Fiscal decentralisation and IDC economic growth: An empirical investigation

The impact of fiscal decentralisation on LDC economic growth and development is widely debated in the development literature. Notwithstanding this, there has been little attempt to test systematically this relationship. Accordingly, in this note we present an empirical examination of the relationship between the level of fiscal decentralisation and economic growth rates across a sample of twenty‐three LDCsfrom 1974 to 1991. We fail to find, however, any strong, systematic relationship between the two among our sample of LDCs.     

Microcycles of Violence: Evidence from Terrorist Attacks by ETA and the FMLN

Recent research has demonstrated that individual crimes elevate the risk for subsequent crimes nearby, a phenomenon termed “near-repeats.” Yet these assessments only reveal global patterns of event interdependence, ignoring the possibility that individual events may be part of localized bursts of activity, or microcycles. In this study, we propose a method for identifying and analyzing criminal microcycles; groups of events that are proximate to each other in both space and time. We use the Global Terrorism Database (GTD) to analyze over 4,000 terrorist attacks attributed to the FMLN in El Salvador and the ETA in Spain; two terrorist organizations that were both extremely active and violent but differed greatly in terms of history, grievances and motives. Based on the definition developed, we find strong support for the conclusion that many of the terrorist attacks attributed to these two distinctive groups were part of violent microcycles and that the spatio-temporal attack patterns of these two groups exhibit substantial similarities. Our logistic regression analysis shows that for both ETA and the FMLN, compared to other tactics used by terrorists, bombings and non-lethal attacks are more likely to be part of microcycles and that compared to attacks which occur elsewhere, attacks aimed at national or provincial capitals or areas of specific strategic interest to the terrorist organization are more likely to be part of microcycles. Finally, for the FMLN only, compared to other attacks, those on military and government targets were more likely part of microcycles. We argue that these methods could be useful more generally for understanding the situational and temporal distribution of crime.     

Judicial Representations of Scientific Evidence

A critical examination of the way that courts handle scientific expert evidence reveals inconsistency in strategies for validating and legitimating the weight attached to it. This inconsistency is rooted in perceptions of the nature of scientific knowledge. The essay considers the implications of this analysis for the Woolf reforms of civil procedure with respect to expert evidence.     

Field Experiments on Eyewitness Identification: Towards a Better Understanding of Pitfalls and Prospects

The Illinois pilot program on lineup procedures has helped sharpen the focus on the types of controls that are needed in eyewitness field experiments and the limits that exist for interpreting outcome measures (rates of suspect and filler identifications). A widely-known limitation of field experiments is that, unlike simulated crime experiments, the guilt or innocence of the suspects is not easily known independently of the behavior of the eyewitnesses. Less well appreciated is that the rate of identification of lineup fillers, although clearly errors, can be a misleading measure if the filler identification rate is used to assess which of two or more lineup procedures is the better procedure. Several examples are used to illustrate that there are clearly improper procedures that would yield fewer identifications of fillers than would their proper counterparts. For example, biased lineup structure (e.g., using poorly matched fillers) as well as suggestive lineup procedures (that can result from non-blind administration of lineups) would reduce filler identification errors compared to unbiased and non-suggestive procedures. Hence, under many circumstances filler identification rates can be misleading indicators of preferred methods. Comparisons of lineup procedures in future field experiments will not be easily accepted in the absence of double-blind administration methods in all conditions plus true random assignment to conditions.