Abstract
This Article analyzes the affirmative and disciplinary duties imposed on tax lawyers that require them to make probability assessments about the merits of a client’s tax position or tax-favored transaction, and to reflect those estimates with numerical precision. It describes how the Treasury Department, Congress, and the American Bar Association (often in concert, occasionally at odds) forged this obligatory standard of care over the last three decades with the shared goal of facilitating accurate advice, accurate tax returns, and compliance with the law. The resulting regulatory standard of care for tax lawyers (which swept aside the old regime of self-regulation) monitors flawed methodological processes, while minimizing psychological biases and misaligned incentives that could distort professional judgment. In this way, the standard of care for tax lawyers—particularly its emphasis on improving accuracy and reducing errors by updating subjective beliefs with new, relevant information—reflects a branch of probabilistic decision theory known as Bayesian reasoning.