The next practical problem in implementing a regulatory budget system is how such a system would actually operate.  This chapter examines the organizations that would be involved in operating a regulatory budget, and how a budget could be workably formulated and executed.  The chapter concludes with a discussion of the special problem of initially setting up a regulatory budget system.

A.  Organization for a Regulatory Budget

A central administrative body would be required to manage the overall regulatory budget system. For convenience, that body is referred to in this study as the Office Regulatory Budget (ORB). 

The role of ORB would be broadly analogous to that of the Office of Management and Budget (OMB) in the fiscal budget process. It would be the responsibility of ORB to develop and administer the detailed procedures needed to operate a regulatory budget, and to manage the formulation and execution of the budget. While the functions of ORB could be performed by some other entity of the Executive Office of the President of comparable rank with OMB, the similarity in responsibilities would argue for establishing ORB as a part of OMB.

A key decision would be whether to merge the regulatory budget operations into the existing fiscal estimates group of OMB. The fiscal estimates staffs possess a wealth of knowledge about the regulatory responsibilities of the agencies that they supervise, and are experienced in bringing to bear on the activities of those agencies the point of view of the Executive Office of the President.  Hence the Fiscal estimates groups could significantly enhance the effectiveness of a regulatory budget process, especially at the outset.

Opposing considerations, however, might make it preferable to establish ORB as a separate element within OMB. Dispersing regulatory budget operations among the various fiscal estimates groups would relegate ORB to the role of a central staff providing mainly procedural guidance. Especially in the initial stages of the regulatory budget, such a role could impede the prompt revision of procedures as experience was gained, and could hamper the dissemination of the revised procedures to the regulatory agencies. Also, dispersion of budget operations would make it more difficult to achieve consistency of treatment among regulatory agencies.

Furthermore, the fiscal estimates groups already have a large and demanding workload that is tied to the inexorable demands of the fiscal budget calendar. Thus, there would be a risk that a new function such as regulatory budgeting would not be able to compete effectively for managerial and staff attention if it was merged into the fiscal estimate divisions.  Finally, regulatory agencies would not necessarily deal with a regulatory budget process through their fiscal budget offices, with which OMB's fiscal estimates groups customarily deal. If so, the existing lines of communication between OMB's fiscal estimates groups and their client agencies would not be the same as the lines of communication needed for the regulatory budgeting process.

On balance, it would appear to be preferable to establish ORB as a separate element of OMB, with responsibility and authority to deal directly with the agencies in the formulation and execution of the regulatory budget. When the regulatory budget system had been operating long enough to have settled down, consideration could be given to integrating the operation of the system into the fiscal estimate groups, retaining for ORB the role of specialized staff to deal with across-the-board procedures and with the aggregation of annual agency authorizations.

B.        The Regulatory Budget Process

For ease of exposition, the discussion of the regulatory budget process follows the budget cycle for a single period, from formulation through execution. In practice, of course, the cycles for successive periods would overlap just as they do for the fiscal budget. In broad outline, the regulatory budget process described here parallels the existing Federal fiscal budget process. The regulatory budget period is assumed to be the traditional Federal fiscal year,[1] which begins on October 1 of the preceding calendar year and ends on September 30 of the current calendar year.

1.         Formulation of, a Regulatory Budget

ORB would initiate the regulatory budget cycle about 21 months before the fiscal year for that budget began. The cycle would start with the issuance to regulatory agencies of detailed procedural guidelines for submitting their requests for compliance-cost budgetary authority. Those guidelines would include tentative ceilings for total agency requests. [2]

The responses of the agencies would be due at ORB about 4 months before the applicable fiscal year began. The agencies' requests would describe the proposed new regulations and the estimated costs of complying with all the regulations that would be in effect during that fiscal year.[3] The expected benefits of the agencies' regulations would also be included in their submissions to ORB, in support of their requests for budget authorizations. However, as discussed in Chapter 1, the benefits of regulations would not enter explicitly into the regulatory budget process.

ORB would review the agencies' requests and recommend modifications to conform with government policy on the total amount and the composition of regulatory activity. The modifications would almost certainly be downward, as most agencies would submit requests for above-ceiling compliance-cost authorizations. About eleven months before the beginning of the applicable fiscal year, ORB would submit to the President its recommendations on both aggregated, government-wide budget totals, and on the agency-by-agency dollar amounts of authorized compliance costs.

The President, aided by his staff, would review ORB's recommendations prior to making the final decision on his proposal to the Congress for the next year's regulatory budget. As part of the review, individual agencies could appeal ORB's modifications of their requests for budget authorization to the President himself, as happens occasionally in the fiscal budget process.

The results of the President's decision would be communicated by ORB to the agencies, which would then prepare detailed submissions to send to the Congress. Early in the new congressional session, the President's regulatory budget for compliance-cost authorizations for the next fiscal year, both total and agency-by-agency, would be forwarded to the Congress. Shortly thereafter, the individual agencies would submit their detailed budget requests to the Congress.

The formulation of a regulatory budget would end with the enactment into law of the President's proposals as modified by the Congress. More detailed discussion of congressional involvement in the regulatory budget process is presented in Section D below.

2.         Execution of a Regulatory Budget

Once the fiscal year had begun, the regulatory activities of Federal agencies would be constrained by the compliance-cost ceilings set in the budget for that year. The constraints would need to be sufficiently flexible to permit the agencies to adapt to changing circumstances in a timely manner. Thus, a regulatory agency would not be limited to the promulgation of only those regulations that were specifically included in its original request to ORB, nor would the agency be required to issue every regulation that was included in its request.[4]  Also, if an agency revoked some of its existing regulatory requirements, it could be allowed to increase its regulatory budget authorization by the amount of compliance costs thereby saved.  The procedures for proposing and promulgating new or revised regulations under a regulatory budget (or for revoking current regulatory requirements) would resemble the existing procedures, but it would expand on them in significant ways:

·        At the time of proposal, an agency would publish a draft economic impact statement, much as it is required to do now. The draft statement would include a preliminary estimate of the anticipated cost of complying with the new or revised regulations. The comments received on the proposal would aid in preparing the subsequent comprehensive estimate of compliance costs.

 

·        When the agency formally promulgated the new or revised regulation, it would be required to publish its final economic impact analysis. That analysis would contain a comprehensive estimate of the additional compliance costs that would result from the regulations.

 

·        The final economic impact analysis would be open to public comments for a period of 90 days. The comments would be submitted to ORB, with a copy to the promulgating agency.[5]

 

·        After the comment period ended, ORB would resolve differences between the agency's estimated compliance costs and those in the comments received.  ORB would reach a final decision on the compliance cost figures to be charged to the agency's regulatory budget authorization within 90 days of the close of the comment period (six months after promulgation of the regulations). That decision would be the official estimate of compliance costs and would be so certified by ORB.[6]

 

·        The certified estimate would be charged against the promulgating agency's regulatory budget authorization.

C.        Enforcing a Regulatory Budget

The existence of a regulatory budget would impose a new constraint on Federal agencies that would not be welcomed by regulatory officials. Efforts to avoid the constraint should therefore be anticipated. 

Three principal problems in enforcing a regulatory budget merit attention: (1) an incentive to overstate compliance-cost estimates during budget formulation; (2) an incentive to understate those estimates in promulgating regulations[7]; and (3) preventing the overspending of regulatory budget allocations. These three problems are discussed in order.

1.         Overstatement of Compliance-Cost Estimates during Budget Formulation

In preparing their requests to ORB in the early part of the budget cycle, Federal agencies would have an incentive to overstate their estimates of the compliance costs of new or revised regulations in order to secure the largest possible budget authorizations.

ORB's ability to identify overstatements at this stage of the regulatory budget process would be limited. The reason is that even the agencies would rarely know so far in advance how the subjects of a proposed new requirement would actually respond to it. Only after the agencies had further developed their regulations and received comments on them would they possess the data to support more reliable estimates.

Overstatement of estimated compliance costs during budget formulation would not, however, be a fatal defect in a regulatory budget system. ORB's recommendation to the President for allocations of budget authority would depend only in part on preliminary estimates of compliance costs by the agencies. Also important would be high-level policy guidance given to ORB examiners on how much in additional national resources could be devoted to the various social goals represented by the regulatory agencies' missions. In addition, the agencies themselves would not want to make outrageous initial requests, lest they weaken their subsequent credibility with ORB.

2.         Understatement of Compliance-Cost Estimates during Promulgation

In promulgating regulations, it would be in an agency's interest to understate estimated compliance costs, in order to minimize the share of the available authorization devoted to any one regulation. During promulgation, however, ORB would be in a much better position than during budget formulation to evaluate the agencies' estimates. For one thing, ORB would have available the supporting data in the detailed economic impact analyses, as well as the data submitted as part of the public comments. 

Secondly, agencies would have an offsetting incentive not to understate their compliance-cost estimates. By systematically understating the costs of all new regulations for a given period, the agency would run the risk that ORB would systematically certify estimates greater than the agency's figures. This would increase the agency's chances of exceeding the regulatory budget authorization for that period. The resulting deficit would prompt sanctions (as discussed below), or would at least be deducted from the agency's budget authorization for the next fiscal year, which would already have been established. An agency would presumably have difficulty justifying supplemental allocations to cover deficits caused by its own poor compliance-cost estimates.

3.         Controlling Overspending of Regulatory Budget Allocations

Enforcing the budget ceilings of regulatory agencies would require both appropriate bookkeeping procedures and sanctions that could be imposed when a budget ceiling was exceeded.

a.         Bookkeeping

The bookkeeping procedures for a regulatory budget would be conceptually similar in form to those for the fiscal budget, but would be simpler and less costly to carry out:

·        Authorized expenditure limits and (certified) compliance costs saved by retiring old regulations would be entered as credits on the agencies' ledgers. Compliance costs spent on new or revised regulations would be entered as debits.

 

·        The scale of bookkeeping needed for a regulatory budget would be far smaller than for the fiscal budget. The huge number of individual fiscal transactions each year would dwarf the sum total of the separate regulatory actions taken each year by Federal agencies.

b.         Sanctions

It would not be possible under a regulatory budget to impose sanctions for overspending parallel to those used under the fiscal budget. Thus, new kinds of sanctions would have to be devised to make a regulatory budget work.  The authority to commit appropriated fiscal funds is extensively delegated to relatively low administrative levels. That is necessary because of the vast number of individual transactions involved. The traditional enforcement tool is to hold individual certifying officers personally liable, financially, for funds committed in excess of their allocations. The threat of personal financial liability has proven widely effective in achieving observance of fiscal budget limits.[8]  In addition, modern management information systems have made possible secondary cross-checks of whether fiscal obligations exceed authorizations.          

It has been suggested that a serious defect in a regulatory budget system would arise from the inability to hold anyone personally liable, financially, for exceeding authorized compliance cost allocations. The implication is that a regulatory budget could not be effective because it would deal only with funny-money, not the real money with which the fiscal budget deals.[9]

In fact, the so-called funny-money problem is more apparent than real. The heart of the issue of controlling the overspending of any budget authority is the level at which the discipline takes place. Under the fiscal budget, as already noted, obligating authority is delegated to low administrative levels. In contrast, the regulatory budget equivalent of fiscal obligating authority would not be nearly so widely dispersed. The reason is that most regulations are promulgated over the names and by the authority of Presidentially-appointed heads of Federal agencies.[10]

Thus, enforcing regulatory budget ceilings would be primarily a political, not an administrative, problem.  Without exception, the officials who would authorize regulatory-budget expenditures—by promulgating regulations—would be at politically responsible levels. Regulatory budget discipline would have to be imposed through the political process.

In starkest terms, to prevent the overspending of regulatory-budget allocations, the President would have to stand ready to dismiss an agency head who failed to stay within his or her budget authorization. Since most officials who promulgate regulations serve at the pleasure of the President, regulatory-budget discipline should pose no problem for a President who was determined to obtain it.[11]

D.        Congress and a Regulatory Budget

A regulatory budget would be a major change in Federal policy towards regulation. For that reason, it was argued in Chapter 2 that the President should not attempt to establish a regulatory budget system unilaterally, by executive order, even though constitutionally it probably could be done.  Rather, such a system should be set up under the authority of a law passed by the Congress.[12]

Beyond merely authorizing a regulatory budget, the Congress would need to participate regularly in the budgeting process itself. This would require that the Congress be kept informed of three key decisions regarding the regulatory budget: (a) the aggregate compliance-cost authority proposed for the budget year; (b) the proposed budget allocations to individual agencies (which the Congress would review, modify, and approve); and (c) proposed increases during the current year in specific agencies' budget allotments.

How the Congress organized itself to deal with the President's regulatory budget proposals would be vitally important. The primary purpose of the regulatory budget system would be to set overall limits on the compliance costs resulting from regulatory requirements. The purpose would not be to improve decision making on individual regulations.  To accomplish the primary purpose, the Congress in its action on the regulatory budget would need to take an overview of the broad economic impact of regulation and avoid being distracted by individual regulatory requirements.

It would be important, therefore, that the Congress deal with the President's budget proposals on a unified basis—for example, through its Budget Committees or the Joint Economic Committee. While the other, more specialized congressional committees could advise on specific matters, it would not be appropriate for those committees to set the regulatory budget authorizations for their client agencies.  As the originators of legislation giving rise to regulations, the specialized committees could not reasonably be expected to be any more objective than the regulatory agencies about the overall limits of total compliance costs in the areas of their particular interest.

To the extent that the Congress elected to become involved in individual regulatory issues, it could do so through its existing substantive, appropriations, and oversight committees. But the regulatory budget could not be successful as a management tool if the Congress were to inject itself into substantive regulatory details when it acted on the President's requests for authority to impose compliance costs.

E.        Public Participation in the Regulatory Budget Process

Providing an opportunity for public participation in government decision processes has in the past several years become a matter of major concern and effort. Thus the issue of public participation in the regulatory budget process needs consideration. As before, it is useful to distinguish between the formulation and execution stages of that process.

1.         In the Formulation Process

There would be no more reason to have the public participate in the formulation of a regulatory budget, prior to its submission to the Congress, than in the formulation of the fiscal budget. Traditionally, the fiscal budget has been kept confidential until the President submits it to the Congress. Members of the public then have their say during the congressional deliberations on the final form of the budget.

A regulatory budget would best be handled in the same manner. Until summary data were available on the compliance costs to be imposed on the economy during a future fiscal year, there could be little useful public discussion of whether the relative and absolute levels of those costs were appropriate. Such data would be available for the first time when the President's regulatory budget proposals were sent to the Congress. The public could then submit comments at the hearings that the Congress would hold on the proposed regulatory budget.

2.         In the Execution Process

Public participation in the execution of a regulatory budget would involve two key issues:

·        which regulations would be acted upon when the budget limit forced a choice; and

 

·        whether the compliance-cost estimates for the proposed or promulgated regulations were valid.

The choice among competing new regulations would be as amenable to public participation under a regulatory budget system as it is currently without it. Public interest groups already have ample ways to communicate their concerns on new regulations to politically responsible regulatory officials. Thus, no special further procedures for public participation would seem necessary here. 

Public comments on the validity of compliance cost estimates, in contrast, would involve at least new forms, if not new channels, of public participation. Under a budget for regulatory compliance costs, various segments of the public would have a stake in the cost estimates finally certified by ORB. The Federal agencies' own estimates would tend to appear, on the one hand, too low to those likely to bear the costs and, on the other hand, too high to those advocating more stringent regulation.

There is a possible problem of equity in public access to government under a regulatory budget. Most of the public comments on the compliance-cost estimates would come from groups with relatively large stakes in the certified cost figures. Groups with smaller stakes—for instance, small businesses or widely dispersed groups such as consumers or the poor—might be discouraged from participating by the costs of submitting comments.

This is not the place to debate the validity of the foregoing argument.[13]  It is pertinent to note, however, that the same problem exists under the present system of government regulation. The existence of a regulatory budget would add a dimension to the problem, in that compliance-test estimates would play a larger role in determining what is regulated and what is not. But having a regulatory budget could also reduce the costs of access to the public decision-making process by providing a formal structure where now only informal procedures are used.

F.         Special Problems of Introducing a Regulator Budget System

Three special, interrelated problems would be encountered in starting up a regulatory budget system:

·        How should a regulatory budget be phased in?

 

·        Which agencies should be included in the introductory phases?

 

·        Should a regulatory budget initially cover only the compliance costs of new and revised regulations, or the total costs of complying with all regulations?

1.         Phase-In

A regulatory budget could not be installed overnight.  A large amount of preparatory work would be needed to make even a skeleton regulatory budget viable. Lead time would be required to build staffs and train operating personnel in both ORB and the agencies. ORB would need staff to develop the budget procedures, without which the budget formulation process could not begin. The agencies would need to acquire and train staff to develop the estimates of compliance costs.

The lead time required before agencies could be subjected to regulatory budget constraints would be at least two years—none too generous an amount of time to allow ORB to be formed and initially staffed, and then to develop guidelines for the agencies on how to prepare their budget requests. To illustrate: If the final decision (represented, for example, by enabling legislation) to proceed with a regulatory budget was made in July 1980, the first period during which the new constraint applied would be fiscal year 1983 (October 1, 1982 - September 30, 1983).  The agencies would send their initial requests for regulatory budget authorizations to ORB in the summer of 1981.  The first budget allocations would be issued to the agencies in about September 1982, to take effect beginning October 1, 1982.

Once an initial schedule was adopted, interim procedures would be needed to head off agency attempts to promulgate as many regulations as possible before the regulatory budget went into effect. One procedure would be to count the compliance costs of regulations promulgated during the lead-time period against the budget allocation for the year of operation. In the above illustration, a1l regulations issued after October 1980 would be charged to an agency's fiscal 1983 budget authorization.

2.         Initial Agency Coverage

The philosophy that it is easier to start something new by trying it out first on a limited scale would argue for beginning a regulatory budget with only a few agencies. Its coverage could be expanded later, once some experience had been gained.

It would be several years, however, before any experience with regulatory budgeting could be analyzed and lessons drawn from it. Considering the lead times required, it would be six or seven years after the initial decision to try regulatory budgeting before additional agencies would be included (two years to implement the initial trial; two or three years of operating experience and evaluation; and two years to implement the expansion itself).

Starting small, moreover, could lend an air of experiment that would encourage the agencies initially covered to do everything in their power to see that the system failed.  In addition, officials in the trial agencies would feel discriminated against; the resulting resentment would further motivate them to sabotage the system.[14]  Finally, partial initial coverage would provide an opportunity for both legislative and executive policy makers to shift regulatory actions to the exempt agencies.  On balance, therefore, it would appear preferable to apply a regulatory budget simultaneously to all regulatory agencies right from the start.

3.         New and Revised vs. Total Compliance Costs

As discussed in Chapter 1, a regulatory budget operating at full scale would deal with the total compliance costs imposed by Federal regulations. However, the magnitude of the task of estimating the total continuing compliance cost of all regulatory requirements would make it impractical to cover them in the first few years of regulatory budget operation.

A workable compromise would be to begin regulatory budgeting with only the compliance costs of new and revised regulations, and to shift to a total compliance-cost base after a period of a few years. The cost data for new and revised regulations that would be generated during the first few years of budget operation would ease the eventual shift to total compliance-cost budgeting.