To save this page as a PDF, click this button and choose the PDF destination.

G2.10: Interest rates

10:30 - 11:30 Sunday, 24th June, 2018

Wellman 115 (Ground floor)

Track G2: Financial

[No author data]

643 Population Aging and the Real Interest Rate in the Last and Next 50 Years -- A tale told by an Overlapping Generations Model –-

Nao Sudo, Yasutaka Takizuka
Bank of Japan, Tokyo, Japan

Abstract

Population aging, along with a secular decline in real interest rates, is an empirical regularity observed in developed countries over the last few decades. Under the premise that population aging will deepen further in coming years, some studies predict that real interest rates will continue to be depressed further to a level below zero. In the present paper, we address this issue and explore how changes in demographic structures have affected and will affect real interest rates, using an overlapping generations model calibrated to Japan's economy. We find that the demographic changes over the last 50 years reduced the real interest rate. About 270 out of the 640 basis points decline in real interest rates during this period was attributed to declining labor inputs and higher saving, which themselves stemmed from the lower fertility rate and increased life expectancy. As for the next 50 years, we find that demographic changes alone will not substantially increase or decrease the real interest rate from the current level. These changes reflect the fact that the size of demographic changes in years ahead will be minimal, but that downward pressure arising from the past demographic changes continue to bite in the years ahead. As Japan is not unique in terms of this broad picture of changes in demographic landscapes over the last 50 years and in the next 50 years, our results suggest that, sooner or later, a demography-induced decline in real interest rates may be contained in other developed countries as well.


885 Trend and uncertainty in the long-term real interest rate: Bayesian exponential tilting with survey data

Taeyoung Doh
Federal Reserve Bank of Kansas City, Kansas City, USA

Abstract

Long-term interest rates have remained low in the U.S. since the Great Recession of 2007-9 with slow growth and low inflation compared to past cycles. This episode renewed interest in the secular stagnation hypothesis that tightly links the permanent downward shift in the real interest rate with the decline in growth and inflation. Some argued that this decline was fairly predictable at the 1980s because it was mainly driven by demographic changes. However, the uncertainty surrounding changes in the trend real interest rate varies a lot among different studies, making it difficult to reach consensus on this argument. In this paper, I set-up a flexible time-series model to examine the time-variation of both level and uncertainty of the trend long-term real interest rate. While the level of the decline in the median estimate of the trend long-term real interest rate is mostly in line with previous literature, the associated uncertainty is much higher. In addition, uncertainty is not symmetric with downside (upside) risk more prominent during certain periods. When model-implied forecasts are tilted to match both means and cross-sectional dispersions of forecasts from the Survey of Professional Forecasters (SPF), the downside tail risks for growth and inflation significantly decrease during the period of 2007-2016. Overall, empirical findings in this paper cast doubt on the particular version of the secular stagnation hypothesis that predictable factors played a large role in the decline in the trend real interest rate.