Formal and Empirical Analyses of Bureaucratic Delay: The Case of FDA Drug Review

官僚拖延的形式和实证分析：FDA 药品审查案例

• 批准号: 0334889
• 负责人: Daniel Carpenter
• 金额: $ 2.15万
• 依托单位: Harvard University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2002
• 资助国家: 美国
• 起止时间: 2002-09-01 至 2003-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0334889&HistoricalAwards=false
• 关键词: Formal; Empirical; Analyses; Bureaucratic; Delay

Why do bureaucrats delay? Why do regulatory choices made under identicaladministrative procedures exhibit highly varying decision times? Theinvestigator conducts a formal and empirical analysis of the duration ofbureaucratic decisions, with applications to drug approval times by theU.S. Food and Drug Administration (FDA). The speed with which drugs arereviewed is now one of the most controversial issues in federal regulation,and is arguably the most salient case where the duration, more than thecontent, of bureaucratic decision making is at issue. While the politicalcontrol of agencies has been well studied, relatively little is known aboutwhy agencies wait, and whether the duration of agency decisions is subjectto political control and from what sources.The project tackles this problem through the elaboration of ageneralizable formal model of the approval decision. The investigator modelsproduct approval as an optimal stopping decision by an uncertainregulator subject to political demand for the drug. The regulator seeks topreserve a reputation for consumer protection and sees drug approvals asirreversible (or reversible only at a significant reputation cost). Thepayoff of approval is also a function of the political "demand" for drugapprovals from consumers (patients) and producers (firms). The modelpredicts that even under risk-neutrality, quick approvals are highlyunlikely, not all drugs will be approved, and that larger firms receive"protection" (quicker approvals) even when firms are not politicallyorganized.The investigator tests this model using maximum likelihoodduration analyses of FDA drug review times. The formal model suggests thatthe proper hazard function for estimation is the inverse Gaussiandistribution, which has a non-monotonic hazard function. Drug approvaltimes will be modeled as a function of political preferences, budgetaryshifts, media coverage and congressional hearings on the diseases treatedby the drug, and characteristics of the submitting firms. Preliminaryresults on a partial sample of 67 drugs approved from 1985 to 1995 supportthe model's predictions, and also suggest that political "control" in drugapproval occurs less through "principals" (Congress and the presidents)than through organized interests (patients and firms) and the media.

官僚们为何拖延？ 为什么在相同的行政程序下做出的监管选择表现出高度不同的决策时间？ 调查人员进行了正式和实证分析的持续时间的官僚决定，与申请药物批准时间由美国。美国食品药品监督管理局（FDA）。 药品审查的速度现在是联邦监管中最具争议的问题之一，也是官僚决策的持续时间而不是内容最有争议的问题。虽然对行政机关的政治控制已经有了很好的研究，但对行政机关为什么要等待，行政机关的决策期限是否受到政治控制以及政治控制的来源却知之甚少。本项目通过对批准决策的可推广的正式模型的阐述来解决这一问题。 研究者将产品批准建模为不确定性监管机构受药物政治需求影响的最佳停止决策。 监管机构寻求保留消费者保护的声誉，并认为药品批准是不可逆的（或只有在付出重大声誉成本的情况下才能可逆）。 批准的回报也是消费者（患者）和生产者（公司）对药物批准的政治“需求”的函数。 该模型预测，即使在风险中性下，快速批准也是极不可能的，并非所有的药物都将获得批准，即使公司没有政治组织，大型公司也会得到“保护”（更快的批准）。研究人员使用FDA药物审查时间的最大可能性持续时间分析来测试该模型。 形式化模型表明，用于估计的风险函数是具有非单调风险函数的逆高斯分布。 药物批准时间将被建模为一个功能的政治偏好，自愿性转变，媒体报道和国会听证会上的疾病治疗的药物，和提交公司的特点。 对1985年至1995年批准的67种药物的部分样本的初步结果支持了模型的预测，并且还表明，药物批准中的政治“控制”较少通过“校长”（国会和总统），而不是通过有组织的利益（患者和公司）和媒体。