Breaking the diabolical loop: regulating banks’ holdings of sovereign debt

Post with image

One of the defining features of the crisis that engulfed the eurozone at the beginning of this decade was the emergence of what became known as a “diabolical loop“.

The fate of Europe’s banks and sovereigns had become increasingly intertwined, creating fertile breeding ground for feedback effects to emerge where bad news about one spelt bad news for the other. This vicious cycle contributed to the amplification of stress during the acute phase of the European debt crisis and breaking this diabolical loop has therefore been a key objective of policymakers ever since. This article is about one of the key outstanding reforms that is required to sever that loop, and the progress towards implementing it.

There were lots of factors which contributed to the co-dependent relationship between European banks and their sovereigns. Investors believed that sovereigns stood behind their banks, so the credit worthiness of a bank was in part a function of the capacity of its sovereign to bail it out, and the expected likelihood and cost of future bank bail-outs weighed on the health of the public finances. Moreover, the banks and the sovereign both influence and are influenced by the state of the economic cycle, such that if either one attempts to repair its balance sheet – if banks lend less or governments save more – they will tend to depress economy activity, which will then impact on the balance sheet of the other. Our focus here is on an additional source of interdependence between the banks and the sovereign which played a prominent role in the crisis: the domestic banking system’s holdings of bonds issued by its sovereign.

From one perspective, the willingness of the domestic banks to invest in the debt issued by their sovereign can be seen as a positive. The banks can act as a buyer of last resort, helping to stabilise the market when transitory shocks lead to a buyer’s strike among overseas investors. One could also argue that ‘incentives’ which ‘encourage’ domestic financial institutions to hold the debt of their sovereign are one way to manage a public debt problem (this argument is usually referred to as ‘financial repression’). However, a system in which the banks are heavily exposed to their sovereign has a knife-edge property where if the shock is big enough then the implied losses on sovereign exposures can threaten the solvency of the banking sector, kick-starting the feedback loop discussed above. The banks come under pressure to take privately rational but potentially socially destructive defensive actions – fire-selling assets that are held by their peers, including perhaps some of their war-chest of sovereign bonds, and turning off the taps on new lending – and that drags the economy down.

By any reasonable yardstick, these portfolios of sovereign debt are large. In parts of the periphery sovereign holdings can account for 10% of the assets of the domestic banking system. If those investments were driven by fundamentals and funded in a prudent manner then these portfolios might not be such a concern. Unfortunately, that does not appear to be the case. There is a flaw in the regulatory regime under which the banks operate which favours holdings of sovereign debt relative to other assets which has contributed to the current position. At the current juncture banks may be able to fund their entire portfolio of sovereign bonds through debt (or put another way, the risk weight attached to these assets may be zero) and the usual large exposure limits which prevent banks from having too much exposure to any one credit do not apply. Banks can hold very large portfolios of sovereign debt with no capital on the other side of the balance sheet to absorb any losses. Indeed, it is possible to find banks in the periphery with exposures to their sovereign in excess of 200% of tier one capital.

This issue has been known about and discussed by the regulatory community for years. The answer seems clear: introduce sensible risk weights for sovereign bonds (that is, require banks to fund part of their portfolios through capital) and impose large exposure limits which would prevent the size of these portfolios getting out of hand. There are just two small snags. First, for certain banks in the periphery the adjustment could be painful. Just removing the zero risk weight exemption could reduce capital ratios in some quarters by over a percentage point – and typically where there is little buffer to absorb the hit. Second, and less immediately, there may be a concern about benching the buyer of last resort whilst the stock of public debt is still high, and governments will have to continue issuing large amounts of paper for the foreseeable future.

The consequences for pockets of the European banking system and to a lesser extent the sovereign bond market are thought by many investors to be so dire that policymakers will not be able to make any progress in this area. Comments by policymakers that appear to corroborate those concerns – such as those by the Governor of the Bank of Italy, Ignazio Visco, on 2 April 2016 – only serve to cement market expectations that no progress can be made on this front:

“The imposition of risk weights or, worse, tight concentration limits could be particularly disruptive for banks’ ability to act as shock absorbers in the event of sovereign stress. I doubt that further changes in prudential regulation are the right instrument for addressing the sovereign bank nexus. My personal view is that the potential benefits of a reform are uncertain, while the potential costs could be sizeable.”

We would caution jumping to the conclusion that the reforms will not be implemented. We know that this topic is on the agenda for Europe’s finance ministers. Whilst some politicians argue that reform would be particularly onerous for high-debt countries, others appear to consider progress on this front a necessary condition if they are to give ground on a European deposit insurance scheme. Moreover, it appears that the key decision-takers within the central bank and supervisory community are basically sold on making this change. The Chair of the Supervisory Board at the ECB, Danièle Nouy, is emphatic on this point:

“I think it is very simple. We learned through the crisis that there is no such thing as zero credit risk for assets – so we have to address this issue regarding sovereigns. We have to have capital requirements based on risk weights for sovereign exposures …. For me, it is not only an issue of capital requirements for the sovereigns, because most of the sovereigns are good quality assets. (…) There is a European rule regarding large exposures, which says that no single borrower can represent more than 25% of a bank’s net own funds. So let us apply this rule to sovereign bonds too.”

The official view appears to be that the change to a prudent treatment of sovereign risk needs to happen but it ought to be phased in gradually and that the change should be made at the global level rather than via a unilateral change by the European authorities. However, once it is clear that a transition will be made there is always the risk that the market will identify who stands to lose the most from the change and those institutions will come under immediate pressure. In effect, the market could compress the official transition process.

The only silver lining is the fact that the ECB is currently engaged in purchasing large quantities of government debt. At the very least, European banks can exit their positions at historically low yields. Banks would be well advised to take advantage of the window of opportunity that ECB quantitative easing affords. If they wait too long and that window has closed before they all start selling the same bonds to meet a demanding large exposure limit then the impact on market conditions – and ultimately their capital position – could be material. [divider] [/divider]

 

Ververidis Vasilis / Shutterstock.com

Richard Barwell

Head of Macro Research

Leave a reply

Your email adress will not be published. Required fields are marked*