In this paper we analyze transition data by means of the minimum-chi-square (MCS) method instead of the more commonly used maximum-likelihood (ML) method. The analysis includes exits to multiple destinations and unmeasured heterogeneity. In the empirical application, turnover in the welfare system in Belgium is found to be very high. Median duration is 4.5 months for men and 7 months for women, although these figures overstate turnover in that exits out of welfare include those occurring as a consequence of recipients moving to another municipality while remaining on welfare.

This paper examines change in property-liability insurers' risk-taking around enactments of state guaranty fund laws. Our evidence suggests that the risk of insurers' asset portfolios increases following enactments. But this increase in risk is significant only for stock insurers. Our evidence of increased risk-taking following guaranty-fund adoptions suggests that the way these funds are organized creates counterproductive investment incentives, especially for stock companies. Because these laws were enacted by states over the period 1969–1981, our evidence on changes in risk-taking helps resolve statistical problems that have been troublesome for studies of bank deposit insurance.

Abstract Stock dividends which increase outstanding shares by less than 25 percent require a transfer from retained earnings of the market value of the new shares, a much larger transfer than that required for stock dividends of 25 percent or more. Choosing a distribution factor near, but below, 25 percent may be an indication of management optimism that future income will replenish retained earnings, avoiding constraints on future cash distributions. In this study, firms declaring 20 percent and 25 percent stock dividends are compared. The 20 percent stock dividend firms exhibit significantly greater announcement-period abnormal returns and significantly greater post-declaration cash dividend growth. These effects are greatest for firms incorporated in states where the level of retained earnings more strictly constrains the payment of cash dividends.

[Stock dividends which increase outstanding shares by less than 25 percent require a transfer from retained earnings of the market value of the new shares, a much larger transfer than that required for stock dividends of 25 percent or more. Choosing a distribution factor near, but below, 25 percent may be an indication of management optimism that future income will replenish retained earnings, avoiding constraints on future cash distributions. In this study, firms declaring 20 percent and 25 percent stock dividends are compared. The 20 percent stock dividend firms exhibit significantly greater announcement-period abnormal returns and significantly greater post-declaration cash dividend growth. These effects are greatest for firms incorporated in states where the level of retained earnings more strictly constrains the payment of cash dividends.]

[This paper reports the results of an experiment that examines the influence of uncertainty about the probability that a future loss will occur ("ambiguity") on auditors' and financial statement users' judgments about appropriate reference to contingent losses in audit reports. Application of Einhorn and Hogarth's (1985) ambiguity model suggests that, with respect to losses of low (high) probability, both auditors and users will act as if an ambiguous probability of loss is higher (lower) than a precise probability of the same magnitude, thus demonstrating a conservative (unconservative) reaction to ambiguity. In addition, since auditors may jeopardize client relations when they unnecessarily make audit report reference to contingent losses, auditors may react less conservatively to ambiguity than do users. The results of the experiment support these predictions.]

[This paper examines the relationship between the incidence of litigation events with potentially large damage awards and managers' accounting choices. We argue that the size of damage awards is a function of reported net income and net worth, and that this relationship provides management an incentive to manipulate accounting numbers. Our results indicate that managers of oil firms facing potentially large damage awards choose income decreasing non-working capital accruals relative to managers of other oil firms. Further, the results indicate that the management of these firms makes accounting choices that result in lower non-working capital accruals during the litigation period than in other years. These negative non-working capital accruals appear to result from the under-estimation of new reserves.]

This paper studies the problems of estimation and inference in the linear trend model yt = α + βt + ut, where ut follows an autoregressive process with largest root ρ and β is the parameter of interest. We contrast asymptotic results for the cases | ρ | < 1 and ρ = 1 and argue that the most useful asymptotic approximations obtain from modeling ρ as local to unity. Asymptotic distributions are derived for the OLS, first-difference, infeasible GLS, and three feasible GLS estimators. These distributions depend on the local-to-unity parameter and a parameter that governs the variance of the initial error term κ. The feasible Cochrane–Orcutt estimator has poor properties, and the feasible Prais–Winsten estimator is the preferred estimator unless the researcher has sharp a priori knowledge about ρ and κ. The paper develops methods for constructing confidence intervals for β that account for uncertainty in ρ and κ. We use these results to estimate growth rates for real per-capita GDP in 128 countries.

Using an expanded version of the purchasing-power-parity condition we construct simultaneous equation models for three key exchange rates which incorporate meaningful long-run equilibrium relationships and complex short-run dynamics. We show that fully dynamic out-of-sample forecasts from these models are capable of significantly outperforming those of a random walk model over horizons as short as 3 months, and that they are also more accurate than the vast majority of professional forecasts.

The Memoranda of Discussion provide detailed records of Federal Open Market Committee (FOMC) meeting deliberations. Procedures are developed for coding the textual data in the Memoranda and assessing the reliability of those codings. The codings are then used in the estimation of parameters of individual FOMC members' reaction functions. Data from the 1970 to 1976 period are employed in the estimation. In the future, similar methods could be used to analyze newly released transcripts of FOMC meetings held after 1976.

In Imbens and Ingrist (1994), Angrist, Imbens and Rubin (1996) and Imbens and Rubin (1997), assumptions have been outlined under which instrumental variables estimands can be given a causal interpretation as a local average treatment effect without requiring functional form or constant treatment effect assumptions. We extend these results by showing that under these assumptions one can estimate more from the data than the average causal effect for the subpopulation of compliers; one can, in principle, estimate the entire marginal distribution of the outcome under different treatments for this subpopulation. These distributions might be useful for a policy maker who wishes to take into account not only differences in average of earnings when contemplating the merits of one job training programme vs. another. We also show that the standard instrumental variables estimator implicitly estimates these underlying outcome distributions without imposing the required nonnegativity on these implicit density estimates, and that imposing non-negativity can substantially alter the estimates of the local average treatment effect. We illustrate these points by presenting an analysis of the returns to a high school education using quarter of birth as an instrument. We show that the standard instrumental variables estimates implicitly estimate the outcome distributions to be negative over a substantial range, and that the estimates of the local average treatment effect change considerably when we impose nonnegativity in any of a variety of ways.