Director of the Centre for European Policy Studies
It has now been nearly half a year since the European Central Bank declared its intention to buy some 1.1 trillion (US$1.3 trillion) worth of eurozone bonds. When it first announced the so-called extended asset-purchase program in January, the ECB emphasized that it was only expanding an existing program, under which it had been buying modest quantities of private-sector bonds, to cover government paper. But this pretence of continuity was just that: a pretence.
In reality, by purchasing large volumes of government bonds, the ECB was crossing the Rubicon; after all, it is explicitly prohibited from financing governments. The ECBs defence was that that the program was the only way to move inflation closer to its target of approximately 2%. Moreover, it pointed out, it was merely following the example of other major central banks, including the Bank of England, the Bank of Japan, and especially the US Federal Reserve, whose program of quantitative easing (QE) entailed the purchase of more than US$2 trillion worth of long-term securities from 2008 to 2012.
Legal uncertainty aside, whether the ECBs decision to pursue QE can be justified ultimately depends on its impact. But, after six months, that impact remains difficult to assess.
One reason for this is that long-term interest rates are affected not just by the actual bond purchases, but also by financial markets expectations about future monetary policy. Indeed, just one day after the ECB made its announcement and weeks before the purchases began interest rates fell by a fraction of a percentage point throughout the eurozone.
When the purchases began, rates did continue to fall for a few weeks, so much so that many were concerned that there would not be enough German bonds to meet the ECBs country-debt quota (determined according to eurozone member states GDP and population). But rates have since risen again, and have now returned, in real terms, to pre-QE levels. In this sense, the ECBs bond-buying program has been a failure.
Other indicators, however, paint a different picture. Notably, price growth turned positive last month, suggesting that the threat of deflation has been eliminated. This has prompted a modest uptick in expected inflation the ECBs favourite measure of price stability not for the immediate future, but in five years, and then for five years.
In concrete terms, the ECB is measuring its policys success today according to the expected inflation rate in 2020-2025. This figure, calculated from the prices of different types of indexed and non-indexed five- and ten-year bonds, is based on the somewhat heroic assumption that all of the markets for these bonds work efficiently.
This presents a fundamental contradiction. QE is supposed to work via portfolio balance effects, which implies that markets are not fully efficient: purchases of longer-term bonds affect financial conditions by changing the types and quantity of financial assets the public holds. How can one use market prices as an indicator of inflation far into the future and simultaneously justify QE by claiming that most investors stick to certain asset classes and thus do not follow market signals efficiently?
Another problem with using the five-year, five-year-forward rate of expected inflation to gauge QEs effectiveness is that the rate is correlated with oil prices. Indeed, when oil prices fell last year, so did inflation expectations (measured however imperfectly). And the ECBs QE announcement coincided with the beginning of the oil-price recovery, which could be expected to increase inflation expectations somewhat.
With such coincidences and contradictions arising in nearly every aspect of the QE debate, it seems that evaluating the policys effectiveness is more of an art than a science. Unfortunately, this leaves plenty of room for distortion and bias.
Most glaring is QE supporters tendency to ascribe any decline in interest rates before the policy was announced to market participants expectations that QE would be coming. Yet they do not apply the same reasoning to the decline in inflation expectations that occurred during the same period. They have followed the same logic since the purchases began, ignoring recent increases in interest rates, while lauding the small rise in inflation expectations as proof of QEs effectiveness.
In reality, of course, QE whether the expectation or execution of it would affect both interest rates, and the inflation expectations that are derived from them. So, if ECB President Mario Draghi wants to highlight the fact that nominal interest rates are lower today than last August, he must also acknowledge that, given low inflation expectations, real interest rates have moved little.
But such pragmatic thinking has been sorely missing from discussions about QE. Instead, each side has caricatured the other: supporters emphasize that QE has nowhere led to runaway inflation, while opponents point out that nowhere has QE alone reignited robust growth.
In fact, there has been neither inflation nor growth: central banks can seemingly pour hundreds of billions of dollars, euros, or yen into the market with little discernible effect. So QE basically consists of an exchange of two low-yielding assets long-terms bonds and central-bank deposits. In broad and efficient markets, that exchange does not mean much.