Harald Uhlig's latest paper "Sovereign Default Risk and Banks in a Monetary Union" (CEPR DP9606, August 2013, http://www.cepr.org/pubs/dps/DP9606) "seeks to understand the interplay between banks, bank regulation, sovereign default risk and central bank guarantees in a monetary union".
The rationale for the paper is that the "European Monetary Union is in distress. Mechanisms that were meant to safe-guard key institutions and to assure stability have become sources of balance sheet risk for these very institutions. Liquidity provision within
the European Monetary Union rests upon repurchase agreements, by which banks guarantee the repurchase of assets deposited with the ECB. If either the bank fails or the asset fails, but not both, this mechanism safe-guards the repayment to the ECB, since it can either rely on the repurchase by the bank or sell the asset. However, when both fail as well as the bank home country fails, the ECB incurs a loss."
Abstracting away from the (important) debate about the implications of such a 'loss', the theoretical framework described by Uhlig is insightful and interesting. The author assumes "that banks can use sovereign bonds for repurchase agreements with a common central bank, and that their sovereign partially backs up any losses, should the banks not be able to repurchase the bonds."
Furthermore, "In the model, banks pursue their investment strategy voluntarily: it is up to regulators to potentially constrain them. Other explanations are conceivable, of course". This is different from the currently dominant views, as per Reinhart (2012a) as well as Claessens and Kose (2013). Specifically, it is distinct from Reinhart (2012b) argument as to why banks hold bonds of their home country. Reinhart argues that in a “financial repression” setting the regulators "make
[the banks] hold the sovereign bonds, perhaps with strong-arm tactics, perhaps in exchange for “looking the other way” concerning weak portfolios of commercial loans and mortgages, or simply as a “favor” in a long, ongoing relationship. Since the banks could potentially refuse, though at considerable cost, it still must ultimately be preferable to them to hold own-country bonds rather than invest elsewhere or to close: so, in some ways, this paper may also be understood as a model of financial repression." Another view for the system by which the banks end up holding rising exposures to domestic sovereign bonds is a political economy argument: "if sovereign bonds are held by home banks, it makes it politically harder to default on these bonds, as this will hurt domestic banks and savers. If so, then such a portfolio arrangement might serve as a commitment device for the government in trouble."
Uhlig's (2013) paper is not covering the underlying reasons for the holding of the bonds.
Overall, "the issue of sovereign default risk, bank portfolios and the role of the central bank has received considerable attention in the recent literature. Acharya and Steffen (2013) is a careful empirical analysis of the “carry trade” by banks, which fund themselves in the wholesale market and invest in risky sovereign bonds. They document, that “over time, there is an increase in ’home bias’ – greater exposure of domestic banks to its sovereigns bonds – which is partly explained by the ECB funding of these positions"… Relatedly, Corradin and Rodriguez-Moreno (2013) show that USD-denominated sovereign bonds of Euro zone countries became substantially cheaper (i.e., delivering a higher yield) than Euro-denominated bonds during the Euro zone crisis, and ascribe it to the usefulness to banks of Euro-denominated bonds as collateral vis-a-vis the ECB, while USD-denominated bonds do not offer this advantage." In addition, "Drechsler et. al. (2013) document “a strong divergence among banks’ take-up of” Lender-of-Last-Resort assistance “during the financial crisis in the euro area, as banks which borrowed heavily also used increasingly risky collateral”. They test several hypothesis and argue that their “results strongly support the riskshifting explanation”…"
The above supports the Uhlig (2013) model that concludes that:
-- "…Regulators in risky countries have an incentive to allow their banks to hold home risky bonds and risk defaults, while regulators in other “safe” countries will impose tighter regulation."
-- "…Governments in risky countries get to borrow more cheaply, effectively shifting the risk of some of the potential sovereign default losses on the common central bank."
-- "As a result, the monetary union has become a system engineered to deliver underpriced loans from country banks to their sovereigns, and to implicitly shift sovereign default risk onto the balance sheet of the ECB and the rest of the Eurosystem."
The last sentence is the key to it all: the euro system is now "engineered to deliver underpriced" credit "from country banks to their sovereigns", while shifting "sovereign default risk onto… the ECB and the rest of the Eurosystem".