Jim Tozzi has for several decades been among the most dedicated and perceptive commentators on the White House Office of Information and Regulatory Affairs (OIRA). Indeed, Tozzi is arguably better suited than anyone to be the keeper of the OIRA flame: from his position in charge of the Office of Regulatory and Information Policy, OIRA’s predecessor organization, Tozzi contributed to the development of Executive Order (EO) 12291 and then stepped in as OIRA’s first Deputy Administrator. Since that time, he has been a steadfast defender of OIRA’s critical role in bringing rigorous analysis, quality control, and policy discipline to Executive Branch regulation. I share Tozzi’s view, and indeed that of all former OIRA officials I have spoken with, of OIRA’s importance to the regulatory state. Preservation of the strength and independence of the office is to me the principal criterion by which policy proposals that effect OIRA should be judged.
In his new article, “Office of Information and Regulatory Affairs: Past, Present, and Future,” Tozzi (Reference Tozzi2019) intertwines an enlightening history of the office with three policy proposals. First, Tozzi argues that there should be a class of EOs designated as “iconic,” to which subsequent administrations should give substantial deference. He argues persuasively that EO 12866, President Clinton’s successor to 12291 that to this day governs much of OIRA’s activity, should be so designated. Second, Tozzi proposes a new EO that would give OIRA authority to review, in advance of presidential signing, any EO that has regulatory consequences. Tozzi’s third proposal is that OIRA oversee implementation of an EO that imposes a “regulatory budget” to constrain the size of the administrative state. I agree with some of Tozzi’s ideas but have questions about aspects of his proposals that might, unintendedly, work against the criterion of preserving OIRA’s strength and independence in the future.
As a starting point, it is important to recognize that, regardless of what EOs might govern and empower OIRA, the institutional positioning of the office within an administration will be as important, if not more so, to OIRA’s effectiveness. An OIRA with substantial powers on paper will do little to improve regulatory outcomes if the office can be ignored within the Executive Office of the President or bypassed by Departments and agencies. EO 12866 might indeed have a history and purpose that make it “iconic” but, unless the OIRA Administrator and the mission of the office have strong backing from the White House, political expediency and agency ambitions will erode the regulatory quality control that OIRA is set up to exercise. For EO 12866 to be accorded particular deference and importance from administration to administration, it is important that the order be closely associated with an effective process that produces better regulation. For OIRA to achieve that goal, I think three things are essential: (i) a reporting relationship for the OIRA Administrator that includes direct communication with the Chief of Staff’s office, (ii) a clear mandate from the Chief of Staff to executive departments and agencies that elevation of disagreements with OIRA to the Office of Management and Budget Director or senior West Wing officials should be a last resort, and (iii) inclusion of OIRA in any process or meeting to resolve any disagreements a Secretary or senior agency official does elevate.
It is important to clarify that the foregoing comment about institutional position does not mean that OIRA must always win or have the last word. In my experience as OIRA Administrator from 2013 to 2017, the three institutional features listed above ensured serious engagement with OIRA by executive branch departments and agencies, established a presumption against attempts to bypass OIRA in the event of disagreements, and made sure that OIRA’s views were fully considered when serious disagreements arose. The result was that most issues got resolved between OIRA and the regulatory agencies and, for those that did not, the discussion among senior officials to resolve disputes were anchored in the principles established by EO 12866 and related EOs, as well relevant OMB circulars. Of the issues that did elevate during my tenure, OIRA won some and lost some. However, the losses always occurred with full airing of OIRA’s concerns and a good faith effort through the process to adhere to core principles of careful analysis, use of best available science and economics, and ensuring that the regulation’s benefits justified its costs.
Coming back to Tozzi’s proposal for elevating the status of EO 12866, I agree that the order is deserving of particular deference and preservation by future administrations. However, I think the likelihood of achieving that deference will be greater if there is also some mechanism for ensuring that OIRA’s role cannot be ignored, compromised, or bypassed; only in that case can EO 12866 have its greatest impact. I would therefore recommend supplementing Tozzi’s idea with an additional proposal: an EO establishing the process of engagement and dispute resolution with OIRA that will ensure OIRA’s strong and central role in regulatory quality control even when agencies challenge the office’s recommendations or conclusions. An alternative way to achieve the objectives of such an EO, as well as Tozzi’s goals of preserving EO 12866, is for Congress to further codify OIRA’s role and underlying EOs. Congress has in recent years considered such legislation, and any codification of EO 12866 and other relevant documents has pros and cons. Although a discussion of those pros and cons is beyond the scope of this commentary, the topic is an important one and it would be interesting to hear Tozzi’s view on the question of codification.
Tozzi’s second proposal is that OIRA’s role be expanded to reviewing any proposed EO, before the President issues it, which relates to regulation. As a practical matter, there is already some circulation and review of draft EOs to relevant stakeholders in the administration. However, I agree with Tozzi that it would be beneficial for OIRA to have a more formal and prominent role in EOs that relate in any way to rulemaking by executive branch agencies. It seems unlikely that the White House issued any of the prominent recent orders on regulation, like EO 13771 and EO 13777, without substantial input from OIRA. To the extent OIRA can have a more central role not just in orders that affect how OIRA itself operates, but in orders that have regulatory implications more broadly, that strikes me as a constructive development. OIRA is in a particularly good position to identify any inconsistencies with prior EOs or sound regulatory practices and to identify beneficial changes that an EO could introduce into the regulatory process.
My principal reservations relate Tozzi’s third proposal, which is that OIRA have the mandate, and greater authority, to enforce a “regulatory budget.” I do not disagree with the importance of regulatory discipline. Rules can easily accumulate, conflict with each other, and saddle particular industries with unnecessary burdens. To prevent such results, it is important to have a regulatory process that: (i) produces individual rules that efficiently achieve benefits that justify their costs; (ii) makes sure those rules do not conflict with existing rules; and (iii) reviews the stock of rules to ensure that, even if benefits outweigh costs on a rule-by-rule basis, rules do not accumulate on particular stakeholders or sectors in a way that flips that calculus. There are a number of ways to achieve these objectives. One option, which Tozzi proposes, is to have a regulatory budget that places a limit on rulemaking activity, for example by capping the total regulatory costs or the number of rules imposed by a given agency. An alternative but not mutually exclusive approach is to engage in ongoing retrospective review of rules already on the books. A third option, also not mutually exclusive of the other two, is to require agencies to account not only for the incremental costs and benefits of a new rule, but also for whether the cumulative costs of regulations on a particular set of actors remains justified by the cumulative benefits. To Tozzi’s credit, he devotes substantial discussion in his paper to the complexities of a regulatory budget and the alternative institutional approaches to achieving a limit on regulatory expenditures. In the end, however, he advocates an administrative solution through an EO that places OIRA in the central role of implementing and enforcing an overall cap on regulatory costs.
As the third part of Tozzi’s article suggests, all of the foregoing approaches and regulatory process objectives are subject to debate. They all have pros and cons and entail distinct tradeoffs. One important dimension in which the alternative approaches differ is in how well OIRA can implement them, and here is where I have a concern with Tozzi’s reliance on the office, and on EO authority, to enforce a regulatory budget. Enforcement of regulatory budgets ultimately requires tradeoffs on the margin among alternative social benefits. Even if one accepts the premise that a regulatory budget is a good idea (a premise that warrants discussion), it is an extremely difficult exercise to determine which rules should be allowed and which should be removed to remain within the budget constraint, however defined.
A proposed new regulation might have the same costs but much different benefits than some existing regulation. If the new rule’s incremental costs would push cumulative regulatory expenditures above the budget constraint, what should happen? Should the new rule be blocked or should an existing rule be repealed? Optimally, the rule with the lower benefits should be discarded. However, to the extent the benefits of these respective hypothetical rules accrue to different stakeholders, or are difficult to quantify, any decision about which rule to allow and which to block will entail an important policy—and political—decision. Although OIRA might, in principle, be in a position to assess when cumulative regulatory costs have exceeded some level, I think the office is less well positioned to enforce tradeoffs among rules. To the extent that OIRA’s principal mechanism for enforcing a regulatory budget constraint would be to refuse to accept a budget-busting rule for review, OIRA would effectively be delaying the benefits of the new rule in favor of the benefits (even if lower) of incumbent rules. To be sure, it might be possible for OIRA to put a particular agency on notice that its regulatory costs were running up against a constraint, so that the agency would know that it should simultaneously propose a rule to repeal to offset the new rule being proposed. But if either the benefits or the costs of the rule to be repealed accrue to different stakeholders than those affected by the new rule, OIRA could be seen to be picking winners and losers from the regulatory process. Such an outcome could create the perception that OIRA is making substantive policy tradeoffs for which it can claim neither expertise nor political or statutory legitimacy. The inevitable criticisms from stakeholders and regulatory agencies, and the increased politicization of OIRA that would likely result, could do more to weaken than to strengthen the office.
I therefore think that, even were Congress or the President to impose a regulatory budget of some kind, OIRA’s role should be limited to publicly reporting on when that budgetary constraint is being approached or exceeded. Putting OIRA in the further role of enforcing the constraint is a much riskier proposition; such an enforcement role is practically and politically better left elsewhere, most likely with Congress.
That said, I think OIRA nonetheless has a very important role in enforcing regulatory discipline. I return to two other approaches mentioned above: requiring agencies to engage in retrospective review of regulations on the books, and requiring agencies to identify not just incremental costs and benefits of rules, but cumulative burdens and benefits on affected stakeholders. Both of these approaches have practical and conceptual challenges. Identifying those rules on the books whose repeal would yield net benefits is analytically difficult, and the constituency for repeal might be quite weak. Calculating cumulative costs on a particular economic segment might not be straightforward, as some costs should arguable not count toward a cap (e.g., those a regulated firm incurs to avoid causing harmful externalities) while other costs more arguably should count toward a cap (costs of correcting for harms the regulated firm did not directly cause). But these difficulties aside, both retrospective review and requiring agencies to account for accumulated effects on parties affected by a new rule are more consistent with OIRA’s historic role and expertise, and is more reliant on the kinds of analysis OIRA typically conducts. Whereas Tozzi puts significant emphasis going forward on EO 13771, the 2017 “one in, two out” order, I would instead put more emphasis implementing EO 13777, the 2017 order further mandating and institutionalizing retrospective review, modification, and repeal of existing regulations.
OIRA had in fact made significant progress toward institutionalizing retrospective review of existing regulations even before EO 13771. President Obama made retrospective review a priority, and beginning in 2011, all executive branch Departments filed regular reports to OIRA on their plans for rules to review and on their progress in completing reviews of rules listed in previous plans. By the middle of 2016, OIRA was able to report that the resulting repeal and modification of rules to that point would already generate cost savings of $37 billion over 5 years.Footnote
1
That result came from significant engagement by agencies and departments in establishing internal processes and assigning staff accountable for achieving beneficial outcomes from regulatory review. Those steps were, to be sure, just a beginning, and EO 13777 is well designed to implement yet further progress in retrospective review going forward.
Although one can debate alternative ways for OIRA to achieve good regulatory outcomes—and reasonable minds can certainly differ on what outcomes count as “good”—OIRA plays and in the future should continue to play a critical role in bringing rigor and quality to federal rulemaking. To that end, Jim Tozzi has written a thought provoking article on how to preserve, strengthen, and extend OIRA’s role in the administration of the regulatory state. I agree in all respects with Tozzi’s view of OIRA’s importance and in his objectives of ensuring the office remains independent and effective through future administrations. Although my views on the role OIRA should play in enforcing any regulatory budget constraints differ from Tozzi’s (and putting to one side altogether the underlying conceptual and practical issues related to “regulatory budgets”), I think he has made important and thoughtful proposals that will no doubt be the subject of serious discussion and consideration going forward.
You have
Access
Jim Tozzi has for several decades been among the most dedicated and perceptive commentators on the White House Office of Information and Regulatory Affairs (OIRA). Indeed, Tozzi is arguably better suited than anyone to be the keeper of the OIRA flame: from his position in charge of the Office of Regulatory and Information Policy, OIRA’s predecessor organization, Tozzi contributed to the development of Executive Order (EO) 12291 and then stepped in as OIRA’s first Deputy Administrator. Since that time, he has been a steadfast defender of OIRA’s critical role in bringing rigorous analysis, quality control, and policy discipline to Executive Branch regulation. I share Tozzi’s view, and indeed that of all former OIRA officials I have spoken with, of OIRA’s importance to the regulatory state. Preservation of the strength and independence of the office is to me the principal criterion by which policy proposals that effect OIRA should be judged.
In his new article, “Office of Information and Regulatory Affairs: Past, Present, and Future,” Tozzi (Reference Tozzi2019) intertwines an enlightening history of the office with three policy proposals. First, Tozzi argues that there should be a class of EOs designated as “iconic,” to which subsequent administrations should give substantial deference. He argues persuasively that EO 12866, President Clinton’s successor to 12291 that to this day governs much of OIRA’s activity, should be so designated. Second, Tozzi proposes a new EO that would give OIRA authority to review, in advance of presidential signing, any EO that has regulatory consequences. Tozzi’s third proposal is that OIRA oversee implementation of an EO that imposes a “regulatory budget” to constrain the size of the administrative state. I agree with some of Tozzi’s ideas but have questions about aspects of his proposals that might, unintendedly, work against the criterion of preserving OIRA’s strength and independence in the future.
As a starting point, it is important to recognize that, regardless of what EOs might govern and empower OIRA, the institutional positioning of the office within an administration will be as important, if not more so, to OIRA’s effectiveness. An OIRA with substantial powers on paper will do little to improve regulatory outcomes if the office can be ignored within the Executive Office of the President or bypassed by Departments and agencies. EO 12866 might indeed have a history and purpose that make it “iconic” but, unless the OIRA Administrator and the mission of the office have strong backing from the White House, political expediency and agency ambitions will erode the regulatory quality control that OIRA is set up to exercise. For EO 12866 to be accorded particular deference and importance from administration to administration, it is important that the order be closely associated with an effective process that produces better regulation. For OIRA to achieve that goal, I think three things are essential: (i) a reporting relationship for the OIRA Administrator that includes direct communication with the Chief of Staff’s office, (ii) a clear mandate from the Chief of Staff to executive departments and agencies that elevation of disagreements with OIRA to the Office of Management and Budget Director or senior West Wing officials should be a last resort, and (iii) inclusion of OIRA in any process or meeting to resolve any disagreements a Secretary or senior agency official does elevate.
It is important to clarify that the foregoing comment about institutional position does not mean that OIRA must always win or have the last word. In my experience as OIRA Administrator from 2013 to 2017, the three institutional features listed above ensured serious engagement with OIRA by executive branch departments and agencies, established a presumption against attempts to bypass OIRA in the event of disagreements, and made sure that OIRA’s views were fully considered when serious disagreements arose. The result was that most issues got resolved between OIRA and the regulatory agencies and, for those that did not, the discussion among senior officials to resolve disputes were anchored in the principles established by EO 12866 and related EOs, as well relevant OMB circulars. Of the issues that did elevate during my tenure, OIRA won some and lost some. However, the losses always occurred with full airing of OIRA’s concerns and a good faith effort through the process to adhere to core principles of careful analysis, use of best available science and economics, and ensuring that the regulation’s benefits justified its costs.
Coming back to Tozzi’s proposal for elevating the status of EO 12866, I agree that the order is deserving of particular deference and preservation by future administrations. However, I think the likelihood of achieving that deference will be greater if there is also some mechanism for ensuring that OIRA’s role cannot be ignored, compromised, or bypassed; only in that case can EO 12866 have its greatest impact. I would therefore recommend supplementing Tozzi’s idea with an additional proposal: an EO establishing the process of engagement and dispute resolution with OIRA that will ensure OIRA’s strong and central role in regulatory quality control even when agencies challenge the office’s recommendations or conclusions. An alternative way to achieve the objectives of such an EO, as well as Tozzi’s goals of preserving EO 12866, is for Congress to further codify OIRA’s role and underlying EOs. Congress has in recent years considered such legislation, and any codification of EO 12866 and other relevant documents has pros and cons. Although a discussion of those pros and cons is beyond the scope of this commentary, the topic is an important one and it would be interesting to hear Tozzi’s view on the question of codification.
Tozzi’s second proposal is that OIRA’s role be expanded to reviewing any proposed EO, before the President issues it, which relates to regulation. As a practical matter, there is already some circulation and review of draft EOs to relevant stakeholders in the administration. However, I agree with Tozzi that it would be beneficial for OIRA to have a more formal and prominent role in EOs that relate in any way to rulemaking by executive branch agencies. It seems unlikely that the White House issued any of the prominent recent orders on regulation, like EO 13771 and EO 13777, without substantial input from OIRA. To the extent OIRA can have a more central role not just in orders that affect how OIRA itself operates, but in orders that have regulatory implications more broadly, that strikes me as a constructive development. OIRA is in a particularly good position to identify any inconsistencies with prior EOs or sound regulatory practices and to identify beneficial changes that an EO could introduce into the regulatory process.
My principal reservations relate Tozzi’s third proposal, which is that OIRA have the mandate, and greater authority, to enforce a “regulatory budget.” I do not disagree with the importance of regulatory discipline. Rules can easily accumulate, conflict with each other, and saddle particular industries with unnecessary burdens. To prevent such results, it is important to have a regulatory process that: (i) produces individual rules that efficiently achieve benefits that justify their costs; (ii) makes sure those rules do not conflict with existing rules; and (iii) reviews the stock of rules to ensure that, even if benefits outweigh costs on a rule-by-rule basis, rules do not accumulate on particular stakeholders or sectors in a way that flips that calculus. There are a number of ways to achieve these objectives. One option, which Tozzi proposes, is to have a regulatory budget that places a limit on rulemaking activity, for example by capping the total regulatory costs or the number of rules imposed by a given agency. An alternative but not mutually exclusive approach is to engage in ongoing retrospective review of rules already on the books. A third option, also not mutually exclusive of the other two, is to require agencies to account not only for the incremental costs and benefits of a new rule, but also for whether the cumulative costs of regulations on a particular set of actors remains justified by the cumulative benefits. To Tozzi’s credit, he devotes substantial discussion in his paper to the complexities of a regulatory budget and the alternative institutional approaches to achieving a limit on regulatory expenditures. In the end, however, he advocates an administrative solution through an EO that places OIRA in the central role of implementing and enforcing an overall cap on regulatory costs.
As the third part of Tozzi’s article suggests, all of the foregoing approaches and regulatory process objectives are subject to debate. They all have pros and cons and entail distinct tradeoffs. One important dimension in which the alternative approaches differ is in how well OIRA can implement them, and here is where I have a concern with Tozzi’s reliance on the office, and on EO authority, to enforce a regulatory budget. Enforcement of regulatory budgets ultimately requires tradeoffs on the margin among alternative social benefits. Even if one accepts the premise that a regulatory budget is a good idea (a premise that warrants discussion), it is an extremely difficult exercise to determine which rules should be allowed and which should be removed to remain within the budget constraint, however defined.
A proposed new regulation might have the same costs but much different benefits than some existing regulation. If the new rule’s incremental costs would push cumulative regulatory expenditures above the budget constraint, what should happen? Should the new rule be blocked or should an existing rule be repealed? Optimally, the rule with the lower benefits should be discarded. However, to the extent the benefits of these respective hypothetical rules accrue to different stakeholders, or are difficult to quantify, any decision about which rule to allow and which to block will entail an important policy—and political—decision. Although OIRA might, in principle, be in a position to assess when cumulative regulatory costs have exceeded some level, I think the office is less well positioned to enforce tradeoffs among rules. To the extent that OIRA’s principal mechanism for enforcing a regulatory budget constraint would be to refuse to accept a budget-busting rule for review, OIRA would effectively be delaying the benefits of the new rule in favor of the benefits (even if lower) of incumbent rules. To be sure, it might be possible for OIRA to put a particular agency on notice that its regulatory costs were running up against a constraint, so that the agency would know that it should simultaneously propose a rule to repeal to offset the new rule being proposed. But if either the benefits or the costs of the rule to be repealed accrue to different stakeholders than those affected by the new rule, OIRA could be seen to be picking winners and losers from the regulatory process. Such an outcome could create the perception that OIRA is making substantive policy tradeoffs for which it can claim neither expertise nor political or statutory legitimacy. The inevitable criticisms from stakeholders and regulatory agencies, and the increased politicization of OIRA that would likely result, could do more to weaken than to strengthen the office.
I therefore think that, even were Congress or the President to impose a regulatory budget of some kind, OIRA’s role should be limited to publicly reporting on when that budgetary constraint is being approached or exceeded. Putting OIRA in the further role of enforcing the constraint is a much riskier proposition; such an enforcement role is practically and politically better left elsewhere, most likely with Congress.
That said, I think OIRA nonetheless has a very important role in enforcing regulatory discipline. I return to two other approaches mentioned above: requiring agencies to engage in retrospective review of regulations on the books, and requiring agencies to identify not just incremental costs and benefits of rules, but cumulative burdens and benefits on affected stakeholders. Both of these approaches have practical and conceptual challenges. Identifying those rules on the books whose repeal would yield net benefits is analytically difficult, and the constituency for repeal might be quite weak. Calculating cumulative costs on a particular economic segment might not be straightforward, as some costs should arguable not count toward a cap (e.g., those a regulated firm incurs to avoid causing harmful externalities) while other costs more arguably should count toward a cap (costs of correcting for harms the regulated firm did not directly cause). But these difficulties aside, both retrospective review and requiring agencies to account for accumulated effects on parties affected by a new rule are more consistent with OIRA’s historic role and expertise, and is more reliant on the kinds of analysis OIRA typically conducts. Whereas Tozzi puts significant emphasis going forward on EO 13771, the 2017 “one in, two out” order, I would instead put more emphasis implementing EO 13777, the 2017 order further mandating and institutionalizing retrospective review, modification, and repeal of existing regulations.
OIRA had in fact made significant progress toward institutionalizing retrospective review of existing regulations even before EO 13771. President Obama made retrospective review a priority, and beginning in 2011, all executive branch Departments filed regular reports to OIRA on their plans for rules to review and on their progress in completing reviews of rules listed in previous plans. By the middle of 2016, OIRA was able to report that the resulting repeal and modification of rules to that point would already generate cost savings of $37 billion over 5 years.Footnote 1 That result came from significant engagement by agencies and departments in establishing internal processes and assigning staff accountable for achieving beneficial outcomes from regulatory review. Those steps were, to be sure, just a beginning, and EO 13777 is well designed to implement yet further progress in retrospective review going forward.
Although one can debate alternative ways for OIRA to achieve good regulatory outcomes—and reasonable minds can certainly differ on what outcomes count as “good”—OIRA plays and in the future should continue to play a critical role in bringing rigor and quality to federal rulemaking. To that end, Jim Tozzi has written a thought provoking article on how to preserve, strengthen, and extend OIRA’s role in the administration of the regulatory state. I agree in all respects with Tozzi’s view of OIRA’s importance and in his objectives of ensuring the office remains independent and effective through future administrations. Although my views on the role OIRA should play in enforcing any regulatory budget constraints differ from Tozzi’s (and putting to one side altogether the underlying conceptual and practical issues related to “regulatory budgets”), I think he has made important and thoughtful proposals that will no doubt be the subject of serious discussion and consideration going forward.