Jim Tozzi is an activist institutional economist. During his 19-year career in the federal civil service, he was a pertinacious institution builder, armed with a PhD in economics but never flaunting it. He gained a reputation, richly deserved in my experience, as a supreme bureaucratic tactician. But he applied his skills to antibureaucratic purposes. Incessantly, and occasionally at professional risk, he promoted and protected internal executive-branch procedures that used economic analysis, and measures of administrative effectiveness, against the incessant forces of political entropy, agency parochialism, and special-interest capture.
Tozzi’s purposes and methods are vividly on display in “Office of Information and Regulatory Affairs: Past, Present, and Future” (Reference Tozzi2019). He is impressed by the durability of White House review of agency regulations under an increasingly explicit cost-benefit (“maximum net benefits”) standard. The practice is wholly discretionary, with no statutory basis and no particular congressional or political constituency, yet it has been followed since the early 1970s by nine presidents of both parties and all points on the political spectrum. That is something of a puzzle, and also a caution. The practice has been indubitably beneficial and has come to play a central role in regulatory policy-making – but it draws the Office of Management and Budget (OMB) into an endless stream of highly charged political controversies, and it could be extinguished by the stroke of a presidential pen or by simple neglect and desuetude.
This leads Tozzi to recommend three measures for buttressing White House regulatory oversight. The first is that the National Archives officially recognize “Iconic Executive Orders,” beginning with Ronald Reagan’s Executive Order 12291 (1981). That order was certainly the decisive regulatory innovation, first establishing the cost-benefit analysis requirement and maximum net benefits standard as presidential policies that were to long endure. But, at the Archives, there would be considerable competition for iconic status, and regulatory efficiency would jostle with other executive purposes. Abraham Lincoln’s Executive Order 95 of 1863, the Emancipation Proclamation, would surely rate a gold star, as would George Washington’s Thanksgiving Day Proclamation of 1789 (unnumbered). Dwight Eisenhower’s school desegregation Executive Order 10730 of 1957, and Franklin Roosevelt’s banking holiday Proclamation 2039 of 1933, might get honorable mention. I hereby nominate E.O. 6102 (FDR’s 1933 confiscation of privately held gold coins), E.O. 10340 (Harry Truman’s 1952 seizure of the steel mills), and E.O. 10998 (John Kennedy’s 1962 authorization of public-employee collective bargaining) as the three rock-bottom worst. But I doubt that the National Archives – one of the least politicized agencies in Washington, standing tall and solid above the swamp – would want to enter this fray. It does have three established icons – the Declaration of Independence, the Constitution, and the Bill of Rights – and these days we should be content if it is able to maintain allegiance to them.
Tozzi’s second proposal is that JFK’s little known E.O. 11030 (1962), which provides that the drafting and vetting of executive orders be managed by OMB, should be amended to give OIRA (a division of OMB) a leading role in the process. This is an excellent idea. The issuance of executive orders has become rather ad hoc and chaotic in recent years, and increasingly controversial. The Department of Justice opines on the form and legality of draft orders, but that is inadequate. The things that presidents attempt to accomplish by executive order, often inspired by a passing contretemps or staff brainstorm, typically have a long and complicated history within the executive branch going well beyond matters of form and law, and also going well beyond the budget-oversight and related functions of OMB’s other divisions. Tozzi wants to fortify OIRA’s position within the executive branch, to help sustain its oversight of agency rulemaking, but his proposal makes sense on its own terms. Executive orders are mainly internal regulations – directions from the executive branch’s principal to his far-flung agents – and regulation is OIRA’s specialty.
Tozzi’s first two institution-building proposals are warm-ups for his third, much more ambitious one: that the cost-benefit standard for individual rules be augmented or supplanted by a regulatory budget capping the total incremental costs of all new rules. (Budgets would presumably be set on an annual basis, and could be applied at the level of programs, agencies, or the executive branch as a whole.) He helpfully characterizes the two procedures as alternative means of addressing the principal-agent problem in government. Missionary regulatory agencies – single-mindedly devoted to promoting environmental quality, energy efficiency, safe and healthful products and workplaces, financial stability and fairness, nondiscrimination, and a host of other worthy causes – are empowered to command substantial private expenditures on behalf of their missions. These expenditures may be on the order of a hundred million dollars for a single rule, and in the aggregate are much greater than the agencies’ own operating budgets, but they are subject to none of the disciplines of public finance (taxation, appropriation, budgeting) that apply to the government’s own operations and spending programs. Left to themselves, regulatory agencies will be inclined to order expenditures on their statutory missions that take insufficient account of competing claims and alternative uses of private resources; the agencies’ principal, be it the President, the Congress, or the general public, will be inclined to take a more capacious, balanced view.
In Tozzi’s formulation, cost-benefit analysis is an application of welfare economics and decision theory to the principal-agent problem, while regulatory budgeting is an application of institutional economics and optimal-delegation theory. The former were in vogue, in academic economics and government, in the 1970s when the White House review procedures were first established – but the latter are older (going back to a delicious passage in The Wealth of Nations) and better suited to the problem at hand, and have recently been enjoying a revival.
He points in particular to an important article by Yair Listokin, “Bounded Institutions.”Footnote
1
In “unbounded institutions,” agents make individual decisions according to rules or standards handed down by principals; in “bounded institutions,” principals set a quantitative cap on agents’ decisions. Each approach has advantages and disadvantages in different circumstances. The bounded approach tends to dominate where (a) agents decide among more numerous candidates for action (e.g., grant awards, regulatory interventions, student grades, legal prosecutions); (b) the range of quality among candidates is narrower; (c) principals have a better general idea of the number and average quality of candidates; and (b) agents are more biased, compared to the preferences of principals, in judging the quality of candidates.
Listokin’s analysis suggests important advantages of a regulatory budget (bounded) over a cost-benefit standard (unbounded), especially in the context of an agency such as the Environmental Protection Agency (his example) with a strong missionary bias and numerous candidates for pollution controls of varying degrees of strictness. Under a cost-benefit standard, EPA will systematically overvalue the benefits of its rules; every rule that meets the standard will be issued, resulting in excessive pollution-control investments from a government-wide (or society-wide) perspective. Under a regulatory budget, individual rules compete not with the standard but with each other – EPA, notwithstanding its benefits bias, will issue rules with the highest net benefits until it reaches its cost budget.
But there are shortcomings as well, or at least challenges. A regulatory budget relies on estimates of private costs, which are much spongier and more open to manipulation than the hard dollar sums of a spending budget (but this is a problem for the cost-benefit standard too).Footnote
2
The principal – the regulatory budget-setter – may have a poor idea of the distribution of potential regulatory strategies and outcomes, and so may set a cost cap that is too high or too low to achieve the benefits it desires (so OIRA may be a better budget-setter than Congress, because it is more knowledgeable about regulatory strategies and outcomes – although Congress might acquire a serviceable grasp of them through a regular budget exercise responding to itemized proposals from the executive).
Tozzi notes that the idea of a regulatory budget was in circulation in academic and government circles in the late 1970s when the cost-benefit standard was adopted instead. Now that institutional economics is catching up with welfare economics, he thinks the time may be ripe for moving to a regulatory budget. The Trump Administration has begun to do so through E.O. 13771 (January 30, 2017) and subsequent OIRA directives. These have required agencies to rescind at least two existing rules for every new rule they issue and to hold the net costs of their actions to an annual cost budget – and have set budget caps of zero or negative incremental costs. Tozzi would build on these steps, and the initial experience with them, by issuing a proposed revised executive order for public notice and comment, and by encouraging new optimal-delegation research building on the existing literature. The notice-and-comment process, and accompanying public hearings, would, like similar agency procedures, have the not-incidental purpose of building a political constituency for OIRA and its work.
I have a somewhat different, but complementary, view of the history of White House regulatory oversight and the cost-benefit standard. It casts additional light on the question of institutional durability and the prospects for a fully realized regulatory budget.
The emergence of White House regulatory oversight in the 1970s, and its continuance through both liberal and conservative administrations, are not puzzling, and will surely continue. The 1970s was the decade of the rulemaking revolution, when the regulatory agencies were breaking out of the cocoon of narrow case-by-case adjudication and discovering the power of informal notice-and-comment rulemaking affecting entire economic sectors. And there were many more such agencies, with the establishment of EPA in 1970 and many other new agencies concerned with health, safety, consumer protection, nondiscrimination, and other matters.
Suddenly, numerous agencies throughout the executive branch were issuing rules with costs and benefits of tens and hundreds of millions of dollars, often accompanied by great political controversy. The controversies were reaching the White House; the president would be praised or blamed, usually both, for the rules and their consequences; the White House staff, and auxiliary staffs in the Executive Office of the President such as OMB and the Council of Economic Advisors, needed to be in the loop. They got there at first intermittently, during the Nixon, Ford, and Carter years, and then systematically beginning with Reagan. The White House will remain in the loop, in one form or another, for as long as the “administrative state” continues to be an important, powerful feature of our government.
That White House oversight began and continued in the form of a cost-benefit standard, as opposed to a regulatory budget or other apparatus, is more complicated. At the beginning, as Tozzi emphasizes, CEA economists were familiar with cost-benefit analysis from their academic work, and OMB officials knew of it through their work on defense and public works projects. But there was another initial appeal: it was not only a decision procedure but a reporting procedure. White House officials lead hectic lives, consumed by a flurry of incommensurate problems, close calls, and ambiguous situations that may or may not become crises. When they receive a complaint from a political supporter or lobby group or member of Congress, alerting them to a heretofore obscure regulatory proceeding and looming decision, they need a prompt, succinct summary of the issue and its magnitude. What is the agency trying to do? How much will it cost? What about these complaints – might the project be moderated or improved, or just abandoned? A 10-minute West Wing briefing is an informal, top-line cost-benefit analysis, augmented with distributional and political details. The cost-benefit standard fits these needs well. In contrast, a regulatory budget leaves most programmatic details and discretion at the agency level – that is one of its advantages in organization theory. But, standing alone, it misses the political-governmental requirement for actionable centralized information and occasional policy countermanding.
Another, deeper advantage has emerged over time. The cost-benefit standard is a constraint, but an elastic constraint, and is a procedure not only for guiding decisions but also for justifying decisions. It places broad limits on regulatory initiatives – decisively vindicating very good ones and scuttling very bad ones – and has done so in every administration. But it also leaves ample room for contestable and subjective assumptions on many matters, especially in estimating the costs of foregone opportunities, the benefits of non-traded public goods, and the consequences of very-low-frequency events and very-low-exposures to pollutants; and it is highly sensitive to financial assumptions concerning discount rates and costs of capital.
These protean qualities have permitted cost-benefit analysis to serve as a constraint during antiregulation Republican administrations and a propellant during pro-regulation Democratic administrations. And they have facilitated OIRA’s dual role as neutral economic overseer and activist policy overseer for the current president. In every administration, Democratic as well as Republican, OIRA has regularly disputed agency cost-benefit analyses and moderated agencies’ tendencies to overregulation. Also, in every administration, OIRA has regularly approved agency initiatives it thought were unjustified on the cost-benefit merits, following presidential priorities or acceding to congressional or other political pressures. In the extreme, it has approved frankly paternalistic rules such as energy-efficiency standards where the benefits are entirely personal – consisting of presumed better purchasing decisions involving no externalities, no market failures, and no health, safety, or environmental improvements at all – and where aggressive calculations of net benefits have then been deployed as public-relations tools.Footnote
3
OIRA is, of course, criticized in the news and in academic journals for throttling beneficial rules and approving harmful ones. However, its conflicted institutional role as both economic and political overseer requires that it maintain case-by-case flexibility. A striking feature of the public pronouncements and writings of OIRA administrators, especially while in office, is that they downplay the importance of the cost-benefit standard and emphasize broader, hazier functions such as promoting regulatory transparency, democratic accountability, and inter-agency cooperation.
Would OIRA’s position change if it were overseeing a regulatory budget? OMB’s overseers of expenditure budgets do not need or want case-by-case flexibility – they are ministerially enforcing spending limits that the president and Congress have agreed to (with their input).
OIRA, as overseer of regulatory budgets, would be more like them – but probably not by much. The budget metric would still be estimates of private costs, not dollars residing in Treasury accounts, and these would entail OIRA-agency disagreements and negotiations, over methods as well as sums, both at budget-setting time and case-by-case during the year. OIRA oversight of agency estimates of regulatory benefits would, in theory, be dispensed with – but, in practice, not entirely, because regulatory costs and benefits are often conflated.Footnote
4
Moreover, the need for centralized information, at least for major and politically controversial rules, would continue.
And, because major rules are almost always challenged in court, followed by lengthy and often unpredictable judicial review, there would be a need for many exceptions and continuous revisions to regulatory budgets. The accounting for the Trump Administration’s prototype regulatory budget, and the agencies’ adherence to its two-for-one rule and incremental-cost budget caps, have been heavily dominated by the agencies’ revisions to a handful of energy, environmental, and financial rules issued late in the Obama Administration; these have been matters of keen White House interest, and the ultimate results and budget impacts will not be known for several years when the courts have had their say. Still, there seems to have been more attention to eliminating old rules (many of them minor but pestiferous) than in previous administrations.Footnote
5
Although President Trump came to office with a strong animus against new regulation, he has, like his predecessors, come to use regulation for his own purposes in response to new political developments – for example, banning bump stocks that convert semi-automatic firearms into automatic firearms, and proposing to ban categories of vaping products. If he wins a second term, regulatory budgeting will become much more demanding all down the line, from the White House to OIRA to the agencies, and we shall see whether it was merely a first-term emergency brake or the beginning of a sustainable institution of executive government.
The regulatory budget is well worth pursuing for the long term. Even with its incorrigible measurement problems, and even with budget caps that may be highly political and detached from formal principal-agent optimization, it could produce better results than unbounded cost-benefit analysis. Along the way, it could permit some relaxation of centralized case-by-case reviews and the institutional conflicts they create for OIRA. But not their elimination: The activity of commandeering private resources for public ends by executive fiat will remain highly contentious and vigorously contested in the individual case, so that the particulars will often be as important to the principals (Congress as well as the president) as to the agents.
The achievable goal of the imperfect regulatory budget should be to improve the internal culture of the regulatory agency – to cause missionary regulators to realistically monitor the effectiveness of their stock of rules, to weigh them against the prospective results of new rules, and to make choices that maximize net benefits at the margin. That could be as politically durable as the current program has been, but here there is a great unknown. Under the cost-benefit standard, the elasticity that has preserved the program through successive administrations has been backstage and politically shrouded – in the more-or-less constraining methods employed for judging and justifying individual rules. Under the regulatory budget, the elasticity is front-and-center – in the more-or-less constraining budget caps publicly established from year to year and administration to administration. Would a newly arrived left-progressive administration be prepared to blow the lid off the regulatory budget of its conservative predecessor – in effect announcing that it was about to levy billions or trillions of dollars in new taxes – or would it prefer to abolish the “antiregulation” budgeting procedure and focus instead on the great social benefits it was about to bestow? But I have put the question in loaded terms, and can imagine that budgeting would indeed survive, especially if it was as embedded as President Reagan’s review program had become when President Clinton took office. An aggressive budget might assure the new president’s supporters that he or she was indeed going to get tough with the corporations and special interests; or budgets might be set differentially for favored and disfavored regulatory programs; or the procedure might be used for reassurance – to advertise, and practice, financial discipline in the pursuit of costly progressive goals.
But policy scholars should not be guided by political speculation. They – including today’s Jim Tozzis in the executive service – should stick to the intrinsic integrity and improvement of the institutions they study and administer. In this light, the greater political transparency of regulatory budgeting may be seen as a benefit, not a cost.
You have
Access
Jim Tozzi is an activist institutional economist. During his 19-year career in the federal civil service, he was a pertinacious institution builder, armed with a PhD in economics but never flaunting it. He gained a reputation, richly deserved in my experience, as a supreme bureaucratic tactician. But he applied his skills to antibureaucratic purposes. Incessantly, and occasionally at professional risk, he promoted and protected internal executive-branch procedures that used economic analysis, and measures of administrative effectiveness, against the incessant forces of political entropy, agency parochialism, and special-interest capture.
Tozzi’s purposes and methods are vividly on display in “Office of Information and Regulatory Affairs: Past, Present, and Future” (Reference Tozzi2019). He is impressed by the durability of White House review of agency regulations under an increasingly explicit cost-benefit (“maximum net benefits”) standard. The practice is wholly discretionary, with no statutory basis and no particular congressional or political constituency, yet it has been followed since the early 1970s by nine presidents of both parties and all points on the political spectrum. That is something of a puzzle, and also a caution. The practice has been indubitably beneficial and has come to play a central role in regulatory policy-making – but it draws the Office of Management and Budget (OMB) into an endless stream of highly charged political controversies, and it could be extinguished by the stroke of a presidential pen or by simple neglect and desuetude.
This leads Tozzi to recommend three measures for buttressing White House regulatory oversight. The first is that the National Archives officially recognize “Iconic Executive Orders,” beginning with Ronald Reagan’s Executive Order 12291 (1981). That order was certainly the decisive regulatory innovation, first establishing the cost-benefit analysis requirement and maximum net benefits standard as presidential policies that were to long endure. But, at the Archives, there would be considerable competition for iconic status, and regulatory efficiency would jostle with other executive purposes. Abraham Lincoln’s Executive Order 95 of 1863, the Emancipation Proclamation, would surely rate a gold star, as would George Washington’s Thanksgiving Day Proclamation of 1789 (unnumbered). Dwight Eisenhower’s school desegregation Executive Order 10730 of 1957, and Franklin Roosevelt’s banking holiday Proclamation 2039 of 1933, might get honorable mention. I hereby nominate E.O. 6102 (FDR’s 1933 confiscation of privately held gold coins), E.O. 10340 (Harry Truman’s 1952 seizure of the steel mills), and E.O. 10998 (John Kennedy’s 1962 authorization of public-employee collective bargaining) as the three rock-bottom worst. But I doubt that the National Archives – one of the least politicized agencies in Washington, standing tall and solid above the swamp – would want to enter this fray. It does have three established icons – the Declaration of Independence, the Constitution, and the Bill of Rights – and these days we should be content if it is able to maintain allegiance to them.
Tozzi’s second proposal is that JFK’s little known E.O. 11030 (1962), which provides that the drafting and vetting of executive orders be managed by OMB, should be amended to give OIRA (a division of OMB) a leading role in the process. This is an excellent idea. The issuance of executive orders has become rather ad hoc and chaotic in recent years, and increasingly controversial. The Department of Justice opines on the form and legality of draft orders, but that is inadequate. The things that presidents attempt to accomplish by executive order, often inspired by a passing contretemps or staff brainstorm, typically have a long and complicated history within the executive branch going well beyond matters of form and law, and also going well beyond the budget-oversight and related functions of OMB’s other divisions. Tozzi wants to fortify OIRA’s position within the executive branch, to help sustain its oversight of agency rulemaking, but his proposal makes sense on its own terms. Executive orders are mainly internal regulations – directions from the executive branch’s principal to his far-flung agents – and regulation is OIRA’s specialty.
Tozzi’s first two institution-building proposals are warm-ups for his third, much more ambitious one: that the cost-benefit standard for individual rules be augmented or supplanted by a regulatory budget capping the total incremental costs of all new rules. (Budgets would presumably be set on an annual basis, and could be applied at the level of programs, agencies, or the executive branch as a whole.) He helpfully characterizes the two procedures as alternative means of addressing the principal-agent problem in government. Missionary regulatory agencies – single-mindedly devoted to promoting environmental quality, energy efficiency, safe and healthful products and workplaces, financial stability and fairness, nondiscrimination, and a host of other worthy causes – are empowered to command substantial private expenditures on behalf of their missions. These expenditures may be on the order of a hundred million dollars for a single rule, and in the aggregate are much greater than the agencies’ own operating budgets, but they are subject to none of the disciplines of public finance (taxation, appropriation, budgeting) that apply to the government’s own operations and spending programs. Left to themselves, regulatory agencies will be inclined to order expenditures on their statutory missions that take insufficient account of competing claims and alternative uses of private resources; the agencies’ principal, be it the President, the Congress, or the general public, will be inclined to take a more capacious, balanced view.
In Tozzi’s formulation, cost-benefit analysis is an application of welfare economics and decision theory to the principal-agent problem, while regulatory budgeting is an application of institutional economics and optimal-delegation theory. The former were in vogue, in academic economics and government, in the 1970s when the White House review procedures were first established – but the latter are older (going back to a delicious passage in The Wealth of Nations) and better suited to the problem at hand, and have recently been enjoying a revival.
He points in particular to an important article by Yair Listokin, “Bounded Institutions.”Footnote 1 In “unbounded institutions,” agents make individual decisions according to rules or standards handed down by principals; in “bounded institutions,” principals set a quantitative cap on agents’ decisions. Each approach has advantages and disadvantages in different circumstances. The bounded approach tends to dominate where (a) agents decide among more numerous candidates for action (e.g., grant awards, regulatory interventions, student grades, legal prosecutions); (b) the range of quality among candidates is narrower; (c) principals have a better general idea of the number and average quality of candidates; and (b) agents are more biased, compared to the preferences of principals, in judging the quality of candidates.
Listokin’s analysis suggests important advantages of a regulatory budget (bounded) over a cost-benefit standard (unbounded), especially in the context of an agency such as the Environmental Protection Agency (his example) with a strong missionary bias and numerous candidates for pollution controls of varying degrees of strictness. Under a cost-benefit standard, EPA will systematically overvalue the benefits of its rules; every rule that meets the standard will be issued, resulting in excessive pollution-control investments from a government-wide (or society-wide) perspective. Under a regulatory budget, individual rules compete not with the standard but with each other – EPA, notwithstanding its benefits bias, will issue rules with the highest net benefits until it reaches its cost budget.
But there are shortcomings as well, or at least challenges. A regulatory budget relies on estimates of private costs, which are much spongier and more open to manipulation than the hard dollar sums of a spending budget (but this is a problem for the cost-benefit standard too).Footnote 2 The principal – the regulatory budget-setter – may have a poor idea of the distribution of potential regulatory strategies and outcomes, and so may set a cost cap that is too high or too low to achieve the benefits it desires (so OIRA may be a better budget-setter than Congress, because it is more knowledgeable about regulatory strategies and outcomes – although Congress might acquire a serviceable grasp of them through a regular budget exercise responding to itemized proposals from the executive).
Tozzi notes that the idea of a regulatory budget was in circulation in academic and government circles in the late 1970s when the cost-benefit standard was adopted instead. Now that institutional economics is catching up with welfare economics, he thinks the time may be ripe for moving to a regulatory budget. The Trump Administration has begun to do so through E.O. 13771 (January 30, 2017) and subsequent OIRA directives. These have required agencies to rescind at least two existing rules for every new rule they issue and to hold the net costs of their actions to an annual cost budget – and have set budget caps of zero or negative incremental costs. Tozzi would build on these steps, and the initial experience with them, by issuing a proposed revised executive order for public notice and comment, and by encouraging new optimal-delegation research building on the existing literature. The notice-and-comment process, and accompanying public hearings, would, like similar agency procedures, have the not-incidental purpose of building a political constituency for OIRA and its work.
I have a somewhat different, but complementary, view of the history of White House regulatory oversight and the cost-benefit standard. It casts additional light on the question of institutional durability and the prospects for a fully realized regulatory budget.
The emergence of White House regulatory oversight in the 1970s, and its continuance through both liberal and conservative administrations, are not puzzling, and will surely continue. The 1970s was the decade of the rulemaking revolution, when the regulatory agencies were breaking out of the cocoon of narrow case-by-case adjudication and discovering the power of informal notice-and-comment rulemaking affecting entire economic sectors. And there were many more such agencies, with the establishment of EPA in 1970 and many other new agencies concerned with health, safety, consumer protection, nondiscrimination, and other matters.
Suddenly, numerous agencies throughout the executive branch were issuing rules with costs and benefits of tens and hundreds of millions of dollars, often accompanied by great political controversy. The controversies were reaching the White House; the president would be praised or blamed, usually both, for the rules and their consequences; the White House staff, and auxiliary staffs in the Executive Office of the President such as OMB and the Council of Economic Advisors, needed to be in the loop. They got there at first intermittently, during the Nixon, Ford, and Carter years, and then systematically beginning with Reagan. The White House will remain in the loop, in one form or another, for as long as the “administrative state” continues to be an important, powerful feature of our government.
That White House oversight began and continued in the form of a cost-benefit standard, as opposed to a regulatory budget or other apparatus, is more complicated. At the beginning, as Tozzi emphasizes, CEA economists were familiar with cost-benefit analysis from their academic work, and OMB officials knew of it through their work on defense and public works projects. But there was another initial appeal: it was not only a decision procedure but a reporting procedure. White House officials lead hectic lives, consumed by a flurry of incommensurate problems, close calls, and ambiguous situations that may or may not become crises. When they receive a complaint from a political supporter or lobby group or member of Congress, alerting them to a heretofore obscure regulatory proceeding and looming decision, they need a prompt, succinct summary of the issue and its magnitude. What is the agency trying to do? How much will it cost? What about these complaints – might the project be moderated or improved, or just abandoned? A 10-minute West Wing briefing is an informal, top-line cost-benefit analysis, augmented with distributional and political details. The cost-benefit standard fits these needs well. In contrast, a regulatory budget leaves most programmatic details and discretion at the agency level – that is one of its advantages in organization theory. But, standing alone, it misses the political-governmental requirement for actionable centralized information and occasional policy countermanding.
Another, deeper advantage has emerged over time. The cost-benefit standard is a constraint, but an elastic constraint, and is a procedure not only for guiding decisions but also for justifying decisions. It places broad limits on regulatory initiatives – decisively vindicating very good ones and scuttling very bad ones – and has done so in every administration. But it also leaves ample room for contestable and subjective assumptions on many matters, especially in estimating the costs of foregone opportunities, the benefits of non-traded public goods, and the consequences of very-low-frequency events and very-low-exposures to pollutants; and it is highly sensitive to financial assumptions concerning discount rates and costs of capital.
These protean qualities have permitted cost-benefit analysis to serve as a constraint during antiregulation Republican administrations and a propellant during pro-regulation Democratic administrations. And they have facilitated OIRA’s dual role as neutral economic overseer and activist policy overseer for the current president. In every administration, Democratic as well as Republican, OIRA has regularly disputed agency cost-benefit analyses and moderated agencies’ tendencies to overregulation. Also, in every administration, OIRA has regularly approved agency initiatives it thought were unjustified on the cost-benefit merits, following presidential priorities or acceding to congressional or other political pressures. In the extreme, it has approved frankly paternalistic rules such as energy-efficiency standards where the benefits are entirely personal – consisting of presumed better purchasing decisions involving no externalities, no market failures, and no health, safety, or environmental improvements at all – and where aggressive calculations of net benefits have then been deployed as public-relations tools.Footnote 3
OIRA is, of course, criticized in the news and in academic journals for throttling beneficial rules and approving harmful ones. However, its conflicted institutional role as both economic and political overseer requires that it maintain case-by-case flexibility. A striking feature of the public pronouncements and writings of OIRA administrators, especially while in office, is that they downplay the importance of the cost-benefit standard and emphasize broader, hazier functions such as promoting regulatory transparency, democratic accountability, and inter-agency cooperation.
Would OIRA’s position change if it were overseeing a regulatory budget? OMB’s overseers of expenditure budgets do not need or want case-by-case flexibility – they are ministerially enforcing spending limits that the president and Congress have agreed to (with their input).
OIRA, as overseer of regulatory budgets, would be more like them – but probably not by much. The budget metric would still be estimates of private costs, not dollars residing in Treasury accounts, and these would entail OIRA-agency disagreements and negotiations, over methods as well as sums, both at budget-setting time and case-by-case during the year. OIRA oversight of agency estimates of regulatory benefits would, in theory, be dispensed with – but, in practice, not entirely, because regulatory costs and benefits are often conflated.Footnote 4 Moreover, the need for centralized information, at least for major and politically controversial rules, would continue.
And, because major rules are almost always challenged in court, followed by lengthy and often unpredictable judicial review, there would be a need for many exceptions and continuous revisions to regulatory budgets. The accounting for the Trump Administration’s prototype regulatory budget, and the agencies’ adherence to its two-for-one rule and incremental-cost budget caps, have been heavily dominated by the agencies’ revisions to a handful of energy, environmental, and financial rules issued late in the Obama Administration; these have been matters of keen White House interest, and the ultimate results and budget impacts will not be known for several years when the courts have had their say. Still, there seems to have been more attention to eliminating old rules (many of them minor but pestiferous) than in previous administrations.Footnote 5
Although President Trump came to office with a strong animus against new regulation, he has, like his predecessors, come to use regulation for his own purposes in response to new political developments – for example, banning bump stocks that convert semi-automatic firearms into automatic firearms, and proposing to ban categories of vaping products. If he wins a second term, regulatory budgeting will become much more demanding all down the line, from the White House to OIRA to the agencies, and we shall see whether it was merely a first-term emergency brake or the beginning of a sustainable institution of executive government.
The regulatory budget is well worth pursuing for the long term. Even with its incorrigible measurement problems, and even with budget caps that may be highly political and detached from formal principal-agent optimization, it could produce better results than unbounded cost-benefit analysis. Along the way, it could permit some relaxation of centralized case-by-case reviews and the institutional conflicts they create for OIRA. But not their elimination: The activity of commandeering private resources for public ends by executive fiat will remain highly contentious and vigorously contested in the individual case, so that the particulars will often be as important to the principals (Congress as well as the president) as to the agents.
The achievable goal of the imperfect regulatory budget should be to improve the internal culture of the regulatory agency – to cause missionary regulators to realistically monitor the effectiveness of their stock of rules, to weigh them against the prospective results of new rules, and to make choices that maximize net benefits at the margin. That could be as politically durable as the current program has been, but here there is a great unknown. Under the cost-benefit standard, the elasticity that has preserved the program through successive administrations has been backstage and politically shrouded – in the more-or-less constraining methods employed for judging and justifying individual rules. Under the regulatory budget, the elasticity is front-and-center – in the more-or-less constraining budget caps publicly established from year to year and administration to administration. Would a newly arrived left-progressive administration be prepared to blow the lid off the regulatory budget of its conservative predecessor – in effect announcing that it was about to levy billions or trillions of dollars in new taxes – or would it prefer to abolish the “antiregulation” budgeting procedure and focus instead on the great social benefits it was about to bestow? But I have put the question in loaded terms, and can imagine that budgeting would indeed survive, especially if it was as embedded as President Reagan’s review program had become when President Clinton took office. An aggressive budget might assure the new president’s supporters that he or she was indeed going to get tough with the corporations and special interests; or budgets might be set differentially for favored and disfavored regulatory programs; or the procedure might be used for reassurance – to advertise, and practice, financial discipline in the pursuit of costly progressive goals.
But policy scholars should not be guided by political speculation. They – including today’s Jim Tozzis in the executive service – should stick to the intrinsic integrity and improvement of the institutions they study and administer. In this light, the greater political transparency of regulatory budgeting may be seen as a benefit, not a cost.