Factors Influencing Risk Management Decision of Small and Medium Scale Enterprises in Ghana

This research seeks to study the factors that enhance or preclude owners of SMEs in Ghana in making risk management decisions. The study was conducted with managers of SMEs in four regions in Ghana. The researchers adopted a quantitative approach and employed STATA 10 and SPSS version 20 in the analysis. Stratified and simple random sampling techniques were used to select the sample units. The probit model was used in the analysis of data. A total of 447 SMEs were sampled for the study, with at least 111 from each of the selected regions. The probit results show that the demographic factors indicate a positive influence on the likelihood that managers will take risk management decisions. All of the business related demographic factors are significant at various levels and positive, except for risk-loving. The economically related factors, such as the estimated amount at risk, the estimated cost of risk management and the estimated total monthly income after tax all have a positive influence on risk management decision making. However, government and tax policies are perceived to negatively influence risk management decisions by managers. We recommend that institutions working closely with SMEs acquire the expertise to train the managers of SMEs on risk management practices.


Introduction
Risk management can be described as the process of determining the maximum acceptable level of overall risk for engaging in a proposed activity. It involves using risk assessment techniques to determine the initial level of risk and, if it is excessive, to develop a strategy to ameliorate appropriate individual risks until the overall level is reduced to an acceptable level. Risk management approaches differ from one firm to the next, which, in part, reflects different risk management goals.
According to Saeidi et al. (2013), a paradigm change has occurred in how organizations view risk management, with the current view being more holistic, rather than viewing risk management from a silo-based per-Board for Small Scale Industries (NBSSI) was also established within the then Ministry of Industry, Science and Technology to meet the needs of small and medium scale businesses.
Basing its assessment on the number of employees, the Ghana Statistical Service (GSS) defines small scale enterprises as firms with fewer than 10 employees, and medium and large-sized enterprises as those with over 10 employees. Generally, SMEs in Ghana include: • Micro enterprises: Employ up to 5 employees with fixed assets (excluding realty) not exceeding the value of $10,000 • Small enterprises: Employ between 6 and 29 employees with fixed assets up to $100,000 • Medium enterprises: Employ between 30 and 99 employees with fixed assets of up to $1 million.
Small and medium scale enterprises have been recognized as the engines to achieve the growth objectives of developing countries because they mobilize idle funds, are labor intensive, employ more labor per unit of capital than large enterprises, promote indigenous technological know-how, are able to compete (but behind protective barriers), use mainly local resourcesthus have less foreign exchange requirements -cater to the needs of the poor and adapt easily to customer requirements (flexible specialization).
According to Gélinas & Bigras (2004), a typical SME is dominated by one person, with the owner/manager making all major decisions. These owner/managers are often characterized by limited formal education, limited access to and use of new technologies, limited market information, limited access to credit from the banking sector, and weak management skills. In addition, SMEs experience extreme working capital volatility and the managers lack modern technical knowhow and their inability to acquire modern skills and technologies impede SMEs growth opportunities.
Despite their dynamic role in the country's development, operators of small and medium enterprises face external and internal risks in their businesses, which, threaten the performance, profitability and sustainability of the business. The following external risks are among those faced: natural disasters (e.g., flooding and earthquakes), wars, political crises, and government policies. The following are internal risks pertaining to the running of the business: risk of reduced demand for products and services, risks to ability to compete in the marketplace, high labor turnover, injury, and risks to financial profitability and growth. It is noteworthy that some of the risks that SMEs face can be controlled through the use of appropriate actions, whereas others are unpredictable and uncontrollable. The occurrence of any of these risks may have a disastrous effect on the entrepreneurs' effort for business success, which may lead to bankruptcy and thus deny the country the expected contribution to national growth.
The ability of managers of SMEs to address the dynamics of the emerging global market is also largely influenced by their ability to carefully identify and analyze the type of risks their business faces and then to examine the factors that need to be taken into account to manage them. Management of an enterprise, including managing the risks the enterprise is exposed to and providing other support services, is perceived to be cost prohibitive and non-value adding (Mambula, 2002). Additionally, institutional and legal structures that facilitate the management of risk by SMEs are lacking (Mensah, 2004). This is the context for the current research, which seeks to study the factors that enhance or preclude owners of SMEs in Ghana in making risk management decisions. This study is especially relevant because it provides policy makers and owners of SMEs with criti-Electronic copy available at: https://ssrn.com/abstract=2548430 Factors Influencing Risk Management Decision of Small and Medium Scale Enterprises in Ghana cal information pertaining to factors that influence the extent of managers' decisions to undertake risk management practices. The study also contributes to the existing literature on risk management in Ghana, as well as building and testing generic models that can be adapted and applied to ascertain the risk management decision behavior of managers in other developing countries.
The following are factors that enhance or preclude owners of small and medium scale enterprises in Ghana in managing risk to which their enterprises are exposed: demographic, economic, government policies, social, and business characteristics, among others.
More specifically, the influence of the following factors will be determined in this research:

Model
This study modeled the business manager's choice of risk management strategies in an expected utility framework. This model is adopted from Velandia et al. (2009) in their study to find the factors affecting farmers' utilization of agricultural risk management tools.
This framework assumes that different business managers assess their end-of-period expected utilities from their own business specific risk and risk preferences.
This approach further assumes that the risk management decision fundamentally affects the net return distribution of each business manager.
The business manager then examines his or her net return distribution by considering the certainty equivalent for the risk management decision and calculates its associated reservation cost. The reservation cost is the amount that would make the business manager indifferent to making a risk management decision.
The business manager then compares the reservation cost with the actual cost of adoption of a risk management strategy and makes a decision to adopt a risk management strategy if the reservation cost is larger than the actual cost. This is equivalent to having a larger certainty equivalent net return with the risk management strategy relative to the projected return without the risk management strategy.
More formally, consider a business manager deciding whether to adopt a risk management strategy I ( i = 1, ..., m). This business manager evaluates each of these risk management strategies by considering its effect on the returns distribution to a set of assets, A, used in production. These assets have a stochastic rate of return A r , with mean r _ , and variance 2 A σ , reflecting the overall business risks. Financial risk is introduced through the use of debt capital. Utilizing the accounting identity that assets are equal to debt plus equity, (A = D + E) and assuming a fixed cost of debt, CD, the expected rate of return to equity ( ) E r and the variance of the return to equity ( 2 E σ ) can, respectively, be expressed as: Given the stochastic environment above, the business manager's certainty equivalent of end-of-period wealth can be approximated as follows (under known sufficient conditions): Where CE W is the business manager's certainty equivalent of end-of-period wealth (W), W _ is the mean of W, and 2 W σ is the variance of W, and ρ is the parameter reflective of risk preferences. Maximizing the certainty equivalent rate of return to equity ( CE r ) is equivalent to maximizing CE W , which can be defined as: From "Equation 1" and "Equation 2", the expression in "Equation 4" can be rewritten as: The effects of using risk management strategies are then assumed to be reflected in the changes in the mean and the variance of the asset return distribution and in the costs (C) of using these strategies for managing risks. Given this cost, the effect of using a particular risk management strategy is to reduce the rate of return to equity by C E . Taking this reduction into account, for every risk management strategy i available to the business manager, the certainty equivalent rate of return to equity can then be redefined as: The amount that implicitly equates the expected utilities from using and not using the risk management strategy is the highest cost that a manager is willing to incur for the use of risk management strategies (i.e., the reservation cost * i C ). Hence, per "Equation 5" and "Equation 6" the reservation cost can be calculated based on: Solving for * i C , we obtain the following expression; help determine the manager's reservation cost for i.
Using "Equation 8, " it is assumed that the manager will then decide to adopt a risk management strategy if the difference between the reservation cost and the actual cost of using i is greater than zero ( Where i Y = 1 if the manager adopt a risk management strategy and 0 i Y = , otherwise. The formulation in "Equation 9" makes it empirically tractable to estimate the factors influencing the adoption of a risk management strategy. In other words, once a risk management strategy is adopted, it implies that a risk management decision has been made.

Literature Review
The following section reviews research that has been conducted on the factors influencing the risk management decisions of managers/owners of SME's. Kouamé (2010) used a multivariate probit approach to show the importance of individual risk aversion, farm size, household size, head of household, and literacy as factors that increase the likelihood of adopting risk management strategies. Velandia et al. (2009) found age to decrease the likelihood of adopting a risk management tool but found farm size to increase the likelihood of adopting a risk management tool.

Individual Demographic Factors
Education was found to both increase and decrease the likelihood of adopting a risk management tool. HA 4 : Gender: Males are more likely to undertake risk management decisions than their female counterparts.
HA 5 : Family size has a positive effect on the decision to undertake risk management.

Business related demographic factors
It has been posited that risk perception is an indis-  (Pfeifer, 2008) and are more likely to become entrepreneurs (Ahn, 2010). The least risk averse are apparently those who can best assess and manage risks (Cho & Orazem, 2011).
Hypotheses; HB 1 : The number of years as a manger in the business has a positive influence on the managerial decision to undertake risk management.
HB 2 : Managerial knowledge in risk management has a positive effect on the decision to undertake risk.

Economic Factors
It is envisaged that the larger the capital base of the enterprise, the more likelihood will its risk be managed because the amount that would be lost could be large. Also depending on the type of risk, the risk might not affect the entire capital base. If the capital at risk is large, it will positively influence the likelihood of adopting a risk management strategy. In addition, depending on where and how the business sourced its capital, the owner may decide either to manage risk or not. If the capital was sourced with collateral, then this is expected to have a positive influence on risk management decision. It is also expected that daily sales/ income influence risk management decision. The larg-

Government Policies
Government policies may either positively or nega-

Business Characteristics
Exposure to some risks is also dependent on the location of the enterprise. If the area is risk prone, then it will positively influence the adoption of a risk management strategy. On the issue of the number of staff, the number of workers employed depicts the capital base of the enterprise. Therefore, a higher number of staff has a positive influence on risk management decision.
If the type of business is less risky, it is also expected this will have positive effect on the decision to undertake risk management.
Hypotheses; HE 1 : Business location has a positive effect on the decision to undertake risk management.  Ranong & Phuenngam (2009)

Methodology
The study was conducted in four regions in Ghana including Greater Accra Region, Ashanti Region, Western Region and the Northern Region. The population for the study includes all SMEs in these regions.

Research Design
This research was designed to take into account the

Sample Selection and Technique
To run a regression analysis, a total of 10 cases for each The sampling distribution is presented in figure 1.
A pilot test of the questionnaire was initially conducted on 20 SMEs and the questionnaire was revised based on the information gathered from the field. After administering the questionnaires, the data were passed through a series of scrutiny and cleaning using the statistical software package (SPSS), after which the descriptive phase of the data analysis was complete.
Out of 448 questionnaires administered, 447 were retrieved representing a response rate of 99.8%.

Data Analysis
Using "Equation 9, " the outcome of interest is risk management decision and this is captured as a binary variable. Both the logit and probit models are often motivated in terms of a latent variable specification, but the choice of models depends on whether the error term is assumed to have a standard logistic distribution or a standard normal distribution. These models assume that there is some continuous latent variable y* that determines the decision to adopt a risk management strategy. We can think of y* as the business manager's decision to adopt a risk management strategy. If y* is positive, the business manager will choose to adopt a risk management strategy and the observed binary outcome equals 1. Otherwise, the business manager will not adopt a risk management strategy and the observed value equals 0. Then, the latent variable y* is modeled by a linear regression function of the independent variables xi and it is assumed that the error term in this equation has a standard normal distribution. Therefore, the probit model is estimated by the method of maximum likelihood estimation.
More specifically, the model is of the form; Where the dependent variable ( ) i Y represents whether a risk management decision has been made, and the independent variables ( ) i X include individual demographic, business related demographic economic fac-  Table 1.

Data characteristics
The data analyzed show that 281 ( With business security, 5% blamed it on taxation and 95% on bad policy. Eleven respondents, representing 100%, attributed the difficulty in loan acquisition to bad policy. It is also worth noting that 52 respondents, representing 11.6% of the entire sample, indicated a positive impact on their business due to perceived good government policy.
To ascertain the knowledge of managers in taking risk management decisions, respondents were asked whether they take risk management decisions.
Approximately 54% of the respondents positively affirmed to taking risk management decisions. A further probing question asked whether risk management as-sessment has ever been conducted on their businesses.
This question revealed that approximately 36% of the managers had a formal risk assessment conducted for their business, which informed them appropriately on the needed mitigating measures to put into place. The study also elicited from managers the extent of their knowledge of risk management practices on the scale of 1 (lowest knowledge) to 5 (highest knowledge). Approximately 25% rated their extent of knowledge on risk management practices as high, approximately 5% rated it lowest and almost 23% did not rate, which is an indication of no knowledge of risk management. This situation implies that there is still a substantial gap in the knowledge base of managers of SMEs as far as risk management is concerned.

Descriptive Statistics of the Variables used in the Regression Analysis
The descriptive statistics of the variables used in the analysis is shown in

Probit Regression Results and Discussion
To differentiate between occupational safety risk measures and business risk measures, the responses on the extent of knowledge of risk management practices, the conduct of risk assessment of the enterprise and the adoption of a risk management strategy (see Table 3) were scored on the scale of 1 to 4. Any score above 2 implies the manager has been taking risk management decisions. Additionally, once a risk management strategy is adopted, it implies a risk management decision has been made. coefficient means that managers with those attributes are less likely to take risk management decision, and a positive coefficient means they are more likely to take risk management decision.
All the demographic factors show a positive influence on the likelihood of the managers to take risk management decisions. Apart from the years of education and gender, the other demographic factors are not significant. Years of education is significant at 10%, meaning that managers with more years of education are more likely to take risk management decisions than those with less years of education. Male managers are more likely to take risk management decisions than their female counterparts, and this is significant at 1%. These findings are consistent with earlier findings (Kouamé, 2010;Shapiro & Brorsen, 1988;Valentia et al. 2009).
All the business related demographic factors are significant at various levels and positive, with the exception of risk loving. The number of years as manager is significant at 1%, indicating that managers with more experience are more likely to take risk management decisions. Knowledge of risk management is also significant at 1%, meaning that managers with some level of knowledge of risk management are more likely to take risk management decision. At the 1% level of significance, managers who own other businesses elsewhere are inclined to take risk management decisions. These managers are most likely to be risk averse and as part of their risk management plan, decide to establish other businesses elsewhere.
Managers who are risk loving are found to be less likely to adopt a risk management strategy. These assertions are consistent with that posited by Cho & Orazem (2011).
Economic factors, such as the estimated amount at The results of government policies and taxes indicate that these factors negatively influence the risk management decision taking of managers. Government tax and government policies affect the decision to manage risk negatively at the 1% and 10% level of significance, respectively. An increase in the tax businesses pay on their annual profits (corporate tax) will not lead to a change in output and prices in the short run, all other things being equal. In the long run however, an increase in tax will put firms out of business if they were earning just normal profits before the tax increment, and this will negatively influence risk management decisions. A tax stamp is a statutory tax collected from small-scale selfemployed persons in the informal sector on a quar-  with respect to their bases must not be ignored in tax research (Wolswijk, 2007). This research will help in

Results of Hypotheses Testing
The factors hypothesized to affect the risk manage-

Conclusion and Policy Recommendations
Small and medium scale enterprises are the engine of growth of the economy and a good provider of employment, and, therefore, they have been one of the major areas of concern for many policy makers. 6. To avoid distorting tax revenue inflows, to ensure high tax elasticities and to prevent distortions to economic decisions on saving, investment, consumption and other business related variables, the Revenue Authority is encouraged to adopt a broad base, low rate approach to taxation. This approach will help managers of SMEs to manage business risk and will contribute to the growth of SMEs in Ghana.
7. Government prudential policies that have an effect on SMEs should be carefully formulated and these policies should be well articulated and explained to stakeholders so that they are not regarded as deterrents to business development in Ghana.

The Bank of Ghana through the Monetary Policy
Committee should collaborate with the financial institutions to reduce interest rates in order to ameliorate the negative effect of high interest rates on business investment and growth. 9. Information on interest rates by the various financial institutions should be made available by the National Board for Small Scale Industries to prospective business men and women so that they can make informed decisions on when and where to borrow.