Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 Information Technology in Services: Implications for Public Policy Information technology (IT) has contributed to the growing economic importance of services, and that contribution is expected to increase sub- stantially. This report has documented, for example, the dramatic growth in electronic transactions, yet the fact that most business transactions still in- volve direct cash or paper checks suggests that what has been seen to date is only the beginning of a long-term transition in the conduct of services. Continued advances in computing and communications technologies, devel- opment of new features and applications, and increases in their affordability and ease of use will drive the further integration of IT into services. Pac- ing that integration will be growth in the understanding of how best to select, introduce, support, and manage IT. Inasmuch as people learn from early successes and failures, future applications are expected to be more successful on average than those of the past. Experimentation will con- tinue; there will still be failures as well as successes in the use of IT in services. The spread of IT to date owes much to market forces there has been no explicit national policy aimed at promoting the use of IT in services. As discussed in Chapter 2, conditions in different industries have motivated different types and rates of IT application; as discussed in Chapter 3, spe- cific applications arise from conditions, options, and approaches perceived by individual management teams. The accumulated experience with IT in services now raises questions about potential market inefficiencies and im- plications for social welfare, concerns that may argue for public policy 193
94 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY intervention. How rapidly and smoothly can adjustments be made to take place? How can the benefits and costs be widely and fairly distributed? How can positive impacts be facilitated and negative impacts ameliorated? How can policymakers better track and anticipate relevant trends? This chapter first discusses some important policy issues and then introduces some specific policy options related primarily to (1) employment shifts associated with the use of IT, (2) investment in IT, and (3) research relating to the application of IT and the changing structure of the economy that the committee believes should be seriously considered. The emphasis is on identifying areas of need rather than detailed specification of recommended actions. Implications for management action are discussed in Chapter 5. IMPLICATIONS FOR MACROECONOMIC AND FISCAL POLICY As this report is being written, the slow economic recovery is motivat- ing public policy interest in economic growth. One important contributor to economic growth is investment, and one vehicle often suggested for stimu- lating investment is tax incentives.2 However, this study indicates that there is no evidence that significant lack of investment in IT has slowed growth in productivity, while there is evidence that a number of (sometimes measurable) benefits have accrued in several industries. Moreover, con- tinuing improvements in IT's functionality and affordability suggest that IT will be used to help lower entry barriers in some markets as well as open possibilities for new kinds of business.3 Any investment stimulus en- acted by government should not discriminate against investment in IT (software and support as well as hardware). The market should decide how to allocate investment dollars among alternative uses of capital. The findings of this study also suggest that the benefits of investment in IT within the U.S. will not all be captured domestically. IT is being used to facilitate the globalization of business, enabling enterprises to manage ef- fectively over a wider geographical area and to shift some activities and associated employment overseas. Greater availability of capital overseas and the worldwide integration of markets and capital resources through IT are other factors promoting globalization. Thus, in the absence of mecha- nisms to reward local spending, there is no guarantee that policies designed to encourage investment or demand will necessarily increase jobs within the United States. Because of its collateral effects, investment overseas is not necessarily a macroeconomic problem, but it may create problems for local communities in which job losses may occur. Local job creation may require incentives for local investment. One type of IT-related investment with many local benefits is invest- ment in the domestic telecommunications and information infrastructure.4
IMPLICATIONS FOR PUBLIC POLICY The proliferation of IT to date has relied in part on a strong national tele- communications network (specifically, a complex of local and national net- works). Large enterprises have built their own, private networking capa- bilities to contain costs, secure advanced capabilities, and gain competitive advantages, but smaller enterprises, in particular, depend on the public in- frastructure that is provided by telecommunications service companies.5 A second arena likely to benefit from a powerful telecommunications and information infrastructure is the private home. Widespread access to networked IT in private homes is likely to lead to the creation of large home-based markets for new products (e.g., devices or services based on or enabled by multimedia technology). Consistent with the pass-through-of- benefits phenomenon discussed in Chapter 3, these new markets will in all likelihood generate public or private financial gains that far exceed those realized by the first companies entering these new markets. The increasing integration of computing and communications, the as- similation of IT as part of the infrastructure of an enterprise, and the grow- ing dependence of enterprises on IT all indicate that advanced information infrastructure will be important to achieving the benefits of IT in service activities. Information infrastructure can also assist in the delivery of train- ing, as recommended above; there is growing interest at all levels in the use of electronic networks to deliver educational services. To date, market forces have generally worked well in providing access to IT benefits. But in some geographical areas, selectively stimulating the growth of information infrastructure particularly for small business, edu- cational institutions, medical care systems, and the home- could enhance interactions among all these units, promote expansion in both the number and types of jobs in the service sector, and possibly lead to the creation of entirely new service industries. Given the potentially broad impacts, information infrastructure including investment in relevant research and development_leserves special consideration in public policy.6 However, wherever possible, the United States should allow the market to opti- mize the allocation of resources. The country has an opportunity to ex- pand on the relative advantage it currently has in many areas of telecommu- nications. A number of countries (e.g., newly industrializing and central European nations) have obviously weak infrastructures; but virtually all countries have targeted information infrastructure for improvement as a vehicle for economic development and growth.7 195 BACKGROUND ON EMPLOYMENT ISSUES RAISED BY INFORMATION TECHNOLOGY IN SERVICES It is impossible to isolate the effects of IT on employment from the many other factors that affect it. In general, the application of new technol
96 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY ogy may initially give rise to job displacement and job change because by definition new technology changes the way output is produced as well as the output itself. Historically, the evidence indicates that the displacement effects have been temporary: over the long term, new technology and production processes tend to promote productivity, competitiveness, and economic growth, all of which contribute to job growth over timed MIT's Robert Solow and others have estimated that 70 to 90 percent of economic growth has depended on the application of new technologies and techno- logical innovations. Thus, although job opportunities diminished in agri- culture and more recently in manufacturing due to automation and other factors, the number of people employed overall grew, particularly in service industries. Lower prices and the introduction of new products contributed to growth in employment, particularly in services; service-occupation em- ployment, for example, experienced substantial growth between 1980 and 1990.9 It is precisely this history that troubles some analysts: Will the combined strengthening and slimming of services now trigger (or accom- pany) new economic growth somewhere, or does the diffusion of IT and other technologies from agriculture through manufacturing to services im- ply that there is nowhere left for sufficient growth in jobs to occur? In addressing this question, several issues must be considered. As discussed in the macroeconomic analysis of Chapter 1, IT applica- tions in services particularly affect white-collar jobs. Bureau of Labor Statistics forecasts, executives interviewed by the committee, committee members, and a number of studies all anticipate slower growth in some white-collar occupations in the near future, resulting (in part) from increased use of information technologies. The first to be affected have been lower- level, administrative and clerical occupations; more recently, the number of paraprofessional and technician occupations has been contracting in some industries. IT is also enabling the ongoing reductions in middle-manage- ment positions, and it is expected to diminish growth in sales positions as sales-related applications proliferate.~° For example, the increased use of IT is likely to have contributed to the higher displacement rates evident in retail trade and "other services," compared to manufacturing, during the l980s. I! Although the executives interviewed for this study were asked prima- rily about strategic and technical issues related to the introduction of new IT systems, many indicated that reductions had occurred in the size of their company's work force. For example, Larry Bacon of the Travelers Compa- nies described the elimination of 200 positions following the introduction of automation systems (Chapter 4~; at McKesson, the use of PCs by customers has resulted in large cuts in the number of order takers, salespeople, buyers, and functional managers (Chapter 3~; Ford reduced its accounts payable staff by 75 percent through implementing IT systems (Chapter 4~. Certainly
IMPLICATIONS FOR PUBLIC POLICY 197 in the short term, and in particular job categories, firms, or industries, intro- duction of new IT can have significant effects on employment that may not be fully offset by growth elsewhere. Some of the observed displacement may reflect subtle and indirect shifts of labor effort, including the shifts to efforts on the part of customers men- tioned in Chapters 3 and 4. Such shifts reflect not only managers' decisions about how to reorganize work within an organization, but also changing levels of expertise and changing preferences within the general popula- tion.~2 Increased use of IT-based systems may also have fostered the recent growth in part-time and temporary employment, some of which may be among people who would prefer full-time employment (and are therefore underemployed). At least some observed and anticipated displacement reflects slow mod- ernization of relatively low-technology industries. Service job growth in the past decades was concentrated in industries that have lagged in their (overall) use of computing and communications (e.g., health care, legal services, and retail).~3 The comparative inefficiency of these industries casts doubt on the notion that they will continue to support significant job growth. Complicating the problem of assessing changes in employment levels is the fact that IT is used to facilitate the relocation of jobs at all levels, both within the country and between countries. Even if the absolute amount of work or jobs remains constant, people do not tend to move with jobs, and so job movement can have the effect of (local) job displacement.~4 Companies with heavy data-entry requirements (e.g., airlines, insurance companies, and some database service providers) have used foreign clerical workers for several years to reduce costs.~5 More recently there has been growth in the use of foreign professionals for software development, especially for lower- level coding but increasingly for higher-level work. While evidence of these trends is largely anecdotal, that evidence suggests growth in the use of lower-wage foreign labor in service activities over the past 5 years. Thus, while the absolute fraction and number of white-collar jobs trans- ferred to foreign citizens is still small, the trend raises questions about both the number and quality or distribution of job opportunities that will remain available to U.S. citizens over time. Entry-level jobs are the first to go offshore, but higher-level jobs are also moving. On the other hand, jobs are created in the United States by foreign-owned (as well as U.S.-owned) enterprises. All of these trends are facilitated by improvements in commu- nications technologies, and all are part of a pattern that has led some econo- mists to posit an emerging global division of labor and economic activity.~7 It is difficult to forecast where new jobs will be created, which leads some analysts to predict a rise in unemployment or a job shortage. How- ever, overall, job growth is reasonable to predict. The past decade alone
198 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY witnessed growth in opportunities in such areas as health care, environmen- tal services, and, particularly relevant, IT-related services (software devel- opment, network-based services, training in the use of IT, systems integra- tion, remote IT facilities management, and so on). Although job creation has been concentrated in small, entrepreneurial firms (across all industries), these firms and their employment patterns are relatively difficult to track. The most obvious trends suggest principally that job growth (and job levels) among traditional large employers will be depressed in the near term. However, the implications for overall employment levels are not clear. A1- though large firms continue to be large and important employers, the percent- age of the working population that larger firms employ has been fallingi9 and the identity of the largest firms has changed over time. Simultaneously, new companies and the new jobs associated with them have been growing. The enabling effects of using IT in these new enterprises have not been measured. An added problem lies in labeling: the conventional division of the economy into agriculture (plus mining and forestry), manufacturing, and services suggests that we have run out of sectors. But the service sector is so heterogeneous that only a more fine-grained identification of subsectors or new sectors is likely to illuminate underlying changes in employment patterns. It is possible that new kinds of businesses will emerge from both the manufacturing and service sectors that would give rise to yet another kind of sector. However, it is unlikely that such developments will be recognized quickly, given the difficulty of detecting them in economic data. Whether or not there is a temporary rise in unemployment, the content of jobs is changing. IT-based systems have been associated with a shift toward intellectual activities, manipulation of information about things or people (as opposed to manipulation of physical materials), and increased attention to the information content or corollaries of products (e.g., airline reservations and frequent flyer services as well as air transport, customization of insurance policies, and so on). With computer systems focused on the mechanics of collecting, storing, processing, and retrieving information, people have more time to do what computers do not do as well-people-to- people interaction, creating new ideas, and so on making use of informa- tion delivered by technology. In addition, as the technology delivers more and more data, people will be called on to use and work with it, in some cases to be more analytical. These are the changes in most jobs that lie behind observations about the rise of the "knowledge worker." Ironically, the rise of the knowledge worker is associated with renewed attention to process engineering: the essence of reengineering is the gathering of infor- mation and the conduct of analysis to help improve "production" processes (food preparation and delivery, transport of people and cargo, and so on). The average complexity of service jobs appears to have risen; by vary- ing definitions, there are more knowledge workers, and higher levels of
IMPLICATIONS FOR PUBLIC POLICY 199 education are required for entry-level jobs. Flatter organizations and the popularity of team organizations, both enabled by the use of IT, imply that the work force of the future will need better communication and coordina- tion skills. These changes in jobs are evident in many industries; they are occurring more rapidly and extensively in some industries (e.g., financial services) than in others (e.g., retailing). Other changes in skill requirements are predicted by management analysts the need for a broader set of skills and more problem-solving ability to meet a wider set of responsibilities but it is not clear how broadly these changes are taking hold in practice. THE NEED FOR POLICY INTERVENTION TO EASE EMPLOYMENT TRANSITIONS Although this report has concentrated on impacts in companies and industries that use IT, its positive benefits at these levels may be accompa- nied by negative side effects, most notably impacts on the labor force. Employment impacts may be direct or indirect, depending on how, where, and when IT is used. As discussed in Chapters 3 and 4, IT is used as a tool by managers to systematize or reorganize production processes and enter- prises, effect transitions from old to new lines of business, and shift activi- ties within and between enterprises, locations, and time periods. Layers are being eliminated from the job structures of enterprises, implying broader job definitions, while the mix of activities within firms is often becoming more focused, implying a narrower band of functions for personnel in such enterprises. All of these changes are occurring in the context of limited economic growth and structural changes in service industries, as discussed in Chapter 2, plus domestic and international competitive pressures that motivate businesses to seek greater efficiency.20 The trends listed above raise the specter of job displacement (the elimi- nation of specific jobs or the reduction in growth for such jobs compared to what would have been under the original conditions) and job change (changes in the nature and mix of tasks that make up jobs). These changes are likely to continue, because computer-based technologies and their applications are often reconfigured over time and because people learn over time how to use these technologies better. It is beyond the scope of this report to quantify potential displacements associated with IT in services or their duration. Broad trends and the diffi- culty of interpreting them are outlined above. Based on those trends, the committee concludes that some displacement, including significant loss of current types of jobs, is inevitable, although new and different jobs are likely to be created. As with past technology-induced displace- ments, public policy intervention may be necessary to ameliorate im- pacts on communities and individuals.
200 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY The changes in service activities addressed in this report are part of a larger pattern of changes in employment opportunities that has been charac- terized in Workforce 20002~ and other studies; the inferences from this report are in harmony with other, more broadly based analyses that create an expectation for a more volatile job market. This study reinforces and amplifies concerns about the transitions in employment that are likely to take place over the next 5 to 10 years. A central concern for public policy is how to ease worker dislocation, reduc- ing negative social impacts (such as unemployment, underemployment, and reduc- tion in household income) and facilitating the economic transition. The current economic system is characterized by short-term notification of displacement, loss of health insurance and pension plans with job loss, limited outplacement assis- tance, and limited unemployment insurance and employment service programs. Absent change (such as movement toward portable, universal health insurance), the conditions described in this report suggest that a growing number of people may be faced with loss of insurance benefits as well as loss of income. Such programs as advance notice, retraining, employment search assistance, portability of benefits, and temporary income support during job search have been suggested as appropriate by various parties. These measures imply interdependent but sepa- rate roles for employers, government, and individuals. This committee did not take a specific position on the relative merits of these proposals. As has been noted by others, measures (such as training, retraining, and other types of adjustment assistance) will probably be needed;22 the chal- lenge is to develop innovative and effective approaches. Also important will be encouragement of job creation through economic growth, including the investment in new, entrepreneurial businesses and information infra- structure (see "Implications for Macroeconomic and Fiscal Policy" above). Calibrating policy measures will require closer monitoring and analysis of trends (see "The Need for Additional Research to Guide Policy Making" below), to permit timely if not proactive responses. Reinforcing conclusions developed by other studies, the committee be- lieves that there is a clear need for increased training of many types. Both retraining for existing members of the labor force (in anticipation of new jobs and a new mix of jobs) and education and training to make prospective entrants to the labor force more versatile (in anticipation of ongoing change in jobs and careers and periodic retraining) will be needed. Especially important, given changes observed in service activities (see Chapter 4), will be the development of skills relating to the handling and use of information, including a solid foundation of basic skills ranging from reading, writing, and mathematics to problem-solving skills. Developing such skills for indi- viduals already in the labor force implies a departure from the job-specific, task-oriented training typical of employer-based programs. In particular, it implies a need for relatively frequent or ongoing training.23
IMPLICATIONS FOR PUBLIC POLICY 201 This study underscores the need for training at all levels within organi- zations. In particular, job trends associated with the use of IT raise added concern about training for the people at the bottom of the labor force, the people with the least skills who would be suited for some of the jobs most likely to be displaced or changed through the use of IT. Inequalities in the earnings distribution have widened recently, reflecting greater growth at lower wage levels and reductions in middle-level opportunities due in part to shifts in the mix of jobs by occupation and industry.24 Inasmuch as information-related and problem-solving skills remain restricted to a small subset of the population, further shifts toward "knowledge work" will make it more difficult for people with limited or weak skills to obtain good jobs.25 This study thus adds to concerns raised by others about the quality of basic (~-12) education (and also vocational education); the majority of the labor force consists of people who have at most a high school diploma. Looking at the other end of the employment spectrum, it should also be recognized that training will be needed for professional and managerial personnel to facilitate their own use of IT, prepare them to manage IT better, and convey new approaches to the organization of work. Questions arise as to the degrees of responsibility held by individuals, employers, and government for training and retraining. Many of the changes discussed in this report are beyond the ability of individual members of the labor force to influence. A fundamental question is how to assist the em- ployee in making transitions, with help from the employer as well as sup- port from government (at all levels). As suggested above, portability of benefits illustrates one mechanism for facilitating transitions. Since will- ingness as well as ability to pay for these measures will affect what is done, incentives may be needed for individuals and employers. THE NEED FOR ADDITIONAL RESEARCH TO GUIDE POLICY MAKING A recurring theme in this report has been that limited data yield limited understanding. Problems in collecting and aggregating data are significant reasons why available macroeconomic data neither prove nor disprove claims about the level of impact investment in IT has had on performance in services. The conventional distinctions between manufacturing and services, which underlie the collection, analysis, and presentation of publicly available data, undermine useful analysis. Available data do not help to track the changes and blurring of industry structures and identities resulting from shifts in who does which activities, where, and when (e.g., the roles of both hospi- tals and insurance companies in providing managed care, the retailer-like roles of transportation service providers, and so on). And the data that are available obscure the fact that production of goods and services involves
202 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY many comparable activities. For example, data regarding employment from major companies that produce goods may be tallied only within manufactur- ing, thus obscuring large volumes of service activity (such as research, marketing, and training) also produced within those companies.26 Because each industry has different performance indicators, developing a comprehensive body of measurements is a huge task that is far from being accomplished.27 Even with perfect measures of output, it may still not be possible to separate out the effects of various factors, including the use of IT, that affect output and other dimensions of performance. In addition to technology, those factors include changes in the skills of management, the education and training of the work force, economies of scale and scope, technology change in general, government regulations, collective bargain- ing, and so on. Detailed microeconomic studies building on the kinds of issues revealed by the committee's interviews, macroeconomic studies, and organizational studies are necessary to better measure and under- stand the contributions of multiple factors and how they interact. Given the apparent significance of small firms to growth in the economy and in employment, valuable insights could be gained from more research specifically focused on how smaller enterprises use IT and with what effects. Insights avail- able through the literature and through interviews tend to be limited to the experi- ences of large firms researchers tend to find it easier to identify and work with larger films, and that tendency may bias understanding.28 Although the committee did not have the resources to investigate smaller- company practices systematically, members cited examples of individual smaller companies that had developed extremely fast response capabilities, strong cross-functional team-oriented cultures, team management and re- ward systems, and very productive desktop IT systems for enhancing com- munications. These enterprises already exhibit many of the characteristics larger companies seek through more disaggregated organization and activ- ity-based competitive structures. Smaller enterprises also differ in other significant ways. They can often rely on direct person-to-person communi- cations and more informal information systems. Because of the increasing importance of small enterprises in creating jobs, introducing IT innovations, and utilizing IT in new ways, special study of small enterprises' use of IT would seem a worthy goal for future research. Finally, more systematic attempts to compare experiences with IT inter- nationally may be useful, in part to benefit from different perspectives on elements of performance, concepts and measurements, and trade-offs in the use of IT and in part to capitalize on the extensive tradition in other coun- tries (especially in Europe) of studying social and organizational impacts of technology. Since use of IT will be a critical factor in international competition, it would be prudent to systematically compare and moni- tor progress and impacts in other countries.
IMPLICATIONS FOR PUBLIC POLICY 203 Improving Federal Macroeconomic Data Gathering and Analysis There is a need for improved statistics about existing activities and for statistics that capture the development of new services. Policymakers, in particular, need better data to monitor and plan for changes in the economy. Better statistics could provide a clearer picture of the economy, illuminating shifts in investment, employment, products, productivity, and other dimen- sions relevant to government and industry. In particular, there is a need for an expansion in scope, i.e., data about more industries. As was noted, detailed data are available for only a fraction of the service-oriented economy. Fundamental to these is a need for a more refined classification of activi- ties, firms, and industries that would guide the collection and analysis of data. For example, it would be useful to have more specific data on se- lected skill classifications and occupations within each industry class. It would also be useful to separate out significant service business compo- nents within firms that predominantly produce goods. The complications and costs of obtaining more detailed data are well recognized. Broader and improved capture of changes in quality and measurement of other factors that contribute value for customers could also help to put statistics on productivity into perspective. Existing measures focus prima- rily on industries whose outputs are captured by gross national product data. The service-sector interviews undertaken by the committee plus both anec- dotal and more formal evidence suggest that many of the performance gains in services particularly those associated with intangible improvements in quality or those realized as cost savings, greater convenience, or increased flexibility or variety may not be captured by the industry producing the benefit.29 At a gross level, these shifts may be reflected in the financial performance of a company. But where competition within an industry is heavy and financial gains are passed through to customers or suppliers as savings, there may not be an obvious correlation between benefits to cus- tomers and the firm's or industry's financial performance.30 Progress in developing and using hedonic price indexes suggests that there are some opportunities to develop better measurements of output quality and cus- tomer benefits, although delivering against those opportunities may take time and be difficult to realize. The need for better economic statistics has been recognized for some time; it is made more urgent by the cumulative impact of cutbacks in fed- eral statistical programs during the 1980s. A number of recommendations for improving statistics on productivity and services were made in an earlier National Research Council report (Box 6.11. Although progress has been made in implementing several of the Rees Panel's recommendations, more work is needed, and the committee therefore endorses those
204 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY BOX 6.1 Rees Panel Recommendations for Better Statistics Recommendation 3: "The Panel recommends that the Bureau of Labor Statistics arid the Bureau of Economic Analysis explore methods for esti- mating the implications of error reduction In component measures for the reduction of overall error in product;~;ty measures beyond that corrected by routine revisors.'' (p. 73 Recommendation 4: "The Panel recommends that the Bureau of Economic Analysis (BEA) and the Bureau of Labor Statistics (BLS) seek to improve their existing price indexes and to develop auxiliary measures of price change. These new auxiliary measures should take into account more adequately the types of quality change that are not now mea- sured. ~ Among the adjustments that could be Incorporated In the new mea- sures are adjustments ~ . . for changes in value to users resulting from the introduction of improved products; estimates of the value to users of improvements in performance that are achieved without increases in real costs; and estimates of the present value of future savings in operating eH;ciensy made possible by design changes and improvements.... '` (p. 8) Recommendation 6: 'I. the Parcel does agree that for the study of many important social problems-for example, improvement of the health sta- tus of the population definitions of output and input that go well beyond those currently used to measure productivity are re- qu~red~ . . . 6` (~. l0) Recommendation I4: '`The Panel endorses . . . calling for the Census I Bureau to collect, as an integral part of each economic census, data or, the purchases of intermediate services as well as materials by estab lishments.', (p. 13) Recommendation 17. '`The Pane! recommends that government agencies support research aimed at improving knowledge about the sources of product',Yity change. These agencies should be especially attentive to research that focuses on measuring technical and organizational change and new product and serv',te innovation...." (p. ~ S) NOTE: Emphasis added SOURCE: Natfonai Research Council, Panel to Review Productivity Sta- tisties 1979. Measurement and Interpretation of Produ~`v,ty, National Academy of Sciences, Washington, O.C.
IMPLICATIONS FOR PUBLIC POLICY 205 recommendations. Specific proposals were also captured in the recent Presidential Economic Statistics Initiative (the Boskin Initiative). That ini- tiative called for a 5-year, $230 million effort, a significant portion of which was aimed at improving service industry statistics.3~ Such an initiative, if funded, would help to ameliorate some of the problems of concern to the committee; momentum in this area must not become a casualty of the tran- sition. It is hoped that the Clinton administration will enhance and build on these recommendations with its own program. Note also that the need for better statistical data is largely independent of issues related specifically to IT and productivity. The committee urges the federal administration to promote a sig- nificant upgrading of statistical programs. At a time when more and more enterprises are recognizing the value of information to the perfor- mance and expansion of their businesses, it behooves the government to collect and make use of better information about the changing economy. Improving Data and Accounting Principles Related to Investments in Information Technology Available data provide information on investments in computing and communications hardware, which is treated for tax and national statistics purposes as a capital investment. However, IT hardware represents only a fraction of the investment needed to make IT effective. Committee mem- bers (based on their experiences in developing, installing, operating, and maintaining systems) and other analysts have noted that hardware may con- stitute only about one-third of the total investment in a system (and that proportion has been declining). Other components of IT system cost in- clude, in particular, the cost to develop, purchase, or maintain software plus associated training, support, integration, upgrading, and other services. All of these costs are expenses under current accounting principles and prac- tices and are not captured in a firm's capital accounts. Increases in the outsourcing of elements of information technology from telecommunica- tions to databases to software development to data center operations- tend to further raise the percentage of IT costs that are expensed rather than capitalized. A consequence of particular concern is that the value of soft- ware in generating income is not obvious to management, investors, or policymakers even when a specialized software system, such as a comput- erized reservation system, is critical to strategic advantage. Software has become a major asset to corporations that is not reflected in their balance sheets.32 Improving the valuation of software will be difficult. There are differ- ent and sometimes conflicting interests among analysts, investors and own- ers, managers of organizations using software, managers of organizations
206 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY producing software, and tax authorities, among others. These different per- spectives have come out recently in the development of legislation that would affect the tax accounting treatment of software, as one of several "intangible assets";33 similar issues arise concerning the treatment of re- search and development spending. Discomfort with capitalizing software over a long period of time arises from difficulties in predicting with reason- able precision the value and life of a software system; there is concern that any piece of software may have a relatively brief useful life, and it may be replaced or upgraded through an ongoing process of "maintenance" that blurs the identity of the software in question. The committee recommends that the Financial Accounting Standards Board (FASB) and the Con- gress look at ways to improve the usefulness and monitoring of account- ing data on investments in both purchased and internally developed IT systems (especially software and databases) to better support fiduciary management and performance measurement.34 Increasing Awareness of and Investments in Research Related to Information Technology in Services and Service Quality Measurements Committee interviews with executives suggest that corporate R&D on the use of IT to develop new service products and processes is underappreciated, underestimated, and probably underfunded.35 Committee deliberations and interviews with executives of large companies plus published materials re- veal a pattern of experimentation that takes place in service industries, in which companies develop and try out new applications, software, and sys- tems.36 Such projects or programs are not necessarily classified or analyzed as research by the companies that undertake (and pay for) them, but they contribute to a larger pattern of trial, error, and success that drives new applications of IT in services and shapes the competitive landscape within industries. The committee concluded from its assessment that more emphasis needs to be placed on research and development relating to services, including the applications of IT in services. This greater emphasis is needed across the board in firms, institutions, and government. Such emphasis should include both process and product innovation. It is not possible to estimate how much relevant R&D takes place in or for services, but there is anecdotal evidence that support from both corpo- rate budgets and federal programs for academic research for these efforts is limited. This situation reflects a larger problem of limited support for R&D explicitly focused on process or organizational issues. Academic analysts, including committee members, have observed that it is difficult to get fund- ing for such research. Enterprises should not be discouraged by accounting
IMPLICATIONS FOR PUBLIC POLICY 207 conventions from experimenting with these technologies, given their proven contributions to competitiveness; rather, financial incentives for such ex- perimentation should be considered. Although R&D within an industry is typically funded by the industry directly, broader, longer-term, or academic research often presumes federal government support. This form of broad stimulus would fall within the realm of macroeconomic and fiscal policy, discussed above. Another vehicle for motivating more innovation among service in- dustries may be an increased effort to apply the Baldrige Award to services. According to literature describing the award, up to two awards annually can be given in the three categories of manufacturing, services, and small business. However, of the awards presented through 1992, ten went to manufacturing companies, four to small businesses, and only three to service firms; many consider the criteria for the award to be aimed largely at manufacturing.37 Although the Baldrige Award program has generated some controversy, it has increased attention to critical elements of perfor- mance, especially quality. It has helped spread quality-oriented attitudes, terminology, and measurements. The perceived availability of a special award in services could facilitate experiments and more widespread mea- surement of quality at the level of enterprises or firms. OTHER POLICY ISSUES IDENTIFIED BY THIS STUDY The increasing use of IT in services raises other concerns that should motivate examination of existing public policies or consideration of new policy. Those concerns relate to such issues as privacy, remote work, com- petition policy, and intellectual property. Although detailed exploration of such issues was beyond the scope of this report, the committee believed it important to identify them, to increase awareness and to signal that further analysis would be useful to determine where private actions may be insuffi- cient to meet public interests. One area where further policy action is already under debate is that of protection of an individual's rights to limit access to personal data, building on the base provided by the Privacy Act of 1974 (which was directed to the public sector). Policy measures in this area will affect choices service and other organizations can make in selecting and imple- menting information technology. For example, as discussed in Chapter 3, electronic databases are being used by more and more organizations to store personal data to support customer service operations, marketing, credit ap- provals, and other service activities. The proliferation of applications, equipment, networks, and network interconnections means that growing numbers of people have potential access to kinds of data (such as health, income, and credit records) heretofore physically less accessible.38 These circumstances
208 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY imply a greater risk of unauthorized access to or use of personal data. In addition to these potentials for abuse, there are questions about the collec- tion and review of data by employers about their employees, a capability facilitated by the use of IT, some of it in the context of electronic monitor- ing of employee behavior. This practice has already raised concern among some employees, unions, and privacy advocates as well as policymakers. Privacy is not the only consideration. The ability of service provid- ers to assemble information regarding so many different aspects of an individual's life so as to provide better service also provides the capa- bility to deny service. For example, caller-identification telephone tech- nology enables a called telephone to display the number of the calling telephone. Such technologies can be used to call up a customer profile based on telephone numbers associated with previous orders, and it can also be used to ignore or give lower priority to service requests that come, for example, from low-income regions of a city. Insurance companies can use detailed information about an individual to generate a customized insurance policy for that individual, but they can also use that information as the basis for denying insurance policies (e.g., for health care) to individuals deemed "high-risk" (e.g., those who test HIV-positive). Another impact of the use of IT with social ramifications is the facilitation of remote work.39 This effect can result in the shifting of work to other countries.40 It can also have more positive effects locally, in the context of "telecommuting" and the movement of work to where people are rather than the converse. Telecommuting whether via satel- lite offices, home offices, or other field locations will change the social structure of organizations, the geography of industries (domestically and internationally), the demand for office space, and so on. Presumably, these practices will grow because they are perceived to have performance benefits and because falling costs for IT will make them increasingly affordable and easy to implement. There may also be environmental, safety, and energy conservation benefits inasmuch as work-related travel is diminished. The potential for more home-based work could affect a wide range of people, including individuals who have difficulty getting to or functioning in con- ventional work environments; such individuals would include the disabled, people with dependent care responsibilities, and those without access to adequate transportations Shifts in activities among industries engendered by the use of IT raise questions about laws and regulations aimed at constraining the competitive behavior (e.g., pricing, collaborative ventures, mergers and acquisitions, and so on) of firms. Developments within the service sector point to increased competition across industries (and sectors), making it harder to target programs to specific industries. Increases in outsourcing and in electronic linkages to trading partners raise questions about the boundary
IMPLICATIONS FOR PUBLIC POLICY 209 between cooperation and collusion. For example, recent government atten- tion to the use of computerized reservation systems as a possible mecha- nism for collusion on pricing illustrates some of the disagreement about how information technology can and should affect competitive conduct.42 Since electronic communication among businesses and across industries is expected to expand, and since IT is expected to enable continued reorgani- zation of businesses, it may be that the goals of competition policies (such as antitrust policy) may be better served by a focus on specific kinds of transactions rather than on institutions. This is an issue that warrants fur- ther exploration by Congress and appropriate executive branch agencies.43 Another area of business (and individual) behavior where current policy might need reexamination is intellectual property rights. Protec- tion of intellectual property will be a factor affecting the rate, nature, and ownership of investments in information infrastructure, including informa- tion- and network-based service businesses. Software has already given rise to both uncertainty and assumptions regarding the existence or lack of intellectual property protection.44 Particularly relevant to services are the implications of increased electronic networking for the protection of intel- lectual property associated with materials made available electronically over networks. The debate has just begun in this area. Finally, IT could be better used to enhance the policy-making pro- cess itself. IT can be used to facilitate both the collection of highly refined data by government and the use of that data for faster and better decision making and therefore potentially better and more re- sponsive services to citizens. Such government activities as approval of eligibility for entitlement and assistance programs could be made more re- sponsive and accurate, drawing on experiences elsewhere; indeed, ongoing federal agency efforts to modernize their systems show that the potential is recognized. Improving government effectiveness by better use of IT is an area of high potential payoff already recognized by the Clinton administra- tion.45 It deserves special in-depth study.46 NOTES AND REFERENCES lIT will be present not only on its own, but also in the form of embedded components in a wide variety of equipment and products that may themselves be linked to more conventional or obvious IT. Can option contemplated by the new administration is tax credits for incremental invest- ments in equipment. See Greenhouse, Steven. 1992. "Economists Back Clinton on Invest- ment Tax Credits," New York Times, November 16, pp. D1-D2. Also, see Landau, R. 1988. "U.S. Economic Growth," Scientific American 258(June):44-52. 3Birch, David L. 1989. "Statement by David L. Birch Before the U.S. House of Repre- sentatives Small Business Committee," transcript of testimony, September. 4It must be recognized that information infrastructure in any one locale is increasingly likely to be linked to such infrastructure elsewhere, through regional, national, and interna- tional connections.
210 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY 5It should be noted that most private networking by large enterprises involves the use of "public" network facilities. Through leasing of circuits and ports and through software, real or "virtual" allocations of facilities are made. Virtual private networks, enabled by software to preempt access to certain facilities, are expected to become more available to smaller users over time. Nevertheless, conventional public infrastructure is expected to continue to provide a lifeline for smaller users who need occasional access to trading partners and service providers. 6Public policy attention to information infrastructure has already expanded considerably during the Clinton administration, and both the 102d and 103d Congresses have introduced related legislation. In addition, both the legislative and executive branches have taken steps to develop and implement a national research and education network program (the NREN compo- nent of the High Performance Computing and Communications program), which is expected to contribute to and interconnect with a broader national information infrastructure. The HPCC program is premised in part on a recognition that technology development and infrastructure are interrelated. A new fifth component is called Information Infrastructure Technology and Appli- cations. 7National Telecommunications and Information Administration. 1991. The NTIA Infra- structure Report: Telecommunications in the Age of Information, NTIA Special Publication 91-26, U.S. Department of Commerce, Washington, D.C., October. 8Cyert, Richard M., and David C. Mowery (eds.). 1987. Technology and Employment: Innovation and Growth in the U.S. Economy, National Academy Press, Washington, D.C. 9Rosenthal, Neal. 1992. "Evaluating the 1990 Projections of Occupational Employ- ment." Monthly Labor Review 115(8):32-48. 10Note that one form of sales that is fundamentally a creation of IT capability and capacity, telemarketing, appears to be growing. Meanwhile, the technology behind telemarketing (switches, storage units, call distributors, and so on) is becoming more sophisticated. 1lPodgursky, Michael. 1992. "The Industrial Structure of Job Displacement, 1979-89." Monthly Labor Review 115 (9): 17 -25. 12The epitome of this phenomenon is the transformation of employment for telephone operators. It is by now a cliche that, absent automation, two-thirds (or more) of the population might have to work as telephone operators to sustain contemporary calling volume. But rather than simply eliminate operator tasks, many were shifted to telephone company customers- telephone users are operators, based on old definitions of operator jobs. Two developments made this shift possible. One was, indeed, computerization of the telephone system, especially the installation of computer-controlled switches, computer-based databases for customer and billing information, and so on. The other was the slow but inexorable training of hundreds of millions of people to memorize huge strings of digits for input to the telephone system through a crude terminal, the touch-tone telephone. This training, or more specifically, the knowledge in the population of telephone customers regarding how to make direct-dial calls, is a vital complement to the technology in the displacement of operators. 13Baumol, William, Sue Anne Batey Blackman, and Edward N. Wolff. 1989. Productiv- ity and American Leadership: The Long View, MIT Press, Cambridge, Mass. 14An exception may apply for highly skilled occupations and individuals. There was reportedly job growth, especially in IT-related occupations, among high-skilled immigrants during the late 1970s and early 1980s. 15Carey, Susan. 1992. "Airlines Seek to Cut Back-Office Costs by Establishing Off- shore Operations," Wall Street Journal, November 30, p. B6D. 16Computer Science and Telecommunications Board, National Research Council. 1993. Computing Professionals: Changing Needs for the 1990s, National Academy Press, Washing- ton, D.C. 17Wolff, Edward N. 1991. "Productivity Growth, Capital Intensity, and Skill Levels in the U.S. Insurance Industry, 1948-86,', The Geneva Papers on Risk and Insurance 16(59, April): 173- 190.
IMPLICATIONS FOR PUBLIC POLICY 211 18Although the diffusion of IT into smaller enterprises could slow job growth within small firms, the launching of new businesses should continue to create new jobs both directly and indirectly. See Birch, David L. 1989. "Statement by David L. Birch Before the U.S. House of Representatives Small Business Committee," transcript of testimony, September. 19Note that the Fortune 500 share of total nonfarm employment has been decreasing from a peak of around 20 percent during the late 1960s and early 1970s. That share is now about 10 to 11 percent. See "Where the Jobs Aren't," Wall Street Journal, August 10, 1992, table drawing on Kemper Financial Services data. 20The most recent employment conditions reflect a recession and slow recovery. The concern in this discussion is with secular or long-term trends as opposed to the inevitable short-term shifts associated with the business cycle. 21Johnston, William B., and Arnold E. Packer. 1987. Workforce 2000-Work and Work- ers for the 21st Century, Hudson Institute, Indianapolis, Ind., June. 22See Cyert and Mowery, 1987, Technology and Employment. 23For further discussion of this point, see Zuboff, Shoshana, 1988, In the Age of the Smart Machine: The Future of Work and Power, Basic Books, New York. 24For a discussion of how many factors contribute to the growth in earnings inequality, see Grubb, W. Norton, and Robert H. Wilson, 1992, "Trends in Wage and Salary Inequality, 1967-1988," Monthly Labor Review 115(6):23-37. Note that some of the trends associated with IT could entail shrinkage in current low-wage jobs; this is already evident in forecasts for slower growth in clerical jobs, for example. Questions arise about the types of jobs that may be available for the people who would have taken such jobs. 25Kutcher, Ronald E. 1988. "Growth of Services Employment in the United States," pp. 47-75 in Technology in Services: Policies for Growth, Trade, and Employment, Bruce R. Guile and James Brian Quinn (eds.), National Academy Press, Washington, D.C. 26What is tallied where depends on whether the services represent final, as opposed to intermediate (component), output. Resources for final output are tallied directly; resources associated with intermediate output are reflected in final goods and services. This practice supports the objective of measuring trends in final goods and services, but it obscures interme- diate-production trends of interest for understanding qualitative changes in the structure of the economy. 27An exhaustive body of data may neither be practical nor justify the expense. Also, note that even if that body of measurements existed, one could not simply add industry-specific measures together to generate an aggregate economy-wide measure. But more detail than is currently available is needed to understand how activities, enterprises, industry, and the economy are changing. 28Indeed, the relative growth in employment among smaller firms and, historically, smaller firms' limited success with IT may be another factor explaining why measured service perfor- mance has been weak. 29Available data do not (and may never) capture adequately the contributions to stan- dards of living that IT may support. As measured, growth in productivity contributes to increases in the standard of living, but it is only one element. The creation of whole new industries and the increased flexibility and variety of products and services available are other elements. 30For example, computerized reservation systems for airlines and automated teller sys- tems for banks provide greater customer convenience, but, although they could not provide competitive service without these tools, even airlines and banks with leading CRS and ATM systems have had financial difficulties over the past few years. 31If funded, the Boskin Initiative would have emphasized Bureau of Economic Analysis (BEA) programs (macroeconomic data on investment and output) and Bureau of Labor Statis- tics (BLS) programs (employment and productivity measurement), as well as Bureau of the Census programs that contribute to BEA and BLS statistics. Planned Census Bureau improve
212 INFORMATION TECHNOLOGY IN THE SERVICE SOCIETY meets, for example, are aimed at better measuring the output of industries that provide mul- tiple services, better measuring the purchases of services by manufacturing firms, and better estimation of the emergence of new services. 32The committee conducted many discussions on the shift from physical to intellectual capital. Many factors obscure the value of software in the eyes of executives, investors, and policymakers more accustomed to and familiar with physical capital: software's intangibility and inscrutability to the layperson, its character as intellectual capital, and the seeming lack of durability of software, which is associated with its ease of duplication and potential to be quickly outmoded or bested by rival products. The situation has been compounded by tradi- tional perspectives in the accounting community, which are oriented toward the cost and proceeds of physical property, plant, and equipment and which reflect concern about the poten- tial for misleading financial reporting as a product of estimating the value of intangibles or making changes in financial metrics. All of these concerns are valid, but all militate against a recognition of the shift from physical and money capital to intellectual capital. 33In the case of software acquired through business acquisitions, that legislation provides for writing off the cost over 14 to 16 years. The software industry has objected that that long period will raise costs and adversely affect international competitiveness, since the useful life of software is only a few years at best. 34According to FASB literature, "The mission of the Financial Accounting Standards Board is to establish and improve standards of financial accounting and reporting for the guidance and education of the public, including issuers, auditors, and users of financial infor- mation.... The FASB develops broad accounting concepts as well as standards for financial reporting. It also provides guidance on implementation of standards...." FASB has come under some criticism lately, due in part to its support for a proliferation of accounting stan- dards, but it remains the logical entity to study this issue. 35Inasmuch as service innovation is associated with entrepreneurial firms while research funding tends to be concentrated in larger firms and universities, there may be a systemic constraint on service research funding. 36For example, Kmart developed and tested a new system to track traffic in stores and relate traffic patterns to purchase patterns. See Schwadel, Francine. 1991. "Kmart Testing Radar to Track Shopper Traffic," Wall Street Journal, September 24, pp. B1 and B5. On a larger scale, Federal Express's pioneering attempt to introduce a fax network service and AT&T Bell Laboratories' early attempts to introduce cellular telephony represented experi- ments in the application of IT. 37Although no service awards were made in 1991, awards went to the Ritz-Carlton Hotel Company and AT&T Universal Card Services in 1992 and to Federal Express in 1990. 38The public uproar over the Equifax-Lotus proposal to distribute customer profile data and controversy over the accuracy of and access to credit-agency records are but two indicators of growing public concern about data, although in general the public may not fully understand how much data about individuals is being collected and becoming available. 39CSTB has launched a new study of technology for telecommuting. 40Although some shifting from domestic to offshore employment may take place, that phenomenon may convey positive benefits in the form of improvements in foreign standards of living and diminution of immigration pressures. However, such benefits would be realized over a much longer term than the employment shifts that produce them. 41The Comprehensive National Energy Policy Act of 1992 (P.L. 102-486) directed the Department of Energy to conduct an analysis of telecommuting and its social and environmen- tal impacts, impacts that would be fostered by IT. 42Tolchin, Martin. 1992. "U.S. Sues 8 Airlines Over Fares," New York Times, December 22, p. D1. 43Those agencies would include the Office of Management and Budget, the Department of Justice, and the Federal Trade Commission.
IMPLICATIONS FOR PUBLIC POLICY 213 44Computer Science and Telecommunications Board, National Research Council. 1990. Intellectual Property Issues in Software, National Academy Press, Washington, D.C. 45Making government more efficient and more responsive, in part through the use of information technology, is one of the priorities of the technology program introduced by the federal administration in February 1993. See Clinton, William J., President, and Vice Presi- dent Albert Gore, Jr. 1993. "Technology for America's Economic Growth, a New Direction to Build Economic Strength," February 22. IT use in government was given special consider- ation in Vice President Albert Gore's National Performance Review-From Red Tape to Re- sults, distributed electronically over the Internet on September 7, 1993. 46The potential for using IT to enhance government is not new to the federal government or lower levels of government. However, the stream of reports issued by the General Account- ing Office and even by the Computer Science and Telecommunications Board attests to the significant opportunity for improvements in the acquisition, use, and management of IT in government.