Loan refusal has been a problem facing many loan applicants at the household level and
this problem is not new to loan applicants in Ghana. Despite this knowledge, researchers
passively discuss loan refusal and do not consider the intensity of this problem. This study
analyses the effect of household income and savings on loan refusal and the intensity of loan
refusal in Ghana using the fifth round of the Ghana Living Standards Survey (GLSS-5). The
study employs the direct elicitation approach to identifying credit constrained (loan refused)
households and makes use of the Logit and Poisson regression to regress the loan refusal
variable on other covariates. The Logit model is applied to loan refusal as a binary variable
(refused and not refused) while the Poisson is applied to loan refusal as a count variable
(number of times of loan refusal). The econometric analysis of 1,600 and 1,591 households for
the loan refusal and intensity of loan refusal respectively shows that income and savings
inversely relate to loan refusal and the intensity of loan refusal at their respective significance
levels. It is also shown that low-income and low-savings households are more likely to be
discouraged from loan applications than their counterparts in high-income and savings
households. Financial institutions are called upon to generally widen their coverage and to
extend their activities more into the rural areas so as to increase the stock of loanable funds
available to rural dwellers. This will reduce the vulnerability of rural dwellers when it comes to
loan refusal.