Technology and Policy Research Initiative conference July 23 2018 Nancy L Rose MIT amp NBER Overview Innovative set of papers that inform the debate Patterns of rising markups across OECD countries industries and firms highest for the top firms in the most digitallyintensive secto ID: 712044
Download Presentation The PPT/PDF document "Firm markups and industry concentration:..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Firm markups and industry concentration: Is there a problem?
Technology and Policy Research Initiative conference
July 23, 2018
Nancy L. Rose, MIT &
NBERSlide2
Overview
Innovative set of papers that inform the debate:
Patterns of rising markups across OECD countries, industries, and firms, highest for the top firms in the most digitally-intensive sectors (
Calligaris
,
Criscuolo
,
Marcolin
)
The role of intangible capital in understanding investment and productivity trends (in retail,
Crouzet
and
Eberly
)
Especially the impact of specialized IT (embodied human capital?) may have in generating important trends in concentration, firm scale, “superstar” firms, and margins
Why these trends may be less strong in the EU, due to greater political independence of EC competition enforcers (Gutierrez and
Philippon
)Slide3
My takeaways:
These papers contribute important insights to the debate
And in the case of GP, raise provocative hypotheses
As do many of the subsequent papers we will see today
We need to work hard to establish
The basic facts, especially at relevant units of observation
A link to
competition
as opposed to
bigness
Mechanisms
Why? Because implications
and
mechanisms matter– LOTS– for policy Slide4
Concentration trends
What do we mean by “concentration”?
Industrial organization: Market share distribution among competitors for consumers
Concentration debate: Size distribution of firms in aggregated industries (e.g. 3-digit
NAICS
)
Crouzet
and
Eberly
, Fig 1, 3-digit NAICS
Gutierrez &
Philippon, Fig 1, KLEMS industriesSlide5
What’s an industry?
Example:
KLEMS
: 10-12
Food products, beverages and tobacco
/
NAICS
3-digit: 311: Food ManufacturingWhat’s in this “industry”?
311111: Dog and Cat Food Manufacturing311230: Breakfast Cereal Manufacturing31135: Chocolate and Confectionery
Manufacturing311511: Fluid Milk
Manufacturing31161: Animal Slaughtering and Processing3117: Seafood Product Preparation and Packaging311942
: Spice and Extract ManufacturingNeither rivals for most consumers nor likely similar in productionSlide6
What’s a market? Geography matters. See airlines
Azar et al.
JoF
, forthcoming, Fig 1. Average route level
HHI
,
MHHI
(common ownership)Slide7
In most cases, we can’t just add across firms
Firms typically compete in many product markets with different rivals and competitive conditions
DowDuPont
(2017 merger) plans split into 3 companies: Agricultural, Materials Science, and Specialty Products. Even these are not
markets/products
.
Materials Science: Plastics
, Materials and Chemicals, Infrastructure and Consumer Solutions
And often in many geographiesUsing Compustat
or other firm-level data to assign firm revenue to a sector tells us little about what is happening at the market levelSlide8
Margins and profits
Hal
described difficulties in measuring margins
Crouzet
/
Eberly
&
Bessen
papers work to unpack productivityC-E highlight the role of intangibles Growth of intangible capital correlated with labor revenue productivityMay reflect omitted variables (IT, brand K, etc.) Also M&A activity (market power?)Bessen
highlights the role of proprietary ITReturns to scale in IT specialization and from IT specialization Slide9
Gutierrez & Philippon: How EU Markets Became More Competitive
Not sure what facts we have to explain, bu
t f
ocus
on the theory
In addition to above, concern about measuring enforcement (see cartels
)
Political Economy story of greater independence in EU to preclude captured regulators/enforcers from favoring one country’s industry
BUT how is the model of European nations v. EU logically distinguished from US states v. Federal government? Especially at inception of antitrust policy in US (Sherman 1890)?
Perhaps the paper is focused on the wrong institutional difference?Will argue from anecdote, with apologies, from my 2.5 years at DOJSlide10
A view from the trenches: What might we be missing?
Judicial oversight and burden of proof
Especially relevant for mergers
US enforcers can’t block mergers or stop conduct without a trial to convince a federal judge
US economists, Chicago School lawyers, and judges have ratcheted up proof burdens
Main cost of
trial loss
for enforcers is not reputation, but bad precedent (e.g.
Amex, potential for ATT/TW appeal)
Asymmetry with EC– no effective appeal path in mergersAnd abuse of dominant position has much lower burden then Sec 2.
Facts and causes drive effective responses, particularly for academic economists