Insurance uptake among small and medium-sized tourism and hospitality enterprises in a resource-scarce environment.

Small and medium-sized tourism and hospitality enterprises (SMTHEs) are often susceptible to various hazards, which result in risk concerns. Insurance is recognised as one of the risk management strategies, but evidence indicates that insurance uptake among SMTHEs has been low. Yet, researchers have hardly researched into the factors that influence insurance uptake among SMTHEs. Two-hundred and fifty (250) respondents were selected using a multi-stage sampling technique. Confirmatory factor analysis, multivariate logit and probit regression techniques were used to determine factors underlying SMTHEs' insurance uptake. Risk concerns, the firm's characteristics, the perceived benefits of insurance and other informal risk coping mechanisms, as well as insurance service provision concerns were identified as determinants of insurance uptake. This is one of the first papers to offer a holistic understanding of the factors influencing SMTHEs' insurance subscription in a resource-scarce destination of Sub-Saharan Africa. The practical and theoretical implications of the paper are discussed.

Background

Small and medium-sized enterprises (SMEs) have been recognised and often referred to as the ‘lifeblood’ and ‘economic muscle’ of most countries because of their significance in supporting economic growth and livelihood through job creation (OECD, 2017). SMEs are not only predominant in the hospitality and tourism (H&T) sector, but also play a critical role as direct providers, brokers and distributors of products and services. Retention of tourism revenue within local economies and associated multiplier effects are high among SMTHEs. Overall, the promotion of SMTHEs contributes to a wider community involvement and development, and to the redistribution of wealth (Garay & Font, 2012). However, the majority of such SMTHEs do not survive beyond a decade after being established while for others, surviving is a constant struggle (Gbireh, 2013; Ngoasong & Kimbu, 2016; Ngoasong & Kimbu, 2019). While several factors are associated with the collapse of SMEs, the inability to properly manage risks is the most critical (Gbireh, 2013). For instance, to Gbireh (ibid), the general distrust about the timely payment of claims by insurers, low knowledge about insurance as well as the dependence on friends and family and personal saving against uncertainties (in lieu of insurance), negatively affect risk management by SMEs.

The tourism and hospitality sector is susceptible to all forms of threats, such as political instability, natural disasters, health pandemics and economic crises. These unexpected events interrupt business activities resulting in significant financial burdens and psychological stress among management and employees. This makes it vital for businesses to develop ways of cushioning themselves against such adversities – through various strategies, including but not limited to insurance uptake (Rejda, 2012; Williams & Baláž, 2015).

Furthermore, due to the low financial capabilities of SMEs in resource-scarce emerging destinations of Sub-Saharan Africa, micro rather than standard insurance is deemed suitable for them in dealing with risk within the informal sector. Resource-scarce is used here to emphasise destinations that are characterised by a high proportion of people with low purchasing power, growing but untapped market opportunities and a high number of active entrepreneurs (Boojihawon & Ngoasong, 2018), a large proportion of whom are operating in the informal sector. The informal business sector is characterised by ease of entry, family owned, small scale operations, unregulated and competitive markets and is also labour intensive especially in developing countries (Kofi & Gemegah, 2011). One essential factor that encourages informal sector workers and businesses to demand micro-insurance is the flexibility in payments and collection of affordable premiums. Accordingly, having an insurance cover is one strategy for protecting SMTHEs against possible losses associated with hazards due to its risk transfer and indemnification role (Williams & Baláž, 2015). The characteristics of the H&T product including intangibility and perishability as well as the high susceptibility of the industry to both economic and environmental volatilities, make the uptake of insurance a useful strategy. Insurance offers owners of businesses the peace of mind to help them achieve higher productivity goals (Rejda, 2012). It provides security – helping to safeguard businesses against financial hardship and make them recover from losses much faster (Atkins & Bates, 2008). Yet, despite the usefulness of insurance as a risk management tool and the concomitant high demand for it in developed economies, its subscription has been generally low in developing countries (Gbireh, 2013). In the tourism literature, a few studies (see Caponecchia & Tan, 2019; Law, 2006; Leggat & Leggat, 2002; Olya, Alipour, Peyravi, & Dalir, 2019; Olya & Altinay, 2016; Williams & Baláž, 2015) have provided insights on tourists' perspectives on the relevance of various insurance covers, particularly travel, weather, climate and health insurance. However, it is as yet undetermined why firms especially SMTHEs subscribe or not to insurance especially from an emerging economy perspective. Even though theories, such as the Cumulative Prospect theory and Expected Utility theory broadly offer some insights on why and how consumers guard against risk in a given situation, the H&T literature is currently bereft vis-à-vis the actual factors influencing the uptake of insurance (or otherwise) among SMTHEs. Besides, this study is a timely response to a call by Olya and Altinay (2016) that future studies should examine the potential constraints and triggers of insurance uptake among consumers especially SMTHEs. Besides, Olya et al. (2019) studied tourism climate insurance, concentrating on tourists' intention to purchase climate insurance and further recommend a similar study among other stakeholders from the supply-side such as SMTHEs, hence the current study. The study thus becomes pertinent because its findings for the first time, enable insurance companies targeting SMTHEs to clearly understand the wide-ranging factors that affect their insurance uptake in resource-scarce but emerging destinations of Africa and possibly design tailored insurance packages based on such factors.

This study, therefore, explores and critically analyses the factors driving the uptake (and non-uptake) of insurance among SMTHEs in Ghana, a Sub-Saharan African country that has been experiencing steady growth in its H&T sector within the last decade but is plagued by resource scarcity, mismanagement and high levels of debt, a situation that is observed across other countries in the region (Calderon, Kambou, Korman, Kubota, & Cantu, 2019; Deloitte, 2019). The study unpacks the risk related issues, firm characteristics, structural and contextual issues, as well as other coping mechanisms that play a role in SMTHEs deciding to take up insurance or otherwise. Consequently, a study of this nature will add a fresh perspective on insurance uptake in the H&T literature in terms of what makes firms subscribe to insurance or otherwise especially in resource scare developing country destinations.

Literature review

Conceptualising SMEs

The conceptualisation of SMEs is unquestionably unsettled in the extant literature. The lack of universality in the characterisation of SMEs is ascribable to the subjective and qualitative understanding of what is meant by small, medium and large sized businesses (see Platteau, De Bock, & Gelade, 2017). Nonetheless, most definitions take into consideration indicators, such as the number of employees in the business, location, assets, turnover, management style or market share. But these pointers also vary by countries and institutions. For instance, the United Nations Development Programme [UNDP] (2015) proposed a working definition for SMEs as firms that employ between 5 and 20 employees while those employing between 21 and 99 employees qualify as medium sized enterprises. In Africa, the definitions are also based on the indicators presented earlier. In Uganda for instance, firms that are classed as SMEs employ between 5 and 50 employees and wield less than US$ 50 million worth of assets, as well as a working capital and turnover of between US$ 10 and 50 million. On the other hand, medium-sized businesses are considered as firms that employ between 50 and 100 workers (Malgwi, 2005).

In Ghana, the definition of SMEs by various government and quasi-governmental institutions point to a certain degree of accord in the structure and nature of SMEs in the country. Suffice it to emphasise, however, that these definitions focus on the number of employees as a major criterion for distinguishing SMEs from other enterprises. Also, the definitions seem to be divided in terms of the cut-off point for the number of employees. For instance,Osei-Assibey and Baah-Boateng (2012) note that the employee cut-off point for SMEs in Ghana is 30. Contrarily, the Ghana Statistical Service [GSS] (2015) suggests that firms with fewer than 10 workers are characterised as small-scale enterprises while those with more than 10 employees typify medium and large-sized enterprises. Moreover, considering assets as part of its criteria, the National Board for Small Scale Industries [NBSSI] (see Oppong, Owiredu, & Churchill, 2014) recognises small-scale firms as those employing up to 9 workers with plant(s) and machinery value not exceeding 10 million Ghana Cedis.1 The challenges with using assets as a criterion for determining SMEs is the quality of valuing assets and the fluctuation of currencies (in the present case the cedi before other major trading currencies like the US dollar and British pound). In addition, Gbireh (2013) observed that the diversity of the SMEs sector in Ghana calls for a rethink of past definitions along geographic, sectoral or other context in which SMEs are becoming relevant. Notwithstanding the variations in parameters characterising SMEs and for that matter SMTHEs, this paper adopts the definition by Osei-Assibey and Baah-Boateng (2012).

Insurance as a risk management tool

According to Brown and Churchill (1999), insurance decreases losses as households substitute the uncertain prospect of large losses with the certainty of making small, regular premium payments. Insurance is a contract involving one party's offer of financial indemnity to another party (i.e. insuree) against possible damages or losses in the future in return for a premium (Rejda, 2012). By paying a premium, possible losses and the related financial encumbrances of businesses are transferred to an insurance company based on an insurance policy agreed upon by the parties. The insurer reallocates the cost of losses among other insured individuals and/or firms in the pool (Williams & Baláž, 2015). Principally, insurance companies use the pooling concept, where light losses are borne by many so that heavy losses are not borne by a few. Put differently, the insurer spreads the risk of economic loss among as many subscribers as possible who are similarly imperilled by the same kind of risk situation. The insurer does so through a variety of insurance initiatives broadly as either general or life insurance (Baranoff, Brockett, & Kahane, 2009; Outreville, 1996). Thus, the management of risk by transferring financial responsibility to another party is, so far, the most common method businesses employ (Platteau et al., 2017).

Guha-Khasnobis and Ahuja (2004) note that even though informal and formal sector businesses face analogous risks including illness, death, accidents, loss of property, fires and theft among others, informal sector businesses are more susceptible to risky events than formal sector ones. Therefore, micro-insurance plays an important role in dealing with such stressors because of the peculiarities of SMEs. Micro-insurance is unique due to the size of its policyholders, who tend to be SMEs with low but also seasonal incomes. As a correlate, premiums of micro-insurance are low for affordance and delivery channels are widespread for ease of access due to the widespread nature of the informal economy, and persons who are the target. These channels include credit unions, rural banks, community-based scheme and non-governmental organizations (Churchill, 2006). Thus, micro-insurance is a contractual agreement where insurers offer protection to SMEs against specific risks in return for regular premium payments proportionate to the probability and severity of the risk involved.

Theoretical framework and development of hypotheses

This section offers an empirical insight on factors that influence insurance demand among businesses, especially SMEs. It explicates the main factors reported in the literature as drivers of insurance uptake. Specifically, the review covers the theoretical and empirical basis of perceived risk, perceived usefulness and cost of insurance, constraints (including scheme related factors) and other coping methods as drivers of insurance uptake gleaned from the literature. The literature review was also informed by the self-reporting of issues that impacted SMEs demand for insurance. This means that we retrospectively fed back the results into the review section after analysing the data. We then develop some expectations by way of hypotheses.

Perceived risk and insurance uptake

The cumulative prospect theory (CPT) and ambiguity aversion theory (AAT) are two major micro-economic theories that explain why insurance uptake is a function of perceived risk. The CPT assumes that people are risk averse in some cases and risk receptive in other cases. It also seeks to explain how people make rational decisions regarding uncertainties (Tversky & Kahneman, 1992). The theory also assumes that people are generally risk averse and will initiate strategies to deal with such aversion. The theory is characterised by two main qualities including ‘diminishing sensitivity’ and ‘loss aversion’ – used to describe decisions about risks and uncertainty. Diminishing sensitivity implies that subjects have trepidations relative to gains and risk seeking regarding losses, while loss aversion suggest that when it comes to taking decisions under risky and uncertain circumstances, losses appear larger than gains. Tversky and Kahneman (1992) contend that people exhibit loss-aversion, in other words, they experience more disutility for a loss, than they experience utility for a gain of the same amount (Platteau et al., 2017). The applicability of the CPT has been examined within travel and tourism. Antón, Camarero, and Laguna-García (2017) adopted this theory in explaining the impact of heritage tourist motivations on satisfaction and loyalty. The study (ibid) uncovered internal factors including seeking of experiences, travelling with family and friends to boost relationships, and getting away from stress and external factors such as the enjoyment of cuisine and sightseeing. Similarly, in studying the antecedents of space travellers' behavioural intention, Olya and Han (2019) assumed nonlinear interactions between risk and motivations in the prediction of behavioural intention of space travellers. The study reported that perceived risk of the respondents can consistently and amply under weigh their motivations, in formulating behavioural intention.

Loss-aversion can have an impact on insurance behaviours as businesses would find insurance as a buffer against losses associated with hazards. Thus, the extent to which insurance is perceived as effective in guarding against losses would influence uptake. This suggests that insurance becomes important in optimally safeguarding against risk, instead of simply reducing it. Though the CPT recognises that consumers' decision making is associated with risk and uncertainty, this idea is limited in a sense that it assumes perceived risk is linked with only negative outcomes (Olya & Al-ansi, 2018). A review of empirical literature points evidently to the fact that risks may result in positive outcomes (see Belanche, Casalo, & Guinalíu, 2012; Malazizi, Alipour, & Olya, 2018; Olya & Al-ansi, 2018; Olya & Altinay, 2016; Tangeland, Vennesland, & Nybakk, 2013). For instance, Belanche found that risk perception positively influenced consumer satisfaction. Similarly, Tangeland et al. (2013) also reported that social risk increased second-home owners' intention to buy nature-based tourism services. A more recent study by Malazizi et al. (2018) realised that psychological risk had a strong positive relationship with satisfaction, continuous intention and the intention to recommend Airbnb services to others. In a related study, Olya and Al-ansi (2018) also reported on the positive influence of health, psychological, environmental and quality risks on consumer satisfaction of halal products and services. The study also found that financial risk also had a positive influence on continuous intention and intention to recommend such products and services to others. Thus, while perceived risk may trigger negative outcomes and the resultant strategies to deal with such outcomes, it could also have a positive influence on consumers' attitudes and behaviours that need to be recognised.

An alternative explanation to insurance demand is found in the ambiguity aversion theory. The theory suggests that the majority of individuals tend to dislike uncertain outcomes of events (Elabed & Carter, 2015). That is, they dislike being uncertain about the probability with which events take place. Ambiguity aversion has been employed to investigate insurance subscription, and it is argued that ambiguity aversion is significantly related to less chances of insurance uptake (Elabed & Carter, 2015). In developing countries, insurance demand is characterised by a lot of uncertainties ranging from the uncertainty of loss occurrence, and trustworthiness of the insurer to pay out in case of losses. Each of these uncertainties limits the demand for insurance since people would want to avoid them (Platteau et al., 2017). The hesitations that surround insurance make insurance itself risky and thus can motivate its avoidance and consideration of alternatives. Nevertheless, one important limitation with these economic theories in explaining insurance demand is that the unit of analysis tends to focus more on the individual other than an entire organisation. For large H&T operations, insurance coverage may not be the same depending on the respondents' equity position. The potential loss of an event would appear to be lower for someone without equity relative to a non-equity respondent. Given that the sample in this current study involves SMTHEs, whose day-to-day operational decisions are determined by the owner managers, the study implicitly asserts that the risk-transfer behaviour of individuals (i.e. the decision to purchase insurance) manifests itself in business entities without distortion.

The category of manager in charge of the business has a significant influence on insurance uptake.

Several risk factors including operational risk, technological risk, transport risk, environmental risk, political unrest and economic risk, have been identified in the literature as potential drivers of insurance demand (Shaw, Saayman, & Saayam, 2012). First, operational risk relates to risk emanating from the day-to-day activities of SMTHEs which can result in adverse impacts or losses. These are often endogenous in nature and include but not limited to injuries, deaths of employees, thefts, burglaries, fires, and financial embezzlements (Lorde & Jackman, 2013). Second, the past decade has witnessed a remarkable interest in the application of information technology (e.g. social media and artificial intelligence) to the operations of businesses in the industry (Ayeh, Au, & Law, 2013). However, its increasing application is beset with threats of fraud, hackers, malwares, and infrastructural distortions that lead to huge losses (Malgwi, 2005). Third, the profitability and survival of SMTHEs are indissolubly linked with the quality of the environment (Mensah & Mensah, 2013). Yet, the probability of environmental disasters either of natural or anthropogenic sources, are issues H&T businesses must contend with on a daily basis. Events including floods, storms, tsunamis, earthquakes, bushfires, and landslides can be catastrophic (Manaktola & Jauhari, 2007) to business operations often resulting in huge losses. Fourth, the literature recognises H&T as a vulnerable sector in the face of unrest, such as terrorism, strikes or wars (Bowen, Fidgeon, & Page, 2012). These events have been found to have serious implications and consequential costs on businesses that are directly affected, not least, the collateral damage on other indirect businesses, and proximal destinations to the affected areas (Henthorne, George, & Smith, 2013). These events make businesses apprehensive and try to use various preventive and recovery measures against such adversities. Fifth, the industry is not immune to economic threats. Volatilities like currency depreciation, high interest rates, and inflation (Baranoff et al., 2009) prey on businesses and consumers. These, no doubt, become impediments to investments leading to retardation of employment opportunities and wealth creation. Finally, transportation is very central to H&T in not just moving tourists from one location to another but supporting the operations of businesses in the sector, particularly around haulage and supply but there exist risks (Nutsugbodo, 2013). Transport related accidents involving employees and clients as well as traffic jams within urban areas and central business districts are on the increase in Ghana (Awunyo-Victor, 2012). Therefore, it appears risk is inevitable, what matters then is the reaction to it. Consequently, the current study hypothesises that (see Fig. 1 ):

Fig. 1

Proposed hypothesised relationships.

Perceived operational risk has a positive influence on insurance uptake.

Perceived technological risk has a positive influence on insurance uptake.

Perceived health risk has a positive influence on insurance uptake.

Perceived environmental risk has a positive influence on insurance uptake.

Perceived unrest has a positive influence on insurance uptake.

Perceived economic risk has a positive influence on insurance uptake.

Perceived benefits and constraints as determinants of uptake of insurance

The expected utility theory (EUT) argues that people try to optimise their expected utility when deciding whether to take an insurance cover. It is obvious that business owners are uncertain about tomorrow's outcomes (Williams & Baláž, 2014), but are aware they will encounter some form of risk and perhaps will be able to evaluate what is risky based on the likelihood of the event and severity in terms of its impact on their current wealth should it occur (Baranoff et al., 2009). Given that the utility of SMTHEs depends on the wealth accumulated from the business and perhaps expected wealth, the topmost goal of every business owner would be to optimise utility from investments made. Rationally, one will seek protection against potential threats to the flow and sustenance of this utility for example through insurance which has been found to be a risk transfer mechanism.

The role and usefulness of insurance has long been established (Atkins & Bates, 2008; Rejda, 2012). The main reason to insure is compensation against future financial losses. It also helps to attain a sense of security that one will remain safe from financial hardship and makes recovery from an insured loss much faster (Atkins & Bates, 2008). Other perceived benefits cited about micro-insurance are its ability to facilitate easy access to loans, enhancing business image, and helping with a quick revamping of businesses after eventualities (Manning, 2004; Rejda, 2012; Banerjee, Duflo, & Hornbeck, 2014). Therefore, perceived usefulness of micro-insurance is a major motivation for SMEs to seek protection through insurance uptake. Accordingly, the following hypotheses are proposed.

The perception that insurance provides compensation for losses has a positive influence on insurance uptake.

The perceived restorative ability of an enterprise through insurance after eventualities has a positive influence on insurance uptake.

The perception that insurance boosts business security has a positive influence on insurance uptake.

In most developing economies, a critical factor affecting the demand for insurance is the generally low level of information and awareness about these schemes (Giesbert, Steiner, & Bendig, 2011; Giné, Townsend, & Vickery, 2008). In Uganda, the lack of awareness about the benefits of insurance has been found as a significant factor that constrains insurance adoption among SMEs in the country (United Nations Conference on Trade and Development [UNCTAD], 2007).

Osunsan (2015) also indicated that the level of public awareness about insurance services is still low. In Kampala, he found that the majority of SMEs in general could only name a few insurance products, but most had no insurance cover. In Nigeria, there have been incidents of the collapse of several promising SMEs; this has been attributed to the lack of awareness to obtain adequate and appropriate insurance coverage (Osunsan, 2015). Kesper (2000) further pointed out that SME operators are mostly not adequately equipped with the requisite knowledge about insurance policies and hence prefer to remain uninsured. Giesbert et al. (2011) in their study on the perception of a micro-insurance scheme also noted that respondents did not fully understand the insurance policy and its terms and conditions. Thus, this study conjectures that:

The perceived difficulty in understanding an insurance policy has a negative relationship with insurance uptake.

Apart from awareness, there are various other supply related factors that also affect the demand for insurance. For instance, insurance premium, has been found as a key driver of insurance uptake among businesses (De Bock & Gelade, 2012). Insurance premium is the amount of money paid to an insuree as claim for a given insurance policy (De Bock & Gelade, 2012). It is determined by the insurance company taking into consideration the likelihood of the insured event happening and the severity, when it does happen. These premiums are put into a pool from which various people are compensated in the event of losses. Other factors taken into consideration include expenses in relation to survey, underwriting, record keeping, and cost connected to accident prevention, claims handling among others. However, SMEs have often complained that premiums are expensive and the premium payment method is inflexible (Gbireh, 2013). De Bock and Gelade (2012) found that SMEs do not have the resources, time or manpower to cope with such high premiums.

The perceived costly nature of premiums has a negative relationship with insurance uptake.

The perceived inflexibility in the payment terms of premiums has a negative relationship with insurance uptake.

Besides, Osunsan (2015) argued that some of the negative attitudes towards insuring are based on experiences and beliefs, especially in developing countries. People generally think that insurance companies only collect premiums and never pay at a time of loss (Osunsan, 2015). This conviction is because of the bureaucratic processes applied when losses occur – posing challenges when making claims. Furthermore, while some think that insurance is not needed since God will protect them from calamities (Chodokufa & Chiliya, 2014), others also indicate that insurance companies do not offer the kind of policies they need (Banerjee et al., 2014).

The perceived time taken to process claims has a negative relationship with insurance uptake.

Alternative risk coping strategies as determinants of insurance uptake

Several other risk coping measures, self-insurance and informal group-based mechanisms, which could either serve as surrogates or complementarities of insurance have been identified in the literature as barriers to micro-insurance demand (Churchill, 2006; Platteau et al., 2017). These are remittances from friends and relatives, precautionary savings, credit and informal risk sharing agreements (Carter, Elabed, & Serfilippi, 2015). On the tenets of the crowding-out and crowding-in hypotheses, these strategies could crowd-out insurance. This occurs when the above alternative measures are substituted for insurance by helping businesses to cope with the consequences of hazards without the need for insurance. However, these alternative measures could be limiting in various ways as risk coping measures. Saving as a precautionary measure may fail to provide a business with the necessary buffer when several shocks occur within a short period of time with catastrophic consequences greater than the amount saved against such perils (Carter et al., 2015). These issues notwithstanding, there can also be complementarities between the alternative measures and insurance uptake (crowding-in effect). Insurance cover can enhance the chances of an SME acquiring loans from financial institutions (Banerjee et al., 2014). Likewise, savings and remittances could enhance future insurance premium renewals and help cope with shocks if the insurance fails to pay-out on time (Platteau et al., 2017). Therefore, we hypothesise that:

Saving against unfortunate events has a negative relationship with insurance uptake.

Support from friends and relatives has a negative relationship with insurance uptake.

Additionally, even though the literature remains silent on the extent to which various organisational characteristics can influence insurance uptake among SMTHEs, we have decided in this study to explore how much impact the sector type, ownership type and position held in the business influences the chances of insurance uptake.

Research methods

Study setting

Our research setting is Ghana, an emerging but under-researched destination with an institutional context that creates opportunities but poses many challenges and risks for SMTHEs. The National Insurance Commission [NIC], 2018 disclosed that apart from South Africa (16.9% in 2018), the average insurance penetration rate in Sub-Saharan Africa is circa 1% relative to a global penetration rate 5%. This paints a picture of an underdeveloped insurance market across SSA. The NIC (National Insurance Commission [NIC] (2018)) also reported that Ghana's insurance penetration rate as at 2017 was 1.12% and has further dropped by 1% in 2018. This low penetration rate in Ghana and across the continent is generally ascribed to low disposable income and literacy, lack of trust and poor public perception, low capacity of the insurance sector, and alternatives to insurance such as community-based schemes (Deloitte, 2019). These make Ghana suitable for understanding the drivers and inhibitors of insurance uptake among SMTHEs.

Sampling

Involved in the survey were accommodation facilities, restaurants, transport agencies and artisans in three cities (Accra, Cape Coast and Kumasi metropolis) of Ghana collectively referred to as ‘the tourism triangle’. The three cities host most of the attractions and tourist accommodation facilities in the country (Boakye, 2010). About 60% of all the businesses surveyed were from the hospitality industry, notably accommodation and restaurants. A multi-stage sampling procedure was used to reach out to either managers/supervisors or owners (where appropriate) of the various enterprises involved in the study. These categories of staff were targeted because they were largely responsible for the day-to-day running and charting of strategic directions for their firms (i.e. SMTHEs) in Ghana (Mensah & Blankson, 2016) and for that matter are more likely to be taking important decisions regarding insurance uptake in their organizations. First, 10% (i.e. 300) of the total registered SMTHEs and the corresponding addresses (i.e. 120 from Greater Accra, 100 from Cape Coast and 80 from Kumasi) were randomly drawn from the Ghana Tourism Authority list of registered SMEs. Second, a stratified sampling procedure was used to divide the sample into the four strata of sub-sectors as noted above, followed by a proportional allocation of the sample for each stratum. Hence, the sample size for each stratum was as follows: 102 accommodation facilities, 75 restaurants, 64 artisans, and 59 transport businesses. These facilities were then contacted for the survey. Of the 300 questionnaires distributed, 250 were found valid for analysis after discarding incomplete and spoiled ones giving a response rate of 83% for the survey.

Research instrument

The questionnaire (please see attached) was divided into three (3) parts comprising both closed and open-ended questions, as well as Likert scale type of measurements. The first part elicited information on the background characteristics of respondents and their firms such as the type of business sector, position of respondents in the business, number of employees in the business and ownership type. The second part examined the level of awareness about insurance. This section elicited perceptions about insurance in terms of its usefulness to SMTHEs and challenges to its uptake using open ended questions. The third part was made up of 20 statements related to the impact of perceived risk on a ranked scale (No threat = 0; extremely low threat = 1; low threat = 2; moderate threat = 3; high threat = 4 and extremely high threat = 5). The measurement items on the impact of perceived risk on insurance uptake were generated from the literature (see Table 1 ) and modified to fit the context of this study. Questions on ‘insurance uptake status’ were covered in the last part but also a dummy scale involving ‘yes’ or ‘no’ responses. Apart from insurance, respondents were also asked to indicate alternative measures used to handle uncertainties in their businesses through self-reporting.

Table 1

Items drawn from literature and modified for this study.

Items	Source
Operational risk	Shaw et al. (2012)
Theft of property
Employee turnover
Damage to business property by Fire, Flood, burglaries etc
Injury or death of an employee in course of work
Embezzlement by employees
Injury, death to third parties that come into your business
Illness
Technological risk	√
Cybercrime, hackers and fraud
Technological changes
Transport risk	√
Rood traffic accidents
Loss of goods or cash in transit
Environmental risk	√
Water pollution
Air pollution
Seasonality
Unrest	√
Political instability
Acts of terrorism
Strikes
Civil riots
Economic risk	√
Inflation
Depreciation of currency
High interest rates

The questionnaire was pretested on a sample of 30 SMTHEs in the Accra Metropolis in February 2017. This city was selected because it hosts most of the SMTHEs in the country, such as hotels, restaurants and tour operators among others. The pre-testing of the instrument enabled the revision of irrelevant and inappropriately phrased questions and statements. Following the pretesting of the data instrument, instructions and questions relative to the determinants of insurance uptake were revised for clarity.

Data analysis

Before the data analysis, missing data was treated by using the series mean option in SPSS to replace missing data. For that matter, a paired sample t-test was conducted to ensure there were no great differences between values of the original variables and estimated mean scores. Using STATA 14, first, basic descriptive statistics comprising frequency and mean scores were used to assess the distribution of the responses. Second, the CFA in AMOS was used to examine how well the data fitted the model as well as the validity and reliability of the measurement items generated from the literature.

Common Method Bias (CMB) assessment of perceived risk

CMB can result in measurement errors, impacting negatively on the validity of conclusions because of inflations in the true correlation estimates among constructs (Agag & El-Masry, 2016). To avoid such a situation in this study, the Harman's single-factor approach was used to check for CMB (Park & Tussyadiah, 2016; Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The nonexistence of CMB was obvious from this result as the first (and largest) factor accounted for 17% of the total variance which was lower than the cut-off point of 50%. This criterion confirmed the absence of CMB in the data collected.

Reliability, validity and model fitness

The internal consistency of the measurements used, their accuracy (i.e. validity) and overall model fitness were determined in this study. As can be seen in Table 2, the composite reliability (CR) and Cronbach's alpha scores for the six measurement constructs were above the lower limit of 0.7 (Bagozzi & Yi, 1988; Kline, 2005). All indicators for the respective constructs loaded above 0.5 (Hair, Hult, Ringle, & Sarstedt, 2017) which meant that the interrelationships between indicators and their associated dimensions were high thus unidimensionality was achieved among all the perceived risk constructs.

Table 2

Means, standard deviations, reliability and model fitness of perceived risk measures.

Constructs	Mean	SD	Factor Loadings	CR	Cronbach's Alpha	AVE
Operational risk	3.64	0.74		0.84	0.71	0.66
Theft of property	4.22	0.59	0.71
Employee turnover	4.27	0.54	0.82
Damage to business property by Fire, Flood, burglaries etc	4.26	0.50	0.76
Injury or death of an employee in course of work	2.29	1.16	0.67
Embezzlement by employees	2.60	1.06	0.71
Injury, death to third parties that come into your business	4.22	0.59	0.87
Technological risk	3.41	0.58		0.81	0.80	0.57
Cybercrime, hackers and fraud	4.56	0.61	0.76
Technological changes	2.25	0.54	0.83
Health risk	4.09	0.64		0.86	0.76	0.74
Illness	4.75	0.56	0.85
Road traffic accidents	3.42	0.71	0.76
Environmental risk	3.02	0.73		0.91	0.84	0.56
Water pollution	1.03	0.77	0.77
Air pollution	3.15	0.68	0.76
Seasonality	4.89	0.74	0.65
Unrest	0.28	0.65		0.82	0.73	0.75
Political instability	0.24	0.70	0.81
Acts of terrorism	0.43	0.63	0.83
Strikes	0.27	0.64	0.83
Civil riots	0.17	0.62	0.67
Economic risk	4.54	0.55		0.90	0.86	0.67
Inflation	4.67	0.55	0.77
Depreciation of currency	4.22	0.52	0.80
High interest rates	4.74	0.58	0.83

x2/df (2.583), SRMR (0.06), CFI (0.936), AGFI (0.900), TLI (0.940), GFI (0.901), RMSEA (0.058).

The following guidelines for assessing model fitness were also used in this study: Chi-Square (χ2/df) less than 3, Goodness-of-Fit Index (GFI) = ≥ 0.90, Comparative Fit Index (CFI) = ≥ 0.90, Adjusted Goodness-of-Fit Index (AGFI) ≥ 0.90, Turker-Lewis Index (TLI) ≥ 0.95, Root Mean Square Error of Approximation (RMSEA) ≤ 0.08 and Standardised Root Mean Square Error Residual (SRMR) ≤ 0.08 (Bagozzi, Yi, & Philipps, 1991; Fornell & Larcker, 1981). The results underneath Table 2 show that the model on perceived risk had a tolerable global fitness.

Furthermore, in examining the convergent validity of measures, the Average Variance Extracted (AVE) scores were found to be greater than 0.5 (Fornell & Larcker, 1981). This means that all latent factors explained more than half of the variances in their relevant items indicating negligible errors in the scales used. In determining whether the constructs uniquely measured what they intended to measure, in other words, whether they were indeed different from each other, the Fornell and Larcker criterion was used (Fornell & Larcker, 1981). This criterion suggests that for there to be discriminant validity, the square root of the AVE of a construct should be greater than the intercorrelation between that construct and another as depicted in Table 3 .

Table 3

Correlation matrix of risk dimensions.

	Constructs	AVE	1	2	3	4	5	6
1	Operational risk	0.66	0.81
2	Technological risk	0.57	0.38	0.75
3	Health risk	0.74	0.54	−0.10	0.86
4	Environmental risk	0.56	0.21	0.24	−0.21	0.75
5	Unrest	0.75	0.30	−0.13	0.45	0.21	0.87
6	Economic risk	0.67	0.61	0.36	−0.07	0.32	0.42	0.81

Note: Square roots of AVEs are shown diagonally in bold.

Third, binary multi-variate logistic and probit regression models were used to probe for the explanatory variables for insurance demand by SMTHEs in Ghana. Insurance uptake is considered a binary utility function, which is modelled as a Bernoulli outcome equal to one (1) for currently insured and zero (0) for otherwise. The equation is as follows:

Model for insurance adoption decision

The probability of an H&T business insurance is expressed as:

By substituting and making e the subject, the following equation is obtained:

Applying natural logarithm to both sides of Eq. (2) leads to the following:

Following from Eq. (3), the empirical estimable econometric model is therefore specified as:where IU is the dependent variable; X is a vector of explanatory variables considered to affect insurance adoption; α is a vector of unknown parameters to be estimated and ε is the stochastic error term.where IU is insurance uptake. X1i-X3i represents sector of operation, X4i-X5i ownership type; X6i is ownership type, X7-X12 represent risk concerns (operational, technological, transport, environmental, unrest and economic risk);…X26i is support from friends and relatives. ε is the error term assumed to be normally distributed with zero mean and constant variance.

We used the two-alternative multivariate binary models to ascertain the consistency of the signs and magnitudes of the coefficients of effects across the two analytical models (Boakye, Annim, & Dasmani, 2013). This not only guaranteed the validity of the results given that there is yet no consensus in the literature as to whether probit models are superior to logit models, but also the robustness of the results since different link functions were tested.

Results

Descriptive statistics

A range of SMTHEs participated in the study. In a descending order, accommodation facilities (34.3%), restaurants (25.0%), artisans (21.2%) and transport (19.5%) were included in the survey. It was found that most businesses (64.0%) were locally owned, 9.3% foreign owned while more than a quarter (26.7%) had mixed ownership i.e. local and foreign ownership. About two-thirds (64.0%) of the respondents were managers or supervisors while 36.0% were owner-managers. Slightly, more than half (53.3%) of the SMTHEs employed between 5 and 10 employees and another 46.7% of them employed between 10 and 20 employees. In terms of their location in the country, 43.3% were in Accra, 30.7% in Kumasi and more than a quarter (26.0%) in Cape Coast. Accra is home to many SMTHEs, not least because it is the capital city of Ghana, and like many capital cities and towns in the world, inhabits a disproportionately higher number of businesses in the country.

The study found that about half (51.0%) of the number of SMTHEs investigated had adopted one form of insurance to manage their risk while the remaining had no insurance cover. Fig. 2 shows that 27.5% motor insurance 23.3% had health insurance and 17.1% of them had property insurance. The least were those who had adopted professional indemnity covers (2.1%).

Fig. 2

Insurance packages used by SMTHEs in Ghana.

Table 2 shows the statistics on the average (mean) ratings and standard deviations (SD) of the items used in the measurement of perceived risk. The most rated risk dimension was economic risk (mean = 4.54; SD =0.74). Health risk emerged the second most rated risk dimension (mean = 4.09; SD = 0.64), and the least rated was unrest (mean = 0.28; SD = 0.65). Table 4 shows the benefits of insurance, constraints and alternative risk coping strategies among SMTHEs in managing risk. The majority (93.30%) of respondents did indicate that insurance provides compensation for losses and business security (94.70%). However, about 72.71% mentioned that premiums were too high while 60.71% of them were of the view that premium payment method was not flexible. With regard to alternative means by which SMTHEs manage their business risk, about 45.20% of respondents indicated that they saved for unfortunate events while nearly half (52.40%) said they got support from family members and friends. Another 12% were also of the view that God watched over their businesses.

Table 4

Self-reported benefits, constraints and alternative risk coping strategies among SMTHEs.

Indicators	Frequency	Percent
Insurance provides compensation for losses	140	93.30
Insurance helps revamp business after eventualities	127	56.00
Insurance contributes to business security	237	94.70
Enables access to loans	115	46.00
Promotes positive business image	127	50.81
Premiums are expensive	177	72.71
Premium payments are not flexible	152	60.71
Making claim/accessing payments in times of disaster is difficult	216	86.31
Insurance contracts are difficult to understand	92	36.62
Insurance companies do not offer the kind of insurance we need	150	60.00
Buying insurance means inviting bad luck	94	37.60
We save for unfortunate events	113	45.20
God watches over the business	30	12.00
Friends and relatives provide support during emergencies	131	52.40

Determinants of insurance uptake

As per Table 5 , both the probit and logistic regression models showed that about 23–27% of the variance in SMTHEs' insurance uptake was explained by factors considered in the model above. Similarly, the individual coefficients from the two models were qualitatively similar which reinforces the predictive relevance of the factors involved in explaining insurance uptake. To interpret the results, references were made to the probit model. The link test [(p > |z| = 0.012) and hatsq (p > |z| = 0.450)] and Hosmer-Lemeshow (χ2 = 6.81; p = .758) indicated that both regression functions were specified and fitted (Cameron & Trivedi, 2013). The results show that hypotheses H1, H2a, H2b, H2c, H2d, H3, H4, H5, H6, H7, H9, H10 and H11 were supported by the study.

Table 5

Multivariate binary models on determinants of insurance uptake among SMTHEs.

	Probit			Logit
Drivers	Coef.	% change in coefficients for unit increase in X	p	Coef.	% change in coefficients for unit increase in X	p
Sector (ref. Artisans)
Accommodation	7.60	56.43	0.00⁎	6.40	54.52	0.00⁎
Restaurant	1.60	0.03	0.06	1.40	0.07	0.07
Transport	5.60	58.93	0.03⁎	5.40	57.93	0.02⁎
Ownership type (ref. locally owned)
Foreign owned	8.41	89.34	0.00⁎	7.97	87.34	0.00⁎
Local and foreign owned	6.80	67.99	0.00⁎	7.00	67.43	0.00⁎
Position in the business (ref. Non-owner managers)
Owner-managers	5.10	47.21	0.04⁎	5.08	34.24	0.05⁎
Firm size (Number of employees used as a proxy)	0.53	0.04	0.94	0.49	0.05	0.95
Risk concerns
Operational risk	6.80	65.45	0.00⁎	6.72	63.44	0.00⁎
Technological risk	7.41	54.31	0.00⁎	7.49	52.11	0.00⁎
Health risk	4.81	42.34	0.00⁎	4.25	43.42	0.00⁎
Environmental risk	6.52	23.40	0.00⁎	2.12	22.43	0.00⁎
Unrest risk	0.63	0.08	0.63	0.65	0.06	0.63
Economic risk	1.54	0.07	0.78	1.56	0.06	0.78
Self-reported determinants
Insurance provides compensation for losses	5.81	68.45	0.00⁎	6.10	69.32	0.00⁎
Insurance helps revamp business after eventualities	4.41	45.63	0.00⁎	4.08	43.60	0.00⁎
Insurance contributes to business security	4.82	34.22	0.00⁎	3.79	34.78	0.00⁎
Enables access to loans	0.95	0.00	0.55	0.93	0.01	0.54
Promotes positive business image	0.89	0.03	0.78	0.84	0.03	0.76
Premiums are expensive	−10.63	78.45	0.00⁎	−10.64	76.43	0.00⁎
Premium payments are not flexible	−4.54	7.40	0.78	−4.56	6.54	0.79
Making claim/accessing payments in times of disaster is difficult	−11.24	71.32	0.00⁎	−11.22	100.21	0.00⁎
Insurance contracts are difficult to understand	−5.01	68.54	0.00⁎	−4.99	67.23	0.00⁎
Insurance companies do not offer the kind of insurance we need	−3.33	31.21	0.04⁎	−3.11	34.21	0.04⁎
Buying insurance means inviting bad luck	0.11	0.10	0.56	0.09	0.09	0.56
We save for unfortunate events	−6.89	70.81	0.00⁎	−7.86	70.23	0.00⁎
God watches over the business	−5.05	23.13	0.00	0.03	21.41	0.00
Friends and relatives provide support during emergencies	−1.96	0.11	0.05⁎	−1.98	0.13	0.04⁎
Constant	0.31		0.00⁎	0.43		0.00⁎

p-value < .05; R2 for probit model = 0.23; R2 for Logit model = 0.27; link test [(p > |z| = 0.012); hatsq (p > |z| = 0.450)]; Hosmer-Lemeshow (χ2 = 6.81; p = 0.758).

Table 5 shows that accommodation and transport related SMTHEs were about 56% and 59% (respectively) more likely to have at least an insurance cover when compared to artisans. For ownership, locally owned SMTHEs had less chances of taking up insurance cover in comparison to those owned by foreigners or by both locals and foreigners. With regard to the position in the business, owner-managers were about 47% more likely to engage in risk transfers via insurance than businesses ran by non-owner managers.

Perceived risk generally emerged as a significant determinant of insurance uptake. As evidenced in Table 5, four (4) out of the six (6) specific risk dimensions increased the odds of SMTHEs' insurance uptake. In order of magnitude, these included operational (65.54%), technological (54.31%), health (42.34%) and environmental (23.40%) risk dimensions. For SMTHEs that agreed that insurance provided compensation for losses, revamped businesses during disasters, and contributed to business security, the probability of uptake was 68.45%, 45.63% and 34.22% respectively. However, the difficulty in making and receiving claims (71.32%) and expensive premiums (78.45%) significantly decreased the odds of purchasing insurance. Similarly, when insurance contracts are difficult to understand, and companies fail to offer the kinds of insurance SMTHEs require, it reduces the odds of uptake by 68.54% and 31.21% respectively. In addition, believing in God (23.13%), saving towards unfortunate events (70.81%) and reliance on friends and relatives (11.00%) to offer support during disasters were also uncovered as inhibitors of insurance demand and uptake.

Discussion

About half of the SMTHEs involved in this study purchased insurance as a risk mitigation measure. This finding resonates with the CPT, which assumes that enterprises guard against possible threats to their operations using various strategies to maintain the flow and sustenance of wealth. Different types of risks affect insurance uptake decisions differently (see Olya & Al-ansi, 2018). The regression model indicated that operational, technological, environmental and health risks were concerns that significantly influenced insurance enrolment by SMTHEs. The more risk-averse businesses are regarding these hazards, the greater the likelihood of an insurance uptake (Schneider, 2004).

Though operational risk relates more to the internal business environment, implying some level of control over it, its high predictive relevance for insurance demand among SMEs is due to the potential losses it holds for businesses (Manaktola & Jauhari, 2007; Rittichainuwat & Chakraborty, 2008). The food service industry, for instance, is estimated to lose a third of business revenue each year to theft and fires as found in the work of Ghiselli and Ismail (1998). Regarding technology risk, cyber fraud has the tendency to reduce the credibility of doing business via the Internet or even hamper the adoption of information technology by tourism businesses (Buhalis & Foerste, 2015). Cybercrimes can be very destructive to tourism business operations since the Internet has become pivotal in facilitating marketing and transactions in the industry. Thus, SMTHEs were aware of the possible losses associated with information technology. Akin to the work of Olya and Al-ansi (2018), health risk also came up as a strong predictor of insurance demand by SMTHEs. Issues regarding road traffic accidents, poor work environment and safety are regarded as major menaces in resource-scarce low and middle-income countries including Ghana (see NIC, 2018). In March 2020, for example, the global COVID-19 pandemic affected many businesses in Ghana like other parts of the world especially in the service industry leading to several employee lay-offs and business fold-ups. The national policies in Ghana including the National Health Insurance Scheme (NHIS) and the mandatory vehicle insurance (Awunyo-Victor, 2012; NIC, 2018) which are both geared towards ensuring that individuals have insurance covers for illness and road accidents could have moderated the positive influence of health risk on insurance enrolment among SMTHEs. The study further demonstrated the relevance of environmental hazards (which are external to the business environment) in driving insurance demand. For instance, floods and storms tend to have a low probability of occurrence but their associated losses to property and life can be very huge. In 2015, for example, the city of Accra got hit by floods leading to hundreds of deaths and innumerable property losses. Restaurants, hotels, resorts and tour operations among others were affected. Though within the context of risk assessment relative to food products and services, Olya and Al-ansi (2018) found that environmental risk was a salient predictor of consumer behaviour regarding insurance uptake.

The study further revealed that the odds of adoption are likely to be high if enterprises can envisage that an insurance would indemnify future losses and contribute to business security and continuity. The perceived usefulness of insurance by SMTHEs could compensate for their concerns about high premiums when marketing insurance to them. As a deterrent however, high premiums are often cited as constraints to insurance demand as reported byGbireh (2013). This was evidenced in this study by nearly 72.7% of SMTHEs' claiming that insurance premiums were expensive, hence they were less likely to purchase insurance. Besides, the NIC (2018) and Deloitte (2019) report that the inability to buy insurance is a major reason for its low penetration in Ghana and most of Africa. It also does appear that SMTHEs were uncertain about the timeliness in the payment of claims by insurance companies. Awunyo-Victor (2012) argued that when clients perceive difficulties in making claims or accessing payments in times of disaster, it hinders their likelihood of insuring their properties.

Giesbert et al. (2011) and Gbireh (2013) opine that insurance is still an elusive concept to many, especially to small-scale business entities. SMTHEs who noted that insurance contracts were difficult to understand were less likely to subscribe to it. An examination of the level of awareness of insurance among owners of SMTHEs showed that they were not well informed about how insurance operates. For instance, it was evident that a greater number of operators (circa 88.00%) did not understand the pooling concept of insurance. SMTHEs operators were oblivious of the fact that their premiums could be used to pay for claims for other businesses that experience losses. More so, a substantial number (81.50%) of respondents were of the view that premiums are refundable once a claim has not been made. Misunderstandings concerning insurance and loss payment could lead to dissatisfaction and distrust among clients and thereby impacting negatively on uptake (Deloitte, 2019; NIC, 2018).

Moreover, when SMTHEs save against uncertainties and are convinced friends and relatives will provide support during emergencies, it crowds out the demand for insurance. Most of the SMTHEs indicated that they saved against unfortunate events but also believed that family members and friends will provide support during disaster. Especially, in Ghana, a local saving mechanism called ‘susu’ is an interest-free saving and loaning method that gives contributors quick turnaround loans for business start-ups and possibly, to absorb losses resulting from disasters. Research indicates that SMTHEs, especially those in Africa, depend mainly on personal savings, family and friends, and other informal microfinance mechanisms to finance their business operations (Ngoasong & Kimbu, 2016).

It was also interesting to note that the sub-sector type, nature of ownership and one's position in the business influence the chances of insurance uptake. SMTHEs in accommodation and transport services were more likely to insure their businesses than other sub-sectors. This is no surprise in terms of the transportation sub-sector because vehicles owners are enjoined by law in Ghana to insure their vehicles without which legal action will be evoked (see NIC, 2018). It could also be ventured that the relatively huge amount of capital required for investment within the accommodation sector in comparison to artisans and restaurants in Ghana means that insurance is a measure of safeguarding investments in that sector. Furthermore, the fact that foreign-owned businesses as well as both local and foreign-owned ones wielded higher odds of insuring their businesses than locally-owned ones suggest that these businesses were not only well-informed about the need to insure but they most likely have the necessary monetary resources needed to continually insure their businesses against potential risks. It was also not surprising that owner-managers had higher likelihoods of insuring than non-owner managers. This is because unlike non-owner managers, owner-managers are default equity holders/financiers of their facilities thus the need to ensure that their businesses are safeguarded from insolvency and disasters among others through risk transfers (Manaktola & Jauhari, 2007; Rittichainuwat & Chakraborty, 2008).

Conclusions and implications

Theoretical contributions

The purpose of this study was to understand the range of factors that influence the uptake (and non-uptake) of insurance among SMTHEs in emerging African destinations such as Ghana. The study found that broadly, perceived risks (as with the case of Olya and Al-ansi (2018) and Olya and Han [2019]) perceived benefits, scheme related factors and other coping mechanisms are important factors affecting insurance uptake. To the best of our knowledge, this study is uniquely positioned in the hospitality and tourism literature in that it has unpacked the drivers and inhibitors of insurance uptake by SMTHEs especially from an emerging resource-scarce destination perspective. Essentially, the uptake or otherwise of insurance by SMTHEs should be understood as a complex intersection of perceived risks, perceived benefits and personal coping mechanisms that need to be considered when designing micro-insurance products for SMTHEs in the sector. These findings diverge from those of previous studies (such as Williams & Baláž, 2015; Olya & Altinay, 2016; Caponecchia & Tan, 2019; Olya et al., 2019) that investigate insurance uptake, concentrating more on tourists, overlooking the supply-side of the travel and tourism industry such as SMTHE.

Specifically, the study showed that the demand for insurance among SMTHEs is largely influenced by operators perceived operational, technological, health, and environmental risks concerns. The findings on risk concerns support the CPT which assumes that firms guard against possible threats to maintain the flow and sustenance of wealth. Another important but unique contribution to the literature is that the study unearthed some of the inhibitors of insurance subscription by SMTHEs in a resource-scarce destination. For instance, expensive premiums, difficulty in understanding insurance contracts and making claims, not offering bespoke insurance services, beliefs in God and other informal coping mechanisms such as savings (susu) and reliance on friends and family were found as inhibitors. The study supports Chodokufa and Chiliya (2014) and NIC (2018) who report that SMTHEs do not have the wherewithal to cope with high premiums. An important contribution to the current literature is the finding that structural factors, which are service scheme related factors, are the main constraints to insurance uptake by SMTHEs in Ghana. Aside from perceived risk, this study’s observation of the material effect of the benefit of insurance, constraints and other alternative risk coping strategies on insurance uptake constitute a significant extension of the theorisations of the widely used CPT and AAT for predicting insurance demand. Both theories emphasise only perceived risk and ambiguity as determinants of insurance adoption.

The study also demonstrated that the sector type, ownership and one's position in the business influenced the uptake of insurance by SMTHEs. In this study, the accommodation and transport sub-sectors, foreign owned and partnered businesses, as well as owner-managers, had higher odds of insuring their businesses than others.

Managerial implications

The conclusions hold important management implications for SMTHEs as well as the insurance sector in Ghana and other sub-Saharan African destinations experiencing similar challenges. Whilst Sub-Saharan Africa is not homogeneous, Ghana is a representative sample because of its 1% insurance penetration rate which mirrors the rest of Sub-Saharan Africa except South Africa; Secondly, even though Sub-Saharan Africa is made up of many different countries, evidence points to the facts that the problems prevalent in the insurance sector cut across the different countries in Sub-Saharan Africa. As such a good number of countries in the region are likely to face similar problems like Ghana when it comes to insurance uptake. The reasons for low or no uptake of insurance in Ghana are very similar to those of other African countries across the continent. Deloitte in their 2019 study of the African insurance market noted the following common factors prevalent across African countries: inappropriate distribution channels; products not suited to market; lack of trust and other challenges including poor literacy, low income, alternatives to insurance. Finally, the characteristics of SMEs across the sub-region including but not limited to the firm's size, number of employees, firm's capital among others appear to be similar across the Sub-Saharan Africa (see Malgwi, 2005; (Osei-Assibey, Baah-Boateng, 2012)).

The study established various concerns and perceptions that affect insurance uptake by SMTHEs such as perceived risks, perceived benefits, scheme related factors and other coping mechanisms. Thus, there is a need for insurance companies in Ghana and other sub-Saharan African countries to design bespoke insurance packages and marketing campaigns to address such perceptions and concerns. Likewise, it is necessary for insurers to specifically target restaurants, artisans and non-owner managers because of their reluctance towards insurance uptake. For instance, while promotional packages or tailored packages may be relevant to these sub-sectors, more sensitisation is required to let non-owner managers appreciate the need to insure their businesses against likely volatilities in the industry.

The study also revealed that the demand for insurance among SMTHEs was positively influenced by the perceived usefulness of insurance covers, that is, 1) insurance providing compensation for losses, 2) helping to revamp businesses and 3) contributing to business security. It is, therefore, encouraged that education/sensitization is continued in this regard to enable many businesses to take advantage of insurance.

Furthermore, the cost of premiums appeared as an inhibitor to insurance uptake by SMTHEs. Hence, insurers in the region could aim at developing tailored packages that would suit the needs of SMTHEs at affordable premiums. If insurers can target SMTHEs as groups, they could reduce administrative costs and hence reduce premiums. Also, the incorporation of flexible payment terms for premiums as well as advice on risks control could reduce the severity of losses associated with disasters hence help keep premiums from soaring.

Finally, the reliance on personal savings and support from family, kin and friends as a means of financing businesses should be noted as an unsustainable coping mechanism by SMTHEs. This is because it can incapacitate businesses during disasters especially when savings are not enough to absorb a shock. Therefore, the possible risk associated with such an indigenous savings culture could be buffered using suitable insurance covers by SMTHEs.

Limitations and further research

There are some limitations in this study that provide pathways for future research. First, the study was limited for not being able to provide or develop a structure for operation of insurance relative to SMTHEs and for being unable to offer empirically supported legislative and legal solutions to insurance services within its context, areas future studies may consider investigating. Second, akin to Olya and Altinay's (2016) direction for future research, the willingness of actors to offer specific insurance against uncertainties such as operational, technological, health and environmental risks should be considered a future study. Third, the findings of the study need to be extrapolated with caution since it concentrated on Ghana. Future researchers should consider involving more regions especially in other Sub-Saharan African countries to offer a much stronger basis for the generalisability of findings. Finally, using a qualitative lens to delve deeper into how ‘culture’ affects the uptake of insurance among SMTHEs is another area worth considering for future research.