投资结构变化与货币政策的威力（英）

N O .1 1 9 0MA R C H 2 0 2 6 Justin Bloesch|Jacob P. Weber Structural Changes in Investment andthe Waning Power of Monetary PolicyJustin BloeschandJacob P. WeberFederal Reserve Bank of New York Staff Reports, no.1190March2026 Abstract We argue that secular change in both the production and composition of investmentgoods has weakenedinvestment’s role in the transmission of monetary policy to laborearnings and consumption. We showanalytically that fluctuations in the production ofinvestment goods amplify the response of consumptionto monetary policy shocks byvarying labor income for hand-to-mouth agents. We document three secularchangesweakening this channel: (i) labor’s share of value added in investment goods productionhasdeclined, (ii) the import share of investment goods has risen, and (iii) thecomposition of investment has JEL classification:E21, E22, E32, E52, F41Keywords:monetary policy, investment, labor income, marginal propensity to consume Weber: Federal Reserve Bank of New York (email:jake.weber@ny.frb.org).Bloesch:Cornell University(email:jb2722@cornell.edu).The authors thankGabriel Chodorow-Reich,Emmanuel Farhi, EmiNakamura, Christina Romer, David Romer, Jón Steinsson, Ludwig Straub,and seminarparticipants atHarvard, UC Berkeley, and Washington University in St. Louis for valuable comments anddiscussion. This paper presents preliminary findings and is being distributed to economists and other interestedreaders solely to stimulate discussion and elicit comments. The views expressed in this paper are those ofthe author(s) and do not necessarily reflect theposition of the Federal Reserve Bank of New York or the 1Introduction Growing evidence suggests that monetary policy shocks have smaller effects on economicactivity now than in the past, even putting aside issues of an effective lower bound oninterest rates. Multiple authors, using various empirical techniques, report declining respon- This paper proposes a partial explanation: secular change in both the production andcomposition of investment goods has weakened private investment’s role in the transmissionof monetary policy to labor earnings and consumption.The importance of investment in driving consumption fluctuations in heterogeneous agent models where some householdshave high marginal propensities to consume (MPC’s) out of labor income has recently been We revisit this mechanism in a parsimonious, two-agent framework that links the con-sumption of hand-to-mouth agents to investment. We depart from the analyses of the pre-vious authors by studying an open economy environment, revealing an important role for imports. We show that the response of consumption by hand-to-mouth agents to changes inthe real interest rate depends on both (i) the responsiveness of investment and (ii) the ex- policy shocks (intellectual property products, such as R&D and software). Viewed throughthe lens of our calibrated model, these shifts imply large declines for the transmission of We begin by reviewing changes in the composition of investment and consumer durablessince 1947. The most notable change is the rise in “Intellectual Property Products” (IPP)which has grown from less than 1% of GDP around 1950 to over 5.5% of GDP by the begin-ning of 2024, now accounting for more than a fifth of nominal spending on investment anddurables. At the same time, investment in tangible goods such as equipment, durable goods, Next, we measure the domestic labor content of investment spending and its subcom-ponents using publicly available Input-Output (I-O) tables. The domestic labor content ofinvestment and durable goods has fallen from 58 cents on the dollar to 46 cents on the dollar To study the effects of these observed trends in a general equilibrium setting, we developa two-agent, five-sector, small open economy New Keynesian model.Households are splitbetween intertemporally optimizing “Ricardian” agents and hand-to-mouth agents. Produc-tion is partitioned between three domestic investment good sectors, a domestic consumption To derive our headline results, we calibrate the model separately to reflect a “1960s”U.S. economy and a “2020s” U.S. economy, matching observed changes in the domestic laborcontent of all final demand components and the composition of gross investment. We thencompute the cumulative 12-quarter response of labor income, consumption, and investmentto the same real interest rate shock under each calibration. This experiment suggests a 23% Lastly, we find that an increased trade share of the U.S. economy and the possibility of astronger exports channel of monetary policy does not offset our findings. Indeed, the model predicts that in response to an expansionary monetary policy shock, net exportsdeclinesin the short run—funded by net borrowing vis a vis the rest of the world—and later turnspositive. This is because rising demand for imports immediately following the shock—driven The remainder of the paper is organized as follows. Sec