Search

This study examines the information content of the sovereign credit ratings assigned by 11 international credit rating agencies to 55 countries for the period from 1989 to 1997. The empirical work is divided into three parts, which separately investigate the determinants of: (1) foreign currency sovereign ratings; (2) local currency sovereign ratings; and, (3) transfer risk -as proxied by the difference between local and foreign currency ratings. In particular, the study explores the extent to which these determinants can be explained by: quantitative and qualitative variables; by macroeconomic and balance-sheet variables; and, by average and lagged values of macroeconomic variables. It also examines the systematic differences across rating agencies and across geographic regions. It compares the robustness of ordered probit analysis and OLS regression to explain sovereign credit ratings; and, it quantifies the impact of the preferred model of sovereign creditworthiness indicators on credit ratings and on transfer risk. With respect to the analysis of foreign currency sovereign ratings, this empirical work differs from previous studies in the following ways: it uses a larger sample and more recent data than any study to date; it expands the set of explanatory variables to include balance-sheet variables; it pays attention, for the first time, to the lagged structures of explanatory variables; it utilises a more appropriate statistical model, ordered probit, given the ordinal, discrete nature of credit ratings; and, it provides accuracy measures at the rating notch level for the models while comparing ordered probit and OLS performance. Additionally, this study expands the existing literature by presenting the first systematic analyses of the determinants of local currency sovereign ratings and the determinants of transfer risk. Finally, this study comprises the first attempt to quantify the impact of the determinants of sovereign credit ratings and transfer risk. The findings reveal that there are significant differences in the determinants of foreign and local currency sovereign ratings and, that the agency assessing the sovereign's creditworthiness and the sovereign's geographic region play an important role in the determination of credit ratings. The results also suggest that uses of ratings which presuppose comparability may prove inadequate, and that differences across agencies may provide sovereigns with an incentive to select those which will provide potentially more favourable ratings. Further, ordered probit analysis proved more successful at modelling sovereign ratings than the linear specification. Important differences are also found regarding the statistical and quantitative significance of the explanatory variables.

Financial institutions have battled with handling and determining the creditworthiness of their clients in the recent past. The ever-increasing customer base makes it hard for financial institutions, especially banks to follow due process of determining if a customer qualifies for a loan or not depending on one's credit history manually. As a result, there have been delays in processing customer loans, making banks and other financial institutions inefficient. Automation of these tasks has come as a factor of necessity, to improve the speed, cost, and efficiency of processing loans. Thus, an AI web-based application that predicts the probability of borrowers’ failure to repay the loans is a handy solution for this time. The system will auto-collect historical borrowing and repaying data for that particular borrower within the shortest time with high precision whenever an individual uploads the personal data. The prototype will apply cloud cutting-edge AI and machine learning services to analyse the borrowers' creditworthiness and apply the recommendation to achieve the following: Identifying personal information of the proposed borrower, evaluating the prerequisite information for loan approval or decline, determining the credibility, notifying the lender if any loan default history, and recommending approval or disapproval based on the history of the borrower. This application will save financial institutions stress and time, avoid losses in the lending business, reduce the loan process time, decrease the likely risks associated with loans, and save the costs of the admission department.