13 November 2018

ECB: Government debt and banking fragility: the spreading of strategic uncertainty

This paper examines the conditions for the existence of the diabolic loop between banks and sovereigns. The analysis identifies three key components. First, the valuation of government debt reflects the probability of sovereign default. Second, banks hold government debt and their operations are affected by fluctuations in the price of this asset. Third, the government has an incentive to bailout banks which get into solvency problems.

Authors focus on the role of beliefs in generating the sovereign-bank loop. In this case, pessimism about government default reduces the value of government bonds. Because banks hold sovereign debt, this hurts banks balance sheets and threatens the solvency of the financial system. Because large-scale bank failures are very costly, this induces the government to bail out banks in order to prevent them from collapsing.

The government’s choice, however, increases the likelihood of a sovereign default. This is the diabolic loop in its purest form - the government’s bailout decision saves the banks from the immediate consequences of pessimism but also validates the initial pessimism itself.

Their mechanism relies on the fact that banks hold the debt of their own government and do not issue sufficient equity to protect their solvency from declines in the price of government debt. This is indeed the case in the data. Banks tend to hold (their own) government debt while treating it as a riskless asset when calculating how much bank capital they should maintain in order to protect depositors against loss.

The model has implications which are fully consistent with this feature of the data. Banks have no incentive to issue equity because they anticipate a government bailout if they get into trouble. In fact, they increase their holdings of risky government bonds because they expect to be compensated if losses occur.

Authors identify plausible conditions, related to the costs of sovereign default and a breakdown of the banking system, that lead governments to bailout banks in the event of insolvency thus validating the initial bailout expectations. Consequently, banks are incentivised to buy a lot of government bonds without issuing equity.

Importantly, the analysis makes clear that bailouts of domestic banks by national governments contribute to, rather than offset, the instability of debt markets. This is made explicit in their model as sovereign debt fragility arises due to a strategic complementarity between the buyers of government bonds, operating through government default, as in Calvo (1988). Since the government’s ability to repay debt depends inversely on the real interest rate it has to pay, this opens up the possibility of self-fulfilling pessimistic equilibria in which the high interest rate needed to compensate bond holders for high expected default risk weakens the government’s solvency and validates the pessimistic default expectations.

There are actions by banks and governments to break this vicious circle. On the banking side, equity cushions can break the adverse feedback between banks and sovereigns. Banks that hold adequate capital against potential sovereign risks can absorb losses from government default thus becoming completely insulated from developments in debt markets. However, when banks expect bailout assistance to be provided ex post, the incentive for them to self-insure by building up equity buffers against losses disappears. In this way, the analysis contains an important moral hazard problem: the incentive for banks to hold government debt without equity buffers is impacted by an anticipated bailout choice of the government. On the sovereign side, if the resolution of a collapsed financial system is not very costly, then governments will choose not to provide a bailout in our model, thus severing the diabolic loop. Further, a sovereign with the power to commit ex ante would choose not to do so, leaving it to the banks to protect their solvency through equity buffers.

Working paper

© ECB - European Central Bank