Legacy and credit scores in South Africa

This paper assesses the impact of legacy and credit scores on access to bank credit in South Africa. Typically, credit data focuses on socio-economic information. However legacy variables, factors beyond the control of individuals, have not been well considered. We find that although credit scores clearly influence access to credit, legacy has a statistically significant impact on credit scores and on the amount of credit granted. While these results can be interpreted as bias against those with negative legacy, they can also be explained in terms of information asymmetry and the relative ability to enforce contracts with clients who have positive legacy.     

Studying (Small) businesses with the Michigan Employment Security Commission longitudinal data base

This paper addresses the usefulness of a longitudinal data file constructed from records on employers from the Michigan Employment Security Commission. We describe the main features of the data file, which includes quarterly (and in some cases, monthly) data from the third quarter of 1978 through the first quarter of 1983, plus the fourth quarters of 1983–87. We then illustrate the uses of the data with two examples: (1) studying changes in the Michigan economy, in particular the early growth and survival of new units of different sizes; and (2) studying the behavior of wages and employment following changes in ownership.Judith Connor and Steven Heeringa are primarily responsible for Section II and the Appendix, John Jackson for Section III, and Charles Brown for Section IV.     

More on public information and stock price

In a recent issue of this journal [2] McCain and Millar examined whether “favorable” and unfavorable” stock analyses appearing in the Wall Street Journal column “Heard on the Street” could be used to predict one-day, seven-day, and six-month price movements of the affected stocks. In this note we question the conclusions of that study based upon methodological grounds.     

Modeling the Conditional Probability of Foreclosure in the Context of Single-Family Mortgage Default Resolutions

Both empirical and pricing-simulation models of mortgage default focus on foreclosure in a one-step decision framework. Such models are misspecified to the extent that mortgage default and foreclosure are two separate decisions or events, where foreclosure is but one outcome of a default episode. This study examines the dynamics of mortgage borrower default episodes using a large sample of FHA-insured single-family mortgages. We estimate the influence of borrower characteristics, mortgage terms, and economic conditions on probabilities of various resolutions, highlighting under what conditions foreclosure is more likely to result from mortgage default.     

Generalized densities of order statistics

Let X ₁, ... , X _n be independent identically distributed random variables with distribution F . We derive expressions for generalized joint 'densities' of order statistics of X ₁, ... , X _n , for arbitrary distributions F , in terms of Radon–Nikodym derivatives with respect to product measures based on F . We then give formulae for conditional distributions of order statistics and use them to derive results concerning Markov properties of order statistics, formulae for distributions of trimmed sums, and other useful representations. Our approach leads to simple and natural expressions which appear not to have been given before.     

Clustering at the Movies

Weekly box office revenues for approximately 100 successful motion pictures are analyzed by use of a finite mixture regression technique to determine if regular sales patterns emerge. Based on an exponential decay model applied to market share data, four clusters of movies, varying in opening strength and decay rate, are found. Characteristics of the clusters and implications for future research are discussed.     

Informed Trading Around Earnings Announcements: Another Look

This study is an empirical test of the Easley, O'Hara, and Srinivas (1998) multimarket sequential trade model of stock and option markets. We employ two approaches to determine the information content of signed stock and option trades executed around quarterly earnings announcements. The first approach expands the vector autoregression (VAR) technique of Hasbrouck (1991a) to include signed option trade volumes and inter‐trade durations. Estimates from the VAR models provide insight into whether both equity and option trades are viewed as informative by the equity specialist. The second approach focuses on the information content of the earnings releases to determine whether signed equity and option trades executed prior to the announcements are informed. Results indicate that although informed traders prefer to transact in both markets around earnings announcements, option market transactions contain no incremental information.     

Capacity utilization and depreciation-in-use

In this paper, the effect capacity utilization has on the depreciation of capital goods is studied, starting from a quadratic approximation to a normalized short-run cost function. Expressions for the optimal rate of capacity utilization, the demands for labour and energy, and for investment are obtained. Investment is a function of past capital stock, expected future capital stock, expected future relative prices and expected future depreciation rates. Expectations are modelled via instruments. The model is tested using US total manufacturing quarterly data. Depreciation dependent upon usage is found to be both statistically and economically significant.