With the expansion of the federal government in the twentieth century, presidents became increasingly involved in managing the administrative state and its myriad regulatory agencies. Acting as "manager in chief," presidents devised two methods for gaining control over the agencies: politicization and centralization (Moe 1985b). Politicization involves appointing politically loyal subordinates not only at the top of the agencies but deep within them as well (Lewis 2008). Centralization manifests itself in an expansive role for the White House, attempting to manage or even micromanage the agencies directly. Obvious examples include centralized budgeting (Tomkin 1998), the formulation of a unified legislative program (Rudalevige 2002), and direct command via executive orders (Howell 2003; Mayer 2002).
One of the newest and most dramatic examples of centralization is the Office of Information and Regulatory Affairs (OIRA), located in the Executive Office of the President. OIRA is empowered to review nearly all rules proposed by federal agencies, excepting the independent regulatory commissions (IRCs). OIRA gives presidents a powerful vantage point from which to scrutinize, delay, and effectively remand and revise proposed rules. Sometimes dubbed "the most powerful office no one has heard of," OIRA is the point of the president's spear in his battle to control the content of federal regulations (Anonymous 2011). In many respects, the agency is an exemplar of the centralizing tendencies of the American presidency. In considering the operation and impact of OIRA, three questions stand out:
* Targeting: Which regulations does OIRA target for regulatory review?
* Revisions: What happens to the regulations after OIRA selects them for review?
* Incentives: What incentives does OIRA's auditing regime create in the agencies? In particular, does a hostile OIRA damp down the production of significant rules in the agencies--in other words, does OIRA's auditing create a chilling effect on the production of regulations? Does regulatory auditing induce agencies to engage in "OIRA avoidance," substituting several smaller regulations for a single big-ticket one in order to escape regulatory review? Does it slow down the production of regulations, an "ossifying" effect?
In this article, we present an early empirical analysis of the "incentives" question. To do so, we examine biannual data on the production of regulations in 18 executive agencies and seven independent regulatory commissions, between 1995 and 2010, as well as newly collected data on OIRA auditing rates by agency. We examine economically significant regulations (ones with a price ticket over $100 million) as well as other regulations.
Uncovering OIRA-induced effects on the production of important regulations is not easy. Part of the problem is that one needs a baseline for comparison: what would have happened, absent OIRA targeting of the agency? The baseline we consider comes from estimating "regulation production functions" for agencies, relating the volume of regulations produced to theoretically plausible covariates. The state-of-the-art in empirically estimating regulation production function remains rudimentary (we review relevant studies below). Nonetheless, we estimate empirical regulatory production functions for 25 executive and independent regulatory agencies, focusing on the volume of economically significant regulations.
But another knotty problem is that presidential direction of OIRA and presidential control of agency appointments are completely collinear. In other words, "centralization" and "politicization" (in Moe's terminology) go hand in hand. Hence, even if one finds a change across Democratic and Republic administrations in the volume of regulations issued by an agency, it is difficult to know whether the effect was due to OIRA's auditing regime or to the president's appointments in the agency. …