The purpose of this economic assessment is to discuss and, where possible, quantify the costs and benefits of the electronic and information technology accessibility standards issued by the U.S. Architectural and Transportation Barriers Compliance Board (Access Board) under the authority of section 508 of the Rehabilitation Act Amendments of 1998. The standards are intended to improve the accessibility of electronic and information technology used by the Federal government. The standards will benefit Federal employees with disabilities, as well as members of the public with disabilities who seek to use Federal electronic and information technology to access information and services. The standards will be incorporated into the Federal Acquisition Regulations (FAR). Federal agencies, including the United States Postal Service, must comply with the standards when they develop, procure, maintain, and use electronic and information technology, unless an undue burden would be imposed on the agency. There is an exemption for national security systems. Failure of a Federal agency to comply with the standards when procuring electronic and information technology may result in an administrative complaint or a civil action seeking to enforce compliance with the standards.

The analysis for the final rule is, for all intents and purposes, identical to the analysis prepared in support of the proposed rule. The original analysis was a very broad estimate of the costs and benefits of the standards. None of the substantive changes made in the final rule were of sufficient magnitude to effect these estimates.

Need for Standards

The improvements in accessibility expected to result from the standards are intended to address three issues that may persist in the absence of the standards.

  • Potential market failure that can arise because transaction costs of organizing persons with disabilities to send appropriate market signals to manufacturers of electronic and information technology for accessible products are too costly.
  • Failure of the Federal government procurement system to place a value on accessible features when purchasing electronic and information technology.
  • Civil rights issues that may call for a solution that goes beyond maximization of market efficiency.

By incorporating the standards into the FAR, the Federal government will be placing a very high value - limited only by the undue burden provision - on accessibility. By doing so, the potential exists that the government may purchase more accessibility than necessary to maximize economic efficiency. However, the civil rights benefits are difficult to capture using traditional economic techniques, as they represent improvements in equity considerations rather than improvements in efficiency.

Baseline of Federal Spending on Electronic and Information Technology

According to Office of Management and Budget (OMB), the Federal government spent $37.6 billion on information technology in fiscal year 1999. The defense agencies appear to have the highest information technology budgets, while civilian agency information technology budgets are expected to increase rapidly. However, it is not possible to disaggregate this data in a manner that it is useful for purposes of this assessment. Instead, annual sales data collected from the General Services Administration (GSA) Federal Supply Service (FSS) for fiscal year 1999 were used to estimate Federal spending on electronic and information technology affected by the standards.

Using the GSA data, this assessment estimates that the Federal government spends approximately $12.4 billion annually on electronic and information technology affected by the standards. This estimate likely understates the actual spending by the Federal government because agencies do not always make purchases through the GSA FSS. Thus any items that are purchased directly from suppliers are not accounted for in the GSA data. As a result, Federal expenditures for electronic and information technology affected by the standards may actually be higher than the GSA data would indicate. The degree to which the potential understatement of baseline spending leads to an understatement of the costs of the standards is unclear. Some components of the estimated costs of the standards rely heavily on Federal spending levels, while others do not. On the other hand, some portion of this $12.4 billion is likely to reflect spending on electronic and information technology that will not be affected by the standards. For example, the GSA sales data include expenditures for wiring and file servers. Neither of these items will be affected by the standards. Thus, their inclusion may have the effect of overstating Federal expenditures on electronic and information technology affected by the rule.

The exact relationship between the OMB data and the GSA data is unclear because the products and services reflected in the two estimates are likely to be significantly different. OMB data on information technology obligations from fiscal years 1994 through 1998 were examined. Two scenarios were used to develop an upper and lower bound representing the proportion expected to be potentially affected by the standards. The average proportion of the total obligations for information technology potentially affected by the standards ranged between 25 percent and 50 percent. The $12.4 billion estimate based on the GSA data falls within this range, representing 33 percent of the total fiscal year 1999 information technology obligations of $37.6 billion. One limitation of these ranges is that they are based on gross classifications of information technology obligations and do not provide the level of disaggregation necessary to parallel the GSA data. As a result, the two scenarios likely include expenditures on products and services that would not be affected by the standards to a higher degree than the GSA data.

Estimated Costs of Standards

Accessible versions of some electronic and information technology products, particularly computers and software, are available to satisfy the requirements of the standards. However, there are many products that will require modifications to meet the standards. The standards are to be applied prospectively and do not require agencies to retrofit existing electronic and information technology. The analysis includes discussions of both direct and opportunity costs associated with the standards. Major sources of cost include:

  • Costs of modifying electronic and information technology to meet the standards.
  • Training of staff, both Federal and manufacturers, to market, support, and use accessible products.
  • Translation of documentation and instructions into alternative formats.

The direct costs that were quantified for this assessment based on fiscal year 1999 data are shown in Table ES-1. The total estimated costs to society range from $177 million to $1,068 million annually. The Federal proportion of these costs is estimated to range between $85 million and $691 million. The ability of manufacturers, especially software manufacturers, to distributes these costs over the general consumer population will determine the actual proportion shared by the Federal government. Assuming that the addition of accessible features adds value to the products outside the Federal government, we expect these costs to be distributed across society thereby setting a lower bound cost to the Federal government of $85 million. If manufacturers do not distribute these costs across society, the upper bound of the Federal cost will increase to an estimated $1,068 million.

These costs must be placed in appropriate context by comparing them with the total Federal budget. In fiscal year 1999, the Federal government spent $1.703 trillion, of which S37.6 billion was spent on information technology. By comparison, the lower and upper bound of the costs of the standards represent 0.01 percent to 0.06 percent of the total Federal budget, 0.23 percent to 2.8 percent of the amount spent on information technology.

Table ES-1 Estimated Costs (millions)

Electronic &Info Technology Lower Bound and Upper Bound Cost Estimates
General Office Software $ 110 - $456
Mission Specific Software $ 10 - $52
Compatible Hardware Products $ 0 - $337
Document Management Products $ 56 - $222
Microphotographic Products $ 0.1 - $0.4
Other Misc. Products $ 0.2 - $1
Total Social Cost $ 177 - $1,068
Estimated Federal Proportion $ 85 - $ 691(1)

Estimated Benefits of Standards

The benefits associated with the standards result from increased access to electronic and information technology for Federal employees with disabilities and members of the public with disabilities who seek to use Federal electronic and information technology to access information and services. This increased access reduces barriers to employment in the Federal government for persons with disabilities, reduces the probability that Federal workers with disabilities will be underemployed, and increases the productivity of Federal work teams. The standards may also have benefits for people outside the Federal workforce - both with and without disabilities - as a result of spillover of technology from the Federal government to the rest of society.

Two methods are presented for evaluating the quantifiable benefits of the standards. The first is a wage gap analysis that attempts to measure the difference in wages between the general Federal workforce and Federal workers with targeted and reportable disabilities. While this analysis is limited to white collar Federal workers due to data constraints, the potential change in productivity is measured by the difference between the weighted average salary for all white collar Federal workers and Federal workers with targeted and reportable disabilities. This assumes that an increase in accessibility will help diminish this wage gap by increasing worker productivity.

The alternative is a team based approach for measuring the productivity of Federal workers. This approach is based on the assumption that a Federal worker's wage rate reflects their productivity and the scarcity of their skills in the labor market. However, this may not apply to Federal wage rates, thus the average productivity of a Federal team is assumed to be equivalent to the average Federal wage rate. Based on this rate, it is assumed that the standards will produce an increase in productivity ranging between 5 percent and 10 percent. Since no data have been identified to support the increase in productivity in the team based approach, we have retained the wage gap analysis to represent the benefits generated by the standards.

The quantified benefits of the standards represent the value added by the standards in closing the wage gap between the general Federal workforce and Federal workers with targeted and reportable disabilities. Keeping in mind certain data limitations with this analysis that are explained further in Chapter 5, it is estimated that the benefits of the standards will produce a maximum upper bound benefit of $466 million to Federal workers with disabilities. This figure does not account for benefits that may be accrued by the general public or other Federal workers due to spillover effects of the standards.

Table ES-2 Estimated Benefits (millions)

Productivity Increase Aggregate Benefits Range
Lower Bound $ -
Upper Bound $ 466

Given the uncertainties associated with the estimates of the quantifiable costs and benefits in this analysis, it is not possible to conclude that the benefits are greater than the costs. However, there is significant overlap in the estimates of the quantified costs and benefits in the analysis, suggesting that there is a non-zero probability of such a state of affairs. All the civil rights benefits of the standards are not quantifiable. These nonquantifiable civil rights benefits are considerable and weighed heavily in Congress's initial enactment and strengthening of section 508, which require the standards to be issued.