Quantitative Analysis Report Mean Differences

Abstract

This empirical study pursued three objectives. First, it sought to determine whether there is a statistically significant difference in environmental spending between coastal and not coastal counties. It also investigated whether there is a statistically significant difference in the intergovernmental revenue growth rate (IGR) based on county type (metro/suburban/rural), and how this changes controlling for political orientation. T-test analysis found that coastal counties spend significantly more on environmental protection efforts than their non-coastal counterparts. Further, from the surveyed counties, one-way analysis of variance (ANOVA) and ANCOVA tests showed that there was no statistically significant difference in intergovernmental revenue growth by county type (metro/suburban/rural) even after adjusting for political orientation. The study used data from the Florida County Government survey, which incorporated 30 counties ranging from metro, to suburban and rural. The findings of the study will go a long way towards informing policy decisions on intergovernmental revenue at both local and state levels.

A study on Environmental Spending and Intergovernmental Revenue Growth by County Type: A Case of Florida

The past few decades have seen a trend towards decentralization of revenue-raising and spending responsibilities from the national government to subnational levels, which are state and local governments (Minassian, 2003). Sources identify several reasons for this trend. First, Minassian (2003) argues that such decentralization enhances more participatory and democratic forms of government, making political leaders at the subnational level more accountable and responsive to the people who elected them. When revenue-raising and spending decisions are made at the subnational level, elected leaders have more room to influence the composition, quality, and quantity of public goods and services in line with the needs of their electorate (Minassian, 2003). Proponents also argue that such decentralization fosters efficiency in the allocation of resources (Minassian, 2003).

At the same time, decentralization has its share of challenges. First, subnational governments have an edge in the provision of public services, while the central government has an edge when it comes to revenue-raising as it has access to more productive revenue sources (Kim & Smoke, 2001). Thus, subnational governments face a mismatch between available local resources and expenditure that the public expects them to undertake (Kim & Smoke, 2001). Further, subnational jurisdictions have different revenue-raising capacities (Kim & Smoke, 2001). Thus, if decentralized units were to operate exclusively on revenues generated locally, less wealthy states and counties would be disadvantaged and may be unable to provide essential services and infrastructure to their electorate (Kim & Smoke, 2001). This raises equity concerns and affects the efficient use and allocation of public resources. Thirdly, due to limitations in revenue-raising sources, subnational units may not adequately meet basic national priority needs such as provision of sanitation facilities, roads, education, and health, which may ultimately increase poverty rates (Kim & Smoke, 2001).

All these challenges point to the need for subnational governments to increase their revenue sources to be more effective in meeting the needs of their electorate. According to the United States Census Bureau, one way by which subnational jurisdictions could increase their revenue is through growing their intergovernmental revenue (US Census Bureau, n.d.).

Using the State of Florida as a case study, this study seeks to determine the factors influencing intergovernmental revenue growth in county governments. The findings will go a long way towards informing policy at the state and national government levels. Policymakers will have clearer insights on where to focus their efforts to ensure that counties move at par in intergovernmental growth and are able to meet basic national priority needs. The national priority need selected for analysis in this study is environmental sustainability.

The study seeks to answer three research questions:

RQ 3: Is there a significant difference in the percent of total spending that is environmental spending between coastal counties and non-coastal counties?

RQ 4: Is there a significant difference in the intergovernmental revenue growth rate (IGR) based on county type (metro/suburban/rural)?

RQ 5: Controlling for political orientation (political), is there a significant difference in the intergovernmental revenue growth rate (IGR) based on county type (metro/suburban/rural)?

Literature Review

This literature review is divided into three sections. The first section covers the definition of intergovernmental revenue and what categories of funds are reported as such revenue. The second section reviews literature on the factors influencing government spending on environmental protection, while the final section reviews literature on the factors influencing revenue mobilization and growth in counties.

What is Intergovernmental Revenue?

The US Census Bureau defines intergovernmental revenue as all funds received from other jurisdictions, including shared taxes, grants, proceeds of reimbursement for services performed for other governments, advances from other governments towards particular functions, advances from other governments received as general financial support, and contingent loans received from other governments to run projects in the receiving jurisdiction (US Census Bureau, n.d.). Funds received from other governments for utility services and property sales are not categorized as intergovernmental revenue as they are reported in a different category of revenue (US Census Bureau, n.d.). Intergovernmental revenue does not also include insurance trust funds or funds paid to the recipient government as the employer share of pension contributions (US Census Bureau, n.d.). Governments classify intergovernmental revenue by the origin of funds - that is local, state, or federal (US Census Bureau, n.d.). Funds received by local governments from the federal government through the state government are recognized as state intergovernmental revenue (US Census Bureau, n.d.).

Factors Influencing Attitudes towards Environmental Protection and Spending

Studies identify different factors that influence a jurisdiction’s attitudes towards the environment and consequently, environmental spending. In one of the earliest studies, Foster and McBeth (1996) measured attitudes towards environmental quality of life and economic development among 187 development officials drawn from a range of urban and rural cities. The study characterized an urban city as one with a population exceeding 50,000 and rural cities as those with populations below 50,000. Respondents answered a series of questions assessing the importance of the environment in relation to economic development. Results of the chi-square test of association showed a statistically significant difference between developers in rural and urban areas, with the former demonstrating a greater appreciation for environmental protection than their urban-based counterparts. The study found the finding interesting given that urban-based developers were younger and more educated, and hence, expected to be more concerned about the environment (Foster & McBeth, 1996).

A primary weakness of the study by Foster and McBeth (1996) is that it covers policy developers, rather than actual rural and urban populations. Salka (2003), partly addresses this weakness in their study that focused on identifying why some counties put more emphasis on environmental protection than others when voting. The study compared respondents’ concern for environmental protection based on whether they lived in an urban or rural county, individual attributes, party affiliation, economic conditions, and age. Regression results showed that all the factors strongly predicted residents’ focus on environmental protection. Respondents were drawn from counties across five states – Oregon, Michigan, Florida, Colorado, and California. Differently from Foster and McBeth (1996), this study found that urban counties were more supportive of environmental issues than their rural counterparts. The researchers attribute this finding to higher levels of education among urban residents. However, the effect of a county’s rurality was minimal compared to other factors such as economic conditions and individual attributes.

To some extent, these findings mirror those of a study by the Duke Nicholas Institute for Environmental Policy Solutions, which sought to understand the attitudes of rural residents towards environmental conservation (Bonnie et al., 2020). Using survey data gathered from 606 urban-based voters, 1,005 rural-based voters, as well as interviews with 36 community leaders from rural counties; the study measured the extent to which rural citizens prioritize environmental protection issues relative to their urban counterparts. Results of ordinary least square regression showed that generally, Democrats paid more attention to environmental conversation than Republicans and independents. However, the study did not find any significant differences in environmental attitudes between urban and rural voters. Thus, the researchers concluded that an individual’s place of residence did not influence their attitudes towards environmental conversation. Like Salka (2003), this study concluded that county of residence was a weaker predictor of environmental attitudes than individual attributes such as political affiliation.

Other studies have gone beyond residents’ characteristics to focus on entire jurisdictional or county economies. Zhang et al. (2019), for instance, conducted a study to determine how foreign direct investment (FDI) influences jurisdictions’ spending on environmental protection. The study compared expenditure on environmental protection (EPEE) between 2007 and 2016 with the FDI trend over the same period for 30 administrative regions in China. Using the spatial correlation test, the study found a positive correlation between FDI and government spending on environmental protection, both in terms of quality and quantity (Zhang et al., 2019). Regions with high FDI equally spend more on environmental protection. The study thus concluded that local governments could enhance the quality and quality (efficiency) of their environmental spending by pursuing FDI (Zhang et al., 2019).

Factors Influencing Revenue Mobilization and Growth in Local Governments

Rurality and Proximity to Urban Areas

Dewees et al. (2003) posits that as intergovernmental relationships change, local governments have to develop innovative strategies to enhance the well-being of their communities. The study sampled 222 local governments on the rural-urban continuum around the Ohio River Valley to identify the extra-economic and local economic activities that the local governments undertake to mobilize revenues locally. The final sample was made up of 148 county officials, representing a 67 percent response rate. 35 percent of respondents identified their counties as rural-adjacent, 32.4 percent as rural non-adjacent, and 33 percent as metro. The study identified that local governments run a variety of local development activities to generate incomes and enhance well-being. These included business incubators, worker training, revolving loan fund, industrial parks, and tax abatement programs, among others. Extra-economic activities across the surveyed counties included industrial foundations, county fairs, annual festivals, community beautification clubs, county strategic plans, and community visioning programs (Dawees et al., 2003).

Chi-square tests of association found significant differences in local development and extra-economic activities between the three groups of counties. Metro counties were more likely to report engaging in the above activities than rural adjacent and rural non-adjacent counties. Logistic regression results showed that urban counties were more likely than non-urban counties to implement economic development programs. Additionally, rural-adjacent counties performed better than rural non-adjacent counties in running economic development programs. The study attributes its observations to lower levels of poverty and high literacy levels in urban counties as compared to rural counties (Dawees et al., 2003).

These findings mirror those of Veneri and Ruiz (2016), who used data from the OECD regional database gathered between 2000 and 2008 to analyze how proximity to urban centers influences economic growth in rural regions. Using a cross section of OECD countries, regression results showed a backwash effect, where rural areas close to urban areas benefit from growth that takes place in the neighboring urban areas. The study found a negative relationship between an area’s proximity to an urban centre and both its population and economic growth rate, with the association weakening with growing distance (Veneri & Ruiz, 2016). Thus, while rurality weakens access to revenue mobilization, this relationship weakens with an area’s proximity to an urban area. Rural-adjacent areas have better opportunities for growth and economic development than rural non-adjacent areas.

Political Alignments

Baskaran and Hessami (2015) conducted a study to test the extent to which political alignment influences intergovernmental transfers. Using the case of a German state, the study hypothesized that state governments’ revenue allocations to county governments are based on two goals: helping municipalities aligned to a particular political side win the next election, and ii) influence unaligned municipalities that may interfere with the policy agenda of the state government (Baskaran & Hessami, 2015). Analytical tests carried out on the data finds that political alignment is a strong predictor of intergovernmental revenues that local counties obtain from state governments. Correlation tests show a significant positive relationship between alignment with local political parties and state intergovernmental support. Thus, “aligned local governments always tend to receive larger transfers from state governments” (Baskaran & Hessami, 2015, p. 1).

In conclusion, a lot of literature exists on intergovernmental revenue, factors influencing government spending on environmental protection, and factors influencing revenue mobilization in local governments. Rurality and proximity to urban areas, as well as political affiliations emerge as significant predictors of revenue mobilization and growth in counties. However, studies focused specifically on counties in Florida are limited, making it a knowledge gap for policymakers in the state. On the question of whether rural counties pay more attention to environmental conservation than their urban counterparts, studies give varied results, opening avenues for studies such as this to explore and offer more insights. The findings of this study will a long way towards informing policy decisions on where to focus more efforts in both environmental conservation and intergovernmental revenue to enhance communities’ welfare.

Methods

The study uses secondary data collected from the Florida County government survey. The sample includes 60 counties, 30 of which are categorized as metro, 12 as suburban, and 18 as rural. The independent variables for the three research questions are county type (coastal/not coastal) for RQ3, and county type (metro/suburban/rural) for RQ4 and RQ5. The dependent variables are percent of total spending that is environmental spending for RQ3, and intergovernmental revenue growth rate (IGR) for RQ4 and RQ5. Political orientation is a covariate explanatory variable to the relationship between intergovernmental growth rate and county type in RQ5. RQ3 is analyzed using independent t-tests samples t-test, while RQ4 and RQ5 are analyzed using one-way analysis of variance (ANOVA) and ANCOVA tests respectively.

Results

RQ 3. 1. Is there a significant difference in the percent of total spending that is environmental spending between coastal counties and non-coastal counties?

Analysis

Figure 1: Error Bar Plot for Average Environmental spending in coastal and not coastal counties

The error plot provides a view of what to expect in the t-test. The different lengths of the plots indicates that the groups have different variances. If the groups had equal variances, the plots would be of the same length. Thus, one would expect the t-test to yield a significant result. From the plots, it is evident that there is more variability in environmental spending in coastal areas than in not coastal areas.

Table 1: Group Statistics

Coastal area or Not

N.

Mean

Std. Deviation

Std. Error Mean

Average % Envir.

Coastal area

34

.14184

.063135

.0.010828

Spending in Total Spending

Not coastal area

34

.09404

.037583

.006644

The sample includes 34 zones categorized as coastal areas and 34 categorized as not coastal areas. The mean environmental spending in coastal areas is 14 percent of total spending, as compared to a mean of 9.4 percent in not coastal areas. To check whether the difference in means is significant, one should look at the independent samples test presented in table 2 below.

Table 2. Independent Samples Test

Independent Samples Test

Leven’s Test for Equality of Variances

t-test for Equality of Means

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

Lower

Upper

Average % Envir.

Equal variance assumed

14.242

.000

3.708

64

.000

.047804

.12891

.022051

.0073557

Spending in Total Spending

Equal variance not assumed

3.763

54.32

.000

.047804

.012703

.022339

.73270

Since the p-value of the Levene’s test for Equality of variances is p

RQ 4: Is there a significant difference in the intergovernmental revenue growth rate (IGR) based on county type (metro/suburban/rural)?

Figure 2: Error Bar Plot for Intergovernmental Revenue Growth (IRG) and County Type

The different lengths of the plots indicates that the groups have different variations around the mean. If the groups had equal means, the plots would be of the same length. From the plots, it is evident that there is more variability in environmental spending in rural and suburban counties, as compared to metro counties. However, the differences in means are small between suburban and rural counties, one would expect the one-way ANOVA test to yield an insignificant result. However, the test may produce a statistically significant difference in means when comparing metro and suburban counties. .

Table 3: ANOVA

Sum of Squares

df

Mean Square

F

Sig,

Between Groups

.304

2

.152

2.749

.072

Within Groups

3.149

57

.055

Total

3.453

59

Table 4: Tests of Between-Subjects Effects for Intergovernmental Revenue Growth (IRG) and County Type

IGR Growth Rate

95% Confidence Interval for Mean

N

Mean

Std. Deviation

Std. Error

Lower Bound

Upper Bound

Minimum

Maximum

Metro County

30

-.07685 4

.1591382

. 0290545

-.136277

-.017431

-.4791

.3888

Suburban County

12

.033722

.2423555

.0699620

-.120263

.187708

-.2804

.5725

Rural County

18

.080411

.3225618

. 0760285

-.079995

.240817

-.4370

.8397

Total

60

-.007559

.2419240

-.0312323

-.070055

.054936

-.4791

.8397

The ANOVA analysis in table 3 shows that there is no statistically significant difference between the group means. The significance value p =.072 is greater than p = 0.05, showing that there is no statistically significant difference in the mean IGR between counties. At this point, the study accepts the null hypothesis and concludes that there is no statistically significant difference in IGR based on county type (metro/suburban/rural). Although this is helpful information, it does not say where the biggest differences between groups lie or which groups account for the greatest share of the insignificant results. To obtain this information, one would need to run a post-hoc test, such as the Tukey post hoc test.

Table 5: Post hoc tests: Multiple Comparisons

Dependent Variable: IGR Growth Rate

95% Confidence Interval

(I) Metro, Suburban or Rural County

(J) Metro, Suburban or Rural County

Mean Difference (I-J)

Std. Error

Sig.

Lower Bound

Upper Bound

Turkey HSD

Metro County

Suburban County

Rural County

.1105764

.1572651

.0802867

.0700799

.359

.072

-.303780

-.325907

.082627

.011377

Suburban County

Metro County

Rural County

.1105764

.0466887

.0802867

.0875999

.359

.855

-.082627

-.257491

.303780

.16113

Rural County

Metro County

Suburban County

.1572651

.0466887

.0700799

.0875999

.071

.0855

-.011377

-.164113

.325907

.257491

The study compares the group p-values with the significance level of p=.05. None of the p-values is less than p=.05, implying that the mean differences are not statistically significant. These results confirm that the intergovernmental revenue growth rate (IGR) does not differ based on county type (metro, suburban, or rural).

RQ 5. Controlling for political orientation (political), is there a significant difference in the intergovernmental revenue growth rate (IGR) based on county type (metro/suburban/rural)?

Figure 3: Error Bar Plot for Intergovernmental Revenue Growth Rate (IRG) and County Type, Controlling for Political Orientation

From the plots, it is evident that there is more variability in IGR growth rate between liberals in suburban counties than in other counties. The different lengths of the plots indicates that the groups have different variations around the mean. If the groups had equal adjusted means, the plots would be of the same length. However, the differences in means across all groups are small, one would expect the one-way ANCOVA test to yield a statistically insignificant result.