Federal Economic Development Transfers to State and Local Governments

Abstract

This study examines the contribution of the federal government to community and regional economic development by exclusively analyzing data on intergovernmental transfers to state and local governments. The study covers the years 1940 through 2010. The intent is to explore and define the patterns and trends within the community and regional development data and establish the extent to which incrementalism served as a basis for setting aside resources for accomplishing economic development objectives. Our findings indicate that the data patterns, trends, and relationships are nonlinear and unsupportive of the “incrementalist” perspective to resource allocation commonly found in major funds within the public sector.

Keywords

regionalism intergovernmental transfers local economic development local economic development strategies federal grants

The recent domestic and global market disturbances, recession, and economic shortfalls continue to negatively affect households, businesses, and public entities. The “Great Recession” serves to remind us of how government efforts to restart economic activity to support national, local, and regional growth constitute a continuing policy problem. Different jurisdictions have different problems and needs during good or bad economic times, but governments do share in the efforts to provide economic development for their communities. States and municipal governments have always been constructively engaged in promoting economic growth and development in their jurisdictions. To attract and retain local industries and prospective employers, these governments adopt strategies for creating conducive environments for retaining short- and long-term capital and for developing capacities aimed at revamping nonperforming sectors of their local and regional economies.

The strategies vary and are not mutually exclusive in application. A government might adopt a targeted strategy that focuses on enhancing growth through business attraction, retention, and promotion (Blakely & Bradshaw, 2002; Buss, 2001; Center for Urban Economic Development, 1987; Greenwood & Holt, 2010; Koven & Lyons, 2006; Markusen, 2007; Zheng & Warner, 2010). There are advantages to targeted programming in prospecting for businesses, but there are also associated bottlenecks (Bartik, 2003, 2005; Lynch, 2004; Markusen & Nesse, 2007; Zheng & Warner, 2010). Sometimes the strategy provides incentives (Buss, 2001; Koven & Lyons, 2006) and a supportive environment for sustaining economic activities. In other instances, governments might favor strategies that promote public-private partnerships and the development of entrepreneurships (Porter, 2000; Sabel, 1992). Other strategies promote direct and indirect investments in infrastructure that are aimed at addressing specific socioeconomic problems while enhancing the quality of life of citizens (Florida, 2002; Greenwood & Holt, 2010; Portney, 2003; Shuman, 1998; Warner, 2001). A strategy might address poverty reduction in deprived economic areas through job creation programs aimed at reducing high rates of unemployment (Bennett & Giloth, 2008; Greenwood & Holt, 2010).

State and local governments are not alone in the economic development policy arena; federal efforts for managing the economy and attracting economic development have existed since the New Deal era. There are private actors involved in economic development as well. For instance, a quarter century ago, Levy (1990) counted between 15,000 and 18,000 organizations promoting economic development in the United States.

Being able to track economic development efforts and funding over time can present a challenge. A case in point that is of particular importance to this study is the participation of the federal government in the economic development policy arena—in particular the extent to which federal policy has contributed over the years toward the promotion of growth and development in regions and local communities through its transfer programs. Though often considered the domain and responsibility of state and local governments, revamping regional and local economies for enhanced economic performance is equally beneficial to the federal government. For instance, flourishing industries in a region generate employment opportunities, income, and taxes that benefit all three levels of government. Furthermore, in a well-performing economy, unemployment, poverty, and crime rates are kept to a minimum. Subsequently, significant reductions in rates of unemployment, poverty, and crime alleviate pressures on state and local government budgets for social and public safety services. Through exerting joint efforts, all three levels of government engage in policies to contribute to economic growth while enhancing societal well-being and quality of life.

There are numerous federal economic development policies. One category of federal activity has been transfers of funds to state and local governments. This interaction between federal, state, and local governments in addressing public policy and problems is called “intergovernmental management.” The transfer of federal funds to other governments is called “fiscal federalism,” in which federal policy goals are associated with budget allocations (Nivola, 2002; O’Toole & Christensen, 2012; Walker, 2000). There are different amounts of centralized control of federal transfers, both from the federal funding authority and from state control of distribution of funds. During the Nixon administration, devolution of spending authority to local governments occurred as part of a general revenue-sharing initiative (Walker, 2000).

Federal prescriptions and mandates accompanied federal policies and federal funds, and often mandates are unfunded (Conlan & Posner, 2008; Posner, 1998). Tracking funds, or what the federal government budgets for, is a rough way to determine priorities and preferences. One major category of federal transfers to state and local governments is for economic development, and this has been tracked since the 1940s by the U.S. Census Bureau.

This study evaluates how these transfers were incremental over the years. Ever since Lindblom (1959) proposed “limited successive comparisons” as a more realistic model of decision-making than the rational comprehensive approach 55 years ago, incrementalism has been a major theory of considering policy making over a period of time. Incrementalism, as applied to budget expenditure decisions, asserts that the prior year’s expenditures are used as a starting point, or base, and adjustments are made for the present budget (Wildavsky, 1964). Thus, small or marginal changes in budgets based on inflation are considered normal or stable budgeting. Incrementalism is seen as a reflection of the stability and conservative nature of budgeted amounts.

There is some dispute as to what an increment is and what a normal budget is. Some suggest that moderate changes in budgeted amounts are seen as evidence of change in the environment (Hayes, 1992). Others see incrementalism, with the small or marginal changes over time, as a normal or steady state, and they consider large changes are evidence of innovation (F. D. Berry & Berry, 2007). An alternative approach rejects simple incremental or nonincremental budget categories. It suggests that within budget decisions over time, the pattern of expenditures can show both incremental and nonincremental decision making, and that punctuations—which are large changes in budgeted amounts—arise from crises or dramatic events (Robinson, Caver, Meier, & O’Toole, 2007; True, Jones, & Baumgartner, 2007). Frequently, critiques of incrementalism begin with lack of consensus on the size of an increment. In addition, incrementalism is seen as descriptive but not useful for forecasting (W. D. Berry, 1990; Hayes, 1992).

A major attribute of incrementalism is that policies are relatively stable and that funding for these policies is affected by prior decisions. The theory of incrementalism is used to explain expenditure decisions made over a period of years. In fiscal federalism the federal government is expressing and imposing its policy preferences (Volden, 2007) and the effect of federal grant transfers have been studied for one city (Hendrick, 1998). The use of incremental models to evaluate economic development grants to state and local governments considers whether there is a stable and incremental federal policy in funding over a period of time.

This study assesses the extent of the federal government’s participation in local and regional economic development through its grants-in-aid programs during the years 1940 to 2010. This goal is accomplished by analyzing the growth patterns and trends within the aggregate data on intergovernmental transfers for the years covered by the study, and by determining whether the transfers to state and local governments reflected a long-term “incrementalist” approach to resource allocation, which is typical of marginal and upward changes in budgeting over time.

This study argues that normal historical progressions of federal grants-in-aid to state and local governments for economic development will reflect incremental patterns in the fiscal data. There are mutually-shared national benefits associated with well-functioning local and regional economies, the study argues, and the pattern of federal funding will demonstrate a stable path over time. Furthermore, it is assumed that the data will exhibit the continuing interest of the national government in sustaining regional and local economic development efforts. This latter argument forms the guiding theme in our study and provides the basis for concentrating solely on intergovernmental transfer programs to state and local governments for promoting economic and regional development.

It is well known that the contributions of the national government to regional and local economic development efforts are not limited to intergovernmental transfers. It is not the goal of this study to survey and assess the array of programs utilized by the national government to promote economic growth and development in regions and local communities. The decision to concentrate on economic development grants-in-aid funding for this study is intentional, and it is aimed at maintaining a sharper focus.

The Government, Economic Development, and Regionalism

The governmental role in encouraging economic development has an impact on human progress and well-being. The important governmental role in regional and local economies is long-standing. Similarly, the role of individual states in promoting economic development is not recent. For instance, Bingham and Mier (1993) assert that the development of industrial development bonds in Mississippi in 1937 marks the beginning point for economic development as a public activity. The literature has benefited enormously from the concerted efforts of contributors to the field, including Bendavid-Val (1991); Bingham and Mier (1993); Downs (1979); Fainstein, Fainstein, Hill, Judd, and Smith (1983); Greenwood and Holt (2010); Koven and Lyons (2006); Levy (1990); and Lynch (2004), to mention a few. The work of these authors contributes toward a better understanding of growth and development, and of how growth in particular is measured. On the topic of economic growth accounting, the pioneering work of Abramovitz (1956), Solow (1957), and Denison (1962) created a series of models and frameworks for tracking the performance of economies. Similarly, much has been accomplished in the area of growth estimation (Solow, 1957), and on how to establish the relative importance of the respective factor inputs needed for production (Denison, 1962; Griliches & Jorgenson, 1966, 1967).

Because of the vastness of the literature on economic and regional development in particular, the current discussion focuses on two objectives: (a) emphasizing the importance of the national government in promoting economic development and (b) considering incremental theory.

Conceptually, and for the purposes of this study, a regional development perspective allows us to pursue these objectives; it also provides the building blocks for arguing in support of the national government’s funding of economic activities in regions and local communities. However, the numerous themes on regionalism results in a multiplicity of tenets, properties, constructs, and interpretations that are associated with local economic activities and states’ roles in development. As a result, several versions of regionalism are identifiable, each of which focuses on the various dimensions making up the major themes in regional development. These dimensions are sometimes dominated by the north-south argument discussing the differences in regions or geography in development. This theme is often extended to incorporate elaborations on urban core periphery or on old industrial areas and fast-growing areas and the relationships to economic cooperation and regional development. The focus of this study is not on the north-south and core-periphery arguments in regional development. Instead, the focus is on the relevance of induced change by a national government in the structural composition of regional and local economies, with the intent of accomplishing specific economic and societal objectives within a predefined time frame.

The major thrust in this framework, as well as the thrust in the regionalist argument involving growth and development in regions and local communities, is provided succinctly by Nelson, who argues that regionalism is about “change in regional productivity as measured by population, employment, income, and manufacture value added. It also means social development such as the quality of public health and welfare, environmental quality, and creativity” (1993, p. 27). Both national and subnational governments have mutually-beneficial interests in ensuring the accomplishment of desired changes and increases in productivity measures in local and regional economies.

Specifically, the mutually-shared benefits are illustrated in the following manner. Growth in local and regional economies provides employment opportunities that generate income for residents and taxes on income and sales for subnational governments. The federal government benefits as well through increased tax revenues on income. The ability to sustain these resources depends on stable and increasing population over time. Persistent increases in value-added in production and manufacturing will in the long run enhance overall output levels, generate additional wealth, and benefit the region as a whole. Furthermore, growth in local employment, income, and production value-added reduces social problems and strains on budgets of subnational governments. For instance, poverty and crime rates may fall as a result of increased job opportunities for residents. These reductions entail significant cost savings that may enhance efficiency. Both national and subnational governments therefore have a stake in ensuring and promoting well-performing economies, as argued by Bingham and Mier, who assert, “It is the role of the public sector to facilitate and promote the creation of jobs and wealth” (1993, p. vii).

Finally, regionalism often creates a “domino effect” in trade in particular (Baldwin, Forslid, Martin, Ottaviano, & Robert-Nicoud, 2003), in that areas not initially targeted for change eventually gain from the spillover effects created by growth from well-targeted regional areas or sectors of the economy. Therefore, the federal government’s role in economic development activities is justifiable from a regionalist perspective. Furthermore, its transfers to state and local governments are indicators of its efforts to induce change without resorting directly to the varied microstrategies for attracting and retaining businesses, and for promoting economic growth and development in general.

Research Questions and Methods

The efforts of the national government to induce change in regional and local economies are examined by seeking answers to the following questions:

How important were the federal government’s transfers to states and municipalities for local and regional economic development for the years covered in the study?

How did these transfers compare with other federal grant-in-aid programs to states and municipalities?

What was the long-term growth or decline of the transfers to states and municipalities during the years covered by the study?

Was the overall growth of these transfers reflective of incrementalism as a basis for resource allocation?

This study suggests that there are regular statistical patterns to federal transfers of funds to state and local governments for the purpose of local and regional economic development. To determine whether these transfers are incremental, the study begins by considering the following assumptions and criteria for evaluation. Specifically, budget allocations including transfers of funds are considered incremental when the following occur:

Funds are allotted by category or expenditure type over an extended period of time.

There is a trend toward those budget allocations that consists of gradual or marginal changes over an extended period of time.

The relationship between the budgeted amounts over time exhibits an upward trend that is positive.

The upward trend resulting from the budgeted amounts is linear.

This positive relationship is sustained over periods covering several observations.

The data for transfers of grants-in-aid from the federal government to states and municipalities are regularly reported by the U.S. Census Bureau. Beginning in 1940 and continuing through 2010, the data in this study cover 71 years. The federal transfers to state and local governments include five major categories, including community and regional development, income security, health, transportation, and education and training. Two other variables used are population and urban price indexes. The sources for these data are the Census Bureau’s yearly population data for the years 1940 through 2010 and Bureau of Labor Statistics publications on all urban indexes. Additional information regarding definitions and technical notes for these variables can be found in Appendix A.

The data treatment and analytical orientation are as follows. As a first step, the raw data were adjusted for population and prices to obtain real per-capita values. The base period adopted for calculating these values was 1982–1984. This period was chosen as an approximate midpoint in the data series. Furthermore, the initial transfers were extremely low and insignificant. Finally, the adjusted data were subsequently divided into seven periods, referred to in this study as “decennial partitions.” These separations are introduced to facilitate easy referencing and provide an avenue for decennial comparisons. In effect, the observations (n = 71) are divided into seven periods, six of which consist of 10 years each, starting from 1940. The seventh period is made up of 11 years instead of 10—2000 through 2010, inclusive. The respective partitions covering the years studied are presented in Table 1. These partitions are used in the discussion phase of the article.

Table 1.

The Seven Decennial Partitions.

Decennial partitions	Partition identification numbers
1940–1949	1
1950–1959	2
1960–1969	3
1970–1979	4
1980–1989	5
1990–1999	6
2000–2010	7

Source. Authors’ calculations.

Note. The decennial partitions are evenly distributed with 10 years per partition, except partition number 7, which has 11 years. The extra year will not significantly influence the final outcome of the analysis.

Techniques for Analysis

In incrementalism, considering changes in funded amounts over time involves the assessment of slopes and bumps in the data. The population and inflation-adjusted data are first summarized using proportions, graphs, and charts for numerical and graphical descriptive purposes. A dominant interest in the study is an assessment of the slopes and bumps embedded in the functional data over time. Ramsay (2002) defines slopes and bumps as “rates of change [that] . . . require at least 2 observations to define. . . . Bumps [are] features having the (a) amplitude of a maximum or minimum, (b) width of the bump, and (c) its location.” A typical bump must integrate all three features, and in studying the functional or longitudinal data, the slopes are as important as the bumps.

Three propositions are tested in this effort. First, given the effects of time on the adjusted data, it is presumed that a first-level model with a linear trajectory might be appropriate. This is derived from this study’s initial contention that the growth emerging would reflect an incrementalist approach to federal funds transferred to state and local governments for community and regional development. An expected incremental rate of change would be indicated in positive upward changes within the observed data. Though the intent to introduce all three propositions in the estimation process appears a bit exploratory, the decision is geared toward a step-by-step assessment of the patterns of the historical data.

The first proposed model is a straightforward first-order equation of the form,

δ_{j} = τ_{o} + τ_{1} {(θ)}_{j} + µ_{j},

where τ_o is the expected effect on δ_j when (θ)_j = 0;

δ is a dependent variable consisting of the transfers;

τ₁ is between-periods expected effects;

(θ)_j is time, with equal distances between observations; and

µ_j is the residual effect for the unexplained, where µ ~ N(0, σ).

The second proposition is based on the presumption that a more curvilinear model is appropriate for the functional data. In certain instances, polynomials or second-order models provide sufficient and relatively superior results when compared with their first-order counterparts. The general form of a dependent polynomial with one independent predictor variable is defined as

δ j 1 = τ_{o 1} + τ_{11} {(θ)}_{j} + τ_{12} {(θ)}^{*}_{j} + \dots . + µ_{j} .

The coefficients are as defined previously, with τ₁₂ and (θ)^*_j representing the added dimensions for the parabola. τ and (*) are nonnegative constants.

According to Kleinbaum, Kupper, Muller, and Nizam, adding high-order terms like (θ)^* and beyond to Equation (1), “which are simple functions of a single basic variable, can be considered equivalent to adding new independent variables” (1998, p. 281).

In the third proposition, the inverse of the second-order model presented in Equation (2) involving the introduction of the natural log is applied to the treated data. This procedure is “useful for removing varying amplitude over time” (SPSS Base 9.0, 1999, p. 523). This model is of the following form:

ln δ_{j 2} = τ_{o 2} + τ_{1} ln {(θ)}_{j 2},

where the dependent variable is the natural log of community and regional development grants represented by the symbol δ (delta);

The numerals are subscripts incorporated for model identification; (θ) still represents the previously-defined independent variable; and τ₀ and τ₁ are the intercept and coefficient, respectively.

Finally, the decennial partitions are used in the discussion section for data periodicity for the years covered. The periods are evenly distributed, 10 per partition, except Period 7.

Results

Relative Importance of Transfers

The results here indicate that the five major categories of the federal government’s grant-in-aid programs constitute approximately 94% of all transfers to state and local governments. These programs include the following five categories: a) health, b)income security, c) education, d) transportation and e) community development. Table 2 shows data on these five categories for selected years.

Table 2.

Historical Representation of Data for Selected Years.

Selected years	Federal outlays for all programs (in $ millions)	Five grant-in-aid programs	Amounts (in $ millions)	Five programs as percentage of total outlays
2010	461,317	Health	218,025	96.13%
		Income security	96,102
		Education	58,904
		Transportation	51,216
		Community dev.	19,221
		Total	443,468
1990	135,325	Health	48,890	93.54%
		Income security	36,768
		Education	21,780
		Transportation	19,174
		Community dev.	4,965
		Total	126,577
1970	24,065	Health	3,849	93.25%
		Income security	5,795
		Education	6,417
		Transportation	4,599
		Community dev.	1,780
		Total	22,400
1950	2,253	Health	70	75.77%
		Income security	1,096
		Education	99
		Transportation	434
		Community dev.	8
		Total	1,707

Source. U.S. Census Bureau and Office of Management and Budget (OMB).

Note. The total excludes figures for minor grants. The differences between outlays and totals for grants are due to the omitted grants. The figures for grants as a percentage of the total for the five categories in 2010 are obtained from the following formula: 19,221 ÷ 443,468 = 0.0433 × 100, which equals 4.33%.

Although four of the categories listed are not of interest to the current study, they are included for comparative purposes and for establishing the relative importance of the variable under study. Specifically, in 1950, the five categories accounted for 76% of the federal government’s total outlays for grants. In that year, transfers for community and regional development represented 0.47% of the total involving the five categories; a very small amount. In 1970, the transfers were 93% of the federal government’s total outlays for grants, with community and regional development accounting for 7.95% of the aggregates involving the five major transfer programs. Though these five categories still accounted for 96% of the total outlays in 2010, community and regional development’s share of the total involving the five categories declined to 4.33%.

Finally, the price- and population-adjusted data for the years 1940 through 2010 are used in Table 3 to illustrate the relative percentage composition of the grant-in-aid programs. As indicated earlier, the base year for the price adjustments is 1982–1984.

Table 3.

Real Per Capita Grants in 1982–1984 Dollars.

	Community & regional development	Transportation	Education & training	Health	Income security
Distribution	5.0%	15.71%	18.07%	33.71%	27.5%
Real per capita dollars 1982–1984	1,085	3,410	3,924	7,319	5,971

Source. Authors’ calculations. Data source: U.S. Census Bureau and Office of Management and Budget (OMB).

Note. Percentages in first row do not sum to 100 because of rounding.

As shown in Table 3, transfers for health and income security are the largest, accounting for approximately 60% of the total. Education and training constitute 18%, whereas community and regional development account for 5%, after adjustments for prices and population. Hence, transfers for community and regional development represent the smallest of the five programs.

Data Patterns, Slopes, Bumps, and Autocorrelations

The real per capita funds are first presented in a scatter diagram, Figure 1, for the period under review, from 1940 through 2010. The relationships sketched by the scatter plot between the time period and the adjusted per capita data do not provide stable patterns for straightforward interpretations. In the diagram’s current form, the patterns and relationships are difficult to decipher.

Figure 1.

Scatter diagram for community and regional development G-I-A, 1940–2010, in real 1982–1984 prices.

To improve on this initial graphical result and better project the expected slopes and bumps, a sequential plotting procedure is applied first to the unadjusted data and subsequently to the adjusted data. The sequence plot of the unadjusted data is presented in Figure 2. This is followed in Figure 3 by the sequence plot for the adjusted data.

Figure 2.

Sequence plot of the historical unadjusted community and regional development data, 1940–2010.

Figure 3.

Sequence plot of the historical adjusted or per capita community and regional development data, 1940–2010.

As compared to the graphing of the gross transfers of funds in Figure 2, Figure 3 differs significantly in that it emphasizes the effects of both size and prices over the years, and it amplifies the slopes and bumps defined earlier. At X < 0, which represents the period prior to the adjusted data reaching a peak (where X = 0), transfers were on the aggregate positive. In effect, the slope was increasing between points L and M and was at its greatest just before reaching the peak. At point L the slope was insignificant, though not negative. At X = 0, the slope disappears, and this was followed by a precipitous decline. Therefore, at N and beyond, the slope was negative until it stabilized at point 0. Importantly, there was discontinuity in the slope at point X = 0.

A difficulty often associated with time series analysis is that the data might be serially correlated and lead to misinterpretations. These problems are verified here through checks for autocorrelations. In the case of the adjusted data represented in Figure 3, with the exception of the initial phase in the series, the rest of the data fell neatly into the 95% confidence interval, thus indicating no autocorrelation. The verifications are presented in Appendix B.

Statistical Assessment of Overall Data for 1940–2010

The results of Equations (1) and (2) for 1940 through 2010 are summarized in Table 4. These results are presented to define the contribution of (θ)_j when used in the model alone, and to isolate the effects of the second-order variable (θ)^* through a lack-of-fit test. Based on the sum of squares contributed by (θ) alone, the straight-line presumption holds, as (θ) is found to be statistically significant at the 1% level. Its t-value is equally significant with an associated R-square of .474. With θ already in the equation, θ^* does not contribute much to the sum of squares. Still representing the second-order variable in the model, its effects are not statistically significant. In effect, the addition of (θ)* to Equation (2) does not provide any improvement to the model. The proportion of the variation defined by the R-square due to the introduction of (θ)* changed from .474 to .484, yielding a net difference of only .010. Furthermore, the estimated coefficients for θ^*did not change as a result of its introduction. Its t-value, −1.543, is equally insignificant at the 5% level. Therefore, the polynomial or quadratic function stipulated in Equation (2) did not provide additional evidence to adequately characterize the estimated slope.

Table 4.

Variations Explained by Equations (1) and (2), 1940–2010.

Source	Sum of squares	df	Mean square	F-statistic	p-value
(θ)	4,668.946	1	4,668.95	63.12	<.000
(θ)\|(θ)*	172.628	1	172.63	2.38	>.05
Residual	4,857.647	67
Total	9,699.221	(n – 1)
Adjusted R-square:
(θ)	=.474
(θ)\|(θ)*	=.484

Source. Authors’ calculations.

Note. The dependent variable is the population and price-adjusted community and regional development data for the period 1940 through 2010. The results are for the entire period.

The results of Equation (3) are summarized below. The estimated model and its constant and related statistics are presented first. This is followed by the presentation of the estimated model without a constant. The results are as follows:

\begin{array}{l} 1 . ln δ_{j 3} = 1.286 + 0.052 ln {(θ)}_{j 3} \\ (. 350) (. 007) \end{array}

Adjusted R - square = 0.475

t - value (θ) = 7.968

(p < 0.000)

The t-value for (θ) is statistically significant (p < .000), with an adjusted R-square of .475, which does not differ from the result obtained under the first-order specification involving Equation (1). The variation explained is 48%.

\begin{array}{l} 2 . ln δ_{j 3} = 0.057 ln {(θ)}_{j 3} \\ (0.003) \end{array}

Adjusted R - square = 0.821

t - value (θ) = 17.921

(p < 0.000)

The above results indicate a significant improvement over the statistics obtained in the use of the second-order model under the specifications of Equation (2). A more reliable and relevant statistical result is obtained after the omission of the constant in the model, whereby τ_o2 = 0. This exclusion mimics very well the characteristics of the data series used in the study, in that the initial stages reflect the relatively insignificant nature of the transfers: in fact, they are very close to zero after adjustments for prices and population.

Figure 2 depicts these patterns in the data series quite well. Furthermore, as a result of the exclusion of the intercept, the adjusted R-square increased from .475 to .821, almost doubling in size, with p < .000. These results also indicate a significant improvement over the findings obtained under the second-order model using the specifications of Equation (2). Furthermore, the results permit the use and application of percentages in interpretation in view of the embedded natural log element. Specifically, a unit change in (θ)_j3 is associated with a 0.052 or 5.2% change in the dependent variable. (θ)_j3 is time measured in years, with equal distances between observations. In effect, the real per capita transfers (δ_j3) changed at an annual rate after adjustments for prices and population of approximately 5.2%. The estimated change increases to 0.057 after adjusting for a constant or y-intercept, which equals zero.

These findings are further corroborated by the results obtained from the application of an average growth rate estimation formula, expressed as

r = {[Y \div X]}^{1 / n} - 1,

where: Y = the ending value in the data series;

X = the beginning value in the data series;

n = the number of years covered in the study;

1 = a numeral or scalar; and

r = the average growth rate.

Solving for r, we obtain

{[60.949869 \div 2.096547]}^{0.0143} - 1

= 0.0494 or, 4.94 %, % \approx 0.05 .

Next, Figure 4 shows the scatter plot of the adjusted data with an estimated line obtained on the basis of the results of Equation (3), defined earlier as ln δ_j = τ_o + τ₁ln(θ). In Figure 4, this estimated line defines the slope within the data better. It also facilitates a better reading of the bumps within the funds.

Figure 4.

The estimated curve for 1940–2010: Real per capita grants for community and regional development.

The Functional Data and Theory: A Reassessment of the Initial Findings

As graphed, transfers of funds present unusual patterns even after adjustments, as exhibited by the scatter and sequence plots (Figures 1, 2, and 3). The initial findings from applying Equation (1) to the data indicate that the expected slope does not support the theory-based prediction of a straight line in incremental discussions of government expenditures. What is surprising here is that the findings indicate a more exponential growth pattern from the application of Equation (3). However, the bumps and amplitude embedded in the data are not fully accounted for. Though the proportion of the variation in the real per capita transfers explained through the transformation process increased to more than 80%, these percentages could still be better. In fact, other plausible statistical measures are available for realigning and integrating the initial findings, the functional data, and the bumps with theory for a more meaningful interpretation. Therefore, an addition could be made to the procedures adopted for the initial findings. For instance, certain disjointed analytical models could be used in view of the data’s unusual patterns and bumps around the 1970s and 1980s.

An alternative approach includes a splitting of the overall adjusted functional data into two parts along the horizontal axis. The distribution presented by the sequence plot in Figure 3 is used as a basis for splitting the data into two unequal periods. As indicated earlier, the slope is discontinuous at X = 0. This discontinuity at point X = 0 provides a justification for examining the two periods separately. The first period includes the years 1940 through 1982. This first period also includes the peak within the distribution, where X = 0. The second period includes the years 1983 through 2010. The adjusted data for the two periods are again tested using Equations (1) and (2). The statistical findings are presented in Tables 5 and 6.

Table 5.

The Effects and Slopes for 1940–1982: Results of Equations (1) and (2).

Fixed effects	Coefficient	Standard error	t-ratio	Adjusted R-square
Equation 1
Dependent, δ_j
Intercept, τ_o	₋6.213	2.591	−2.398**	.619
(θ), τ₁	0.836	0.102	8.224**
Equation 2
Dependent, ^δj1
Intercept, τ_o1	8.170	2.581	3.166**	.837
(θ), τ₁₁	−1.097	0.270	−4.063**
(θ)^*, τ₁₂	0.044	0.006	7.387**

Source. Authors’ calculations.

Note. The vertical axis (δ) represents community and regional development grants and is defined as a function of (θ) and (θ)^* on the horizontal axis.

Statistical significance at the 5% level or greater.

Table 6.

The Effects and Slopes for 1983–2010: Equations (1) and (2).

Fixed effects	Coefficient	Standard error	t-ratio	Adjusted R-square
Equation 1
Dependent, δ_j
Intercept, τ_o	13.181	1.821	7.238**	.476
(θ), τ₁	0.555	0.110	5.055**
Equation 2
Dependent, ^δ*j
Intercept, τ_o1	20.478	2.220	9.224**	.685
(θ), τ₁₁	−0.905	0.353	−2.564**
(θ)^*, τ₁₂	0.050	0.012	4.262**

Source. Authors’ calculations.

Note. The vertical axis (δ) represents community and regional development grants and is introduced as a function of (θ) and (θ)^* on the horizontal axis.

Statistical significance at the 5% level or better.

For each of the two periods, the quadratic function in Equation (2) fits the data perfectly well. Both spreads are curvilinear, and the proportions of variations explained by the new procedure increased significantly. These proportions are defined by the respective adjusted R-square. The estimated rate of change for the first period (1940–1982) is .044. The rate for the second period (1983–2010) is .05. The graphs from both periods are presented in Figures 5 and 6.

Figure 5.

The estimated curve for 1940–1982: Real per capita grants for community and regional development.

Figure 6.

The estimated curve for 1983–2010: Real per-capita grants for community and regional development.

Discussion

This study looked at census data where state and local governments reported their receipt transfers of funds by broad categories. The federal transfers, or grants-in-aid, to state and local governments constitute approximately 5% of the major grant categories over the years. This proportion does not provide evidence regarding the importance of federal government grants in community and regional development; that question was not considered here. This study did not look at amounts budgeted for economic development or all federal programs or policies that could affect economic development. Certainly the role and contributions of the federal government to local economic development are not limited to direct transfers of grants to states and municipalities. The various economic development objectives that were accomplished over the years by state and local governments through these transfers also were not discussed in this study.

The overall involvement of the federal government in local and regional economic development activities over a 71-year period were not addressed in this study. Based on the findings of the study, the growth in the overall fund transfers for the 71-year period averaged an annual change of approximately 5%, as was derived through the application of Equation (3); but simply stating this average is misleading because it did not occur in a linear pattern as presented in Figure 5. Yet the growth was significant enough to override the combined effects of prices and population changes. In effect, the slope that emerged from the data is not equal to zero, and the average rate of change, which is 5%, is a real rate of change in constant 1982–1984 dollars. Further, the distribution of the adjusted data reached a peak in the early 1980s and subsequently declined, thus making it impossible for a straight line estimation model to yield a reliable outcome. Instead, an exponential model is suitable for the overall data and provides slightly superior statistical results.

As an alternative, the data were split into two periods, resulting in what Figure 7 shows as a double dip. The results of the divided data suggest that additional data, including presidential and congressional policy initiatives, should be considered in evaluating the per capita spending. The results indicate an initial surge in the mid-1950s that expanded during the 1960s and 1970s and peaked in the 1980s. The period of decline in the early 1980s is consistent with the Reagan administration’s New Federalism approach, and is followed by a new surge. Therefore, splitting the adjusted data into two parts offers important policy insights. It also facilitates a reassessment and redefinition of the initial findings, thus simplifying the interpretation of the patterns and bumps found in the observed data. In effect, the remarkable increases experienced in the early 1960s are supported by various national policy initiatives of the late 1940s that provided opportunities for enhancing efforts aimed at promoting community and regional growth and development.

Figure 7.

A nonlinear and nonincremental relationship.

In summary, based on the study’s assumptions and criteria presented earlier, the statistical tests and results portrayed above and in Figure 7 do not favor incrementalism. Perhaps it is expecting too much to show steady incrementalism over a period of 71 years. Even if the precepts of punctuated equilibrium theory were considered, where there are steady incremental periods with occasional punctuations or shocks, this does not comport with the graphic presentation of the data. It is not linear, and it is nonincremental in the long term.

On the other hand, significant policies came into existence during the study period, and though the study cannot exhaust the list of national programs that were enacted from the onset, presentation of a select few is in order. The Urban Renewal Program was created as part of the Urban Housing Act of 1949. Though the initial intent of the act was to promote urban housing, the flexibility provided by the provisions of the act made it possible for municipalities to apply the funds toward accomplishing economic development objectives. Federal funding for these programs continued until 1973, when it was terminated; it was replaced in 1974 by the Urban Development and Action Grant (UDAG) program. Crafted to be more competitive than its predecessor program, UDAG enabled municipalities to acquire land, develop infrastructure, and improve sites for economic development purposes. These policy initiatives explain in part the surge in transfers observed between the 1950s and early 1980s. These policy efforts and initial surge are further illustrated by the decennial partitions presented in Figure 8.

Figure 8.

The decennial partitions, 1940–2010.

Conclusion

Federal programs for funding community and regional development are important. This study considered transfers of federal funds to state and local governments for promoting community and regional development and determined whether these transfers were incremental in nature. It yields several findings. First, the economic development transfers represent approximately 5% of grants to state and local governments, which is lower than the other four major categories of health, income security, transportation, and education and training. Second, between 1940 and 2010, prices and population played a critical role in determining the shape and form of the transfers. However, the initial hypothesis of incremental growth in resource allocation is not supported by the data and findings. An incremental data pattern would show steady linear growth and that did not emerge here. It may be asking too much of incremental theory to assume steady conservative growth over a period of 71 years.

But the graphic results showed something beyond a lack of linear incremental funding. The results showed exponential relationships. There was a double U-pattern with upward changes in the 1960s and 1970s, reaching a peak in the early 1980s, followed by periods of decline in the late 1980s and 1990s. These periods were followed by gradual increases during the mid-1990s through the year 2010.

When constant dollars are used and adjustments made based on population size, there were significant effects. These effects were graphically visible across the two sequence plots of the unadjusted and adjusted data. At the same time, from a resource allocation perspective, real changes occurred in the data over the years studied. The increases were significant enough to overshadow the influences of prices and population size on the data.

From a policy analysis perspective, the staggered nature of the data plots and lack of consistency in trajectory underscore the variability in policy orientation and priorities over the years analyzed. This study did not consider these specific policy changes, and this remains an open question for further inquiry. Importantly, as expressed in 1982–1984 prices, there were eras of real physical changes or increases, followed by decreases stretching over 25 years. These years were subsequently followed by a surge in the mid-1990s, thus making the resulting slope nonlinear in the long term.

In considering the policy relevance and prospects for future studies, there are further issues to explore. This study began with using existing federal data regarding transfers, but the available data are rather lumpy. The policy relevance of federal transfers can be seen as distinct in different eras. The question of how these funds were useful or beneficial to cities, regions, or states is a question yet to be considered. Here, the use of regionalism as a framework provided the context for examining the contribution of the national government to the growth and development efforts of states and local governments.

Recent trends of global influences on cities, regions, and states raise interesting policy questions for further analysis. The question in this study considered the extent to which federal government transfers were made to supplement growth and development efforts in the regions and local communities; it concentrated exclusively on intergovernmental transfers to states and local governments. This study suggests that considering a framework for justifying the state’s role in intergovernmental transfers for local and regional economic development is also a question for further study.

From a policy standpoint, the study’s finding of an average rate of change of 0.052 or 5.2% over the years examined constitutes a significant contribution to the literature and to our understanding of the role of the national government in promoting growth in regions through its transfer programs. From a regionalist perspective, and considered over 71 years, the cumulative effects of these annual changes are measurable in terms of productivity change, population growth, employment, and income creation. These effects coincide with the major categories of economic influence within the regionalist argument.

Two relevant policy and research questions can be drawn from the current study for future inquiry. First, both policymakers and practitioners might be interested in assessing the federal government’s effects on regional and local economic development efforts of subnational governments through studies of how the grants were used. This area of inquiry can enhance our understanding of the overall contribution of the national government to the promotion of growth and development in regions and local communities. This study focused on intergovernmental transfers, which constitute only a segment of the federal government’s contribution toward the promotion of growth and development.

Second, a related study might examine the specific mechanisms often applied to induce and promote growth in regions and local communities. These mechanisms are directly the result of the federal government’s intergovernmental transfers. Finally, federal funds for economic development are a policy choice that shows that federal transfers affect regional and local governments in a broad sense. The more detailed and nuanced manner of how the funds are used is also an important concern. In effect, studies that incorporate analysis of the types of programs funded and how the aid is used at state and local levels would advance the understanding of how the aid achieved policy goals.

Footnotes

Appendix A

Appendix B

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Francois K. Doamekpor is an associate professor in the Department of Public Administration and Urban Studies. He earned his Ph.D. in public and international affairs from the University of Pittsburgh, Pennsylvania. His teaching and research interests include public budgeting and finance, statistics and methods, fiscal analysis, and comparative management. His research studies have appeared in several journals, including Public Finance and Management, The Journal of Public Budgeting, Accounting and Financial Management, International Economic Journal, and the Journal of Developing Societies.

Julia Beckett is an associate professor in the Department of Public Administration and Urban Studies. She earned a J.D. from Washington University in St. Louis and her Ph.D. in public administration from the University of Colorado at Denver. Her research interests include public budgeting and finance, government accountability, democratic governance, and public law. She has written two books on public law and public administration. Her work has appeared in The American Review of Public Administration; Public Finance and Management; The Journal of Public Budgeting, Accounting and Financial Management; Public Administration Review; 21st Century Political Science: A Reference Handbook; and The Encyclopedia of Public Policy and Public Administration, 2nd ed.

References

Abramovitz

(1956). Resource and output trends in the United States since 1970. American Economic Review, 46, 5–23.

Baldwin

Forslid

Martin

Ottaviano

Robert-Nicoud

(2003). Economic geography and public policy. Princeton, NJ: Princeton University Press.

Bartik

T. J.

(2003). Local economic development policies. Kalamazoo, MI: W.E. Upjohn Institute for Employment.

Bartik

T. J.

(2005). Solving the problems of economic development incentives. Growth and Change, 36, 139–166.

Bendavid-Val

(1991). Regional and local economic analysis for practitioners (4th ed.). New York, NY: Praeger.

Bennett

Giloth

R. P.

(2008). Economic development in American cities: The pursuit of an equity agenda. Albany, NY: SUNY Press.

Berry

W. D.

(1990). The confusing case of budgetary incrementalism: Too many meanings for a single concept. Journal of Politics, 52(1), 167–196.

Berry

F. D.

Berry

W. D.

(2007). Innovation and diffusion models in policy research. In Sabatier

P. A.

(Ed.), Theories of the policy process (2nd ed., pp. 223–260). Boulder, CO: Westview.

Bingham

R. D.

Mier

(Eds.). (1993). Theories of local economic development: Perspectives from across the disciplines. Thousand Oaks, CA: Sage.

10.

Blakely

E. J.

Bradshaw

T. K.

(2002). Planning local economic development: Theory and practice (3rd ed.). Thousand Oaks, CA: Sage.

11.

Buss

T. F.

(2001). The effect of state tax incentives on economic growth and firm location decisions: An overview of the literature. Economic Development Quarterly, 15, 90–105.

12.

Conlan

T. J.

Posner

P. L.

(2008). Intergovernmental management for the twenty-first century. Washington, DC: Brookings Institution.

13.

Center for Urban Economic Development. (1987). Community economic development strategies: A manual for local action. Chicago: University of Illinois at Chicago.

14.

Denison

E. E.

(1962). Sources of economic growth in the United States and the alternatives before us (Suppl. 13). New York, NY: Committee for Economic Development.

15.

Downs

(1979). Key relationships between urban development and neighborhood change. Journal of the American Planning Association, 45(4), 462–572.

16.

Fainstein

Hill

Judd

Smith

M. P.

(1983). Restructuring the city: The political economy of urban development. New York, NY: Longman.

17.

Florida

R. F.

(2002). The rise of the creative class . . . and how it’s transforming work, leisure, community, and everyday life. New York, NY: Basic Books.

18.

Greenwood

T. D.

Holt

R. P. F.

(2010). Local economic development in the 21st century: Quality of life and sustainability. Armonk, NY: M. E. Sharpe.

19.

Griliches

Jorgenson

D. W.

(1966). Sources of measured productivity change: Capital input. American Economic Review, 55, 50–61.

20.

Griliches

Jorgenson

D. W.

(1967). The explanation of productivity change. Review of Economic Studies, 34(99), 249–283.

21.

Hayes

M. T.

(1992). Incrementalism and public policy. White Plains, NY: Longman.

22.

Hendrick

(1998). The impact of federal grants and other funds on general fund expenditure decisions: A detailed analysis of one city. Journal of Public Administration Research and Theory, 8(3), 353–390.

23.

Kleinbaum

D. G.

Kupper

L. L.

Muller

K. E.

Nizam

(1998). Applied regression analysis and other multivariable methods (3rd ed.). Pacific Grove, CA: Duxbury.

24.

Koven

S. G.

Lyons

T. S.

(2006). Economic development: Strategies for state and local practice. Washington, DC: International City/County Management Association.

25.

Levy

(1990). Economic development programs for cities, counties, and towns (2nd ed.). Westport, CT: Praeger.

26.

Lindblom

C. E.

(1959). The science of muddling through. Public Administration Review, 19, 79–88.

27.

Lynch

R. G.

(2004). Rethinking growth strategies: How state and local taxes and services affect economic development. Washington, DC: Economic Policy Institute.

28.

Markusen

A. R.

(2007). Reining in the competition for capital. Kalamazoo, MI: Upjohn Institute for Employment Research.

29.

Markusen

A. R.

Nesse

(2007). Institutional and political determinants of incentive competition. In Markusen

A. R.

(Ed.), Reining in the competition for capital (pp. 1–41). Kalamazoo, MI: W. E. Upjohn Institute for Employment Research.

30.

Nelson

A. C.

(1993). Theories of regional economic development. In Bingham

R. D.

Mier

(Eds.), Theories of local economic development: Perspectives from across the disciplines (pp. 27–60). Thousand Oaks, CA: Sage.

31.

Nivola

P. S.

2002. Tense commandments: Federal prescriptions and city problems. Washington, DC: Brookings Institution.

32.

O’Toole

L. J.

Jr. Christensen

R. K.

(2012). American intergovernmental relations (5th ed.). Washington, DC: CQ Press.

33.

Porter

M. E.

(2000). Location, competition, and economic development: Local clusters in a global economy. Economic Development Quarterly, 14, 15–34.

34.

Portney

K. E.

(2003). Taking sustainable cities seriously: Economic development, the environment, and quality of life in American cities. Cambridge, MA: MIT Press.

35.

Posner

P. L.

(1998). The politics of unfunded mandates. Washington, DC: Georgetown University Press.

36.

Ramsay

J. O.

(2002). Multilevel modeling of longitudinal and functional data. In Moskowitz

D. S.

Hershberger

S. L.

(Eds.), Modeling individual variability with repeated measures data: Methods and applications (pp. 87–102). Mahwah, NJ: Lawrence Erlbaum Associates.

37.

Robinson

E. S.

Caver

Meier

K. J.

O’Toole

L. J.

Jr. (2007). Explaining policy punctuations: Bureaucratization and budget change. American Journal of Political Science, 51(1), 140–150.

38.

Sabel

C. F.

(1992). Studied trust: Building new forms of cooperation in a volatile economy. In Pylee

Sengenberger

(Eds.), Industrial districts and local economic regeneration (pp. 215–250). Geneva, Switzerland: International Institute for Labour Studies.

39.

Shuman

(1998). Going local: Creating self-reliant communities in a global age. New York, NY: Free Press.

40.

Solow

R. M.

(1957). Technical change and the aggregate production function. Review of Economics and Statistics, 39, 312–320.

41.

SPSS Base 9.0. (1999). SPSS user’s guide. Chicago, IL: SPSS Inc.

42.

True

J. L.

Jones

B. D.

Baumgartner

F. R.

2007. Punctuated equilibrium theory: Explaining stability and change in American policymaking. In Sabatier

P. A.

(Ed.), Theories of the policy process (pp. 97–116). Boulder, CO: Westview.

43.

Volden

(2007). Intergovernmental grants: A formal model of interrelated national and subnational political decisions. Publius: The Journal of Federalism, 37(2), 209–243.

44.

Walker

D. B.

(2000). The rebirth of federalism (2nd ed.). New York, NY: Chatham House.

45.

Warner

M. E.

(2001). Local government support for community-based economic development. In The municipal yearbook (pp. 21–27). Washington, DC: International City/County Management Association.

46.

Wildavsky

(1964). The politics of the budgetary process. Boston, MA: Little, Brown.

47.

Zheng

Warner

(2010). Business incentive use among U.S. local governments: A story of accountability and policy learning. Economic Development Quarterly, 24(4), 325–336.