Over the last decade, central banks have faced economic crises that have required unusual responses. For example, the Bank of England (BoE) cut policy rates to extremely low levels; the US Federal Reserve (Fed) reduced them to zero per cent; and in the euro zone, Japan and Switzerland, official rates were even driven below zero. But that didn’t appear to be enough. Considerable amounts of cash were injected into the economy through either full allotments on commercial banks’ refinancing operations at the central bank, or through the purchase of financial assets (commercial paper, corporate and government bonds, and even shares through ETFs by the Bank of Japan (BoJ)).
As a result, the Fed now holds more than 15% of US government debt, equivalent to 13% of GDP, the BoE owns 23% of the gilt market and the European Central Bank (ECB) owns the equivalent of 18% of eurozone GDP in securities thanks to its asset purchase programme. The BoJ had the largest programme and now holds 76% of GDP equivalent in securities, mainly government bonds. Given the colossal amounts involved, and the widespread use of these stimulus policies, some believed quantitative easing (QE) had become irreversible.
Impact of quantitative easing
These security purchase policies have had a direct impact on interest rates, causing them to fall sharply, but indirectly, they also improved conditions for issuers. Since central banks purchase this debt on a hold-to-maturity basis (without any provisions for unrealised capital losses at the end of the year) and finance them at the deposit rate (currently set at -0.4% for the eurozone), they can generate a non-negligible profit, with the exception of the Bundesbank. This remains true as long as the interest rate at purchase remains higher than the deposit rate. Any profit is remitted to the state in the form of dividends from the central bank. Taking the case of Italy, the Bank of Italy buys EUR 9 billion of government bonds (BTPs) every month and has been keeping the paper on its books. To date, the gain for the Italian Treasury has amounted to the equivalent of 0.25% of GDP.
Extrapolating this, and assuming that the deposit rate at the ECB will remain low for a long time, Italy’s debt-to-GDP ratio can be expected to drop from 133% to less than 120%. Thus, QE indirectly facilitates slow deleveraging even as the direct impact has lowered yields through monetisation.
The Fed has now signalled it will stop reinvesting repayments from maturing government and mortgage-backed securities (MBS) (without specifying the date) and the ECB is preparing the financial markets for the end of its purchasing programme. Consequently, the System Open Market Account (SOMA) of the US Central Bank, where the state securities are held, could drop by more than 30% in the first three years. The Fed thus shows that its purchasing programme is reversible and that the economy is not addicted to asset purchases. Accordingly, QE becomes a normal policy instrument that can be used again later. The success of this approach will require, however, that the economy does not weaken dramatically in the future.
Meanwhile, the ECB has been sending messages to the markets that it would like to begin unwinding its bond purchase programme early next year. Given what the markets have already seen from the Fed, investors should start to anticipate the reversibility of the ECB’s programme, which is comparable in size to that of the Fed.
Japan is a different case. The BoJ’s asset purchases amount to a much larger percentage of GDP and the size of the government’s debt stood at 234% of GDP at the end of 2016. Reversibility does not seem easy. At the annual meeting of central bankers in the Portuguese city of Sintra, the BoJ’s representative said that the end of central bank security purchases was still some way off.
Market expectations of an end to the ECB’s purchasing programme, initially led to a rise in interest rates across the entire curve. We expect this rise to continue in a slow and uneven way. The Fed had pre-announced its tapering in May and June 2013, making an official announcement in December 2013 and actually starting in January 2014. The ECB’s timetable could be similar. The rise in euro rates, however, should be less given the US experience and the extended period over which the Fed has been raising its key rates. Nevertheless, the yield curve in Germany is less protected than the US curve was in 2013. Consequently, we expect the euro curve to steepen.
As far as BBB-rated eurozone countries are concerned, a reduction in quantitative easing is likely to widen yield spreads over core member state bonds as it reduces debt sustainability. However, this effect could be offset by reasonable economic growth, as long as it is sustained.
Swap spreads can be expected to tighten with the disappearance of a major buyer – the ECB – and ‘proxy swaps’ such as covered bonds and agency debt are likewise expected to see narrowing spreads. Indeed, the large investors in agency debt with maturities of less than seven years, namely banks managing their liquidity coverage ratio, deal primarily with swaps. The addition of quantitative easing to the panoply of conventional monetary policy measures should reassure the markets, while increasing moral hazard” instead of “but at the same time, it increases moral hazard.Download to read more