Indicators of State Failure: Phase 2

Abstract : As noted in the Canada First Defence Strategy and reiterated in the more recent US Quarterly Defence Review, instability and state failure in distant lands can directly affect our own security and that of our allies. Development of a predictive model has become both a topical issue and an increasingly important area of research in academic and policy communities. This is the second report documenting CAE's support to DRDC's continuing efforts to develop an Early Warning Model (EWM) of state instability. The conceptual framework for an EWM was developed in a previous project, though without a data set to validate assumptions and the general hypothesis. The focus of the current project was to collect and code events data and integrate it with structural data that will ultimately be used to calibrate and validate the conceptual model. A descriptive framework was established and incidences of failure identified using the methodology developed by Carleton University's Country Indicators for Foreign Policy (CIFP) project. Events data were then collected for the 24 months preceding these instances of state failure. Periods of relative stability for these states were also identified and events data collected for these periods. These events were distinguished as eroding or bolstering state Authority, Legitimacy or Capacity, including the severity of the challenge recorded. An inter-coder reliability test was conducted to confirm coding consistency. The results were compared with data available through Virtual Research Associates (VRA) , thus affording an opportunity to gauge the merits of human (versus machine) coding. The research effort concluded that human coding is more discriminating but also considerably more time consuming. An extensive data base has been developed and analysis commenced, which will continue beyond the submission of this report.