DETERMINANTS OF HOTEL EMPLOYEES’ ELECTRICITY SAVING INTENTION: EXTENDING THE THEORY OF PLANNED BEHAVIOUR

The hospitality industry including hotels makes a significant contribution to job creation and the gross domestic product of many countries. However, the negative environmental effects of hotels include high electricity consumption and emission of green house gases. Employees can help to improve workplace pro-environmental behaviour but the determinants of their electricity saving behaviour has not received thorough empirical investigation. Grounded on on the theory of planned behaviour (TPB), the study examined the determinants of hotel employees’ electricity saving intention (ESI). The study extended the TPB by adding two constructs (environmental concern and organisational climate) to the three TPB constructs. Data was collected through the cross-sectional survey method and the Partial Least Square Structural Equation Modelling was used for data analysis. The results showed significant positive relationships between two TPB constructs (attitude and perceived behavioral control) and hotel employees’ ESI. In addition, the effects of the two added constructs are significant. Theoretically, the study extended the TPB by adding two constructs and linking them to employees’ ESI in the hospitality industry. Recommendations include workplace training on electricity saving.


Introduction
Environmental sustainability has emerged as a vital component of human and business survival. Human needs for natural resources have doubled in the past fifty years with negative environmental impact. Environmental risks continue to be in the forefront of the results of the annual Global Risks Perception Survey (GRPS) and in 2018 accounted for three of the five major risks by likelihood and four by impact (Bahadure, 2017;World Economic Forum, 2018). In South Arica, electricity generation increased by 2.2% between July 2017 and July 2018 and electricity consumption increased by 1.2% within the same period (Statistics South Africa, 2018;International Energy Agency, 2019). Carbon emissions associated with energy production especially from fossil fuels and consumption are a major part of total emissions and an important driver of global climate change (Zierler, 2017;Tvaronavičienė and Ślusarczyk, 2019). The speed of diversity loss has accelerated and the abundance of species has declined by about 60% since 1970 with negative impact on health and social-economic development (Norton et al. 2015;Khan and Chang, 2018). In addition, energy prices have increased in South Africa and around the world and global energy supplies have become less stable. In South Africa, it is anticipated that the price of electricity will rise by 9.41% in 2019/20 and this is expected to double the expected inflation rate of 4.5% (Akpan and Akpan, 2012;Tang et al. 2019;National Energy Regulator of South Africa, 2019). Also, public concern about environmental issues has increased and this has stimulated eco-friendly products and services. Green consumption is on the increase as many customers are aware of the effect of their buying decisions on the environment (Chen and Tung, 2014;Verma and Chandra, 2017). Due to external pressure from customers and government and the increase in the cost of electricity, many firms have started to introduce energy saving measures. The participation of employees is crucial for the success of a company's energy saving activities. Energy saving behaviour can be described as the reduction of energy use by individuals. Electricity saving intention can be described as the self-commitment of an individual to participate in electricity saving behaviours. Reducing energy use through conservation is one of the more cost effective ways to significantly decrease greenhouse gas emissions. Electricity saving is a pro-environmental behaviour (Pollard, 2015;Sony and Mekoth, 2018;Ru et al. 2018).
Hotels are energy intensive facilities, with high energy costs. The consumption of energy by hotels is higher than other commercial buildings. Energy costs account for 3-6% of overall operating costs of hotels. Hotels are associated with high energy wastage and low energy efficiency with 42% of the energy used to heat and cool spaces in hotels wasted. However, rising energy costs, attention of guests to sustainability and the rise of green movement in the hotel industry have increased the focus of hotels on energy management initiatives (Mensah and Blankson, 2013;Prud'homme, and Raymond, 2016;Cingoski and Petrevska, 2018).
The theory of planned behaviour (TPB) is the most frequently used theory to predict pro-environmental intentions and behaviours (Pollard, 2015;Ru et al. 2018). The TPB by Ajzen (1991) posits that intention positively affects actual behaviour and behavioral intention is influenced by three factors namely attitude, subjective norms and perceived behavioural control. Chen and Tung (2014) argue that despite the broad effectiveness of the TPB in explaining pro-environmental behaviour, many studies have enriched the explanatory effect of the theory by adding other constructs. This study extends the TPB by adding two new constructs (environmental concern and organisational climate) to develop a model of electricity saving purchase intention for hotel employees. Researchers on sustainability issues and pro-environmental behaviour should develop a multidisciplinary viewpoint that includes both micro and macro factors (Joshi and Rahman 2015;Sobiegalla et al. 2018;Sarma et al. 2019).
The study aims to examine the effect of these five variables on hotel employees' electricity saving intention (ESI). The study will contribute to knowledge in the following ways. First, the electricity saving behaviour of individuals has become an emerging area of research. However, extant research at the individual level has focused 88 on households rather than the workplace (Gao et al. 2017;Chang et al. 2018). Furthermore, while individuals need to pay for electricity consumption in households, electricity consumption in the workplace is almost free of charge to the employee. This makes electricity more easily wasted in workplaces compared to households (Gao et al.2017;Wesselink et al. 2017). Second, there is a research gap with respect to employees' electricity saving intention and behaviour in organisations and studies that have used the extended TPB to examine hotel employees' electricity saving intention in South Africa are scarce.

Electricity saving behaviours
Electricity saving behaviour can be described as the behaviours performed by individuals to reduce overall electricity use and can be broadly divided into two categories.
(1) Habitual energy-saving behaviours: These focus on continuous efforts to reduce electricity use by curtailment measures. Examples of habitual electricity saving behaviours include reducing or avoiding the usage of air-conditioners and turning the power off when appliances are not used (2) one-shot purchasing behaviours: This involves the replacement of old technology with high electricity use with new technology with low energy use and the purchase of more-efficient technology.

Theory of planned behaviour and electricity saving intention (ESI)
ESI can be described as the self-commitment of an individual to participate in electricity saving behaviours. The TPB is the most commonly used theory to predict pro-environmental behaviour (Pollard, 2015;Ru et al., 2018). The TPB by Ajzen (1991) contends that intention predicts actual behaviour and intention is influenced by attitude, subjective norms and perceived behavioural control. In addition, other variables can be included in the TPB as long that they can be shown to improve the explanatory power of the model and are reasonable to explain a range of behaviour (Ajzen, 1991;Tommasetti et al.2018). Two additional constructs (environmental concern and organisational climate) are added as predictors of employees' ESI.

Attitude and electricity saving intention
Attitude towards the behaviour determines the extent to which a person has a favourable or an unfavourable assessment of a certain behaviour. A more favorable attitude towards a certain behavior by an individual should lead to a stronger intention to perform the behavior (Lin et al. 2015). Empirical literature is not conclusive about the effect of attitude on pro-environmental behaviour of individuals. Ha and Janda (2014) show that the attitude towards a green product has a strong effect on intention. Greaves et al. (2013) find that attitude has a significant positive relationship with the intention to engage in pro-environmental behaviour. The findings of the study by Wells et al. (2016) reveal that attitude has a positive effect on environmental behaviour both at home and in the workplace. Wang et al. (2014) show that environmental attitudes significantly influence energy-saving behavior. However, studies such as Kaiser et al. (1999), Park and Yang (2012) report a weak relationship between environmental attitude and environmental behaviour. If an individual considers electricity saving behaviour in the workplace as beneficial, he/she will hold a positive attitude and this can influence the intention to save electricity (Gao et al. 2017;Ru et al. 2018). Therefore, a more favorable environmental attitude by an individual should lead to a stronger intention to engage in electricity saving behaviour. Consequently, it is hypothesised that (H1): attitude towards energy saving positively affects employees' ESI.

Subjective norms (SNs) and energy saving intention
SNs indicate the possibility that individuals or groups that are important to an individual will like or dislike the performance of a particular behaviour (Ajzen, 1991). Greaves et al. (2013) find that SNs positively affect employees' intention to switch off computers when leaving their desk for more than one hour and recycle waste at work. The results of the study by Wang et al. (2014) show that SNs significantly affect the energy saving intention of city residents. Zierler (2017) argues that the relationship between SNs and intentions and behaviours ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 8 Number 2 (December) http://doi.org/10.9770/jesi.2020.8.2 (5) is a subject of much debate with varying empirical findings and Armitage and Conner (2001) remark that SNs tend to vary considerably across behaviours. Abrahamse and Steg (2011) find that SNs do not contribute to the explanation of intentions when attitudes and perceived behavioural control are controlled for. However, the opinions of a person or group of importance to an individual may influence the intention of that individual to engage in electricity saving as an employee in the workplace. It is hypothesised that (H2): SNs positively affect employees' ESI.

Perceived behavioural control (PBC) and electricity saving intention
PBC can be described as the perceived difficulty or ease of conducting a behaviour (Ajzen, 1991). There is a significant positive relationship between PBC and intention to use energy savings devices (Pollard (2015;Lin et al. 2015). However, Park and Yang (2012) and Kranz and Picot (2012) did not find a significant association between PBC and consumers' intention to adopt smart metering technology. The availability of resources and skill about electricity saving should positively influence behavioural intention. It is hypothesised that (H3): PBC positively affects employees' ESI.

Organizational electricity saving climate (OESC) and electricity saving intention
Organisational climate is a construct that has major implications for understanding human behaviour in organisations. Organisational climate can be described as the shared perception of employees about their work environment, particularly how policies and procedures are translated into tacit guidelines and practices. Proenvironmental organisational climate depicts the perception of employees about their organisations' proenvironmental policies, procedures, and practices (Castro and Martin, 2010;Norton et al. 2012). Zientara and Zamojska (2015) find that green organisational climate has a direct impact on organisational citizenship behaviour for the environment in hotels. However, Zhang, Wang and Zhou, (2014) find that OESC does not significantly affect electricity saving intention. This study argues that the positive perception of employees about the pro-environmental organisational climate of their organisation can positively impact on electricity saving intention. It is hypothesised that (H4): OESC positively affects employees' ESI. Dunlap and Jones (2002, p 484) define EC as "the degree to which people are aware of environmental problems and support efforts to solve them and/or indicate a willingness to contribute personally to their solution". Li et al.

Environmental concern (EC) and electricity saving intention
(2019) find that EC is significantly positively correlated with the willingness to purchase energy-efficient appliances. EC positively affects people's intention to use a park-and-ride facility (De Groot and Steg, 2007) and consumers' intention to adopt hybrid electric vehicles (Wang et al. 2016). EC has a significant positive effect on environmental knowledge, behavioral intention and environmentally friendly behaviour ((Pagiaslis and Krontalis, 2014;Newton et al. 2015). It is hypothesised that (H5): EC positively affects employees' ESI.

Research methodology
The study followed the quantitative research method and the data was collected through the cross-sectional survey method from participants in the Gauteng Province of South Africa. Participating hotels were conveniently selected from the website of the Tourism Grading Council of South Africa and were contacted by the researcher through a formal letter that explained the purpose of the study. Self-administered questionnaires were completed by employees of the participating hotels (see Appendix). Each participant was given a month to complete the questionnaire and was reminded weekly through emails and phone calls obtained during questionnaire distribution. The researcher pre-tested the questionnaire and the results led to the removal of sensitive information such as the name of the participant or hotel to ensure anonymity. Descriptive analysis and the Partial Least Square Structural Equation Modelling were used to analyse data.

biographical characteristics
Thirty-three out of the forty-two hotels contacted particiopated in the survey. 660 (20 per hotel) questionnaires were distributed to the participants and 342 questionnaires were returned.  Table 1 depicts the biographical characteristics of the survey participants. The results indicated that the majority of the respondents are males, in the 31-40 age bracket and with more than five years work experience.   Table 3 and table 4 depict the discriminant and the convergent validity. Four constructs (attitude, perceived behavioural control, organisational climate and energy saving intention) have all items greater than 0.708. However subjective norm and environmental concern have one item each with loading below 0.708. The two items were deleted. The results also indicate that the composite reliability is between 0.70 and 0.95, Cronbach's ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 8 Number 2 (December) http://doi.org/10. 9770/jesi.2020.8.2(5) 92 alpha values for the constructs are above 0.70 and the the square roots of AVEs are higher than the correlations among the latent variables ( (Hair et al. 2019).

Structural model assessment
Following the suggestions by Hair et al. (2019), the likelihood of common method bias (CMB) was examined. The VIFs for the constructs of the study were lower than 3.3 suggesting that the model is not constrained by CMB . The R 2 obtained by the study is 0.521 indicates a high level of predictive accuracy of the model. The original TPB model accounted for 44.9% of the variance. This indicates that inclusion of environmental concern and organisational energy saving climate increased the explained variance by 7.2%. The value of the goodness of fit (GIF) is 0.59 suggesting a good model fit and the (Q 2 ) using the cross validated communality is 0.56 which supports a predictive model. The effect size, f 2 , ranged from 0.01 to 0.12 indicating small to medium effect sizes and the standardised root mean square residual of 0.03 is indicative of a good model fit. The results of the path coefficients and T-statistics using the bootstrapping technique are depicted in table 5. The results H1 (β =0.307, T= 6.244, p<.001). H3 (β= 0.263, T= 7.216, p<.001) and H3 (β =0.263, T= 6.852, p<.001) support significant positive relationships between attitude and perceived behavioural control and electricity saving intention. Thus H1 and H3 are not rejected. The results for H2 (β =0.108, T= 0.559, p>.0.05) is not significant. Thus H2 is rejected. Environmental concern (β=0.225, T=3.775, p<0.01) and organisational energy saving climate climate (β =0.173, T =3.284, p<0.05) have significant positive relationships with electricity saving intention. Thus H4 and H5 are not rejected.

Discussion
Electricity saving behaviour can be described as the reduction of electricity use by individuals and electricity saving intention is the plan of an individual to participate in electricity saving behaviours. The TPB is the most commonly used theory to predict pro-environmental behavioural intentions including electricity saving intention. This study extends the TPB by adding two new constructs (environmental concern and organisational climate) to the three constructs of the TPB. The findings indicated that there is a significant positive relationship between attitude and PBC and ESI. H1 and H3 of the study are supported. The study did not find a significant relationship between SNs and ESI and H2 is not supported. The results also indicated significant positive relationships between two additional constructs (EC and OESC) and ESI. Therefore, H4 and H5 of the study are supported. The findings of this is supported by both theoretical and empirical literature. The TPB shows that attitude and perceived behavioral control can predict behavioral intention. This is supported by the findings of this study. However, the effect of SNs is not significant. Pollard (2015) finds that attitude toward sustainability at work has a strong association with the use of energy savings devices. The study by Wells et al. (2016) revealed that attitude has a significant positive positive effect on environmental behaviour both at home and in the workplace. There is a significant positive relationship between PBC and to intention to use energy savings devices (Pollard (2015; Lin  Gao et al. (2017) find that SNs have an insignificant effect on intention to save energy in workplaces. Organisational climate has a positive effect on hotel employees green behaviour in the workplace (Chou, 2014). Li et al. (2019) find that environmental concern is positively correlated with the willingness to purchase energyefficient appliances.

Conclusion
This study extended the TPB by adding two new constructs (environmental concern and organisational climate) to the three constructs of the TPB. The study investigated the effect of three TPB constructs and two additional constructs on hotel employees' ESI. The findings indicate that there is a significant positive relationship between attitude and PBC and employees' ESI. The study did not find a significant relationship between SNs and ESI. The findings also indicate significant positive relationships between two additional constructs (environmental concern and organisational electricity saving climate) and ESI.
The findings of the study have some policy implications. Employees must develop a more favorable environmental attitude. Therefore, factors such as turning off lights and air conditioners when leaving the office can help to save electricity. In addition, attending workplace training on electricity saving mechanisms will be a proactive way to reduce electricity consumption. Hotels must also make available resources, knowledge and skills about electricity saving. This can be achieved by the replacement of old technology with high electricity use with new technology with low energy use. The reward of employees must take pro-environmental behaviour into consideration. To improve, organisational electricity saving climate, hotels must develop and communicate their sustainability policy to employees. The limitations of the study include the use of convinience sampling which may lead to sampling bias and the focus on 342 hotel employees in one province limits the generalisability of the findings. Other studies can examine the effect of environmental passion and workplace spirituality on employees' ESI.