We develop a bottom-up approach to estimating the slope of the primitive form of the New Keynesian Phillips curve, which features marginal cost as the relevant real activity variable. Using quarterly microdata on prices, costs, and output from the Belgian manufacturing sector, we estimate dynamic pass-through regressions that identify the degrees of nominal and real rigidities in price setting. Our estimates imply a high slope for the marginal cost-based Phillips curve, which contrasts with the low estimates of the conventional unemployment or output gap-based formulations in the literature. We reconcile the difference by demonstrating that, although the pass-through of marginal cost into inflation is substantial, the elasticity of marginal cost with respect to the output gap is low. We also illustrate the advantage of a marginal cost-based Phillips curve for characterizing the transmission of supply shocks to inflation.

Investors face reduced incentives to finance technological change that devalues their legacy investments. We formalize this “asset overhang” and apply our framework to the climate-banking nexus. Leveraging (1) firm-level data on green innovation & diffusion and (2) the sets of product & technology market peers, we implement a shift-share design that identifies banks’ credit facilities impacted by green firm activities. We find that green firms imposing an asset overhang across all lenders are 3 to 7 p.p. more likely to report tight credit supply conditions. The presence of legacy-free investors mitigates the asset overhang problem, thereby facilitating technological change.

Combining administrative records on firm-level output prices and quantities with quasi-experimental variation in credit supply, we explore the interconnection between the productivity and pricing effects of financial shocks. We show that a tightening of credit conditions affects firms’ technical productivity growth (TFPQ) in the long-run and but also induces firms to change their pricing policies. As a result, commonly adopted estimates that rely on revenue-based productivity measures (TFPR), which conflate the pricing and productivity effects, offer biased predictions regarding the effects of financial shocks on firms’ real productivity growth. Moreover, we document that the nominal adjustments triggered by these shocks have real implications. By adjusting prices, firms can leverage the product market to relieve pressure to cut expenditures on productivity-enhancing activities, thereby mitigating the contraction in future productivity growth.

We study cost-price dynamics in normal times and during inflation surges. Using microdata on firms’ prices and production costs we construct an empirical measure of price gaps—the deviation between a firm’s listed and optimal price. We then examine the mapping between gaps and price changes in the cross-section of firms and derive implications for inflation dynamics in the time-series. In the microdata, pricing policies display state-dependence: firms are more likely to adjust prices as their price gap widens, a mechanism that becomes quantitatively significant when large aggregate cost shocks occur. In normal times, adjustment probabilities are approximately constant and the microdata conform with the predictions of time-dependent models (e.g., Calvo 1983). Conditional on a path of aggregate cost shocks extracted from the data, we show that a generalized state-dependent pricing model accounts well for the pre-pandemic era’s low and stable inflation and the nonlinear surge observed during the pandemic.

This paper provides novel theory and evidence on the dynamic pricing behavior of firms, their process of belief formation, and the aggregate implications for monetary non-neutrality. We combine almost three decades of monthly microdata on prices, costs, and survey expectations for the Belgian manufacturing sector to quantify the roles of nominal, real, and information frictions in accounting for the incompleteness of the pass-through of cost shocks into prices. We find that the three rigidities are all quantitatively important, and in particular that firms exhibit a high average monthly discounting of about 0.8 attributable to the information friction. We further show that the discounting is state-dependent, just below one when firms are hit by large shocks and lower in normal times, consistent with a model in which firms update their beliefs faster in response to large disturbances. At the aggregate level, these findings imply a non-linear Phillips curve and a central role for heterogeneity across industries via the elasticity of inflation to aggregate shocks. Because the state-dependent information friction operates as an amplification mechanism of cost shocks, the model can explain a larger share of inflation volatility than the full-information benchmark, suggesting a quantitatively relevant channel by which incomplete information increases the neutrality of money.

In a production network, shocks originating in individual sectors do not remain confined to individual sectors but permeate through the pricing chain. The notion of “pipeline pressures” alludes to this cascade effect. In this paper we provide a structural definition of pipeline pressures to inflation and use Bayesian techniques to infer their presence from quarterly U.S. data. We document two insights. (i) Due to price stickiness along the supply chain, we show that pipeline pressures take time to materialize which renders them an important source of inflation persistence. (ii) As we trace their origins to 35 disaggregate sectors, pipeline pressures are documented to be a key source of headline/disaggregated inflation volatility. Finally, we contrast our results to the dynamic factor literature which has traditionally interpreted the comovement of price indices arising from pipeline pressures as aggregate shocks. Our results highlight the (often underappreciated) role of sectoral shocks – joint with the production architecture – to understand the micro origins of disaggregate/headline inflation persistence/volatility.

Bank lending shocks affect firm behaviour and permeate, via firm interactions, throughout the real the economy. In this paper, we explore how the production architecture of the real economy determines the aggregate real impact of shocks that originate from individual banks. Confidential data on the universe of (i) firm–to–firm transactions and (ii) bank—firm borrowing relations of the Belgian economy allows us to reconstruct (i) the firm–level input—output production architecture of the Belgian economy and (ii) the bank credit network supporting it. We use these objects to structurally document the extent to which individual banks support value creation in the economy either directly or indirectly through endogenous network acceleration mechanisms. We then study how the current structure of the Belgian economy affects the size of aggregate real GDP fluctuations induced by shocks from individual banks. Lastly, we revisit the Lucas (1976) argument – i.e. the rate at which this aggregate effect vanishes as the number of banks in the economy increases. We show that our analysis speaks to key research questions related to financial sector competition policy, strategic bank lending and selected macro prudential topics.

Any modern economy is characterized by an interlinked production architecture in which firms rely on each other for goods and services as inputs for production. Since bank credit is vital to support this production process, shocks to credit availability of individual firms propagate throughout this production network. In this paper we investigate the nature and extent of these general equilibrium effects using, inter alia, a novel and confidential VAT database which documents the universe of firm–to–firm transactions in the Belgian economy for over more than a decade. ML spatial econometric techniques are used to structurally trace the change in firm–level performance to (i) its own credit supply shock and (ii) credit supply shocks to other firms that work their way through the identified production architecture and trade credit network. We show that during 2003 − 2014 on average 53% (and during crisis periods up to 78%) of the aggregate real impact of bank credit supply shocks to individual firms arises through spillovers via upstream and downstream mechanisms. We identify endogenous responses in price setting and demand for supplier trade credit as the relevant drivers underlying the downstream and upstream propagation effects, respectively.