Financial market integration is a key economic goal of European integration. Knitting together EU capital markets should foster efficient capital allocation and better risk pooling. Both of these should improve Europe’s investment climate and thus be pro-investment and pro-growth. As far as removing formal barriers is concerned, Europe has come a long way. Starting with the abolition of capital controls under the 1992 Single Market Programme and moving forward with the creation of the EMU and the Financial Services Action Plan, the EU has steadily removed the legal barriers.
The empirical evidence on EU capital market integration is mixed. One strand of evidence suggests tight integration of money markets and government bond markets as having been achieved (Ehrmann et al., 2007; ECB, 2007). By contrast, a Huizinga and Jonung (2005) study on the internationalization of asset ownership in Europe concludes that in spite of all the reforms, the process of European financial integration is far from complete. Both measures of integration are fraught with difficulties.
A very recent line of research proposes an alternative measure that looks at capital market integration as more of a macro problem and therefore avoids the limitations of price or ownership based measures. Interestingly, results from this approach yield two important findings:
- Apart from Ireland, EU nations are not tightly integrated;
- The determinants of integration go far beyond institutional measures that the EU could fix with legislations; a great deal of the variation in integration across EU regions depends upon the degree of trust and confidence.
Measuring financial market integration
A new line of research focuses on an indirect way of measuring “deep” capital market integration (Kalemli-Ozcan, Reshef, Sørensen, and Yosha, 2006). If a nation’s capital has a geographically diverse ownership, then we should see a separation between regional jumps in output and regional jumps in capital income. For example, consider regions that are small relative to the aggregate market. If ownership is perfectly diversified, almost all capital income should come from other regions while almost all capital income generated in the region should go to other regions. Given that about 30% of output goes to capital, a (small) region that sees an output growth of 100 million euros should send 30 million euros to other regions if its capital ownership is perfectly diversified. In other words, the change in income following growth – keeping output of other regions constant – should be 70 million or 0.7 times the output shock (assuming labour income is not diversified). A test of this prediction of “deep” integration using data from U.S. states finds that the prediction is met, i.e. the U.S. seems to be deeply integrated (Kalemli-Ozcan, Reshef, Sørensen, and Yosha, 2006).
Recent work we’ve undertaken with Mehmet Ekinci tests the same prediction for EU nations and regions and finds little evidence of integration between EU countries except for Ireland, which is well-known to be integrated into the world economy. More surprisingly, we find that not even regions within European countries are integrated in the deep sense (Ekinci et al., 2007).
Why so little integration among regions within nations?
An obvious answer is that country-level financial institutions account for the difference, but the data reject this. The question then remains: our regional dataset is ideal for hunting for alternative answers since they allow us to compare European regions – such as Sardinia or Bavaria – to each other, isolating the countrywide legal systems and institutions that cause differences between Italy and Germany.
Social scientists have long been aware that levels of “social capital” may differ within countries – southern versus northern Italy is a well-known case. This may have consequences for financial integration as also argued by Guiso et al. (2006). People will likely invest less via formal markets if they trust others less (outside of family and close acquaintances), or if they have low confidence in institutions – financial institutions may be an obvious example, but the courts that ultimately are the enforcers of contracts may be even more important.
We explore this using data on trust and confidence measured in the World Value Survey. Figure 1 and Figure 2 display the data on relative (to the country average) degrees of trust and confidence; darker colours indicate higher levels of trust or confidence. The differences are marked, e.g. the level of trust is higher in northern than in southern Germany while the level of confidence is higher in western than in eastern Germany.
We find that these differences in trust and confidence are strong determinants of “deep” financial integration. It appears that regions with low level of confidence and trust do not invest in other regions and/or do not receive investments from other regions.
Figure 1: Trust within EU
Regional capital flows
In addition to investigating changes in regional capital income following output shocks, Ekinci et al. (2007) also examine the determinants of the level of net capital flows. Net capital flows into a region equals physical investment minus saving and may be positive or negative. There is no reason for net capital flows to be near zero (meaning within a percent or two of output). In particular, capital ought to flow to regions or countries with high growth. Feldstein and Horioka (1980) pointed out long ago that net capital flows between countries appear to be much smaller than one should expect in financially integrated economics.
Blanchard and Giavazzi (2002) interpret recent data, showing that countries such as Greece and Portugal have received large capital inflows, as the “end of the Feldstein-Horioka puzzle” within the EU. However, our results indicate the process of “deep‘” financial integration between countries may need some time before capital adjusts as freely as between US states. In countries, where all regions are at a similar level of development, high growth regions will also typically be high output regions.
Figure 2: Confidence within EU
Note that this observation goes against the intuition of many economists because the world has witnessed many instances of poorer countries – such as the Asian “tigers” or Ireland – growing fast and catching up with the richer countries. This is also what would be predicted by the standard neoclassical theory. Kalemli-Ozcan, Sørensen, and Turan (2007) study indicators of catch-up growth for European countries and US states and find that while the southern US states displayed high catch-up growth before 1960, the catch-up phase appears to have run its course within the US. However, the less developed countries in the EU, such as Greece and Portugal, are still in the catch-up phase and, therefore, display different patterns than US states. For highly integrated regions within EU countries, one might expect to see capital flow to regions with high growth and output. We find that the prediction that high output regions have been recipients of outside investment and, therefore, have negative net asset positions (are net debtors) holds in northern Europe but not in southern Europe. In the South (regions of Portugal, Italy, and Spain), there is little evidence that savings are channelled from poorer low growth regions to regions with relatively high output.
Our findings suggest that Europe has a long way to go before its capital markets are as integrated as the U.S. market is internally. However, our work also suggests that much of the fragmentation stems from things that the EU cannot directly affect in the short run. Trust and confidence are things that evolve slowly. Policies that reward transparency and punish corruption may help but this is likely to take generations as exemplified by the low level of confidence in East Germany.
Banfield, Edward (1958), The Moral Basis of a Backward Society, New York: The Free Press.
Blanchard, Olivier and Francesco Giavazzi (2002), “Current Account Deficits in the Euro Area: The End of the Feldstein-Horioka Puzzle,” Brookings Papers on Economic Activity 2, 147–186.
European Central Bank (2007), “Financial Integration in Europe,” March 2007 Report
Ehrmann, Michael, Marcel Fratzscher, Refet S. Gurkaynak, and Eric T. Swanson (2007), “Convergence and Anchoring of Yield Curves in the Euro Area,” CEPR WP. 6456.
Ekinci, Mehmet, Sebnem Kalemli-Ozcan, and Bent Sørensen (2007), “Capital Flows within EU Countries: The Role of Institutions, Confidence and Trust,” In: NBER International Seminar on Macroeconomics, (eds) Richard H. Clarida, Jeffrey A. Frankel, Francesco Giavazzi, and Kenneth D. West, MIT Press, forthcoming.
Feldstein, Martin, and Charles Horioka (1980), “Domestic Savings and International Capital Flows,” Economic Journal 90, 314–29.
Frankel, Jeff (2001), “Comment on: Long Term Capital Movements by Lane and MilesiFerretti,” in B. S. Bernanke and K. Rogoff, editors ,” NBER Macroeconomic Annual. Cambridge, Mass. MIT Press.
Guiso, Luigi, Paola Sapienza and Luigi Zingales (2006), “Does Culture Affect Economic Outcomes?” NBER WP. 11999.
Huizinga, Harry and Lars Jonung (2005), The Internationalisation of Asset Ownership in Europe, (Eds.) New York: Cambridge University Press.
Kalemli-Ozcan, Sebnem, Ariell Reshef, Bent Sørensen and Oved Yosha (2007), “Why Does Capital Flow to Rich States?” NBER WP. 11109.
KalemliOzcan, Sebnem, Bent Sørensen and Belgi Turan (2007), “Where does Capital Flow? A Comparison of US States and EU Countries 1950–2000.” University of Houston WP.
Putnam, Robert (1993), Making Democracy Work. Civic traditions in modern Italy, Princeton, NJ: Princeton University Press.