Introduction Data Empirical strategy Results Conclusions Asymmetric Price Adjustments in Airlines Diego Escobari Department of Economics & Finance The University of Texas-Pan American University of Alabama December 3, 2010 Diego Escobari Asymmetric Price Adjustments in Airlines
24
Embed
Asymmetric Price Adjustments in Airlines - UTRGVfaculty.utrgv.edu/diego.escobari/Slides/Slides_mde.pdf · Asymmetric Price Adjustments in Airlines Diego Escobari ... Markov-switching
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
IntroductionData
Empirical strategyResults
Conclusions
Asymmetric Price Adjustments in Airlines
Diego Escobari
Department of Economics & FinanceThe University of Texas-Pan American
University of AlabamaDecember 3, 2010
Diego Escobari Asymmetric Price Adjustments in Airlines
Outline
1 IntroductionMotivationDynamic pricing in airlinesContribution
2 DataConstructionRealized demand
3 Empirical strategyDecomposition of PricetTransitory component PriceTtMarkov-switching with FTP and TVTPAsymmetries specifications: Signt , Sizet , Wkndt , and Summt
4 ResultsSignificance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
5 Conclusions
IntroductionData
Empirical strategyResults
Conclusions
MotivationDynamic pricing in airlinesContribution
Motivation
Large price dispersion in airlines. Borenstein and Rose (JPE 1994),Gerardi and Shapiro (JPE 2009).
Positive demand shifts drive prices up:
Basic prediction of economic theory.
Common view among travelers and the media:
“The inevitable outcome of limited seats and stronger demandwill be higher fares.”
New York Times, May 31, 2010
Do prices fall as a response to negative demand shifts?
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
MotivationDynamic pricing in airlinesContribution
Motivation
Potential explanations for the asymmetric response:
Simple convex supply curve.
Models with costly capacity and demand uncertainty: Prescott (JPE1976), Eden (JPE 1990), Dana (Rand 1999)
Menu costs. Mankiw (QJE 1985), Ball and Mankiw (1994)
‘Asymmetric pricing.’ (changes in costs) Peltzman (JPE 2000)
Collusion / Market power.
Uninformed consumers. Tappata (Rand 2009)
Inventories. Borenstein and Shepard (Rand 2002)
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
MotivationDynamic pricing in airlinesContribution
Motivation: Dynamic pricing in airlines
Key characteristics:
Fixed capacity.
Perishable good.
Aggregate demand uncertainty.
Advance sales.
Carriers exploit ‘fences’ such as:
Saturday-night-stayover.
Advance purchase discounts.
Minimum- and maximum-stay.
Refundable tickets.
Frequent flier miles.
Blackouts.
Volume discounts.
Fare classes (e.g. coach, first class)
Airlines have the most sophisticated pricing systems in the world.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
MotivationDynamic pricing in airlinesContribution
Contribution of the current paper
First to explain price variation over different departure dates.
Finds strong evidence of an asymmetric response.
Combines different sources of asymmetries.
Positive cost shifts have a positive effect, but negative demand shiftshave no effect on prices.
Cost shifts have larger effect on prices during summer travel.
Less evidence that the shifts are related to the size of the demandshift or weekend and holiday travel.
Importance of capacity constraints as a source of asymmetricpricing.
Importance of asymmetric pricing to stabilize demand fluctuations.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
ConstructionRealized demand
Construction of the Data
Minimum available non-refundable fare from expedia.com.
Controls for more expensive refundable fares.
126 days (18 weeks) for 48 flights departing between Tuesday June 9and Monday October 12, 2009. Keeping the same flight-number.
Controls for time-invariant specific characteristics.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Decomposition of PricetTransitory component PriceTtMarkov-switching with FTP and TVTPAsymmetries specifications: Signt , Sizet , Wkndt , and Summt
Asymmetries specifications
Four specifications for the vector zt :
Signt = 1, if the demand shift is positive.
Sizet = 1, if the demand shift is more than one standard deviationaway from its mean.
Wkndt = 1, if the departure date is during a weekend or holiday.
Summt = 1, if the departure date is during the summer.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Significance of the regime-switch
Markov-switching state-space representation:
Kim (1994) filter.
Significance of the regime-switch.
Hansen (1992): H0: γj,0 = γj,1 for all jp-value of 0.01.
Note: The ML estimate of c0 appears on the boundary, violating regularity conditions.Hence, to calculate the standard errors, c0 = 0 was imposed to calculate the secondderivatives of the log likelihood.
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
PriceTt
Positively skewed PriceTt
0.1
0.15
0.2
0.25
0.30.3 -
0.25 -
0.2 -
0.15 -
0.1 -
-0.1
-0.0
0
0.05
Sa
t, J
un
27
Tue
, Ju
n 3
0
Fri
, Ju
l 0
3
Mo
n,
Jul
06
Th
u,
Jul
09
Su
n,
Jul
12
We
d,
Jul
15
Sa
t, J
ul
18
Tue
, Ju
l 2
1
Fri
, Ju
l 2
4
Mo
n,
Jul
27
Th
u,
Jul
30
Su
n,
Au
g 0
2
We
d,
Au
g 0
5
Sa
t, A
ug
08
Tue
, A
ug
11
Fri
, A
ug
14
Mo
n,
Au
g 1
7
Th
u,
Au
g 2
0
Su
n,
Au
g 2
3
We
d,
Au
g 2
6
Sa
t, A
ug
29
Tue
, S
ep
01
Fri
, S
ep
04
Mo
n,
Se
p 0
7
Th
u,
Se
p 1
0
Su
n,
Se
p 1
3
We
d,
Se
p 1
6
Sa
t, S
ep
19
Tue
, S
ep
22
Fri
, S
ep
25
Mo
n,
Se
p 2
8
Th
u,
Oct
01
Su
n,
Oct
04
We
d,
Oct
07
Sa
t, O
ct 1
0
0.05 -
0 -
-0.05 -
-0.1 -
Figure: PriceTt . Shaded areas: Signt = 1
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
State dependent Impulse Response Functions
Simulate the path of PriceTt+j as captured by: γ0,0, γ1,0, γ0,1, and γ1,1.
PriceTt−1 = PriceT
t−2 = 0, εt+j = 0, ∀j and Costt−j = 0, j 6= 0.
Costt = 5.5%
6%
10%
14%S(t)=1, S(t+1)=1
S(t)=0, S(t+1)=0
S(t)=1, S(t+1)=0
S(t)=0, S(t+1)=1
-2%
2%
0 1 2 3 4 5
Figure: State dependent IRF of PriceTt . TVTP: Sign
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Transition Probabilities
Determine the transition probabilities as captured by:
Positive demand shifts are more likely to have a large effect on pricesthan negative demand shifts.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Combined Asymmetries
Combined asymmetries.
Robustness of the Summ specification.Variation within the summer departure dates.
Table: Model Selection
Elements of zt SIC AIC Log likelihood LR testa LR testb
Note: a p-value for a test of the null of the FTP model. b p-value for a test of the null ofthe Summ model.
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
State Dependent Impulse Response Functions
State dependent IRF.
6%
10%
S(t)=1, S(t+1)=1
S(t)=0, S(t+1)=0
S(t)=1, S(t+1)=0
S(t)=0, S(t+1)=1
-2%
2%
0 1 2 3 4 5
Figure: State dependent IRF of PriceTt . TVTP: Summ, Sign×Summ
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Significance of the regime-switchMaximum Likelihood EstimationTransitory component of price: PriceTtState dependent Impulse Response FunctionsTransition probabilities
Diego Escobari Asymmetric Price Adjustments in Airlines
IntroductionData
Empirical strategyResults
Conclusions
Conclusions
Strong evidence of response asymmetries.
Prices are more sensitive to demand fluctuations during summer.
Positive demand shifts are more likely to have a positive effect.
Importance of capacity constraints as a source of ‘asymmetricpricing.’
Importance of pricing to stabilize demand fluctuations.
Results are consistent with:
Simple convex supply curve.Models with costly capacity and demand uncertainty: Prescott(1976), Eden (1990), Dana (1999)Menu cost models. Ball and Mankiw (1994)‘Asymmetric pricing.’ Peltzman (2000)
Diego Escobari Asymmetric Price Adjustments in Airlines