Suchergebnisse

In recent years efforts to account for changing patterns of alliance relationships have incorporated a growing number of insights and propositions derived from the analysis of "global interdependence." These efforts have been complicated, on the one hand, by a lack of precision in providing operational definitions of the independent variables and, at a more fundamental level, by considerable ambiguity with respect to those types of changes which may be best explained in terms of these variables.An analysis of Washington-Bonn relations reveals a reasonably coherent pattern of changes with regard to four specific aspects of the relationship: 1) the nature of the alliance agenda; 2) the structure of the relationship (i.e. the characteristic distribution of influence between alliance partners); 3) the operative procedural norms which regulate bilateral bargaining and negotiation; and 4) the institutional arrangements which have evolved for the coordination of alliance policy.

The "guns versus butter" literature focuses mainly on the industrialized capitalist democracies and has reported little success in empirically substantiating the existence of durable trade-off relationships between military spending and particular civilian programs during peacetime. This article identifies structural features of Soviet-type politicoeconomic systems that make them more likely to display such durable trade-offs and then demonstrates the point by multiple regression time-series estimation of the effects of changing rates of growth in Soviet military spending on 13 major civilian programs during the period 1951-1980, controlling for demographic, economic growth, and leadership succession effects. Despite considerable noise in the Soviet military spending data, substantial and robust trade-off phenomena are demonstrated only for housing construction and the production of durable consumer goods, whereas ideologically favored programs for productive investment and social spending (except perhaps old-age pensions) appear to have been much less affected by marginal changes in military spending.

Trade offs between military spending and spending on consumer durables, housing construction, social security, and soft goods. Structural features of Soviet-type politicoeconomic systems that make them more likely to display such durable trade offs; demonstrates the point by multiple regression time-series estimation of the effects of changing rates of growth in Soviet military spending on 13 major civilian programs.

AbstractPlanning for community resilience through public infrastructure projects often engenders problems associated with social dilemmas, but little work has been done to understand how individuals respond when presented with opportunities to invest in such developments. Using statistical learning techniques trained on the results of a web‐based common pool resource game, we analyze participants' decisions to invest in hypothetical public infrastructure projects that bolster their community's resilience to disasters. Given participants' dispositions and in‐game circumstances, Bayesian additive regression tree (BART) models are able to accurately predict deviations from players' decisions that would reasonably lead to Pareto‐efficient outcomes for their communities. Participants tend to overcontribute relative to these Pareto‐efficient strategies, indicating general risk aversion that is analogous to individuals purchasing disaster insurance even though it exceeds expected actuarial costs. However, higher trait Openness scores reflect an individual's tendency to follow a risk‐neutral strategy, and fewer available resources predict lower perceived utilities derived from the infrastructure developments. In addition, several input variables have nonlinear effects on decisions, suggesting that it may be warranted to use more sophisticated statistical learning methods to reexamine results from previous studies that assume linear relationships between individuals' dispositions and responses in applications of game theory or decision theory.

AbstractThe concepts of vulnerability and resilience help explain why natural hazards of similar type and magnitude can have disparate impacts on varying communities. Numerous frameworks have been developed to measure these concepts, but a clear and consistent method of comparing them is lacking. Here, we develop a data‐driven approach for reconciling a popular class of frameworks known as vulnerability and resilience indices. In particular, we conduct an exploratory factor analysis on a comprehensive set of variables from established indices measuring community vulnerability and resilience at the U.S. county level. The resulting factor model suggests that 50 of the 130 analyzed variables effectively load onto five dimensions: wealth, poverty, agencies per capita, elderly populations, and non–English‐speaking populations. Additionally, the factor structure establishes an objective and intuitive schema for relating the constituent elements of vulnerability and resilience indices, in turn affording researchers a flexible yet robust baseline for validating and expanding upon current approaches.