Content uploaded by Victor Olorunsola
Author content
All content in this area was uploaded by Victor Olorunsola on Mar 21, 2024
Content may be subject to copyright.
Empirical Article
Tourism Economics
2023, Vol. 0(0) 1–22
© The Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/13548166231207132
journals.sagepub.com/home/teu
The predictive roles of financial
indicators and governance scores on
firms’emission performance in the
tourism and hospitality industry
Victor Oluwafemi Olorunsola
Eastern Mediterranean University, Turkey
Mehmet Bahri Saydam
Eastern Mediterranean University, Turkey
Hasan Evrim Arici
Kastamonu University, Turkey
Mehmet Ali K ¨
oseoglu
Metropolitan State University, USA
Abstract
The tourism and hospitality (T&H) industry can significantly impact the environment by using
natural resources and waste generation. Measuring and improving firms’emission performance in
this industry is becoming increasingly important. However, the prediction of financial indicators and
governance scores on firms’emission performance is still poorly established. Drawing on the
agency and stakeholders’theory, our research explores whether financial indicators and gover-
nance scores predict firms’emission performance in the T&H industry. Data on all companies that
are publicly traded was acquired from Thomson Reuters Eikon using 485 publicly traded tourism
firms as of the end of the fiscal year 2021. Our findings suggest that governance pillar score (GP),
management score (MS), board size (BS), and market capitalization (MC) are the best four pre-
dictors of carbon emission reduction scores in T&H companies. Theoretical and practical im-
plications are presented, and directions for further research are provided.
Keywords
emission score, financial indicators, governance score, machine learning algorithms, tourism and
hospitality
Corresponding author:
Mehmet ali K ¨
oseoglu, College of Management, Metropolitan State University, 1501 Hennepin Avenue, Minneapolis, MN
55403-1897, USA.
Email: koseoglu.mehmetali@gmail.com
Introduction
The need for environmentally friendly products and services has skyrocketed during the past two
decades (Elkhwesky et al., 2022;Sharma et al., 2020). From a business standpoint, there is a
growing demand for businesses of all sizes to enhance their environmental performance by adopting
novel environmental initiatives that consider pressing societal and ecological problems (Mishra
et al., 2022). Therefore, tourism and hospitality (T&H) organizations strive to be financially and
environmentally sustainable (Shafiullah et al., 2023;Sharma et al., 2020). Thus, innovative sus-
tainable revenue models, sustainable service delivery systems, eco-friendly technology capabilities,
and environmentally evolved management practices have come to be known as “eco-innovations”
in T&H (Songur et al., 2022). As concerns about climate change continue to grow, the pressure on
industries to reduce their carbon emissions increases (Yu et al., 2022). This pressure is especially
intense in the T&H industry, which is acknowledged as a vital contributor to global emissions
(Hamrouni et al., 2023;Mishra et al., 2022).
The T&H industry, being a service-oriented industry, has garnered attention for its potential to
minimize carbon emissions over extended periods (Sarkodie, 2021). In order to facilitate com-
prehensive policymaking for the T&H industry, it becomes imperative to routinely examine its
influence on greenhouse gases and air pollutants. It’s worth noting that different subsectors within
the T&H realm exhibit distinct performances and varying contributions to environmental con-
tamination within the shared business landscape. Although the T&H industry significantly drives
global economic expansion, this growth comes at the cost of environmental deterioration and
pollution (Farooq et al., 2023;Schubert and Schamel, 2021). The industry’s reliance on energy
consumption for transportation and day-to-day operations notably contributes to the environmental
burden. Encompassing diverse subindustries such as accommodation, entertainment, and casinos,
the T&H industry assumes a multifaceted nature. Relevant data underscores that the T&H industry
is accountable for approximately 5% of annual carbon emissions, underscoring its ecological
footprint (Wang, 2023).
Therefore, it is essential to identify the factors that may influence the emissions performance of
firms in T&H. Financial indicators and governance scores are two critical factors that play a role in
firms’emissions performance. Financial indicators such as profitability, leverage, and liquidity are
commonly used to assess a firm’sfinancial health and performance (Atayah et al., 2022). Gov-
ernance scores, on the other hand, measure a firm’s adherence to good governance practices,
including transparency, accountability, and ethical behavior (ElGammal et al., 2018).
Extant literature reveals outcomes of both financial indicators and governance scores on firms’
performance. For example, Rajesh and Rajendran (2020) collected data from 1840 firms from
Thomson Reuters. They found a positive and significant relationship between environmental,
social, and governance (ESG) scores, including governance scores, as a predictor of the sus-
tainability performance of firms. Burhan and Rahmanti (2012) analyzed the sustainability per-
formance with the financial performance of Indonesian companies from 2006 to 2009 utilizing
Return on Assets. Vinayagamoorthi et al. (2015) examined how corporates’sustainable efficiency is
impacted by their income as well as endeavors toward safeguarding the environment. Using the data
of 30 Airline firms from Thomson Reuters, Kuo et al. (2021) revealed that corporate social re-
sponsibility practices increase financial performance. Boutabba (2014) found that financial success
exerts a long-run active effect on carbon emissions in India. Yang et al. (2022) found that financial
development has an apparent active impact on carbon emissions. Although the aforementioned
studies give a clue regarding the financial and governance scores and their outcomes, there has not
been a study on whether they predict firms’emission performance among global firms.
2Tourism Economics 0(0)
Our research makes several contributions to the existing T&H literature. First, we establish the
validity of the theory’s function in studying the link between financial indicators, governance scores,
and emission performance by employing machine learning (ML) algorithms with a comprehensive
application rather than surveys or text analysis. Compared to applying a survey-based or text-mining
analysis of user-generated content (UGC), assessing not just the linear correlation but non-linear
correlations between prospective predictors and the results gives a rigorous evaluation method.
Second, by understanding the predictive roles of financial indicators and governance scores on
emissions performance, firms in the T&H industry can make informed decisions about investing in
sustainability and reducing their environmental impact. This can help them meet the demands of
stakeholders, comply with regulations, and maintain their competitive edge in an increasingly
environmentally conscious market.
Finally, because the majority of the correlations that this research looks at have never been
studied before, our findings provide insight into how financial indicators and governance scores
jointly affect emissions performance in T&H organizations. By targeting the best determinants of
emissions performance, ML solutions offered substantial statistical evidence for the knowledge.
Multinational businesses gradually recognize the importance of greenhouse gas emissions and
overreliance on carbon-based fossil fuels. Our paper’s current proposed analytical technique
provides a thorough knowledge of how carbon emissions are mitigated.
Literature review
Theoretical underpinning
This research adopted agency and stakeholders’theories to understand the roles of financial indicators
and governance scores on firms’emission performance in T&H industry. According to agency theory
(Jensen and Meckling, 1976), the interests of the principal (i.e., the shareholders) and the agent
(i.e., management) may not always align. Akin to agency theory, improving emission performance may
ensure the synchronization of management and the interests of shareholders, hence avoiding agency
conflicts (Guthrie and Parker, 1990). Shareholders may emphasize profit maximization in the context of
emission performance, but management may prioritize social responsibility. Revenue, profitability, and
leverage can all serve as incentives or restrictions for management to reduce emissions.
For example, a firm with high leverage may have limited financial resources to invest in emission
reduction initiatives. If a company is prosperous, it may have more significant resources to engage in
emission reduction methods, which can improve its reputation and attract socially conscious investors.
As a result, financial indicators can influence management behavior and, as a result, the firm’s emission
performance. Furthermore, according to Ben-Amar et al. (2017), carbon performance disclosure has
proved to be an effective communication technique for ensuring accountability and transparency.
Akin to stakeholder theory, a firm requires theassistance of its stakeholders toguaranteelong-range
achievement (Roberts, 1992). As a result, corporations alter their operations to balance stakeholders’
competing interests and demands (Ansoff, 1965). Firms may be responsible for decreasing their
environmental effect to fulfill their stakeholders’expectations in the context of emission performance.
Corporate governance may play an essential role in ensuring businesses meet their environmental
obligations. Firms with greater governance ratings (e.g., are more likely to have improved monitoring
and control processes, which can enhance their environmental performance).
To fulfill the expectations of their stakeholders, such businesses are more likely to implement
ecologically beneficial activities such as employing renewable energy and decreasing waste.
Therefore, governance scores can be used as an indicator of a firm’s environmental responsibility,
Olorunsola et al. 3
which can attract socially responsible investors and customers. This scientific research deployed a
dual theoretical base to explain this theoretical framework exploring the predictive roles of financial
indicators and governance scores on firms’emission performance in T&H industry. Hence, financial
indicators can serve as incentives or constraints for management to reduce emissions, and gov-
ernance scores can be used to assess a company’s environmental responsibility.
Financial indicators
Over the last 30 years, academic fascination with evaluating T&H performance has grown.
However, although it has gained recognition as a legitimate area of scholarly investigation, and
numerous articles have been published in this timeframe, the complete definition of T&H
performance remains incomplete (Sainaghi et al., 2017). The T&H industry contributes sig-
nificantly to global greenhouse gas emissions. As a result, there is growing interest in dis-
covering indicators that might predict a company’s emission performance (Huang et al., 2015).
T&H industry is a diverse and complex sector that includes hotels, restaurants, bars, resorts, and
other related businesses (Khan et al., 2021). Financial management is crucial to these orga-
nizations’success, enabling them to make informed decisions about pricing, revenue man-
agement, cost control, and investment (Hayden et al., 2022). Financial indicators have been
validated as a predictor of financial performance (Kabir and Chowdhury, 2023). Revenue per
available room (RevPAR) is one of the most significant financial measures in the hotel industry
(Chattopadhyay and Mitra, 2019). RevPAR is a fundamental indicator hotel managers use to
assess the efficacy of their pricing strategies and revenue management techniques, as it
quantifies the average revenue generated per room. A higher RevPAR shows that a hotel may
charge higher prices and/or fill more rooms, resulting in good hotel revenue performance and
higher profits (Barreda et al., 2017). Another commonly used financial indicator in the hos-
pitality industry is gross operating profit per available room (GOPPAR), which takes into
account not only room revenue but also revenue from other departments such as food and
beverage, spa, and conference facilities (Mun et al., 2019;Lee et al., 2019). GOPPAR is a
profitability metric that is derived from gross operating profit, specifically EBITDA (Earnings
Before Interest, Taxes, Depreciation, and Amortization). Unlike RevPAR, which considers only
the revenue generated per available room, GOPPAR takes into account both the revenue and
certain costs associated with the hotel’s operations (Lee et al., 2019). Hotel management may
enhance their hotel’sfinancial performance by reviewing GOPPAR and identifying areas where
they can raise revenue or cut expenditures. Aside from RevPAR and GOPPAR, other key fi-
nancial metrics in the hotel business include average day rate (ADR), occupancy rate, and net
operating income (NOI). Although the abovementioned research and information are given,
none focused on the predictive roles of financial indicators on firms’emission performance in
the T&H industry. This research lacuna also appeared in the past and recent review studies done
by Altin et al. (2018) and Assaf and Josiassen (2012) study. Hence, our study will fill this
research gap by exploring predictive role of financial indicators on firms’emission performance
in T&H industry.
Governance score
Governance scores are a widely used technique in several organizations for analyzing the efficacy of
governance activities. There is currently a profusion of governance scores and their usage in re-
search and policymaking (Gisselquist, 2014). Investors, governments, and international
4Tourism Economics 0(0)
organizations can use these ratings to assess the risk associated with investing or conducting
business in a certain country (Kaufmann et al., 2011). Governance ratings that try to capture diverse
dimensions of governance performance have proliferated in recent years (Knack and Keefer, 1995).
The Corporate Governance Quotient (CGQ), established by Institutional Shareholder Services
(ISS), was one of the first and most extensively used corporate governance ratings (Epps and
Cereola, 2008). The CGQ assesses a company’s governance policies by examining board com-
position, CEO remuneration, and shareholder rights. Institutional investors use the score to measure
the strength of a company’s governance processes and to inform their investment decisions.
Governance mechanisms are typically classified into internal and external categories (Gillan,
2006). Internal mechanisms are intended to foster a shareholder-oriented mindset and effective
censoring of managers, and they aim to bring into line the interests of managers and shareholders (Li
and Singal, 2022). The board of directors is at the pinnacle of internal governance. The board is
elected by shareholders to oversee management and possesses the authority, ability, and expertise to
“fire,”“hire,”“compensate,”“monitor,”and advise top management on their behalf (Li and Singal,
2022). On the other hand, external governance mechanisms are outside of the control of share-
holders and the board, and they either complement or substitute for internal governance mechanisms
(Brown et al., 2011). The role of the board of directors is an important part of corporate governance.
According to research, the board of directors’membership may considerably influence a company’s
success. According to Adams and Ferreira (2009), businesses with more independent members on
their boards have higher values and perform better.
Previous research has investigated the relationship between governance and sustainability efforts.
For instance, Goud (2022) examined the impact of board characteristics on carbon emission per-
formance and found that “board meetings,”“board size,”“CEO duality,”“board gender diversity,”and
“ownership concentration”have a negative relationship with carbon emission performance.
Carbon emission score
Climate change has become a worldwide issue for humanity’s future since carbon emissions have risen
substantially (Denton et al., 2020). T&H industries’carbon emissions continue to rise (Wang et al.,
2023). T&H is estimated to generate 5%–8% of total carbon emissions that cause climate issues, with
hotel services accounting for 20% (Kim, 2020). Carbon emissions from the lodging sector harm the
environment, consequently leading greenhouse effect, particularly in metropolitan areas (G¨
ossling and
Scott, 2018). As a result,there is a needto research predictors of carbon emission performance. Hence,
many studies have been conducted to investigate the impact of carbon emissions in this sector and the
role that hotels and other tourism-related enterprises may play in lowering their carbon footprint.
Some businesses enthusiastically joined initiatives to decrease carbon emissions, create climate-
related technology, and use limited resources to demonstrate successful sustainability results such as
higher carbon emissions performance (Alkurdi et al., 2023). Furthermore, schemes such as the
carbon offset program and sustainable hospitality and tourism practices were introduced into T&H
to bolster the sustainability strategy. According to Denton et al. (2020), these schemes efficiently
reduce carbon emissions in the hotel business. In the same light, implementing sustainable hos-
pitality and tourism practices minimizes the negative impact of tourism on the environment and
local communities. Over the past few years, the carbon footprint has become one of the most
important environmental protection indicators (
ˇ
Cuˇ
cek et al., 2012). Carbon footprint usually stands
for the amount of carbon and other greenhouse gases emitted over the full life cycle of a process or
product (Perˇ
ci´
c et al., 2020).
Olorunsola et al. 5
Research design
Sample and data
Drawing on past investigations (e.g., Torres and Augusto, 2021;Uyar et al., 2022), the data utilized in
this analysis were obtained from the Thomson Reuters Eikon database. The database includes, along
with many other items, reports on financial markets, financial indicators for businesses, data on CEOs,
roles of executive directors, and statistics on governance mechanisms for many publicly traded en-
terprises in over 150 countries. The primary sample for this article includes 652 firm-year data from
2019, 2020, and 2021. The years 2020 and 2021 were profoundly impacted by the global COVID-19
pandemic, which introduced unprecedented challenges and disruptions to the business environment.
Three primary criteria were used to get the functional sample for the current investigation: (1)
solely businesses involved considering the tourism industry were acknowledged (Kılıç et al., 2021);
(2) solely businesses with financial- and governance mechanism-related scores calculated by
Thomson Reuters were evaluated; and (3) records with missing data were excluded. Due to this
filtering process, the final sample contained 485 publicly traded firm-year data from 2019 to 2021.
Variable selection
In our study, we assessed potential predictors that could effectively estimate firms’emission scores.
These predictors were drawn from financial indicators and governance scores. On the financial side,
we considered indicators such as EBITDA Margin (EM), Operating Margin (OM), Net Margin
(NM), Asset Turnover (AT), Current Ratio (CR), % LT Debt to Total Capital (DT), and ROIC (RC).
On the governance front, we incorporated Governance Pillar Score (GP), Management Score (MS),
and Shareholders Score (SS) as key scores.
Additionally, we explored the impact of certain firm characteristics on the data analysis. These
characteristics encompassed variables like year (TI), market capitalization (MC), independent board
members score (IB), and BS. It’s important to note that financial data consisted of various financial
ratios used as predictor variables. Furthermore, governance scores were assigned on a scale from
zero to 100, with 100 representing the highest rating achievable for a company (refer to Table 1).
Method
We used R programming and basic forward selection techniques to find significant variables to run
multiple regression and best subset regression. The regularized regression model, including lasso
regression, was the second method we used to include predictors. For selecting relevant forecasting
components and lowering the accuracy of variables within a collection, lasso regression is a useful
technique (Wang et al., 2007). Based on the conventional regression approach, the lasso model
computes estimations by reducing the following mechanism β
j
:
X
n
i¼1yiX
j
xijβj2þλX
p
j¼1
βj
where iis the indicator of each witness in the analysis and nrepresents the overall amount of
incidences. The normalizing component is integrated into the reduction procedure specified in the
formula in order to retain the moderate regression coefficients β
j
.“normalization”refers to how
much the punishment level λhas been raised. When λ= 0, the lasso regression generates an estimate
using the ordinary least squares (OLS) method. Some coefficients are considered as zero due to the
6Tourism Economics 0(0)
penalty component, which is embodied in the equation’s constant. As a result, lasso regression is a
widely used approach for estimate. The value of a lasso regression is substantial (Tian et al., 2021).
With the aid of lasso regression, the model was then completed.
Second, the standardized regression analysis attempts to forecast the relationship between a
dependent variable Yand a feature vector Xby taking into consideration a number of predictor-
related characteristics, X
1
,X
2
, ..., X
p
from the predictor X:
ϵ
where ϵis the error term. Carbon emission reduction score is the outcome variable Yin our formula
and the features Xare EM, OM, NM, AT, CR, DT, RC, GP, MS, SS, TI, MC, IB, and BS.
According to Svanberg et al. (2022), using a ML technique allows for less subjectivity in the
study of relative importance and predictive validity. No other approach we are aware of may
combine even the majority of the variables we analyze and discover patterns in those linked with the
chance to achieve a carbon emission performance. For example, multicollinearity makes regression
in explanatory modeling problematic, but this issue does not limit the predictive capacity of
Table 1. Variables.
Variables Definition
Predictor variables
EBITDA margin (EM) The ratio of earnings before interest, taxes, depreciation, and amortization.
Net margin (NM) The ratio of net income generated from a company’s revenue.
Current ratio (CR) The ratio that indicates if a company has enough resources to meet its short-
term obligations.
Asset turnover (AT) The ratio of total sales or revenue to average assets.
LT debt to total
capital (DT)
The ratio that measures how much financial leverage a company has and if it’s
funded mainly through debt.
Return on invested
capital (RC)
The ratio of return a company makes on the cash it invests in its business.
Operating margin (OM) The ratio measuring revenue after the deduction of operating expenses.
Governance pillar
score (GP)
It measures the company’s governance performance based on verifiable
reported data in the public domain.
Management score (MS) It measures a company’s commitment and effectiveness towards following best
practice corporate governance principles.
Shareholder score (SS) It measures a company’s effectiveness towards equal treatment of shareholders
and the use of anti-takeover devices.
Outcome variables
Carbon emission score This is the emission reduction score that measures a company’s commitment
and effectiveness towards reducing environmental emission in the production
and operational processes.
Business characteristics
Board size (BS) The total number of executives on the board of each sample company.
IBMS (IB) The percentage of independent board members (%)
Market capitalization
(MC)
The total value of all a company’s shares of stock.
Year (TI)
Note. The variables and their definitions were adopted from the Thomson Reuters Eikon. Source: https://www.esade.edu/itemsweb/
biblioteca/bbdd/inbbdd/archivos/Thomson_Reuters_ESG_Scores.pdf
Olorunsola et al. 7
algorithmic modeling (D’Amato et al., 2022;Vaughan and Berry, 2005), allowing ML algorithms to
contain a diverse collection of variables (Svanberg et al., 2022). Thus, as an alternative to traditional
approaches like linear modeling, we use a ML method to estimate the parameter f(.) and examine the
usefulness of different approaches and ensemble methods (bagging, random forest (RF), and
boosted regression). By pooling the estimates of many predictors, ensemble approaches aim to
improve the generality and resilience of a single estimate. They are sometimes separated into two
categories: standard techniques and boosting approaches. The latter (for instance, RF) separately
develops a number of estimation methods and combines their predictions. The ensemble estimator is
typically better than any single estimator since it has lower volatility. The latter creates fundamental
estimators incrementally to lessen error (for instance, boosted regression). Several weak estimation
techniques are combined to form the ensemble estimator. Here, we provide a quick explanation of
the algorithms used.
ML systems’reliability and efficiency were improved by the development of bagging. This
method creates a lot of bootstrap samples from the training data and gives each sample a weak
learner. It eventually collects the low scores by averaging them all. Predictor evaluation varies from
bagging in the RF. The method prevents prominent predictors from overwhelming the subdivisions
of each tree by selecting a random subset of determinants from the final prediction set as subdivision
contenders at each splitting. A RF introduces a random disturbance into the learning system to
identify the trees and combine their predictions using an aggregation strategy (Breiman, 2001). The
RF algorithm is
_
fRF ðXÞ¼1
B
B
b1b
fDTðXjbÞ, where Bis the overall amount of bootstrap samples, and
b
fDTðXjbÞis the decision trees (DT) estimator over the b2Bsample.
Moreover, the methods evaluate the relevance of factors using Mean Decrease Accuracy (%
IncMSE) and Mean Decrease Gini (IncNodePurity). By comparing the variable’s prohibited test set
estimate to the actual estimate, %IncMSE is determined (normalized by the standard error).
Additionally, IncNodePurity assesses the integrity of a split for each parameter (node) in a tree using
the Gini Index. A higher IncMSE and IncNodePurity percentage denotes the variable’s importance
(Soner et al., 2018). If both %IncMSE and %IncNodePurity are high, the predictor is seen to be very
meaningful.
We evaluate each predictor’s predictive capability in elucidating the outcome variable Yusing the
RF technique and the subsequent modeling.
Y∼EM + OM + NM + AT + CR + DT + RC + GP + MS + SS + TI + MC + IB + BS
We first utilize the outcome variable, Y
Emission
, for the emission reduction score. b
Ystands for the
result variable’s RF method. In order to identify the appropriate factor to select, we perform hyper-
parameter tuning after first considering a set of 100 random seeds for the pseudo-random engine
and a reasonable number of trees (trees = 2722). The two hyper-parameters that need to be
simplified are node size and try, which stand for the lowest cut-off of a single node and the number
of forecasters selected at each splitting node, respectively. For example, the setting may =
13 denotes that for every split, one of the 13 parameters is chosen at random from a candidate
pool. The seed, try, and node size combination that yields the lowest mean squared residuals is
chosen, MSE ¼1
J:njPjϵJPiϵRjðyib
yRjÞ2, with n
j
being the number of observations in the area R
j
,
and largest rate of explained variance RSS ¼PjϵJPiϵRjðyib
yRjÞ2.
We divided the data set into training and test sets using the 80% (396 observations) - 20%
(89 observations) splitting rule. The following adjustments are made following the criterion tuning:
try = 13 and node size = 1 are the values for the outcome variable, Y
Emission.
As can be seen in
Table 2, R-squared estimates are used to calculate each model’s estimated coefficients. The best
8Tourism Economics 0(0)
degree of accuracy in predicting the outcome—emission reduction score—is offered by the en-
semble methods, boosted regression algorithms.
Friedman (2001) used boosted regression trees using gbm method, also known as stochastic
gradient boosting, which uses fixed-sized decision trees as weak forecasting analytics (Leathwick
et al., 2006). Unlike the RF, which parallelizes the tree-building procedure, the forecast is created by
a linear program. Each tree is modified upon the findings of the previous trees to improve the
existing fit. A boosted regression trees approach can estimate connections between predictors,
manage numerous types of regression models and missing data, and is unaffected by extreme
outliers or the insertion of unnecessary factors (Friedman and Meulman, 2003).
Finally, although the aforementioned algorithms - bagging, RFs, and boosted regression - have
become increasingly prevalent nowadays in automated forecasting, particularly when coping with
massive empirical records that are contrary to the rigid presumptions enforced by conventional
statistical methods (e.g., OLS regression that presumes a linear manner, uniformity, and normality),
understanding the outcomes of such equations and explaining them to executives may prove
difficult (Greenwell, 2017). To address this worry, we also used partial dependency plots (PDPs),
which are low-dimensional visual illustrations of the predicting functionality, to help understand the
link between the result and predictions of concern.
Findings
To examine the link between each pair of factors in the dataset, we display their correlation in the
correlogram shown in Figure 1 (positive relationships are shown in blue, while negative correlations
are shown in red). The intensity of the color is related to the correlation coefficient. As is obvious,
we find positive connections between MC and GP, which are then connected with carbon emission
performance. Negative associations are quite rare. The correlogram detects the linear relationship
between a collection of variables, but ML algorithms may detect non-linear patterns and hidden
associations (D’Amato et al., 2022).
Subset regression models (forward step selection)
First, we utilized the basic variable selection method—the forward step. To formulate the best
model, we consider two selection criteria of subset regression models—Bayesian Information
Criterion (BIC) and Adjusted R2 Criterion (ADJR2) (see Table 3). The results show that for BIC, six
predictors of emission reduction score give the best model (MC, BS, AT, GP, MS, and SS), while for
ADJR2, 11 predictors of emission reduction score give the best models (TI, MC, BS, EM, NM, AT,
DT, RC, GP, MS, and SS).
Table 2. R2 values for test data.
Carbon emission score
Model BIC ADJR2 Lasso-optima Relaxed lasso-optimal Bagging RF BR
R2 0.62 0.58 0.59 0.58 0.39 0.35 0.75
Note. BIC is the Bayesian information criterion, ADJR2 is adjusted R-squared, RF is the random forest, and BR is boosted
regression trees.
Olorunsola et al. 9
Lasso regression
After these two forward selection methods, we employed lasso regression optimal. The results are
presented in Figure 2. According to these findings, the best three predictors of carbon emission
reduction score in T&H companies are RC, EM, and NM. Among them, RC and NM are negative
predictors, while EM has positively estimated carbon emission reduction scores in T&H companies.
This shows that the carbon emission performance of T&H firms has been positively affected by EM
scores but negatively affected by NM and ROIC scores. Regarding governance mechanisms,
interestingly, only the GP score has positively estimated emission performance, while other
mechanisms included (MS and SS) have negatively predicted emission scores in these firms.
Overall, most of the influential predictors have a negative effect on carbon emission reduction scores
in T&H companies.
Ensemble methods
The tree-based algorithms were run on the data to examine the effectiveness of the predictors
defined on carbon emission reduction scores. Aside from strong forecast accuracy, the algorithms
used show the most critical aspects of predictors’effectiveness. The outcomes of the algorithms
were contrasted to illustrate the competence of bagging, RF, and boosted regression on the
Figure 1. Correlation matrix.
10 Tourism Economics 0(0)
prediction of the carbon emission scores. Hence, Figures 3–6depict the variable importance of
bagging, RF, PDP, and boosted regression for the T&H firms’carbon emission performance.
By the bagging algorithm, concerning the %IncMSE, MC, GP, and BS are the most effective
predictors of carbon emission reduction scores (Figure 3). Other five influential predictors that
importantly estimate carbon emission reduction ratings are AT, IB, CR, SS, and MS.
In Figure 3, according to IncNodePurity, GP has the best-estimated carbon emission scores,
followed by MS and BS. Other influential predictors estimating the carbon emission performance of
T&H companies are IB, AT, MC, DT, CR, SS, EM, NM, RC, OM, and TI, respectively, according to
IncNodePurity.
Both plots of bagging and RF algorithms (Figures 3 and 4) indicate that the MC, GP, BS, and MS
are strongly related to the median value of emission reduction ratings. The issue then becomes,
“What are the characteristics of these associations?”To assist in explaining this, consider carbon
emission performance on MC and BS, both separately and combined. A partial dependence plot
(PDP) is “especially useful for visualizing the relationships discovered by complex ML algorithms
such as a bagging”(Greenwell, 2017, p. 423). Thus, we adopted PDP visualization to perfectly
represent how the two best predictors explored by the ensemble methods (i.e., bagging) function in
estimating the carbon emission performance of H&T companies (see Figure 5). For a given
combination of MC and BS, the PDP plot reflects the estimated emission reduction rate interval. For
example, for a T&H company with a MC rate of around 30 and BS 15, the predicted emission
reduction score is more than 65% based on the algorithm (see the left panel in Figure 5). Another
plot based on the bagging and RF algorithms is also shown in the middle panel to provide additional
evidence of the relationship between MC and BS. To continue to demonstrate the use of PDPs, the
right panel in Figure 5 shows joint PDPs for carbon emission performance for its two most important
variables: MC and BS. Because they display the joint partial dependence on two variables in a single
plot, the plots are now three-dimensional, showing the different interactions of the pairs of variables
involved. According to the 3-D plot, the carbon emission performance is at its highest rate when the
MC is low, but BS is high (see light green and yellow color in the plot). In other words, when the
Table 3. Forward selection results.
BIC ADJR2
Predictors
Outcome Outcome
Carbon emission score Carbon emission score
Intercept 14.44 8.15
MC 4.10 4.76
BS 1.63 1.71
AT 7.77 4.59
GP 3.75 3.75
MS 2.45 2.45
SS 0.75 0.76
TI 1.37
EM 11.91
NM 5.35
DT 4.29
RC 13.28
Olorunsola et al. 11
business size increases in a relatively small MC, the probability of reporting a high emission
reduction score increases.
Furthermore, our research demonstrated the variable importance to the extent that the best
algorithm, the boosted regression (Figure 6). This plot shows that the most important variable in
explaining the emission reduction score selected by the boosted regression algorithm is GP.). The
other robust predictors significantly estimating emission ratings in T&H companies are MS, CR, SS,
AT, MC, and IB.
We aim to discover how the predictors defined affect the carbon emission performance of T&H
companies. All the algorithms have discovered similar variables (i.e., GP, MS, MC, CR) as the
strongest predictor in estimating emission ratings in these companies.
Discussion and conclusion
Most of the literature examining the predictive roles of financial indicators and governance scores
on firms involves theoretical propositions and limited empirical examination of the firm perfor-
mance (Bahadori et al., 2021;Kotane and Kuzmina-Merlino, 2012). Therefore, our research was
inspired by the dearth of robustness analysis of the link between financial indicators and governance
scores and by the lack of research on firms’emission performance by using agency theory and
stakeholders’theory as a theoretical underpinning. Broadly speaking, as among the first in the field,
our research has great potential to contribute to the firms’emission performance literature. We found
support for the predictive roles of financial indicators and governance scores on the firms’emission
performance in T&H firms. However, our findings deserve further explanation. For instance,
bagging and RF algorithms were able to find predictor roles of financial indicators and governance
Figure 2. Lasso regression results.
12 Tourism Economics 0(0)
scores on emission performance. According to the bagging algorithm, the best three predictors of
carbon emission reduction scores in T&H companies are MC, GP, and BS. RF algorithm, on the
other hand, shows that GP has the best-estimated carbon emission scores, followed by MS and BS.
As seen, GP and BS are the most influential estimators of carbon emission performance in global
T&H companies. Moreover, PDP visualization has perfectly represented how the two best pre-
dictors (MC and BS) explored by %InsMSE RF algorithm in estimating the carbon emission
performance of T&H companies. This visualization endorses that when the business size increases
in a relatively small MC, the probability of reporting a high emission reduction score increases.
We also performed the best algorithm (i.e., boosted regression) to select the most important
variable in estimating carbon emission performance. The plot generated by the boosted regression
algorithm has confirmed the previous algorithms by revealing GP as the most important estimator of
carbon emission performance in global T&H companies.
Furthermore, our view of financial indicators as well as governance scores, which is depicted by
several “formalized”and “primary”edges commenced voluntarily by a company’s administration
(Rajesh and Rajendran, 2020;Sachin and Rajesh, 2022), analyzed via the companies’freely
published data. By operationalizing our measure using such data, this study provides an alternative
way to assess the carbon emission performance of T&H firms using two predictors of carbon
emission performance: financial indicators and governance scores. Past and recent investigations
have comprehensively pinpointed the financial indicators in the performance outcomes (e.g.,
Iacuzzi, 2022;Ziaei, 2015). In contrast, our innovative approach reveals that it is conceivable to
conduct a wide-ranging investigation of several features of financial indicators using T&H firms’
reports.
Figure 3. Bagging algorithm results.
Olorunsola et al. 13
Sustainability reporting and performance are still limited and fragmented, with little im-
provement in sustainable performance, despite the substantial growth in sustainability reporting
literature over the past few decades (Ong and Djajadikerta, 2020). Corporate governance mech-
anism, which comprises the system of rules, practices, and processes by which a company is
Figure 4. Random forest algorithm results.
Figure 5. PDP visualization of the interaction effect of two predictors (MC and BS) on carbon emission
performance.
14 Tourism Economics 0(0)
directed and controlled, plays a crucial role in the quality of sustainability reporting and perfor-
mance, according to previous research (Kumar et al., 2022). However, does the same hold true for
the predictive power of financial indicators on global T&H firms’emission performance? We
endeavor to shed some light on this question.
Previous literature on the relationship between corporate governance and firm performance has
been mixed, with some finding a positive, some a negative, and some no relationship at all (e.g.,
Aguilera et al., 2021;Kumar et al., 2022). A recent review of the methodology used to study
corporate governance revealed that some contradictions across studies in the corporate governance
literature may result from corporate governance operationalization (Aguilera et al. 2021). However,
this study found that support for predictive corporate governance roles remained the same when
using ML algorithms. Our findings support the previous studies empirically revealing the significant
relationship between corporate governance and emission performance (Elsayih et al., 2021;Goud,
2022;Oyewo, 2023).
This study analyses the feasibility of utilizing ML to generate a model of optimal financial
predictors and corporate governance scores of emission performance scores in global T&H
companies. Such a model would have to predict whether financial indicators and governance scores
have a predictor role on emission performance scores alone. As mentioned above, our findings
revealed the predictive roles of financial indicators and governance scores on T&H firms’emission
performance in T&H firms. Broadly speaking, the common best predictor of carbon emission
performance in T&H companies is GP according to three ensemble models. Other important
predictors are MC, MS, and BS. This shows that the carbon emission performance of T&H firms has
been mostly estimated by governance mechanisms, including GP score and MS, and business
characteristics, such as business size and MC. Among the financial indicators, CR, AT, and DT have
Figure 6. Variable importance for carbon emission performance in boosted regression.
Olorunsola et al. 15
been found as significant predictors of carbon emission performance according to bagging and RF
algorithms. However, the boosted regression algorithm has revealed EM and NM as the best fi-
nancial predictors of the outcome variable. A possible reason behind the finding could be related to
that higher ratios in these financial indicators could imply more financial resources available for eco-
friendly policies, investments, and practices in global T&H companies.
Three ML algorithms with extraordinary complication-handling capabilities (forward selection,
lasso regression, and RF) outperform the other prediction models in gaining knowledge to predict
emission performance, which is a methodological contribution. Algorithms that can reflect non-
linearity, indicator interaction, and multidimensionality produce the best predictions of emission
performance. This shows that the fundamental structure provides a difficulty that explanatory
modeling and typical statistical approaches might find challenging.
In addition to this study’s theoretical and methodological contributions, it also brings new
implications for practice. Our research highlights the importance of financial indicators and
governance scores in predicting firms’emission performance. Therefore, firms in the T&H industry
should develop and implement sustainability strategies that focus on improving their financial and
governance performance, which can lead to better emission performance. First, T&H firms should
regularly monitor their financial and governance performance to identify improvement areas. This
can involve tracking financial indicators such as profit margin, return on investment, debt-to-equity
ratio, and governance scores such as board independence, executive compensation, and stakeholder
engagement. Second, our research depicts that financial indicators and governance scores can
predict firms’emission performance. Hence, organizations should consider employing environ-
mental performance metrics in their performance evaluations to incentivize and reward sustainable
practices. Third, collaboration with stakeholders is essential, considering the complex nature of
sustainability issues in the T&H industry. Organizations should collaborate with industry asso-
ciations, NGOs, governments, and local communities to address sustainability challenges and share
best practices. Lastly, transparency and disclosure of sustainability-related information can improve
accountability and trust among stakeholders. Organizations should improve their sustainability
reporting practices, including reporting on their financial, governance, and emission performance, to
boost transparency and disclosure.
Limitations and future research directions
The study has a restriction in terms of its transferability to non-included enterprises and specific
industries. Further research may look at financial and corporate governance scores in small- or
medium-sized enterprises, which could yield conflicting outcomes and consequences because they
do not have as many financial resources as bigger companies. The report suggests various research
directions. First, how financial and corporate governance scores influence the possibility of social
development may be explored to acquire knowledge, like “are enterprises with higher financial
scores and corporate governance scores support or opposed to such development?”Furthermore,
unlike our study, scientific investigations may go deeper into the roles of ESG initiatives on
emission performance. Is there a difference between the predictor roles of these initiatives, a lack of
competence on emission metrics, or they diminish the emission performance in the businesses or a
limited understanding of how to achieve these three mechanisms together in international cor-
porations? Third, the pandemic wrought unprecedented disruptions across industries, impacting
financial decisions, corporate strategies, and governance practices in unforeseen ways. The resulting
volatility in market conditions and changes in business operations during this period could have
influenced the observed relationships among variables, potentially confounding our findings.
16 Tourism Economics 0(0)
Hence, conducting a more extended longitudinal analysis, spanning both pre-pandemic and post-
pandemic periods, could help isolate the pandemic’s effects and provide a clearer view of how
financial and governance factors impact emission performance across different market contexts.
When conducting research that spans across a period affected by a significant event like the COVID-
19 pandemic, there are several potential biases to consider. The pandemic may have dispropor-
tionately impacted certain industries or companies. For example, industries related to travel and
hospitality were severely affected, while some technology companies thrived. This could lead to
selection bias if the sample is not representative of the broader business landscape. In addition,
government policies and interventions, such as stimulus packages and lockdowns, may have
influenced corporate performance and governance practices during the pandemic. Failure to account
for these policy-related factors could introduce bias into the findings. Complementing quantitative
data with qualitative case studies could provide a deeper understanding of how individual firms
navigated the challenges posed by the pandemic. This could offer insights into how specificfi-
nancial and governance strategies were adapted to address emission performance in the face of
unprecedented disruptions. Lastly, we recommend a specific analysis of the causal factors triggering
the impact of financial indicators on firms’diversity performance. As an illustration, consider how a
firm’s diversity performance is impacted by decisions on human rights, the workforce, product
responsibility, community, shareholders, management, and corporate social responsibility strategy.
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
Mehmet ali Koseoglu https://orcid.org/0000-0001-9369-1995
References
Adams RB and Ferreira D (2009) Women in the boardroom and their impact on governance and performance.
Journal of Financial Economics 94(2): 291–309. doi: 10.1016/j.jfineco.2008.10.007
Aguilera RV, Aragón-Correa JA, Marano V, et al. (2021) The Corporate Governance of Environmental
Sustainability: A Review and Proposal for More Integrated Research. Sage Publications. DOI: 10.1177/
0149206321991212
Alkurdi A, Al Amosh H and Khatib SF (2023) The mediating role of carbon emissions in the relationship
between the board attributes and ESG performance: European evidence. EuroMed Journal of Business.
doi: 10.1108/EMJB-08-2022-0144
Altin M, Koseoglu MA, Yu X, et al. (2018) Performance measurement and management research in the
hospitality and tourism industry. International Journal of Contemporary Hospitality Management 30(2):
1172–1189. doi: 10.1108/IJCHM-05-2017-0251
Ansoff HI (1965) Corporate Strategy: An Analytic Approach to Business Policy for Growth and Expansion.
McGraw-Hill Companies.
Assaf AG and Josiassen A (2012) Identifying and ranking the determinants of tourism performance: a global
investigation. Journal of Travel Research 51(4): 388–399. doi: 10.1177/0047287511426337
Olorunsola et al. 17
Atayah OF, Dhiaf MM, Najaf K, et al. (2022) Impact of COVID-19 on financial performance of logistics firms:
evidence from G-20 countries. Journal of Global Operations and Strategic Sourcing 15(2): 172–196. doi:
10.1108/JGOSS-03-2021-0028
Bahadori N, Kaymak T and Seraj M (2021) Environmental, social, and governance factors in emerging
markets: the impact on firm performance. Business Strategy & Development 4(4): 411–422. doi: 10.1002/
bsd2.167
Barreda AA, Zubieta S, Chen H, et al. (2017) Evaluating the impact of mega-sporting events on hotel pricing
strategies: the case of the 2014 FIFA world cup. Tourism Review 72(2): 184–208. DOI: 10.1108/TR-02-
2017-0018
Ben-Amar W, Chang M and McIlkenny P (2017) Board gender diversity and corporate response to sus-
tainability initiatives: evidence from the carbon disclosure project. Journal of Business Ethics 142(2):
369–383. doi: 10.1007/s10551-015-2759-1
Boutabba MA (2014) The impact of financial development, income, energy and trade on carbon emissions:
evidence from the Indian economy. Economic Modelling 40: 33–41. doi: 10.1016/j.econmod.2014.03.005
Breiman L (2001) Random Forests. Machine Learning 45(1): 5–32. doi: 10.1023/A:1010933404324
Brown P, Beekes W and Verhoeven P (2011) Corporate governance, accounting, and finance: a review.
Accounting and Finance 51(1): 96–172.
N Burhan AH and Rahmanti W (2012) The impact of sustainability reporting on company performance.
Journal of Economics, Business, and Accountancy 15(2): 257. doi: 10.14414/jebav.v15i2.79
Chattopadhyay M and Mitra SK (2019) Determinants of revenue per available room: influential roles of
average daily rate, demand, seasonality and yearly trend. International Journal of Hospitality Man-
agement 77: 573–582. doi: 10.1016/j.ijhm.2018.09.001
ˇ
Cuˇ
cek L, Klemeˇ
s JJ and Kravanja Z (2012) A review of footprint analysis tools for monitoring impacts on
sustainability. Journal of Cleaner Production 34: 9–20. doi: 10.1016/j.jclepro.2012.02.036
Denton G, Chi OH and Gursoy D (2020) An examination of the gap between carbon offsetting attitudes and
behaviors: role of knowledge, credibility and trust. International Journal of Hospitality Management 90:
102608. doi: 10.1016/j.ijhm.2020.102608
D’Amato V, D’Ecclesia R and Levantesi S (2022) ESG score prediction through random forest algorithm.
Computational Management Science 19(2): 347–373. doi: 10.1007/s10287-021-00419-3
ElGammal W, El-Kassar A and Canaan Messarra L (2018) Corporate ethics, governance and social re-
sponsibility in MENA countries. Management Decision 56(1): 273–291. doi: 10.1108/MD-03-2017-0287
Elkhwesky Z, Salem IE, Varmus M, et al. (2022) Sustainable practices in hospitality pre and amid COVID-19
pandemic: looking back for moving forward post-COVID-19. Sustainable Development 30(5):
1426–1448. doi: 10.1002/sd.2304
Elsayih J, Datt R and Tang Q (2021) Corporate governance and carbon emissions performance: empirical
evidence from Australia. Australasian Journal of Environmental Management 28(4): 433–459. doi: 10.
1080/14486563.2021.1989066
Epps RWand Cereola SJ (2008) Do institutional shareholder services (ISS) corporate governance ratings reflect
a company’s operating performance? Critical Perspectives on Accounting 19(8): 1135–1148. doi: 10.
1016/j.cpa.2007.06.007
Farooq U, Tabash MI, Saleh Al-Faryan MA, et al. (2023) The nexus between tourism-energy-environmental
degradation: does financial development matter in GCC countries? Tourism Economics: The Business and
Finance of Tourism and Recreation. doi: 10.1177/13548166231174812
Friedman JH (2001) Greedy function approximation: a gradient boosting machine. The Annals of Statistics
29(5): 1189–1232. doi: 10.1214/aos/1013203451
Friedman JH and Meulman JJ (2003) Multiple additive regression trees with application in epidemiology.
Statistics in Medicine 22(9): 1365–1381.
18 Tourism Economics 0(0)
Gillan SL (2006) Recent developments in corporate governance: an overview. Journal of Corporate Finance
12(3): 381–402. doi: 10.1016/j.jcorpfin.2005.11.002
Gisselquist RM (2014) Developing and evaluating governance indexes: 10 questions. Policy Studies 35(5):
513–531. doi: 10.1080/01442872.2014.946484
G¨
ossling S, Scott D and Hall CM (2018) Global trends in length of stay: implications for destination
management and climate change. Journal of Sustainable Tourism 26(12): 2087–2101. doi: 10.1080/
09669582.2018.1529771
Narsa Goud N (2022) Corporate governance: does it matter management of carbon emission performance? an
empirical analysis of Indian companies. Journal of Cleaner Production 379: 134485. doi: 10.1016/j.
jclepro.2022.134485
Greenwell BM (2017) PDP: an R Package for constructing partial dependence plots. The R Journal 9(1): 421.
Guthrie J and Parker LD (1990) Corporate social disclosure practice: a comparative international analysis.
Advances in Public Interest Accounting 3: 159–175.
Hamrouni A, Karaman AS, Kuzey C, et al. (2023) Ethical environment, accountability, and sustainability
reporting: what is the connection in the hospitality and tourism industry? Tourism Economics 29(3):
664–695. doi: 10.1177/13548166211062649
Hayden MT, Mattimoe R and Jack L (2022) Sensemaking and financial management in the decision-making
process of farmers. Journal of Accounting and Organizational Change 18(4): 529–552. doi: 10.1108/
JAOC-11-2020-0186
Huang K, Wang JC and Wang Y (2015) Analysis and benchmarking of greenhouse gas emissions of luxury
hotels. International Journal of Hospitality Management 51: 56–66. doi: 10.1016/j.ijhm.2015.08.014
Iacuzzi S (2022) An appraisal of financial indicators for local government: a structured literature review.
Journal of Public Budgeting, Accounting and Financial Management 34(6): 69–94. doi: 10.1108/
JPBAFM-04-2021-0064
Jensen MC and Meckling WH (1976) Theory of the firm: managerial behavior, agency costs, and ownership
structure. Journal of Financial Economics 3(4): 305–360. doi: 10.1016/0304-405X(76)90026-X
Kabir MA and Chowdhury SS (2023) Empirical analysis of the corporate social responsibility and financial
performance causal nexus: evidence from the banking sector of Bangladesh. Asia Pacific Management
Review 28(1): 1–12. doi: 10.1016/j.apmrv.2022.01.003
Kaufmann D, Kraay A and Mastruzzi M (2011) The worldwide governance indicators: methodology and
analytical issues. Hague Journal on the Rule of Law 3(2): 220–246. doi: 10.1017/S1876404511200046
Khan A, Bibi S, Lyu J, et al. (2021) COVID-19 and sectoral employment trends: assessing resilience in the US
leisure and hospitality industry. Current Issues in Tourism 24(7): 952–969. doi: 10.1080/13683500.2020.
1850653
Kılıç M, Uyar A, Kuzey C, et al. (2021) Drivers and consequences of sustainability committee existence?
Evidence from the hospitality and tourism industry. International Journal of Hospitality Management 92:
102753.
Kim M (2020) A systematic literature review of the personal value orientation construct in hospitality and
tourism literature. International Journal of Hospitality Management 89: 102572. doi: 10.1016/j.ijhm.
2020.102572
Knack S and Keefer P (1995). Institutions and Economic Performance: Cross-Country Tests Using Alternative
Institutional Indicators. St. Louis: Federal Reserve Bank of St Louis.
Kotane I and Kuzmina-Merlino I (2012) Assessment of financial indicators for the evaluation of business
performance. European Integration Studies 0(6). doi: 10.5755/j01.eis.0.6.1554
Kumar K, Kumari R, Nandy M, et al. (2022) Do ownership structures and governance attributes matter for
corporate sustainability reporting? an examination in the Indian context. Management of Environmental
Quality: An International Journal 33(5): 1077–1096. doi: 10.1108/MEQ-08-2021-0196
Olorunsola et al. 19
Kuo T, Chen H and Meng H (2021) Do corporate social responsibility practices improve financial perfor-
mance? A case study of airline companies. Journal of Cleaner Production 310: 127380. doi: 10.1016/j.
jclepro.2021.127380
Leathwick JR, Elith J, Francis MP, et al. (2006) Variation in demersal fish species richness in the oceans
surrounding New Zealand: an analysis using boosted regression trees. Marine Ecology Progress Series
321: 267–281
Lee J, Jeong JH, Jung JH, et al. (2019) RevPAR vs. GOPPAR: property- and firm-level analysis. Oncology 97:
180–188. doi: 10.1016/j.annals.2019.04.006
Li Yand Singal M (2022) Corporate governance in the hospitality and tourism industry: theoretical foundations
and future research. Journal of Hospitality & Tourism Research 46(7): 1347–1383. doi: 10.1177/
10963480211011718
Mishra HG, Pandita S, Bhat AA, et al. (2022) Tourism and carbon emissions: a bibliometric review of the last
three decades: 1990–2021. Tourism Review 77(2): 636–658. doi: 10.1108/TR-07-2021-0310
Mun SG, Woo L and Paek S (2019) How important is F&B operation in the hotel industry? empirical evidence
in the U.S. market. Tourism Management (1982) 75: 156–168. doi: 10.1016/j.tourman.2019.03.010
Ong T and Djajadikerta HG (2018) Corporate governance and sustainability reporting in the Australian re-
sources industry: an empirical analysis. Social Responsibility Journal 16(1): 1–14. doi: 10.1108/SRJ-06-
2018-0135
Oyewo B (2023) Corporate governance and carbon emissions performance: international evidence on cur-
vilinear relationships. Journal of Environmental Management 334: 117474. doi: 10.1016/j.jenvman.2023.
117474
Perˇ
ci´
c M, Vladimir N and Fan A (2020) Life-cycle cost assessment of alternative marine fuels to reduce the
carbon footprint in short-sea shipping: a case study of Croatia. Applied Energy 279: 115848. doi: 10.1016/
j.apenergy.2020.115848
Rajesh R and Rajendran C (2020) Relating environmental, social, and governance scores and sustainability
performances of firms: an empirical analysis. Business Strategy and the Environment 29(3): 1247–1267.
doi: 10.1002/bse.2429
Roberts RW (1992) Determinants of corporate social responsibility disclosure: an application of stakeholder
theory. Accounting, Organizations and Society 17(6): 595–612. doi: 10.1016/0361-3682(92)90015-K
Sachin N and Rajesh R (2022) An empirical study of supply chain sustainability with financial performances of
Indian firms. Environment, Development and Sustainability 24(5): 6577–6601. doi: 10.1007/s10668-021-
01717-1
Sainaghi R, Phillips P and Zavarrone E (2017) Performance measurement in tourism firms: a content analytical
meta-approach. Tourism Management 59: 36–56. doi: 10.1016/j.tourman.2016.07.002
Sarkodie SA (2021) Failure to control economic sectoral inefficiencies through policy stringency disrupts
environmental performance. The Science of the Total Environment 772: 145603. doi: 10.1016/j.scitotenv.
2021.145603
Schubert SF and Schamel G (2021) Sustainable tourism development: a dynamic model incorporating resident
spillovers. Tourism Economics 27(7): 1561–1587. doi: 10.1177/1354816620934552
Shafiullah M, Khalid U and Okafor LE (2023) Do birds of a feather flock together? analyzing environmental
performance and tourist behavior using a gravity approach. Tourism Economics 29(5): 1139–1163. doi:
10.1177/13548166221097025
Sharma RK, Sodhi GPS, Bhakar V, et al. (2020) Sustainability in manufacturing processes: finding the en-
vironmental impacts of friction stir processing of pure magnesium. CIRP Journal of Manufacturing
Science and Technology 30: 25–35. doi: 10.1016/j.cirpj.2020.03.007
Soner O, Akyuz E and Celik M (2018) Use of tree-based methods in ship performance monitoring under
operating conditions. Ocean Engineering 166: 302–310.
20 Tourism Economics 0(0)
Gunduz Songur A, Turktarhan G and Cobanoglu C (2022) Progress on green technology research in hotels: a
literature review. Journal of Hospitality and Tourism Insights. doi: 10.1108/JHTI-10-2021-0280
Svanberg J, Ardeshiri T, Samsten I, et al. (2022) Corporate governance performance ratings with machine
learning. Intelligent Systems in Accounting, Finance and Management 29(1): 50–68. doi: 10.1002/isaf.
1505
Tian F, Yang Y, Mao Z, et al. (2021) Forecasting daily attraction demand using big data from search engines and
social media. International Journal of Contemporary Hospitality Management 33(6): 1950–1976. doi: 10.
1108/IJCHM-06-2020-0631
Torres P and Augusto M (2021) Attention to social issues and CEO duality as enablers of resilience to
exogenous shocks in the tourism industry. Tourism Management 87: 104400. doi: 10.1016/j.tourman.
2021.104400
Uyar A, Kuzey C, Kilic M, et al. (2022) Does firms’CSR engagement support tourism sector development?
Moderation effect of CSR committee and CEO duality. Tourism Economics 28(8): 1973–1997.
Vaughan TS and Berry KE (2005) Using Monte Carlo techniques to demonstrate the meaning and implications
of multicollinearity. Journal of Statistics Education 13(1): 1–9. doi: 10.1080/10691898.2005.11910640
Vinayagamoorthi V, Murugesan S and Kasilingam L (2015) Impact of firms’profitability on environmental
performance: evidence from companies in India. Mediterranean Journal of Social Sciences. doi: 10.5901/
mjss.2015.v6n1p109
Wang L (2023) Low carbon management of China’s hotel tourism through carbon emission trading. Sus-
tainability 15(5): 4622. doi: 10.3390/su15054622
Wang H, Li G, Jiang G, et al. (2007) Robust regression shrinkage and consistent variable selection through the
LAD-Lasso. Journal of Business & Economic Statistics 25(3): 347–355.
Wang J, Wang S, Ru X, et al. (2023) Understanding the acceptance of carbon offset programs among hospitality
consumers: an application of the extended parallel process model. Journal of Sustainable Tourism 1–18.
doi: 10.1080/09669582.2023.2188149
Yang J, Sun Y, Sun H, a, et al. (2022) Role of financial development, green technology innovation, and
macroeconomic dynamics toward carbon emissions in china: Analysis based on bootstrap ARDL ap-
proach. Frontiers in Environmental Science 10. doi: 10.3389/fenvs.2022.886851
Yu P, Hao R, Cai Z, et al. (2022) Does the emission trading system achieve the win-win of carbon emission
reduction and financial performance improvement? —evidence from Chinese A-share listed firms in the
industrial sector. Journal of Cleaner Production 333: 130121. doi: 10.1016/j.jclepro.2021.130121
Mahdi Ziaei S (2015) Effects of financial development indicators on energy consumption and CO2 emission of
European, east Asian and Oceania countries. Renewable and Sustainable Energy Reviews 42: 752–759.
doi: 10.1016/j.rser.2014.10.085
Author biographies
Dr. Victor Oluwafemi Olorunsola, Ph.D., is a Senior Lecturer in the Faculty of Tourism at Eastern
Mediterranean University in Mersin, Turkey. (E-mail: victor.olorunsola@emu.edu.tr).
Dr. Mehmet Bahri Saydam, Ph.D., is a Senior Lecturer in the Faculty of Tourism at Eastern
Mediterranean University in Mersin, Turkey. His research interests include user-generated content,
organizational psychology, consumer behavior, human resources, and service quality (E-mail:
mehmet.saydam@emu.edu.tr).
Dr. Hasan Evrim Arici is an Assistant Professor in the Faculty of Tourism, at Kastamonu University,
Kastamonu, Turkey. The author’s research foci have been tourist behaviors, Sustainability, Green
Olorunsola et al. 21
marketing, Hotel employees psychology and emotions, hospitality marketing and branding (E-mail:
hasanevrimarici@yahoo.com).
Dr. Mehmet ali Koseoglu is an Assistant Professor at Metropolitan State University. He is the Editor
of the International Journal of Bibliometrics in Business and Management (IJBBM). His research
interests include strategic management and bibliometric analysis in business and management. He
has published in leading journals including Annals of Tourism Research, Tourism Management,
International Journal of Hospitality Management, and International Journal of Contemporary
Hospitality Management.
22 Tourism Economics 0(0)
