New York Fed Research Finds Most Firms Barely Change Investment After Rate Shocks, but a Small Minority React Sharply

Most companies make only modest changes to investment after unexpected shifts in U.S. monetary policy, but a relatively small share of firm-quarter observations shows a much larger response, according to new research published by the Federal Reserve Bank of New York.

The researchers estimate that about 5% of firm-time observations fall into the strongly responsive part of the distribution. For those observations, a monetary-policy surprise equivalent to a 1-percentage-point increase in the federal funds rate is associated with roughly a 4-percentage-point decline in the growth rate of capital. The finding does not mean that a fixed 5% group of companies always reacts sharply. Much of the variation appears within the same firms over time.

The October 7 Liberty Street Economics analysis, by Thomas Drechsel, Daniel Lewis, Davide Melcangi and Laura Pilossoph, is based on Federal Reserve Bank of New York Staff Report No. 1210. The authors set out to measure the full distribution of firms’ investment responses to monetary policy rather than sorting companies in advance by a single characteristic such as size, age or leverage.

Most firm-quarter responses are small, but the tail is much larger

The study measures monetary-policy surprises using changes in the federal funds rate implied by federal funds futures in a narrow window around policy announcements. Firm investment is measured as the quarterly growth rate of capital stock for publicly listed companies in Compustat. The researchers then use a clustering-regression framework to group each firm in each period according to whether its investment response is relatively high or low.

The resulting distribution is heavily concentrated around modest reactions. The authors report that most firms in most quarters cut investment only slightly when monetary policy tightens. The estimated responses are generally negative, so the result is not that the majority of companies are completely insulated from higher rates. Instead, the average-looking response conceals a small set of firm-quarter observations in which capital spending reacts much more strongly.

At the more sensitive end of the distribution, the estimated effect becomes economically large. For about 5% of firms and quarters, a surprise that raises the federal funds rate by 1 percentage point is associated with a 4-percentage-point drop in capital-growth investment. That contrast is central to the paper because a single average response can obscure the size of the reaction among the most sensitive observations.

The authors’ approach differs from studies that begin with a particular firm trait and ask whether that characteristic explains monetary-policy sensitivity. By first estimating the distribution and then examining what predicts the stronger responses, the paper is designed to capture interacting observable characteristics as well as differences that cannot be readily observed in standard firm data.

Rate sensitivity appears to move within firms over time

One of the study’s more important findings is that the strongest responses are not concentrated in a stable set of companies. The researchers decompose variation in firms’ investment sensitivity into permanent differences across firms, aggregate changes over time and firm-specific changes over time. They find that relatively little of the variation is common to all firms and that permanent cross-firm differences explain only a limited share.

Most of the variation instead occurs within firms over time. A company may react strongly to a monetary-policy surprise in one quarter and then show a much smaller response in later periods. That pattern makes the 5% figure different from a statement that 5% of companies are structurally rate-sensitive. It describes observations tied to particular firms at particular times.

The result also complicates attempts to identify rate-sensitive companies using a single balance-sheet characteristic. Firm traits matter, but they do not provide a complete explanation for why investment becomes unusually responsive in a particular quarter. The paper’s framework is intended to capture that shifting sensitivity rather than force each company into one permanent category.

This distinction matters when interpreting how monetary policy reaches the real economy. An average investment response may look modest even when a smaller group of firms is making much larger adjustments at the same time. The New York Fed researchers argue that the full distribution therefore contains information that is lost when analysis focuses only on the mean effect.

Smaller and younger firms tend to react more, but much remains unexplained

After estimating the distribution, the authors examine characteristics associated with stronger reactions. Smaller firms and younger firms tend to cut investment more after a monetary tightening, and the two characteristics interact: firms that are both small and young show particularly negative responses in the authors’ estimates.

Debt structure also matters. The staff report finds that companies with shorter debt maturities tend to be more sensitive to monetary policy. That result is consistent with the idea that firms facing nearer-term refinancing needs may feel changes in financing conditions more quickly than companies whose debt is locked in for longer periods.

The researchers also identify correlations involving firms’ perceptions of discount rates and the cost of capital. Companies reporting greater optimism tend to reduce investment less after an interest-rate increase. These relationships do not provide a single causal explanation for the pattern, however, and the authors emphasize that firm characteristics can operate jointly rather than in isolation.

A large portion of the variation remains unexplained by the observable characteristics examined in the study. That is one reason the authors frame their contribution around heterogeneity itself rather than around one preferred mechanism. Their estimates provide evidence that investment responses can differ substantially across firms and across time even after accounting for traits that previous research has linked to monetary-policy transmission.

The research does not imply that the Federal Reserve should target particular companies or that the strongest estimated response applies to every tightening cycle. The Liberty Street Economics post states that the views are those of the authors and do not necessarily represent the position of the Federal Reserve Bank of New York or the Federal Reserve System. Its narrower conclusion is empirical: most firm investment responses to monetary-policy surprises are modest, a small share are much larger, and much of that sensitivity changes within firms over time.