VoxEU Column EU policies International Finance

Banking integration: Friend or foe?

Is a banking union the answer to Europe’s woes? This column argues that banking union is no panacea – and it may actually make monetary policy harder. It urges Europe’s policymakers to re-evaluate their proposals.

To address the instability of the EZ, many analysts advocate closer financial integration among EZ members. While proposals differ in their details, some involve bold initiatives that would ‘complete’ the monetary unification via the establishment of a banking union. Recently the European Commission has released specific proposals initiating the policy debate.

Will a banking union ‘complete’ monetary unification, helping to narrow within EZ imbalances and synchronise the asymmetric cycles of member states? Academic research has not been helpful in answering this question since, according to the theory, financial integration may lead to a higher level of business cycle synchronisation but it can also cause a ‘decoupling’ of business cycles between these interconnected economies1. The key issue is the nature of shocks that hit the economies.

Take the case of two integrated economies, where one of the two is hit by a negative shock. If the shock hits the banking sector, then problems in one country are likely to spread to others, as global banks (operating in both countries) will probably pull funds from the non-affected country to cover their losses or maintain the regulatory capital requirements. This in turn will lead to contagion, making the business cycles of the interlinked countries more synchronised. Recent research indeed shows that many global banks pulled out from many European countries and emerging markets after the fall of 2008, when the financial crunch in the US capital markets intensified (Cetorelli and Goldberg 2011).

If, however, the negative shock hits the productivity of firms, businesses, and entrepreneurs in a country, while banks stay (relatively) healthy, then the return to capital falls and banks pull funds out from the effected country, amplifying the initial shock; this in turn makes the business cycles (and especially the investment component of the national account) of financially interconnected economies to diverge. Take the case of Greece nowadays. Given the negative shock to the productivity of the economy it should not be that surprising that European financiers pull out their capital. The huge reallocation of capital from the countries of the European south to the core of the EZ can be explained through this mechanism that operates via integrated banks during tranquil times. Before the crisis, during 2001-2007, the same mechanism seems to have been in place: but during this period capital was allocated from the European core to the periphery as Greece, Spain, Portugal, and Ireland experienced a productivity boom.

Recent empirical findings

In recent work with Jose-Luis Peydro and Fabrizio Perri (Kalemi-Ozcan et al. 2012a, 2012b, both forthcoming), we examine the role of financial integration on business cycle synchronisation, explicitly allowing for the possibility that its effect may differ in tranquil and in financial crisis times. For our analysis we exploit a confidential dataset from the Bank of International Settlements (BIS) that covers all international bilateral banking activities for the 20 largest economies over the past three decades.

We document that across country pairs, there is a significant positive correlation between financial integration, measured as cross-border banking exposures, and output synchronisation. This should come as no surprise. The business cycle of the US economy is both more synchronised and more financially linked with Canada than with Germany or France. The reasons for such a result can be many, such as socioeconomic ties and close distance between certain country pairs compared to others.

In sharp contrast to the positive cross-sectional correlation, when we examine the response within country pairs of output synchronisation to financial integration before the 2007-2009 crisis we find a significantly negative association. This emerges both when we use cross-border banking exposures to proxy financial integration and when we employ an index of integration that reflects legislative-regulatory harmonisation policies in financial services among EU member states2. This implies that in tranquil times increases in financial integration within country-pairs (say Canada-US or France-Germany) over time are associated on average with less synchronised, more divergent, output cycles. The negative within country-pair association between bilateral financial linkages and business cycle co-movement is in line with the standard textbook models implying that in the absence of financial sector shocks, cross-border financial integration should magnify total-factor-productivity shocks and make output patterns diverge.

Yet, the negative correlation within country pairs between financial integration and output synchronisation turned positive or zero between 2007 and 2009. This result aligns with theoretical models and the conventional wisdom that financial integration facilitates co-movement via contagion during periods of financial meltdown.

In the same vein we find that during 2007-2009 a significantly positive association emerges between banking linkages to the US and business cycle synchronisation with the US. Interestingly this association emerges only when we use a broad measure of financial links to the US financial system that on top of direct exposure to the US also incorporates indirect exposure via the Cayman Islands, the main off-shore financial centre of the US economy.

Policy implications

These findings are important not only because they shed light on the delicate role played by financial globalisation on the synchronicity of economic activity, but because they bridge two bodies of research in international macroeconomics and finance on the implications of financial integration, one that looks at its effect on international business cycles and another that focuses on financial contagion3. Specifically, the negative association within country pairs between bilateral financial linkages and business cycle co-movement in regular times is in line with the workhorse models that most central banks use. The policy message is clear: in the absence of financial sector shocks, cross-border financial integration magnifies total-factor-productivity shocks making output patterns diverge4.

The results are of utmost significance for policymakers since they imply that the conduct of monetary policy becomes significantly harder within financially integrated currency areas. This is because financial integration magnifies output fluctuations across countries therefore making monetary policy instruments inappropriate for all regions. This problem is clearly illustrated nowadays in the EZ5. The high degree of banking and capital market integration between EZ member countries has amplified idiosyncratic (country-specific) asymmetric shocks thus leading to huge divergence in economic activity (and real interest rates to firms and entrepreneurs) between the countries of the south and the core; and as global banks and other financial intermediaries can easily pull capital out of the periphery the historically low policy rate of the ECB is not channelled to Italy, Spain, Portugal, and Greece. At the same time, real rates in Germany or the Netherlands are currently negative, rationalising (to some extent at least) the fears of the Bundesbank for inflation.

What is the way forward? The standard textbook theories of financial integration also have implications about consumption co-movement and risk sharing. Financial integration makes output cycles of integrated economies to diverge but under standard assumptions consumption cycles may converge as financial integration facilitates international diversification and risk sharing. While research shows that there are some positive effects of financial integration on risk sharing and consumption smoothing, the economic magnitude is small or moderate at best for countries, though sizeable within federations6. While we have not studied this effect yet for European economies from causal inference it seems like consumption cycles are not converging at the moment: southern European countries are getting more improvised with lower levels of consumption and higher levels of unemployment relative to the northern countries. Hence it might be indeed true that financial integration is not complete and can be complete with a banking union that allows debt mutualisation and a central supervision of all banks.

Of course, the key difference between Europe and the US is that, the US banking union not only works via central supervision and common deposit insurance but also via existence of institutions such as FDIC that orchestrate the resolution process by taking up failed banks. Most importantly in the US the federal government buffers asymmetric (state-level) shocks via fiscal transfers. It is thus still unclear how a currency union with an almost perfect degree of banking integration but without fiscal backstops and transfers can operate. Economic research, both theoretical and empirical, suggests that by itself banking union (or complete financial integration) will not dampen business cycles of member states7. Actually a higher degree of financial integration may lead to more divergent output cycles, in spite of potentially converging consumption cycles in the future. Moreover financial integration and banking union in particular may be a destabilising force when shocks hit the financial sector in a crisis.

Our arguments should not be taken as being against to the recent proposals for enhanced regulation-supervision of financial institutions towards the creation of a ‘banking union’ within Europe. We are simply pointing out the fact that academic research suggests, by itself banking union is not a panacea to Europe’s economic woes and may make the conduct of monetary policy harder. This implies that European policymakers must re-evaluate proposals towards the creation of some form of fiscal union.


Backus, D, P Kehoe, and F Kydland (1992), “International Real Business Cycles”, Journal of Political Economy, 100(4):745-775.

Cetorelli, N and L Goldberg (2011), “Banking Globalization and Monetary Transmission”, Journal of Finance, forthcoming.

Farhi, E and I Werning (2012), “Fiscal Unions”, mimeo Harvard University and MIT.

Frankel, J and A Rose (1998), “The Endogeneity of the Optimum Currency Area Criterion”, Economic Journal, 108(6):1009-1025.

Holmstrom, B and J Tirole (1997), “Financial Intermediation, Loanable Funds, and the Real Sector”, Quarterly Journal of Economics, 112(3):663-691.

Kalemli-Ozcan, S, BE Sørensen, and O Yosha (2001), “Regional Integration, Industrial Specialization and the Asymmetry of Shocks across Regions”, Journal of International Economics, 55(1):107-137.

Kalemli-Ozcan, S, BE Sørensen, and O Yosha (2003), “Risk Sharing and Industrial Specialization: Regional and International Evidence”, American Economic Review, 93(3): 903-918.

Morgan, DP, B Rime, and P Strahan (2004), “Bank Integration and State Business Cycles”, Quarterly Journal of Economics, 119(3):1555-1585.

Kalemli-Ozcan, Sebnem, Elias Papaioannou, and Fabrizio Perri (2012b), “Global Banks and Crisis Transmission”, Journal of International Economics, forthcoming.

Kalemli-Ozcan, Sebnem, Elias Papaioannou, and Jose-Luis Peydro (2010), “What Lies Beneath the Euro’s Effect on Financial Integration? Currency Risk, Legal Harmonization, or Trade?”, Journal of International Economics, 81(1): 75-88.

Kalemli-Ozcan, Sebnem, Elias Papaioannou, and Jose-Luis Peydro (2012a), “Financial Regulation, Financial Globalization, and the Synchronization of the Economic Activity”, Journal of Finance, forthcoming.

Mundell, R (1961), “A Theory of Optimum Currency Areas”, American Economic Review, 51(4):657-665.

Papaioannou, Elias, Sebnem Kalemli-Ozcan, and José-Luis Peydró (2009), “What is it good for? Absolutely for financial integration”, VoxEU.org, 20 June.

Perri, F and V Quadrini (2012), “International Recessions”, USC Working Paper.

1 Most theoretical works focus on either tranquil times or during periods of financial turmoil. For theoretical models where both countervailing mechanisms are in place see, among others, Holmsrom and Tirole (1997), Morgan, Rime, and Strahan, (2004), Quadrini and Perri (2011), and our recent research Kalemli-Ozcan et al. (2012b, forthcoming).

2 In Kalemli-Ozcan et al. (2012b) we show that EU-wide policies that aimed to harmonise the regulatory and legislative framework of financial intermediation have crucially contributed to the spur of cross-border financial transactions in the past decade. Extrapolating from these findings implies that if the current proposals to further harmonise banking regulation across the EU will further increase gross banking transactions across member states. See our review on Vox (Papaioannou et al. 2009).

3 In Kalemli-Ozcan et al. (2012b) we develop a dynamic stochastic general equilibrium model with global banks that helps reconcile our empirical findings.

4 See for example Backus et al. (1992).

5 See Mundell (1961) for the classical treatment of the issue and Frankel and Rose (1998) for pointing out the endogenous nature of the currency area and the associated monetary policy to trade and financial integration.

6 The literature shows that within fully financially integrated federations, such as the US and the UK, output cycles diverge but consumption cycles converge as predicted by the theory (Kalemli-Ozcan et al. 2001, 2003). The key point is that most of this effect is delivered by capital markets and not by the federal government during regular times.

7 See Farhi and Werning (2012) for a theoretical exposition on the interplay between financial integration and fiscal transfers within a currency union.

5,985 Reads