1、 外文翻译 原文 Looking into the Crystal Ball: Performance Management over the Next Decade Material Source: Public Administration Review; Dec2010 Supplement, Vol. 70, p s 2 0 8 - s 2 11 , 4 p A u t h o r : H a t r y, H a r r y P Introduction This essay focuses on the performance management elements of publ
2、ic administration. It addresses both performance measurement (measurement that focuses on outcomes as well as outputs) and the use of performance information in management (called performance management). As of 2010, this “new” performance measurement has been widely touted around the world, particu
3、larly in the United States. The 1993 Government Performance and Accountability Act has been a major catalyst for this, bringing the big elephantthe federal governmentinto the movement. A major formal performance measurement effort was started many years earlier in local government. A few cities, suc
4、h as New York City and Charlotte, North Carolina, have very long sustained performance measurement efforts, ones dating back to the early 1970s. State governments began formal performance measurement work in the early 1990s, with Texas and Oregon leading these state efforts. Some estimate that at le
5、ast 35 to 40 states currently have some form of legislated performance measurement requirement (see Lu, Willoughby, and Arnett 2009). Not clear is the extent to which governments have used performance information on a regular basis to help guide agency planning, budgeting, and operating management d
6、ecisions. The use of performance information appears to be particularly weak, especially at the federal and state government levels (see, e. g., GAO 2008, 2009). The following is one persons crystal-ball gazing on the next decade. The prognostications first address performance measurement and then t
7、he use of performance information, a major step that transforms performance measurement into performance management. Performance Measurement Managers and their staff s will have an enormous amount of data at their fingertips. Technology advances will continue at breakneck speed, in areas such as com
8、puting and communications. For example, performance information will be readily accessible on small geographic areas such as neighborhoods, census tracts, and even street blocks. Performance information will be readily accessible to supervisors on each supervisors own coverage area, such as precinct
9、s, divisions, field offices, and so on. GIS and mapping software are already making much more detailed and more analytically based information available. Comparison information will be readily available from many other similar public agencies in the United States and other countries as well. Getting
10、 systematic citizen feedbackfrom all segments of the populationwill become accepted as a major way to obtain reasonably reliable data on service quality and outcomes. Federal agency “Performance and Accountability Reports” already have begun to contain such indicators. Public officials will be able
11、to survey samples of citizens at almost a moments notice. Most citizens will have one or more communication devices enabling this to occur. However, this will raise issues about how representative the surveys are. (Citizens will become more accustomed to providing input to public agencies, but still
12、 will be highly selective as to which government surveys they respond to.) Performance data in the future will be able to be processed continuously and in real time. Managers will be able, for emergency and last-minute situations such as in budget hearings, to respond almost immediately to issues ra
13、ised. Tracking of clients after they have left services will become more fashionable. Currently, public service agencies, particularly human services agencies, resist follow-ups with clients as too difficult, too expensive, and beyond their responsibility. But how can programs know whether what they
14、 are doing is working unless such tracking is done? Agencies and their funders will increasingly recognize this need to follow up on the progress of past clients. Better technology and follow-up procedures will enable programs to do this more easily. For human services, software packages will enable
15、 programs to track an individual clients progression through many services and agenciesand to link outcome information to individual clients. A number of software firms are already developing workable case management systems. Such systems record the types and amounts of interventions provided to eac
16、h individual client and link them to the clients demographic data. Using such procedures as follow-up surveys of clients will enable programs to link service and demographic information to outcomes. Program managers will be considerably better able to identify what works well and what does notand fo
17、r what types of clients. The new technology will continue to raise privacy concerns. Such problems probably will not have been fully solved by 2020. For example, personal health data will be readily available to appropriate medical personnel, but such access will likely be at least partly controlled
18、 by each citizen. Citizens will be able to file complaints and requests for services in considerably more convenient ways. Advanced 311 systems will become the norm. Information on the number, type, and resolution of complaints and requests will be much better tracked and reported for individual ser
19、vice agencies. Public agency websites will become easier to use, especially as the public becomes more proficient in their use. Performance data posted on these websites will be updated much more frequently and will be more time. The websites will have a large variety of readily available links for
20、citizens who want details. Considerably more attention will be paid to across-agency, across-sector issues. More formal “performance partnerships” will be formed. Partners will agree on the outcomes to track and the targets for these outcomes, and they will identify each partners role and responsibi
21、lities in producing those outcomes. The cost of services will be a normal part of management information systems, with cost being related to outputs and outcomes. Cost information will become commonly used to evaluate the efficiency and cost-effectiveness of services. Public administrators will be a
22、ble to obtain performance information on successful and unsuccessful practices from an increasing number of program evaluations and meta-program evaluations. Meta-evaluations, which systematically examine multiple evaluations on the same topic, have begun to be used, beginning in the health and crim
23、inal justice fields. The cost of services will be a normal part of management information systems, with cost being related to outputs and outcomes. Cost information will become commonly used to evaluate the efficiency and cost-effectiveness of services. Managers in public service agencies will begin
24、 testing new, or significantly revised, practices and policies using randomized assignment of customers whose outcomes would then be tracked and compared. For example, social service agencies will be able to compare alternative ways to deliver services, such as whether individual or group interventi
25、ons work best or which service delivery technology is most effective. This will provide managers with stronger evidence for program choices. Comparisons of outcomes across jurisdictions will commonly be made and include examinations of service-relevant characteristics to provide more informative (an
26、d fairer) comparisons. Interagency comparisons have begun to sprout. This interest will expand considerably, especially as more performance related data become available. Few services will be able to escape such comparisons. In this authors experience, many, if not most, government managers and elec
27、ted officials are interested in comparisons with other similar types of organizations. However, they are not enthusiastic about having such comparisons reported externallyunless they feel they would do well. More than 10 years ago, the International City/County Management Association introduced its
28、comparative performance measurement process, covering a number of basic local government municipal services. The North Carolina Benchmarking Project and the Florida Benchmarking Consortium are among other efforts to compare local government performance. Possibly the granddaddy of all U.S. comparison
29、s is the Federal Bureau of Investigations local crime report comparisons. Along with this movement, there will be major pressure for agencies to report results on common sets of core indicators for each individual service. The development of core performance indicators will mushroom. Funders, both g
30、overnmental and nongovernmental organizations such as foundations, are increasingly pressing nonprofit service organizations for outcome information that they can use to compare the past performance of applicants for funding. A primary issue is obtaining agreement on which indicators and data collec
31、tion procedures to use. (Dropout rates and response times for emergency calls are classic examples of definitional problems.) Perfect comparability is not possibleexcept when a government, particularly the federal government, collects a common set of data from jurisdictions, as it currently does thr
32、ough its many national surveys. More focus will be given to developing analytical tools to help public officials make decisions about the future. In recent decades, most analysis has focused on measuring the outcomes of past performance. Performance measurement and program evaluation are “looking ba
33、ck” procedures. However, decisions are about the future. Methods for looking into the future (often called policy or program analysis) are currently sparse. Since the 1960s, the U.S. Department of Defense has used its “systems analysis” office to employ analytics to estimate the effectiveness of its
34、 future weapon systems. However, in nondefense departments, systematic examination of the future has been much less practiced. Indeed, it is enormously difficult to predict the future. Decisions require assumptions about the future environment, future outcomes, and future costs. Data on past perform
35、ance have been the major source of information used by government agencies to estimate future outcomes and costs. For short-term decisions, such as next years budget, recent performance information is highly useful. However, for decisions with important out-year implications, historical data become
36、increasingly less reliable. A variety of statistical techniques are used to help make projections, such as regression analysis, trend analysis, and simulations. In the future, with increasingly powerful computer technology, we can expect more sophisticated analytical techniques developed to help pub
37、lic officials see into the futurethough still with mixed results. Using Performance Information: Moving to Performance Management Performance information can be used for a number of managerial purposes, such as day-to-day resource allocation, motivating employees, motivating contractors, developing
38、and justifying capital and operating budgets, communicating to the public, and, ultimately, improving the effectiveness of services. The following are prognostications about obstacles and progress. Public administrators will face major problems and pressures in handling, sorting through, using, and
39、channeling the enormous amount of data available through many different media. This will be a major challenge. Many books and articles will be written that will attempt to address this growing information overload. The burgeoning availability of performance information to the media and public indica
40、tes that public administrators will need to steel themselves even more for regular media second-guessing and citizen and citizen-group criticism. The exponential increase in performance information will help public administrators make improved choices, but also will lead to much misinterpretation an
41、d misuse of that informationboth intentional and unintentional. We may see a growth in “truth squads” to sort out poor use and misinterpretation, creating a new profession of truth-tellers. The medias interest in reporting bad, rather than good, performance news will inevitably continue. How to mana
42、ge bad news will occupy lots of the attention of public administrators. There will be a full emergence at all levels of government of “How Are We Doing?” processes, the movement currently labeled the “Stat” approach. The basic characteristic of this approach is that a public official meets regularly
43、 with his or her staff to discuss the latest performance data to identify what is working, what is not working, and what can be done to improve services. This simple concept has considerable potential for helping managers build into their organizations a focus on getting results and continual servic
44、e improvement. A number of local governments have already introduced this process. Most well known are the New York City CompStat and Baltimore CitiStat processes. State governments have begun to embrace it (most notably, thus far, Washington and Maryland). This is a natural progression in the use o
45、f performance information as such information becomes more readily available and in a more timely way. “How Are We Doing?” approaches are likely to be introduced in most public service agencies, and not just at the top levels (such as by mayors and governors). However, the approach seems equally val
46、uable at lower levels, as done by a number of New York City departments. The process will likely become a subject for schools of public administration. The current Stat movement emphasizes elaborate data reporting and displays. A stepped-down version without the sophisticated machinery is likely to
47、appear, which will be practical for most public agencies, large or small. “What Works/Effective Practices” information, using performance data to help identify what works, will be used by public administrators. The federal focus, thus far, has been on obtaining evidence from in-depth program evaluat
48、ions. These evaluations are very important. However, because of their costs and time requirements, they provide highly limited coverage of the many public services for which “what works” information is needed. The data from regular performance measurement efforts, thus, will likely be the major sour
49、ce of “what works” information, even though that evidence is weaker than that from in-depth program evaluations. Federal government elected officials, and some state and local governments, will continue to push for performance-based employee pay systemsand will continue to face employee resistance and difficulties in making a strong case that these systems are cost-effective. More experiments will be made with rewards based primarily on outcome information. The federal government has wrestled with pay-for-
