How Do Firm Financial Conditions Affect

Product Quality and Pricing?

Gordon M. PhillipsUniversity of Maryland and NBER

andGiorgo Sertsios

University of Maryland

Motivation Effect of financial structure and in particular financial

distress on real side with respect to investment has been studied extensively (Asquith, Gertner and Scharfstein (1994), Hotchkiss (1995), Andrade and Kaplan (1998), Maksimovic and Phillips (1998)).

Effect on product quality is unknown – additional claimant to a firm in form of an after-purchase claim.

Firms (and their managers) face different incentives in financial distress and bankruptcy and thus financial distress and bankruptcy can cause different product market behavior and different quality choices

Motivation Firms face different incentives in financial

distress and bankruptcy (chapter 11) and thus financial distress and bankruptcy can cause different product market behavior and differential quality choices as modeled by Maksimovic and Titman (1991).

In financial distress firms may be struggling to avoid bankruptcy.

Firm time horizon: If firm managers maximize shareholder interests this can give them short-term incentives.

Focus on short-term cash flows: H1: Firms may reduce quality of experience goods

(Maksimovic and Titman 1991) during financial distress.

H1 part 2: If there was an implicit collusion, financial distress can cause a deviation from the collusive agreement reducing prices.

Motivation In Chapter 11 bankruptcy, the firm managers

and new potential equity holders wish to reorganize and emerge and has to convince customers and suppliers along with the bankruptcy judge and bondholders that it is viable.

Time-horizon may be longer term. In this situation they may increase quality. The effect on prices is not straight forward.

Under automatic stay provision interest payments are deferred, so prices might stay low.

Prices might stay low to attract customers. Prices might increase to convince the

bankruptcy judge of the firm's viability.

Airline Industry The airline industry provides the appropriate conditions to

study the effects of financial condition on product quality and prices.

We observe many financially distressed and bankrupt firms.

Information about quality of experience goods is available: On time performance and mishandled baggage. Also Morrison and Winston (1996) and Busse (2002) have documented that the airline industry is characterized by implicit collusive agreements and price wars.

Previous Literature

Financial conditions affect airline’s accident rates: Rose (1990), Dionne et al (1997) and Noronha

and Singal (2004).

Financial conditions affect airline’s prices: Borenstein and Rose (1995), Busse (2002).

Bankruptcy: Ciliberto and Schenone (2009, 2010) (examine

price, capacity, on-time performance).

Contributions1. We allow for financial distress and bankruptcy

to have different effects on firm quality provision and pricing.

2. We explicitly control for the endogeneity of financial distress and bankruptcy. Our main instrument is the tangibility of firm’s assets.

3. We have a more precise measure of a firm’s financial distress as we use Merton’s default probability.

4. We simultaneously estimate the firm’s pricing, quantity, quality decision and its financial condition.

Data• Unbalanced quarterly panel of 21 airlines from the first

quarter of 1997 to the fourth quarter of 2008.

• 647 firm-quarter observations. 59 of them in bankruptcy.

• Information was obtained from Transtats (BTS), Air Travel Consumer Report (BTS), Compustat, CRSP and BEA.

• We construct 2 measures of firm product quality. One related to mishandled baggage (1 divided by the mishandled baggage rate) and another related to on time performance (1 divided by “Late” where “Late” is how late are late flights, on average, for each airline).

• Yield is our measure of price. It is interpreted as the average price per mile that an airline charges.

Carrier Quarters in the sample Bankruptcy episodes

ATA Airlines 15 1

Air Tran Airways 23 0

Alaska Airlines 36 0

America West Airlines 36 0

American Airlines 48 0

American Eagle Airlines 31 0

Atlantic Southeast Airlines 23 0

Comair 19 1

Continental Airlines 48 0

Delta Air Lines 48 1

ExpressJet Airlines 22 0

Frontier Airlines 14 1

Hawaiian Airlines 20 1

JetBlue Airways 23 0

Mesa Airlines 11 0

Northwest Airlines 48 1

SkyWest Airlines 23 0

Southwest 48 0

Trans World Airways* 15 0

United Airlines 34 1

US Airways 41 2

Total Airlines: 21 Total firm-quarters: 647 Bankruptcy episodes: 9

Carriers and bankruptcies

Summary StatisticsVariables 10th Pctile Mean 90th Pctile Std. Dev. NQuality: Inv. of Mishandled Baggage 0.102 0.209 0.313 0.080 647

Quality: On time Performance 0.062 0.106 0.154 0.061 647

Yield 106.5 143.5 206.9 45.8 647

Total Emplaned Passengers 2.532 9.501 19.284 6.787 647

Default Probability 0.000 0.138 0.692 0.293 586

Bankruptcy 0.000 0.091 0.000 0.288 647

Decongestion 0.718 0.766 0.815 0.038 647

Average Miles per Flight 487.4 886.5 1225.9 276.4 647

Oil Fuel Cost 70.54 153.31 258.30 81.01 647

Competition 1.224 2.013 2.977 0.693 647

Efficiency 0.411 0.578 0.705 0.115 647

% liquidable assets 0.228 0.34 0.429 0.091 647

Income 72547 79197 85261 4968 647

Unemployment 0.041 0.051 0.062 0.085 647

Market Share 0.014 0.066 0.150 0.050 647

Preliminary Evidence I

Evolution of Quality Relative to Bankruptcy

.18

.2.2

2.2

4In

v of

MH

B

-10 -5 0 5Quarters Relative to Bankruptcy

Figure 1A: Inverse of Mishandled Bags

.1.1

2.1

4.1

6.1

8O

n-tim

e P

erfo

rman

ce-10 -5 0 5

Quarters Relative to Bankruptcy

Figure 1B: On-time Performance

Preliminary Evidence II

Evolution of Prices Relative to Bankruptcy

.14

.15

.16

.17

Yie

ld

-10 -5 0 5Quarters Relative to Bankruptcy

Figure 1C: Yield

Preliminary Evidence III

Detrended Quality and Yield according to

Financial Situation

Pre-

distress Distress Bankruptcy

Detrended Within Variation

Quality: Inv. of Mishandled Baggage 29.02 -105.72*** 165.53***

(41.29) (46.87) (56.36)

Quality: On time Performance 3.92 -66.069** 16.25

(24.66) (30.20) (58.06)

Yield -7.51 -50.873*** -15.07

(14.38) (14.62) (13.89)

N 103 32 59 Significance level (p-value): *<0.1, **<0.05, ***<0.01

Empirical Strategy

Default probability and Bankruptcy are mutually exclusive events in theory and practice.

We divide the empirical strategy in two:1. We get rid of the bankruptcy observations and show how the

firm’s supply decisions are affected by the default probability.2. We compare distressed firms that end up in bankruptcy to study

the differential effect of bankruptcy relative to financial distress.

As a robustness check we estimate both conditions simultaneously by imposing a default probability of zero when the firm is bankrupt.

Financial Distress Simultaneous equation system for quality, price, quantity

and demand. Considering only non-bankrupt firm-quarters. 1a) Sit=h(Qit, Def_Prob it, Yit) 2a) Pit=g(Qit, Def_Prob it, Xit) 3a) Qit=f(Def_Prob it, Pit, Sit ,Wit) 4a) Def_Probit=j(Qit, Pit, Sit, Zit)

S is quality, P is price, Q is TEP. Y, X, W, Z are exogenous.

Y: {Decongestion}

X: {Decongestion, Average Miles per Flight, Oil Fuel Cost, Efficiency}

W: {Competition, income, unemployment}

Z: {% liqudable assets}

Instruments satisfy the exclusion restriction.

Financial Distress As we do not have instruments for quality we

replace equation 1a) into the others. Thus, we estimate: 1a’) Sit=h(Qit, Def_Prob it, Yit) 2a’) Pit=g(Qit, Def_Prob it, Xit) 3a’) Qit=f(Def_Prob it, Pit, Yit, ,Wit) 4a’) Def_Probit=j(Zit, Qit, Pit, Yit)

(1) (2) (3) (4) (5) Quality: Inv. of Quality: Total Mishandled On time Emplaned Default Variables Baggage Performance ln(Yield) Passengers Probability Default Probability -0.0979*** -0.0562** -0.4717*** 9.6066*** (0.0330) (0.0259) (0.0667) (1.7962) Total Emplaned 0.0033* -0.0018 0.0147*** 0.0189** Passengers (0.0019) (0.0015) (0.0049) (0.0089) ln(Yield) -2.3552 -0.4044**

(2.1830) (0.1650) Decongestion 0.4303*** 0.0656 -0.6494*** 10.9724** -0.8305* (0.0903) (0.0706) (0.1917) (4.9263) (0.4451) Average Miles per Flight -0.0005*** (0.0001) ln(Oil Fuel Cost) 0.1786*** (0.0335) ln(Efficiency) -0.1788*** (0.0328) Competition -2.3256*** (0.3121) ln(Income) 42.7624*** (9.6782) Unemployment -7.6814 (32.0400) % Liquidable Assets 0.3626** (0.1579) Firm fixed effects Yes Yes Yes Yes Yes Time fixed effects Yes Yes Yes Yes Yes "R-squared" 0.7692 0.7046 0.8717 0.8831 0.5818 N 577 577 577 577 577

Financial Distress (3SLS)

Bankruptcy We use a bankruptcy dummy rather than the

default probability.

We only consider firms that entered into bankruptcy at some point. We restrict our non-bankrupt firm-quarters to have a default probability of at least 10%.

5a) Sit=h(Qit, bankit, Yit) 6a) Pit=g(Qit, bankit, Xit) 7a) Qit=f(bankit, Pit, Yit ,Wit) 8a) bankit=j(Zit, Qit, Pit, Yit)

(1) (2) (3) (4) (5) Quality: Inv. of Quality: Total Mishandled On time Emplaned Variables Baggage Performance ln(Yield) Passengers Bankruptcy Bankruptcy 0.0965*** 0.0597*** 0.0498 0.9023* (0.0169) (0.0175) (0.0596) (0.4777) Total Emplaned -0.0049 -0.0266** -0.0912** 0.2859*** Passengers (0.0122) (0.0124) (0.0391) (0.1022) ln(Yield) 0.0814 -0.9035

(2.7596) (0.8027) Decongestion 0.8262*** 0.3973 0.6088 13.3163** -8.9621*** (0.2479) (0.2544) (0.6225) (6.0715) (1.9681) Average Miles per Flight -0.0005*** (0.0001) ln(Oil Fuel Cost) -0.0412 (0.2260) ln(Efficiency) -1.0257***

(0.3952) Competition -0.1557 (0.2831) ln(Income) -18.5841* (9.8941) Unemployment 30.5765 (51.6761) % Liquidable Assets 1.1190*** (0.4007) Firm fixed effects Yes Yes Yes Yes Yes Time fixed effects Yes Yes Yes Yes Yes "R-squared" 0.7137 0.7017 0.86 0.9674 0.1669 N 121 121 121 121 121

Bankruptcy

Conclusions Quality goes down in financial distress, but

it goes up in bankruptcy relative to the distress situation.

Prices go down with financial distress in agreement with Borenstein and Rose (1995) and Busse (2002).

The End