Image: libertystreeteconomics.newyorkfed.org · rights & removal

Firm Heterogeneity and the Response of Investment to Monetary Policy

Reporting by Liberty Street Economics (NY Fed)Read the original at libertystreeteconomics.newyorkfed.org

Executive Summary

Investment by most firms in most time periods responds little to changes in monetary policy. However, for some firms in specific periods, investment is very responsive to interest rate changes. The analysis of the distribution of investment responses (RIMP) shows that while not all firms react strongly, a small subset exhibits significant sensitivity. The variation in investment response is largely accounted for by permanent differences across firms and time variation within individual firms over time, suggesting that responsiveness tends to be transient rather than constant across the board.

Facts Only

* Investment by most firms in most time periods responds little to monetary policy changes.
* All firms still cut investment when interest rates increase.
* For about 5 percent of all firms and quarters, a one percentage point rise in the federal funds rate induces a four percentage point drop in capital growth rate.
* The variation in investment response is largely explained by permanent differences across firms and time variation within firms over time.
* Small and young firms exhibit the most negative RIMP, cutting investment the most when faced with monetary tightening.
* Firms with shorter debt maturity are more sensitive to monetary policy.
* Perceptions of discount rates and the cost of capital correlate with the RIMP.

Full Take

The analysis shifts focus from average effects to the distribution of responses, revealing that large-scale aggregation masks significant firm-specific dynamics. The finding that responsiveness is transient—a firm can be highly sensitive in one quarter but unresponsive afterward—suggests that monetary policy transmission is complex and path-dependent rather than a static relationship between macro variables and firm behavior. The fact that the most significant variation in investment response lies within firms over time, rather than across distinct groups, implies that idiosyncratic firm characteristics modulate the macroeconomic signal. The concurrent influence of smallness, youth, and debt structure suggests that policy effects are not uniform but are filtered through heterogeneous structural constraints, indicating a systemic gap between aggregate macro observations and micro-level reality. The unobserved factors explaining the residual variation underscore the need to move beyond simple correlation models to understand investment decisions fully.

From the original · Liberty Street Economics (NY Fed)

It is well known that monetary policy affects firms’ investment decisions. But which firms are the most responsive to changes in interest rates?

Read the full story at libertystreeteconomics.newyorkfed.org

Sentinel — provisional

Raw score, 0 to 1 (uncalibrated)0.10

No strong signs of machine writing were found in the source article. Provisional estimate, not a finding that a person wrote it.

The model's own note (unverified)

The text exhibits the structure and analytical depth of a human-authored economic research summary, focusing on methodological novelty rather than simple recitation of facts.

This looks only at the wording of the original source article, not at this page's AI-written sections. A small local AI model made this estimate. It has not been checked against known human and machine texts, so treat it as provisional. It cannot show who wrote an article.

What the model noticed in the source article

Moderate sentence length variance and use of nuanced academic phrasing; transitions are functional rather than purely mechanical.

Strong logical flow moving from the general question to the specific methodology (RIMP) and back to the implications, demonstrating sustained focus.

Consistent use of cited literature and clear articulation of empirical findings, typical of peer-reviewed synthesis.

Citations and technical descriptions appear grounded in established economic modeling language; no obvious markers of LLM confabulation.

Signs the model read as against machine writing

Specific, dense methodological description tied to empirical findings (e.g., 'clustering regression framework'), suggesting domain expertise beyond general synthesis.

Incorporation of specific, contemporary academic references and disclosure statements enhances authenticity.