Savagar, Anthony (2018) Measured Productivity with Endogenous Markups and Economic Profits. Discussion paper. University of Kent (KAR id:70466)
PDF
Publisher pdf
Language: English |
||
Download (793kB)
Preview
|
|
|
This file may not be suitable for user of assistive technology. Request an accessible format. |
||
Official URL https://ideas.repec.org/p/ukc/ukcedp/1812.html |
Abstract
I develop a model of dynamic firm entry, oligopolistic competition and returns to scale in order to decompose TFP fluctuations into technical change, economic profit and markup fluctuations. I show that economic profits cause short-run upward bias in measured TFP, but this subsides to upward bias from endogenous markups as firm entry adjusts. I analyze dynamics analytically through a nonparametric DGE model that allows for a perfect competition equilibrium such that there are no biases in measured TFP. Given market power, simulations show that measured TFP is 40% higher than technology in the short-run, due solely to profits, and 20% higher in the long-run due solely to markups. During transition both effects contribute upward bias: initially the profit effect dominates, but by 5 quarters the two effects contribute equally, and by 10 quarters only the markup effect persists. The speed of firm adjustment ('business dynamism') will determine these timings and therefore the importance of each bias.
Item Type: | Monograph (Discussion paper) |
---|---|
Additional information: | Discussion Paper KDPE 1812 |
Uncontrolled keywords: | Endogenous Markups; Dynamic Firm Entry; Endogenous Productivity; Endogenous Entry Costs |
Subjects: | H Social Sciences |
Divisions: | Faculties > Social Sciences > School of Economics |
Depositing User: | Anthony Savagar |
Date Deposited: | 29 Nov 2018 19:00 UTC |
Last Modified: | 12 Feb 2020 04:10 UTC |
Resource URI: | https://kar.kent.ac.uk/id/eprint/70466 (The current URI for this page, for reference purposes) |
Savagar, Anthony: | ![]() |
- Export to:
- RefWorks
- EPrints3 XML
- BibTeX
- CSV
- Depositors only (login required):