Two recent empirical studies highlight the risk that inflation expectations in the euro area are becoming de-anchored, similar to Japan. De-anchoring means that short-term price shocks can change long-term expectations. Importantly, the papers suggest medium- and short-term measures to track this de-anchoring. De-anchoring increases the risk of actual deflation and may add to the risk premia on equity and credit.
Buono, Ines and Sara Formai (2016), “The evolution of the anchoring of inflation expectations”, Banca D’Italia, Occasional Papers No. 321, March 2016.
Galatia, Gabriele, Zion Gorgi, Richhild Moessner and Chen Zhou (2016), “Deflation risk in the euro area and central bank credibility”, De Nederlandsche Bank, Working Paper No. 509, April 2016.
This post ties in with our summary of non-conventional monetary policies and their role as essential tools for containing deflationary dynamics. The post also provides measures for tracking macroeconomic trends, particularly those related to broad economic uncertainty.
The below are excerpts from the papers. the first two sections are based on the Bank of Italy paper, the third on the DNB paper. Headings, links and cursive text have been added.
Why de-anchoring of inflation expectations is a great risk
“Anchoring inflation expectations is a cornerstone of central banks’ monetary policy strategy. Once incorporated into agents’ wage- and price-setting decisions, expectations drive actual inflation…Inflation expectations are firmly anchored if long-run expectations are stable over time, exhibit little cross-sectional dispersion, and are insensitive to macroeconomic news.”
“Recent evidence…has shifted the debate towards the risks entailed by actual inflation rates being persistently below the central banks’ target. The resulting low inflation expectations may lead to destabilizing dynamics and self-fulfilling liquidity traps. This is a concern especially when monetary policy is constrained by the zero lower bound: lower inflation expectations result in higher real interest rates, causing a monetary tightening that central banks are unable to counter by lowering nominal rates.”
“The de-anchoring of inflation expectations…may explain why the Japanese authorities, stuck in a liquidity trap, were unsuccessful in their battle against deflation.”
Survey-based evidence of de-anchoring of inflation expectations
“We find that starting from the second half of 2008, inflation expectations in the euro area, unlike in the US and in the UK, have shown signs of a de-anchoring.”
“[The Bank of Italy paper] analyzes the anchoring of inflation expectations for the four largest advanced economies [the euro area, U.S., Japan and the UK] from 1990 to March 2015…Our empirical analysis is based on semi-annual data on expected inflation, taken from the Consensus Economics surveys released in April and October of each year…We study the inflation expectation pass-through, defined as the link between short- and long-term inflation expectations, given respectively by one-year-ahead and five-years-ahead expectations. If expectations are well anchored, then changes in short-run expectations should not be correlated with changes at longer horizons…The change in short-run inflation expectations is a proxy for macroeconomic news relevant for inflation.”
“Inflation expectation pass-through is quite different across economies, at all horizons. For Japan the co-movement is always positive and statistically significant. A 1% increase in one-year-ahead inflation expectations is associated with an increase of 0.49% for two-years-ahead inflation expectations and of 0.27% for five-years-ahead expectations.”
“We investigate whether the intensity of the inflation pass-through has changed over time in different economies…In the euro area the co-movement of expectations is significantly different from zero between 1990 and 2000…The co-movement then decreases starting from 1996, and becomes null in 2001, highlighting the key role played by the adoption of the common currency and the credibility of the ECB in keeping inflation expectations well anchored. The pass-through becomes again significantly different from zero starting from the second half of 2008. At the end of the sample, a 1 percentage point decrease in one-year-ahead is estimated to trigger a 0.2 percentage point decrease in long-run expectations: therefore, the reduction of one-year-ahead inflation expectations observed from the second half of 2012 implies a reduction by 0.3percentage points of long-run expectations.”
“The picture appears different for the US, where the co-movement, never statistically different from zero, decreases from 2007 and remains fairly constant in the last part of the sample… Expectations in the US and in the UK appear firmly anchored, particularly in more recent years.”
Options-based evidence of de-anchoring of inflation expectations
“[The Dutch National Bank] paper investigates how the perceived risk that the euro area will experience deflation has evolved over time, and what this risk implies for the credibility of the ECB. We use a novel dataset on market participants’ perceptions of short- to long-term deflation risk implied by year-on-year options on forward inflation swaps… Our analysis suggests that the anchoring properties of euro area inflation expectations have weakened, albeit in a still subtle way.”
“Although a relatively new market, the inflation options market has been evolving rapidly since the onset of the financial crisis… The demand for inflation caps and floors is largely driven by investors with future cash flows that are exposed to inflation risk… The empirical analysis is based on option-implied probabilities that were provided to us by the Royal Bank of Scotland.”
“All in all, there is a clear ongoing trend in the distribution of inflation expectations by professional forecasters and market participants, which has become increasingly skewed towards a scenario of low and possibly negative inflation over the medium-term.”
“We investigate whether long-term inflation expectations have become de-anchored, by studying whether long-term deflation risk has been affected by changes in oil prices and by short-term deflation risk… If the central bank is fully credible, they should not.”
“We find that [from 2010 to 2015] short-term deflation risk (2 years ahead) reacts significantly and most strongly to oil prices. Medium-term deflation risk (3, 5, and 7 years ahead) also reacts significantly to oil prices, but less strongly than at shorter horizons, while long-term deflation risk (10-years ahead) does not react…However, some reaction of deflation risk at the 7-year horizons may give an indication that inflation expectations have not remained strongly anchored.”
“We study whether the anchoring of inflation expectations has changed over time, by estimating the reactions of deflation risk to oil price changes separately for the first half of the sample period (January 2010 to December 2012) and the second half (2013 to December 2015)…In the first half of the sample, only short-term deflation risk (2 and 3 years ahead) reacted significantly to oil prices, while medium and long-term deflation risk (5, 7 and 10 years ahead) was not significantly affected. By contrast, in the second half of the sample, deflation risk 5 and 7 years ahead reacted significantly to oil prices.”