Determinants of Capital Structure Choices by Listed Firms in Zimbabwe under Hyperinflation and Dollarization

Thumbnail Image
Mbulawa, Strike
Journal Title
Journal ISSN
Volume Title
The paper uses panel data, from 2000 to 2013, to examine the key determinants of capital structure choices for Zimbabwe listed firms under hyperinflation and dollarization by: (1) providing a reduced form model which isolates the key factors consistent with the unique situation for Zimbabwe; (2) testing the existence of a nonlinear relationship between leverage and capital structure factors; (3) ascertaining the significance of marginal effects of explanatory variables due to inflation; and (4) establishing how the behaviour of firm managers influenced the choice of leverage. In an inflationary environment, the main factors explaining the choice of debt were profitability, non-debt tax shield, payout ratio, ownership structure, hyperinflation dummy variable, growth opportunities and asset structure. Under dollarization, leverage was explained by changes in revenue, firm size, short-term liquid assets, dividend payout ratio, taxation and the industry dummy variable. Information asymmetries, the use of short-term debt and the strong influence by firm managers on the choice of leverage were prevalent during the period of inflation. Firm size and liquidity explained use of long-term debt during dollarization. Debt finance had a nonlinear relationship with firm size and managerial ownership. A reversed pecking order of finance is suggested by evidence in this study. The study shows that the composition and level of debt are important under the review period. The key implications for policy require the streamlining of access and use of bank finance and funds from capital markets. Access to capital and money markets by firms can be supported by improving the flow of quality information and efficient credit rationing policies.
Pecking order hypothesis , Zimbabwe , Capital structure , Hyperinflation , Static trade-off theory , Agency theory , Panel data