Garciano and Posner (2005) make an example of the failures and ineffectiveness of the United States intelligence system in this examination of the tradeoffs intelligence organizations ordinarily face. They assess the organizational economics of an intelligence agency through the analysis of key factors that prevent or handicap efforts to combat terrorism (or any given mission, for that matter).
The authors chiefly examine the “herding” or groupthink phenomenon intelligence organizations will face at the analysis stage, its cause, and the sacrifices organizations must make to avoid it. The potential largest root of groupthink is the flawed set of incentives for agents, which will also result in an unsound sharing system; the disincentives for governmental agencies to share information far outweigh the incentives. Incentive problems can range from an agent’s desire to mimic the opinions of superiors, as the agent’s future career depends on evaluations from those superiors, to larger, more organizational problems stemming from its hierarchy.
Top management has little time to evaluate the plethora of information agents gather, so each piece of information must go through a hierarchical filter on its way from the bottom to the top. Agents and managers on lower levels want to make fewer mistakes (false positives), and so fewer false information pieces are accepted and passed along the chain; however, this is as the cost of failing to act on a good piece of information (false negative). A looser hierarchy would inevitably produce a wider and more diverse set of data, though not as reliable.
The authors cite the U.S.-led invasion of Iraq based on flawed intelligence as evidence of both the snowballing effect of groupthink and the failure of an organizati ...