Determinants of Financial Decision Making Rationality in Deposit-taking Savings and Credit Cooperatives in Kenya

Abstract

This thesis quantitatively established the effect of selected determinants on the financial decision making rationality exercised by SACCO members and their SACCO managements, for the complete rational man assumption – homoeconomicus, does not always hold. Rationality bounds in financial decision making as espoused in bounded rationality theory, whose analytical proof was incidental, were determined to reflect how much SACCOs actually rationalize their financial decisions. The objectives were: to establish the effects of first, prior knowledge about a financial decision; secondly, prospects of posting wealth increase after an irrational decision; thirdly, prospects of posting wealth increase after a rational decision as the determinants; and lastly, effect of wealth movement on the determinants as inputs into a binomial multi-period Bayesian decision model, using longitudinal design. Both frequentist and Bayesian paradigms applied where wealth movement modeled by geometric Brownian motion through Monte Carlo simulation acted as the observable dimension model with total assets (wealth) as the observable dimension variable and financial decision making rationality as the unobservable dimension variable in the process. Stochastic discrete and continuous time cases in were examined. This behavioural decision research used diffusion finance and market microstructure models to explore human intrinsic determinants of financial decision making rationality on a 0 – 100% open interval scale. The three biggest Kenyan Savings and Credit Cooperative Organizations namely, Unitas, Mwalimu National and Stima were purposively targeted. Two questionnaires were administered; one on the members and the other on management staffs. Simple random sampling was used on members amounting 271 and 46 management staffs. Hypotheses about population mean and about differences between populations means carried out tested significant. The results showed that prior knowledge affected financial decision making rationality positively and should therefore be maximized, but prospects of posting a wealth increase after a rational decision affected positively up to local rationality maximum, beyond which negative effect set in, suggesting that it should be maximum at this point and minimum at 0.6321. Prospects of realizing a wealth increase after making an irrational decision on the other hand, affected financial decision making rationality negatively and more adversely in the entire range; hence xxiv should be minimized. Wealth increase and decrease was found to increase and decrease financial decision making rationality respectively, hence increases are favourable. On average, a SACCO member updated 11 times in the last 10 years; translating into an updating period of 10.91months. Only about 58.81% of financial decisions rationalization is exercised; the least educated group posting the highest rationality level of 76.34%. The rest of decisions were largely made on the basis of heuristics; meaning that standard finance and classical economic theories need urgent revision. The model used was validated through psycho-social economic equation which presents invaluable lessons in risk management. Further, that the rationality learning age limit is 67.27 years. Portfolio diversification problem entropy concept featured prominently coupled with the principle of minimum guess work. The concept of generational entropy-q substitution rate emerged. It is recommended that financial decision making entropy be adopted as a superior criterion to education for employee selection and credit rating by SACCO managements. Entropy-q rationality theory and model were proposed, including related problems solution tables; that are likely to find a broad range of applications in industry.