Content uploaded by Xiaohu Wang
Author content
All content in this area was uploaded by Xiaohu Wang
Content may be subject to copyright.
ARPA/September2002Wang/ASSESSINGADMINISTRATIVE ACCOUNTABILITY
ASSESSING ADMINISTRATIVE
ACCOUNTABILITY
Results From a National Survey
XIAOHU WANG
University of Central Florida
This article presents the results of a national survey on accountability practices in U.S. municipal
administrations. It also examines possible causes and effects of accountability. It finds that city adminis-
trations use a variety of political and legal accountability tools. They reveal more financial information
than performance information to their stakeholders. Accountability appears to be enhanced through
eliminating government workers’concerns about exposing their performances. Responsiveness, public
consensus, and stakeholder trust are strengthened in accountable administrations. However, stake-
holder fiscal commitments and administrative service capacities arenot associated with accountability.
Administrative agencies in the U.S. government are facing increasing pressures
for accountability. For one thing, recent mismanagement practices in governments
have attracted public attention. Poor investment practices in Orange County, Cali-
fornia, led to the bankruptcy of that government in 1994. Corruption charges were
brought against public officials in Miami, Florida, in 1997. More recently, in the
2000 presidential election, government was blamed for ballot design errors, count-
ing-machine malfunctions, and voting irregularities. Inside government, legisla-
tures are demanding administrative outcome data, which require new accountabil-
ity tools such as performance measurement and citizen surveys. Finally, new
technologies such as e-mail and the Internet have been used in government. How-
ever, few systematic and empirical studies have been conducted to examine these
technologies in the context of accountability.
Using survey data, this research first examines the prevalence of accountability
practices in U.S. city governments. It then discusses possible reasons and potential
effects of these practices. This research seeks to help public managers and theorists
know what is being done nationwide for accountability purposes. More important,
it provides useful information about developing new activities, tools, and proper
strategies to make governments more accountable.
In government, accountability often refers to a relationship between elected leg-
islators and appointed administrators. Administrators need financial resources,
Initial Submission: October 30, 2000
Accepted: October 20, 2001
AMERICAN REVIEW OF PUBLIC ADMINISTRATION, Vol. 32 No. 3, September 2002 350-370
© 2002 Sage Publications
350
administrative flexibility, and decision-making discretion in public service provi-
sion. Legislators exercise control over administrators’ behaviors. The tension
between legislative control and administrative discretion intensifies as bureaucracy
grows in size, complexity, technicality, and centralization (Krislov & Rosenbloom,
1981). In this research, accountability refers to a relationship between those
entrusted with the specific tasks (stewards) and those having power of review
(reviewers). The stewards make efforts to fulfill the tasks specified by reviewers,
who are concerned with how and to what extent expected tasks are fulfilled
(Browder, 1971). Accordingly, administrative accountability concerns the extent to
which an administrative agency is answerable to its supervisory constituencies for
the tasks assigned to it. The supervisory constituencies (or stakeholders) of an
administration include not only elected officials but also the general public and the
business community. Administrative tasks include compliance with legal, organi-
zational, managerial, and financial rules and regulations, healthy financial condi-
tions, and the accomplishment of organizational goals.
FRAMEWORK
Accountability Tools
Accountability is a guiding principle in administrative decision making and ser-
vice delivery. The accountability literature has centered on two questions. One is
how to enhance accountability (accountability tools). One debate in the 1930s was
between Carl Friedrich, who emphasized the need of internal professional stan-
dards and codes in ensuring administrative responsibility, and Herman Finer, who
believed in external checks and balances (Shafritz & Hyde, 1997). Noticing that
legislation is often vague, Herring (1936) stressed professional and ethical behav-
iors in administrative discretion. Kaufman (1969) pointed out the importance of
public involvement and decentralization of decision making in ensuring account-
ability and public representation, a concept further examined and advocated by oth-
ers (Frederickson, 1971; Krislov, 1974; Krislov & Rosenbloom, 1981). Romzek
and Dubnick (1994) integrated different approaches in a model that classifies
accountability into four categories based on the source and degree of control in pub-
lic agencies. According to this model, hierarchical accountability is realized
through fulfillment of internal administrative rules and mandates. Professional
accountability enhances administrative accountability through self-imposed pro-
fessional values such as expertise and ethical codes. The external control and
request for administrative accountability come from legal accountability mecha-
nisms, which emphasize a fiduciary relation between governmental agencies and
their trustees, as well as from political accountability, which mainly concerns gov-
ernmental stakeholders’ oversight.
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 351
Accountability Contents
Unlike accountability tools, accountability contents concern what should be
accountable. In a more traditional view, administrations implement laws and poli-
cies designed by political decision makers. The contents of accountability include
the administrative compliance of laws and policies and the efficient use of
resources in administrative operations. In this review, an agency’s budgetary and
financial practices are important indicators of administrative compliance. Budget-
ary actions are monitored for their compliance with legislation. Financial transac-
tions, account activities, and fund balances are scrutinized for law violations, waste
of resources, abuse of power, loss of public properties, and malfunction of service
provisions. The ultimate goal of this financial accountability is legitimacy and effi-
ciency in the use of public resources. The tools in financial accountability include
financial monitoring, reporting and auditing, independent financial reports, and
evaluations (Meyer, 1985; Sheldon, 1996). Measures to evaluate financial account-
ability include financial ratios and service workload indicators.
However, for a long time, adherence to administrative compliance has been
blamed for inefficient operations, creation of unnecessary rules and procedures
(red tape), and irresponsible behaviors of government workers (Merton, 1997).
These problems were targets of many previous managerial reforms including per-
formance budgeting, management by objective (MBO), and total quality manage-
ment (TQM). The most recent wave of criticism is from the so-called new public
management approach (NPM) that argues that administrative compliance is guilty
for the lack of administrative initiatives and entrepreneurship. According to NPM,
administrations should be held accountable for service achievements and results,
not just operational processes and financial records. Accountability goals are speci-
fied by administrative outcome measures. Accordingly, tools and practices that
enhance outcomes are encouraged. They include quality management, perfor-
mance measurement, citizen surveys, and balanced scorecard and outcome pro-
gram evaluations (Barzelay, 1992; Osborne & Gaebler, 1992). Managers are
encouraged to pursue service outcomes in a system in which decision-making
authority is delegated to frontline managers, and rules and regulations to micro-
management and operations are minimized or eliminated (Bowsher, 1992; Osborne &
Plastrik, 1997). Since the 1990s, efforts in this performance accountability have
been intensified with many U.S. governments having adopted outcome-oriented
performance measurement systems (General Accounting Office, 1999; Melkers &
Willoughby, 1998; Poister & Streib, 1999).
Reasons for Accountability
Accountability can be used as an administrative strategy to respond to external
pressures. Political divisiveness among different groups may motivate administra-
tions to open up their decision-making process to reduce the potential unpopular or
questionable decisions (Langton, 1978a, pp. 6-8). The exposure of administrative
352 ARPA / September 2002
information may be the result of political conflicts of different parties seeking
issues for debate. A suspicious legislature may demand an open process to evaluate
budgetary and financial practices. Also, a cynical media can stimulate the account-
ability conversation by making more people aware of problems associated with
administrations (Langton, 1978a, pp. 6-8). Finally, the size of bureaucracy may be
associated with administrative accountability as the government stakeholders may
want to know more about large administrations as they are afraid of losing “per-
sonal contact” and control to a large bureaucracy (Creighton, 1981).
The tension between accountability and administrative discretion is well estab-
lished in the research literature (Krislov & Rosenbloom, 1981). Government work-
ers with the discretion to implement a policy are held accountable for the policy
consequence. Accountability is facilitated when government workers are willing to
be held accountable and when their worries for negative consequences are
addressed. Accountability is improved when government workers are active in pre-
senting new ideas to improve accountability. Finally, accountability should be
enhanced with the availability of accountability support capacities such as manage-
ment information systems and cost-based accounting systems.
Accountability Effects
Many groups have interests in government. Their demands to government vary.
Consequently, many administrative agencies have multiple and sometimes contra-
dictory goals. An open system allows government stakeholders to voice their prior-
ities and expectations to administrations. Stakeholders’choices provide legitimacy
for administrations to develop stakeholder-supported goals, missions, service pri-
orities, and performance standards (Langton, 1978b, pp. 13-24). In this sense,
accountability leads to consensus on administrative goals, service priorities, good
performances, and fiscal commitments. In addition, with more information about
available administrative resources, stakeholders may be more realistic and
acceptive about administrative capacities, performance standards, and decision
limitations (Pateman, 1970). Another accountability effect concerns responsive-
ness to public needs. A more open process allows administrations to interact more
with the public. The accountability process (such as public hearings, community
meetings) that reveals information to the public also becomes a process for people
to express their concerns and expectations to governments. The interaction helps
administrations better identify, assess, and thus satisfy public needs (King & Sti-
vers, 1998). Finally, public trust in government may improve in a more responsive
administration with strong stakeholder consensus. Improved public trust can lead
to a greater public willingness to pay for services (Denhardt & Gray, 1998; Gray &
Chapin, 1998).
Accountability can also affect administrative service capacities and decision
making. First, stakeholders’ willingness to pay for services as a result of account-
ability can enhance administrative fiscal capacities to maintain service production
levels. Also, accountability may lead administrations to adopt different decision-
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 353
making systems. The emphasis of finance accountability on financial transparency
often requires more written rules, policies, and tight internal control. This emphasis
may eventually lead to a centralized decision-making system. Conversely, the focus
on service outcomes in performance accountability requires swift responses to
stakeholder demands. Response speed is improved in a decentralized decision-
making system in which decision-making powers are delegated to frontline manag-
ers close to customers/clients. Consequently, the emphasis on performance
accountability requires a decentralized decision-making system (Bowsher, 1992).
METHODOLOGY
The Survey
A survey was conducted in late 1999 and early 2000. It was sent to all chief
administrative officers in U.S. cities with populations of more than 50,000. The cit-
ies were identified through an address file from the National League of Cities. The
names and addresses were also verified with International City/County Manage-
ment Association’s (ICMA) Municipal Year Book (1998). Among 541 surveyed
cities, 249 returned the surveys with valid information, for a response rate of 46%.
Among the respondents, 63.5% were city managers (or chief administrators) or
assistant city managers. Of respondents, 15% were chief finance or budget officers
(finance directors or budget directors). Other respondents (21.5%) include senior
management analysts, budget analysts, directors of administrations, directors of
planning, and other high-level city officials.
Three comparisons were conducted to check the nonresponse bias. First, the
population distributions of respondent cities and nonrespondent cities were exam-
ined. The population distribution of the respondent cities was compared with the
population distribution of the cities in the ICMA’s yearbook. Except for cities with
over 1,000,000 population, of which 7 of 8 responded to the survey (response
rate = 87.5%), cities within other population categories exhibited similar response
rates: 47.0% for the cities with populations from 500,000 to 1,000,000; 48.7% for
cities with populations from 250,000 to 499,999; 51.1% for cities with populations
from 100,000 to 249,999; 41.6% for cities with populations from 50,000 to 99,999.
Second, the government form of the respondent and nonrespondent cities was also
examined. Of respondents, 179 reported a council-manager form (response rate
52.7%). The response rates in other governmental forms were significantly lower
(28.7% for the mayor-council form, 12.5% for the commission form, and 20.0% for
the township form), which suggests that this sample was more representative of
council-manager governments. Third, telephone surveys were conducted to 50 ran-
domly selected city officials who did not respond to the survey (about 20% of
nonrespondents). These officials were asked two selected survey questions that
included 32 survey items. Their answers were compared with respondents’
354 ARPA / September 2002
responses. No response bias was found in this process. To ensure the validity of the
responses, the researcher also conducted follow-up telephone interviews in which
respondents were asked to verify their responses through specific examples in their
organizations. Few changes were made as the result of these phone interviews.
To gain insights into accountability practice in cities, the researcher also con-
ducted 20 in-depth interviews during the study. Interviewees were survey respon-
dents who were selected based on their comments on an open-ended question in the
survey. All of these interviewees had made extensive written comments on the item
“Please comment on the city’s achievements in accountability and the areas for fur-
ther improvement in accountability.” Interviewees were asked questions concern-
ing the achievements and obstacles in using accountability tools such as perfor-
mance measurement in their cities. Interviewees were asked to give specific
examples to elaborate their points. The results of these interviews are used to fur-
ther explore the findings of the mailed survey.
Measurement of the Variables
Measuring Accountability
In measuring financial accountability, respondents were asked to assess on a 5-
point Likert-type scale (strongly agree, agree, neutral, disagree, strongly disagree)
whether their administrations inform their stakeholders of 17 important budgetary,
financial, legal, and operational activities and trends (the items are listed in
Table 1). The exposure of these activities and trends is strongly recommended in the
Government Finance Officers Association’s (GFOA) guidelines for budgetary
practices (Government Finance Officers Association, 1993, 1998), the Govern-
mental Accounting Standards Board’s (GASB) (1999) guidelines for financial and
activities reporting, and other prominent public administration and management
literature (Rubin, 1996). A financial accountability index was then constructed
by averaging the values of the 17 items. The value of the index ranged from 1 to 5
(5 = strongly agree, 4 = agree, 3 = neutral, 2 = disagree, 1 = strongly disagree). It
indicates the extent of financial accountability. For example, a score of 4 = agree
with all finance accountability items on average. The mean of this index is 4.10; the
standard deviation is .54. The measure of internal reliability shows that this index is
highly reliable (Cronbach’s α= .92).
Similarly, performance accountability was measured through nine survey items
that describe the steps in design and implementation of a performance measure-
ment system (items listed in Table 1). The items were extracted from the perfor-
mance measurement literature (Center for Accountability and Performance, 1998;
Hatry, 1999). A performance accountability index was created through computing
average values of these nine items. Similar to the finance accountability index, the
value of this index ranges from 1 to 5. The index measures the extent of
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 355
performance accountability. The mean of this index across all respondents is 3.68;
the standard deviation is .72. The index has a high internal reliability (Cronbach’s α=
.90). This index is associated with the finance accountability index (Kendall’s τ-c =
356 ARPA / September 2002
TABLE 1: Financial and Performance Accountabilities
Association Association
With With
% Financial Performance
Survey Item Description Agreement Index Index
“Our administration informs residents, elected officials,
and the business community about...”
Financial accountability
Major capital projects, changes in capital budgets (n= 248) 97.2 .474
Significant budgetary issues, resource choices (n= 248) 96.4 .486
Major expenditure sources, estimates, trends (n= 248) 94.4 .554
Major revenue sources, estimates, trends (n= 248) 94.4 .584
Financial trends/activities (n= 248) 93.2 .499
Debt obligations, levels, and limits (n= 248) 88.4 .606
Significant changes in fund balances (n= 247) 83.1 .619
All funds subject to appropriation (n= 247) 81.1 .583
State or federal mandates (n= 247) 81.1 .502
Legal guidelines for budget balance (n= 245) 75.5 .570
Statutory/charter constraints (n= 245) 75.1 .576
Supplementary appropriation (n= 241) 69.9 .601
Procurement activities (n= 246) 66.7 .383
Major fund transfers (n= 247) 65.1 .648
Accounting basis (n= 246) 61.8 .588
Investment information (n= 243) 61.8 .596
Personnel changes (n= 247) 45.8 .399
Financial accountability index 61.0 1.000
(mean = 4.1; standard deviation = .54; Cronbach’s α= .92)
Performance accountability
Organization-wide policy priorities/goals (n= 247) 84.3 .563
Department/program goals and objectives (n= 247) 77.9 .587
Department/program functions, activities (n= 246) 76.7 .557
Citizen survey results (n= 234) 65.9 .465
Department/program output measures (n= 243) 51.4 .774
Department/program outcome measures (n= 243) 49.0 .739
Department/program narrative performance
information (n= 240) 48.6 .696
Trends of performance measures (n= 238) 39.4 .734
Interjurisdictional comparisons of performance (n= 242) 29.3 .511
Performance accountability index 37.0 1.000
(mean = 3.68; standard deviation = .72 ; Cronbach’s α= .90)
NOTE: A Likert-type 5-point scale was used in the survey (strongly agree, agree, neutral, disagree,
strongly disagree). The % Agreement indicates the percentage of respondents who agree or strongly
agree with the statement. The measure of association is Kendall’s τ-c, which estimates association
between two ordinal variables. The statistic has a range from –1 (perfect negative association) to 1 (per-
fect positive association). All associations are statistically significant at the .001 level.
.342, p< .001), which indicates that finance-accountable administrations also tend
to be performance-accountable.
In measuring the use of accountability tools, respondents were asked to evaluate,
on the same Likert-type 5-point scale, whether their administrations used each of
24 accountability tools to inform their stakeholders about accountability activities.
The selection of these tools was based on the accountability typology presented by
Romzek and Dubnick (1994). In this typology, administrations are held account-
able by “hierarchical accountability” and “professional accountability” tools (such
as municipal budgets, comprehensive annual financial reports, and audit reports),
as well as “legal accountability” and “political accountability” tools (such as public
hearings, judicial review, community newsletters, cable television, Internet). Table
2 presents the use of these 24 tools (Cronbach α= .90) and their associations with
the financial and performance accountability indices. An accountability tool
index was created to include all these items. The values of the index range from 1 to
5; 1 = strongly disagree to the use of the accountability tools, whereas 5 = strongly
agree to the use of the tools. The index has a mean of 3.75, with a standard deviation
of .56.
Measuring Accountability Reasons and Effects
Survey items were also developed to measure possible reasons and effects of
accountability. The items to measure the reasons for accountability are listed in
Table 3. They include 5 items measuring external pressures, 11 items concerning
government workers’ motives, and 4 items measuring organizational support
capacities. Respondents were requested to grade these items on a 5-point Likert-
type scale (strongly agree, agree, neutral, disagree, strongly disagree).
Survey items measuring accountability effects are listed in Table 4 (effects on
stakeholders) and Table 5 (effects on administrations). Items used to construct com-
posite indices are also presented in the table notes. All items use the same 5-point
Likert-type scale. In the survey, the researcher measured stakeholder consensus in
goal/missions (2 items, Cronbach’s α= .89), service priorities (1 item), expected
performances (1 item), and budget negotiation (1 item). In measuring responsive-
ness to the public needs, the researcher asked respondents to indicate whether their
administrations can identify, assess, or satisfy public need. Specific survey items
were derived from the literature. Three items were used to measure need identifica-
tion (Cronbach’s α= .61), two items for need assessment (Cronbach’s α= .78), and
three items for need satisfaction (Cronbach’s α= .84).
The items measuring stakeholder trust were from the recent public trust litera-
ture (Berman, 1997; King, Feltey, & Susel, 1998). This research measures stake-
holder trust perceived by public administrators. Stakeholders include citizens,
elected officials, and business communities. A citizen trust index (7 items,
Cronbach’s α= .92), an elected official trust index (4 items, Cronbach’s α= .92),
and a business trust index (3 items, Cronbach’s α= .96) were created. Also, survey
items were developed to measure stakeholders’ fiscal commitment, defined as a
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 357
government’s ability to increase taxation, debts, and budget appropriations (Hildreth,
1997). The accountability effect literature was used in measure development (King
et al., 1998; O’Toole & Marshall, 1988). The items are included in Table 4.
Efforts were also made to measure accountability effects on administrative
capacities and decision-making structures (items included in Table 5). Service pro-
duction capacities concern an administration’s ability to maintain and update its
levels of public services (Levine, 1981). The measures for service production
capacities are recommended by ICMA (Valente, 1994) to evaluate financial health
358 ARPA / September 2002
TABLE 2: Accountability Tools
% Finance Performance
Survey Item Description Agreement Accountability Accountability
“To inform residents, elected officials, and the business
community about our activities and performance,
weuse...”
Hierarchical and professional tools
Municipal budgets (n= 249) 98.8 .267** .103
Annual financial reports (n= 247) 90.8 .323** .135
Financial audit reports (n= 249) 79.5 .396** .133
Program evaluation reports (n= 239) 40.6 .340** .481**
Performance audit reports (n= 237) 39.8 .293** .341**
Performance monitoring report (n= 236) 30.1 .304** .490**
Performance appraisal reports (n= 238) 27.3 .277** .411**
Codes of ethics (n= 236) 50.6 .278** .268**
Political and legal tools
Council (commission) meetings (n= 249) 98.4 .236** .107
Legislative standing committees (n= 230) 47.4 .251** .132
Public hearings (n= 249) 96.8 .289** .180*
Community/neighborhood meetings (n= 246) 86.3 .176* .199**
Citizen advisory boards (n= 245) 80.3 .299** .244**
Citizen survey reports (n= 238) 50.6 .140 .280**
Citizen telephone hotlines (n= 237) 50.3 .147 .308**
Individual citizen representatives (n= 237) 49.0 .265** .247**
Chamber commerce meetings (n= 249) 78.7 .318** .260**
News briefing/conferences (n= 241) 63.5 .302** .163*
Cable television (n= 245) 80.7 .203** .151
Internet (n= 244) 79.9 .145 .258**
Community newsletters (n= 246) 68.3 .104 .299**
E-mail (n= 243) 55.4 .250** .293**
Sunshine laws (n= 223) 44.6 .235** .159*
Judicial review (n= 214) 13.3 .231** .222**
NOTES: A Likert-type 5-point scale was used in the survey (strongly agree, agree, neutral, disagree,
strongly disagree). The % Agreement indicates the percentage of respondents who agree or strongly
agree with the statement. Measure of association is Kendall’s τ-c.
*p< .01. **p< .001.
and service efficiency of local governments. Five items were used to measure short-
term service production capacities (Cronbach’s α= .88). Four items were used to
measure intermediate (or long-term) service capacities (Cronbach’s α= .75).
Finally, the survey measured accountability effects on administrative decision-
making structures in budgeting, personnel, and erocurement. The efforts to decen-
tralize these decision-making systems are highly recommended in the reinventing
government literature, the U.S. sibling of the NPM. The survey items (included in
Table 5) were selected with reference to the National Performance Review (Gore,
1993), a premier work of the reinventing movement. In the survey, respondents
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 359
TABLE 3: Possible Reasons for Accountability
Accountability Financial Performance
Survey Item Description Tools Accountability Accountability
Association with
External pressures
Much political competition among different groups .058 .005 .016
Critical elected officials –.068 –.010 –.099
Cynical media .055 .115 .027
Critical business community –.074 .020 –.024
Size of government (number of full-time
employees) .065 .026 .019
Government workers’ motives
Employees’ willingness to expose performances .140 .116 .230**
Employees’ support for accountability
strategies/actions .294** .186** .323**
Employees active in improving accountability .266** .179** .277**
High ethics among employees .173* .181** .192**
Employees cynical toward administration –.162* –.208** –.195**
Managers’ willingness to expose performances .168* .207** .295**
Managers’ support for accountability
strategies/actions .244** .235** .326**
Managers active in improving accountability .218** .188** .268**
Supervisors’ willingness to expose performances .210** .156* .300**
Supervisors’ support for accountability
strategies/actions .280** .198** .298**
Supervisors active in improving accountability .284** .189** .251**
Internal organizational support capacities
Management information system .241** .201** .185**
Cost-based accounting system .197** .166* .208**
Capable staff to develop goals, objectives, measures .205** .224** .286**
Budgetary surpluses for new ideas .097 .048 .045
NOTES: Except for Size of Government, which is measured by the number of full-time employees, all
other variables are measured on a Likert-type 5-point scale (strongly agree, agree, neutral, disagree,
strongly disagree). The measure of association is Kendall’s τ-c.
*p< .01. **p< .001.
were asked to evaluate 12 survey items concerning decentralization tactics in their
budgeting, personnel, and purchasing systems. A budgeting decentralization index
(four items, Cronbach’s α= .72), a personnel decentralization index (four items,
Cronbach’s α= .65), and a purchasing decentralization index (four items,
Cronbach’ α= .78) were created.
360 ARPA / September 2002
TABLE 4: Possible Accountability Effects on Stakeholders
Accountability Financial Performance
Survey Item Description Tools Accountability Accountability
Association with
Stakeholder consensus on
Service goals and objectives .397** .264** .477**
Service priorities .256** .163* .247**
Expected performances .268** .255** .300**
Budget negotiation .237** .172* .195**
Responsiveness to public needs
Identifying public needs .370** .187** .344**
Assessing public needs .363** .187** .398**
Satisfying public needs .233** .177** .252**
Perceived stakeholder trusts
Citizen trust .210** .176** .201**
Business community trust .174** .158* .175*
Elected officials trust .226** .208** .192**
Stakeholder fiscal commitment
Increasing budget appropriations .032 .066 .125
Increasing taxes without strong resistance .001 –.047 .032
Increasing debts without strong resistance .041 –.051 .060
NOTES: Several concepts in this table are measured by composite indices that include multiple survey
items. Items measuring Consensus on Goals/Objectives include “Our administration has developed
clear goals and objectives for service delivery” and “Our administration has reached consensus on goals
and objectives for service delivery.” Items measuring Identifying Public Needs include “Our administra-
tion can define customers or clients and their needs,” “Our city can provide services the public needs,”
and “Most employees understand public needs.” Items measuring Assessing Public Needs include “Our
administration frequently evaluates the achievement of goals and objectives” and “Our administration
frequently modifies goals to respond to changing public demands.” Items measuring Satisfying Public
Needs include “Our city can provide high quality of public service,” “Our city can satisfy public needs,”
and “Our city can achieve high public satisfaction for services.” Items measuring Citizen Trust include
most citizens believe that local government “meets their needs for services,” “treats citizens fairly,” “is
competent,” “understands their needs,” “can be trusted,” “fulfills its promises,” and “is honest.” Items
measuring Elected Official Trust include most elected officials “believe that the administration is effi-
cient and effective,” “believe that the administration is competent,” “believe that the administration
keeps its promise,” and “trust the administration.” Items measuring Business Community Trust include
most businesses “believe that the administration is efficient and effective,” “believe that the administra-
tion is competent,” and “trust the administration.” Survey items are measured on a Likert-type 5-point
scale (strongly agree, agree, neutral, disagree, and strongly disagree). Measure of association is Kend-
all’s τ-c.
*p< .01. **p< .001.
FINDINGS
How Much Accountability?
Financial information is widely reported to stakeholders (Table 1). More than
90% of administrations report financial information including major capital pro-
jects, major revenue, and expenditure estimates and trends. Other widely reported
information includes debt obligations, changes in fund balances, legal guidelines
for budget balance, and intergovernmental mandates. If high finance accountability
is defined as scoring on average “agree” or higher on all finance accountability
items (i.e., scoring 4 or higher on the finance accountability index), then 61.0% of
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 361
TABLE 5: Possible Accountability Effects on Administrations
Accountability Financial Performance
Survey Item Description Tools Accountability Accountability
Association with
Short-term service production capacities
Maintain existing service level .136 .066 .110
Continue paying existing services .147 .084 .112
Continue funding debts .143 .082 .107
Continue funding pensions .140 .103 .128
Maintain capital facilities .121 .115 .106
Intermediate (or long-term) service capacities
Withstand economic disruptions .163* .162* .139
Withstand effect of losing a major employer .145 .124 .136
Meet needs of population growth .134 .120 .130
Meet needs of population and income decline .131 .127 .093
Decision-making structures (decentralization)
Budgeting .112 .010 .157*
Personnel .194** .113 .190**
Purchasing .250** .246** .202**
NOTES: Several concepts in this table are measured by composite indices that include multiple survey
items. Items measuring Decentralization in Budgeting include “We allow departments to move funds
among line items,” “We consolidate accounts to minimize restrictions,” “We minimize legislative
restrictions such as line items and earmarking,” and “We allow departments to use their savings for
bonus and other incentives.” Items measuring Decentralization in Personnel include “We empower
departments to hire, promote, and fire employees,” “We abolish central and standard job application
forms,” “We give departments flexibility in job classification and pay,” and “We allow departments to
design their own performance appraisal, reward systems.” Items measuring Decentralization in Pur-
chasing include “We simplify the procurement process from rigid rules to guiding principles,” “We
allow departments to buy where they want,” “We allow departments to purchase through a simplified
process,”and “We delegate authority to departments for purchasing information.”Survey items are mea-
sured on a Likert-type 5-point scale (strongly agree, agree, neutral, disagree, strongly disagree). Mea-
sure of association is Kendall’s τ-c.
*p< .01. **p< .001.
administrations (139 of 229) are highly finance-accountable. This finding indicates
the important role of financial information in governmental reporting.
Fewer administrations reveal performance information to stakeholders (Table 1).
Although 93.2% of administrations reveal financial trends, only 39.4% report per-
formance trends. Only 48.6% have narrative performance information. Of adminis-
trations, 37.0% (83 of 224) are highly performance-accountable, defined as scoring
on average “agree” or higher on all performance accountability items. Highly per-
formance-accountable administrations tend to be highly finance-accountable. Of
highly performance-accountable administrations, 83.1% (69 of 83) are also highly
finance-accountable.
The relationship between city size (measured by the number of full-time
employees) and accountability may not exist. The measure of association (Kend-
all’s τ-c) between city size and the financial accountability index is .026. The mea-
sure between city size and the performance accountability index is .019. None of
these measures is statistically significant at the .01 significant level. Large cities do
not necessarily have high accountability.
Of 69 highly finance- and performance-accountable administrations, 76.8%
(53) have the council-manager form, 17.4% (12) have the mayor-council form. Of
administrations without high finance and performance accountabilities, 71.2%
(111 of 156) have the council-manager form, and 25.0% (39 of 156) have the
mayor-council form. The associations between government form and finance
accountability (λ= .028, p= .302) and performance accountability (λ= .025, p=
.195) are not statistically significant at the .01 level.
This study also finds that city location is not significantly associated with
accountability. Cities in the north central area score 4.16 on the financial account-
ability index, compared with 4.12 in Northeast cities, 4.05 in Southern cities, and
4.10 in Western cities.1These differences are not statistically significant (F= .456,
p= .714 for a one-way analysis of variance test). Cities in the South score slightly
higher on the performance accountability index (3.75) than cities in the Northeast
(3.54), north central (3.55), and West (3.73). However, these differences are also
not statistically significant (F= 1.177, p= .319 for a one-way analysis of variance
test).
What Accountability Tools In Use?
City administrations use a variety of accountability tools (Table 2), classified in
this research according to Romzek and Dubnick (1994). Among hierarchical and
professional accountability tools, financial documents such as budgets, financial
reports, and financial audit reports are widely used. The use of tools is positively
associated with financial accountability (Table 2). Evidently, these tools deliver
financial information to stakeholders. On the other hand, use of these financial tools
is not significantly associated with the performance index, which suggests the limi-
tation of using financial tools to carry performance information.
362 ARPA / September 2002
Performance tools (i.e., program evaluation reports, performance audits, perfor-
mance monitoring reports, and performance appraisal reports) are not as popular as
financial tools (Table 2). For example, fewer than one third of cities (27.3%) report
performance appraisal results to stakeholders. Nevertheless, the use of perfor-
mance tools is positively associated with both finance and performance accounta-
bilities. Administrations using these tools have higher financial and performance
accountabilities. In fact, administrations that use performance-monitoring reports
are 8.5% higher in finance accountability and 20.9% higher in performance
accountability than those not using the reports (measured by the financial index and
the performance index scores). This finding suggests a need to intensify the use of
performance tools to improve accountability.
Among political and legal tools, legislative meetings and standing committees
are commonly used. Use of them is associated with financial accountability.
Administrations appear to use these meetings to reveal financial information
required by legislatures. The use of citizen-oriented tools in accountability (public
hearings, community/neighborhood meetings, citizen advisory boards, and others)
varies from 49.0% in individual citizen representation to 96.8% in public hearings.
About half of administrations use citizen survey reports. All citizen-oriented tools
are positively associated with the performance index and the relationships are sta-
tistically significant. Administrations using these tools have higher performance
accountability. For example, administrations that use citizen advisory boards are
15.3% higher in performance accountability than administrations not using this
tool (measured by performance index scores). With the increase in use, citizen-
oriented tools are becoming a major vehicle to deliver performance information. It
is expected that the administration-to-citizen communication is characterized by
performance conversations. Administrations should ponder a way to communicate
effectively with citizens about administrative performance.
City size (measured by the number of full-time employees) is not significantly
associated with the accountability tool index (Kendall’s τ-c = .065, p= .174). This
finding indicates that larger cities may not always use more accountability tools.
Also, the government form is not significantly associated with the accountability
tool index (λ= .018, p= .081). Council-manager governments score 3.77 on the
accountability index, whereas mayor-council governments score 3.70, and other
governments score 3.90. Finally, cities in the Northeast score 3.82 on the account-
ability tool index, compared with 3.73 in north central cities, 3.82 in Southern
cities, and 3.70 in Western cities. These differences are not statistically significant
(F= .585, p= .625 for a one-way analysis of variance test). City location is not sig-
nificantly associated with accountability.
Why Accountability?
Table 3 demonstrates the possible reasons for administrative accountability.
Stakeholder external pressures have no clear influences on administrative account-
ability. No relationship has been found between accountability and political
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 363
divisiveness, critical elected officials, or the cynical media. Also, the relationship
between the size of government and accountability indices is positive but not statis-
tically significant. The argument that accountability increases in large administra-
tions because stakeholders are afraid of losing control to large bureaucracies is not
strongly supported by the data.
The data, however, do show that accountability is associated with government
workers’ motives. All three accountability indices (tool, financial, and perfor-
mance) are associated with the government workers’desires, supports, and actions
in accountability. Accountability information is more likely to be revealed in
administrations with workers motivated for accountability. The associations
between government workers’ motives and the performance accountability index
are particularly high (from .192 to .323). For example, cities with “employees sup-
porting for accountability strategies and actions” score 3.97 in the performance
accountability index (n= 102), 15.7% higher than the cities not seeing such support
(score 3.43 for n= 112). These findings indicate an important role of public
employees in performance measurement. In a follow-up interview, a senior admin-
istrative officer described an example in his city that the police employees changed
their attitudes toward performance measurement from angry to acceptable as they
understood the need for the system. The police employees now submit performance
reports on a regular basis, and the performance information is pooled and specifi-
cally applied to measuring policing.
Table 3 also shows a significant relationship between accountability and an
administration’s support systems. Management information systems and cost-
based accounting systems appear to increase accountability. Capable staff mem-
bers with abilities to develop goals, objectives, and performance measurement are
important properties for administrative accountability. Cities with capable staff
members score an average 3.72 on the performance accountability index, 18%
higher than the cities without such staff (3.15).
What are Accountability Effects?
Table 4 shows possible effects of accountability on stakeholders. Accountability
indices are found to be positively associated with stakeholder consensus on goals,
service priorities, and expected performances. Of respondents from highly perfor-
mance-accountable cities, 81.9% believe that their cities can achieve consensus on
service priorities, compared to only 55.3% of respondents who say so in the cities
without high performance accountability. Similarly, 79.5% of respondents from
highly performance-accountable cities say their cities can achieve public consen-
sus on good service performance. Only about 50.0% of respondents from the cities
without high performance accountability believe so. These findings may indicate
that cities feel it is easier to reach consensus with an informed stakeholder, and that
an informed stakeholder is more likely to adjust his/her demands to administrative
limitations and capacities.
364 ARPA / September 2002
Accountability is also positively associated with administrative responsiveness
to public needs (Table 4). Accountable administrations are more responsive to pub-
lic needs. For example, 85.0% of the respondents from highly performance-
accountable cities believe that their cities can define customers/clients and their
needs, compared to 56.0% of the respondents who say so in the cities without high
performance-accountability. Also, 79.4% of the respondents from the cities that
use citizen surveys (an accountability tool) say that their cities frequently modify
goals to respond to changing public demands (n= 126). Only 61.8% of the respon-
dents from the cities without doing citizen surveys say so (n= 55). In a follow-up
interview, a city manager gave an example that frequent communication kept the
city informed about citizen needs in snow removal and street safety, and that citizen
needs directed the city funding. It appears that accountable administrations show
stronger abilities to identify, assess, and satisfy public needs. Indeed, the account-
ability process (public hearing, community meetings, citizen advisory boards, and
so forth) that reveals information to the public is also the process for the public to
channel their concerns and expectations to administrations. Public needs are better
identified and assessed through this process.
The findings show accountability is associated with stakeholders’trust (Table 4).
Highly accountable administrations perceive high stakeholder trust. More specifi-
cally, for example, elected official trust (measured by the elected official trust
index) is 11.0% higher in high finance-accountable administrations and 7.8%
higher in performance-accountable administrations. In high performance-account-
ability administrations, 98.8% of cities agree or strongly agree that most elected
officials believe that the administration is competent, compared to 85.8% of cities
without high performance-accountability saying so. Also, there is some evidence
that elected officials may become less critical to administrations in accountable
administrations. The relationship between the performance-accountability index
and the survey item indicating most elected officials are very critical to the adminis-
tration, may exist (Kendall’s τ-c= –.100, p< .10, n= 243). These findings provide
evidence that accountability enhances stakeholders’confidence in administrations
and thus improves stakeholder trust toward administrations.
However, accountability efforts may not result in stakeholder fiscal commit-
ments in budget appropriations, taxation, and debts. The relationship between fis-
cal commitments and accountability indices is not statistically significant (Table 4).
Accountability efforts do not necessarily lead to taxpayers’ willingness to pay and
improved fiscal conditions in cities. Organizations that target accountability should
not expect improved fiscal capacities as an immediate return.
This study also examines the possible effects of accountability on administrative
service capacities and decision-making structures (Table 5). No strong relationship
has been found between accountability and service capacities (short-term and long-
term). Accountability does not lead to an immediate improvement in service pro-
duction capacities (Table 5). This result should not be surprising, given that
accountability is not associated with stakeholder fiscal commitments.
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 365
Finally, this study measures decentralization of administrative decision making
and its relationship with accountability. It finds that cities use a variety of decentral-
ization practices from 3% abolishing central and standard job application forms to
79.6% allowing departments to move funds among line items. It also finds that
decentralization is associated with performance accountability (Table 5). High per-
formance-accountable administrations tend to have more decentralized decision-
making systems. For example, 25.3% of high performance-accountable cities agree
or strongly agree that they allow departments flexibility in job classification and
pay. Only 12.1% of administrations without high performance accountability say
so. These findings may indicate that performance accountability efforts, such as
developing performance-oriented goals, increase an administration’s confidence in
its lower level decision units (departments). Indeed, the power shift to lower level
decision units from the central management should become easier when these units
can be held accountable for their performance.
CONCLUSION
This study examines administrative accountability. It finds that financial infor-
mation is revealed more than performance information in U.S. cities. Financial
documents are more common than performance reports. Various political and legal
channels are used for accountability purposes. Accountability appears to be pro-
moted by government workers’willingness, support, and action in accountability.
Responsiveness, public consensus, and stakeholder trust are improved in account-
able administrations. Elected officials are more confident and trustful in account-
able administrations. Decision-making power is delegated to lower level decision
units in performance-accountable administrations. However, the improvement of
stakeholder fiscal commitments and administrative service capacities is not a result
of accountability.
How to enhance administrative accountability? This research finds a significant
relationship between accountability and government workers’ motives. On the
other hand, accountability is not associated with stakeholder pressures at a signifi-
cant level. This finding indicates an important role of government workers in
accountability enhancement. Stakeholder pressures may force administrations to
open up temporarily. The media exposure, the elected officials’ inquiry, and inde-
pendent auditors’ investigations may force administrations to reveal some informa-
tion. The real challenge in accountability enhancement, however, is to win govern-
ment workers’support in accountability action. This research finds that only 34.1%
of respondents believe that their employees are willing to expose their job perfor-
mances to public scrutiny and only 45.8% say that their employees support the
strategies and actions that make them more accountable. Indeed, there is room for
improvement in this regard.
These findings suggest a management strategy that may enhance administra-
tive accountability. The strategy focuses on motivation of government workers. It
366 ARPA / September 2002
requires the elimination of employees’ concerns for exposing their job perfor-
mances through development of realistic performance targets, articulation of expected
performance, and creation of due process for employee complaints. Ethics training
is needed to promote the idea that public interest is a meaningful concept that can be
realized through accountability practices and that it is important to balance expedi-
ency and fairness, as well as discretion and responsiveness in service delivery. Man-
agers and supervisors should be leaders in exposing their job performances and
responsible for overall performance targets. An effective strategy also requires a
free environment in which employees can discuss performance issues with their
supervisors and managers.
What to expect from accountability? Accountable administrations experience
more stakeholder consensus on service goals, priorities, and expected perfor-
mances. They also perceive enhanced stakeholder trust. However, they should not
expect that their fiscal capacities in taxation, debt, and budget be dramatically
improved as the result of accountability efforts. There were some anecdotes in the
case study portion of this research that link accountability and fiscal capacity. In a
city that had citizen approval for a sales tax increase for a new public safety center,
the city manager contributed the favorable citizen vote to administrative efforts in
bringing citizens in on the feasibility study. However, the linkage may not exist for
many other governments. Stakeholder fiscal commitment and an administration’s
service production capacity may be influenced by factors other than administra-
tions’ accountability efforts. Indeed, accountability is a goal of management, not a
means to achieve fiscal or service production competency. Governments should
look for other ways to improve fiscal and service production capacities. The moti-
vation for administrative accountability rests on the desire to serve the public better
rather than on its possible immediate financial benefits.
Performance accountability has been advocated in government recently. This
study shows that performance accountability may be better accomplished in decen-
tralized decision making in which frontline managers are responsible for their
performance achievements. However, the drive for decentralization poses a chal-
lenge for accountability. As micro-management rules and regulations are mini-
mized, managers are accountable for their performance results. How to ensure that
empowered managers use limited resources properly? How to keep managers flexi-
ble in managing for results and, at the same time, to ensure that organization-wide
goals are not being compromised? Generally, how to achieve entrepreneurship in
government with a proper level of control? One interviewee advocated the use of
intensified monthly evaluation on the most important projects in an administrative
agency, suggesting that such evaluation should be conducted by central manage-
ment offices (city manager office, finance office, or management office), which are
released from micro-management duties in a decentralized system. The evaluation
should focus on both financial and performance data. It should provide managers
prompt feedback concerning goal achievement and law conformity. Indeed, we
have seen a serious drive for decentralization in privatization and reinventing
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 367
government items. However, failures in controlling and feedback can result in
financial disasters (Jereski, 1996) and loss of lives (Romzek & Ingraham, 2000).
This study targets city administrations. The generalization of these findings to
other levels of government should be made with caution. For example, the sheer
size of many federal agencies and the constant party politics in Washington may
force these agencies to focus more on the certain accountability tools (such as pub-
lic hearings, the Internet, e-mail) and less on others (such as citizen/customer meet-
ings). This research finds no significant effect of government form on accountabil-
ity. However, as this sample has a relatively low response rate for mayor-council
and commission-form governments, further efforts are still needed to verify the
findings for these governments.
NOTE
1. Cities were classified according to the states in which they were located. International City/
County Management Association’s (ICMA) classification is used. States covered in the Northeast
include Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylva-
nia, Rhode Island, and Vermont. States covered in the north central area include Illinois, Indiana,
Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and
Wisconsin. Southern states include Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky,
Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas,
Virginia, and West Virginia. States in the West include Alaska, Arizona, California, Colorado,
Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming. See The
Municipal Year Book (International City/County Management Association, 1999). The South pro-
vided 30.9% of cities in the study,the West 35.4%, the Northeast 10.8%, and the north central 22.9%.
REFERENCES
Barzelay, M. (1992). Breaking through bureaucracy. Berkeley: University of California Press.
Berman, E. M. (1997). Dealing with cynical citizens. Public Administration Review,57(2), 105-112.
Bowsher,C. A. (1992). Performance measurement: An important tool in managing for results.Washing-
ton, DC: General Accounting Office.
Browder, L. H. (1971). Introduction: Emerging patterns of administrative accountability—A point of
view. In L. H. Browder Jr. (Ed.), Emerging patterns of administrative accountability (pp. 1-25).
Berkeley, CA: McCuthchan.
Center for Accountability and Performance. (1998). Performance measurement: Concepts and tech-
niques. Washington, DC: American Society of Public Administration.
Creighton, J. L. (1981). The public involvement manual. Cambridge, MA: Abt.
Denhardt, R. B., & Gray, J. E. (1998). Targeting community development in Orange County, Florida.
National Civic Review,87(3), 227-235.
Frederickson, H. G. (1971). Toward a new public administration. In F. B. Marini (Ed.), Toward a new
public administration (pp. 309-331). London: Chandler.
General Accounting Office. (1999). Managing for results:Opportunities for continued improvements in
agencies’ performance plans (GAO/GGD/AIMD-99-215). Washington, DC: Author.
Gore, A. (1993). Creating a government that works better and costs less: Reports of the national perfor-
mance review. Washington, DC: Government Printing Office.
368 ARPA / September 2002
Governmental Accounting Standards Board. (1999). Statement No. 34 of GASB: Basic financial state-
ments and management’s discussion and analysis—for state and local governments. Norwalk, CT:
Author.
Government Finance Officers Association. (1993). Distinguished budget presentation awards program:
Awards criteria. Chicago: Author.
Government Finance Officers Association. (1998). Recommended budget practices: A framework for
improved state and local government budgeting. Chicago: Author.
Gray, J. E., & Chapin, L. W. (1998). Targeted community initiative: Putting citizens first.In C. King &
C. Stivers (Eds.), Government is us: Public administration in an anti-government era. Thousand
Oaks, CA: Sage.
Hatry,H. P. (1999). Performance measurement: Getting results.Washington, DC: Urban Institute Press.
Herring, E. P. (1936). Public administration and the public interest. New York: McGraw-Hill.
Hildreth, W.B. (1997). Financial management: The centrality of the fiscal in local government and poli-
tics. In J. J. Gargan (Ed.), Handbook of local government administration (pp. 159-190). New York:
Marcel Dekker.
International City/County Management Association. (1998). The municipal year book. Washington,
DC: Author.
International City/County Management Association. (1999). The municipal year book. Washington,
DC: Author.
Jereski, L. (1996). How a rescue mission failed, just barely, in Orange County. In R. Stillman II (Ed.),
Public administration concepts and cases. Boston: Houghton Mifflin.
Kaufman, H. (1969). Administrative decentralization and political power. Public Administration
Review,29(1), 3-15.
King, C. S., Feltey,K. M., & Susel, B. O. (1998). The question of participation: Toward authentic public
participation in public administration. Public Administration Review,58(4), 317-326.
King, C. S., & Stivers, C. (1998). Government is us: Public administration in an anti-government era.
Thousand Oaks, CA: Sage.
Krislov, S. (1974). Representative bureaucracy. Englewood Cliffs, NJ: Prentice-Hall.
Krislov, S., & Rosenbloom, D. (1981). Representative bureaucracy and the American political system.
New York: Praeger.
Langton, S. (1978a). Citizen participation in America: Current reflections on the state of arts. In S. Langton
(Ed.), Citizen participation in America (pp. 1-12) Lexington, MA: D. C. Heath.
Langton, S. (1978b). What is citizen participation? In S. Langton (Ed.), Citizen participation in America
(pp. 13-24). Lexington, MA: D. C. Heath.
Levine, C. H. (Ed.). (1981). Managing fiscal stress (p. 12). Chatham, NJ: Chatham House.
Melkers, J., & Willoughby, K. (1998). The state of states: Performance-based budgeting requirements in
47 of 50. Public Administration Review,58(1), 66-73.
Merton, R. K. (1997). Bureaucratic structure and personality. In J. M. Shafritz & A. C. Hyde (Eds.),
Classics of public administration (4th ed., pp. 100-108). New York: Harcourt Brace.
Meyer, M. W. (1985). The concept of rational administration. In J. S. Ott, A. Hyde, & J. Shafritz (Eds.),
Public management: The essential readings (pp. 174-188). Chicago: Lyceum Books/ Nelson-Hall
Publishers.
Osborne, D., & Gaebler, T. (1992). Reinventing government. New York: Addison-Wesley.
Osborne, D., & Plastrik, P. (1997). Banishing bureaucracy: The five strategies for reinventing govern-
ment. Reading, MA: Addison-Wesley.
O’Toole, D. E., & Marshall, J. (1988). Citizen participation through budgeting. The Bureaucrat,17(2),
51-55.
Pateman, C. (1970). Participation and democratic theory. Cambridge, United Kingdom: University of
Cambridge Press.
Poister, T., & Streib, G. (1999). Performance measurement in municipal government: Assessing the
state of the practice. Public Administration Review,59(4), 325-335.
Wang / ASSESSING ADMINISTRATIVE ACCOUNTABILITY 369
Romzek, B. S., & Dubnick, M. (1994). Issues of accountability in flexible personnel system. In
P. W. Ingraham & B. S. Romzek (Eds.), New paradigms for government: Issues for the changing
public service (pp. 263-294). San Francisco: Jossey-Bass.
Romzek, B. S., & Ingraham, P. W. (2000). Cross pressures of accountability: Initiatives, command, and
failure in the Ron Brown plane crash. Public Administration Review,60(3), 240- 253.
Rubin, I. (1996). Budgeting for accountability: Municipal budgeting for the 1990s. Public Budgeting
and Finance,16(2), 112-132.
Shafritz, J. M., & Hyde, A. C. (1997). Classics of public administration (4th ed.). Fort Worth, TX: Har-
court Brace.
Sheldon, D. R. (1996). Achieving accountability in business and government: Managing for efficiency,
effectiveness, and economy. Westport, CT: Quorum.
Valente, M. G. (1994). Evaluating financial condition: A handbook for local government. Washington,
DC: International City/County Management Association.
XiaoHu Wang is a faculty member in the Department of Public Administration at the University
of Central Florida. His research and teaching interests include public budgeting and financial
management, performance management, and comparative public administration. His research
has appeared in many academic journals.
370 ARPA / September 2002
