Fed study: Households ignore interest rate changes

A new study by Edmund Crawley and William L Gamber at the Federal Reserve Board concludes that household spending shows no direct response to changes in interest rates, effectively eliminating a direct interest rate channel for monetary policy.

This finding is derived from a Bayesian procedure applied to a Heterogeneous-Agent New Keynesian (HANK) model, disciplined by empirical impulse responses to ten distinct macroeconomic shocks.

The researchers employed two primary estimation approaches: a structural model incorporating sticky expectations for both income and interest rates, and a non-parametric estimation of the consumption-to-interest-rate Jacobian.

Both methods consistently indicate that households do not adjust their consumption decisions based on interest rate fluctuations, even at varying time horizons.

This suggests that any impact of monetary policy on household spending operates entirely through its effects on labor income, such as employment and wages, rather than direct intertemporal substitution.

The paper's findings significantly challenge the traditional New Keynesian framework, which posits that interest rate changes directly influence households' intertemporal consumption decisions.

In contrast, more recent HANK models emphasize an indirect channel, where monetary policy affects consumption by altering labor income through firms' production and hiring.

The study's conclusion that the direct channel is non-existent reinforces the importance of this indirect mechanism.

Furthermore, this result has direct implications for the 'forward guidance puzzle,' where announcements about future interest rates generate implausibly large immediate effects in standard models.

If households are unresponsive to future interest rates, this puzzle largely disappears, simplifying theoretical explanations.