Predictors of Post Start-Up Survival of SMEs in a Strained Economy: A Case of Blantyre, Malawi.

Abstract

A study was conducted to establish the determinants of SMEs’ survival beyond start-up in a strained economy. The study was conducted in Blantyre targeting SMEs in both rural and urban areas. The study design was cross sectional and followed quantitative research methods of data collection and analysis. A total of 191 participants were sampled using simple random sampling. A structured questionnaire was used to collect data on the following variables: age, gender, business type, education, source of income, Employment status, employment category, job satisfaction, business survival, business life span, turnover, capital, loan access, repayment problems, market opportunity, target market, incubation challenge, constraints and strategies on copying mechanism. The data were analyzed using SPSS version 20.0, in which a binary regression model with “business survival” as response variable, was fitted into the significantly correlated / associated predictor variables. Business survival was significantly associated with gender, r=0.64, p=0.001 and inversely associated with age of participants, r=-0.63, p=0.001 and the business life span r=-0.70, p=0.001. A male owned business was more likely to survive (Adjusted OR = 1.0, c2 =25.091, df=2, CI=0.20-0.83, P=0.014) than that owned by a female participant. Compared to a business owner who is aged less than 25 years, a business owner aged between 25 and 40 years was 2.57 times more likely to survive the business within the first 5 years (Adjusted OR=2.57; c2 =25.091, df=2, CI=1.07-6.21, P=0.035) and participants aged more than 40 years were 9.16 times more likely to survive their businesses beyond the first 5 years of startup (Adjusted OR=9.16; c2 =25.091, df=2, CI=2.98-28.08, P=0.001). The results show that business lifespan (r=0.70, P=0.001) and copying strategies were not significantly associated with Business survival because they were also significantly correlated with gender. In addition, access to loan (r=0.54, P=0.001), reason for banks to refuse granting loans (r=0.58, P=0.001) and target market (r=0.88, P=0.001) were also significantly correlated with gender. Hence these predictors did not add value when the effect of gender was already accounted for in the model. Therefore participants’ gender and age were important predictors of business survival in the final reduced model.