coordination to preserve the long-term stability of the Union.

While this study is descriptive and does not establish causality, it offers an updated empirical picture of cyclical interdependencies within the EU and identifies key asymmetries across member states. Future research could further investigate the direction of causality between financial and business cycles, explore non-linear transmission channels, and assess the implications of evolving EU institutions for cyclical dynamics.