The demand for low-cost carriers: an empirical micro analysis
Juan Muro, Cristina Suárez and María del Mar Zamora
Universidad de Alcalá
Fac. CC. Económicas y Empresariales
Dpto. Estadística, Estructura Económica y O.E.I.
Plza. de la Victoria, 2
28.802 Alcalá de Henares (Madrid)
e-mail: [email protected]
ÁREA TEMÁTICA: El turismo como factor de desarrollo regional y rural
In Europe low-cost carriers (LCC) is a fast growing industry. We empirically
investigate the microeconomic determinants of the demand for LCC, a subject we
know very little about. We adopt a reduced form demand for transportation model,
extended to incorporate possible selectivity biases stemming from interactions
between unobserved individual heterogeneity associated with specific
transportation choice. The model is estimated using a very rich dataset from
EGATUR (Encuesta de Gasto Turístico), the Spanish Foreign Tourist Expenditure
Survey. The sample allows us to explore the influence of price and income related
variables as well as personal characteristics on LCC’s demand. Price and income
results are consistent with theory. Unobserved individual heterogeneity linked with
travelling by air, as opposite to road travel, is significantly and inversely correlated
to unobserved individual heterogeneity related with travelling by air with a LCC, as
contrary to full service airline travel, reflecting travel choice interdependence
through unobserved. This primary attempt to apply selectivity models in
investigating LCC’s demand offers insights and raises new empirical and theoretical
questions.
PALABRAS CLAVE: Demand for transportation, low-cost carriers, binary choice
model with selectivity, individual unobserved heterogeneity, models with many
qualitative variables.
CÓDIGOS JEL: C25, L83
1
1. Introduction
In Europe, there has been a significant growth in the number of operating low-cost
carriers (LCC). The fast development of LCC is related to three factors, Dobruszkes
(2005): first, the demand for air transport is connected with economic cycles with
significant fixed costs linked to aeroplanes ownership; second, the price of air transport
that often remains a limiting factor for a large portion of the population; and third,
airline liberalizations that allow free creation of new services and can thus encourage
the establishment of new airlines. The LCC by offering lower prices is the answer to
these elements.
Related to the airline deregulation in Europe has been the role of new airports, Barret
(2004). The combination of low-cost airlines and low-cost airports has been significant
in terms of increasing gains to the passengers that can be summarised as facing lower
air fares, using smaller airports with shorter waiting times for baggage, shorter walking
times at airports, etc.
Given that the relatively new phenomenon of LCC has important repercussions for
tourist travel decisions, the aim of this paper is to analyse the microeconomic
determinants of the tourists’ LCC choice for foreign tourism arriving to Spain. The
literature on this subject is very scarce. It reduces to some mainly descriptive papers,
see, for example, O’Connell and Williams (2005), Mason (2001), or others on very
specific questions, see, for example, Proussaloglou and Koppelman (1999), Carlsson
and Löfgren (2006).
This paper regards the tourists’ mode of travel choice in a travel choice modelling
framework by assuming that they compare the stochastic utility of several transport
alternatives and select the one that maximizes their utility. In a three alternative
setting, road, LCC, full service airline, we use a probit model with sample selection in
which the probability of a LCC choice is conditional to travel by air.
The model has been estimated with Spanish data on foreign tourism. We utilize the
2004 wave of a very rich database coming from Egatur (Encuesta de gasto turístico)
the Spanish Foreign Tourist Expenditure Survey. The survey is a questionnaire
answered by more than a sixty thousand foreign tourists visiting Spain and requests
information on tourists’ socioeconomic characteristics, attributes of the trip and other
relevant variables including the airline choice.
The paper is organized as follows: we outline a conceptual framework that allows one to
explain tourists’ choice of transportation mode in section 2. We describe in section 3 the
database and we present the empirical results and analyse the main determinants of
LCC choice. Finally, in section 4 we conclude with our conclusions on LCC’ experience in
Spain.
2. The model
We analyse the microeconomic determinants of LCC demand in a travel demand
modelling framework, Warner (1962), Ben-Akiva (1973), McFadden (1974), Domencich
and McFadden (1975). The utility function of the representative tourist is
2
U=U(q1, …, qk, qk+1, …, qn, z, ttrans, ttour),
where qtrans=(q1, …, qk) represents the vector of passenger transport services, qj =(qk+1,
…, qn) denotes all consumer goods (excluding passenger transport services), z stands
for characteristics that define the holiday, ttrans is travel time to the destination and ttour
the length of stay.
The consumer balance will be reduced to
Max U=U(q1, …, qk, qk+1, …, qn, z, ttrans, ttour);
Subject to:
ptransqtrans + pjqj = Y
ttrans+ ttour = T,
where ptrans are the vector of passenger transport services prices, pj are the vector of
the rest of the prices, and Y represents the income level.
In this setting each tourist is assumed to have to make a choice between three travel
modes, road, full service airline and LCC. Due to the cross-sectional nature of our
database we assume a myopic behaviour. For any given tourist, defined by means of
individual observed characteristics, his/her utility is derived from a number of observed
trip attributes and travel features and a set of unobservables.
The probability that a tourist i will choose to travel by LCC equals the probability
associated with a positive difference in the comparisons between the utility derived
from travelling by LCC and the utilities related with road travel and a full service airline
travel. The difference between LCC and full service airline can be represented as an
unobserved latent variable Yi*. So
Yi* = Xi’β + Wi’δ + ui,
[1]
such that one observes only the binary outcome,
Yi = 1 if Yi*> 0 and
Yi = 0 if Yi*≤ 0.
However, one only observes Yi for observation i if tourist travel by air (Ci =1), where Ci*
follows
Ci* = Xi’γ + Zi’π + εI,
[2]
where
Ci = 1 if Ci*> 0 and
Ci = 0 if Ci*≤ 0.
Xi is a vector of individual characteristics; Wi is a vector of variables that are not
included in [2] and Zi is a vector of variables that are not included in [1]. ui and εi are
the error terms for equations [1] and [2], respectively, distributed as bivariate normal
3
with mean zero, unit variance, and ρ = Corr(ui,εi). After controlling by observables our
model allows for correlation between unobservables in equations [1] and [2].
As is well known, when ρ ≠ 0, standard probit techniques applied to equation [1] yield
biased results, and the probit model with sample selection provides consistent,
asymptotically efficient estimates for all the parameters in such models.
3. Empirical analysis
In this section we present results from a probit model with selectivity, equations [1]
and [2] above, in which the primary equation represents the probability of travel with a
LCC (versus a full service airline) conditional on the probability of travel by air (versus
travel by road choice). The model is estimated by maximum likelihood. We do not
report results from previous estimation of a multinomial logit specification but it shows
IIA assumption does not hold1.
All specifications include variables Xi related with tourists’ characteristics that can
influence tourist’ travel choices and the possibility to undertake certain activities. They
include elements such as age, level of education, labour market status, country of
residence, level of income, purpose and organization of the trip. Wi contains trip
attributes (size of travel group, main destination, length of stay and type of
accommodation) and other control variables that can affect the selection between LCC
and a full service airline, for example seasonality, fidelity and a dummy variable for the
use of internet in order to reserve the travel. This last variable measures the influence
of Internet sales on the demand for LCC. Zi includes variables that can affect the choice
between travel by air or by road, for example, a dummy variable which indicates the
proximity to the border with Spain, this variable drives travels by road choice.
The data has been collected from the 2004 wave of EGATUR (Encuesta de Gasto
Turístico) the Spanish Foreign Tourism Expenditure Survey, whose main objective is the
quantification of non resident visitors coming to Spain and of their travel expenditure. It
provides a very rich data set to estimate the empirical model in order to answer why to
choose to fly in low cost carrier.
To highlight the importance of the means of entering the country (road or airport)
Figure 1 shows the percentage of the tourist by the mode of transport. It is important
to remark the importance of the airport in general as a way to coming to Spain, with
more than eighty per cent of tourist travelling by air, if we analysed these air tourists,
we show that the greater number of those correspond to full service carrier, but also
the number of LCC tourist are important with a percentage of 21.3%.
1
A generalized test of the IIA assumption rejects the null.
4
Figure 1: Tourist by mode of entering the country
Percentage of Tourist
100%
Road
Air
Low Cost Carrier 80%
Full Service Carrier
18,5%
21,3%
60%
40%
81,5%
60,2%
20%
0%
In general independent variables have been defined as dummy variables which take a
value of 1 if the tourist belongs to the category specified and 0 otherwise2. Table 1
presents the percentage of the LCC tourist analysed in the whole sample and in the air
traveller sample and we distinguish three types of independent variables: tourist’
characteristics, trip attributes and other control variables. While most of the
characteristics are similar across the samples, it is important to remark the growing
importance of the LCC tourist in the air sample versus the total sample of the
Netherlands tourist, retired, with low level of income, without package tour, with a
number of visits greater than ten and using Internet for booking the trip.
[Insert Table 1]
The results of the estimations of a probit model with sample selection are reported in
the Appendix 2 and Table 2 shows the marginal effects and the pseudo-elasticities
estimate for the model.
[Insert Table 2]
The socioeconomic variables of travellers were introduced to explore differences in
sensitivity to different aspects of air carrier. Differences by origin market show that
French tourists present the lower probability of travelling by LCC. The marginal effect
(which measure the percentage change in probability in relation to the reference
category) are all negatives for travel with a LCC and implies, for example, that French
travellers are about 53% less likely to travel with such companies than the Netherlands.
The greater probability to travel by air corresponds to countries without border with
Spain and reflects the expected pattern for different groups of travellers and the
2
See Appendix for variables definition and a descriptive analysis of the sample.
5
relative importance that travellers place on the additional time spent unproductively at
the travel time to the destination.
The Andalusia and the Community of Valencia are the main destinations for LCC
travellers whilst Madrid and the Canary Islands show the lowest probability for LCC
choice. This result can be explained not only in terms of a shadow price argument but
also by the existence of secondary airports used by LCC or location advantages. From
the supply side, one feature of the LCC is the use of secondary or regional airports to
reduce their costs as much as possible, Warnock-Smith and Potter (2005). Concerning
the selection equation, the Communities without border to other countries present a
greater probability than border Communities of travelling by air, as expected.
Economic variables such as income should also influence the tourists’ choice of air
carriers. It is evident that consumer income does play an important role, as less
wealthy consumers are more price-conscious and hence more susceptible to a switch
between airlines as a result of changes in fares. Therefore, the income coefficients have
the expected signs and when a traveller have more income then increases the
probability of travel with a full service airline versus a LCC, and also travel by air versus
road.
The difference in coefficient values by trip purpose shows business travellers’ have the
highest effect to travel by air or by a full service airline, these results indicate much
more sensitivity to time delays for business travellers that for leisure travellers. Also, it
is important to remark that sun and beach motives have a positive effect to travel by
LCC with a marginal effect or 6.6%, we interpret this result as the higher price
sensitivity of leisure travellers, and, as a consequence, the low cost carrier acts as a
determinant in the decision-making processes of this type of tourist.
In transport planning it is essential to analyse how habits and acquired environmental
knowledge influence travel choice, Gärling and Axhausen (2003). Examples of such
indicators would be the organization of the trip. The trade-off among package holidays
has a positive effect in the probability of travel by air, but a negative effect in the
probability of travel by a LCC, with a marginal effect of 44.5% and -24%, respectively.
The explanation of this result can be found in the different way of planning a travel,
directly or indirectly thorough distribution channels such as travel agents. With the
deregulation of airlines, a new tourism distribution system has emerged that required a
better knowledge of their environments and of the transportation system, and the LCC
have emerged to obtain market advantages of a better understanding of the needs and
wants of individual travellers.
Also evidenced in Table 2 is the fact that the use of Internet for transport reservations
increases the probability of travelling with a low cost airline. This is one of the most
important characteristics of this type of companies, which prefers direct access to a
consumer only through call centers and the Internet
The environmental knowledge also influences the transport choice. It may be assumed
that the cost of searching for new alternatives is generally too high and the expected
gains associated with new alternatives too uncertain. In this situation, travellers reuse
past solutions. If we analyse travellers with ten or more visits to Spain, we can observe
than they prefer to travel with LCC.
6
The lowest the length of the stay the greatest the probability of travel with a low cost
carrier, with positive means marginal effects ranging from 8.6% (between 1 and 3
days) to 0.6% (between 4 and 7 days).
Finally, the statistical significance of the correlation coefficient suggest that controlling
for the likelihood of travel by air versus road is critical to determining the effects of
travel with a LCC. Also, the negative point estimate of ρ implies that the unobserved
factors affecting the probability of travel by air or by low cost carrier are negatively
correlated. In other words, the two outcomes are negatively correlated after controlling
for tourists’ characteristics and the attributes of the destination itself.
4. Conclusions
The impact of low-cost airlines and low-cost airports on fares and passenger numbers in
air transport in Europe has been a significant growth in the last years. Given that the
LCC has important repercussions for tourist travel decisions, the aim of this paper is to
analyse the microeconomic determinants of the tourists’ LCC choice for foreign tourism
arriving to Spain.
The paper has revealed that there are differences between tourist travelling on a LCC
and those on a full service airline. It is clear that the passenger travelling on LCC place
great importance to the use of Internet for booking the trip, to planning the travel
without package, with a trend towards shorter holiday stay and with knowledge of
Spain (more than ten visits). In general, the tourist who visits Spain and chooses travel
by LCC is a traveller with a secondary level of studies, low level of income, which comes
from Netherlands and United Kingdom and goes to the beach in Andalusia and
Community of Valencia.
Also, it is important to remark that tourist who travels by air, including LCC travellers,
come from distance countries (without border with Spain), who has his/her own house
or visit friends or family.
References.
Barret, S.D. (2004), “How do the demands for airport services differ between fullservice carriers and low-cost carriers?”. Journal of Air Transport Management, 10,
33-39.
Ben-Akiva, M. (1973), The structure of passenger travel demand models, unpublished
Ph.D. dissertation, M.I.T. Department of Civil Engineering.
Carlsson, F. and A. Löfgren (2006), “Airline choice, switching costs and frequent flyer
programs”. Applied Economics, 38 (13), 1469-1475.
Coto-Millán, P., J. Baños-Pino and V. Inglada (1997), “Marshallian Demands of Intercity
Passenger Transport in Spain: 1980-1992. An Economic Analysis”. Transport
Research-E, 33(2), 79-96.
Dobruszkes, F. (2005), “An analysis of European low-cost airlines and their networks”.
Journal of Transport Geography, article in press.
Domencich T. and D. McFadden (1975), Urban travel demand: A behavioral analysis.
North Holland, Amsterdam.
7
Gärling, T. and K.W. Axhausen (2003), “Introduction: Habitual travel choice”.
Transportation, 30, 1-11.
Kester, J. G. C. (2003), “Preliminary results for international tourism in 2002: Air
transport after 11 September”. Tourism Economics, 9(1), 95-100.
Mason, K. (2001), “Marketing low-cost airline services to business travellers”. Journal of
Air Transport Management, 7, 103-109.
McFadden, D. (1974), “Conditional logit analysis and qualitative choice behavior” in P.
Zarembka (ed.) Frontiers in Econometrics. Academic Press, New York.
O’Connell, J. and G. Williams (2005), “Passengers’ perceptions of low cost airlines and
full service carriers: A case study involving Ryanair, Aer Lingus, Air Asia and
Malaysia Airlines”, Journal of Air Transport Management, 11, 259-272.
Proussaloglou, K. and F. Koppelman (1999), “The choice of air carrier, flight and fare
class”. Journal of Air Transport Management, 5, 193-201.
Warner, S. (1962), Stochastic choice of mode in urban travel: A study in binary choice.
Northwestern U.P., Evanston.
Warnock-Smith, D. and A. Potter (2005), “An exploratory study into airport choice
factors for European low-cost airlines”, Journal of Air Transport Management, 11,
388-392.
8
Table 1: Percentage of LCC tourist by tourist’ characteristics, trip attributes and other control variables
Total Sample
Tourists’ characteristics
Age
<= 24 years
28,95%
24 < age <= 44
25,77%
44 < age <=64
23,66%
>= 65 years
21,62%
Level of education
Basic education
30,41%
Secondary education
37,62%
University education
31,97%
Labor market status and job category
Employed
19,58%
Student
24,02%
Retired
17,62%
Occupation: Housewife
20,26%
Other
18,52%
Country of residence
France
1,05%
Germany
16,27%
United Kingdom
26,37%
Italy
11,93%
Netherlands
33,26%
Rest of the world
11,13%
Level of income
High
27,51%
Medium
33,18%
Low
39,30%
Purpose of the trip
Sun and beach
35,68%
Work and Business relations
25,74%
Other motives
38,57%
Organization of the trip
With package tour
38,25%
Without package tour
61,75%
By Air Sample
26,26%
22,76%
24,70%
26,28%
30,61%
38,61%
30,78%
18,60%
22,18%
23,83%
17,38%
18,01%
1,85%
11,33%
21,42%
9,11%
47,73%
8,56%
24,33%
28,27%
47,41%
33,86%
20,93%
45,21%
26,03%
73,97%
Total Sample
Trip attributes
Size of travel group
Alone
Couple
More than two
Tourist main destination
Rest of Spain
Andalusia
Balearic Island
Canary Island
Catalonia
Community of Valencia
Madrid
Length of stay
1 < days < 3
4 < days < 7
>= 8 days
Type of accomodation
Freee accomodation
Tourism resort
Other type of accomodation
37,80%
33,52%
28,68%
34,42%
34,62%
30,96%
9,46%
21,97%
18,69%
9,59%
10,44%
21,61%
8,23%
10,54%
23,11%
14,35%
6,02%
14,85%
25,52%
5,61%
27,00%
37,88%
35,12%
32,43%
32,97%
34,60%
44,48%
26,11%
29,40%
46,82%
18,35%
34,83%
Other control variables
Seasonality
First Quarter
27,95%
Second Quarter
24,17%
Third Quarter
23,11%
Fourth Quarter
24,78%
Number of visits
>= 10 visits
49,60%
< 10 visits
50,40%
Use Internet for transport reservations (booking)
Yes
78,59%
No
21,41%
9
By Air Sample
26,62%
23,57%
25,61%
24,19%
57,37%
42,63%
82,04%
17,96%
Table 2: Marginal effects and pseudo-elasticities of the probit model with sample selection
LCC vs Full Service Airline
Age
<= 24 years
24 < age <= 44
44 < age <=64
Level of education
Basic education
University education
Labor market status
Employed
Student
Retired
Occupation: Housewife
Country of residence
France
Germany
United Kingdom
Italy
Rest of the world
Level of income
High
Medium
Purspose of the trip
Work and Business relations
Sun and beach
Organization with package tour
Size of travel group
Alone
Couple
Tourist main destination
Rest of Spain
Andalusia
Canary Island
Catalonia
Community of Valencia
Madrid
Length of stay
1 < days < 3
4 < days < 7
Type of accomodation
Free accomodation
Tourism resort
Seasonality
Second Quarter
Third Quarter
Fourth Quarter
Number of visits >=10
Use Internet transport booking
Marginal Ef.
Ps.-elast.
2.3%
0.0%
0.2%
3.5%
0.0%
0.2%
1.3%
-5.6%
2.0%
-8.6%
2.3%
0.0%
6.0%
1.0%
3.5%
0.0%
9.2%
1.5%
-53.1%
-26.7%
-22.3%
-38.6%
-35.8%
-81.0%
-40.7%
-34.1%
-58.9%
-54.6%
-7.6%
-7.3%
-11.6%
-11.1%
-12.3%
6.6%
-24.0%
-18.8%
10.1%
-36.6%
-2.5%
0.8%
-3.9%
1.2%
-4.5%
8.7%
-12.2%
1.1%
5.0%
-13.2%
-6.8%
13.3%
-18.7%
1.7%
7.7%
-20.1%
8.6%
0.6%
13.0%
0.9%
-0.9%
-2.2%
-1.4%
-3.3%
-5.1%
-3.5%
-6.1%
2.7%
21.0%
-7.7%
-5.3%
-9.3%
4.2%
32.0%
Air vs. Road
Age
<= 24 years
24 < age <= 44
44 < age <=64
Level of education
Basic education
University education
Country of residence with border
Level of income
High
Medium
Purspose of the trip
Work and Business relations
Organization with package tour
Size of travel group <=3
Tourist main destination with border
Length of stay
1 < days < 3
4 < days < 7
Type of accomodation
Free accomodation
Seasonality
Second Quarter
Third Quarter
Fourth Quarter
Number of visits>=10
10
Marginal Ef.
Ps.-elast.
30.9%
27.0%
13.4%
57.9%
50.7%
25.2%
-10.3%
8.0%
-45.9%
-19.3%
14.9%
-86.0%
8.4%
16.9%
15.7%
31.8%
35.6%
44.5%
20.6%
-27.1%
66.8%
83.5%
38.6%
-50.9%
-13.0%
12.6%
-24.5%
23.7%
18.3%
34.2%
-8.5%
-18.3%
-5.0%
-17.6%
-16.0%
-34.3%
-9.4%
-33.0%
Appendix 1
Tourists’ characteristics:
Age: The socio-demographic characteristics have been defined including
information relating to age of the tourist. We have established four categories:
Under 24, between 24 and 44, between 45 and 64 and over 64.
Level of Education: The educational level has been established in three different
categories: Basic, Secondary and University Education.
Labor market status and job category: The occupational situation has been
measured establishing five different situations. These situations are defined by
the following category variables: Employed, Students, (Retired) Pensioners,
Housewife and Other occupation.
Country of residence: We have considered six different origins: France,
Germany, United Kingdom, Italy, Netherlands and rest of the world. Also, we
have used this information distinguishing if the country of residence has border
with Spain or not.
Level of income: This variable considers different income levels which are placed
into the following categories: High income level, Middle (Medium) income level,
and Low income level.
Purpose of the trip: These variables identify tourists whose the principal motives
of the visit Spain is Work and Business relations, Sun and Beaches as travel
motive and others motives.
Organization of the trip: This variable recognizes if the tourist have visited Spain
with a package tour or not.
Trip attributes:
Size of travel group: With this variable we identify if the tourist travels Alone,
with Couple or in a Group of more than two persons.
Tourist main destination: In order to collect the main tourism destinations in
Spain we have defined seven dummy variables: Andalusia, Canary Islands,
Balearic Island, Catalonia, Community of Valencia, Madrid and other
destinations, respectively.
Length of stay: In order to identify the tourist’s fidelity, we have considerate the
categories: More than once in a year, Once in a year and Less than once to refer
the number of times that the tourists visit Spain in one year.
Type of accommodation: We use three different categories: hotels and similar
establishments, free accommodation and other type of accommodation.
Other control variables:
Seasonality: these variables identify the quarter during the trip is made.
Number of visits: In order to identify the tourist’s fidelity, we have considerate
the categories: More than once in a year, Once in a year and Less than once to
refer the number of times that the tourists visit Spain in one year.
Use Internet for transport reservations (booking): This variable reflects the use
of internet by tourist for transport reservations.
11
Appendix 2
Table 3: Estimation of the probit model with sample selection
Coef.
Std. Err.
Air vs. Road
LCC vs Full Service Airline
Age
<= 24 years
0,063
24 < age <= 44
-0,001
0,004
44 < age <=64
Level of education
0,035
Basic education
University education
-0,149
Labor market status
Employed
0,063
0,000
Student
Retired
0,170
0,026
Occupation: Housewife
Country of residence
-1,551
France
Germany
-0,684
-0,571
United Kingdom
-1,015
Italy
-0,931
Rest of the world
Level of income
-0,200
High
Medium
-0,191
Purspose of the trip
Work and Business relations
-0,320
0,188
Sun and beach
Organization with package tour
-0,614
Size of travel group
Alone
-0,068
0,022
Couple
Tourist main destination
Rest of Spain
-0,119
Andalusia
0,252
-0,317
Canary Island
0,030
Catalonia
0,141
Community of Valencia
-0,340
Madrid
Length of stay
0,246
1 < days < 3
0,016
4 < days < 7
Type of accomodation
-0,025
Free accomodation
Tourism resort
-0,058
Seasonality
Second Quarter
-0,134
-0,093
Third Quarter
Fourth Quarter
-0,162
Number of visits >=10
0,076
Use Internet transport booking
0,703
Constant
0,400
Correlation coefficient
-0,445
Number of obs.
60011
Log pseudolikelihood
-37462.8
(0,051)
(0,043)
(0,036)
(0,024)
(0,017)
(0,054)
(0,061)
(0,060)
(0,065)
***
***
(0,146)
(0,032)
(0,032)
(0,043)
(0,036)
***
***
***
***
***
(0,062)
(0,062)
***
***
(0,041)
(0,021)
(0,040)
***
***
***
(0,029)
(0,024)
**
(0,035)
(0,033)
(0,021)
(0,036)
(0,026)
(0,036)
***
***
***
(0,027)
(0,017)
***
Age
<= 24 years
24 < age <= 44
44 < age <=64
Level of education
Basic education
University education
Country of residence with border
Level of income
High
Medium
Purspose of the trip
Work and Business relations
Organization with package tour
Size of travel group <=3
Tourist main destination with border
Length of stay
1 < days < 3
4 < days < 7
Type of accomodation
Free accomodation
Seasonality
Second Quarter
Third Quarter
Fourth Quarter
Number of visits>=10
Constant
Coef.
Std. Err.
0,920
0,771
0,349
(0,041)
(0,030)
(0,028)
***
***
***
-0,259
0,203
-1,526
(0,030)
(0,019)
(0,019)
***
***
***
0,214
0,449
(0,069)
(0,067)
***
***
1,139
1,936
0,557
-0,721
(0,033)
(0,049)
(0,023)
(0,020)
***
***
***
***
-0,329
0,328
(0,025)
(0,021)
***
***
0,487
(0,020)
***
-0,215
-0,468
-0,125
-0,449
0,083
(0,027)
(0,026)
(0,027)
(0,019)
(0,078)
***
***
***
***
***
***
(0,036)
(0,031)
*
(0,021)
(0,024)
(0,020)
(0,019)
(0,016)
(0,113)
(0,114)
***
***
***
***
***
***
***
Censored obs.
Uncensored obs.
a
11106
48905
Individual reference: Netherlands, Balearic Island, less than ten visits, first quarter, without package tour, more
than 64 years old, other type of accommodation, size of travel group over two, secundary education, low level of
income, other labour market status and job category, no use internet for booking, length of stay over 8 days.
***Level of significance 1%, **level of significance, 5%, *level of significance 10%.
12