United States v. H&R Block: Market Definition in Court since the 2010 Merger Guidelines

• Date: 01 September 2014
• Published date: 01 September 2014
• Author: Frederick R. Warren-Boulton,Marc Remer
• DOI: 10.1177/0003603X1405900310
• Subject Matter: H&R Block: A Merger in a Consumer Good—Article

THE ANTITRUST BUL L E T I N :Vol. 59, N o. 3/Fall 2 014 :599

United States v. H&R Block:

Market definition in court since

the 2010 Merger Guidelines

BYMARC REMER *AND

FREDERICK R. WARREN-BOULTON**

United States v. H&R Block has b een viewed as a contest between old

and ne w app roac hes t o merg er an aly sis: a s tru ctur al ap proa ch

centering on market definition versus an effects analysis involving

merger simulation. The economists on the DOJ team, however, saw

the two approaches as nonconflicting and complementary. We focus

here on two issues in market definition that were disputed at trial:

first, in what order should products be added to the provisional

market; and second, w hen should that proces s stop. With respect to

the r an kin g of “ clo se s ubs ti tut es ,” we s ug ges t “n orm al izi ng ”

diversion ratios to account for shares or output, and perhaps also

for m ar gi ns. Wi th r es pec t to t he l att er, w e sh ow t hat m er ger

simulati on can be use d to estimat e not only the p rice effec t from

© 2014by Federal Legal Publications, Inc.

* Economic Analysis Group, U.S. Department of Justice.

** Microeconomic Research and Consulting Associates (MiCRA), Wash-

ington, DC.

AUTHORS’ NOTE: The views expressed in this article are entirely ours and do not

purport to reflect those of the U.S. Department of Justice. We, along with Ronald

Drennan, Matthew Magura, William Whalen, and Tor Winston from the Justice

Depa rtm ent , and Ha l Van Gie son a nd Ser dar D alk ir fro m MiC RA, we re the

economics team that developed and presented the economics expert testimony for the

Justice Department in United States v. H&R Block. We would like to thank Greg

Werden and Steve Silberman in particular for helpful comments and editing.

this merger, but also the pric e effect for a merger to mo nopoly in

the proposed market and t hus provide a rigorous impl ementation

of the SSNIP test.

KEY WORDS: United St ates v. H&R Block, a ntitrust , mergers , market

definition, merger simulation, critical loss

I. INTRODUCTION

Two basic approach es to evaluating whet her a merger is l ikely to

lead to a price increase have been followed since the Merger Guide -

lines first appeared in their modern form in 1982. The first is th e tra-

ditional structural methodology in the Guidelines’ “two-step”

approach: Fi rst def ine a rel evant marke t broad e nough s o that a

profit-maximizing hypothetical monopolist (an HM) would impose

at least a small but significant and non-transitory increase in price (a

SSNIP), then see how close to that monopoly a merger would take

yo u. T ha t se co nd s te p de pe nd s on t he p ot en ti al r el at io ns hip

between structure and performance: If the concern is unilateral

effects and the market is best described using the Cournot model for

homog enous p roduc ts, the re is a dire ct relations hip bet ween th e

Herfindahl-Hirschman Index (the HHI) and the price effect, leading

to HHI “thresholds” for a safe harbor. Alternatively, if the concern is

unilateral behavior and the dominant firm model best describes the

market, there is an implied relationship between the market share of

the merged fir m and pr ices th at can b e used to infer t he cri tical

threshold.

To the unin itiated o bser ver, thi s two-ste p proc ess ma y seem

strange: in effect, instead of asking what the distance is from

Washington , DC, to Philadel phia (that is, by h ow much wo uld a

merg er in cre ase p ric es) , we fi rst a sk wh at th e dis tan ce is f rom

Washingto n to New York ( where N ew York repres ents a SSN IP).

Then we ask w here Philadelphi a lies on the l ine between Washing-

ton and New York. Neverthel ess, this t wo-step approach h as been

effective in imposing some consistency, and even rigor, on merger

analysi s. It also provides a sc reen that th e parties can imple ment

themse lves to de termin e if their p roposed m erger will fall into a

safe ha rbor and, if no t, prov ide guidanc e as to how the Dep art-

600 :THE ANT I T R U S T BULLETIN:Vol. 59, No. 3/F all 2014