Can You Find It on the Web? An Assessment of Municipal E-Government Transparency

Abstract

Despite the potential of open government, earlier research has found that local governments vary significantly in their embrace of transparency. In this article, we explore the variability question through the innovative application of an alternative set of transparency indicators. We find that cities are more likely to make information about finance and budgeting and general administration accessible to the public, less likely to place information related to human resources online. We use the literature to derive a model to test five types of promising explanations for a city’s propensity for transparency. Our analysis suggests that community demand and a city’s organizational networks play an important role in fostering transparency, regardless of city size. Important differences do exist between large and small cities: Transparency in larger cities is spurred by political competition; in smaller cities, governmental resources and administrative professionalism influence transparency.

Keywords

transparency e-governance open government local government governance

In 2013, with great fanfare, the city of Los Angeles launched a new website, Control Panel LA, intended to make city finances more transparent to the public. As the city controller said, “Knowledge is power, and this initiative is providing both to the people of Los Angeles . . . We have an historic opportunity to make our government faster, more efficient, transparent to the public, in truly unprecedented ways” (Finnegan & Welsh, 2013). Massive amounts of city financial data—revenues, expenditures, budgets, payrolls, audits, trends—are categorized and presented as lists, charts, and maps on Control Panel LA. The effort to make city government more transparent was expanded later that year when the mayor of Los Angeles issued an Executive Directive establishing an Open Data initiative requiring each of the city’s 41 departments to post online, with few exceptions, any data they collect and generate. A performance tab on the mayor’s website takes the viewer to Data LA, which contains an extensive set of performance data for city departments organized functionally. One can learn how long it took for the city’s Office of Community Beautification to respond to requests to remove graffiti or how many potholes the Street Resurfacing and Reconstruction Division filled and compare those figures with earlier time periods and with performance targets.

The city of Los Angeles has put these data online with the avowed intent of promoting transparency and accountability. The expectation is that access to data will empower citizens, increase their level of engagement in governance, foster creative problem solving, and, ultimately, result in more effective government (Garcetti, 2013). Yet questions remain as to the extent to which city governments have embraced transparency. In this research, we analyze the amount and types of data municipalities post on their websites. Our goal is to determine what explains cities’ propensity for transparency.

In the next sections of the article, we consider the concept of transparency and we develop a model that we use to understand the drivers of municipal transparency. We also introduce the framework developed in the Government Performance Project (GPP) from which we derive the management and performance indicators that produce the transparency scores—the dependent variable in the model. Following that, our methodology for evaluating the websites of more than 200 city governments in Texas is presented. After discussing the results of the website evaluation, we conduct the determinants analysis. Our findings suggest that some of the explanations for transparency are similar across large and small cities, others vary. In larger cities, organizational networks, political competition, and demand for transparency are the strongest predictors. However, in smaller cities, while community demand and organizational networks remain important, governmental resources and the professionalism of the city manager play an important role. We conclude the article with ruminations on the nature of local government transparency.

Transparency in Local Government

The concept of transparency has multiple definitions. Consequently, perspectives vary as to what is—and what is not—governmental transparency (Welch, 2012). The unifying characteristic of the differing definitions is accessibility to governmental information or, from the perspective of government, “the proactive dissemination of information” (Piotrowski, 2011, p. 3). For our purposes, the definition used by Grimmelikhuijsen and Welch (2012) is apropos: “Transparency is defined as the disclosure of information by an organization that enables external actors to monitor and assess its internal workings and performance” (p. 563). An essential aspect of this definition is the notion that the information being released by the organization is relevant to an evaluation of that organization. The release of timely and accurate information about government operations allows external observers to determine whether government is operating within acceptable parameters. “Transparency enables citizens and other stakeholders to watch government and, if transgressions are identified, challenge it through the media, courts or other institutions” (Welch, 2012, p. 109).

Over time, transparency has become a core value in the public sector: “well-functioning government bodies should be not only effective, democratic, and legitimate but also transparent” (Meijer, 2013, p. 432). Thus, President Obama on his first day in office in 2009 signed the Memorandum on Transparency and Open Government, and 5 years later, Congress passed the Digital Accountability and Transparency Act of 2014. At the subnational level, most state governments have mounted their own transparency efforts such as Colorado’s Transparency Online Project and Virginia’s Commonwealth Data Point.¹ As the Los Angeles example shows, local governments also have taken up the cause. For example, New York City encouraged software developers to create apps to address pressing issues or challenges facing the city by making more than 1,100 datasets available online.²

Still, amid the activity, variation remains in “the quality of transparency initiatives and the degree to which transparency is actually adopted” by governments (Meijer, 2013, p. 429). In 2010, the U.S. Public Interest Research Group (U.S. PIRG; 2010) compared the progress of state governments toward what it called “Transparency 2.0,” which is defined as “a new standard of comprehensive, one-stop, one-click budget accountability and accessibility” (p. 7). Similar to the highly interactive nature of the Web 2.0 movement, this second generation approach to transparency seeks to use technology to provide citizens with simplified, direct way to engage their government. To compare the states, U.S. PIRG evaluated the extent to which states provided online access to government spending data. The organization assessed the searchability of each state’s website, the availability of detailed information on agency contracts and subcontracts, tax subsidies, and economic development grants. In the 2010 report, 18 states received a grade of “F” including 14 states that did not have a website to track government spending. Seven states were considered to be leading states (garnering grades of A or B), the remaining 25 states received grades of C or D. Four years later, using a similar methodology and more rigorous criteria, the number of failing states had dropped to three, and more than half of the states received grades of A or B (U.S. PIRG, 2014). Extending their analysis to the country’s 30 largest cities, U.S. PIRG (2013) found that 17 cities had established online databases containing “checkbook-level detail” of their expenditures; 13 cities had not. Even as transparency has become a core value, transparency practices themselves differ across jurisdictions (Piotrowski, 2011).³

The State of Texas, which earned an A grade in the U.S. PIRG’s Transparency 2.0 study, offers an ideal test bed for research on municipal transparency. The state has received numerous plaudits for its leadership in the use of web technology to increase government transparency (Kirkham & Chapman, 2010). In 2009, the Comptroller’s Office launched TexasTransparency.org, a website designed to promote open, accountable government in the state. Since then, the comptroller has encouraged local governments to increase the online availability of, in particular, their budgetary and financial information.

Before moving on, we acknowledge that single state studies like this one may raise legitimate questions about generalizability. However, an important advantage of a single state study is that the same set of laws and rules govern constituent municipalities. This allows us to develop a better specified model with more contextual details and defensible measures, an advantage of single state studies noted by Nicholson-Crotty and Meier (2002). Moreover, Texas cities offer other positive features such as their number (217 cities with populations of 10,000 or more) and their variation in terms of size, location, resource base, demographics, and politics. Texas cities may not be a perfect microcosm of U.S. cities, but they capture numerous relevant municipal characteristics.

Building the Explanatory Model

To date, no grand theory explains why some city governments have embraced transparency while others have been reluctant to do so. Absent grand theory, our task is to assemble a plausible explanation by drawing upon the empirical work that has explored local government actions and decisions related to e-government. For example, Styles and Tennyson (2007) found that larger cities and more affluent communities were more likely to provide financial information online. Interestingly, the structure of city government, operationalized as the presence of a council-manager form of government, did not produce a statistically significant relationship with transparency. Taking a slightly different tack, Armstrong (2011) explained transparency as a result of a jurisdiction’s commitment to public outreach and accessibility as well as the partisan characteristics of the electorate. A study of New Jersey local governments identified five possible explanations for a city’s commitment to openness: a supportive administrative culture; sufficient governmental resources; ample political competition; an active and professional media, particularly local newspapers; and the presence of good government watchdogs (Piotrowski, 2011).

Other exploratory studies have tested related explanations. For instance, Ahn (2011) found that economics and politics play a role in stimulating adoption of e-communication applications in cities while the impact of elected officials and public administrators is more variable and nuanced. Manoharan (2013), in a study of U.S. county governments, showed that factors such as technical capacity, organizational size, and employee support go a long way in explaining county e-government. Beyond the United States, research on Spanish cities produced several different explanations for various aspects of governmental transparency (Albalate del Sol, 2013). Population size was a factor (larger cities tend to be more transparent) as were mayoral ideology (leftist mayors are linked to more transparent local governments) and the community’s status as a provincial capital (which works to depress transparency). Looking at three aspects of transparency related to air quality in Dutch cities—decision making, policy information, and policy outcome—Grimmelikhuijsen and Welch (2012) found the explanations varied from government capacity to ideology to group influence. It is apparent from these earlier studies on e-government and transparency that there is much more ground to be plowed in an effort to identify and test determinants.

Drawing from and expanding upon this prior work, we contend that cities’ propensity toward transparency is a function of five types of drivers: governmental resources, political competition, administrative professionalism, community demand, and organizational networks. Our goal is to construct a parsimonious explanatory model that utilizes valid operationalizations of these drivers and tests their relative importance in municipal transparency.

Governmental Resources

In general, cities with bountiful resources have more ability to provide assorted services, including making information available to citizens. Some of the earliest research on e-government has demonstrated the importance of city resources (e.g., Styles & Tennyson, 2007). However, the recessionary economy in the United States over the past few years has resulted in an environment where municipalities, already under tremendous fiscal pressure, have been forced to make tough choices about how to allocate money in their budgets. According to a recent International City/County Management Association (ICMA; 2011) survey, 42% of local governments identified a lack of financial resources as a top barrier in providing e-government services. Therefore, despite the growing interest in transparency, it is reasonable to assume that less-resourced cities would attach a lower priority to placing information online. Given real constraints imposed by fiscal stress, we expect cities with sufficient resources to be more resilient and, therefore, have a greater commitment to transparency than less-resourced cities.

Political Competition

Cities vary as to how competitive their electoral politics are. An absence of political competition in a community, where few elective offices are contested and little turnover in office-holding occurs, is likely to produce a relatively complacent, status quo-oriented government. In such an environment, we would not expect there to be much interest in trying new policies or approaches. In a politically competitive setting, Ahn (2011) found that more value is placed on the public’s ability to access government information and contact government officials. Without electoral competition, an officeholder has little incentive, other than an idiosyncratic personal commitment, to engage in innovative behavior by pursuing e-government. Piotrowski (2011) developed a similar argument about the importance of political competition in her research on transparency in New Jersey municipalities. Consequently, we contend that cities experiencing higher levels of political competition are more likely to be more transparent.

Administrative Professionalism

From the early days of public administration, education has been one way in which information about best practices and reforms has been disseminated to public sector workers (Recchiuti, 2007; Teasley, 1993; Zhang & Feiock, 2010). One can argue that managers with more formal education would be more aware of the trend toward greater transparency and the ways in which governments are using their websites to share information with the public. As a result, these credentialed professionals will be more cognizant of, and potentially more committed to, managerial best practices and therefore more likely to create the supportive administrative culture that Piotrowski (2011) suggested is important for transparency. Moreover, city managers, as a group, have been found to be receptive to e-government, saying that it has a positive impact on their ability to manage effectively and increases stakeholder involvement (Reddick & Frank, 2007). Therefore, our expectation is that cities led by credentialed managers will display a greater propensity toward transparency.

A manager’s length of service in the profession is another relevant indicator (Zhang & Feiock, 2010); however, its impact on transparency is not clear cut. However, having served in leadership roles for many years suggests effectiveness in accomplishing tasks. A seasoned professional might be expected to embrace transparency as yet another mechanism for delivering high-quality local government services. By the same token, long service in central management positions may produce some skepticism about innovation or a disinclination to change standard operating procedures that appear to be serving the community well. (In a colloquial sense, this could be thought of as an “if it ain’t broke, don’t fix it” ethos.) For example, Zeemering (2009) found that longtime county administrators were less supportive of exploring new opportunities related to service provision than less-seasoned administrators were. Thus, while we expect years of managerial service to matter with regard to transparency, just how they matter is uncertain.

Community Demand for Transparency

Another factor influencing the amount and type of information provided to citizens involves managerial judgments about the perceived demand for such information. If managers perceive that citizens want, or will use, the information, they are more likely to put the data online. Consequently, if managers believe that citizens do not want the information, that segments of the population lack access to technology, or that the information is too complex to be easily digested, then the city would be less likely to place the data online. A Pew Charitable Trust study examined the behavior of several demographic groups in terms of their use of government websites (Smith, 2010). Although differences in gender and race/ethnicity revealed very little in terms of the way that people use the Internet to receive information about government, differences in education and income were much more revealing. According to the report, those with higher levels of income were more likely to access government websites to get information and to use online services than those with lower incomes. Similar differences were found for users with higher levels of education versus those with less education. Therefore, we expect affluence and education to have a similar positive impact on a city’s placement of information on the web.

Demand may also be stimulated by active local media, particularly media with an investigative focus (Piotrowski, 2011). Local newspapers or broadcast television stations often assign reporters to the city-hall beat to ferret out hints of wrongdoing or scandal. Media stories featuring a city government’s misdeeds (e.g., questionable expenditures) or poor performance (e.g., the failure to monitor a contractor) tend to land on the front page. Local media play a transmission role, acquainting the public with aspects of city government and stimulating demand for more information. Therefore, we anticipate a positive relationship between active local media and municipal transparency.

Organizational Networks

The literature on policy diffusion and networks offers another potential explanation for why some cities would place more information online than other cities do. Frequently, managers look to the practices of nearby or peer cities for cues and signals of best practices (Shipan & Volden, 2008). In other words, cities learn from one another. Managers can also use the social ties found in their networks to transfer or gain knowledge (Reagans & McEvily, 2003). This reduces the uncertainty that often accompanies the implementation of new practices.

A regional council of governments (COG) is an institutionalized diffusion mechanism that lets its members interact and create networks. One of the foundational principles of a COG is the facilitation of information sharing among member governments (Marando, 1971). Cities that are members of COGs in which transparency is valued and practiced by member governments are more likely to follow suit as a result of the exposure. A city that is part of a less-transparent COG has fewer opportunities to learn and, consequently, will be less prone to transparency. Our expectation is that COG membership affects a city’s propensity to embrace transparency.

Measuring Municipal Transparency

To measure municipal transparency, we turn to the GPP, sponsored by the Pew Charitable Trust. As discussed above, transparency is not simply about making random information available, it is about making relevant information accessible, information that can assist the public in monitoring and assessing government operations and performance (Grimmelikhuijsen & Welch, 2012). Expenditure data, an unequivocal reflection of government choices and actions, are often of interest to the public. Taxpayers, especially, are curious about how their tax dollars are being used, but few take the time to attend public budget hearings or wade through the Comprehensive Annual Financial Report to acquire the information.⁴ Beyond finances and budgeting, other aspects of local governments’ operations are worthy of public scrutiny. The GPP has identified a reliable set of approaches and tools that capture a full range of government operations of interest to the public.⁵

In the periodic “Grading the States” reports, the GPP analyzed states’ performance in four fundamental areas of government management: money (finance and budget), people (human resources), information (general administration), and infrastructure (physical assets). Each of these dimensions includes indicators that, in the view of scholars and practitioners, are essential to the effective operation of government; many of them have become best management practices such as long-term planning and the use of performance metrics.

The GPP fine-tuned its methodology over time to produce a validated set of indicators that capture a range of important government functions.⁶ These data are of interest to the attentive public; they are what governments should be disclosing if they want to be transparent about their operations and performance. Armed with this information, external actors can monitor and assess the internal workings and performance of governments, which, as noted earlier, is a central feature of transparency. Consequently, we use the GPP’s dimensions and indicators in our analysis. Doing so minimizes concerns about selection bias.

Data Collection Methods

In this study, the websites of Texas cities with populations of 10,000 or more (n = 217) were evaluated on four significant dimensions of their operations and management: finance and budgeting, human resources, physical assets, and general administration, as derived from the GPP. Researchers searched city websites to find evidence of these managerial best practices—the approaches and tools that are widely considered by scholars and practitioners alike to be part and parcel of a well-managed jurisdiction (Pew Charitable Trust, 2008). Our working assumption was that a transparent local government is one that makes information about these managerial best practices easily accessible on its website.

Three of the categories—finance and budgeting, human resources, and physical assets—are represented by five indicators each, while general administration is comprised of six indicators. Therefore, each category contributes nearly equally to an overall assessment of a city’s commitment to transparency. For the finance and budgeting aspect, websites were searched for information on these indicators: the city’s long-term fiscal outlook, its budget process, its structural balance, rules for contracting and purchasing, and financial controls/reporting. In the human resources category, the informational items sought included strategic workforce planning, hiring practices, retention practices, employee training and development, and managing employee performance. The physical assets category covers data on capital planning, project monitoring, maintenance of physical assets, and internal as well as intergovernmental coordination of infrastructure development. For the general administration category, researchers examined websites to find information about the city’s strategic direction, its use of performance-based budgeting, performance management and program evaluation, and its development of e-governance. Table 1 provides a description of these 21 indicators of the managerial best practices and their operationalizations.

Table 1.

Data Collection Guide.

Dependent variable	Indicator	Key word/search term
Finance and budgeting	Budget process	Budget process, citizen input
	Contracting/purchasing	Contract management
	Financial controls/reporting	Audit, comprehensive annual financial report
	Long-term outlook	Forecasting, multi-year plan
	Structural balance	Balanced budget, rainy day fund, reserve
Human resources	Hiring	Employment, jobs, online application
	Managing employee performance	Employee performance, evaluation
	Retaining employees	Retention plan, succession plan
	Strategic workforce planning	Human resource plan, human capital plan, strategic workforce plan, workforce plan
	Training and development	Employee development, employee enrichment, professional development, training
Physical assets	Capital planning	Capital budget, capital improvement plan
	Intergovernmental coordination	Council of governments, department of transportation/inter-local agreements, development council, metropolitan planning organization, regional planning organization
	Internal coordination	Planning meeting, planning minutes
	Maintenance	Infrastructure improvement, maintenance forecasting, preventative maintenance
	Project monitoring	Capital improvement committee, capital projects, project management/measurement
General administration	Digital government	City services, request for services
	Digital democracy	Agenda, public hearings
	Performance-based budgeting	Performance, performance measure
	Performance management	Managing for results, performance management
	Program evaluation	Performance audit, internal audit, controller
	Strategic direction	Comprehensive plan, general plan, strategic plan

Starting from the home page of a city’s official website, coders⁷ were instructed to find evidence of the 21 best practices using a combination of keyword and mouse click searches.⁸ An indicator was scored as a “2,” if the implementation of the best practice was a citywide policy; it was scored a “1,” if it was implemented in some departments and not others. A score of “0” was given, if there was no evidence via the web that the best practice had been implemented. In some instances, the choices were effectively limited to “0” or “2” because the practice was not found or it had been implemented throughout city government. It is important to note that coders were not expected to make determinations about how well the city was performing the best practices, but instead they were asked to assess whether citizens could find evidence of documents or web portals that would indicate that a particular best practice was being used. To ensure accuracy, each indicator was reviewed by two coders. Intercoder reliability for the complete data set, measured by Cronbach’s alpha, was 86.7, a value for agreement that is within accepted standards (Neuendorf, 2002). In the few cases in which coding discrepancies emerged, they were resolved by bringing in a third coder from a different managerial dimension to conduct an independent review of the website. A city’s scores in each of the managerial areas were summed to generate an overall transparency score.

The Penetration of Transparency

The data collection effort revealed a substantial range in the degree to which Texas cities have embraced transparency via the Internet. The scores range from 0 in a city that had not yet developed a website to a high of 34 recorded by Houston and San Antonio. Eighty-eight cities have scores of 20 or higher; 47 cities had scores of 10 or lower. The mean score for all cities is 16.7, out of a maximum possible score of 42, with a standard deviation of 7.2. In other words, on average, a Texas city has posted approximately 40% of the indicators online.

A city’s population size is an important contextual factor in implementing transparency. Preliminary analysis of the data showed that city population and the transparency variable were correlated at .42; simple bivariate regression revealed that population accounted for 17% of the variance in a city’s transparency score. Acknowledging the impact of population size, we split our sample of 217 cities into two population groupings: the 58 cities with 2010 populations greater than 50,000 and the 159 cities with 2010 populations less than 50,000. By doing so, the impact of the other determinants in the analysis can be gauged more clearly. This is particularly important if any of the variables have differing effects that are masked when all 217 cities are analyzed together. Moreover, it is not uncommon for local governments to be differentiated by population size in analyses (Nelson & Svara, 2010; Wang & Hou, 2012). Table 2 displays the mean transparency scores for the two groups of cities as well as the full sample. The overall transparency score for larger cities is 23.4; for smaller cities, it is 14.3. The differences in transparency scores for the two subsets are statistically significant in every category and further underscore the advisability of proceeding with a split sample.

Table 2.

Transparency Scores, Means (Standard Deviations).

Transparency category	Cities > 50,000 population	Cities < 50,000 population	All cities
Finance and budgeting	7.38 (1.8)	4.65 (2.2)	5.38 (2.4)
Human resources	2.66 (1.3)	2.11 (1.1)	2.25 (1.2)
Physical assets	6.07 (2.4)	2.50 (2.7)	3.46 (3.1)
General administration	7.31 (1.8)	5.01 (2.2)	5.62 (2.3)
Overall transparency score	23.41 (4.9)	14.28 (6.4)	16.72 (7.2)
n	58	159	217

Type of Substantive Information on City Websites

Deconstructing the scores into the four managerial areas shows that the finance/budgeting and general administration categories are the most developed on city websites, while the human resources scores lag. These patterns are evident in both large and small cities. Some variation exists with regard to physical assets transparency which is substantially higher in larger cities than in smaller ones. It is not surprising to discover that transparency with regard to finances is comparatively high given the emphasis it has received at the state level, especially from the Texas Comptroller who bestows awards on cities that have placed budgetary and financial documents online. The similarly high score for transparency in general administration is likely a result of the ease with which items such as a comprehensive plan and meeting agendas can be posted, as well as the increasingly common usage of websites as mechanisms for service requests. Some performance measures can be found, especially in larger cities, but they tend to be less prevalent and more difficult to unearth on the websites.

Looking at the individual indicators, we find evidence of several trends. The overwhelming majority of municipalities offer a way for citizens to provide input on city services. Most cities provide a means for citizens to see how closely the organization’s spending is aligned with its revenues, which is typically reflected in the city’s annual budget. Some cities also include data related to a “rainy day fund” they maintain to cover unanticipated shortfalls. While most cities receive low scores in terms of transparency in human resources, almost all of the cities provide a way for people to apply for jobs online. This development is consistent with the public personnel literature that anticipated this trend on the local level (Llorens & Kellough, 2007; West & Berman, 2001).

Along with these positive trends, there are also areas where cities’ efforts fall short. Of the 217 cities reviewed, only one municipality provides any type of employee retention plan on its website. In a similar fashion, very few cities offer information concerning employee development or training programs. Given the competitive environment for hiring and retaining talented employees, the low scores on both of these indicators appear to be a missed opportunity for local government. Also, the practice of placing performance audits online so that the public can evaluate the success or failure of particular programs has not caught on with Texas cities. This may change in the future as demand intensifies for governments at all levels to provide to evidence of program success or failure.

Explaining the Propensity for Transparency

In the earlier section, Building the Model, we identified the five drivers that, based on earlier research, likely affect municipal transparency: governmental resources, political competition, administrative professionalism, community demand, and organizational networks. In the analysis, our measure of governmental resources is city per capita budget data, which we expect to have a positive relationship with transparency in both groups of cities. Political competition is measured by a ratio that reflects the number of city council seats up for election and the number of candidates for those seats averaged over a 3-year period.⁹ A positive relationship between political competition and transparency is expected. We measure administrative professionalism in two ways. One is the education level of the city manager or chief administrator (Zhang & Feiock, 2010). We hypothesize that a city managed by a professional administrator holding a master’s degree will be more transparent than cities in which the manager or lead administrator does not possess such a post-graduate degree. The other indicator of professionalism is the manager’s total years of service as city manager, deputy city manager, or assistant city manager. Because length of service could plausibly have a positive or negative effect on transparency, we do not offer a directional hypothesis for the relationship.

The model includes three different measures of community demand: median income, low education levels, and media presence. Median household income is used as an indicator of community wealth, and we argue that more affluent cities will be more transparent than less affluent cities. A positive relationship with transparency is expected. Considering the issue of demand from another perspective, we posit that cities with a greater preponderance of less-educated residents will be more likely to have lower transparency scores than cities with a more highly educated populace. As explained previously, local managers in these cities may assume there is little interest among the public regarding government operations; therefore, there is little need to post information online. To measure low education, we use the percentage of residents aged 25 and older with less than a ninth grade education. In this instance, we anticipate a negative relationship with transparency. Cities characterized by active local media, measured here by the number of days per week the local newspaper is published, are expected to be more transparent than cities with less active media.

The final driver of municipal transparency is organizational networks, which we operationalize as a city’s membership in a COG. Cities participating in COGs in which transparency is practiced by member governments should become more transparent themselves; those cities in less-transparent COGs will be less likely to do so. We calculate the mean transparency scores of member governments for each of the state’s 24 COGs and use those figures as the variable. The relationship between COG membership and municipal transparency should produce a positive coefficient.

Table 3 displays the variables, their measurement, and our expectations about their impact on city transparency levels. In general, we expect the direction of the relationships to be the same for large and small cities, although the relative importance of specific variables likely differs across the two groups. Descriptive statistics for the variables used in the models are presented in Appendix Tables A1 and A2. Results of the ordinary least squares (OLS) regression model for the overall transparency scores appear in Table 4. Diagnostic tests show an absence of multicollinearity among the independent variables in the model.¹⁰

Table 3.

Independent Variables: Operationalization, Measurement, and Expectation.

Transparency drivers	Operationalization	Measurement	Expected relationship
Governmental resources	City spending levels	City budget per capita, 3 preceding years, averaged	Positive
Political competition	Amount of competition for City Council positions	Number of candidates/available seats 3 preceding election cycles	Positive
Administrative professionalism	1. City manager’s education	Master’s degree (yes/no)	Positive
	2. City manager’s years in management	Total number of years as city manager, deputy manager, and assistant manager	Positive/negative
Community demand	1. Level of affluence	Median household income, preceding 3 years, averaged and logged	Positive
	2. Low education levels	Percentage of population age 25 or more with ninth grade education or less	Negative
	3. Frequency of newspaper publication	Number of days/week local newspaper is published	Positive
Organizational networks	Overall COG transparency	Transparency scores for cities in each COG, averaged	Positive

Note. COG = council of governments.

Table 4.

Explanatory Model of City Government Transparency.

	Overall transparency score large cities	Overall transparency score small cities
Governmental resources
Budget per capita	−0.003 (0.004)	0.004*** (0.001)
Political competition
Candidates/seat	2.84*** (0.708)	0.226 (0.641)
Administrative professionalism
Master’s degree	1.28 (1.27)	1.43* (0.760)
Years in management	0.012 (0.052)	−0.090* (0.037)
Community demand
Median household income (log)	5.44* (2.45)	3.07** (1.21)
% low education	0.129 (0.148)	−0.084^† (0.063)
Frequency newspaper publication	0.554** (0.212)	0.458** (0.147)
Organizational networks
COG membership	0.770*** (0.148)	0.805*** (0.100)
Constant	−56.12** (27.70)	−35.42** (13.30)
Adjusted R²	.46	.54
N	58	153

Note. OLS regression coefficients, standard errors in parentheses. COG = council of governments; OLS = ordinary least squares.

†

p < .10. *p < .05. **p < .01. ***p < .001, one-tailed tests (except years in management variable).

The eight-variable model performs quite well, explaining 46% of the variance in transparency scores in larger cities and 54% of the variance in smaller cities. In most instances, the drivers perform as expected. Although some of the variables have the same impact regardless of city size, each subset of cities has some degree of distinctiveness. Two of the community demand variables—media presence and median income—have the expected positive relationship with transparency regardless of city size. This supports earlier research by Piotrowski (2011) and Smith (2010), respectively. A city’s COG membership also matters, suggesting that cities are taking cues and learning from other jurisdictions participating in the same COG. A city in a COG in which other members have embraced transparency tends to score higher; a city that is a member of a COG with less emphasis on transparency tends to score lower. These three determinants—median income, media presence, and COG membership—affect both groups of cities similarly.

The two groups of cities differ with regard to the other explanatory variables. For example, governmental resources as measured by budget per capita have a positive effect on small cities’ transparency scores but not on larger cities. This finding is at odds with earlier research reporting a positive relationship between government resources and transparency in large cities (see, for example, Styles & Tennyson, 2007). It may be that, over time, transparency has become routine in bigger cities; however, in smaller cities, governmental resources serve as an impetus for putting information online. The role of professionalism is important in smaller cities but does not appear to matter as much in larger cities. Small cities led by a city manager with a master’s degree tend to have higher transparency scores; however, years of service in management ranks have a dampening effect on website transparency. In larger cities, the role of professionalism, be it a master’s degree or years in management, appears to be of little consequence. Transparency in larger cities is greatly influenced by political competition, which is not important in smaller cities. This finding refines Ahn’s (2011) argument concerning the impact of political competition on the public’s ability to access information.

The same model is used for each of the four managerial categories in an effort to uncover the commonalities and differences among them. Decomposing the overall transparency score reveals important distinctions in the explanations, as shown in Table 5. One quarter to one third (or more) of the variance in transparency in finance and budgeting, general administration, and physical assets is explained by the models; for human resources, the models explain less. Transparency in finance and budgeting in large cities is motivated by political competition, community affluence, and, to a lesser extent, COG membership; in smaller cities, it is driven by COG membership, governmental resources, media presence, and education levels. Transparency in general administration is motivated by a similar set of variables except that professionalism has an impact in small cities while media presence and education levels affect larger cities. Placing information about physical assets on a large city’s website is largely a matter of COG membership, political competition, community wealth, and media presence; in smaller cities where the physical assets transparency scores are substantially lower, only COG membership and community affluence seem to matter. The human resources category is somewhat problematic in that such a small portion of the variance is explained by the model. Clearly other factors are at work driving transparency in human resources, the category in which cities have made less progress using the web. In addition to COG membership, for large cities, political competition, a city manager’s experience, and community education levels have an impact; in smaller cities, it is governmental resources, media presence, a city manager’s experience, and income that make a difference.

Table 5.

Explanatory Models for Transparency Categories.

	Finance and budgeting		Physical assets		Human resources		General administration
	Large cities	Small cities	Large cities	Small cities	Large cities	Small cities	Large cities	Small cities
Government resources
Budget per capita	−0.002 (0.002)	0.002*** (0.000)	−0.001 (0.002)	0.001 (0.001)	−0.000 (0.001)	0.001** (0.000)	−0.000 (0.002)	0.001* (0.000)
Political competition
Candidates/seat	1.16*** (0.300)	0.046 (0.257)	0.678* (0.393)	−0.130 (0.329)	0.402* (0.236)	0.012 (0.154)	0.596* (0.300)	0.297 (0.238)
Administrative professionalism
Master’s degree	0.071 (0.536)	0.255 (0.305)	−0.099 (0.703)	0.217 (0.390)	0.156 (0.421)	−0.033 (0.183)	1.15* (0.535)	0.995*** (0.282)
Years in management	−0.028 (0.022)	−0.018 (0.015)	−0.002 (0.029)	−0.026 (0.019)	0.036* (0.017)	−0.019* (0.009)	0.006 (0.022)	−0.028* (0.014)
Community demand
Median household income (log)	1.77** (1.04)	0.036 (0.484)	1.80^† (1.36)	2.10** (0.619)	0.120 (0.817)	0.384^† (0.299)	1.76* (1.04)	0.547 (0.447)
% low education	−0.021 (0.063)	−0.058* (0.025)	−0.062 (0.082)	−0.007 (0.032)	0.067^† (0.049)	0.013 (0.015)	0.145* (0.063)	−0.032^† (0.023)
Newspaper publication	0.044 (0.090)	0.121* (0.059)	0.158^† (0.118)	0.082 (0.076)	0.080 (0.071)	0.103** (0.036)	0.272** (0.090)	0.153** (0.055)
Organizational networks
COG membership	0.088^† (0.063)	0.200*** (0.040)	0.251** (0.082)	0.271*** (0.051)	0.107* (0.049)	0.082*** (0.024)	0.224*** (0.063)	0.252*** (0.037)
Constant	−14.45 (11.74)	0.143 (5.34)	−18.83 (15.39)	−24.85*** 6.82)	−2.77 (9.23)	−4.12 (3.21)	−20.07^† (11.72)	−6.60 (4.93)
Adjusted R²	.29	.33	.29	.35	.10	.15	.28	.45
n	58	153	58	153	58	153	58	153

Note. OLS regression coefficients, standard errors in parentheses. COG = council of governments; OLS = ordinary least squares.

†

p < .10. *p < .05. **p < .01. ***p < .001, one-tailed tests (except years in management variable).

Looking at Table 5 from the perspective of particular variables, some consistency exists across the different managerial categories. For example, COG membership, a measure of networking and learning, matters in both population groups regardless of management category. Its impact is especially strong with respect to general administration. Two of the demand drivers, media presence and income, influence all of the categories in a positive direction, but the degree of impact varies by city size. For transparency in budgeting and finance in large cities, income is the primary demand factor; in smaller cities, demand is created by media presence and education levels. Another distinctive factor is political competition which is an important determinant of transparency in large cities irrespective of the managerial category. Neither of the professionalism variables appears to affect transparency in budgeting and finance or in physical assets, but managers with master’s degrees are consequential in transparency in general administration.

Implications and Conclusions

Increasingly, local governments are using their websites to share information about their managerial practices. Still, as we have discovered here, city governments have a choice about how transparent they will make their operations. We find that variation continues to exist with some cities opening up their operations for greater public scrutiny while others seem reluctant to do so. Members of the public are more likely to uncover data about budgets and finances when they peruse city government websites and less likely to find information about human resources. As mentioned, human resource professionals have been slow to take advantage of e-government beyond putting up job postings and accepting online applications. The inherent tension between privacy and confidentially, a core value on the area of personnel, and the need for openness, a trademark of transparency, may be at the root of this problem.

Building on this, we believe the article advances the study of transparency in several important ways. First, we rely on a widely accepted and validated framework, the GPP, to generate the type and range of information about governmental operations that cities should be making available to the public. This framework can be used to analyze and compare municipal transparency in other states across the United States and in cities around the world. We believe that the framework is applicable to other types of local government, such as counties and school districts.

Also, the approach used to collect the data represents an advantage over self-reported or survey data. Instead of depending on an official’s perception or recollection of the information available on the city’s website, we conduct an empirical examination of these websites. The scores are based on what an average citizen would find, given standard parameters of a website’s ease of use, if he or she accessed a city’s website to obtain some information.

Our model revealed important similarities, and differences, in what drives transparency in terms of large and small cities. In both cases, community demand and organizational networks helped to explain a city’s transparency score. Regardless of size, cities appear to use members of their network, in this case, members of the COG, to share information about what should be made available to the public. This is true, not only for the overall score, but in terms of each managerial dimension. There is clearly a need to understand how the decision makers are influenced by bonds, or ties made through these networks. The variations between large and small cities are interesting as well: Transparency is spurred by political competition in larger cities; in smaller cities, resources and administrative professionalism play a key role.

Finally, the model itself is quite robust, explaining a large proportion of the variance in the overall transparency scores for large and small cities alike, and yielding strong findings for three of the four managerial dimensions. Clearly the drivers included in the model have explanatory power. However, as we acknowledge, the human resources dimension requires a more specialized model.

One compelling question awaiting an answer in future research is whether the availability of information in transparent cities has produced the sort of monitoring and assessing by the public that some scholars have anticipated. After all, it is through the public’s use of the information that the ultimate value of transparency is realized. Social media may provide a promising way to reach the public. Through the use of popular applications, such as Facebook or Twitter, managers have the opportunity to engage the public around many of the indicators discussed here. For instance, one might imagine a prolonged discussion around the merits or challenges involving the city’s strategic plan or its proposed budget. This moves the conversation from the old paradigm of making information available online to a newer paradigm that involves citizens and businesses engaging that information.

Footnotes

Appendix

Table A2.

Descriptive Statistics: Small City Sample (n = 159).

	M	SD	Minimum	Maximum
Overall	14.27673	6.362381	0	32
Finance/budgeting	4.654088	2.187384	0	10
Human resources	2.106918	1.100007	0	6
Physical assets	2.509434	2.67872	0	10
General administration	5.006289	2.15094	0	10
Independent variables
Budget per capita	742.9619	334.6518	50.81	2,263.14
Political competition	1.779434	0.5508454	1	4.2
Master’s degree	0.6100629	0.4892768	0	1
Years in management	13.49686	9.846367	1	44
Median household income^a	54,499.11	28,233.84	12,263	186,703
Low education levels	9.246289	8.135188	0	36.58
Frequency newspaper publication	2.553459	2.59417	0	7
COG transparency level	16.48962	3.924309	8.875	22

Note. COG = council of governments.

Variable logged in the analysis.

Acknowledgements

The authors thank Judson Brown, David Cabrera, Warren Chalklen, Rachael Dahl, Maria Garnett, Michael Hardy, James Hartshorn, Evelyn Liu, Cody Price, Lindsay Taylor, Jeremy Twitchell, and Catherine Jones for their research assistance on this project. They would also like to thank the three anonymous reviewers for their thoughtful and insightful comments.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Domonic A. Bearfield is an Associate Professor in the Bush School of Government and Public Service at Texas A&M University. His research focuses on issues related to governance in public administration and on improving the understanding of public sector patronage. Email: dbearfield@tamu.edu

Ann O’M. Bowman is Professor and Hazel Davis and Robert Kennedy Endowed Chair in Government and Public Service at the Bush School at Texas A&M University. She specializes in state and local politics and management. She is a fellow of the National Academy of Public Administration and is immediate past president of the Southern Political Science Association. Email: annbowman@tamu.edu

References

Ahn

M. J.

(2011). Adoption of e-communication applications in U.S. municipalities: The role of political environment, bureaucratic structure, and the nature of applications. American Review of Public Administration, 41, 428-452.

Albalate del Sol

(2013). The institutional, economic and social determinants of local government transparency. Journal of Economic Policy Reform, 16, 90-107.

Armstrong

C. L.

(2011). Providing a clearer view: An examination of transparency on local government websites. Government Information Quarterly, 28, 11-16.

Coggburn

J. D.

Schneider

S. K.

(2003). The quality of management and government performance: An empirical analysis of the American states. Public Administration Review, 63, 206-213.

Finnegan

Welsh

(2013, October 24). L.A. controller unveils website to make city finances more transparent. Los Angeles Times. Retrieved from www.latimes.com/local/la-me-1024-city-opens-books-20131024,0,3215790.story

Garcetti

(2013). Executive Directive No. 3: Open data. Retrieved from www.lamayor.org/garcetti_directs_city_departments_to_collect_data_for_open_data_initiative

Grimmelikhuijsen

S. G.

Welch

E. W.

(2012). Developing and testing a theoretical framework for computer-mediated transparency of local governments. Public Administration Review, 72, 562-571.

International City/County Management Association (ICMA). (2011). Electronic government. Retrieved from http://icma.org/en/icma/knowledge_network/documents/kn/Document/302947/EGovernment_2011_Survey_Summary

Kendall

K. E.

Kendall

J. E.

(2011). System analysis and design. New Jersey: Pearson Prentice Hall.

10.

Kirkham

Chapman

(2010). Government transparency: Trends in Texas. Texas Business Review. Retrieved from http://repositories.lib.utexas.edu/bitstream/handle/2152/14680/tbr-2010.04.pdf?sequence=2

11.

Knack

(2002). Social capital and the quality of government: Evidence from the states. American Journal of Political Science, 46, 772-785.

12.

Llorens

J. J.

Kellough

J. E.

(2007). A revolution in public personnel administration: The growth of web-based recruitment and selection processes in the federal service. Public Personnel Management, 36, 207-221.

13.

Manoharan

(2013). A study of the determinants of county e-government in the United States. American Review of Public Administration, 43, 159-178.

14.

Marando

V. L.

(1971). Metropolitan research and councils of governments. American Review of Public Administration, 5, 3-15.

15.

Meijer

(2013). Understanding the complex dynamics of transparency. Public Administration Review, 73, 429-439.

16.

Michalec

(2006). A content analysis of art library web sites. Art Documentation: Journal of the Art Libraries Society of North America, 25(2), 46-54.

17.

Nelson

K. L.

Svara

J. H.

(2010). Adaptation of models versus variations in form: Classifying structures of city government. Urban Affairs Review, 45, 544-562.

18.

Neuendorf

K. A.

(2002). The content analysis guidebook. Thousand Oaks, CA: Sage.

19.

Nicholson-Crotty

Meier

K. J.

(2002). Size doesn’t matter: In defense of single-state studies. State Politics & Policy Quarterly, 2, 411-422.

20.

Pew Charitable Trust. (2008, March). Government performance project “grading the states”. Retrieved from http://www.pewtrusts.org/~/media/legacy/uploadedfiles/pcs_assets/2008/gradingthestates2008pdf.pdf

21.

Piotrowski

S. J.

(2011). The operationalization of municipal transparency: Primary administrative functions and intervening factors. Transparencia y Privacidad. Revista Mexicana de Acceso a la Información y Protección de Datos, 1, 1-31.

22.

Piotrowski

S. J.

Van Ryzin

G. G.

(2007). Citizen attitudes toward transparency in local government. American Review of Public Administration, 37, 306-323.

23.

Reagans

McEnvily

(2003). Network structure and knowledge transfer: The effects of cohesion and range. Administrative Science Quarterly, 48, 240-267.

24.

Recchiuti

J. L.

(2007). Civic engagement: Social science and progressive-era reform in New York City. Philadelphia: University of Pennsylvania Press.

25.

Reddick

C. G.

Frank

H. A.

(2007). The perceived impacts of e-government on U.S. cities: A survey of Florida and Texas city managers. Government Information Quarterly, 24, 576-594.

26.

Rogerson

P. A.

(2001). Statistical methods for geography. London, England: Sage.

27.

Shipan

C. R.

Volden

(2008). The mechanisms of policy diffusion. American Journal of Political Science, 52, 840-857.

28.

Shum

Liu

Wagner

(2013, May 18). Measuring the effectiveness of an organizational website to stakeholders (SAIS 2013 Proceedings, Paper 35). Retrieved from http://aisel.aisnet.org/cgi/viewcontent.cgi?article=1034&context=sais2013

29.

Smith

(2010). Government online: The internet gives citizens new paths to government services and information. Washington, DC: Pew Research Center. Retrieved from http://www.pewinternet.org/files/old-media//Files/Reports/2010/PIP_Government_Online_2010_with_topline.pdf

30.

Styles

A. K.

Tennyson

(2007). The accessibility of financial reporting of U.S. municipalities on the internet. Journal of Public Budgeting, Accounting & Financial Management, 19, 56-92.

31.

Teasley

C. E.

III. (1993). Does professionalism really matter? A look at the changing roles of local government executives. Public Administration Quarterly, 16, 478-491.

32.

U.S. Public Interest Research Group. (2010). Following the money: How the 50 states rate in providing online access to government spending data. Retrieved from www.uspirg.org/sites/pirg/files/reports/Following-the-Money-USPIRG.pdf

33.

U.S. Public Interest Research Group. (2013). Transparency in city spending: Rating the availability of online government data in America’s largest cities. Retrieved from http://www.uspirg.org/sites/pirg/files/reports/%232USP_transparent_ciites_v6_screen_2.pdf

34.

U.S. Public Interest Research Group. (2014). Following the money: How the 50 states rate in providing online access to government spending data. Retrieved from www.uspirg.org/sites/pirg/files/reports/Following%20the%20Money%20vUS%20v2_0.pdf

35.

Wang

Hou

(2012). Do local governments save and spend across budget cycles? Evidence from North Carolina. American Review of Public Administration, 42, 152-169.

36.

Welch

E. W.

(2012). The relationship between transparent and participative government: A study of local governments in the United States. International Review of Administrative Sciences, 78, 93-115.

37.

West

J. P.

Berman

E. M.

(2001). From traditional to virtual hr: Is the transition occurring in local government? Review of Public Personnel Administration, 21, 38-64.

38.

Zeemering

(2009). California county administrators as sellers and brokers of interlocal cooperation. State and Local Government Review, 41, 166-181.

39.

Zhang

Feiock

R. C.

(2010). City managers’ policy leadership in council-manager cities. Journal of Public Administration Research and Theory, 20, 461-476.