Japanese Inflation Expectations, Revisited
Benjamin R. Mandel and Geoffrey Barnes
Liberty Street Economics April 22, 2013
An important measure of success for monetary policy is a central bank’s ability to anchor inflation expectations; inflation expectations influence actual inflation and, hence, the achievement of a given inflation goal. This notion has special significance for Japan, where CPI inflation has been intermittently negative since 1994 and where it is widely believed that expectationsof future inflation have been persistently negative (that is, ongoing deflation is expected). In this post, we describe and evaluate an alternative, market-based measure of Japanese inflation expectations based on international price parity conditions. We find that recent inflation expectations have attained a level substantially higher than their previous peaks over the past three years.
By way of background, recent policy action by the Bank of Japan has shone a spotlight on Japanese inflation expectations. On April 4, the Bank announced a program called Quantitative and Qualitative Monetary Easing (QQE), which was a pledge to drastically ramp up asset purchases to increase the monetary base, and to extend the duration of assets held on the Bank’s balance sheet. Since nominal yields on Japanese government bonds have been quite low for some time, a preferred indicator of QQE’s success would be a decline in real interest rates as inflation expectations move closer to the Bank’s recently announced 2 percent price stability target.
How does one go about measuring Japanese inflation expectations? The consensus on this topic is that there is no single reliable measure. A commonly used market-based gauge of U.S. inflation expectations is the difference in yield between nominal and Treasury inflation-protected securities (TIPS)—the breakeven inflation rate. Analogous measures come from over-the-counter derivatives called inflation swaps. In Japan, the market for inflation-protected government bonds, called JGBi’s, is very thinly traded and a majority of the issuance has been bought back by the Ministry of Finance in recent years. These factors have cast doubt on the ability of JGBi prices to convey reliable information about inflation expectations. Swaps suffer from similar liquidity issues.
Alternative extant measures of inflation expectations are available from surveys of households, investors, and professional forecasters. However, survey responses may by formed in a backward-looking manner, making them more responsive to actual inflation than predictive of the future. The range of views offered by market‑ and survey-based measures is illustrated in the chart below. While measures of five- and ten-year expectations have converged somewhere around 1 percent in recent months, in the past analysts would have little confidence of even getting the correct sign of expected inflation by looking at any given measure.
A Measure Based on Purchasing Power Parity
Given these concerns, we consider an additional market-based measure derived from U.S. inflation expectations—for which there are more actively traded inflation‑protected securities and swaps markets—and international price parity conditions. To our knowledge, these tools are not commonly used to make inferences about Japanese inflation expectations, but may provide a useful alternative to Japanese JGBi’s and swaps. One exception is a report by Goldman Sachs Economics Research (“The Market Consequences of Exiting Japan’s Liquidity Trap,” Global Economics Weekly 13/05, February 2013), which uses the thirty-year yen/dollar forward rate to infer Japanese inflation expectations.
The measure relies on purchasing power parity (PPP), which equates the price level in one country to the price level in a second country and the two countries’ nominal exchange rate. PPP has been shown to work relatively well over longer periods as well as in relative terms; that is, it works better in changes than in levels. In the case of Japan, PPP implies that the change in expected future Japanese prices (a close analogue to inflation expectations) is equal to the change in expected future U.S. prices plus the change in the expected future value of the yen. We implement this measure using the corresponding breakeven rate for U.S. inflation implied by TIPS and the yen/dollar forward exchange rate.
The chart below uses daily data and shows the resulting PPP-implied Japanese inflation expectations for five-, seven-, and ten-year horizons since January 2010. Note that the timing of fluctuations in expectations suggests that they are related to policy actions, since each peak over the past three years has followed a major policy innovation. In October 2010, the Bank of Japan’s introduction of the Asset Purchase Program spurred a rise in expectations, which was then undone by mid-2011. In February 2012, the Bank introduced a 1 percent price stability goal, which prompted another, less pronounced, change in inflation expectations that was again undone after a few months. Most recently, inflation expectations have increased following the election of Shinzo Abe as leader of Japan’s Liberal Democratic Party in September 2012 and subsequently as prime minister in December, marking the beginning of a policy regime commonly referred to as “Abenomics.” By our measure, post-Abenomics inflation expectations have attained a level substantially higher than the previous two peaks.
A Check of the Methodology
We can check the robustness of the PPP-implied measure by applying the same logic to different country pairs. If the movements in our measure of inflation expectations are not driven by idiosyncrasies in U.S.-Japan financial markets, then substituting another nation’s forward exchange rate and breakeven inflation rate for the respective U.S. rates should yield similar results. The United Kingdom is a natural candidate for the role of “other nation” here, given its relatively liquid inflation-protected security markets. Hence, we use the pound/yen forward rate and U.K. breakeven inflation rates to produce the PPP-implied measure of Japanese inflation expectations shown in the chart below (plotted alongside the analogous U.S.-based measure from above). While not always perfectly aligned in levels, the two series have been highly correlated—with a correlation coefficient of 0.66–in their day-to-day movements since 2010.
In a similar test of the methodology, we compute a measure of U.S. inflation expectations as implied by U.K. inflation expectations and the pound/dollar forward rate. The chart below plots the resulting implied ten-year U.S. inflation expectations against actual U.S. breakeven inflation rates. With the exception of a few days in late 2012 when the series diverged and then converged again, the two measures are again closely correlated at a daily frequency—with a correlation coefficient of 0.64. These findings suggest that PPP provides a good approximation to U.S. TIPS.
In summary, PPP provides an alternative, market-based view of Japanese inflation expectations, which appear to have been quite responsive to recent monetary policy innovations by the Bank of Japan. Moreover, the similarity of the measures across time periods (five-, seven-, and ten-year) and country pairs (U.S.-Japan, U.K.-Japan, and U.S.-U.K.) gives some comfort that PPP implies a robust reading of inflation expectations more generally.
The views expressed in this post are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.
Benjamin R. Mandel is an economist in the Federal Reserve Bank of New York’s Research and Statistics Group.
Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.