Eurozone bank integration: EU versus non-EU banks

Vincent Bouvatier, Anne-Laure Delatte

14 December 2014



In 2013, cross-border capital flows were 40% of their 2007 level (McKinsey Global Institute 2013). While the reversal was in all broad categories of flows (Forbes and Warnock 2012), the sharpest decline in activity was in international bank loans extended cross-border or by local affiliates (Milesi-Ferretti and Tilles 2011), a fact that has prominently driven contraction in the real economy (Cetorelli and Goldberg 2011, 2012). In a recent CEPR paper, we document the dynamics of international banking activities with the benefit of hindsight and show that the forward march of banking integration has reversed only as far as Eurozone countries are concerned as source or destination countries (Bouvatier and Delatte 2014). In the rest of the world, the decline of international banking activity in the aftermath of the Global Crisis was entirely due to temporary frictions. In sharp contrast with the Eurozone, our results indicate that the banking integration has in fact strengthened in the rest of the world.

Figure 1 below plots the evolution of the consolidated bilateral foreign claims reported by the banks of 14 countries vis-à-vis 196 partners at the Bank for International Settlements (solid line). The dashed and dotted lines plot the same data broken down by banks from Eurozone and non-Eurozone reporting countries.

Figure 1. Consolidated foreign claims, 1995–2012

Source: Authors’ calculations with data from the Bank for International Settlements.

No doubt the Global Crisis dealt international banking a serious blow – in 2008, the solid line decreases significantly and has remained at this lower level since. However, the aggregate situation hides two opposite evolutions. International banking activities reported by non-Eurozone countries were severely hit in 2007, but have recovered quickly since 2008 (dashed line). On the contrary, the activity by Eurozone banks has kept on declining (dotted line). It is worth observing that the slope of the dashed line is steeper before than after the crisis, a fact that means that banking activities have slowed down. In sum, raw statistics suggest a massive retrenchment of European banks and a temporary interruption in the rest of the world with a sluggish recovery.

Explaining the divergence of banking integration inside and outside the Eurozone

Can this heterogeneous situation be entirely attributed to the contrasting economic context inside and outside the Eurozone? In fact, European countries have faced sequential crisis episodes since 2008, and the Eurozone is now one of the few zones where the economy has not yet recovered. During the previous decade, rising institutional linkages have unambiguously accelerated financial integration in the Eurozone. Are we observing a correction after the tremendous acceleration of banking integration at the European level? How does the European situation compare with the rest of the world? To answer these questions, we compute a measure of international banking integration that controls for frictions and time-varying factors that affect banking activity.

In fact, recent empirical research reveals that bilateral financial positions rise proportionately with the economic size of both partners and their institutional linkages, and decline with frictions, including agency problems and information asymmetries proxied by the geographical distance between partners.1 We define banking integration as the changes in international banking activities that are not driven by these standard gravity factors. We use a non-linear time-trend to precisely capture these changes and we estimate four different integration trends depending on source and destination countries. In order to assess the magnitude of the patterns obtained, we compute the contribution of these trends in the model. In other words, by how much have banking activities deviated from their benchmark level justified by standard gravity factors? Our estimates yield the graphs in Figure 2.

Figure 2. Banking integration: Overshooting analysis

Note: The grey area corresponds to the one-standard error band.

First, it is striking that after the crisis, all trends involving the Eurozone as recipient or source countries are downward-sloping (Figures 2a, 2b, and 2c) contrary to the trend in the rest of the world (Figure 2d). In sum, the banking integration process has unambiguously ceased as far as the Eurozone is concerned. Second, we find that only Eurozone banks are experiencing a disintegration process (the curve is below the x-axis in Figures 2a and 2b). In turn, we find that the exposure to the Eurozone of banks from reporting countries outside the Eurozone is precisely at the benchmark level (Figure 2c). In other words, the integration of non-Eurozone banks in the Eurozone has declined, but this is a correction of the pre-crisis 20% overshooting and not a marked retrenchment. In contrast, Eurozone banks have massively reduced their international exposure inside and outside the Eurozone – in 2012, the international banking integration of Eurozone reporting banks vis-à-vis non-Eurozone partners and Eurozone partners was 33% and 37% below its benchmark level respectively. In sum, the banking fragmentation inside the Eurozone documented by the SYNFINT financial integration index of the ECB (ECB 2014) is only one side of a broader disintegration of Eurozone banks. The economic downturn faced by the Eurozone since 2008 is not sufficient to account for the massive retrenchment. And more importantly, this decline is not a correction of previous overshooting but a marked disintegration. Last, Figure 2d which plots the situation in the rest of the world, i.e. exposures outside the Eurozone of non-Eurozone reporting banks, emphasises a stunning difference – not only has banking integration never declined, but on the contrary the trend is even steeper after the crisis. In other words, the decline of banking activities observed in 2008 was entirely due to temporary frictions.

Concluding remarks

We have three comments to conclude. First, the simultaneity of the European retrenchment and of the acceleration of international banking integration of non-Eurozone members raises the questions of a transfer of international banking activities from the Eurozone to non-Eurozone countries. In other words, our findings suggest a loss of global market shares by European banks. Second, it would be interesting to test whether the disintegration uncovered in our work is partly driven by the conditions imposed on the banks by the European Commission to receive state aid. Indeed, as a counterpart to state aid, the European Commission requested banks to downsize and focus on the domestic economy. Third, it is striking that banking integration – one of the main achievements of the single currency union – has already vanished. Without urgent political action, there will soon be no more benefit to maintaining the euro.


Bouvatier, V and A L Delatte (2014), “International Banking: The Isolation of the Euro Area”, CEPR Discussion Paper 10264. Published in European Economic Review.

Cetorelli, N and L S Goldberg (2011), “Global banks and international shock transmission: evidence from the crisis”, IMF Economic Review 59(1): 41–76.

Cetorelli, N and L S Goldberg (2012), “Follow the money: quantifying domestic effects of foreign bank shocks in the Great Recession”, American Economic Review 102(3): 213–218.

ECB (2014), Financial integration in Europe, April. 

Forbes, K J and F E Warnock (2012), “Capital flow waves: surges, stops, flight, and retrenchment, Journal of International Economics 88(2): 235–251.

Head, K and T Mayer (2013), “Gravity equations: workhorse, toolkit, and cookbook”. Handbook of International Economics 4.

Milesi-Ferretti, G-M and C Tille (2011), “The great retrenchment: international capital flows during the global financial crisis”, Economic Policy 26(66): 289–346.

McKinsey Global Institute (2013), “Financial globalization: Retreat or reset?” March.

Portes, R and H Rey (2005), “The determinants of cross-border equity flows”, Journal of International Economics 65(2): 269–296.


[1] Portes et al. (2001) were the first to apply the gravity equations to financial flows. See Head and Mayer (2013) for a literature review of gravity models.



Topics:  Financial markets International finance

Tags:  euro, eurozone, financial integration, capital flows, banking, cross-border capital flows

Associate Professor, University Paris Ouest Nanterre La Défense

Tenured Researcher, French Institute for Scientific Research (CNRS); Fellow, OFCE-Sciences Po; Research Affiliate, CEPR