Miguel Urdanoz | Toulouse Business School (original) (raw)

Papers by Miguel Urdanoz

RePEc: Research Papers in Economics, 2011

Faut-il augmenter ou baisser les prix pour ameliorer la rentabilite d'un grand reseau de tran... more Faut-il augmenter ou baisser les prix pour ameliorer la rentabilite d'un grand reseau de transport? La loi de la demande qui stipule une relation inverse entre la quantite et le prix ne peut a elle seule repondre a la question. D'autres parametres interviennent. Ce dossier propose des modelisations de la fonction " quantite " qui permettent d'evaluer l'elasticite c'est-a-dire les consequences sur celle-ci d'une variation de la demande.

HAL (Le Centre pour la Communication Scientifique Directe), Jun 21, 2017

HAL (Le Centre pour la Communication Scientifique Directe), Jun 10, 2019

HAL (Le Centre pour la Communication Scientifique Directe), Jul 5, 2017

The literature on airlines presents few studies analyzing the airlines network evolution and its ... more The literature on airlines presents few studies analyzing the airlines network evolution and its impact on prices, costs or profitability. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2015, and obtain three network indicators. The first one measures how close the network is to a huband-spoke structure. The second indicator measures the airline's ability to provide alternative routes. The third indicator captures the network size. We analyze how the carriers' network evolution can be described by those indicators. We show that low-cost carriers (LCCs) and legacy carriers' network choices differ for the second indicator, while our results exhibit no difference in strategies for the other two indicators. We also show that economic conditions affect differently the three indicators and the magnitude of the impact depends on the airline type.

Transportation Research Part A-policy and Practice, Mar 1, 2020

The literature on airlines presents few studies analyzing the airlines network evolution. We beli... more The literature on airlines presents few studies analyzing the airlines network evolution. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines' network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2018, and obtain three network indicators. The first one measures how close the network is to a single-center structure. The second indicator measures the airline's ability to provide alternative routes. The third indicator captures the network size. We analyze the indicators evolution across time and show their robustness under different scenarios.

Transport Policy, 2022

Airline alliances have a long history yet there is no academic consensus on how they affect price... more Airline alliances have a long history yet there is no academic consensus on how they affect price levels and their impact on price dispersion has not yet been studied. We address this ques�on using a novel methodology mo�vated by the service homogeniza�on and increased price compe��on in this industry in the recent years. Establishing an equivalence between the online sales process and a reverse English auc�on, we use methods from auc�on econometrics to work in a new way with the standard industry data set: using individual �cket sales where only aggregated prices have been used in the past. Applicable to other industries where sellers compete in prices, this approach allows us to reconsider the effect of airline alliances on the distribu�on of airfares in the US domes�c market. We find lower price mean and dispersion in markets where airlines belong to an alliance as a result of the lower variability of costs. The methodology we apply here can be used to study any distribu�on of individualized prices, which are now prevalent since the advent of the digital economy.

HAL (Le Centre pour la Communication Scientifique Directe), Jun 21, 2017

Airlines can allow for excessive delays by introducing extra time in their schedule in addition t... more Airlines can allow for excessive delays by introducing extra time in their schedule in addition to what is technically required. This is the so-called buffer delay or buffer time, and is a generalized practice for all companies around the world. This paper studies the difference between a situation where a social planner could control time schedule to maximize the total welfare and the unregulated market. It appears that, even at the social optimum there are some delays. Thus, if penalties are to be applied, they should be computed from the difference between the observed delays and what could be considered as optimal delays. With this in mind, we model a small network composed by three cities, one of them operating as a Hub, with a single airline serving the three cities. A calibration exercise is run with daily traffic data of May and June 2004 between Nice, Toulouse and Paris that operates as a Hub. Overall, it appears that the welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease of ticket prices. The model thus advocates that, at least for the connections that are considered, prices rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

The so-called buffer time or buffer delay allows airlines to control for excessive delays by intr... more The so-called buffer time or buffer delay allows airlines to control for excessive delays by introducing extra time in their schedule in addition to what is technically required. . We study the differences between unregulated markets - where airlines are free to fix their buffer times strategically - and a situation where a social planner would control for time schedules, and in particular the buffer time. To do so, we use a calibrated model of a network of three cities - one of them being a hub - served by a single airline. Welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease in ticket prices. The model thus advocates that, at least for the connections that are considered, fares rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

22nd Air Transport Research Society World Conference (ATRS 2018), Jul 2, 2018

This paper proposes a methodology to build indicators to measure airline network structure, based... more This paper proposes a methodology to build indicators to measure airline network structure, based on graph theory and principal component analysis. We apply the methodology to the US domestic market on the period 2005-2015, and obtain three network indicators. The first one measures how close the network is to a hub-and-spoke structure. The second indicator measures the network resilience, stressing the ability of an airline to provide alternative routes. The third indicator quantifies the network size. We analyze how the carriers’ strategies evolution can be described by those indicators. We show that low cost carriers (LCCs) and legacy carriers network choices differ for resilience, while our results exhibit no difference in strategies for the other two indicators.

The DB1B data set has been used in the majority of empirical work studying the competition in the... more The DB1B data set has been used in the majority of empirical work studying the competition in the US airline industry, including analyses of prices and price dispersion, airline alliances, network formation and entry. Motivated by the increased price sensitivity and service commoditization in the airline industry in the recent years, we propose a new methodology to work with this data set. By modelling airlines as competing on price, we are able to take advantage of the observational equivalence between Bertrand competition and the reverse English auction so that we can apply empirical auction methodology on the DB1B data set. Using this methodology, we estimate the effect of market-level covariates on the level and variance of prices at the flight level, one such covariate being alliance membership. We find that alliance membership has a positive effect on prices, making them on average 9 percent higher, or about 40 USD per ticket, in markets where alliances operate. ∗Toulouse Scho...

We study airline price competition on long - haul markets. Our analysis includes the two main det... more We study airline price competition on long - haul markets. Our analysis includes the two main determinant s of airline pricing strategies: average level of price and price dispersion. This joint analysis is particularly relevant on mature market where airlines face an increasing competitive pressure from incumbents and potential entrants. To lower price competition, they tend to differentiate their products and use complex discriminatory practices on the basis of their Revenue Management process . The price level and dispersion are jointly affected by a number of common factors. The level of competition and the date of booking compared to the day to departure appear to have a particular relevance. It’s much noticing that in the airline Revenue Management process these two elements are taken into consideration. Much of the models developed in the empirical literature use the information freely provided by the US DB1B database . This restricts the empirical analysis to the US market, ...

CEPR Discussion Paper Series, 2015

The so-called buffer time or buffer delay allows airlines to control for excessive delays by intr... more The so-called buffer time or buffer delay allows airlines to control for excessive delays by introducing extra time in their schedule in addition to what is technically required. . We study the differences between unregulated markets - where airlines are free to fix their buffer times strategically - and a situation where a social planner would control for time schedules, and in particular the buffer time. To do so, we use a calibrated model of a network of three cities - one of them being a hub - served by a single airline. Welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease in ticket prices. The model thus advocates that, at least for the connections that are considered, fares rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

We follow up Ito and Lee (2007)’s research on the effect of alliances on air tickets prices for t... more We follow up Ito and Lee (2007)’s research on the effect of alliances on air tickets prices for the US domestic market. Our main goal is to check if price differentials of allied and or codeshared products are maintained in the recent years. Even if there has not been much modification on the identities of airlines engaging on alliances, our evidence highlights a gradual change of fare gaps patterns over the past decade. Specifically, we document that price differentials between cooperated and non-cooperated products fade away almost to the point of price uniformization. JEL Classification: R48, L40, L93

The literature on airlines presents few studies analyzing the airlines network evolution and its ... more The literature on airlines presents few studies analyzing the airlines network evolution and its impact on prices, costs or profitability. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2015, and obtain three network indicators. The first one measures how close the network is to a huband-spoke structure. The second indicator measures the airline’s ability to provide alternative routes. The third indicator captures the network size. We analyze how the carriers’ network evolution can be described by those indicators. We show that low-cost carriers (LCCs) and legacy carriers’ network choices differ for the second indicator, while our results exhibit no difference in strategies for the ...

Transport Policy, 2020

We study the impact of airline network design on excess travel times for the main US carriers bet... more We study the impact of airline network design on excess travel times for the main US carriers between 2008 and 2017 and find that network configuration affects excess travel time. Based on graph theory and a principal component analysis we build four continuous indicators to measure the airlines networks. We observe that airlines serving more destinations, organizing flights landings and take offs around banks or moving towards a point to point configuration present higher levels of excess travel time. However, there does not seem to exist a preferred network configuration between hub and spoke or point-to-point configuration to reduce excess travel time. We also find a nonlinear impact of competition measured at the citypair level over excess travel time. These results are robust when analyzing observed delays rather than excess travel time.

Transportation Research Part A: Policy and Practice, 2020

RePEc: Research Papers in Economics, 2011

Faut-il augmenter ou baisser les prix pour ameliorer la rentabilite d'un grand reseau de tran... more Faut-il augmenter ou baisser les prix pour ameliorer la rentabilite d'un grand reseau de transport? La loi de la demande qui stipule une relation inverse entre la quantite et le prix ne peut a elle seule repondre a la question. D'autres parametres interviennent. Ce dossier propose des modelisations de la fonction " quantite " qui permettent d'evaluer l'elasticite c'est-a-dire les consequences sur celle-ci d'une variation de la demande.

HAL (Le Centre pour la Communication Scientifique Directe), Jun 21, 2017

HAL (Le Centre pour la Communication Scientifique Directe), Jun 10, 2019

HAL (Le Centre pour la Communication Scientifique Directe), Jul 5, 2017

The literature on airlines presents few studies analyzing the airlines network evolution and its ... more The literature on airlines presents few studies analyzing the airlines network evolution and its impact on prices, costs or profitability. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2015, and obtain three network indicators. The first one measures how close the network is to a huband-spoke structure. The second indicator measures the airline's ability to provide alternative routes. The third indicator captures the network size. We analyze how the carriers' network evolution can be described by those indicators. We show that low-cost carriers (LCCs) and legacy carriers' network choices differ for the second indicator, while our results exhibit no difference in strategies for the other two indicators. We also show that economic conditions affect differently the three indicators and the magnitude of the impact depends on the airline type.

Transportation Research Part A-policy and Practice, Mar 1, 2020

The literature on airlines presents few studies analyzing the airlines network evolution. We beli... more The literature on airlines presents few studies analyzing the airlines network evolution. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines' network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2018, and obtain three network indicators. The first one measures how close the network is to a single-center structure. The second indicator measures the airline's ability to provide alternative routes. The third indicator captures the network size. We analyze the indicators evolution across time and show their robustness under different scenarios.

Transport Policy, 2022

Airline alliances have a long history yet there is no academic consensus on how they affect price... more Airline alliances have a long history yet there is no academic consensus on how they affect price levels and their impact on price dispersion has not yet been studied. We address this ques�on using a novel methodology mo�vated by the service homogeniza�on and increased price compe��on in this industry in the recent years. Establishing an equivalence between the online sales process and a reverse English auc�on, we use methods from auc�on econometrics to work in a new way with the standard industry data set: using individual �cket sales where only aggregated prices have been used in the past. Applicable to other industries where sellers compete in prices, this approach allows us to reconsider the effect of airline alliances on the distribu�on of airfares in the US domes�c market. We find lower price mean and dispersion in markets where airlines belong to an alliance as a result of the lower variability of costs. The methodology we apply here can be used to study any distribu�on of individualized prices, which are now prevalent since the advent of the digital economy.

HAL (Le Centre pour la Communication Scientifique Directe), Jun 21, 2017

Airlines can allow for excessive delays by introducing extra time in their schedule in addition t... more Airlines can allow for excessive delays by introducing extra time in their schedule in addition to what is technically required. This is the so-called buffer delay or buffer time, and is a generalized practice for all companies around the world. This paper studies the difference between a situation where a social planner could control time schedule to maximize the total welfare and the unregulated market. It appears that, even at the social optimum there are some delays. Thus, if penalties are to be applied, they should be computed from the difference between the observed delays and what could be considered as optimal delays. With this in mind, we model a small network composed by three cities, one of them operating as a Hub, with a single airline serving the three cities. A calibration exercise is run with daily traffic data of May and June 2004 between Nice, Toulouse and Paris that operates as a Hub. Overall, it appears that the welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease of ticket prices. The model thus advocates that, at least for the connections that are considered, prices rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

The so-called buffer time or buffer delay allows airlines to control for excessive delays by intr... more The so-called buffer time or buffer delay allows airlines to control for excessive delays by introducing extra time in their schedule in addition to what is technically required. . We study the differences between unregulated markets - where airlines are free to fix their buffer times strategically - and a situation where a social planner would control for time schedules, and in particular the buffer time. To do so, we use a calibrated model of a network of three cities - one of them being a hub - served by a single airline. Welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease in ticket prices. The model thus advocates that, at least for the connections that are considered, fares rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

22nd Air Transport Research Society World Conference (ATRS 2018), Jul 2, 2018

This paper proposes a methodology to build indicators to measure airline network structure, based... more This paper proposes a methodology to build indicators to measure airline network structure, based on graph theory and principal component analysis. We apply the methodology to the US domestic market on the period 2005-2015, and obtain three network indicators. The first one measures how close the network is to a hub-and-spoke structure. The second indicator measures the network resilience, stressing the ability of an airline to provide alternative routes. The third indicator quantifies the network size. We analyze how the carriers’ strategies evolution can be described by those indicators. We show that low cost carriers (LCCs) and legacy carriers network choices differ for resilience, while our results exhibit no difference in strategies for the other two indicators.

The DB1B data set has been used in the majority of empirical work studying the competition in the... more The DB1B data set has been used in the majority of empirical work studying the competition in the US airline industry, including analyses of prices and price dispersion, airline alliances, network formation and entry. Motivated by the increased price sensitivity and service commoditization in the airline industry in the recent years, we propose a new methodology to work with this data set. By modelling airlines as competing on price, we are able to take advantage of the observational equivalence between Bertrand competition and the reverse English auction so that we can apply empirical auction methodology on the DB1B data set. Using this methodology, we estimate the effect of market-level covariates on the level and variance of prices at the flight level, one such covariate being alliance membership. We find that alliance membership has a positive effect on prices, making them on average 9 percent higher, or about 40 USD per ticket, in markets where alliances operate. ∗Toulouse Scho...

We study airline price competition on long - haul markets. Our analysis includes the two main det... more We study airline price competition on long - haul markets. Our analysis includes the two main determinant s of airline pricing strategies: average level of price and price dispersion. This joint analysis is particularly relevant on mature market where airlines face an increasing competitive pressure from incumbents and potential entrants. To lower price competition, they tend to differentiate their products and use complex discriminatory practices on the basis of their Revenue Management process . The price level and dispersion are jointly affected by a number of common factors. The level of competition and the date of booking compared to the day to departure appear to have a particular relevance. It’s much noticing that in the airline Revenue Management process these two elements are taken into consideration. Much of the models developed in the empirical literature use the information freely provided by the US DB1B database . This restricts the empirical analysis to the US market, ...

CEPR Discussion Paper Series, 2015

The so-called buffer time or buffer delay allows airlines to control for excessive delays by intr... more The so-called buffer time or buffer delay allows airlines to control for excessive delays by introducing extra time in their schedule in addition to what is technically required. . We study the differences between unregulated markets - where airlines are free to fix their buffer times strategically - and a situation where a social planner would control for time schedules, and in particular the buffer time. To do so, we use a calibrated model of a network of three cities - one of them being a hub - served by a single airline. Welfare losses that follow from delays are relatively small as compared to the potential benefits that would follow from a decrease in ticket prices. The model thus advocates that, at least for the connections that are considered, fares rather than delays should be the focus of institutions aiming at enhancing passengers’ welfare.

We follow up Ito and Lee (2007)’s research on the effect of alliances on air tickets prices for t... more We follow up Ito and Lee (2007)’s research on the effect of alliances on air tickets prices for the US domestic market. Our main goal is to check if price differentials of allied and or codeshared products are maintained in the recent years. Even if there has not been much modification on the identities of airlines engaging on alliances, our evidence highlights a gradual change of fare gaps patterns over the past decade. Specifically, we document that price differentials between cooperated and non-cooperated products fade away almost to the point of price uniformization. JEL Classification: R48, L40, L93

The literature on airlines presents few studies analyzing the airlines network evolution and its ... more The literature on airlines presents few studies analyzing the airlines network evolution and its impact on prices, costs or profitability. We believe that this gap is due to the difficulty of capturing the network complexity in a simple manner. This paper proposes new simple and continuous indicators to measure the airlines network structure. The methodology to build them is based on graph theory and principal component analysis. We apply this approach to the US domestic market for 2005-2015, and obtain three network indicators. The first one measures how close the network is to a huband-spoke structure. The second indicator measures the airline’s ability to provide alternative routes. The third indicator captures the network size. We analyze how the carriers’ network evolution can be described by those indicators. We show that low-cost carriers (LCCs) and legacy carriers’ network choices differ for the second indicator, while our results exhibit no difference in strategies for the ...

Transport Policy, 2020

We study the impact of airline network design on excess travel times for the main US carriers bet... more We study the impact of airline network design on excess travel times for the main US carriers between 2008 and 2017 and find that network configuration affects excess travel time. Based on graph theory and a principal component analysis we build four continuous indicators to measure the airlines networks. We observe that airlines serving more destinations, organizing flights landings and take offs around banks or moving towards a point to point configuration present higher levels of excess travel time. However, there does not seem to exist a preferred network configuration between hub and spoke or point-to-point configuration to reduce excess travel time. We also find a nonlinear impact of competition measured at the citypair level over excess travel time. These results are robust when analyzing observed delays rather than excess travel time.

Transportation Research Part A: Policy and Practice, 2020