Foong Soon Cheong

Evidence from observational studies plays a central role in shaping public policy in health, education, and financial regulation, where randomized experiments are rarely feasible. Propensity score matching (PSM) is a widely used method to approximate fair comparisons between treatment and control groups. Incorporating machine learning into the estimation of propensity scores can strengthen prediction and enhance the credibility of findings. However, stronger predictive models create a “predictability paradox”. As predictive accuracy improves, estimated propensity scores for treated and control units become more distinct when treatment assignment is strongly predictable from observed covariates, revealing limited overlap between groups. In the limit, near-perfect prediction produces near-complete separation between groups, rendering traditional matching infeasible and confining inference to a narrow subset of units near the boundary of the propensity score distribution, a setting analogous to a regression discontinuity design (RDD). Researchers thus face perverse incentives to use weaker models for statistically significant but spurious results. These dynamics jeopardize the reliability of evidence for policy. To safeguard decision-making, we propose a simple reform: require that studies using PSM disclose model error rates, including false positive and false negative rates, along with information on overlap and effective sample size.

Financial crises repeatedly reveal organizations that appear internally aligned while failing to recognize accumulating tail risks. This paper argues that cohesion is observationally ambiguous. It can arise from information integration, in which heterogeneous inputs are debated and synthesized, or from exclusion, in which variance is removed through conformity pressure, gatekeeping, and intolerance of dissent. This distinction is formalized using a signal aggregation model in which an organization maintains an anchor belief and achieves agreement through two exclusion channels: report shrinkage toward the anchor and a tolerance rule that discards reports deviating beyond a threshold. Relative to a full inclusion benchmark, exclusion based cohesion jointly produces state contingent bias that is small in normal regimes but grows sharply under displacement, illusory precision in which observed disagreement falls as tail regime estimation error rises, effective concentration of decision inputs below the nominal participant count, and, when the anchor updates from filtered aggregates, dynamic lock in with delayed regime recognition and abrupt correction. External inputs that bypass internal filtering shorten recognition delays. The model yields testable governance diagnostics linking latent fragility to observable patterns in recorded dissent, anonymous to formal voting gaps, scenario set diversity, pipeline and method concentration, and anchor lag. The central implication is that governance systems should treat low internal conflict and unanimity as potentially diagnostic of variance depletion and should monitor whether heterogeneity is integrated or excluded before stress reveals the difference.

Past studies have observed that the low (high) accrual portfolio in the accrual anomaly consists of firms with high (low) credit risk, and have suggested that the abnormal return in the accrual anomaly arises from buying (selling) stocks with high (low) credit risk. In this paper, I first investigate whether the low accrual portfolio is indeed dominated by firms with higher credit risk. I find that this claim is not necessarily true. Next, I regress the abnormal return on both the level of accrual and credit risk. The regression is repeated using both decile ranking and actual values. In both cases, I find that the level of accrual is always statistically significant and negative. Finally, I investigate the claim that the abnormal return in the accrual anomaly is due to taking a long (short) position in stocks with high (low) credit risk. In each year, to control for credit risk, I first rank all firms by both their level of accrual and credit risk. The ranking for accrual and credit risk are independently determined. I require that in each year, the long position (in the low accrual decile) and short position (in high accrual decile) are equally weighted within each credit risk decile. After controlling for credit risk, I find that the abnormal return from Sloan’s accrual trading strategy is still positive, statistically significant and economically significant. I conclude that the accrual anomaly cannot be explained by credit risk. All findings in this paper are robust as to whether credit risk is measured using Altman’s z-score or the Standard & Poor’s credit rating.

We document substantial management of reported and forecast earnings per share (EPS) for analyst-followed U.S. firms, with the extent of management increasing with share price. Managers smooth the volatility of reported EPS by using accruals to offset cash flow shocks. Smoother EPS is easier to forecast, resulting in smaller forecast errors. Managers also differentially guide forecasts to improve accuracy. Whereas unmanaged forecast errors are much larger for high-price firms, they are compressed to the point their magnitudes resemble those for low-price firms. Managers also guide analyst forecasts to generate patterns of forecast walkdowns that again vary with share price. Given the remarkable level of management implied by our results, we conduct additional robustness analyses. The strongest evidence is observed in stock price responses: investors recognize efforts to manage reported and forecast EPS and adjust accordingly. We highlight potential biases caused by researchers being unaware of managerial efforts and investor responses, and offer ways to mitigate those biases. The online appendix is available at https://doi.org/10.1287/mnsc.2017.2832 . This paper was accepted by Shivaram Rajgopal, accounting.

We examine the effect on earnings forecast accuracy when financial analysts add or drop coverage. We find that the accuracy of analysts’ first forecast for a firm (newly added coverage) is lower relative to their peers. In addition, the accuracy of their last forecast (just before coverage is dropped) is lower relative to their peers. Further analysis shows that our results are not driven by the rookie analysts (analysts with less than 1-year experience) or retiring analysts (i.e., analysts who are within their final year before retiring).

This paper finds the weekend effect to be a remarkably robust anomaly and refutes the widespread belief that the weekend effect is due to data-mining or a consequence of some unusual/rare events. Out-of-sample analysis finds both the mean and median return on Monday is lower than that on Friday in nearly all years. It also reconciles and explains how some prior studies reached such an erroneous conclusion.

We document a counter‐intuitive finding regarding analyst forecasts of quarterly earnings per share (EPS): magnitudes of deviations from benchmarks—individual forecasts versus consensus (dispersion) and consensus versus actual (forecast error)—do not vary with scale. Seasonally differenced EPS, or time‐series forecast error, also exhibits substantial lack of variation with scale, but only for firms followed by analysts. This lack of variation with scale is not observed for analyst and time‐series forecasts for (a) EPS for some countries, (b) sales and cash flows for all countries, and (c) stock splits. We propose and investigate different explanations for these puzzling regularities that have important implications for practice and research. We believe the cause is managerial smoothing of EPS designed to reduce across‐firm variation in EPS volatility.