Why Europe’s QE resembles a CDS trade

The European Central Bank's (ECB) €1 trillion quantitative easing (QE) programme is commonly seen as an extra-ordinary effort to reflate the eurozone economy and unleash a recovery. It can also be seen as a smart risk management strategy for the core countries of the eurozone. The programme resembles a series of credit derivatives, in which the protection sellers are the participating national central banks. One of the conclusions of this analysis is that the protection sellers are not being appropriately compensated.

This requires a little explanation. First, it is important to recognise the ECB is not following the best practices for QE established by other central banks, such as the US Federal Reserve and Bank of Japan. Bonds are bought for the Public Sector Purchase Programme (PSPP) via the secondary market, meaning current market value determines the purchase price. Moreover, the rebates of interest payments by the ECB and national central banks to local eurozone governments do not necessarily reflect the spreads on sovereign bonds. Finally, the ECB does not fully accept the credit risk implicit in the PSPP.

These decisions are political and are mainly rooted in the Germanic theory that a more traditional approach to QE encourages moral hazard. If a member country does not have to pay interest on bonds purchased under the programme, it could use the cash to reduce taxes or increase spending – or so the argument goes. By the same token, the purchases have to be made directly by the national central banks to limit the risk of a scenario where a member country defaults on its public debt, and the losses are borne by eurozone taxpayers.

But the moral hazard theory is not supported by the facts. The ECB's stance on interest payments certainly has fiscal policy implications, but for different reasons.

The ECB purchased around €220 billion of bonds issued by eurozone peripheral countries under its Securities Market Programme between 2010 and 2012. In total, an estimated €10 billion of interest payments have been distributed to Eurosystem countries on a pro-rata basis – 18% to Germany, 14% to France, 12% to Italy, and so on. In other words, Germany has received nearly €2 billion of interest from the periphery states. As feared, ECB monetary policy has led to fiscal transfers between members – but from the periphery to the core, not vice versa.

The national central banks are in fact selling a CDS to the ECB on Greek sovereign risk without receiving any compensation

The largest share of sovereign bond purchases under the PSPP will be conducted directly by the national central banks: 80% of the full programme, representing about €800 billion of asset purchases. The purchasers will record these bonds as assets and the debt owed to the ECB as liabilities. As such, national central banks are effectively guaranteeing the value of the public debt of member states participating in the programme.

If a member country was to leave the euro and devalue the public debt purchased under the PSPP by denominating it in the new national currency, it would still be required to repay the full value in euros to the ECB.

This is where the derivatives analogy comes in – national central banks are effectively selling a credit default swap to the ECB, with the interest payments on the bonds taking the place of premiums. But the de facto premiums paid to date are well below those seen in the actual CDS market in recent years.

The ECB's decision to accept only a minimum amount of sovereign risk-sharing – equivalent to only 8% of the PSPP – has a similar impact: hedging the core economies against the risks of the peripheral countries and ring fencing within each member state the risk of its own public debt.

The final 12% of the PSPP entails the purchase of bonds issued by European supranational institutions, including the European Financial Stability Facility (EFSF), which has significant exposure to Greece. Officially, these purchases are part of a risk-sharing regime, but once again national central banks conduct the purchases. Any losses will be borne by participating member states on a pro rata basis, according to their share of the EFSF – so 27% by Germany, 20% by France, 18% by Italy, and so on.

Viewed through the lens of credit derivatives, through the purchases of the EFSF's bonds, the national central banks are in fact selling a CDS to the ECB on Greek sovereign risk without receiving any compensation, given the negligible yields on EFSF bonds – 0.4% in late June for bonds maturing in five years. In other words, the risk of a Greek debt restructuring and the consequent need for recapitalisation of the EFSF by the member states is reinsured through this part of the PSPP.

This is not only a matter of risk mitigation. The ECB's QE also creates room for private bank recapitalisation. These banks bought sovereign bonds when yields were higher and prices were lower. It follows that these same bonds were sold to national central banks when prices were nearing record highs. To date, I estimate this has generated capital of €13 billion for private banks.

The unconventional approach of the ECB is therefore aiding the banking system – presumably under the assumption that by saving the banks, the eurozone's financial system is also saved. This policy works in conjunction with the extraordinary long-term refinancing operations started at the end of 2011 – €1 trillion of "easy money" that private banks of the peripheral member states used to nationalise the public debt sold to them by the core countries to repay loans from German banks.

The design of the ECB's QE programme is therefore not financially neutral, and does not lead to gifts to peripheral countries. In fact, it continues the risk nationalisation process – a political compromise that is strengthening centrifugal forces within the eurozone at the cost of serious unintended consequences.

Marcello Minenna is professor of financial mathematics at Bocconi University and head of the quantitative analysis unit at Italy's markets regulator, Consob. The opinions expressed in the article are those of the author and do not necessarily reflect the views of these institutions. Marcello can be contacted at: marcello@minenna.it