South Africa remains one of the most unequal countries globally, yet the role of household balance sheet composition in shaping wealth inequality remains underexplored. Using data from the National Income Dynamics Study (NIDS), this study applies Recentred Influence Function (RIF) regressions to examine how assets, liabilities, and household characteristics affect the Gini coefficient, percentile wealth shares, and unconditional wealth quantiles. The findings reveal substantial heterogeneity across balance sheet components. Real estate, vehicles, retirement assets, and household possessions generally have equalising effects, whereas business and financial assets are associated with upper-tail wealth concentration. Consumer debt disproportionately reduces wealth among lower-wealth households, whereas non-consumer debt increases wealth concentration through unequal access to productive credit. The results further underscore the importance of income, education, employment, and persistent racial disparities in shaping wealth outcomes. The findings suggest that reducing wealth inequality requires policies that expand access to appreciating assets, affordable productive credit, financial inclusion, homeownership, and small-business development for historically disadvantaged households.