Determinants of loss reserve errors: evidence from the general insurance market in South Africa

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

A loss reserve is the estimated liability for unpaid claims on all losses that occurred prior to the balance sheet date. The loss reserve is the most significant liability on the balance sheet of a general insurance company, often driving its overall financial performance. The loss reserve is calculated to determine the claims liability for published accounts, internal accounts, statutory accounts, business plans and budgets. It is also required for purposes of pricing and in case of a merger or acquisition. The purpose of the loss reserve can affect the methodology used as well as the extent of over-reserving or under-reserving. Additionally, under-reserving and over-reserving can be driven by the intent to smooth the inome of the general insurer, to mask financial weakness or to defer taxes. This study examines the loss reserve errors in the South African general insurance industry. The study estimates the loss reserve errors using annual firm level data on 79 general insurance companies from 2007 to 2014. The study then proceeds to examine the hypothesised effect of firm level characteristics on the estimated loss reserve errors within a panel data framework. The panel data regression models are estimated using the ordinary least squares technique, the random effects technique and the fixed effects technique. The findings suggest that South African general insurance industry is characterised by over-reserving. Specifcally, approximately two-thirds of the sample reported incidence of over-reserving. The results of the panel data regression analysis indicate that tax shield, financial weakness and premium growth are the significant drivers of reserve errors in the market. Tax shield was found to have a positive relationship with loss reserve errors, whereas financial weakness and growth were found to have an inverse relationship with loss reserve errors. Business line diversification and reinsurance were not found to be significant variables in the model. The management of South African general insurers and regulation of the industry should be directed towards ensuring that general insurers do not manipulate reserves to defer taxes, fund growth through more competitive premiums, or manipulate the perceived financial strength. Additionally, this study identified issues relating to the quality of loss reserve information supplied to the regulator. There is scope for improving the quality and consistency of the loss reserve data supplied to the regulator by the general insurers.