< Back to previous page

Project

The Impact of Credit Frictions on the Dynamics of International Trade.

Contemporaneous macroeonomic models invariantly rely on economywide shocks to explain aggregate fluctuations. The role of microeconomic shocks (e.g. to individual firms) has generally been overlooked. The relevance of small shocks for the macroeconomy was downplayed by Robert Lucas’ (1977) renowned essay on real business cycles. His argument is simple but powerful: if shocks that affect individual firms are idiosyncratic (i.e. random), these shocks balance out when we aggregate across all firms in order to study the aggregate economy. His conjecture relies on a “law of large numbers”-type of argument and has inspired decades of macroeconomic modeling. Despite this dismissal, the role of micro—level shocks has recently attracted theoretical attention.

First, a shock to an individual firm has a (much) larger impact on the aggregate economy if it reduces the output of not only this firm, but also that of others firms with which it interacts through input/output interactions. Hence, input/output linkages can neutralize the law of large numbers because shocks that affect firms that are particularly important in the production architecture of the economy will not balance out with that of other firms. Their shocks will then resurface in the aggregate. In a similar vein, when the distribution of firm size has fat tails (implying that shocks hitting the larger firms will not be off-set by shocks affecting smaller firms) the law of large numbers need not apply. Both notions of “size” and “interconnectedness” open the way for sizeable macroeconomic fluctuations originating from individual firms. This thesis takes stock of this insight and contributes in various dimensions to this literature.

Chapter 1: Pipeline Pressures and Sectoral Inflation Dynamics.
(i.s.m. Frank Smets & Jan Van Hove)

Chapter 1 is built around the notion of “pipeline pressures”, a term popularized by recent policy work (e.g. by the European Central Bank, the Federal Reserve System, etc.) and the popular press (e.g., the Wall Street Journal, the Financial Times, etc.). The principle underlying pipeline pressures is simple; inflationary shocks originating in specific sectors do not remain confined to individual sectors but permeate through the pricing chain. Not surprisingly, recent developments in individual sectors are scrutinized by policymakers, professional forecasters and central bankers in order to assess their impact on headline inflation. E.g., (i) shifts in healthcare sector regulation (e.g. Affordable Care Act) (ii) competition in the telecommunications sector, (iii) productivity shocks in the computer and electronics sector, (iv) the shale gas boom in the mining sector, (v) Dieselgate, etc.

Chapter 1 provides a structural definition of pipeline pressures and uses Bayesian techniques to infer their presence from quarterly U.S. data. We document two insights. (i) Due to price stickiness along the supply chain, that pipeline pressures which originate from individual sectors take time to fully permeate throughout the economy. Hence, sectoral shocks can be 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 role of sectoral shocks - joint with the production architecture - to understand the micro origins of disaggregate/headline inflation persistence/volatility.

Chapter 2: Credit Supply Shock Propagation and Amplification in the Real Economy: Firm-level evidence.
(i.s.m. Hans Dewachter & Jan Van Hove)

Chapter 2 is motivated by the recent financial crisis and ensuing recession which are characterized by large drops in credit supply to the real economy. A large body of empirical work has leveraged this economic episode to document how variation in credit supply affects firm–level (i) investment, (ii) exports, (iii) employment, (iv) default probabilities, etc. In this literature, however, the econometric frameworks invariantly focus on the individual firm as the unit of observation, i.e. the impact of variation in credit supply to “firm A” on performance of “firm A”.

Any modern economy, however, is characterized by an interlinked production architecture in which firms rely on each other for goods and services as inputs for production. In addition, as firms typically pay for these intermediates with some delay, a trade credit network financially ties interacting firms. An important feature of these real and financial interactions is that it leaves firms exposed to financial distress originating with other firms. The dominant partial equilibrium take on the economy then, by construction, exposes only a fraction of the total real effect of variation in bank credit.

In chapter 2 we take this concern seriously. 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 firmlevel performance to (i) its own credit supply shock and (i)i credit supply shocks to other firms that work their way through the identified production architecture. We show that a material part of the aggregate real impact of credit supply shocks to individual firms arises through spillovers via upstream and downstream mechanisms. We identify endogenous responses in price setting and demand for trade credit as the relevant drivers underlying the downstream and upstream propagation effects, respectively

Chapter 3: The Amiti Weinstein Estimator: An equivalence result.
(i.s.m. Jan Van Hove)

Chapter 3 builds on a long and well–established tradition in the empirical banking literature of disentangling credit demand and supply shocks. This literature has recently popularized a new methodology to separate both channels using matched bank–firm credit data. The contribution of this methodology is to account for general equilibrium constraints such that the micro and macro features of the data are mutually consistent. It generates a unique set of bank and firm shocks that aggregate exactly to match aggregate bank lending and total firm borrowing. The fact that micro–level estimates are consistent with macro–level data allows one to understand how bank–level shocks affect aggregate real activity. In chapter 4 we relate this estimation framework to a weighted least squares routine.

Chapter 4: The Network Origins of Bank Influence: Evidence from Bank-to-Firm and Firm-to-Firm linkages.
(i.s.m. Hans Dewachter & Jan Van Hove)

Chapter 4 focuses on the role of individual banks in the economy. As bank lending shocks affect firm behavior and permeate, via firm inter linkages, the structure of the real economy determines the aggregate real influence of individual banks. Data on the universe of (i) firm-to-firm transactions and (ii) bank-firm borrowing relations of the Belgian economy allows us to reconstruct (a) the firm-level input-output production architecture of the Belgian economy and (b) the credit network supporting it. These objects allow us to quantify and decompose the network origins of individual bank influence. 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.

Date:1 Oct 2013 →  21 Jun 2019
Keywords:Credit frictions
Disciplines:Applied economics, Economic history, Macroeconomics and monetary economics, Microeconomics, Tourism
Project type:PhD project