MERGERS INCREASE OUTPUT WHEN FIRMS COMPETE BY MANAGING REVENUE

ARTURS KALNINSCORNELL SCHOOL OF HOTEL ADMINISTRATION

LUKE M. FROEB, STEVEN TSCHANTZ+

VANDERBILT UNIVERSITY

28 September, 2010Canadian Competition BureauOttawa Canada

I. Antitrust in Industries whereFirms Manage Revenue

• 1999 Central Parking $585 Million acquisition of Allright.– Divestitures in 17 cities

• Froeb et al. (2002) criticize the Justice Department's enforcement action by arguing that the merger would not have raised price because there is very little uncertainty about parking demand. – Firms price to fill capacity, pre- and post-merger

Antitrust in Industries whereFirms Manage Revenue (I)

• 1999 Central Parking $585 Million acquisition of Allright.– Divestitures in 17 cities

• Froeb et al. (2002) criticize the Justice Department's enforcement action by arguing that the merger would not have raised price because there is very little uncertainty about parking demand. – Firms price to fill capacity, pre- and post-merger

Antitrust in Industries whereFirms Manage Revenue (II)

• 2003, the European Commission (EC) gave their approval to Carnival's $5.5 billion takeover of rival cruise operator P&O Princess– Followed UK and US approvals

• Coleman et al. (2003) summarized the empirical analysis done by the FTC, – no correlation between prices and concentration– no correlation between changes in capacity and changes in

price. – firms were adding capacity, increasing amenities, and

competing on price

Antitrust in Industries whereFirms Manage Revenue (III)

• 2005, six luxury hotels in Paris exchanged information about occupancy, average room prices, and revenue– French competition agency: "Although the six hotels

did not explicitly fix prices, …, they operated as a cartel that exchanged confidential information which had the result of keeping prices artificially high" (Gecker, 2005)

– industry executives insisted that their information sharing was to "to bring more people to the area and to maximize hotel utilization"

Revenue Management: set price before demand is realized

• Firm optimizes expected profit:

• Non linearity of min() function means that capacity constrained firm “shades” price to minimize expected error costs– Over-pricing means unused capacity– Under-pricing means foregone revenue

Figure 1: Deterministic unconstrained profit function

60 80 10 0 12 0 14 0price

50 0

10 00

15 00

20 00

25 00

30 00

35 00

pro fi t

Figure 2: Deterministic profit function with non-binding capacity constraint

60 80 100 120 140price

500

1000

1500

2000

2500

3000

3500

pro fi t

Figure 3: Deterministic profit function w/tightly binding capacity constraint

60 80 10 0 12 0 14 0price

50 0

10 00

15 00

20 00

25 00

30 00

35 00

pro fi t

Figure 4: Expected profit function (solid) w/non-binding constraint

60 80 10 0 12 0 14 0price

50 0

10 00

15 00

20 00

25 00

30 00

35 00

pro fi t

Figure 5: Expected profit function (solid) w/tightly binding constraint

60 80 10 0 12 0 14 0p rice

50 0

10 00

15 00

20 00

25 00

30 00

35 00

p ro fi t

Testable hypothesesMerger Theory Demand

Uncertainty

Capacity Constraint

Prediction for occupancy

Prediction for price Comment

Unilateral effects: price or quantity competition

Not binding

Down, unless outweighed by efficiencies

Up, unless outweighed by efficiencies

P and Q move in opposite directions

Pricing to fill capacity: when demand is known

Low Binding No effect No effect Price to fill capacity, both pre- and post-merger

Stochastic economies of scale: pricing to fill capacity when demand is uncertain

High Binding Up Up, if tightly binding constraint

Jointly managed capacity is easier to fill

Demand externalties:merged firm is able to bid for group business

Varies No effect if capacity constrained; Up if not.

Up, if capacity constrained; no prediction if not.

Demand increases for merged hotel.

Data

• Price and occupancy data from Smith Travel Research (STR).– 32,314 U.S. hotels reported to STR the average room-

night price actually received each day, as well as the total number of rooms available and the number of rooms sold.

– 97 monthly observations from 2001 –2009 for each hotel for occupancy and price.

– These 32,314 hotels represent about 95% of chain-affiliated properties in the United States and about 20% of independent hotels and motels.

Table 2: Analysis of all 2628 mergers

Table 2: All tracts9,305 STR client hotels 2,628 hotels involved in

mergers32,314 data-reporting hotels

DV = Occ. DV = price

DV = Occ. DV = price

DV = Occ. DV = price

1.Within-tract Merger .0041+ 0.53 .0083* .55 .0044** 1.51**(.0023) (.34) (.0033) (.50) (.0017) (.23)

2. Out-of-tract merger .0010 .07 .0023* .07 .0005 .48**(.0010) (.05) (.0011) (.12) (.0003) (.04)

Observations 369,627 93,368 1,826,487FX: Hotel*brand 9,607 2,285 36,139FX: Tract*month 8,975 1,868 42,579

Ho: (1) – (2) = 0 1.93 2.11 3.80+ 1.15 5.6* 20.8**(F-test)

Huber-White standard errors in parentheses, clustered by hotel*brand combination.** p < 0.01; * p < 0.05; + p < 0.10 as per two-tailed tests

Table 3: Market tracts split by capacity constraints and then by uncertainty

Table 3: Split of Markets by likelihood of capacity constraints and by level of uncertainty

Likelihood of Capacity Constraints Uncertainty

Lower Half Upper Half Lower Half Upper Half

Occ. ADR Occ. ADR Occ. ADR Occ. ADR

Within-tract Mgr -.0002 0.81 .0070* 0.34 -.0003 -.273 .0074* 1.13*

(.0032) (.53) (.0031) (.43) (.003) (.356) (.0032) (.51)

Out-of-tract Mgr .0009 .15* .0010+ -.0001 .0004 -.018 .0015** .14*

.0005 (.06) (.0006) (.07) (.0006) (.058) (.0006) (.07)

Observations 184,296 185,331 181,569 188,058

FX: Hotel*brand 4,912 4,695 4,701 4,906

FX: Tract*month 4,583 4,394 4,390 4,587

Within-tract mgr 400 498 415 483

Hotels in mergers 1,123 1,505 1,217 1,411

Average of DV 0.60 $92.72 0.66 $102.98 0.62 $94.05 0.64 $101.48

Ho: (1) – (2) = 0 0.15 1.65 3.87* 0.69 .07 .55 3.57+ 4.07*

Huber-White standard errors in parentheses, clustered by hotel*brand combination.** p < 0.01; * p < 0.05; + p < 0.10 as per two-tailed tests

Table 4.1: High Capacity Constraints & Low/High Uncertainty

Table 4 Part 1

Low Uncertainty Markets High Uncertainty MarketsOcc. ADR Occ. ADR

Market tracts where capacity constraints are likely to bind

In-tract Merger .00018 -0.65 .0114** 0.98+(.0004) (.53) (.004) (.60)

Out-of-tract merger -.00001 -0.10 .0018* 0.07(.0008) (.08) (.0008) (.09)

Observations 84,906 100,425FX: Hotels 2,120 2,575FX: Tract*month 1,986 2,410Within-tract mgr 212 286Hotels in mergers 684 871Average of DV 0.65 $98.91 0.67 $106.30

Ho: (1) – (2) = 0 0.001 1.23 5.20* 2.45

Huber-White standard errors in parentheses, clustered by hotel*brand combination.** p < 0.01; * p < 0.05; + p < 0.10 as per two-tailed tests

Table 4.2: Low Capacity Constraints & Low/High Uncertainty: No signif. Results

Huber-White standard errors in parentheses, clustered by hotel*brand combination.** p < 0.01; * p < 0.05; + p < 0.10 as per two-tailed tests

Low Uncertainty High Uncertainty

Occupancy ADR Occupancy ADR

Market tracts where capacity constraints are unlikely to bind

In-tract Merger -.0013 0.17 .0006 1.40

(.0045) (.46) (.0046) (.94)

Out-of-tract merger -.0008 .06 .0010 .24

(.0008) (.08) (.0008) (.09)

Observations 96,663 87,633

FX: Hotels 2,581 2,331

FX: Tract*month 2,406 2,179

Within-tract mgr 203 197

Hotels in mergers 583 540

Average of DV 0.595 $89.79 0.615 $95.95

Ho: (1) – (2) = 0 .24 1.31 .01 1.70

Conclusions

• Mergers increases in occupancy , and lead to economically significant gains of between $1700 and $3300 per month for a 100-room hotel.

• Effects occur only in capacity-constrained and uncertain markets – Mergers allow hotels to better forecast demand.

• No evidence that mergers decrease occupancy or raise price. – Mergers in “revenue management industries,” should not be

modeled with “traditional” models of price or quantity competition. – The same warning applies to the scrutiny of information sharing by

hotels in same market• The Grand Dame hotels of Paris justification for information sharing might

have increased occupancy.