Special Issue: The Economics of Revenue Management
in Hospitality and Tourism
Tourism Economics
1–19
Predicting the direction of ª The Author(s) 2020
Article reuse guidelines:
dynamic price adjustment in sagepub.com/journals-permissionsDOI: 10.1177/1354816620903900
journals.sagepub.com/home/teu
the Hong Kong hotel industry
Ibrahim Mohammed
University of Professional Studies, Ghana
Basak Denizci Guillet
Hong Kong Polytechnic University, China
Rob Law
Hong Kong Polytechnic University, China
Wassiuw Abdul Rahaman
University of Ghana, Ghana
Abstract
This study analysed dynamic pricing data of Hong Kong hotels within the last-minute 1-week
booking window to determine patterns and direction of room rate changes and their asso-
ciation with hotel characteristics regarding tangible attributes, reputational variables and
contextual factors. Findings show that room rates are more likely to increase than decrease
or stay constant, and that, holding demand and market conditions constant, the likelihood of
price increases (decreases), based on standard binomial probit regression, is positively
(negatively) associated with size (tangible attribute), chain affiliation and star rating (reputa-
tional attributes), and seller density and location accessibility (contextual factors). These
results confirm the importance of differentiation in pricing hotel rooms and indicate how
hotel customers and revenue managers can combine these characteristics with predicted
demand to anticipate the direction of room rate change in the last-minute booking window as
the booking horizon approaches check-in.
Keywords
Hong Kong, intertemporal price discrimination, price change, price patterns
Corresponding author:
Ibrahim Mohammed, Department of Banking and Finance, University of Professional Studies, P. O. Box LG149, Accra,
Ghana.
Email: ibrahim.mohammed@upsamail.edu.gh
2 Tourism Economics XX(X)
Introduction
Nearly four decades ago, the hotel industry begun to implement revenue management (RM), a
strategic pricing approach which allows hotels to deploy information systems and pricing strate-
gies to allocate the right capacity to the right customer at the right price at the right time (Kimes
and Wirtz, 2003). Per this pricing strategy, hotels’ room rates are differentially set according to
demand elasticities and varied according to predicted demand levels so that when demand is high
or inelastic, room rates will correspondingly be high and vice versa. Thanks to new technologies
and Internet-enabled distribution channels, the application of RM has become widespread and
more acceptable (Anderson and Xie, 2016; Kimes and Wirtz, 2003). Through RM software and
technologies, hotels are empowered to vary their room rates more rapidly and efficiently, leading
to heightened uncertainty about the direction of room rate change and the need for predictive
analytics to guide strategic decision-making.
While technology has reduced the cost of price adjustment, price-point movement is not
entirely costless (Kalnins, 2016). Beyond the tactical decision-making aided by RM software,
revenue managers still have to consider other factors, strategic in nature, before deciding to
implement the recommendations of RM software or not. According to Abrate and Viglia (2016),
these attributes can be conceptualized into three main types of variables: tangible variables in
the form of physical objective characteristics, reputational variables in the form of ratings and
contextual variables such as the characteristics of the location and the competitive environment.
By invoking the argument presented by Abrate and Viglia (2016), it is believed that patterns of
price movements could be linked to these variables as well. For example, based on quality
signalling theory and hedonic pricing models, revenue managers of high-quality-rated hotels
may be more likely to increase their room rates in the face of increased demand than lower them
when demand is low (Abrate et al., 2011). In other words, price adjustment could be more
flexible upward and sticky downward.
Examining pricing data from the airline industry, Mantin and Gillen (2011) argued that there are
certain ‘hidden information’ about airlines that can be combined with demand to predict the
direction of airfares. Taking a cue from Mantin and Gillen’s (2011) study and contributing to
filling the lacuna in hotel RM literature, this study sought to examine patterns of price adjustments
in the Hong Kong hotel market with the objective of identifying hotel characteristics regarding
tangible, reputational and contextual attributes that can be combined with demand and market
conditions to predict the likelihood of hotels increasing or decreasing their room rates in the last-
minute booking window. To achieve this stated objective, we collected primary data from the
Hong Kong hotel market and examined the data to: first, determine patterns of price change
(increases or decreases) as the booking horizon approaches check-in. Second, we modelled the
price increases and decreases using probit regression techniques to determine their relationship
with demand and hotel characteristics regarding tangible attributes, reputational variables and
contextual factors.
In the hotel sector, one of the commonest ways of segmenting customers for RM and dynamic
pricing is the booking day or transaction date (Abrate et al., 2012; Hanks et al., 1992). Typically,
intra-weekday (e.g. Tuesday night) bookings are characterized by business travellers while a
weekend day (e.g. Saturday night) stays are more in line with leisure travellers (Balaguer and
Pernias, 2013). In a dynamic context, Abrate et al. (2012) observed statistically significant
opposing price patterns for a mid-week and a weekend day. In view of this, our data collection
targeted two different booking days, an intra-weekday (Tuesday night) and a weekend day
Mohammed et al. 3
(Saturday night) so that we could investigate the price patterns on these days separately to reveal
the subtle differences in dynamic pricing. The findings make a modest contribution to our
understanding of the factors that influence the direction of room rate change given predicted
demand. This knowledge can be used by hotel customers who either book to stay on a weekend
(typically, leisure/tourist customers) or mid-weekday (typically, business customers) and revenue
managers to predict the direction of room rate changes in the last-minute booking window.
Literature review
Following the adoption of RM into the hotel industry, dynamic pricing has become a popular and
prevalent pricing strategy implemented by hotels (Anderson and Xie, 2016). With this pricing
strategy, hotels vary their room rates over the booking horizon to maximize revenue (Guizzardi
et al., 2017, 2019). In its early application, dynamic pricing was viewed by customers as a strategy
used to benefit only firms (i.e. increase firms’ profit) (Dixit et al., 2005; Gabarino and Maxwell,
2010). However, in recent times, this pricing technique is now being accepted as a win–win
strategy for both customers and firms. Dixit et al. (2008) note that under certain boundary con-
ditions, dynamic pricing benefits both customers and managers. When implemented appropriately,
low-valuation customers who are sufficiently patient to wait for sales may get good deals (Su,
2007) while companies get higher revenue than fixed prices will produce (Ingold et al., 2000). On
the contrary, when dynamic pricing is not implemented appropriately, prices may end up being too
high when demand is low or too low when demand is high (Sanjay, 2009). Thus, appropriate
implementation of dynamic pricing helps to balance demand with supply by adjusting prices
accordingly.
Given the widespread application of dynamic pricing in the hotel industry, there has been an
avalanche of research in this area. These series of research have been classified into four main
areas by Abrate et al. (2019). The first set of studies examines dynamic pricing under the term
‘intertemporal price discrimination’ and suggests patterns of price change over time (Abrate et al.,
2012; Alderighi et al., 2015). The second strand of studies (Choi and Mattila, 2004; Heo and Lee,
2011; Kimes, 1994; Viglia et al., 2016) referred to as ‘price fairness’ investigates questions of
whether dynamic pricing is considered to be fair, transparent and acceptable. The third batch of
studies referred to as ‘inventory controls’ investigates models and processes that optimally allocate
available items (e.g. rooms in a hotel) to ensure that they are available till the end of the booking
window (Guo et al., 2013; Ling et al., 2015). The fourth set of studies, labelled as ‘organisational
culture’, focuses on the social, human and infrastructure needed to support the successful
implementation of dynamic pricing (e.g. Aubke et al., 2014; Kalnins, 2016; Melis and Piga, 2017;
Selmi and Dornier, 2011). In the subsequent sections of this literature review, we will devote
attention to intertemporal price discrimination where this study fits.
In economics literature, intertemporal price discrimination is distinguished from behaviour-
based price discrimination (Villas-Boas, 1999) when the only difference in the items sold is the
time of purchase (Stokey, 1979). As far as RM literature is concerned, interest in intertemporal
price discrimination has mainly centred on patterns of price changes over the buying period or
booking window. Studies conducted on this subject matter have indicated that, depending on the
underlying modelling assumptions, the optimal dynamic/intertemporal price pattern can be
increasing, decreasing, fluctuating or constant over time (Chen and Schwartz, 2008; Feng and
Gallego, 1995; Gallego and Van Ryzin, 1994; Su, 2007). In one of the theoretical explanations of
optimal dynamic price patterns, Su (2007) used the composition of customers and patience
4 Tourism Economics XX(X)
(waiting cost) to explain why price pattern may be increasing or decreasing. In his exposition, Su
(2007) contends that price will decrease over time when high-valuation customers are quite
impatient to buy early at higher prices and low-valuation customers are sufficiently patient to wait
for sales. On the other hand, price will increase over time when high-valuation customers are more
patient than low-valuation customers.
In an empirical study on dynamic pricing among European hotels, Abrate et al. (2012) argued
that Su’s hypothesis can simply be tested in the hotel industry by comparing price patterns on
weekdays when business customers (high-valuation less-patient) are dominant to those on
weekend when leisure customers are predominant. On the strength of this argument, they postu-
lated that pricing on weekdays will generally exhibit a downward pattern as the booking period
approaches check-in while pricing on weekend will show upward trend. Their subsequent analysis
of dynamic pricing among European hotels confirmed that the pattern of intertemporal price
discrimination is related to the type of customers, leisure customers or business customers, who
normally book weekend or weekday, respectively. In other studies, it is reported that, although
some slight differences in terms of dynamic price patterns on weekday versus weekend exist,
observing increasing price trend is more common in the tourism industry (Alderighi et al., 2015;
Gaggero and Piga, 2011).
Seeking to demonstrate that dynamic pricing decision in hotel RM is based on tactical as
well as strategic considerations, Abrate and Viglia (2016) offered a framework which iden-
tifies three main types of variables that influence dynamic prices and their evolution. These
variables consist of tangible variables in the form of physical objective characteristics,
reputational variables in the form of ratings and contextual variables in the form of the
characteristics of the location and the competitive environment. Empirical analysis by the
authors using dynamic hedonic price function supported this framework and exposed areas of
further research, one of which is considered in this study. In the main, while Abrate and
Viglia’s (2016) study has been very useful in providing a better understanding on how the
three variables affect price variations, the scope of their study did not allow the authors to
investigate how the tangible, reputational and contextual variables affect the direction of price
variation. To fill this gap, this study proposes the following three hypotheses and tests them
with data from a hotel market where dynamic pricing is prevalent.
Hypothesis 1: Tangible attribute of hotels like number of rooms is associated with the
likelihood of increasing or decreasing room rate pattern over the last-minute 1-week book
window.
Hypothesis 2: Reputational variables such as star rating and chain affiliation are associated
with the likelihood of increasing or decreasing room rate pattern over the last-minute 1-week
book window.
Hypothesis 3: Contextual variables such as seller density, proximity to transport hub,
international airport and attractions are associated with the likelihood of increasing or
decreasing room rate pattern over the last-minute 1-week book window.
The above hypotheses are justified in different ways. The first hypothesis recognizes that hotels
have certain tangible attributes such as size (or number of rooms) that determine the price level
(Abrate et al., 2011). Generally, hotels with larger inventory or capacity tend to have lower optimal
prices (Pan, 2007; Vives and Jacob, 2019). In terms of dynamic pricing, it is argued that, all things
Mohammed et al. 5
being equal, big-sized hotels are more likely to have unsold rooms than small-sized hotels, hence,
they are more likely to dynamically adjust their room rates to avoid empty rooms. Vives and Jacob
(2019) observe that the number of rooms available in a hotel negatively affects dynamic pricing.
Therefore, we expect that the higher the capacity of a hotel establishment, the lower (higher) the
probability of a price increase (decrease).
The second hypothesis emphasizes the role of reputational attributes such as star rating and
chain affiliation in dynamic pricing. In setting hotel room rates, quality has always been an
important consideration (Abrate et al., 2011; Becerra et al., 2013; Yang et al., 2016; Zhang et al.,
2011). As a differentiation factor, higher quality hotels – measured by star rating and chain
affiliation – have higher probability of a price rise (Ropero, 2011). Therefore, we expect that
hotels that differentiate themselves through higher star rating and chain affiliation will have a
higher (lower) probability of increasing (decreasing) their room rates as compared to those with
lower quality.
Lastly, the third hypothesis hinges on the notion that the location of a hotel (or contextual
factors) is paramount in setting room rates (Bull, 1994; Urtasun and Gutiérrez, 2006) as it
determines the attractiveness of the hotel and density or competition it faces (Balaguer and
Pernias, 2013; Yang et al., 2014). Location attractiveness or contextual factors such as
proximity to transport hub and tourism attractions put premium on room rates. Spatial
agglomeration may also affect the probability a price rise (Canina et al., 2005). In a fixed
location, Abrate et al. (2012) showed that one of the contextual factors that affects dynamic
pricing strategy is seller density. As seller density increases in a location, hotels decrease their
room rates in real-time.
In RM pricing, one of the most common ways to segment hotel customers for dynamic
pricing is the booking day or transaction date (Abrate et al., 2012; Hanks et al., 1992).
Typically, leisure customers who are mostly price sensitive tend to book on weekend while
business customers who are less price sensitive tend to book on weekdays (Balaguer and
Pernias, 2013). In an empirical analysis to estimate willingness to pay for hotel rooms
by different customers, Schamel (2012) distinguishes hedonic models for weekday and
weekend to reflect the different customer segments – business travellers and leisure travellers.
Examining the dynamic pricing structure of almost 1000 European hotels, Abrate et al. (2012)
observed that the price patterns for mid-weekday and weekend day are statistically significant
and opposing, with price patterns for a mid-weekday being downward while that of the
weekend being generally upward.
In light of the notable difference in price patterns over weekday and weekend, the current study
proceeds to test the above hypotheses using pricing data for an intra-weekday (Tuesday night) and
a weekend day (Saturday night) to capture two main customer segments of hotels – business
travellers and leisure travellers. By doing so, we will be presenting information on price patterns
and their determinants in a fashion that will be useful to business travellers who normally book on
weekdays and leisure traveller who tend to book on weekends. In addition, the information will be
helpful to deal-seeking customers to know which hotels, by virtue of their tangible, reputational
and contextual attributes, exhibit what price patterns.
Methodology
This section presents the methodology which covers the model specification and variables defi-
nitions and data.
6 Tourism Economics XX(X)
Econometric model specification
To identify the factors that are associated with dynamic price adjustments, we estimate the effect of
several variables representing the tangible, reputational and contextual attributes of hotels as well
as market conditions on the probability of price increase or decrease using standard binomial probit
model. This models has been used in similar studies by Benkovskis et al. (2012) and Ropero (2011)
due to the characteristics of the dependent variable. According to Ropero (2011), the probit models
are preferred over logit models because they do not require the independence of the irrelevant
alternative. In its standard form, the panel probit models for price increase ðyþit Þ and price decrease
yit are expressed mathematically as follows:
yþit ¼ xitb þ ci þ uit ð1Þ
yit ¼ xitb þ ci þ uit ð2Þ
and
 
þ= e
xitbþci
Pr yit ¼ 1jxit; ci ¼ þ bþ ¼ Gðxitb þ c Þ ð3Þ1 ex c iit i
where yþit is price increase for hotel i over a 1-week booking window for time t; y

it is price decrease
for hotel i over a 1-week booking window for time t; xit is the vector of tangible, reputational and
contextual attributes as well as market conditions identified in the literature to influence price
increase or decrease. These variables are listed Table 1. b is the set of parameters to be estimated. ci
is an unobserved time-invariant individual effect; uit is a zero-mean residual uncorrelated with all
the terms on the right-hand side; Gð:Þ is the standard normal cummulative distribution function
(cdf) with zero mean and unit standard deviation.
To understand how the dependent variables are constructed, we follow Chen and Schwartz (2008)
to identify four generic price-change-patterns: increasing, decreasing, fluctuating, and no-change. An
increasing price pattern is one where the observed quoted room rate increases monotonically over the
booking horizon. In contrast, a decreasing price pattern is one where the quoted room rate diminishes
continually over the booking horizon. A fluctuating pattern describes room rates that change irre-
gularly, whereas the no change pattern represents a single quoted rate that does not vary during the
evaluation period. Drawing on an analogous literature on how to evaluate multiple price changes
(Chen and Rao, 2007; Thaler, 1985), we define five possible mutually exclusive and collectively
exhaustive outcomes based on the price patterns identified by Chen and Schwartz (2008):
i. Pure increase: this refers to increasing price pattern and shows that succeeding prices are
higher than the preceding prices in the booking window as you move towards check-in.
Thus, the parentage change in prices are all positive.
ii. Pure decrease: this refers to decreasing price pattern and shows that succeeding prices are
lower than the preceding prices in the booking window as you move towards check-in.
Thus, the parentage change in prices are all negative.
iii. Mixed increase: this refers to a fluctuating price pattern and shows a combined effect of
price changes that yields a real positive outcome. Thus, the net percentage change in prices
is positive.
iv. Mixed decrease: this refers to a fluctuating price pattern and shows a combined effect of
price changes that yields a real negative outcome. Thus, the net percentage change in
prices is negative.
7
Table 1. List of variables, definitions and a priori expectations for logit model.
Variable Definition Operationalization Source Reference
Dependent variable
Price increases or price decreases The BAR within the 7 days preceding a Binary for price increase and price Kayak.com Chen and Schwartz (2008)
target date for check-in decrease
Independent variables
Market Conditions
Occupancy The proportion of the available rooms in Percentages STR
a month sold by hotels
Starting price BAR at the start of last-minute 1-week Natural logarithm of the BAR Kayak.com Chen and Schwartz (2008)
booking window
Tangible attribute
Size The number of rooms in a hotel Categorized into three groups and HKTB/STR Becerra et al. (2013); Hung
operationalized by dummies et al. (2010)
Reputational attributes
Chain Independent or chain-affiliated Dummy variable STR Balaguer and Pernı́as (2013);
Hung et al. (2010)
Star rating The official star category of the hotel Dummy variables Kayak.com Becerra et al. (2013); Abrate
et al. (2011)
Contextual factors
Seller density The number of hotels within 500 m Count of hotels Computed Balaguer and Pernı́as (2013);
radius of a focal hotel Lewis (2008)
District The official administrative district Dummy variables HKTB Balaguer and Pernı́as (2013)
assigned to a hotel by the HKTB
Distance to attractions The sum of the distance between a hotel Mean Harversine distance in Google maps Becerra et al. (2013)
and the top 10 tourist attraction kilometres (km)
Distance to airport The distance between a hotel and the Harversine distance in kilometres Google maps Balaguer and Pernı́as (2013)
Hong Kong International airport (km)
Distance to MTR The distance to nearest Mass Transit Harversine distance in kilometres Google maps Balaguer and Pernı́as (2013)
Railway station (km)
Note: The coordinates were measured in degrees. The top tourist attractions (according to HKTB) are the Avenue of Stars, the Peak, Ocean Park Hong Kong, Hong Kong Disneyland,
Ladies’ Market, Temple Street Night Market, Hong Kong Convention and Exhibition Centre (and Golden Bauhinia Square), Tsim Sha Tsui Promenade, Sik Yuen Wong Tai Sin Temple
and the Clock Tower. BAR: best available rate; HKTB: Hong Kong Tourism Board; MTR: mass transit railway; STR: Smith Travel Research.
8 Tourism Economics XX(X)
v. Stable: this refers to no change pattern and shows that either prices did not vary in the
booking window or the fluctuations yield zero outcome.
Using a simple mathematical model (without losing generality), the net effect of two sequential
changes in price is measured by the overall percentage from the base value as:
ð Þ ¼ ½vð1 þ aÞð1 þ bÞ  vNet effect NE ¼ a þ b þ ab ð4Þ
v
where v is the base value and is greater zero; a and b are the first and second percentage. Based on
the net effect, price increase (comprising pure increase and mixed increase) and price decrease
(comprising pure decrease and mixed decrease) are expressed in binary forms as:

þ 1 if the ith hotel price pattern for time t exhibits a positive NEPrice increase ¼ yit ¼
0 if otherwise
and 
 1 if the ith hotel price pattern for time t exhibits a negative NEPrice decrease ¼ yit ¼
0 if otherwise
The estimation of the parameters in equations (1) and (2) are done by means of maximum
likelihood. Following the examples of Abrate et al. (2011, 2012) and Schamel (2012), each
equation is estimated for the Tuesday and Saturday data as previously intimated.
Variables and data
The dependent and independent variables used in this study are defined in Table 1. The inde-
pendent variables have been classified into tangible, reputational and contextual attributes as well
as market conditions. In addition, the table provides information on the corresponding sources
from which the data were gathered and literature references to justify the selected variables. As
indicated early-on, the data were collected from the Hong Kong hotel market, a leading tourism
destination in Asia with world-class hotels.
At the commencement of the data collection, 186 hotel establishment were targeted for the data
collection based on two selection criteria: (a) establishments that are registered with the Hong Kong
Tourism Board (HKTB, an institution that provides hotel industry statistics in Hong Kong) as hotels
and (b) hotel establishment with advertised room rates on standard twin/double room on third-party
channels accessible on kayak.com, a price comparison site. For all the targeted hotels, the unit of
analysis was the best available rate (BAR) for a single night stay in a standard twin/double room. The
minimum BAR rates across the listed online travel agents (OTAs) were selected. In most cases, there
were price parity across the OTAs. The standard twin/double room was chosen for this study because
it is the commonest hotel room type in the world and the Hong Kong hotel market to be precise. As
such, it constitutes a good basis for meaningful comparison and replication.
To have comprehensive data, the data collection was done over a 6-month period using auto-
mated web-scraping technique that assures timeliness and validity of data. This technique has been
used in similar studies by Balaguer and Pernı́as (2013), Becerra et al. (2013), Abrate et al. (2012)
and Schamel (2012) and found to be reliable and valid. Within the 6-month period of the data
collection, two representative days for business and leisure customers (i.e. Tuesdays and Satur-
days, respectively) were targeted for check-in, and, the BAR rates for these days collected from
Mohammed et al. 9
day 7 prior to check-in to the day/date of check-in. The decision to limit our data collection to the
last-minute 1-week booking window was informed by earlier studies on dynamic pricing that have
reported frequent price adjustment and bookings in this period (Jang et al., 2019). In a report cited
by Jang et al. (2019), more than 60% of hotel bookings in the United States were within 0–7 days
before arrival. Findings from an experiment conducted by Chen and Schwarz (2013) also indicate
that there is a sharp increase in the participants’ propensity to book during the last week before the
date of stay, thus, making it an important window to study.
Given the focus on last-minute booking, hotels that were not available for booking and thus no
rates provided were justifiably omitted as out-of-scope. In the end, a total of 126 hotels were
retained. These hotels consistently had advertised room rates at the start of the last-minute one-
week booking window. Hotels that sold out within the 1-week period were still retained to
calculate the percentage change in price for the days that they had inventory. In the rare case,
where inventories lasted for just day 1 into the booking window, the starting price was treated as
stable. Thus, a total of 3276 patterns of price change for 26 Tuesdays and 26 Saturdays were
generated for the analysis. The composition of the sampled 126 hotels used for the analysis is
described in Figure 1.
Results
The results from the analysis are presented in two parts. The first part presents the findings on the
descriptive analysis showing the patterns of price change or direction of room rate change. The
second part is devoted to the econometric analysis of the factors that are associated with the
likelihood of room rate increase or decrease.
Descriptive results of direction of room rate change
To start, the aim of the descriptive analysis in this section was to provide insights on the direction
of room rate change to advance the understanding on dynamic pricing strategies of Hong Kong
hotels. Figure 2 presents the result for the combined data which shows that, within the last 7 days
to check-in, room rates are varied (increased or decreased) 93.6% of the times. In 60.64% of the
observations, the direction of price change was upward while in the rest of the cases, price
decreases and no-price-change accounted for 33.04% and 6.32%, respectively. From this finding,
it can be deduced that although room rates could go up, down or remain the same during the 7
days prior to check-in, on the balance of probabilities, dynamic price adjustments were more
likely to tilt towards prices increases than price decreases which in turn was more likely than no-
price-change, all else remaining constant. Thus, for customers who may be looking forward to
securing last-minute deals (or lower prices), this finding suggests that securing lower prices is
less probable than having to pay higher prices, hence, early bookers may avoid paying last-
minute higher prices.
Analysing the patterns of price change between the representative days for weekend (i.e.
Saturday) and weekdays (i.e. Tuesday), the findings, reported in Figure 3, are largely consistent
with that of Figure 2. However, one notable difference is that, whereas price reductions were less
likely for both Saturdays and Tuesdays, a slightly higher proportion of room rates on Tuesdays
were more likely to have price decreases than room rates on Saturdays. Similarly, the proportion of
room rates for Saturdays that were dynamically adjusted upward was slightly higher than the
proportion of room rates for Tuesdays that were adjusted upward. What this means is that even
10 Tourism Economics XX(X)
Compositon of hotels by age (in Composition of hotels by quality, 
years) Star rating
45 
More than 20 years 20 
(35.7%) 5-star
(15.9%)
15 
10 to 20 years
(11.9%)
76 
4-star
(60.3%)
30 
5 to 9 years
(23.8%)
30 
36 3-star
Less than 5 years (23.8%)
(28.6%)
Composition of hotels by size, rooms Composition of hotels by opeartion
65 
Large-sized hotels (> 300)
(51.6%)
Independent 67 (53.2%)
38 
Mid-sized hotels (101-300)
(30.2%)
Chain Management 59 (46.8%)
23 
Small hotels (≤ 100)
(18.3%)
Figure 1. Composition of sample. Star rating was obtained from kayak.com; size classification was based on
McCann and Vroom (2010) study; class information was obtained from STR. STR: Smith Travel Research.
though both leisure and business guests can expect to make last-minute deals with lower
probability, the odds are in favour of business guests than leisure customers. In terms of price
stability (i.e. the bars for ‘price unchanged’), Tuesdays’ rates appeared to be more stable than
Saturdays’ rates.
Linking the descriptive results to the extant literature, it can be averred that, the results offer
partial support for the findings of Abrate et al. (2012) who reported that the dynamic pricing
structure of European hotels was related to the booking day and by extension the type of customers.
In their study, the authors determined a decreasing dynamic pricing structure (equivalent to price
decreases in this study) for a weekday; while on a weekend, the inter-temporal price structure was
increasing (equivalent to price increases in this study). Furthermore, the authors noted that the
proportions of their sample maintaining stable prices on Saturday and Tuesday were 7.1% and
10.9%, respectively. This is similar to the finding in this study where the proportion of no-price-
change on Tuesdays (8.42%) is higher than that on Saturdays (6.32%).
Mohammed et al. 11
70
60.64%
60
50
40
33.04%
30
20
10 6.32%
0
price decreases price increases price unchanged
Figure 2. Overall patterns of room rate change.
70 60.64%
60 56.56%
50
40 35.01%
33.04%
30
20
8.42%
10
6.32%
0
price decreases price increases price unchanged
Saturdays Tuesdays
Figure 3. Overall patterns of room rate change on Saturdays and Tuesdays.
Regression results
Having determined the patterns of price change in the previous section, the analysis in this section
focuses on identifying the factors that are associated with the probability of room rate increase
versus the probability of room rate decrease. The analyses were done using the probit models stated
in the methodology, results of which are reported in Table 2.
From Table 2, four regression outputs are reported: price increase on Saturdays, price decrease
on Saturdays, price increase on Tuesdays and price decrease on Tuesdays. Looking at these
regression outputs, there are a couple of ways to pair up the results for meaningful comparison and
interpretation. First, the results for each day (both price increase and price decrease) can be
compared to identify the significant variables. Second, the results for price increase, both Saturday
and Tuesday (or price decrease, both Saturday and Tuesday), can be considered together to identify
the significant variables. While each approach may emphasize a point, we consider the second
Pattern/Direction of price  change (% o) Pattern/Direction of price  change (% o)
12 Tourism Economics XX(X)
Table 2. Probit regression results for price increase and decrease on Saturdays and Tuesdays.
Saturday Tuesday
Variables Price increase Price decrease Price increase Price decrease
Occupancy 0.0893*** (15.74) 0.0821*** (14.01) 0.0487*** (16.56) 0.0388*** (11.89)
Starting price 0.0026*** (7.92) 0.0018*** (6.61) 0.0049*** (8.74) 0.0011*** (5.94)
Seller density 0.0192** (2.41) 0.0899 (1.11) 0.0126** (2.19) 0.0064 (0.96)
Chain 0.0462** (2.18) 0.0338* (1.80) 0.0737** (2.11) 0.0682* (1.83)
4-Star 0.0289 (0.32) 0.0403 (0.54) 0.0619 (0.86) 0.0384 (0.59)
5-Star 0.0411*** (3.31) 0.0278 (0.33) 0.498*** (3.38) 0.204 (0.62)
Medium-sized (101–300 0.213* (1.89) 0.201 (1.09) 0.282** (2.37) 0.251** (2.29)
rooms)
Large-sized (more than 0.228* (1.83) 0.217 (0.88) 0.240** (2.15) 0.295** (2.62)
300 rooms)
Distance to airport 0.0512*** (4.67) 0.0691*** (3.98) 0.0551*** (3.84) 0.0641** (3.72)
Distance to nearest train 0.365*** (4.19) 0.264** (3.71) 0.0429 (0.56) 0.0161 (1.08)
station
Mean distance to top 0.0124 (0.78) 0.0217 (0.34) 0.0062 (1.32) 0.0098 (1.11)
attractions
District dummies (controlled) Yes Yes Yes Yes
Constant 8.647*** (13.64) 8.121*** (12.72) 5.672*** (11.71) 5.013*** (7.61)
Log pseudolikelihood 1794.11 1784.66 1872.94 1861.22
Wald 2 469.18*** 448.81*** 474.63*** 446.28***
Lnsig2u_constant 3.826*** 3.41*** 3.790*** 3.752***
N 3276 3276 3276 3276
Note: t-statistics in parenthesis; standard errors are robust; 3-star is the comparison hotel for star rating; small hotel (less
than 100 rooms) is the reference group for size.
*p < 0.1; **p < 0.05; ***p < 0.01.
approach to be more informative and conducive for highlighting any differences that may be of
interest to either business customers making reservations for mid-week stays (Tuesdays) or leisure
customers making reservations for weekend stays (Saturdays). Therefore, as far as possible, the
interpretation and discussion of the results in the subsequent sections will follow the second option.
Contrasting the results for prices increase on Saturdays with that on Tuesdays, it can be noted
that the significant factors contributing to the probability of price increase on both days are nearly
the same. Similarly, the factors that are associated with the probability of price decrease on both
Saturdays and Tuesdays are largely consistent and conversely related to price increase. For price
increase on Tuesdays and Saturdays, the chief market condition factors are occupancy (repre-
senting demand) and starting price at inception of the last-minute booking window which repre-
sents load factor (Ippolito, 1981). Size (or number of rooms) representing tangible attribute is also
significant in influencing the probability of price increase. In terms of reputational attributes, the
significant variables are chain affiliation and star rating which also differentiate one hotel from
another. The significant contextual variables include seller density, capturing competition and
distances to international airport and train station representing location accessibility or attrac-
tiveness. By contrast, the same factors, except seller density and star rating, are associated with the
probability of price decrease on both Saturdays and Tuesdays.
To proceed with the interpretation of the significant coefficients, it is worthwhile to state that
the coefficients reported in Table 2 are not marginal effects, therefore, the interpretations will be
done with respect to signs and not magnitudes. Starting with the market conditions variables,
Mohammed et al. 13
occupancy is positively associated with probability of price increase, all other factors remaining
constant. In other words, as occupancy increases, the probability of price increase occurring
increases. This means that, in line with demand-based pricing, room rates can be expected to go up
or down when demand is predicted to be high or low, all else being equal. In terms of the second
market condition variable which is starting price at inception of the last-minute 1-week booking,
the coefficient in the price increase model is positive and significant. This implies that at higher
price level, which may be indicative of higher demand, the probability of price increase is higher,
all other factors remaining constant. Conversely, when the last-minute 1-week booking window
starts with a lower price, the probability of price decrease increases. This finding is supported by
the significant negative coefficient of starting price in the price decrease models.
The significant coefficient of hotel size in both the price increase and price decrease models,
which represents an important tangible attribute, offer support to hypothesis one and the a priori
expectation that large-sized hotels, relative to small-sized hotels, should be less (more) likely to
increase (decrease) price in order to avoid unsold inventory, all things being equal. Regarding hotel
characteristics that represent reputational attributes, the results show that chain affiliation is
positively (negatively) associated with the probability of price increase (decrease) but star rating is
only associated, positively, with the probability of price increase. This goes to partially support
hypothesis two. The effects of these two reputational attributes emphasize the importance of
quality differentiation in dynamic price adjustments of room rates. They underscore the fact that
service quality differentiation is important in the hotel industry and can be used to predict why
some hotels may be more likely to increase or decrease room rates given demand and other factors.
Finally, the significant coefficients of some of the contextual variables also offer partial support
for hypothesis three. Particularly, seller density, which captures the extent of localized competition
in the immediate environs of a hotel has a significant negative coefficient in the price increase
models, implying that the more densely populated the location of a hotel is with other hotels, the
lower the probability of price increase, all other factors remaining constant. However, the same
variable is not significant in the price decrease models. The results may be interpreted to mean that
hotels that are located in spatially agglomerated spaces have a lower probability of increasing room
rates, suggesting that spatial competition restrains room rates from going up but not necessarily
down since the seller density variable does not have a significant effect on price decrease.
Still on the coefficients of the contextual variables, distance to international airport is signif-
icantly associated with price increase and decrease in both the Saturday and Tuesday regressions.
However, average distance to main attractions is not significant. Meanwhile, distance to the nearest
train station, the fastest and convenient mode of ground transportation in Hong Kong, is signifi-
cantly associated with price changes on Saturdays and not Tuesdays. Perhaps, the convenience of
the train system in Hong Kong can be used explain why distance to the train station matters for
price increase on Saturdays but distance to attraction do not matter. For leisure customers or
tourists, their concern may be convenience of access to attractions and not necessarily how far they
are. After all, the journey time may add to the touristic experience. In terms of sign, distance to
airport is positively (negatively) associated with price increase (decrease), suggesting that hotels
that are located farther away from the international airport have a higher (lower) probability of
price increase (decreases). This finding is surprising considering that proximity to airport is known
to have premium pricing effect (Bull, 1994; Lee and Jang, 2011, 2012). Besides the location
accessibility variables, the districts where hotels are located were controlled for in the regression
output in order to account for location attractiveness. However, these coefficients are not explicitly
reported in Table 2.
14 Tourism Economics XX(X)
Discussion and implications. The results presented in the previous section have revealed that different
patterns of price adjustments occur in Hong Kong hotel market as a result practicing dynamic
pricing, and these differences are related to the booking day. On both Saturdays and Tues-
days, dynamic price patterns within the last-minute 1-week window could be upward or
downward. However, price increases were more likely (i.e. 60.64% of the times) than price
decreases (i.e. 33.04% of the times). The finding on last-minute price increases can be said to
support the argument of Lott and Roberts (1991) who stated that, in the airline industry, late
bookers are charged higher price because the price includes the opportunity cost arising from
the airlines’ risk of having empty seats. To the contrary, the finding on last-minute price
decreases can be interpreted to mean that because of the perishability of hotel rooms, rates
may be decreased to avoid/minimize revenue loss from empty room, that is, the opportunity
cost of not selling room.
The findings from the econometric analysis, which identifies hotel characteristics relating to
tangible attributes, reputational variables and contextual factors that can be combined with demand
and market conditions to predict the direction of room rate change, can also be linked to existing
literature. Relating these findings to the theories of pricing and price determination, which posit
that the prevailing price of a product or service changes in response to changes in the market
conditions, we can conclude that the findings are in line with the theoretical predictions.
First, the positive (negative) association between occupancy and price increase (decrease)
simply confirms the theory of demand which predicts that, all things being, price increases when
demand goes up, and decreases when demand goes down. Empirically, similar conclusion has been
arrived at in another study on dynamic pricing strategies of European hotels conducted by Abrate
et al. (2012). Using the proportion of hotels in their sample with available room as a proxy for
occupancy rate which was not available to them, the authors found that increasing price patterns
were positively associated with increasing demand. Second, the direct (inverse) relationship
between starting price at inception of last-minute 1-week booking and price increases (decreases)
indicate that for weeks that the starting price is high (presumably because of high demand), the
pattern of price change within the last-minute booking window is upward. However, when the
week starts with a lower price, the pattern of price change can be expected to be downward. An
implication of this finding is that given a reference price in mind, customers and revenue managers
can predict the pattern of price change.
In addition to the market condition variables, the results provide evidence to support the thesis
statement that hotel characteristics in the form of tangible attributes, reputation variables and
contextual factors can be combined with demand to predict the likelihood of dynamic price
increases or decreases of hotel rooms. In terms of contextual factors, the inverse relationship
between seller density and price increase is in consonance with competitive market structures and
the findings of Ropero (2011) in the Spanish hotel market, where increased number of hotels does
not affect the price reductions but is negatively related to price increases. Also, average distance to
the nearest train station is significantly associated with price changes on Saturdays but not
Tuesdays, suggesting that accessibility to ground transportation to facility movement is considered
in adjusting room rates for leisure travellers. Surprisingly, distance to airport is positively
(negatively) associated with price increase (decrease), suggesting that hotels that are located
farther away from the international airport have a higher (lower) probability of price increase
(decreases). This finding is contrary to the expectation that proximity to airport is known to have
premium pricing effect (Bull, 1994; Lee and Jang, 2011, 2012). However, considering that the
central business district and tourist attractions in Hong Kong are located farther away from the
Mohammed et al. 15
airport, this may not be entirely surprising. Besides, Hong Kong is known to have efficient
transport to and from the airport.
Regarding the reputational attributes, the point to note from the findings is that differentiation
matters when it comes to the direction of dynamic price adjustments. Especially, reputational
attributes such as chain affiliation and star rating are associated with price changes. For star
rating, the results show that 5-star hotels, relative to 3-star hotels, have a higher probability of
increasing room rates and not decreasing it. The statistically significant positive coefficient of
the 5-star variable which relates only to the Tuesday regression can be explained as follows.
Considering that the room rates of 5-star hotels are generally higher than that of 3-star hotels and
targeted at business customers, price increase on the mid-weekday (Tuesday) may be more
acceptable to business guests. This finding could also be interpreted to mean that price increases
may be considered to be a signalling tool about quality as predicted by the quality signalling
theory and therefore high-quality hotels indulge in it (Abrate et al., 2011). Similarly, chain
affiliation is related to dynamic price changes but affects both price increase and decrease on
both days. Finally, the coefficient of the representative variable for tangible attributes of hotels
(i.e. number of rooms) confirm the a priori expectation that, medium- and large-sized hotel,
relative to small-sized hotels, have higher probability of price decrease as a result of the large
inventory they have to sell. The evidence, as established, shows that: being medium- or large-
sized hotel contribute negatively to the probability of price increase and positively to the
probability of price decrease, especially on Tuesdays.
The findings of this study have several implications, three of which are briefly discussed as
follows. First, the descriptive results, which highlight the incidences of price increase, decrease
and no-change can be used by both leisure and business guest to anticipate the direction in which
room rate are more likely to go (up, down, or stable) as the booking horizon approaches check-in.
In other words, the descriptive results shed light on patterns of room rate change which can be
exploited by deal-seeking customers. Second, the factors that have been identified to be associated
with prices increase and decrease can be used by both customers and revenue managers to predict
the likelihood of price increase or decrease by hotel establishments with certain characteristics
within the last-minute 1-week booking window. For instance, given demand and competition,
customers and revenue mangers can expect that chain hotels, high-quality hotels and large-sized
hotels will have a higher (lower) probability of price increase (decrease) than their counterpart
independent, low-quality and small-sized hotels. Third, hotel developers can use the results for site
selection since certain locations have high or low seller densities and contribute negatively or
positively to the probability of price increase or decrease.
Conclusion and directions for future
This study has analysed dynamic pricing data of Hong Kong hotels within the last-minute 1-week
booking window to determine patterns and direction of room rate changes and their association
with hotel characteristics relating to tangible attributes (number of rooms), reputation variables
(chain affiliation and star rating) and contextual factors (seller density, distances to main trans-
portation hub and airport). The findings revealed that, on the balance of probabilities, room rates
were generally likely to increase than decrease or stay constant as the booking horizon approaches
check-in. Also, the results showed that in addition to demand and market conditions, the likelihood
of adjusting room rates upward or downward as time passes is associated with chain affiliation, star
rating, size and location accessibility which differentiate one hotel from another. Specifically, the
16 Tourism Economics XX(X)
findings suggest that high-quality hotels (i.e. 5-star) and those belonging to a hotel chain have a
higher (lower) probability of price increase (decrease). Furthermore, large-sized hotel, compared
to small-sized hotels, have a reduced probability of price increase and an increased probability of
price decrease. Proximity to the international airport and train station were positively (negatively)
associated with the probability of price increase (decrease).
Given these findings, it is concluded that the direction of room rate change is not determined
solely by demand and market consideration but also characteristics of hotels and their locations
that differentiate them from others. Therefore, to make informed judgement about the direction of
room rate, it will be imperative to consider these differentiation variables in addition to predicted
demand and market conditions. As it is with most empirical studies, this study is not without
limitations that simultaneously create opportunities for further research. Mainly, the findings are
limited in geographical scope and data collection as its main focus was on only one room type in
Hong Kong hotel market. The booking window was also limited to 7 days before check-in.
Therefore, future research can expand this study along these lines to contribute to the general-
izability or otherwise of these initial findings.
Acknowledgement
The authors wish to thank Smith Travel Research (STR) for providing portions of the data used for this study.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or
publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD
Ibrahim Mohammed https://orcid.org/0000-0002-9583-7078
Rob Law https://orcid.org/0000-0001-7199-3757
References
Abrate G, Capriello A and Fraquelli G (2011) When quality signals talk: evidence from the Turin hotel
industry. Tourism Management 32(4): 912–921.
Abrate G, Fraquelli G and Viglia G (2012) Dynamic pricing strategies: evidence from European hotels.
International Journal of Hospitality Management 31(1): 160–168.
Abrate G, Nicolau JL and Viglia G (2019) The impact of dynamic price variability on revenue maximization.
Tourism Management 74: 224–233.
Abrate G and Viglia G (2016) Strategic and tactical price decisions in hotel revenue management. Tourism
Management 55: 123–132.
Alderighi M, Nicolini M and Piga CA (2015) Combined effects of capacity and time on fares: insights from
the yield management of a low-cost airline. Review of Economics and Statistics 97(4): 900–915.
Anderson CK and Xie X (2016) Dynamic pricing in hospitality: overview and opportunities. International
Journal of Revenue Management 9(2–3): 165–174.
Aubke F, Woeber K, Scott N, et al. (2014) Knowledge sharing in revenue management teams: ante-
cedents and consequences of group cohesion. International Journal of Hospitality Management 41:
149–157.
Mohammed et al. 17
Balaguer J and Pernı́as JC (2013) Relationship between spatial agglomeration and hotel prices. Evidence from
business and tourism consumers. Tourism Management 36: 391–400.
Becerra M, Santaló J and Silva R (2013) Being better vs. being different: differentiation, competition, and
pricing strategies in the Spanish hotel industry. Tourism Management 34: 71–79.
Benkovskis K, Fadejeva L and Kalnberzina K (2012) Price setting behaviour in Latvia: econometric evidence
from CPI micro data. Economic Modelling 29(6): 2115–2124.
Bull AO (1994) Pricing a motel’s location. International Journal of Contemporary Hospitality Management
6(6): 10–15.
Canina L, Enz CA and Harrison JS (2005) Agglomeration effects and strategic orientations: evidence from the
US lodging industry. Academy of Management Journal 48(4): 565–581.
Chen C and Schwartz Z (2008) Room rate patterns and customers’ propensity to book a hotel room. Journal of
Hospitality & Tourism Research 32(3): 287–306.
Chen C and Schwartz Z (2013) On revenue management and last-minute booking dynamics. International
Journal of Contemporary Hospitality Management 25(1): 7–22.
Chen H and Rao AR (2007) When two plus two is not equal to four: errors in processing multiple percentage
changes. Journal of Consumer Research 34(3): 327–340.
Choi S and Mattila AS (2004) Hotel revenue management and its impact on customers’ perceptions of
fairness. Journal of Revenue and Pricing Management 2(4): 303–314.
Dixit A, Braunsberger K, Zinkhan GM, et al. (2005) Information technology–enhanced pricing strategies:
managerial and public policy implications. Journal of Business Research 58(9): 1169–1177.
Dixit A, Whipple TW, Zinkhan GM, et al. (2008) A taxonomy of information technology-enhanced pricing
strategies. Journal of Business Research 61(4): 275–283.
Feng Y and Gallego G (1995) Optimal starting times for end-of-season sales and optimal stopping times for
promotional fares. Management Science 41(8): 1371–1391.
Gaggero AA and Piga CA (2011) Airline market power and intertemporal price dispersion. The Journal of
Industrial Economics 59(4): 552–577.
Gallego G and Van Ryzin G (1994) Optimal dynamic pricing of inventories with stochastic demand over
finite horizons. Management Science 40(8): 999–1020.
Garbarino E and Maxwell S (2010) Consumer response to norm-breaking pricing events in e-commerce.
Journal of Business Research 63(9–10): 1066–1072.
Guizzardi A, Pons FME and Ranieri E (2017) Advance booking and hotel price variability online: Any
opportunity for business customers? International Journal of Hospitality Management 64: 85–93.
Guizzardi A, Pons FME and Ranieri E (2019) Competition patterns, spatial and advance booking effects in the
accommodation market online. Tourism Management 71: 476–489.
Guo X, Ling L, Yang C, et al. (2013) Optimal pricing strategy based on market segmentation for service
products using online reservation systems: an application to hotel rooms. International Journal of Hospi-
tality Management 35: 274–281.
Hanks RD, Cross RG and Noland RP (1992) Discounting in the hotel industry: a new approach. Cornell Hotel
and Restaurant Administration Quarterly 33(1): 15–23.
Heo CY and Lee S (2011) Influences of consumer characteristics on fairness perceptions of revenue
management pricing in the hotel industry. International Journal of Hospitality Management 30(2):
243–251.
Hung W, Shang J and Wang F (2010) Pricing determinants in the hotel industry: quantile regression analysis.
International Journal of Hospitality Management 29(3): 378–384.
Ingold A, Yeoman I and Leask A (2000) Yield Management: Strategies for Service Industries. London:
Thomson.
Ippolito RA (1981) Estimating airline demand with quality of service variables. Journal of Transport Eco-
nomics and Policy 15(1): 7–15.
Jang Y, Chen CC and Miao L (2019) Last-minute hotel-booking behavior: the impact of time on decision-
making. Journal of Hospitality and Tourism Management 38: 49–57.
18 Tourism Economics XX(X)
Kalnins A (2016) Beyond Manhattan: localized competition and organizational failure in urban hotel markets
throughout the United States, 2000–2014. Strategic Management Journal 37(11): 2235–2253.
Kimes SE (1994) Perceived fairness of yield management: applying yield-management principles to rate
structures is complicated by what consumers perceive as unfair practices. Cornell Hotel and Restaurant
Administration Quarterly 35(1): 22–29.
Kimes SE and Wirtz J (2003) Has revenue management become acceptable? Findings from an international
study on the perceived fairness of rate fences. Journal of Service Research 6(2): 125–135.
Lee SK and Jang S (2011) Room rates of US airport hotels: examining the dual effects of proximities. Journal
of Travel Research 50 (2): 186–197.
Lee SK and Jang S (2012) Premium or discount in hotel room rates? The dual effects of a central downtown
location. Cornell Hospitality Quarterly 53(2): 165–173.
Lewis M (2008) Price dispersion and competition with differentiated sellers. Journal of Industrial Economics
56(3): 654–678.
Ling L, Dong Y, Guo X, et al. (2015) Availability management of hotel rooms under cooperation with online
travel agencies. International Journal of Hospitality Management 50: 145–152.
Lott JR and Roberts RD (1991) A guide to the pitfalls of identifying price discrimination. Economic Inquiry
29(1): 14–23.
Mantin B and Gillen D (2011) The hidden information content of price movements. European Journal of
Operational Research 211(2): 385–393.
McCann BT and Vroom G (2010) Pricing response to entry and agglomeration effects. Strategic Management
Journal 31(3): 284–305.
Melis G and Piga CA (2017) Are all online hotel prices created dynamic? An empirical assessment. Inter-
national Journal of Hospitality Management 67: 163–173.
Pan CM (2007) Market demand variations, room capacity, and optimal hotel room rates. International
Journal of Hospitality Management 26(3): 748–753.
Ropero MA (2011) Dynamic pricing policies of hotel establishments in an online travel agency. Tourism
Economics 17(5): 1087–1102.
Sanjay K (2009) Yield management: getting more out of what you already have. Ericsson Business Review
1(2): 17–19.
Schamel G (2012) Weekend vs. midweek stays: modelling hotel room rates in a small market. International
Journal of Hospitality Management 31(4): 1113–1118.
Selmi N and Dornier R (2011) Yield management in the French hotel business: an assessment of the
importance of the human factor. International Business Research 4(2): 58–66.
Stokey NL (1979) Intertemporal price discrimination. The Quarterly Journal of Economics 93(3): 355–371.
Su X (2007) Intertemporal pricing with strategic customer behavior. Management Science 53(5): 726–741.
Thaler R (1985) Mental accounting and consumer choice. Marketing Science 4(3): 199–214.
Urtasun A and Gutiérrez I (2006) Hotel location in tourism cities: Madrid 1936–1998. Annals of Tourism
Research 33(2): 382–402.
Viglia G, Mauri A and Carricano M (2016) The exploration of hotel reference prices under dynamic pricing
scenarios and different forms of competition. International Journal of Hospitality Management 52: 46–55.
Villas-Boas JM (1999) Dynamic competition with customer recognition. The Rand Journal of Economics
30(4): 604–631.
Vives A and Jacob M (2019) Dynamic pricing for online hotel demand: the case of resort hotels in Majorca.
Journal of Vacation Marketing. Epub ahead of print. DOI: 1356766719867377
Yang Y, Mueller NJ and Croes RR (2016) Market accessibility and hotel prices in the Caribbean: the
moderating effect of quality-signaling factors. Tourism Management 56: 40–51.
Yang Y, Luo H and Law R (2014) Theoretical, empirical, and operational models in hotel location research.
International Journal of Hospitality Management 36: 209–220.
Zhang Z, Ye Q and Law R (2011) Determinants of hotel room price: an exploration of travelers’ hierarchy of
accommodation needs. International Journal of Contemporary Hospitality Management 23(7): 972–981.
Mohammed et al. 19
Author biographies
Ibrahim Mohammed (PhD) is a senior research fellow/lecturer and the Director of Research and Consul-
tancy at the University of Professional Studies, Accra (UPSA), Ghana. He has a wide range of research
interests due to his diverse background in Economics, Accounting, Financial Management and Hospitality
and Tourism Management. He is a Chartered Accountant and a Certified Hospitality Educator. Currently, he
focuses his research attention on revenue management and the application of economics to hospitality and
tourism management. His publications have appeared in top-tier journals such as International Journal of
Hospitality Management, International Journal of Contemporary Hospitality Management, Tourism Eco-
nomics, International Journal of Social Economics and International Review of Economics.
Basak Denizci Guillet is an associate professor in the School of Hotel and Tourism Management, Hong Kong
Polytechnic University. Her research interests include hospitality financial management and revenue man-
agement. She has published widely in top-ranking journals in hospitality and tourism. She is a coordinating
editor for International Journal of Hospitality Management and an editorial board member for a host of other
journals including International Journal of Contemporary Hospitality, European Journal of Tourism
Research and International Journal of Hospitality and Tourism Administration.
Rob Law is a professor in the School of Hotel and Tourism Management, Hong Kong Polytechnic University.
His research interests are in technology applications to tourism and information management. Prof. Law is an
active researcher. He has edited four books and published hundreds of research articles in first-tier academic
journals. His publications have received 22,000þ citations. In addition, Prof. Law serves in different roles for
160þ research journals and is a chair/committee member of more than 140 international conferences.
Wassiuw Abdul Rahaman is an Economist and a lecturer at the Department of Economics, University of
Ghana, Legon. He is a graduate of the University of Nottingham, UK (where he obtained his PhD in
Economics) and the University of Ghana, Legon (where he obtained his MPhil Economics and BA Eco-
nomics with Sociology degrees). Wassiuw has 10-years teaching experience at the university level. He has
published in peer-review journals. His research interests include tax reforms in developing countries,
welfare effects of price increases and tax reforms, analysis of household survey as well as poverty dynamics
of commodity price increases.