Appendix B: Initial Regulatory Assessment


As directed by Executive Order 12866, as amended without substantial change to its requirements by Executive Order 13258, the Department is required to conduct an initial regulatory impact analysis (hereinafter "RIA" or "regulatory assessment") in order to assess the economic benefits and costs of its proposed regulations implementing titles II and III of the ADA.  The purpose of regulatory analysis is to inform stakeholders in the regulatory process of the effects, both positive and negative, of the proposed regulations.  In this context, the primary stakeholders are individuals with disabilities who will benefit from using accessible facilities and the owners and developers of covered entities that will incur the costs of compliance.  In addition, as directed by the Regulatory Flexibility Act of 1980, as amended by the Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), as well as Executive Order 13272, the Department is required to consider the potential impact of its proposed regulations on small entities.                                                                               

A key component of the Department's regulatory assessment is a comprehensive benefit-cost analysis of the proposed revisions to the ADA Standards.  OMB Circular A-4 requires Federal agencies to conduct a full benefit-cost analysis for any regulation that is "economically significant"--that is, a regulation that is expected to have an annual impact on the economy of $100 million or more.  Such an analysis must include both quantitative and qualitative measurements of the benefits and costs of the proposed regulation, as well as a discussion of each potentially effective and reasonably feasible regulatory alternative.  OMB Circular A-4 also stipulates that regulatory analyses should only assess those costs and benefits that arise as a result of the proposed regulations themselves --in other words, the incremental impact of the proposed regulations when compared to a baseline of the legal status quo that would continue to apply absent regulatory action.     

Early on in this process, the Department concluded that the economic impact of its adoption of the proposed standards was likely to exceed this $100 million threshold, not only because it would be proposing to adopt several years' worth of revised and supplemental accessibility guidelines at once, but also because the proposed standards would apply to all newly constructed and existing facilities.  Accordingly, the Department has conducted an initial RIA for the proposed standards.  Consistent with the requirements for regulatory analyses, the RIA assumes a 40-year lifecycle for the longest lasting facilities subject to the regulations (here, a typical newly constructed building) before they must be substantially altered, torn down, or rebuilt.  The RIA also assumes that the proposed regulations will remain in force for 15 years, after which time it is presumed they would be superseded by future revisions to the title II and title III regulations. 

In September 2004, the Department issued an Advance Notice of Proposed Rulemaking ("ANPRM") which, among other things, described its proposed methodology for the initial regulatory assessment and solicited public comment on this methodology generally.  See 69 Fed. Reg. 58,768 (Sept. 30, 2004).  Additionally, section IV of the ANPRM entitled "Regulatory Assessment Issues" posed specific questions for public comment relating to the application of the proposed standards to existing facilities, including general sources for benefit and cost data, information on the impact of the proposed rules on small entities and suggestions for regulatory alternatives, and recommended sources of data for certain types of facilities or requirements.  Id. at 58,779-782 (Question Nos. 9-49).  The Department received many comments in response to the ANPRM and it has taken those comments into consideration during the regulatory assessment process.

At the same time, the Department also received many comments expressing the view that economic analysis is irrelevant with respect to the implementation of a civil rights statute.  Under this view, because the ADA is a civil rights statute protecting the rights of individuals with disabilities, regulations designed to implement its protections are necessary regardless of whether quantifiable benefits can be shown to outweigh costs.  As these commenters noted, traditional benefit-cost analysis is not designed to measure the inherent value of civil rights protections or to make judgments about fairness or equity.

The Department is sympathetic to the views expressed by these commenters.  However, the Federal laws and regulations that require agencies to express the benefits and costs of regulations in economic terms do not distinguish between regulations that implement civil rights statutes like the ADA and regulations that implement other kinds of laws.  The Department also believes that there is much to be gained from the comprehensive identification and description of the benefits of accessibility standards, which are, after all, designed to ensure equal access for everyone.  Such benefits include not only the measurable benefits to individuals with disabilities but also the more subtle and far-reaching benefits for society as a whole.  The majority of commenters representing industry groups also expressed the belief that the proposed standards would not confer any measurable benefit on individuals with disabilities, and, consequently, were perceived by some business owners as "punitive."  In fact, not only do the revised requirements confer measurable benefits on individuals with disabilities, in many cases, they also lower the costs for businesses.  By conducting a comprehensive assessment of the benefits and costs of the proposed standards, the Department hopes to promote greater understanding of the ADA and to further compliance with its civil rights protections.

Complete copies of the Department's initial RIA and accompanying Supplementary Results report are available on the Department's ADA web site (  The RIA itself is the work product of HDR|HLB Decision Economics, Inc., the economics firm with which the Department has contracted to conduct its initial regulatory assessment.  The Department has adopted the results of the RIA as its assessment of the benefits and costs that the proposed standards will confer on society.  The Department invites the public to read the RIA and to submit electronic comments by visiting the Department's web site for public comments.  See When the Department publishes a final rule, it will also publish an accompanying final regulatory assessment.  What follows is a general overview of the basic principles of the RIA, as well as the Department's responses to ANPRM comments concerning the methodology for this assessment.

Methodology for Data Collection

Several commenters proposed that the Department measure the relevant inputs for the RIA--such as the types of benefits individuals might realize from using a particular element or space in a facility, the unit costs that facilities will incur to comply with a requirement, or the likelihood that compliance will be readily achievable--by conducting surveys, focus groups, and similar types of studies.  For example, commenters representing industry groups suggested that the Department conduct a nationwide survey of existing facilities representing a range of ages, sizes, and building methods in order to assess the unit costs to existing facilities of complying with the proposed regulations.  Similarly, in order to measure the benefits to users, some commenters proposed that the Department conduct a national survey of people with disabilities using a broad sampling of ages, types of impairments and socioeconomic status.  Other suggestions included interviewing support groups or State health officials and staff at long term care facilities, conducting a nationwide survey using the Social Security mailing list, and adding questions to the U.S. Census questionnaire.

The Department has determined that it would be infeasible to conduct surveys or otherwise collect information from (or about) all facilities and all persons with disabilities nationwide.  Nor would surveys on the "real world" costs of compliance have aided the regulatory assessment; only the incremental costs of compliance are relevant to the analysis. Similarly, the Department also has determined that it would be infeasible to conduct a nationwide survey of individuals with disabilities with respect to the incremental benefits they might be likely to experience from the proposed regulations. 

Instead, the RIA relies on publicly available data sources--supplemented as necessary with estimates generated or verified by expert cost and benefit panels--to calculate the incremental impact of the proposed regulations.  See RIA, Ch. 4.  Public data sources used in the RIA are wide-ranging and include:  the 2002 Economic Census (to estimate the number and types of existing facilities); RS Means publications (to estimate unit costs); Dodge Construction Potential Bulletins (to estimate new construction rates); firm size data compiled by the Small Business Administration's Office of Advocacy (to estimate the total number and sales receipts of small businesses); the Annual Time Use Survey published by the Bureau of Labor Statistics (to estimate facility use and travel time); population surveys by the U.S. Census Bureau (to estimate the percentage of U.S. population with disabilities and types of disabilities); and average hourly wage statistics compiled by the Bureau of Labor Statistics (to estimate the value of time per facility group).   For those aspects of the RIA model that lacked publicly available data, estimates were developed by HDR/HLB or Department architects (as appropriate) and then reviewed by expert cost and benefit panels.  From the cost perspective, estimated values include the number and type of elements per typical facility.  See RIA §§ 4.1.2, 4.1.7.  With respect to benefits, the expert panel developed estimates concerning the time savings due to changes in accessibility, the expected number of uses for each requirement, and the likelihood that persons with disabilities would realize benefits from a requirement.  See RIA §§ 4.2.4, 4.2.6. 

The Access Board's Final Regulatory Assessment - 2004 ADAAG

In July 2004, the Access Board published its final regulatory assessment for the 2004 Americans with Disabilities Act and Architectural Barrier Act Accessibility Guidelines ("2004 ADAAG").  See Regulatory Assessment of the Final Revised Accessibility Guidelines for the Americans with Disabilities Act and Architectural Barriers Act, (July 2004). A few years earlier, the Access Board also issued final regulatory assessments for its supplemental guidelines for play areas (2000) and recreation facilities (2002).[1] The Access Board's final regulatory assessment for the 2004 ADAAG does not, however, incorporate these supplemental guidelines into its economic analysis since the costs of these guidelines had already been addressed in prior regulatory assessments.  

In summary, the Access Board's final regulatory assessment for the 2004 ADAAG used a sampling approach to calculate the costs of the revised guidelines as applied to newly constructed and altered facilities.  In this final regulatory assessment, the Board identified fourteen requirements that were projected to impose higher costs (relative to the 1991 ADAAG) for newly constructed or altered facilities.  From this group of "increased cost" requirements, the Board selected ten requirements for direct economic analysis based on its determination that these requirements were likely to have the greatest cost impact on newly constructed and altered facilities.  The Board then calculated the costs of applying these ten requirements to the new construction and alteration of four representative facility groups:  office buildings; hotels; hospitals and nursing homes; and public (government) housing.  These four facility groups were selected based on the assumption that they would most likely incur relatively higher costs for the ten selected requirements as compared to other facilities.  Using the foregoing methodology, the Board's final regulatory assessment estimated that the aggregate national cost of the ten selected final revised guidelines for newly constructed or altered office buildings, hotels, hospitals and nursing homes, and public housing ranged from $12.6 million (using IBC 2000 & 2003 as the "lower bound" baseline) to $26.7 million (using an "upper bound" baseline of the 1991 ADAAG) annually.

In the ANPRM, the Department stated that it expected to "adopt" the Access Board's final regulatory assessment for the 2004 ADAAG as its assessment of the cost impact that the proposed standards would have on newly constructed and altered facilities.  At the same time, however, the Department recognized that its assessment of the costs for newly constructed and altered facilities would have to be broader than that of the Board.  First, the Department's assessment would have to include the costs associated with the supplemental guidelines, which, because they had been adopted by the Board in earlier rulemaking initiatives, had not been included in the Board's final regulatory assessment of the 2004 ADAAG.  In addition, as the Department noted in the ANPRM, the unit costs estimated by the Board, though they might serve as a starting point, would nonetheless have to be supplemented with indirect costs, balanced with reduced costs, and then spread out over the 40-year lifecycle of the regulations.  Finally, because the Department was undertaking a comprehensive benefit-cost analysis, the Department--unlike the Board--would have to include an assessment of benefits for each requirement.

In response to the ANPRM, several commenters representing industry groups urged the Department not to simply "adopt" the Board's assessment but, instead, to conduct its own assessment of the benefits and costs of the proposed standards for newly constructed and altered facilities.  Questioning the accuracy of the sampling approach employed in the Board's assessment, as well as its decision not to estimate unit costs for requirements it had concluded would impose "reduced cost" or "no or minimal cost," these commenters urged the Department to conduct a comprehensive benefit-cost analysis that would assess the benefits and costs of all requirements as applied to all types of facilities. 

As a practical matter, the RIA does indeed follow the comprehensive benefit-cost approach suggested by these commenters. The Department had long planned to assess the incremental impact of revised and supplemental requirements at existing facilities on a per requirement and per facility basis with respect to barrier removal.  Using a different methodology for newly constructed and altered facilities would have made it impossible to "roll up" the benefits and costs of the proposed regulations for each requirement, each facility group, and for the rule as a whole.  The Department concluded that the most sensible approach would be to use the same methodology throughout its initial regulatory assessment.  Thus, the Department did not "adopt" the Access Board's final regulatory assessment for the 2004 ADAAG, but, rather, conducted its own assessment of the proposed title II and title III regulations. 

Moreover, while the Department suggested in the ANPRM that it might use the Board's unit cost estimates as a starting point for newly constructed and altered facilities, the RIA does not, in fact, rely on the Access Board's cost figures.  Instead, the RIA uses detailed cost estimates for each requirement as provided by an independent professional cost estimator.  See RIA §§ 4.1.3-4.1.6 & App. 3-H.  These unit cost estimates were derived using standard industry practices and published sources for construction costs.  Low, middle, and high unit cost estimates were developed for each requirement and separately applied to new construction, alterations and barrier removal.  As with all data used in the RIA, the Department invites the public to comment on its unit cost estimates and to provide, where appropriate, any supporting information that might be necessary for the Department to properly consider the comment.  Because this is an initial RIA, it will be followed by a final regulatory assessment when the Department publishes a final rule.  The Department will carefully consider all comments relating to the initial RIA during the development of the final rules and final regulatory assessment.

Categorization of Requirements

The Department's RIA assesses the incremental benefits and costs of 110 proposed requirements (or series of closely-related requirements).  For ease of reference, the RIA assigns a number to each proposed requirement.  See RIA, Tbl. 1 & App. 2.  The RIA's requirements largely follow the requirement categories developed by the Access Board in its final regulatory assessment for the 2004 ADAAG.  The Department's categorization of requirements, however, does not track perfectly with the Board's final regulatory assessment for two primary reasons.  First, the two assessments use different primary baselines.  In the Access Board's final regulatory assessment, the 1991 ADAAG served as one of the two primary baselines, whereas the RIA employs the Department's 1991 Standards as the primary baseline.  Second, the Board's final regulatory assessment only directly calculated the cost impact of a limited subset of revised guidelines as applied to four representative newly constructed or altered facility groups. For situations in which either of these considerations altered the incremental substantive or monetary impact of a proposed requirement, the RIA categorizes that requirement differently than the Access Board.  See RIA § 2.2.

Requirements in the RIA are categorized as either "supplemental" or "revised" requirements.  Supplemental requirements represent proposed requirements that have no scoping or technical counterpart in the 1991 Standards.  There are about 44 requirements in the RIA categorized as "supplemental."    See RIA, App. 2 (Req. ## 67-110) & App. 8 (Matrix of Changes).  For the most part, these supplemental requirements come from the supplemental guidelines promulgated by the Access Board for judicial, detention, and correctional facilities (1998), play areas (2000), and recreational facilities (2002).  The Department's title II and title III NPRMs also independently propose a handful of new regulatory requirements applicable to sports stadiums, post-secondary school multi-story dormitory facilities, accessible prison cells, and social service establishments.  See RIA, App. 2 (Req. ## 106 -110) & App. 8 (Matrix of Changes).  In general, supplemental requirements apply to features or elements that are typically found only in specific types of facilities such as courthouses, jails, recreational boating and fishing facilities, golf courses, amusement rides, and playgrounds.  However, a few supplemental requirements (i.e., requirements relating to exercise facilities, swimming pools and play areas) apply to features or elements found in a broader range of facility types. Supplemental requirements in the RIA are assigned requirement numbers 67-110.  See RIA, Apps. 2 & 8.    

The RIA also identifies about 66 proposed requirements as "revised" requirements.  Unlike supplemental requirements, revised requirements apply to features or elements that are currently subject to (or specifically exempted from) scoping or technical provisions in the 1991 Standards.  For the most part, revised requirements apply to elements that are found in a wide range of commonly used facility types, such as restaurants, retail stores, schools, hospitals, and office buildings.  Also categorized as revised requirements in the RIA are requirements applicable to common building elements (such as windows) and commonly used facility types (such as residential dwelling units) that have long been subject to specific accessibility requirements, either through the Uniform Federal Accessibility Standards ("UFAS"), other Federal accessibility standards (such as the Fair Housing Act or Section 504 of the Rehabilitation Act), or the International Building Code (IBC).  Each of the "revised" requirements in the RIA was adopted by the Board in 2004 and is, therefore, also described in the final regulatory assessment accompanying the 2004 ADAAG.  "Revised" requirements in the RIA encompass requirement numbers 1 through 66.  See RIA, Apps. 2 & 8. 

For analytical purposes, the RIA also further divides "revised" requirements into two subcategories:  "more stringent" and "less stringent" requirements.  Generally speaking, more stringent requirements are requirements that have been modified to mandate greater accessibility as compared to the 1991 Standards.  For the most part, the RIA's "more stringent" revised requirements generally correspond to requirements identified by the Board as "no or minimal cost" or "increased cost" requirements in its final regulatory assessment for the 2004 ADAAG.  These differences in terminology arise out of the dissimilar methodologies underlying the respective regulatory assessments--namely, while the Board's final regulatory assessment assessed only the costs of the revised guidelines, the Department's RIA includes both incremental benefit and the cost calculations for each proposed requirement. "More stringent" requirements in the RIA include the following requirement numbers:  2-11; 14-16; 19-24; 27-29; 32; 35-37; 40-42; 45-46; 48-49; 51-53; and, 58-62.  See RIA, App. 8.  Less stringent revised requirements, on the other hand, represent requirements that have been relaxed relative to the 1991 Standards.  Requirements categorized as "less stringent" in the RIA generally equate to "reduced cost" requirements in the Access Board's final regulatory assessment.  In the RIA, less stringent revised requirements are represented by the following requirement numbers: 1; 12-13; 17-18; 25-26; 30-31; 33-34; 38-39; 43-44; 47; 50; 54-57; and, 63-66.  See RIA, App. 8.   

Facilities - Categorization by Group

The Department appreciates the need for a facility categorization scheme that reflects, to the greatest extent possible, the wide range of facilities covered by titles II and III of the ADA.  Accordingly, rather than simply relying on the twelve facility categories enumerated in the ADA, the RIA features over 65 different facility groups.  See RIA, Tbl. 2 & App. 3-A to 3-C.  All public (title II) and private (title III) facilities are assigned separate facility groups.  Additionally, public and private facilities are also grouped according to general similarities in size, in underlying economic characteristics (including the responsiveness of average customers to changes in price), or both.  Some of the resulting facility groups represent single-purpose facilities (i.e., elementary schools or hospitals), while other groups include classes of facilities (i.e., single level stores).  A few facilities--namely, swimming pools and parking garages-- represent both individual facility groups and elements in larger facilities (such as hotels).

While the range of facility groups in the RIA is thus broad, it is not limitless.  No regulatory assessment can account for every nuance across all industries and facility types nationwide.  The Department has nonetheless endeavored to craft as many facility groups as necessary to properly estimate the incremental benefits and costs of the proposed regulations, as well as to afford stakeholders a meaningful opportunity to assess the regulations in terms of their own particular circumstances.  For example, due to the wide variations among transient lodging facilities and the fact that several revised requirements are directly related to the number of rooms in such facilities, places of lodging have been divided into three size-specific groups: "motels,"  "inns," and "hotels." Additionally, both because most of the supplemental requirements relate to specific types of recreation facilities and because such facilities vary greatly by size and features, the RIA includes distinct categories for each of the following public and private recreation-related facility groups:  amusement parks; exercise facilities and health spas; aquatic centers; bowling alleys; golf courses; recreational boating facilities; fishing piers and platforms; miniature golf courses; and shooting facilities.  The RIA does not, however, differentiate restaurants and other eating establishments into multiple facility groups as suggested by some commenters.  Since over 75% of restaurants are owned by small businesses, their respective sizes, features, and elements are relatively homogenous.  See RIA, Ch. 6, Tbl. 17.  Thus, for purposes of the RIA, restaurants and other eating establishments are collectively assigned to a unitary facility group.  The Department, however, welcomes public comment on these and other facility groups used in the initial RIA and will consider such comments carefully when preparing the final RIA.

Facilities - Estimation of Number of Elements Per Facility

The primary building blocks for the RIA's economic analysis are the estimated number of elements in each facility. Elements represent the architectural features, amenities, or spaces that are subject to revised or supplemental proposed requirements.  As noted previously, it was not feasible for the Department to conduct a nationwide survey of all buildings and facilities.  Nor are there published sources that document the number and types of elements--as defined in the RIA--in all facilities across the country.  Estimating the number of elements per facility thus required the development of specifications for each element, as well as a methodology for counting the number of elements in each facility.  These estimates were initially developed by Department architects and HDR and then verified (or, as needed, modified) by a panel of experts with broad experience in architecture, code consulting, and cost estimation across a wide spectrum of facilities.  See RIA §§ 3.1, 4.1.2 & Apps. 3-D, 3-E, & 7.

The end result of this element estimation process is a constructed element count for all types of ADA-covered facilities nationwide.  Within each facility group, the RIA assumes a "typical" or average facility for each facility group that applies to all facilities in that group. See RIA, App. 3-C.  Examples of assumptions about facility size include square footage, number of stories or elevators, and seating capacity.  For each typical facility, in turn, the RIA assumes a specified set of elements.  See RIA, App. 3-E.  As a general rule, larger facilities have more elements, and smaller facilities have fewer elements.  However, the specific number and type of elements in a typical facility are determined by the size and nature of the facility.  For example, the typical restaurant is assumed to potentially have up to the following number of elements subject to change:  valet parking garages (1); passenger loading zones (1); parking spaces (1); urinals (1); water closet clearances in single-user toilet rooms (2); side reach (3); sales and service counters (1); limited access spaces and machinery spaces (1); detectable warnings (1); and small play area (1).  See RIA, App. 3-E1.

In actuality, of course, not every facility will share precisely the same set of elements that are assumed for the typical facility in the facility group.  For example, even though it is estimated that the typical restaurant facility has one passenger loading zone, many restaurants are located on streets, in shopping malls, or other interior spaces where passenger loading zone requirements do not apply.  The RIA takes this uncertainty factor into account by incorporating likelihood values into the model.  That is, each element is assigned a range of values (low, medium, and high) representing the likelihood that the element is both located in the typical facility and subject to change in order to bring it into compliance with applicable revised or supplemental requirements.  See RIA §§ 3.1, 4.1.2 & Apps. 3-F, 3-G.  Continuing with the restaurant example, the "most likely" value for passenger loading zones being located at a particular facility and requiring change is assumed to be 10%, with high and low values equal to plus or minus 5% respectively.  See RIA, App. 3-G. Thus, by quantifying and incorporating likelihoods into the model with respect to facility element counts (and other estimated cost and benefit values), the RIA more realistically addresses some of the inherent uncertainties underlying benefit-cost analyses.  See RIA §§ 3.3, 4.3.1 (discussing "Risk Analysis" approach) & App. 6 (RAP Primer). 

Facilities - Application of Model to Newly Constructed and Existing Facilities

The universe of facilities required to comply with the Department's proposed standards will be divided into mutually exclusive categories--facilities that are "newly constructed" after the effective date, and facilities that are already "existing" as of the effective date.  Facilities constructed after the effective date of the regulations will be required to build in conformance with the requirements governing new construction.  Elements and spaces within existing facilities will be subject to the proposed standards through either alterations or barrier removal requirements.  In the RIA, each of these types of construction is modeled separately with respect to each facility group (and each requirement) so that stakeholders will be able to better assess the impact of the proposed regulations on their own particular facilities or circumstances.

Application of the RIA cost model to new construction is relatively straightforward.  The number of new facilities constructed each year after the effective date of the regulations (up to the 15th year) is generally based on published industry and sector-specific annual growth rates.  See RIA §§ 3.1, 4.1.1 & App. 3-B.  In simplified form, the total incremental cost for a particular facility group in a given year is calculated by multiplying the number of newly constructed facilities for that group for the year by the total number of elements across all newly constructed facilities in that group and the unit cost per element (that includes both initial and recurring costs).  As a general rule, new construction costs are typically lower than the costs for other types of construction.  Indeed, many proposed requirements are expected to have zero costs for new construction either because the cost of the element is negligible, or because it is presumed that architects can "design around" the new requirement in the planning stages with no appreciable increase in design or construction costs.    

For existing facilities, compliance with the proposed standards may come in the form of either alterations or barrier removal.  The alterations requirement is only triggered when an entity voluntarily undertakes an alteration project, and, even then, generally applies only to the particular elements undergoing alteration.  (Alterations affecting "primary function areas" are also required, absent certain circumstances, to ensure that the path of travel to the altered area is accessible to persons with disabilities.)  Moreover, not all existing facilities would be altered within the presumed 15-year lifespan of the proposed regulations.  The RIA thus incorporates a historically derived alterations schedule for each facility group based on published data.  See RIA § 3.4 & App. 3-B.  Based on this alterations schedule, the total incremental alterations cost for a particular facility group are then calculated using the same basic formula as described above for new construction costs.  Alterations costs reflect only the incremental costs necessary to bring the affected element(s) into compliance and exclude costs otherwise attributable to other planned aspects of the alteration.  Overall, alterations costs vary greatly by facility group, with some facilities experiencing minimal alterations costs (or even cost savings) under the proposed regulations (e.g., stadiums, convention centers, airport terminals, depots, ski facilities, bowling alleys, fishing piers, and public amusement parks), and other facilities projected to incur relatively higher alterations costs (e.g., single-level stores, indoor service establishments, offices of health care providers, office buildings, and courthouses).  See Initial Regulatory Impact Analysis -- Supplemental Results ("Supplemental Results"), pp. 14-147.  The variability in alterations costs are largely driven by the mix of affected elements in each respective facility group.

Barrier removal, by contrast, is a continuing obligation that applies to all public areas of existing title III-covered facilities.  For this reason, all elements in these existing facilities-- irrespective of compliance with the current 1991 Standards--potentially would be required to satisfy applicable supplemental or revised proposed requirements to the extent barrier removal was readily achievable.  Factors in the barrier removal calculus include whether elements are subject to more stringent revised requirements and, thereby, potentially exempt from barrier removal under the Department's safe harbor proposal; whether elements are subject to supplemental requirements for which safe harbor protection does not apply; when the facility was originally constructed; whether, or to what extent, elements have been altered; and whether removal of architectural barriers is readily achievable under the 1991 Standards or proposed requirements respectively.

Taking all of the foregoing factors into consideration makes barrier removal cost calculations potentially more complex (or, put another way, more variable-driven) as compared to costs for other types of construction.  Figure 1 in the RIA fully illustrates the various conditions under which particular elements in an existing facility may become compliant and whether the costs associated with such compliance is assessed under barrier removal or alterations.  As a practical matter, however, barrier removal cost calculations in the RIA can be distilled down to two essential considerations.  First, the RIA assumes that elements in existing facilities subject to supplemental requirements may potentially incur barrier removal costs.  Since the Department's proposed safe harbor is conditioned on compliance with the 1991 Standards, elements covered by supplemental requirements--which, by definition, have no counterpart in the 1991 Standards--are necessarily ineligible for safe harbor protection.  Second, with respect to revised requirements, the RIA presumes no barrier removal costs will be incurred by virtue of the safe harbor provision.  (Instead, modifications to existing elements subject to revised requirements proceed on the schedule and are costed accordingly.)

The RIA presents the overall results for barrier removal under two scenarios--a comparison of total net present value ("NPV") under "safe harbor" and "no safe harbor" conditions, and a comparison of varying assumptions about readily achievable barrier removal rates (i.e., 0%, 50%, and 100%).  See RIA, Figures ES-3 & ES-4.  (Total barrier removal costs are also presented for each respective facility group under the "Safe Harbor" scenario in the Supplemental Results.)  In sum, many title III-covered facilities are expected to incur few--if any--costs for barrier removal due to the Department's proposed safe harbor provision.  Indeed, when taking safe harbor into account, one-half of the 38 facility groups comprised of title III-covered (private) facilities are projected to incur no barrier removal costs.  See Supplemental Results, pp. 14-147.  Such facility groups include:  motels; restaurants; movie theaters; single-level stores; shopping malls; museums and libraries; day care centers; and homeless shelters. Other facilities, on the other hand, are expected to incur barrier removal costs under the proposed regulations due to the presence of elements affected by supplemental requirements.  For such existing facilities, barrier removal costs typically run higher than new construction costs because: (1) retrofitting existing buildings or facilities is often more expensive than new construction; and (2) from an economic perspective, the full cost of bringing existing elements into compliance with the proposed regulations is attributable to barrier removal whereas, for new construction, only the incremental cost differential between compliant and noncompliant elements is attributable to new construction.  See RIA § 4.1.3.  Title III-covered facility groups with expected barrier removal costs that are higher relative to their respective new construction costs include amusement parks; exercise facilities; aquatic centers; and golf courses.

Facilities - Assumption of Compliance with Current Law

In accordance with the principle that regulatory analyses should only assess the incremental benefits and costs attributable to proposed regulations, the RIA assumes that elements in existing facilities covered by the ADA are currently in compliance with applicable regulatory standards.  Indeed, if the RIA did not make this assumption, the benefits and costs of entities' noncompliance with their legal obligations would be improperly charged to the proposed regulations.

While the RIA's assumption of compliance has implications throughout the assessment, its impact is most obvious with respect to existing private (title III) facilities subject to barrier removal.  As discussed previously, the Department is proposing a safe harbor provision that would exempt elements in existing facilities that comply with the 1991 Standards from barrier removal that might otherwise be necessary to bring them into compliance with revised standards in the proposed regulations.  In this context, the RIA presumes that existing facilities have already satisfied their legal obligations by removing architectural barriers to the extent readily achievable.  Thus, any remaining barriers are those for which barrier removal has not yet been readily achievable under the 1991 Standards.  Moreover, if barrier removal to date has not been readily achievable under the current Standards (which, by definition, are less stringent than the proposed revised requirements), it is reasonable to assume that barrier removal will also remain beyond reach under more stringent revised requirements.

For existing public (title II) facilities, however, the assumption of compliance with current law plays out differently. Existing public facilities are not subject to barrier removal requirements. Instead, title II-covered public entities must ensure that their programs and services, "when viewed in their entirety," are accessible to individuals with disabilities.  Compliance with program accessibility requirements thus does not necessarily require structural modifications to existing facilities since compliance is determined on a program-wide--rather than element-by-element--basis.

For these reasons, the RIA follows the methodology outlined in the ANPRM and generally does not assess the impact of the proposed regulations on existing public facilities covered by title II. However, there are two limited circumstances in which the regulatory assessment does include existing public facilities in the economic calculus.  First, alterations to existing public facilities must still comply with the proposed regulations irrespective of program accessibility requirements.  Thus, the RIA model assumes that when an existing title II-covered facility undergoes alteration, the incremental costs and benefits of that alteration are included in the regulatory assessment.  Second, the RIA takes into account program access when calculating the estimated incremental impact of the proposed regulations with respect to supplemental requirements relating to existing swimming pools, saunas and steam rooms, and play areas.  The RIA includes program accessibility in the regulatory calculus in the context of these three sets of requirements for several reasons.  Even in the context of program accessibility, compliance with these supplemental requirements would undoubtedly require some structural modifications unless the facilities that compose the program were already--pursuant to program accessibility or otherwise--accessible in the same manner and to the same extent as required by the proposed standards.  Moreover, the Department is proposing certain regulatory exemptions and exceptions that exclusively apply to existing title II-covered facilities with swimming pools, saunas and steam rooms, or play areas.

The Department's statement in the ANPRM that it did not intend to include existing title II-covered public facilities in the assessment generated several objections by commenters.  In summary, these commenters asserted that existing public facilities should be included in the regulatory assessment since they would be affected by the proposed standards in various circumstances, including voluntary efforts to improve access, determinations that compliance with program accessibility requirements could only be met with structural changes or litigation.

As stated previously, however, the purpose of the RIA is to measure the incremental benefits and costs of the Department's proposed regulations.  Because the program accessibility provisions in title II require public entities to ensure access to programs, rather than facilities, the necessity for structural modifications cannot be assumed.[2] (By comparison, the obligation to remove structural barriers in existing private facilities is both mandatory and amenable to assessment on an element-by-element basis.)  Moreover, as with existing private facilities, public facilities newly constructed or altered since the effective date of the 1991 Standards should already be fully or largely accessible, and older facilities--those built before 1993--have been required to meet the program accessibility requirements for at least 15 years, if not longer.  It is thus reasonable to assume that if structural modifications were necessary to provide program access, they likely would have been implemented by now.

Benefits - Public Comments Relating to the Measurement of Benefits

The Department received many public comments with suggestions about how the RIA should measure the benefits of the proposed standards to individuals with disabilities.  With the exception of those commenters who expressed the view that any form of economic analysis is inappropriate for regulations implementing a civil rights statute, commenters were unanimous that the assessment should balance costs against a comprehensive assessment of benefits, both economic and social.  Generally speaking, commenters also recognized that quantifying benefits would be a difficult, if not impossible task, since the paucity of hard data on the economic benefits of accessibility would require the Department to generate such data from scratch.

Most comments relating to the assessment of benefits tended to be global in nature.  That is, rather than suggesting methods for estimating the incremental benefits of the proposed regulations, the majority of proposals appeared better suited to a comprehensive assessment of the overall societal benefits of accessibility itself.  For example, commenters representing disability groups recommended that the Department adopt a process of benefit-based analysis recommended to the President by the National Council on Disability (NCD) in its report entitled "National Disability Policy: A Progress Report, December 2002-December 2003." Recognizing the need for "vastly more data" on the effects of societal decisions on people with disabilities, these commenters urged the Department to analyze the long-term benefits of the proposed regulations for people with disabilities, as well as economic activities foregone by persons with disabilities due to inaccessibility.  As one commenter noted:  "An individual with a disability able to access the local aquatic center will be able to seek physical activity and recreation opportunities that promote healthy living and wellness, reduce the risk for disease and declining health, seek additional opportunities for community participation including employment and thereby reduce reliance on governmental subsidies for housing, welfare or health care."

Other commenters representing disability groups recognized that, while certain short-term benefits could be measured, gauging the more enduring or meaningful benefits of the changes represented by the proposed regulations for people with disabilities and for society as a whole would be very difficult.  For example, determining the incremental impact that one change--or even all of the changes--might have on the earning power of people with disabilities would "require a much more complex exercise than construction cost estimating."  Other unquantifiable benefits noted by commenters included the extent to which the incremental changes reflected in the proposed regulations might lower the liability exposure faced by facilities by making accessible elements and spaces safer for persons with disabilities.

Commenters representing industry groups suggested that the RIA assess the benefits of accessibility on an element-by-element basis in order to establish a "breakeven" value for each proposed requirement--that is, how much benefit an accessible element would need to provide to be worth the cost of making it accessible.  One commenter representing the design and construction industry described this approach as measuring "performance outcomes" (i.e., the quantifiable benefits and costs conferred by each proposed requirement), as compared to other types of analysis that measure "social outcomes" (i.e., the overall impact of the proposed requirement on society).  This comment suggested that "cost effectiveness analyses" focus on quantifiable performance outcomes, while "cost utility analyses" focus on qualitatively describing the range of social benefits and costs.  In the RIA, the Department is doing both-- quantifying the incremental benefits and costs of each proposed requirement to the extent they can be quantified, and, to the extent they cannot, describing the unquantifiable benefits and costs in qualitative terms.

Several commenters representing disability groups or industry groups suggested that the practical effect of accessibility requirements is to redistribute economic resources from society as a whole to the "under served" population of individuals with disabilities.  Commenters representing disability groups hailed the redistribution as an obvious social good, asserting that civil rights regulations need not confer benefits on "society as a whole" to be worthwhile.  By contrast, commenters representing industry groups questioned whether such redistribution was cost-efficient. These commenters referred the Department to Part D of OMB Circular A-4 ("Distributional Effects"), which applies when the benefits and costs of a regulation are unevenly distributed throughout the U.S. population or economy. Distributional effects may be imbalanced for different industrial sectors or regions of the country, or, as urged here, for different subpopulations of people.  As OMB Circular A-4 puts it, the uneven distribution of regulatory impacts occurs when "[t]hose who bear the costs of [the] regulation and those who enjoy its benefits . . . are not the same people."  These commenters urged the Department to recognize that the proposed regulations would have uneven distributional effects because, in their view, those who will purportedly bear all the costs of compliance (facility owners and operators) and those who will enjoy its benefits (people with disabilities) are not the same groups.

From the Department's perspective, however, the redistribution analogy is inapposite.  Accessibility requirements do not represent a transfer of resources from one group of people to another, but, rather, a dedication of shared resources to a particular end.  In contrast to the types of subpopulations mentioned in OMB Circular A-4 (i.e., race, sex, or income level), disability is not a fixed or even relatively static category; rather, it is inherent in the human condition.  The vast majority of individuals who are fortunate enough to reach an advanced age will benefit personally from an accessible environment.  Business owners and people with disabilities are not discrete subpopulations--just as people with disabilities own businesses, many business owners have or will acquire a disability during their lifetime.  Moreover, while the direct costs of compliance with the proposed standards may be incurred initially by businesses, as commenters representing industry groups have repeatedly stated, such costs eventually may be passed along to consumers.  In other words, all members of society will pay the price for accessibility, just as all will benefit from it.  Rather than representing a transfer of resources between distinct groups of people, then, accessibility requirements represent--for all members of society, whether they will benefit from accessibility now or at some point in the future--a choice among different forms of societal benefits.

Benefits - Quantification and Monetization of User Benefits in the RIA

From an economic perspective, the value that people derive from accessibility can be divided into three categories:  "use value" (the value that people derive from using accessible facilities), "option value" (the value that people with and without disabilities derive from the opportunity to obtain the benefit of accessible facilities in the future) and "existence value" (the value that people with and without disabilities derive from the simple existence of accessible facilities including the fulfillment of constitutional guarantees of equal protection and nondiscrimination).  The RIA, however, only quantifies and monetizes the incremental benefits to users (i.e., persons with disabilities) conferred by changes in accessibility due to the proposed regulations.  This is largely due to data constraints.  The overall benefits of the proposed regulations will be experienced by nearly all members of society to a greater or lesser extent during the projected 40-year lifecycle of facilities affected by these regulations.  However, quantification of these benefits is beyond the scope of the Department's regulatory assessment, and, likely, any regulatory assessment.  Instead, the RIA is necessarily limited to assessing the value of specific types of benefits that can be quantified and assigned monetary values (i.e., user benefits) for a demographically defined population of people (i.e., persons with disabilities).  In this sense, the regulatory assessment must be considered conservative since it almost certainly understates the overall value of the proposed regulations to society.

The RIA quantifies and monetizes user benefits in two ways.  First, an expert panel developed estimates of the amount of time persons with disabilities can be expected to save time either gaining access to a facility (e.g., a retail store), waiting to use a particular amenity in that facility (e.g., a restroom), or using an amenity in the facility (e.g., an ATM inside the store) as a result of the proposed regulations.  See RIA §§ 3.2.2, 4.2.6 & Apps. 4-H, 4-K, 4-L, and 4-N.  Second, for proposed requirements--primarily, supplemental requirements--that can be expected to create new users who previously were unable to visit a facility (e.g., fishing piers) or to use a facility amenity independently (e.g., hotel swimming pools), the assessment quantifies the value of the new uses generated by the change in accessibility. See RIA § 3.2.3 & App. 4-I.  Each of these components of user benefits is then monetized using an appropriate "value of time"--namely, an expression of a user's willingness to pay for changes at the facility.  In keeping with common economic assumptions, user benefits associated with accessibility changes are monetized based on the value of the user's time.  See RIA §§ 3.2, 4.2.5 & App. 4-J.

The benefits model in the RIA also places a "premium" on the value of certain types of time savings.  The RIA describes the theory and mechanics of this approach in greater detail.  See RIA § 4.2.5 & App. 4-J.  Briefly stated, the assessment assumes that individuals would be willing to pay more for time saved gaining access to a facility due to improved accessibility than their respective typical uses of the same amount of time.  This presumption derives from studies in the transportation industry concluding that the inherent discomfort of having to wait (as compared to the satisfaction of feeling like one is at least moving in the direction one wants to go) leads people waiting at a bus stop to prefer to have the bus arrive sooner, even if it means that the bus ride itself will take longer (so that the net travel time is the same).  Essentially, people experience the time they spend waiting for the bus as a more negative experience--by a factor of two to one--as compared to the time they spend riding the bus and, consequently, "value" decreasing the time spent waiting more than they would an equivalent amount of bus time.  In the RIA, this premium is applied, as applicable, to the incremental time savings benefit afforded by each revised or supplemental requirement.

In the end, the approach the Department has taken with respect to the assessment of benefits in the RIA is closest to the proposals of commenters representing industry groups.  By calculating the incremental benefits (and costs) for each supplemental and revised requirement, the assessment generates a benefit-cost ratio for each such requirement. Although this approach has allowed the Department to gauge the incremental cost-effectiveness of the change represented by each revised or supplemental requirement as applied to a particular element, it should be understood that it is also fundamentally different from gauging the absolute cost-effectiveness of requiring a given element to be accessible.  Most of the inherent value of an accessible element, as with accessibility generally, derives not from the incremental changes represented by the proposed standards, but from the fact that the element is required to be accessible at all.

Finally, not all of the revised requirements will confer increased benefits on persons with disabilities.  The "less stringent" revised requirements generally reduce both benefits and costs, though such reductions may not be distributed equally.  As a general matter, requirements have been made less stringent to clarify the meaning of the current requirement, or to provide an exception that takes into account special circumstances in specific facilities.  For less stringent requirements that propose reductions in scoping, these revisions were typically based on the Access Board's determination that demand for the affected accessibility feature or communication device was not high enough to warrant the current numerical requirements.  For purposes of the RIA, when less stringent revised requirements confer lower benefits relative to the current requirements, these reduced benefits have been assessed only with respect to new construction and alterations.  Elements in existing facilities subject to less stringent requirements are assumed to be compliant already, either with current (more stringent) requirements or  revised (less stringent) requirements.  Facility owners would have neither a legal obligation nor a financial incentive to undergo barrier removal for such elements in order to "comply" with the revised standard.  The RIA thus assumes that reductions in benefits due to less stringent revised requirements will not be realized for elements in existing facilities unless the affected elements are altered.

Benefits - Nature and Significance of Unquantified Benefits 

In addition to the foregoing monetized user benefits, the RIA acknowledges that the proposed regulations would, if promulgated in final form, undoubtedly confer significant and important benefits on society that defy easy quantification or monetization.  These benefits include the option and existence values discussed previously.  Other benefits would also likely accrue to businesses through reduced administrative costs (from harmonization of the 2004 ADAAG with model codes) or increased worker productivity (due to greater workplace accessibility).  The regulatory assessment discusses these types of benefits in qualitative, rather than quantitative, terms.  See RIA § 5.4.

Perhaps the most significant unquantified benefit is the myriad ways in which the proposed standards--to the extent they make the built environment more accessible--would improve the lives of many persons with disabilities.  Even on an incremental level, the beneficial domino effect of increased access to all types of facilities, for each individual and, ultimately, for society as a whole, simply cannot be measured, much less reduced to monetary terms.  An example related by one commenter referred to the way in which the proposed regulations would enable many individuals with disabilities to begin independently accessing various types of recreational facilities for the first time.  This commenter observed how "[r]egular involvement and participation in recreation, social, and leisure activities plays a significant role in living and maintaining a healthy lifestyle," and ensures that people "remain physically active, develop social skills, and develop the skills necessary to enjoy lifelong leisure activities."  Among the many collateral benefits of access to recreational opportunities are the "prevention of obesity, [a] decrease of secondary conditions, improved social and problem solving skills, promotion of physical and emotional health and decreased likelihood of being hospitalized for another illness," not to mention "increased independent living skills and preparation for employment."

Unquantified benefits from the proposed regulations, moreover, are not limited to those accruing from the increased accessibility of recreational facilities.  The revised requirements would increase accessibility throughout the entire range of public and private facility groups.  For example, one commenter cited a study published in a recent issue of the Journal of Consumer Affairs presenting the perspectives of people with disabilities regarding the effectiveness of the ADA.  Based on a national sample of one thousand noninstitutionalized individuals with disabilities, the study found that respondents who interacted more frequently with the marketplace, or even simply perceived the marketplace as more accessible, were more satisfied with life.  According to this comment, study authors Carol Kaufman-Scarborough and Stacey Menzel Baker stated that their finding "indicates the value behind efforts designed to empower consumers with disabilities by offering services that assist them . . . and by creating environments that enable them to experience full participation in society."  Increased accessibility of the marketplace as a whole, which can be expected to heighten facility use across a wide range of facility groups, will also lead to greater benefits over time.  A commenter representing a State government echoed this theme, citing potentially increased usage of public recreation areas and greater participation in the democratic process.

Additionally, the number of Americans with disabilities is expected to continue increasing over time.  As many commenters pointed out, the proportion of the U.S. population that has a disability not only has been growing steadily over the last forty years, but also is projected to continue growing during the 40-year lifecycle of the regulations.  Data provided by the Disability Statistics Center at the University of California at San Francisco demonstrates that the number of adults who use wheelchairs increased at a rate of 6% per year between 1969 and 1999; by 2010, it is projected that 2% of the adult population in the U.S. will use wheelchairs.  In addition to people who use wheelchairs, in 1999, 3% of adults used crutches, canes, walkers, and other mobility devices; by 2010, that number is projected to have increased to 4%.  Thus, by 2010, up to 6% of the U.S. population is projected to have mobility impairments.  Moreover, because this figure was based on data from 1999, it does not take into account the influence of the current war in Iraq.  This war is creating a new generation of young men and women with disabilities, the majority of whom are returning from war in their early twenties and can be expected to outlive the 40-year lifecycle of any building subject to these proposed regulations.  Just as the original Federal disability rights legislation -- Section 504 of the Rehabilitation Act of 1973--was enacted in direct response to the thousands of disabled war veterans returning home from Vietnam, the need to ensure an accessible built environment is now more critical than ever.

Benefits from the proposed regulations potentially would also extend to the public generally irrespective of disability status.  For some, value may be derived simply from the existence of enhanced accessibility and improved social equity brought on by the proposed regulations.  Others may take "insurance" value from the opportunity to make use of accessible features or facilities in the event they should need them in the future.  Accessible facilities also benefit individuals without disabilities.  Several commenters noted that improved accessibility features might benefit, for example, elderly persons, athletes temporarily on crutches, expectant mothers, or mail carriers using hand carts to deliver large packages.  Moreover, because individuals tend to patronize facilities--especially places of public accommodation like hotels and restaurants--in pairs or groups, the benefits of accessibility also extend to the partners, companions, friends, family members, and personal assistants of people with disabilities.  Finally, although requirements that apply to existing facilities pursuant to the barrier removal requirement are not primarily intended to benefit employees, employees with disabilities will certainly benefit from the accessibility of such features, which, given the importance of employment to the economic vitality of an individual, their family, and society as a whole, magnifies the benefits of accessibility throughout the economy.

Lastly, businesses--as well as State and local governments--would also likely experience benefits from the proposed regulations in ways that are not quantified in the RIA.  Increased harmonization of the revised ADA Standards with model codes and consensus standards will yield substantial benefits to businesses, architects, and State and local governments by eliminating confusion and reducing administrative costs.[3] Harmonization will also make it easier for code-setting governmental entities to have their respective State or local codes certified as meeting or exceeding Federal standards.  Businesses may also experience increased workforce efficiency and productivity as a result of accessibility changes in the proposed regulations.  For example, one commenter representing the design and construction industry pointed out that greater independence for users of facilities confers a "productive" benefit for businesses, whose staff can be redirected from providing assistance to customers with disabilities to potentially more economically rewarding tasks.

Analytical Scenarios - Safe Harbor

The most significant of the regulatory alternatives proposed by the Department is the "safe harbor" for certain existing title III-covered facilities and elements.  As noted previously, the safe harbor proposal exempts covered facilities from barrier removal obligations that might otherwise arise under the proposed regulations so long as the elements therein are in compliance with the 1991 Standards.  The Department has proposed this safe harbor to mitigate the impact of the proposed regulations on existing private facilities.

The RIA results indeed reflect the significant impact of the safe harbor proposal.  In order to both assist the Department with its consideration of the safe harbor provision and inform the public of the benefits and costs of its adoption, the RIA compares the total NPV for "safe harbor" versus "no safe harbor" scenarios.  See RIA, Figures ES-3 & 13.  These comparative scenarios use the 1991 Standards as the primary baseline and assume barrier removal is readily achievable for 50% of the elements in existing facilities.  Based on these assumptions, the RIA shows that there is most likely a $4.3 billion difference in total NPV between the "safe harbor" scenario ($7.6 billion) and the "no safe harbor" scenario ($3.3 billion).

Analytical Scenarios - Barrier Removal

By statute, an action to remove barriers is considered "readily achievable" if, for a particular entity, it is "easily accomplishable and able to be carried out without much difficulty or expense." 42 U.S.C. § 12182(b)(2)(A)(iv).  In practice, what is readily achievable for any given entity with respect to a given element must be determined on a case-by-case basis, and has no monetary or other absolute parameters--it is specific to the individual facility and to the particular time, place, and context in which that facility operates.  The Department's current title III regulations provide a list of factors that should be considered in determining whether an action is readily achievable.  Only one of those factors--the nature and cost of the action--relates to the element itself.  All of the other factors specifically relate to the business entity, including the impact of the action on the operation of the site; the overall financial resources of the entity and any parent corporation; the type of operation of the entity or parent corporation (including the composition, structure, and functions of the relevant workforce); the geographic, administrative and fiscal relationships between the facility, entity, and parent company; and the effect of the action on any legitimate safety requirements that may be necessary for safe operation.

Recognizing the infeasibility of conducting an empirical assessment of the individualized barrier removal efforts by facility owners and operators nationwide, the Department proposed in the ANPRM to develop a computer simulation model that would assess the statistical probability that existing facilities would be required to remove barriers in order to comply with supplemental or revised requirements.  Several commenters expressed concern that the lack of reliable data would make the results of a simulation model useless.  Other commenters suggested that the same indefinite parameters that make compliance with the barrier removal requirement difficult would also complicate any attempt to accurately calculate the likelihood that compliance would be required.  In addition, these commenters stated that modeling readily achievable barrier removal as a function of the financial resources of an entity would underestimate the costs of compliance since entities, faced with an ambiguous definition of "readily achievable," purportedly often spend more on barrier removal efforts than required by the ADA.  Rather than using definite parameters to evaluate an indefinite requirement, these commenters proposed that the Department simply make an honest attempt to quantify the costs of compliance and to describe the distributional impacts of the rule across individuals and industries.

The Department agrees that the lack of reliable data on existing facilities' barrier removal efforts would render any statistical analysis too indefinite to be of value.  Therefore, rather than basing calculations of total incremental benefits and costs on potentially arbitrary assumptions about whether (or to what extent) elements at existing facilities have undergone barrier removal, the RIA takes a more practical approach.  First, with respect to existing elements subject to supplemental requirements, the RIA calculates an expected total NPV based on the assumption that barrier removal would be readily achievable for every element (100%) in a manner that is fully compliant with the new standards.  Second, the RIA then calculates total NPV under two other compliance scenarios (0% and 50%) to show how varying barrier removal rates impact the overall results.  Taken together, these three barrier removal scenarios reflect the range of probabilities of barrier removal obligations that existing facilities would have under the proposed regulations.  Presenting the data this way enables the facility owner who could potentially incur the costs of compliance, as well as the individual with a disability who could potentially benefit from that compliance, to gauge the impact that the proposed standards might have on a particular facility by selecting the scenarios that most closely match the level of compliance and resources of the covered entity.

Primary Baseline

The 1991 Standards serve as the primary baseline for the RIA because they are the only uniform set of accessibility standards that apply to every place of public accommodation, every commercial facility, and every State or local government facility in the country.  According to statistics compiled by the International Code Council (which publishes the IBC), a version of the IBC--either IBC 2000, IBC 2003 or IBC 2006--has been adopted at the State or local level in all 50 States and the District of Columbia.  Nonetheless, there is still variation among states with respect to model code adoption.  For example, because model codes such as the IBC are voluntary, public entities sometimes modify or carve out particular provisions or sections or leave adoption to the discretion of local jurisdictions.  By contrast, because the ADA is a mandatory Federal law, it applies the same standards to every facility in the country, ensuring a uniform level of accessibility--as well as a uniform means of baseline assessment--nationwide.

Because of this uniformity, the 1991 Standards baseline is the only baseline against which the incremental costs and benefits of the proposed regulations are estimated on a requirement-by-requirement and facility-by-facility basis.  The results for the primary baseline are summarized in the main RIA text and presented in full in the accompanying Supplemental Results.  It also bears noting that the primary baseline assumes that facilities subject to the 1991 Standards are not also required to comply with equivalent provisions in model codes (such as the IBC) that have been adopted as State or local building codes--even though compliance with State or local building codes necessarily is compulsory.  In other words, the primary baseline does not take into account the substantial overlap between requirements in the proposed regulations and model code provisions in the IBC. While this approach likely leads to significant overstatement of the costs (and benefits) of the proposed regulations with respect to many requirements, it also nonetheless represents the only means of uniformly assessing the incremental impact of the proposed regulations across all facilities nationwide.

Some commenters representing industry groups expressed the view that the Department should not use the 1991 Standards as a baseline because, in their view, the benefits and costs of the current requirements were not adequately measured when the requirements were first adopted in 1991.  Instead, these commenters propose that the Department assess the absolute benefits and costs of the proposed standards as measured against a zero baseline--that is, the full cost of compliance with the proposed regulations irrespective of the current level of accessibility of facilities due to the 1991 Standards.

The Department disagrees with these comments. OMB Circular A-4 is very clear that regulatory analyses should only account for those incremental benefits and costs that arise as a result of the proposed regulatory action itself.  To assess the absolute (or total) benefits and costs of compliance with the proposed regulations would improperly attribute to the proposed standards all of the benefits and costs of the 1991 Standards, thereby distorting the economic impact of the proposed regulations.  The 1991 Standards are the law of the land and facilities have been subject to the current requirements for 15 years.  Assessing the benefits and costs of the proposed standards as if the ADA had just been enacted would thus drastically overstate both the benefits and the costs of the proposed regulations.  For these reasons, the RIA uses the 1991 Standards as the primary baseline and assesses the incremental impact of the proposed standards accordingly.

Alternate Baselines

While the RIA uses the 1991 Standards as the primary baseline, the assessment nonetheless still accounts for the impact of the widespread adoption of model codes by using alternate IBC baselines for several analyses.  Due to the high degree of overlap between the IBC, the 2004 ADAAG, and the Department's proposed standards, the widespread adoption of various versions of the IBC by State and local jurisdictions means that most buildings and facilities nationwide are already being constructed or altered in compliance with many of the proposed standards.  (Indeed, one of the Access Board's goals in revising ADAAG was to harmonize these guidelines with model codes, such as the IBC, precisely because they form the basis of most State and local building codes.)  Thus, for facilities located in one of the many jurisdictions that have adopted--in whole or in part--a version of the IBC, the Department's adoption of the proposed regulations will have far less impact as compared to other facilities.

For these reasons, several commenters representing disability groups urged the Department to use the IBC, in conjunction with other accessibility standards that have been adopted by States or local governments, as the primary baseline in lieu of the 1991 Standards. Commenters representing industry groups also recognized that versions of the IBC had been adopted in many States and localities, but suggested that the Department only use the IBC as a baseline for those jurisdictions in which its provisions had actually been adopted into law by code-making authorities.

As noted in the Regulatory Framework section of the ANPRM, the Department considered following a State-by-State approach in which the relevant baseline for newly constructed and altered facilities would vary from State to State, depending on which IBC version each State or local jurisdiction had adopted.  Under this approach, the 1991 Standards would only have been used as a default baseline for jurisdictions that had not yet adopted any version of the IBC.  However, the many variations among State and local jurisdictions concerning the extent to which various IBC-related accessibility provisions (i.e., IBC Chapter 11, IBC Appendix E, and ANSI A117.1) have been adopted without revision, adopted in a modified fashion, or carved out completely, make the creation of State-by-State baselines infeasible for every supplemental and revised requirement across all facilities nationwide.  First, given these variations among States, use of State-by-State baselines would effectively require the creation of over one hundred separate baselines in order to accurately reflect which jurisdictions have adopted IBC provisions that are equivalent to each of the revised and supplemental requirements assessed in the RIA.  Moreover, State-by-State baselines would also necessarily require information concerning the precise geographical location, age, and type of occupancy of all existing facilities nationwide.  The Department, however, is not aware of any publicly available "facility census" to provide this requisite information.  Such considerations would have made State-by-State (or, as applicable, locality-by-locality) baselines both extremely time-consuming to create and likely unreliable in application.

Thus, while the RIA applies alternate baselines for three different versions of the IBC (i.e., IBC 2000, IBC 2003, and IBC 2006) to assess the overall impact of the proposed regulations, it employs a simplified approach to the creation of these baselines.  Specifically, the RIA assumes that the applicable version of the IBC applies equally to all facilities nationwide, and that relevant provisions of ANSI A117.1, IBC Chapter 11 and IBC Appendix E have been incorporated by all State and local jurisdictions.  This latter assumption is necessary because these three sources establish most of the accessibility standards that apply under the IBC.  If none of them were assumed to apply, adoption of the IBC by a jurisdiction would tell us little about the accessibility of its facilities, and, if some but not all of them were assumed to apply, predicting which provisions would apply to which facilities would be impossible.  The alternate IBC baselines in the RIA, therefore, do not present the overall results on a State-by-State basis. However, these baselines nonetheless still permit facilities to see how the impact of the proposed standards varies depending on which version of the IBC the State or local code authorities have or might adopt in the future.

The RIA presents the comparative results for the three alternate IBC baselines in summary "rolled-up" fashion that combines all proposed requirements and facility groups.  That is, for each alternate IBC baseline, the regulatory assessment provides a graphic representation (in the shape of a so-called "S-Curve") of the NPV at various likelihoods of occurrence.  See RIA, Figure ES-5 & 15.  Unlike the primary (1991 Standards) baseline, the results for each of the alternate IBC baseline scenarios are not further broken down to show the incremental benefits and costs for each requirement or facility group.  Since requirement-by-requirement and facility-by-facility results are already calculated for the primary baseline, similarly detailed analyses for each IBC baseline effectively would have amounted to conducting four separate regulatory assessments.

Moreover, to further assist stakeholders in assessing the impact of the proposed regulations, the RIA also presents several more limited analyses that assess the incremental impact of four illustrative proposed requirements against requirement-specific alternate IBC/ANSI baselines.  When constructing these four requirement-specific IBC baselines, the Department endeavored to determine (or approximate) the actual extent to which the relevant equivalent IBC provisions have been adopted by every State or local jurisdiction nationwide. The results of these analyses underscore the point that consideration of alternate requirement-specific IBC baselines on a requirement-by-requirement basis would likely lead to markedly lower incremental costs and benefits for many proposed requirements.  For example, the first scenario in the RIA uses requirement-specific IBC baselines to assess the incremental impact of the proposed revisions with respect to two proposed requirements--alterations to existing stairs and elevators--that have equivalent provisions in the "main" IBC chapters (Chapters 10 and 34) and, thus, have been adopted by virtually every State and local jurisdiction nationwide.  See RIA, Table 10.  This first scenario shows that the incremental costs for these two requirements collectively would be reduced by about $ 1.1 billion over the lifespan of the regulations when using the requirement-specific alternate IBC baselines as compared to the primary baseline (1991 Standards).  A second scenario in the RIA employs requirement-specific alternate IBC/ANSI baselines to assess the incremental impact of proposed revisions to two other requirements--relating to side reach and water closed clearances--whose corresponding IBC provisions are only incorporated by reference into the IBC (through Chapter 11 and ANSI A117.1).  See RIA, Table 11.  These incorporated provisions have not been as uniformly adopted as other IBC provisions.  Nonetheless, the incremental costs for these latter two requirements still would be reduced by about $ 660 million over the lifespan of the regulations when using requirement-specific IBC baselines as compared to the primary baseline (1991 Standards).

Regulatory Alternatives - Existing Facilities

As required by the Regulatory Flexibility Act of 1980, as amended by SBREFA, as well as Executive Order 13272, the Department has considered regulatory alternatives that would achieve the same statutory and regulatory goals but impose less cost on society.  With respect to new construction and alterations, the ADA requires the Department to adopt standards that are "consistent with" the minimum guidelines issued by the Access Board.  The Department does not have the statutory authority to modify the 2004 ADAAG.  The Department does, however, have the discretion to determine whether--or to what extent--those guidelines should apply to existing facilities.

The most far-reaching regulatory alternative in the proposed regulations is the safe harbor provision that potentially exempts certain elements at existing facilities from barrier removal obligations under the proposed regulations.  The RIA results demonstrate that this safe harbor proposal is expected to reduce substantially the total monetary impact of revised (more stringent) requirements on existing facilities, whether owned by small entities or larger groups or organizations.  See RIA, Table ES-3.   

Another regulatory alternative being proposed by the Department would--for the first time--place a monetary limit on the barrier removal obligations of qualifying small businesses.  Qualifying small businesses are those small entities that satisfy small business size standards promulgated by the Small Business Administration.  Pursuant to this proposal, a "qualified small business" would have met its readily achievable barrier removal obligations for a given year if, in the preceding tax year, that entity had spent at least one percent (1%) of its gross revenues removing architectural barriers. 

The RIA does not, however, incorporate this monetary cap on barrier removal expenditures for qualifying small businesses into its cost or benefit models.  Assessing the incremental impact of this provision would have required assumptions regarding the number of small businesses satisfying the definition of "qualified small business" in any given year, as well as the nature and extent of barrier removal efforts by such businesses in the preceding year.  For example, even assuming it could be determined (or assumed) that a particular small retail establishment satisfied the "qualified small business" definition in a particular year, several sets of assumptions would nonetheless still be required to model the presumed barrier removal efforts made by that small retailer in the preceding year.  For example, should it be assumed that the small retailer had removed architectural barriers related to a ramp, accessible routes, and accessible parking spaces in the preceding year?  Or had this small retailer instead focused its barrier removal efforts on removing barriers concerning sales and service counters, doorways, and a single-user toilet room?  In either case, did the small retailer's efforts result in complete or partial removal of the affected architectural barriers?  Such questions underscore the difficulty in creating a reliable framework for modeling the individualized determinations that are necessarily part of the barrier removal calculus.  The Department thus determined that incorporating the provision for qualifying small businesses into the RIA would have been neither feasible nor useful.  Nonetheless, interested parties may still get a rough gauge of the potential impact of this proposed safe harbor by reviewing the "Small Business Impact Analysis" in Chapter Six of the RIA.

Lastly, the Department is also proposing several regulatory alternatives directed at lessening the monetary impact of certain supplemental requirements relating to existing play areas, swimming pools, and saunas and steam rooms at public and private facilities.  Smaller existing and unaltered play areas, pools, and saunas (meeting specified size limits) would be exempt from technical and scoping standards in the supplemental requirements.  Facilities exceeding the proposed size threshold would nonetheless have reduced scoping requirements for elevated play components (play areas) or accessible means of entry (swimming pools).  Because there are few sources of reliable data concerning the number and relative size of existing play areas, swimming pools, and saunas and steam rooms in the United States, the RIA does not incorporate this proposed regulatory alternative into the model.  However, to the limited extent such information was available, it is used in the RIA to modify, as appropriate, the likelihood of occurrence or unit cost of the element.  See RIA, Apps. 3-E, 3-G, and 3-H.

Commenters representing small business groups expressed appreciation for the Department's efforts--represented by the foregoing regulatory proposals--to mitigate  the potential impact of the proposed regulations.  These commenters noted that such regulatory alternatives "have the potential to remove much regulatory uncertainty and provide a level playing field for small businesses anxious to provide accessibility to their customers."

Summary of Results - Main Regulatory Assessment

From an economic perspective (as specified in OMB Circular A-4), the primary determinant of whether proposed regulations increase social resources and thus represent a public good is whether monetized benefits exceed monetized costs --that is, whether the regulations have a positive net present value.  The Department's proposed regulations indeed have a positive NPV under each of the four scenarios calculated in the regulatory assessment.

First, the RIA examines the incremental impact of the proposed regulations using the "main" set of assumptions (i.e., assuming a primary baseline (1991 Standards), safe harbor applies, and barrier removal readily achievable for 50% of elements subject to supplemental requirements).  Under this first set of assumptions, the proposed regulations have an expected NPV of $ 31.1 billion (3% discount rate) and $ 7.5 billion (7% discount rate).  See RIA, Table ES-1 & Figure ES-2. 

Second, the RIA calculates the incremental impact of "safe harbor" versus "no safe harbor" scenarios with all other assumptions remaining equal.   The expected NPV for the proposed regulations under a "no safe harbor" scenario would still remain positive, albeit at a significantly reduced level.  See RIA, Table ES-3.

Third, the RIA explores the incremental impact of varying the assumptions concerning the percentage of existing elements subject to supplemental requirements for which barrier removal would be readily achievable.  Readily achievable barrier removal rates are modeled at 0 %, 50%, and 100% levels.  The results of this third scenario show that, while the expected NPV is positive for each readily achievable barrier removal rate, varying this assumed rate has little impact on expected NPV.  See RIA, Table ES-4.

Fourth, the RIA's demonstrates the impact of using three alternate baseline scenarios (i.e., IBC 2000, IBC 2003, and IBC 2006) instead of the primary baseline.  As with the other scenarios, use of these alternate IBC baselines results in positive expected NPVs in all cases.  See RIA, Table ES-5.  These results also indicate that IBC 2000 and IBC 2006 have the respective highest and lowest expected NPVs.  These results are due to changes in the make-up of the set of requirements that are included in each alternative baseline.

Summary of Results - Small Business Impact Analysis

In addition to its benefit-cost analysis of the impact of the proposed standards on all entities subject to titles II or III of the ADA, the Department is required under the Regulatory Flexibility Act ("RFA") to analyze the impact of its proposed regulations on "small entities"-- namely, small businesses, small non-profit organizations, and small governmental jurisdictions with populations of less than 50,000.  If the proposed regulations are projected to have a "significant economic impact on a substantial number of small entities," the RFA requires an agency to prepare and make available for public comment an initial regulatory flexibility analysis ("IRFA").  On the other hand, no IRFA need be prepared should the head of the agency certify that the proposed rules--if promulgated--would not have a such an economic impact on a substantial number of small entities.

The Access Board certified, in both its NPRM and final rule promulgating the 2004 ADAAG, that its revised guidelines would not have a significant economic impact on a substantial number of newly constructed and altered small facilities.  See 64 Fed. Reg. 62,248 (Nov. 16, 1999) (NPRM); 69 Fed. Reg. 44,084 (July 23, 2004) (final rule).  Consequently, the Access Board was not statutorily required to prepare either an initial or final regulatory flexibility analysis for the 2004 ADAAG.

In the ANPRM, the Department encouraged small entities to provide cost data on the potential economic impact of applying specific provisions of the 2004 ADAAG to existing facilities and to recommend less burdensome alternatives.  Small businesses were well represented among ANPRM commenters.  Many commenters representing industry groups of all sizes said that "the possibility of having to modify existing facilities presents the most severe and burdensome compliance scenario for most businesses" and that the biggest potential cost of the proposed standards was represented by the "no safe harbor" scenario.  By contrast, several commenters representing disability groups urged the Department not to adopt a safe harbor, asserting that the "readily achievable" defense provided in the ADA adequately addresses the concerns of small businesses. 

The Department agrees with the commenters representing small businesses that a safe harbor provision is a reasonable means of lowering the potential costs of the regulation and, with these NPRMs, is proposing to adopt the safe harbor scenario.  Because the potential costs of compliance with the proposed standards pursuant to the barrier removal requirement was consistently identified by commenters as their paramount concern, the Department's adoption of the safe harbor should go a long way toward addressing the concerns of small businesses.

Some commenters representing small businesses also suggested that the Department employ a different methodology for its regulatory assessment than the Access Board.  Specifically, these commenters recommended that the Department assess the incremental benefits and costs for all facilities, rather than just a few.  These comments noted that many of the facility groups for which the Board did not provide a direct assessment of costs--including retail stores, restaurants, small manufacturers, and small service providers--are more typically small businesses.  By comparison, as noted previously, the Department's RIA assesses the impact of the proposed regulations on all public and private facilities.  Moreover, the Department's small business impact analysis includes all facility groups (for which statistical information was available) that could potentially be effected by the proposed regulations, including facility groups within which small businesses predominate.

Several commenters representing industry groups pointed to particular revised requirements as likely to have a disproportionate cost impact on small businesses, including the requirement relating to public entrances (which they suggest could impose greater costs on small businesses, which are more likely to have only two entrances, both of which would now be required to be accessible), and the requirement relating to operable windows (which are more typically found in small or rural motels than in large urban high rises). Commenters also noted that small businesses are more likely to be located in older buildings, which cost more to renovate than newer buildings, and discussed the greater marginal impact that any regulation (particularly one as complex as the proposed standards) has on small businesses due to their smaller economies of scale.  The Department notes that the revised requirement relating to public entrances is expected to effect no change for small facilities, and to the extent it effects a change at all, it will be for very large facilities for which it will be "less stringent" than the current requirement.  Similarly, the operable windows requirement can be met using inexpensive add-on hardware (similar to a light switch extension handle).

More generally, with respect to requirements that may impose a fixed cost, several commenters representing small businesses suggested that the Department provide small businesses with a lower cost alternative by permitting equivalent facilitation.  In the proposed regulations for title III, the Department has specifically recognized the continued legitimacy of equivalent facilitation as a means of lowering the potential costs associated with barrier removal.  In all cases, measures to remove barriers are only required when they are readily achievable, but if substantially equivalent access can be provided at less cost through alternative measures, entities are entitled to use them.

Chapter Six of the RIA sets forth the Department's comprehensive assessment of the estimated impact of the proposed regulations on small entities.  For the most part, this analysis uses the same methodology as the underlying "main" regulatory assessment except that some additional publicly-available statistics (from, for example, the Census Bureau and the Office of Advocacy of the Small Business Administration) are incorporated into the model in order to permit particularized calculations for small entities.

In sum, the Department's small business impact analysis uses the following methodological approach.  First, the analysis estimates (by facility group) the total number of facilities owned or operated by small entities and their respective total annual sales receipts.  Since governmental entities typically do not have sales receipts, expenditures--broken down by category (e.g., education, hospitals, parks, museums)--serve as a proxy for "sales receipts" for small governmental jurisdictions.  The resulting figures for small entity-owned facilities and sales receipts are compared to the "typical" facility.  See RIA, Table 17.  Second, the analysis compares the net costs of the proposed regulations on small entities and the "typical" facility for each facility group.  See id., Table 18.   Lastly, the analysis estimates total annual costs and annual costs as a percentage of sales for both small entities and "typical" facilities.  See id., Table 19. 

The results of the Department's small business impact analysis demonstrate that the proposed regulations would not have a significant economic impact on a substantial number of small entities.  See RIA, Ch. 6.  For small government jurisdictions, annualized costs are not expected to be greater than 0.5% of sales for any type of facility. Similarly, for all but a handful of small private entities, annualized costs are not expected to be greater than 0.5% of sales.  Only with respect to two types of facilities owned or operated by small private entities--aquatic centers and miniature golf courses--are annualized costs estimated to exceed 0.5% of sales.  However, as noted previously, the RIA does not incorporate the Department's proposed monetary limit (i.e., 1% of gross revenue) on barrier removal obligations for qualified small entities.  Application of this monetary cap on barrier removal costs for small qualifying businesses that own or operate aquatic centers or miniature golf courses would mitigate the incremental impact of the proposed regulations on these (or any other) qualified small entities.

[1]The Access Board's final assessments for its supplemental guidelines for play areas and recreation facilities are available on its web site.  See Assessment of Benefits and Costs of Final Accessibility Guidelines for Recreation Facilities, (Sept. 2002); Final Accessibility Guidelines for Play Areas - Economic Assessment, (Oct. 2000).  The Board conducted an initial, but not a final, regulatory assessment for its supplemental guidelines for State and local government facilities issued in 1998.

[2] Nor will public entities be required to retrofit elements in existing title II-covered facilities to bring them into compliance with the applicable revised standards so long as such elements presently comply with either the 1991 Standards or UFAS.  To make this clear, the Department is proposing a safe harbor provision for existing public facilities.

[3] While the benefits of harmonization between the ADA Standards and the model codes are clear, a few commenters noted the potential short-run downsides of harmonization.  For example, some commenters complained that it would be expensive for small businesses to purchase copies of the IBC which is privately published by the International Code Council.  Other commenters expressed concern that, since the 2004 ADAAG has a revised organization and format, they will have to learn a whole new regulatory system should the Department adopt these guidelines as the revised ADA Standards.  The Department recognizes that, while harmonization will make ADA compliance easier for all covered entities (including small business owners) over the lifespan of the regulation, this benefit may not be fully realized by all entities immediately.  To assist in the transition to the 2004 ADAAG, the Access Board has published a side-by-side comparison between the 2004 ADAAG and IBC 2003--including the provisions that have been incorporated by reference in the 2004 ADAAG--on its web site (  The ICC offers free downloads of a similarly detailed comparison between the 2004 ADAAG and IBC 2006 on its web site (  The Department is exploring the possibility of publishing a similar side-by-side analysis on its web site that compares the ADA Standards (both current and as revised) to one or more editions of the IBC (including any IBC provisions incorporated by reference) following promulgation of the final regulations.  Additionally, when the proposed regulations become final, the Department will publish small entity compliance guides required by SBREFA and other appropriate technical assistance.

Return to NRPM index

Last updated December 9, 2010