Flexibility or Stability? Analyzing Proposals to Reform the Separation of Powers

The social welfare effects of legislatures in presidential systems, such as the U.S. Congress, are frequently lamented. In response, there are proposals to reform the separation of powers system by giving presidents control of the legislative agenda and weakening rules such as the filibuster. We provide a game-theoretic analysis of the policy and social welfare consequences of a more executive-centric system. Integrating standard assumptions about legislative and executive incentives into a dynamic model of decision making with private investment, we show there are a variety of conditions under which stronger executives do not produce better outcomes. Moreover, we characterize how these conditions depend on factors such as the stability of the policymaking environment or investment fundamentals. Our findings are robust and consistent with empirical observations that U.S. policy outputs are not necessarily worse than those of nations with stronger executives, which more closely approximate prominent proposals by populist-oriented reformers.     

A Poisson model of incidents of international terrorism in the United States

Abstract

Various haphazard, and often uncoordinated, efforts have been devoted to discouraging and combatting terrorism. Unfortunately, there has been a noticeable lack of quantitative studies of the problem of terrorism. Nevertheless, it can be expected that the problem of international terrorism will eventually be a focus for operations researchers. A factor fundamental to any OR study is an understanding of the underlying generating process for incidents of terrorism. Based on incidents of terrorism from 1968 to 1974, this paper notes that the Poisson is a good model for the occurrence of incidents of international terrorism in the United States. Results of both chi‐square and Kolmogorov‐Smirnov tests are presented. Finally, an unusual result, inconsistent with popular beliefs, is noted.     

Direct certification in the national school lunch program expands access for children

Under the policy of direct certification for free school meals, school districts use information shared by state agencies about household eligibility for means‐tested programs in the state in order to determine the potential eligibility for free meals of students enrolled in the district. This information allows districts to automatically approve students in these eligible households for free meals without requiring the household to complete the application process. This paper examines the impacts of direct certification on students' likelihood of becoming certified for free meals, using data from a national survey of school food service directors as well as statelevel administrative data on program participation. A state‐level fixed effects model is estimated to account for the possibility of selection bias. The key finding is that direct certification leads to a statistically significant increase in the number of children getting free school meals. More generally, this finding highlights a promising approach for improving access to means‐tested programs without compromising program integrity. © 2008 by the Association for Public Policy Analysis and Management.     

Melatonin concentrations in the sudden infant death syndrome

To examine a possible relationship between pineal function and the sudden infant death syndrome (SIDS), samples of whole blood, ventricular cerebrospinal fluid (CSF) and/or vitreous humor (VH) were obtained at autopsy from 68 infants (45 male, 23 female) whose deaths were attributed to either SIDS (n = 32, 0.5-5.0 months of age; mean +/- S.E.M., 2.6 +/- 0.2 months) or other causes (non-SIDS, n = 36, 0.3-8.0 months of age 4.3 +/- 0.3 months). The melatonin concentrations were measured by radioimmunoassay. A significant correlation was observed for melatonin levels in different body fluids from the same individual. After adjusting for age differences, CSF melatonin levels were significantly lower among the SIDS infants (91 +/- 29 pmol/l; n = 32) than among those dying of other causes (180 +/- 27; n = 35, P less than 0.05). A similar, but non-significant trend was also noted in blood (97 +/- 23, n = 30 vs. 144 +/- 22 pmol/l, n = 33) and vitreous humor (68 +/- 21, n = 10 vs. 81 +/- 17 pmol/l, n = 15). These differences do not appear to be explainable in terms of the interval between death and autopsy, gender, premortem infection or therapeutic measures instituted prior to death. Diminished melatonin production may be characteristic of SIDS and could represent an impairment in the maturation of physiologic circadian organization.