As former President George W. Bush said once, “the tax code is a complicated mess . . . [and] a million pages long.” The length and complexity of the Internal Revenue Code (Code) is largely the result of the U.S. government's rule-based approach to curtail tax abuse. Taxpayers, aided by literalism, have long found and used language in the tax laws to avoid or minimize their tax obligations. Although complying with the language of the law, abusive tax transactions are nevertheless at odds with the law's spirit. Historically, the government, specifically Congress and the Department of the Treasury (Treasury), has responded by crafting new rules to stem the abuse. Using a rule-based approach to combat tax abuse had its advantages, including certainty in application. With rules, the substance of the law is generally known before an individual acts; hence, rules often work well in the tax arena. Nevertheless, the rule-based approach had its disadvantages. For one thing, it forced the government to be reactive rather than proactive, as innovative taxpayers found new ways to work around the language of each new rule. The government has never been able to craft enough targeted rules to stem all tax abuse. For this reason, both the Treasury and Congress decided to try something new: a standards-based approach. Doctrines meant to prevent abuse of the Code may be better designed as standards, which are more difficult than rules to circumvent.This article examines two standard-based approaches. First, in 1995, the Treasury promulgated a general, anti-abuse regulation applicable to all of subchapter K of the Code. Second, in 2010, Congress codified the economic substance doctrine. In both cases, these governmental entities adopted judicially crafted tax doctrines that required transactions to satisfy the tax code's language as well as its underlying purpose. However, Congress's codification and the Treasury's promulgation differed in significant ways. This article explains why Congress got it right, while the Treasury got it entirely wrong
