euro: no free lunch for the outsiders

A few years ago, Andy Rose of the University of Berkeley started a whole literature on the effects of common currencies on trade. His message was that the effect was really large and this was interpreted as suggesting that the euro would also have a large effect on trade. The effect has since been downsized from very large to modest (see Baldwin, 2006). I do not want to start a debate on this issue again but rather to change the focus to financial markets (see also Lane 2006 on this). After all, there are good reasons to think that the euro has a potentially more direct effect on transaction costs (interpreted in a very broad sense) on financial markets than on goods markets so there is a chance to get a larger effect. And indeed, this seems to be the case. In some recent work with Nicolas Coeurdacier from the London Business School, I looked at the euro’s impact on trade in financial assets (Coeurdacier and Martin 2007). We used two data sets – a cross-country one on bilateral asset holdings (bonds, equity and banking assets) and a Swedish data set on both holdings of foreign assets and outflows. Our estimates based on comparing purchases of assets with similar characteristics suggest that the transaction costs to buy assets from the euro zone are lower by around 17% for equity and 14% for bonds than average. This effect of the euro benefits both those investors that are in and outside of the euro zone. In addition to this effect that benefits all, investors inside the euro zone benefit from an extra decrease of transaction costs for equities and bonds respectively of around 10% and 17%. This is something like a preferential financial liberalization effect. Hence, for an investor inside the euro zone the transaction cost for the cross border purchase of a stock or a euro bond is lower by around 27% and 31% respectively. Overall, this translates in large effects – may be too large to believe for some who may downsize us too in the future– on cross-border asset holdings. The euro increases by 150% bilateral bond holdings between two euro countries while equity holdings rise by around 45%. We try as much as possible to identify a pure euro effect by taking into account a relatively large set of variables that might be correlated with being part of the euro (trade linkages, geography, exchange rate volatility…) but I admit that we will need more data to get a more robust estimate. The effect is much larger for bonds than for equity even though the percentage difference in transaction costs inside and outside the euro zone is estimated to be similar. This actually is not surprising because different bonds are much closer substitutes than different equities so that any reduction in transaction costs has a much larger effect on holdings. One result I find particularly interesting is that contrary to the euro effect on trade in goods, the euro does not seem to generate any decrease of the transaction cost for euro investors of purchasing equity outside the euro zone. In fact, for equities there seems to be substantial diversion taking place in the sense that euro countries buy less equity from outside the euro zone than what is predicted by our financial gravity equations. This evidence is based on comparing asset trade between euro countries and the Nordic countries in (Finland) and out (Sweden, Norway, Denmark) of the euro zone. There does not seem be any diversion effect seems for bonds and our guess is this is because a significant portion of bonds in these countries are issued in euro. The evidence of a diversion effect for cross border equity trade tells us also a bit more about how the euro affected financial flows: it points to a transaction cost story where the relative – not absolute – transaction cost for a euro based investor to buy assets from the rest of the world has increased with the euro. The fact that we find that the euro effect is larger for flows than for stocks in the case of equities and loans also points to a transaction cost story. One of the messages of Baldwin (2006) on the euro trade effect is that countries do not need to be inside the euro zone to benefit from most of its economic gains. This has some intriguing political economy implications on the future dynamics of monetary integration with potential free-rider problems. But for the financial side of the euro story, this message does not hold fully. Outsiders do benefit from lower transaction costs to diversify risk when purchasing euro assets but remember the gain is around half what the insiders get. Moreover, the diversion effect should clearly be detrimental to non euro countries. If assets are imperfect substitutes, the lower demand for non euro equity (the only asset for which some diversion is suggested by our empirical analysis) implies a lower price of non-euro assets relative to euro assets. This diversion effect should therefore translate into an increase in the cost of capital for firms outside the euro zone. So after all the euro is not a free lunch…for outsiders.

3 Responses to "euro: no free lunch for the outsiders"

  1. KS   October 21, 2007 at 5:02 pm

    How would you analyze the experience of the UK? As an international financial center they decided to keep their own currency and their choice seems to have been vindicated so far.

  2. Anonymous   October 25, 2007 at 1:59 am

    If the euro has had the negative financial effects on non-EMU members of the EU that has not translated into lower economic growth. If anything those countries have done very well growth-wise compared to the EMU ones. So the factors you discuss are important and relevant but reduction of transaction costs cannot compensate for some more fundamental economic and policy weaknesses.

  3. Anonymous   October 25, 2007 at 2:01 am

    Great analysis. But do we see welfare benefits of such a larger cross-border portfolio diversification? is cross-country risk insurance of national shocks significantly higher in the EMU members?