The Impact of Economic Development Strategy on Market Conditioned Income Inequality in the American States

Abstract

Langer found that the supply-side (locational) development policy increased income inequality in the American states, while demand-side (entrepreneurial) policy decreased inequality. Recent scholarship emphasizes the impact of unionization on state inequality. This analysis examines economic development policy and unionization in the same model. Using time series data (1983–2004) with an error correction model, it finds that entrepreneurial activism is associated with rising state inequality, while locational activism has no significant relationship. Additionally, entrepreneurial policy exerts a positive impact on inequality as union density rises, contrary to expectations. Skills-biased technical change is proposed to explain these findings.

Keywords

economic development policy unionization income inequality power resources entrepreneurial policy

State governments are making increased efforts to promote economic investment and employment opportunities in an expanding global economy. Much research emphasizes how state development strategy impacts job growth, unemployment, and inward investment (Brace 1993; Atkinson and Andes 2010; Saiz 2001a; Turner 2003). In addition, researchers examine the effect of industrial recruitment development programs on revenues for education and social programs (Burkdull and Tuman 1999; Hansen 2001). Others show how development policy encourages competitive bidding for investment eroding wages and regulatory protections (Saiz 2001b; Hanson 1993; Peters and Fisher 2004).

There is less research on the distributional consequences of these policies. Many studies on job growth do not consider the fairness of employment gains for workers at differing skills levels. Nonetheless, this is of critical importance to public servants. States with low average wages grapple with higher crime, more children in poverty, and lower per capita income (Atkinson and Andes 2010; Hansen 2006). Even when state labor markets shift from manufacturing to high-tech employment, income inequality may rise (Boeckelman 1995). Inequality generates higher crime, lowers public health outcomes (Neckerman and Torche 2007), reduces individual happiness (Hout 2003), and may fuel public demand for redistributive policies (Jacobs and Skopcol 2005). Given such consequences, government officials need to evaluate how their development programs impact income inequality.

Laura Langer’s (2001) analysis directly examined the link between the state economic development strategies and income inequality. She found that strategies designed to produce high-skilled jobs in immerging industry sectors decrease income inequality, and strategies that attract outside investment by keeping wages low increase inequality. She also found higher overall levels of economic policy activism were associated with increased inequality.

Still, other factors in the state political environment—namely, government ideology and unionization—are also important to income inequality levels (Kelly and Witko 2012). These “power resources” affect the degree to which working-class citizens can access the state’s economic policy-making process. Thus, public servants concerned about economic equality need to consider how unionization and left governments relate to development policy.

This article considers all three factors together. First, it retests the hypothesis that the type of development policy used by the states can influence levels of inequality. Its results indicate that entrepreneurial (demand-oriented) strategies to development increase inequality when controlling for unionization levels and government ideology, while locational (supply-side) strategy has no significant impact. Moreover, the overall independent effect of unionization is a decrease in inequality. Second, this article considers how unionization and left government liberalism condition the effect that entrepreneurial development policies have on distributional outcomes. It finds no support for a significant conditioning effect of left government liberalism, but it does find, surprisingly, that the interactive effect of unionization and entrepreneurial policy raises inequality. This suggests that unionization does not act as a working-class power resource in states that emphasize entrepreneurial development policy.

Economic Development Policy Variation and Distributional Consequences

Over the last three decades, states pursued business relocations and new capital growth through a wide variety of government-sponsored programs, including tax incentives, job training, and public–private venture capital funds (Brace 2002; Gray and Lowery 1990; Eisinger 1990; Peterson 1995). Economic development programs accelerated during the eighties due to competitive pressures between states to lure businesses and even more so after the Republican “contract with America” in 1994 (Saiz and Clarke 2012). However, some states utilize more active strategies than others. Higher levels of economic policy activism are associated with rising levels of income inequality according to one study (Langer 2001). Generally, state development efforts prioritize number of new jobs or investment dollars secured, rather than equity-related outcomes of employment growth and increased investment.

Nevertheless, the types of economic development tools are abundant (Friedman and Associates 2006; National Association of State Development Agencies 2002), and states engage in a diverse and varied mix of these programs. The differing types of development strategies utilized by states could result in divergent distributional consequences. Scholars developed several unique economic development typologies, categorizing strategies according to their approach to labor relations, capital accumulation, regulatory enforcement, and degree of state interventionism (Gray and Lowry 1990; Leicht and Jenkins 1994; Grant, Wallace, and Pitney 1995; Saiz 2001a; Langer 2001). Many of these classification schemes are based on Peter Eisinger’s distinction between supply-side and demand-side strategies that emerged during the 1980s (1989).

Supply-side policies, pejoratively known as “smokestack chasing” programs, follow a more traditional approach to development. These programs offer generous incentives to lure out-of-state firms in an attempt to outbid rival states also in search of new investment. These approaches are also known as “locational strategies,” given their focus on business relocation (Saiz 2001a). Supply-side policies offer a low-cost business environment, usually by promoting a base of cheap, available labor with reduced tax burdens. Job credit tax exemptions, subsidized worker training, state-sponsored loan guarantees for construction, and accelerated depreciation of industrial machinery are examples of incentives provided by this “business friendly” strategy (Schweke, Rose, and Damson 1994).

On the other hand, demand-side policies prioritize new capital growth rather than capital relocation. These strategies attempt to build a “wealth-generating economy” by promoting high-tech, finance, “green” energy, or communications industries. The government has a more interventionist role in demand-oriented strategies than it does in supply-side programs. Demand-side development goes beyond the provision of tax relief and other passive administrative incentives; it makes government and business active partners in growing capital (Saiz 2001a). Public–private venture capital funds, business incubators housed at state higher education institutes, and research and development partnerships are features used in the “entrepreneurial,” demand-oriented approach (Langer 2001).

Langer’s time series study of the year 1976–1994 found that the “quality jobs” focus of the entrepreneurial approach lowered inequality, presumably by generating higher wages for workers generally. Supply-side programs generated increased inequality due to their low-wage approach. But variables omitted in the Langer study, in particular unionization, demonstrate a significant relationship to state inequality levels in recent research (Kelly and Witko 2012; Apergis, Dincer, and Payne 2010). A major contribution offered by this article is that it controls for unionization and government ideology, as well as the conditioning effects of both factors on economic development policy, in an updated analysis.

Additionally, some research questions if wealth generation in the state economy actually promotes equality. Earlier studies conducted at the state and county level indicate that states with the highest levels of mean income experience greater inequality, with gains concentrated in the upper half of the income distribution (Nelson 1984; Braun 1991). Growth in “knowledge economy” sectors, which tend to produce highly paid finance and information technology jobs, fueled greater inequality in some states (Boekelman 1995).

Nevertheless, entrepreneurialism has become a preferred economic growth strategy among many politicians and practitioners of economic development (National Governors Association, Center for Best Practices 2004). The costs of competitive bidding wars between states soured many professionals toward the “smokestack-chases” of the supply-side approach, and leaders increasingly promoted entrepreneurial development strategies in the 1980s as a better route (Eisinger 1989). “Growth from within” could replace an outdated and inefficient model that pushed states to “beat out” neighboring states in a mad dash for investment dollars. Moreover, the increased wages offered by entrepreneurial professions offered the promise of richer revenues for public programs, further enhancing the quality of life of citizens. By 2002, associations such as the National Governor’s Association called to end competitive bidding and adopt the entrepreneurial approach (see Brace 2002, 174). While locational programs to development are still widely used, governments recommend entrepreneurial strategies and alternative “third-wave” programs (Brace 2002) as more responsible approaches.

An updated examination of the distributional consequences of varying development strategies is critical. If entrepreneurial development, counter to assumptions, actually exacerbates inequality among workers state leaders may find it advantageous to adjust their programs, similar to their scrutiny of industrial recruitment policies. This article evaluates if entrepreneurial programs really are a “high road” to more equitable growth (quoting Hanley and Douglass 2014) when controlling for unionization and left governments.

Power Resources Theory and Economic Development

Left governments are as a means by which lower-income citizens could push for policies favorable to their class interests. Unions, which organize workers into a cohesive political force, have traditionally provided a vehicle for electing left governments and for raising workers’ wages through collective bargaining. These “power resources,” which give working classes access points to influence governmental and market outcomes, are essential factors to consider in understanding variation in wage inequality (Korpi 1989; Huber, Ragin, and Stephens 1993; Miles and Quadogno 2002). Unionization and left-leaning state governments are associated with lower levels of pretransfer “market conditioned” inequality, even prior to government redistributive spending from programs such as social security or welfare (Kelly and Witko 2012). These critical control variables need to be included in a multivariate model to accurately gauge the independent effect of economic development policy on inequality.

The natural assumption is that unionization acts to exert upward pressure on incomes for workers in the lower levels of the income distribution, thus reducing overall levels of wage inequality. Several studies confirm this view (Pontusson, Rueda, and Way 2002; Radliff and Saiz 1998; Card 1998, 2001). However, given the severe erosion of union density across the American states in the past two decades (Hirsch, Macpherson, and Vronan [2001] 2012), it is clear that unions have lost ground in representing the bulk of the American working class. Data show that in the United States union membership shifted away from the manufacturing sector, historically the anchor of working-class trade unionism, in recent decades. Unions gained members in the service sector, which includes retail, food service, and other low-paid occupations, but only modestly. However, union membership in the professional and management sectors increased substantially (Bureau of Labor Statistics 2012), probably due to rising unionization among governmental workers (Bureau of Labor Statistics 2015; see online supplemental Figure 1 to view shifting patterns in unionization). The assumption that American unions still provide a power resource to lower-skilled workers demands rethinking. It appears that unions are beginning to concentrate their advocacy efforts on middle-income “white-collar” employees rather than lower-income earners in the twenty-first century economy.

Union strength across the different occupational groups is likely related to state-specific policies on labor relations and economic development. Scholars often point to right-to-work legislation, which prohibits mandatory dues deduction from employees covered under a union contract, as a hindrance to union organizing in the private sector (Mayer 2004; Davis and Huston 1995). Collective bargaining protections for public employees also impact labor union strength (Farber 2005), in particular among professional occupations such as public school and university teaching. The American states vary significantly in the enactment or absence of these laws (Amberg 2008). Generally, the Southeast and Mountain West maintain low union density through labor laws unfavorable to union organizing, while the Northeast, industrial mid-West, and Pacific Coast have higher union densities due to more union-friendly laws.

State economic development policy may also significantly affect which classes of workers unionize. States only have partial authority in redistributive and labor relations policy making (Kelly and Witko 2012; Peterson 1995; Witko and Newmark 2010), while economic development policy is largely within their discretion. Different economic development strategies have the potential to grow different skill and income classes of workers (Eisinger 1989). This in turn impacts the types of workers, whether middle class or lower class, who would be the beneficiary of union wage increases across the states. Economic development policy and unionization are likely to interact together in ways that relate to distributional outcomes.

Interaction of State Economic Development and Unionization in Impacting Wage Inequality

The article theorizes that entrepreneurial policies do not equalize the income distribution under some conditions. Entrepreneurial strategies prioritize production of “quality” high-paying jobs, which may actually serve to exacerbate inequality levels among workers, because those with higher skills levels could command a “skills wage premium.” This is especially true if there is no countervailing force present to place upward pressure on the wage rates of lower-skilled workers (Greenspan 1996; Berman, Bound, and Manchin 1998). While quality jobs may increase overall wealth and resources for social redistribution, they may also spread the range of wage rates paid across the labor skills spectrum. Moreover, professional and “knowledge economy” employees are more individualistic in their approach to wage bargaining and demand premium wages, while lower-skilled workers are more likely to accept the wage rates that prevail, without demanding individual premiums (Pontusson, Rueda, and Way 2002). Thus, growth in entrepreneurial occupations could skew wage gains to the upper-end of the income distribution.

Thus, the equalizing impact of entrepreneurial policy depends on the presence of left governments and labor unions acting on behalf of lower-income workers, ensuring that they are the beneficiaries of new economic sectors nurtured. If “blue-collar” unions are active participants in economic policy development, they will likely focus on creation of jobs in sectors which employ lower-skilled workers. However, if unions do not advocate for lower-skilled citizens, or if job creation occurs mostly in sectors such as banking, finance, and information technology with low unionization, there will be less equalization in job gains across the skills spectrum. Moreover, if unions are concentrated in high-skills occupations their upward push on professional-sector wages would intensify the skills premium paid to those workers. When union strength resides among middle- and upper-middle class workers, unionization could interact with entrepreneurial policy activism to raise inequality.

As for locational policy, on first glance it does seem likely to increase inequality because it lures existing manufacturing industries with the promise of cheap labor. Cutting wages at the bottom end of the income distribution would seemingly increase inequality. However, if locational policies attracted jobs where few employment opportunities previously existed for lower-skilled workers, they may have an equalizing impact on distributional outcomes.

This article places the relationship between economic development policy and inequality front and center, but it also contends that power resource variables are necessary to include for two reasons. First, unionization and left governments need to be accounted for as key control variables, given the independent effect they likely have on inequality outcomes. Second, this article considers how these variables interact with economic development policy to affect distributional outcomes. The relationship of economic development policy to distributional outcomes likely depends on the presence of power resources, particularly unionization, and the specific classes of workers who benefit from these power resources.

Given the likely effect of skills-biased wage increases in states that promote an entrepreneurial policy strategy, this study offers the following hypotheses:

Hypothesis 1: Greater use of entrepreneurial policy will be associated with higher inequality.

Additionally, given that locational policy emphasizes the maintenance of low wages to attract businesses, it may depress wages at the lowest tier of the income distribution, therefore this article tests the following:

Hypothesis 2: Greater use of locational policy will be associated with higher inequality.

This article will also test the hypothesis that entrepreneurial development policy and unionization have an interactive relationship to inequality. The standard argument is that as union density rises, economic development policy will be likely to decrease inequality. This holds if unions primarily represent the lower-skilled working class. But in states where unions have limited representation among the working classes, entrepreneurial policy may fail to produce jobs to benefit lower-skilled employees and prioritize higher-skilled professional workers. This “skills premium” effect would exacerbate inequality.

Hypothesis 3: Entrepreneurial policy will increase inequality in lower union density states, but not in higher density states.

However, Hypothesis 3 rests on the assumption that unions articulate lower-class interests generally across all states. If unions have shifted to representing workers in the higher-paid professional classes in states that emphasize entrepreneurialism, they may intensify the “skills premium” effect with an additive “union wage premium” effect. In entrepreneurial-oriented states with substantial unionization in the professional sectors, inequality could actually increase as union density rises. Thus Hypothesis 4 is presented as an alternate hypothesis:

Hypothesis 4 (alternate hypothesis): Entrepreneurial policy will exert a positive impact on inequality in higher union density states.

Finally, while the interactive relationship between unionization and economic development policy is the primary conditional hypothesis this article explores, it is also interesting to consider how state left government liberalism could condition the effect of entrepreneurial policy on inequality. Left-leaning state governments might interact with entrepreneurial policy approaches to equalize income. Government leaders that are more ideologically liberal could promote entrepreneurial programs that benefit the lower half of the income distribution. Thus, this article considers one additional hypothesis:

Hypothesis 5: Entrepreneurial policy will decrease inequality in states with left-leaning governments but will increase inequality in states with right-leaning governments.

Methods and Data

The model estimation method for this study is the same one used in previous studies of inequality (Kelly and Witko 2012; Apergis, Dincer, and Payne 2010). The data used are both cross-sectional (different states) and time series (different years). The assumption of independent observations is critical to ordinary least squares regression; thus, the modeling strategy must adjust for autocorrelation present in time series data. Scholars recommend the use of panel-corrected standard errors along with a lagged-dependent variable to properly estimate over time change (Kelly and Witko 2012; DeBoef and Keele 2008; Beck 2001). ECM equations compute the change in the dependent variable (ΔY) between time points, rather than the actual value of the Y given a value of X. ECMs estimate two coefficients for each independent variable—a differenced value and a lagged value. The coefficient on the differenced value applies to the initial impact of the change in X to the change in Y, which occurs entirely at one point in time. The coefficient on the lagged value applies to the impact of X on the change in Y that is felt over a longer time span. If either the differenced or lagged value of the independent variable is significant, then one can conclude that variable impacts the dependent variable (Kelly and Witko 2012).¹ ECMs are one of the most versatile modeling strategies available for analyzing time series panel data, especially because they are suitable for use with both cointegrated and stationary time series (DeBoef and Keele 2008). ECMs account for both short-term shocks and long-term change dynamics common in time series analysis (see Online Supplemental Text 1 for more information).

Dependent Variable

The data for this study come largely from Kelly and Witko’s (2012) examination. It utilizes their measure of household pretransfer market-conditioned inequality as the dependent variable for years 1983–2004 (see also Kelly 2005). This measure calculates a Gini coefficient using income figures from private earnings collected from the Census Bureau Annual Social and Economic Supplement. It leaves out any income earned through government programs such as social security, welfare, or disability. States have less influence over most national-level transfer programs; thus, the use of the pretransfer measure of inequality is more appropriate in a state-level analysis. A previous study of economic development policy and inequality used the posttransfer Gini as the dependent variable (Langer 2001), so the use of the pretransfer measure is a new contribution. This measure ranges from 0 to 1, varying continuously, with 1 representing maximum inequality.

Key Independent Variables

Economic development policy variables

The major explanatory variables are entrepreneurial policy intensity and locational policy intensity measured by two index variables constructed by Martin Saiz (2001a). Saiz’ constructed his indices by coding attributes of state incentive programs found in the Directory of Incentives for Business Investment and Development in the United States. This catalog is one of the most extensive sources of data tracking economic development programs over a multiyear span available, hence enabling time series analysis. Saiz investigated attributes of all programs listed in the guide for multiple years and coded them as locational or entrepreneurial. Saiz coded an attribute as locational if the program emphasized costreduction through direct financial subsidies, tax relief, and/or low-cost loans. He also coded corporate subsidy programs as locational if they were administratively passive, paid for by foregone revenues instead of direct spending (Saiz and Clark 2012). He excluded programs targeted to specific business sectors or geographic locations within the state from the locational coding. Saiz coded program attributes as entrepreneurial if their incentives limited eligibility to higher-risk start-ups, small businesses, or certain “growth producing” sectors (communications, banking, and education). Programs designed to enhance modernization of production processes as well as programs which raised public–private venture funds were also classified as having entrepreneurial attributes. Saiz then developed an index value for each state, based on the frequency of attributes of each state’s program. A full description of the development of the indices is described in Saiz 2001a, with updates provided by Saiz in 2002 and 2006. Values of the index span the time range 1983–2004; however, indices values were produced intermittently and are not available for every year of this study. This research project used linear interpolation to derive values in years not available. (See Table 1 in the Online Supplemental Materials for more information on years available in the Saiz indices.) Index values range from about 0 to 3, varying continuously and expressed to the tenth of a decimal place. Both policy variables are expected to be positively signed, indicating that they increase inequality.

Power resources variables

The model includes the power resource variables of state union density (in percentage) and state government liberalism as critical control variables. Union density is the percentage of total workers (in both public and private sectors) that are members of unions for all years of study, 1983–2004 (Hirsch, Macpherson, and Vronan [2001] 2012, with updates for subsequent years). Unionization is expected to have a negative independent effect on inequality. As for the interactive effect on entrepreneurial policy, as union density rises, entrepreneurial policy is expected to have a negative effect on inequality. The state liberalism variable, developed by Berry, Ringquist, and Fording, measures left power in state governments by accounting for ideology and policy liberalism of state legislatures, rather than simple partisan control, and is widely used in political science research (Berry et al. 1998; see also Kelly and Witko 2012). This variable ranges continuously from 0 to 1 expressed to the hundredth of a decimal. A full description of the coding and measurement strategy for this variable is found in Berry et al. (1998), with updates provided to current years. Values are available for all years of analysis from 1983 to 2004. Left government liberalism is expected to have a negative independent effect on inequality.

Additional control variables

As did Kelly and Witko, this examination includes national-level governmental variables to control for democratic president and percentage democrats in congress, given the federalist structure of the U.S. political system. Higher values of both of these variables are associated with greater democratic control and are predicted to be negatively related to inequality.

Indicators to account for general economic conditions which could impact inequality levels are supplied in the models. State unemployment rate (in percentage; Kelly and Witko 2012, based on Bureau of Labor Statistics data) is anticipated to be positively related to inequality and gross state product growth (percentage change) is expected to decrease overall inequality. Also included are variables to account for manufacturing proportion of the gross state product as well as proportion of state labor force employment in government. Higher values of either variable are anticipated to be related to less inequality as government and manufacturing sectors may have a more compressed pay spectrum with higher average wages than is observed in the private service sector.

Neither the Langer nor Kelly and Witko studies accounted for international economic openness and its potential impact on inequality levels. Some researchers have questioned the conventional argument that globalization will lead to increased income disparities because capital mobility allows corporations to threaten relocation and force workers to accept lower wage rates and labor rights (Hansen 2006). While data on outward foreign direct investment (FDI) would be ideal to test this argument, only inward FDI jobs data are readily available at the state level (Brady and Wallace 2000 and updates, based on Bureau of Economic Analysis data). Thus, the author includes percentage of total state employment attributable to inward FDI to control for globalization impacts. If the conventional argument is correct, higher values of FDI will be positively related to inequality.

Demographic variables such as proportion nonwhite workforce and proportion elderly population both are expected to lead to more inequality. Minority workers tend to make lower average wages than whites, and elderly residents often rely more on transfer payments (such as social security) than privatesector sourced income, thus potentially creating conditions for greater pretransfer inequality. Although this examination focuses on private sector economic development rather than transfer payments as a determinant of inequality, a measure of welfare share, the fraction of state expenditures devoted to means-tested transfers, is included as a control variable. This is to account for the contention sometimes made on the right that higher levels of social assistance could lead to more inequality in the market through less labor force participation. The variable state imprisonment rate per 100,000 state residents is added (Bureau of Justice Statistics 2010), because previous studies indicate incarceration can act as a driver of income and racial inequality (Western 2002, 2006). State minimum wage, as tracked by Kelly and Witko (2012), is also included as a control variable and measured as the actual dollar amount of the state-mandated minimum wage for the year of observation. States which do not have a specific minimum wage law have been given a value of 0, though by law all states must abide by federal minimum wage standards (see Online Supplemental Table 1 for variable specifications as well as the data source(s) for each).

Table 1.

Error Correction Model—Dependent Variable: Δ State Market Conditioned Inequality, Years 1984–2004.

	Model I	Model II
State market conditioned inequality, t − 1	−.3841*** (.0459)	−.3936*** (.0464)
Δ Locational development intensity	.0085 (.0081)	.0086 (.0081)
Locational development intensity, t − 1	.0024 (.0022)	.0026 (.0022)
Δ Entrepreneurial development intensity	.0029 (.0058)	.0032 (.0058)
Entrepreneurial development intensity, t − 1	.0026* (.0015)	−.0014 (.0023)
Δ Union density (percentage)	.0000 (.0005)	−.0001 (.0005)
Union density (percentage), t−1	−.0003*** (.0001)	−.0007*** (.0002)
Δ State left government liberalism	−.0089** (.0040)	−.0085** (.0039)
State left government liberalism, t − 1	.0064** (.0025)	.0068*** (.0025)
Δ Democrat president	.0048** (.0022)	.0046** (.0022)
Democrat president, t − 1	.0030* (.0018)	.0027 (.0018)
Δ Percentage democrats in congress	−.0019*** (.0002)	−.0020*** (.0002)
Percentage Democrats in Congress, t − 1	−.0012*** (.0002)	−.0013*** (.0002)
Δ State unemployment (percentage)	.0020** (.0008)	.0019** (.0008)
State unemployment (percentage), t − 1	.0030*** (.0005)	.0030*** (.0005)
Δ Gross state product growth	−.0427*** (.0138)	−.0415*** (.0137)
Gross state product growth, t−1	−.0979*** (.0194)	−.0954*** (.0193)
Δ Inward FDI jobs (percentage)	.0008 (.0012)	.0009 (.0012)
Inward FDI jobs (percentage) t − 1	.0004 (.0004)	.0004 (.0004)
Δ Proportion nonwhite population	.5202*** (.1731)	.5340*** (.1719)
Proportion nonwhite population, t − 1	.0119 (.0079)	.0129 (.0079)
Δ Proportion elderly population	−1.1766* (.6574)	−1.0572 (.6475)
Proportion elderly population, t − 1	.3418*** (.0525)	.3531*** (.0536)
Δ Proportion manufacturing, GSP	.0615 (.0423)	.0622 (.0421)
Proportion manufacturing, GSP t − 1	.0310*** (.0090)	.0327*** (.0092)
Δ Proportion government employment	−.0827 (.2183)	−.0737 (.2171)
Proportion government employment, t − 1	.0344* (.0189)	.0381** (.0190)
Δ Welfare share	−.0596** (.0269)	−.0596** (.0268)
Welfare share, t − 1	−.0473*** (.0120)	−.0464*** (.0120)
Δ Imprisoned population (per 100,000 res)	−.0000 (.0000)	−.0000 (.0000)
Imprisoned population (per 100,000 res), t − 1	.0000* (.0000)	.0000* (.0000)
Δ State minimum wage	.0030** (.0015)	.0029* (.0015)
State minimum wage, t − 1	−.0004 (.0004)	−.0004 (.0004)
Lag entrepreneurial × lag union density		.0003** (.0001)
Constant	.1795*** (.0240)	.1879*** (.0244)
Observations	1,000	1,000
R ²	.326	.329
Wald χ²	256.23	262.85

Note: Panel-corrected standard errors are given in parentheses. FDI = foreign direct investment; GSP = gross state product.

***p < .01.

**p < .05.

*p < .1.

Model and Results

The first model in Table 1 tests Hypotheses 1 and 2 about the independent effect of the two economic development strategies on state inequality, including the critical control variables of unionization and left government liberalism. The coefficients for locational policy are not significant, thus failing to support the hypothesis that it raises inequality. However, entrepreneurial policy has a significant positive impact on inequality in the long term, confirming Hypothesis 1. Moreover, the independent effects of the power resources of state union density and left government liberalism are significant. Union density is related to decreased inequality but only in the long term. State government liberalism appears to have a negative initial impact on inequality, but over time higher levels of Left strength actually appear to increase inequality. This is counterintuitive, as Left-leaning governments are expected to produce economic policies in the interests of lower-class citizens. However, previous research supporting the equity role of Left policy liberalism reveal a limited relationship constrained to years following 1994 and only in the short term (Kelly and Witko 2012). This article does not find that Left policy liberalism substantively increases the state-level equality over the longer time span of 1984–2004.

The national-level control variables indicate that democratic congresses do decrease state-level inequality overall, but a democratic president, counter to expectations, has a positive impact on state-level inequality in both the lagged and differenced terms. Interestingly, only one Democratic President, William Jefferson Clinton, is represented in the data. Some of the other control variables conform to expectations. Unemployment raises inequality, as does higher percentage of nonwhite residents in the short term and elderly population in the long term, while higher rates of economic growth are linked to lower levels of inequality in both the long term and short term. Interestingly, both higher percentage of workers in manufacturing and government sectors appear to be significantly related to increases in inequality. This was unexpected. Additionally, the findings suggest that the initial implementation of state minimum wage laws raise inequality, but that over time they do not have a significant impact.

Inward FDI appears to have no significant relationship to state inequality levels, suggesting that domestic forces, rather than international trade and investment, relate more directly to income inequality. Contrary to some claims from the Right, welfare generosity appears to reduce, rather than intensify inequality, although it is important to note that in the mid-nineties essential changes were made to promote labor force participation as a condition of receiving temporary means-tested assistance. In any case, states with higher shares of welfare spending as a part of their overall expenditures appear to have declining or lower rates of inequality. Higher rates of imprisonment are significantly related to rising inequality, though the effect appears to be very small, given the near 0 value on the coefficients.

In an effort to subject these findings to additional scrutiny, supplementary models were estimated using the Atkinson index, an alternate measure of income inequality (see Atkinson 1970). This measure calculates a social welfare function to measure inequality, based on varying values of an equality aversion parameter (∊). The author estimated a model with an identical set of independent variables as included in Table 1, model I, substituting the differenced and lagged values of the Atkinson’s index for the Market Gini (using data from Frank 2014a and 2014b; see Online Supplemental Text 2 for robustness checks). This alternate model confirmed the positive relationship of entrepreneurial policy to inequality and the negative independent relationship of unionization to inequality (see Online Supplemental Table 3 for a model specification using the Atkinson’s index).

As a final robustness check, this project estimated a regression model substituting mean inflation-adjusted income of the bottom income quintile (US Census, Current Population Survey 2014) for the Gini coefficient. This examines whether the observed increase in inequality was also accompanied by an absolute decrease in the mean income of lower-income citizens. Lower-income earners may still experience a rise in income over time, even as their share of income relative to higher-income earners decreases. Results indicate that the long-term relationship of entrepreneurial and locational policy to mean income of the bottom quintile is negative; however, these coefficients do not meet the threshold of statistical significance. The independent effect of unionization, however, is a significant increase in mean income of the bottom quintile. (See Online Supplemental Table 4 for a model estimation using mean income of the bottom quintile).

So far, this article finds that the independent effect of unionization is equity inducing overall. However, unionization may not equalize incomes in those states which heavily utilize entrepreneurial strategies. Model II in Table 1 estimates an interaction term composed of unionization and entrepreneurial policy as a final independent variable. Four pairs of interaction terms were created from the lagged and differenced variables, but only the interaction term for the lagged values is listed in the table, because it was the only one found to be significant. It is necessary to take caution when interpreting the coefficients and significance levels on the separate constitutive terms making up the interaction term (Brambor, Clark, and Golder 2005). In model II, the coefficient and sign on the lag of entrepreneurial activism indicates its effect on inequality only when the lag of union density = 0. The significant coefficient of the interaction term does suggest that entrepreneurial policy’s effect on inequality changes according to the value of union density. As union density increases, entrepreneurial policy’s effect on inequality becomes positive and significant, confirming the alternate Hypothesis 4. Unionization does not mitigate the inequality produced by entrepreneurial policy, rather it reinforces it.²

As for the conditioning effect of state government liberalism on entrepreneurial policy, the author created factor variables for all pairs of differenced and lagged variables (not shown in Table 1 due to space needs), and none of them met significance levels.³ While left government liberalism has a significant independent effect on inequality, it does not condition the relationship of entrepreneurial policy to inequality. Thus, Hypothesis 5 receives no support.

Conclusion

This article shows that power resources are important control variables to include alongside economic development to explain state income inequality. Overall, union density exerts a significant negative effect on inequality over the long term, and state government liberalism equalizes income in the short term, though it might increase inequality in the long run.

Although unions act to decrease inequality in general, this study finds that entrepreneurial policy increases inequality at higher levels of union density, not at lower levels as expected. Unions fail to effectively articulate lower-class interests in states that engage heavily in entrepreneurial development. The American labor movement struggles to recruit lower-wage workers in a hostile regulatory and business environment (Bronfenbrenner 2009) and has shifted its advocacy to “white-collar” workers in some state economies as a result.

The most important finding of this article is that entrepreneurial development does not equalize state income distributions. The dot-com boom in the 1990s and expansion of investment and financial markets around the turn of the millennium reconfigured the job market. The “dark side” of entrepreneurial development is a “skills premium” effect that drives income inequality among the different classes of employees. Moreover, institutions such as unions and left-leaning governments largely fail to counteract the inequality that entrepreneurial policies produce. State leaders concerned about distributional outcomes should reconsider how entrepreneurialism and other economic development strategies can best promote economies of widely shared wealth. Economic development policy offers the promise of wealth generation, but more thoughtful program design may enable all classes of workers to reap its rewards.

Footnotes

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

Supplemental Material

The online data supplements are available at .

References

Amberg

Stephen

. 2008. “Liberal market economy or composite regime:” Institutional legacies and labor market policy in the United States. Polity 40:164–96.

Apergis

Nicholas

Dincer

Oguzhan C.

Payne

James E.

. 2010. The relationship between corruption and income inequality in U.S. states: Evidence from a panel cointegration and error correction model. Public Choice 145:125–35.

Atkinson

A. B.

1970. On the measurement of inequality. Journal of Economic Theory 2:244–63.

Atkinson

Robert

Andes

Scott

. 2010. The 2010 state new economy index: Benchmarking economic transformation in the states. Washington, DC: Information Technology and Innovation Foundation.

Beck

Nathaniel

. 2001. Time-series-cross-section data: What have we learned in the past few years. Annual Review of Political Science 4:271–93.

Berman

Bound

Machin

. 1998. “Implications of skill-based technological change: International evidence”. Quarterly Journal of Economics 113:1245–80.

Berry

William D.

Ringquist

Evan J.

Fording

Richard C.

Hanson

Russell L.

. 1998. Measuring citizen and government ideology in the American states, 1960-93. American Journal of Political Science 42:327–48.

Boeckelman

Keith

. 1995. The American states in the postindustrial economy. State and Local Government Review 27:182–87.

Brace

Paul

. 1993. State government and economic performance. Baltimore: Johns Hopkins University Press.

10.

Brace

Paul

. 2002. Mapping economic development policy change in the American states. The Review of Policy Research 19:161–78.

11.

Brady

David

Wallace

Michael

. 2000. Spatialization, foreign direct investment, and labor outcomes in the American states, 1976-1998. Social Forces 79:67–105.

12.

Brambor

Thomas

Clark

William

Golder

Matt

. 2005. Understanding interaction models: Improving empirical analysis. Political Analysis 13:1–20.

13.

Braun

Denny

. 1991. Income inequality and economic development: Geographic divergence. Social Science Quarterly 72:520–36.

14.

Bronfenbrenner

Kate

. 2009. No holds barred: The intensification of employer opposition to organizing. EPI Briefing Paper # 235, Economic Policy Institute, Washington, DC.

15.

Bureau of Economic Analysis. 2012. Wage and salary employment by industry 1983–2011. Data provided upon request.

16.

Bureau of Justice Statistics. 2010. Rate (per 100,000 resident population) of sentenced prisoners under jurisdiction of State and Federal Correctional Authorities on Dec. 31. Published as a table in the “Sourcebook of Criminal Justice Statistics Online.” http://www.albany.edu/sourcebook/csv/t6292010.csv (accessed December 31, 2012).

17.

Bureau of Labor Statistics. 2015. “Union Members Summary. Union Members–2014.” http://www.bls.gov/news.release/union2.nr0.htm (accessed November 1, 2015).

18.

Bureau of Labor Statistics, using Current Population Survey Data. 9 2012. Unpublished data provided request by BLS in September 24, 2012. http://www.bls.gov/cps/cpsoccind.htm (Accessed September 24, 2012)

19.

Burkdull

John

Truman

John P.

. 1999. Texas and the International Economy. State and Local Government Review 31:106–22.

20.

Card

David

. 1998. Falling union membership and rising wage inequality: What’s the connection? NBER Working Papers 6520, National Bureau of Economic Research, Inc, Cambridge, MA.

21.

Davis

Joe

Huston

John H.

. 1995. Right-to-work laws and union density: New evidence form micro data. Journal of Labour Research 16:223–29.

22.

DeBoef

Suzanna

Keele

Luke

. 2008. Taking time seriously: Dynamic regression. American Journal of Political Science 52:184–200.

23.

Eisinger

Peter

. 1989. Rise of the entrepreneurial state. Madison: University of Wisconsin Press.

24.

Eisinger

Peter

. 1990. Do the American states do industrial policy? British Journal of Political Sciences 20:509–35.

25.

Farber

H. S.

2005. Nonunion wage rates and the threat of unionization. Industrial and Labor Relations Review 58:335–52.

26.

Frank

Mark W

. 2014a. U.S. state-level income inequality data. Stata file of “Other Inequality Measures.” http://www.shsu.edu/eco_mwf/inequality.html.

27.

Frank

Mark W.

2014b. A new state-level panel of annual inequality measures over the period 1916-2005. Journal of Business Strategies 31:241–63.

28.

Friedman and Associates. 2006. Directory of incentives for business investment and development in the United States: A state-by-state guide. CD-ROM.

29.

Grant

D. S.

Wallace

Pitney

. 1995. Measuring state-level economic development programs, 1970-1992. Economic Development Quarterly 9:134–45.

30.

Gray

Virginia

Lowry

David

. 1990. The corporatist foundations of state industrial policy. Social Science Quarterly 71:3–23.

31.

Greenspan

Alan

. 1996. Address: Job insecurity and technology. In Technology and growth conference proceedings, ed. Furher

J. C.

Sneddon Little

, 173–81. Boston, MA: Federal Reserve of Boston Press.

32.

Hanley

Caroline

Douglass

Michael

. 2014. High road, low road, or off road? Economic development strategies and the American states. Economic Development Quarterly 28:220–29.

33.

Hansen

Susan B.

2001. The impact of a low-wage strategy on state economic development. State Politics & Policy Quarterly 1:227–54.

34.

Hansen

Susan B.

2006. Globalization and the politics of pay: Policy choices in the American states. Washington, DC: Georgetown University Press.

35.

Hanson

Russell

. 1993. Bidding for business: A second war between the states? Economic Development Quarterly 7:183–97.

36.

Hirsch

Barry T.

Macpherson

David A.

Vronan

Wayne G.

. (2001) 2012. Estimates of union density by state. Monthly Labor Review 124:51–55. www.unionstats.com (accessed October 1, 2012)

37.

Hout

Michael

. 2003. Money and morale: What growing inequality is doing to Americans’ view of themselves and others. Working Paper, University of California – Berkeley Survey Research Center. http://ucdata.berkeley.edu/ (accessed November 1, 2015).

38.

Huber

Evelyne

Ragin

Charles

Stephens

John D.

. 1993. Social democracy, christian democracy, constitutional structure, and the welfare state. American Journal Sociology 99:711–49.

39.

Jacobs

L. R.

Skocpol

Theda

, eds. 2005. Inequality and American democracy: What we know and what we need to learn. New York: Russell Sage Foundation.

40.

Kelly

Nathan

. 2005. Political choice, public policy, and distributional outcomes. American Journal of Political Science 49:865–80.

41.

Kelly

Nathan J.

Witko

Christopher

. 2012. Federalism and American inequality. The Journal of Politics 74:414–26.

42.

Korpi

Walter

. 1989. Power, politics, and state autonomy in the development of social citizenship: Social rights during sickness in eighteen OECD countries since 1930. American Sociological Review 54:309–28.

43.

Langer

Laura

. 2001. The consequences of state economic development strategies on income distribution in the American states, 1976-1994. American Politics Research 29:392–415.

44.

Leicht

K. T.

Jenkins

J. C.

. 1994. Three strategies of state economic development: Entrepreneurial, industrial recruitment, and deregulation policies in the American states. Economic Development Quarterly 8:256–69.

45.

Mayer

Gerald

. 2004. Union membership trends in the United States. Congressional Research Service Report for Congress, Washington, DC.

46.

Myles

John

Quadagno

Jill

. 2002. Political theories of the welfare state. Social Service Review 76:34–57.

47.

National Association of State Development Agencies. 2002. Directory of incentives for business investment and development in the United States: A state-by-state guide. Washington, DC: National Association of State Development Agencies.

48.

National Governors Association, Center for Best Practices. 2004. A governor’s guide to strengthening state entrepreneurial policy. www.nga.org (accessed November 1, 2015).

49.

Neckerman

Kathryn M.

Torche

Florencia

. 2007. Inequality: Causes and consequences. Annual Review of Sociology 33:335–57

50.

Nelson

Joel I.

1984. Income inequality: The American states. Social Science Quarterly 65:854–60.

51.

Peters

Alan

Fisher

Peter

. 2004. Commentary; the failures of economic development incentives. Journal of the American Planning Association 70:27–37.

52.

Peterson

Paul E.

1995. The price of federalism. New York: Twentieth Century.

53.

Pontusson

Jonas

Rueda

David

Way

Christopher R.

. 2002. Comparative political economy of wage distribution: The role of partisanship and labour market institutions. British Journal of Political Science 32:281–308.

54.

Radcliff

Benjamin

Saiz

Martin

. 1998. Labor organization and public policy in the American states. Journal of Politics 60:113–25.

55.

Saiz

Martin

. 2001a. Using program attributes to measure and evaluate state economic development strategies. Economic Development Quarterly 15:45–57.

56.

Saiz

Martin

. 2001b. Politics and economic development: Why governments adopt different strategies to induce economic growth. Policy Studies Journal 29:203–14.

57.

Saiz

Martin

Clarke

Susan E.

. 2012. Economic development and infrastructure policy. In Politics in the American States: A Comparative Analysis, 9th ed., ed. Gray

Virginia

Hanson

Russell L.

, 501–32. Washington, DC:CQ Press.

58.

Schweke

Rose

Damson

. 1994. Bidding for business: Are cities and states selling themselves short? Washington, DC: Corporation for Enterprise Development.

59.

Turner

Robert C.

2003. “The political economy of gubernatorial smokestack chasing: Bad policy and bad politics?” State Politics and Policy Quarterly 3:270–93.

60.

United States Census Bureau, Current Population Survey, Annual Social and Economic Supplements. 2014. Data provided by request. ftp://ftp2.census.gov/programs-surveys/cps/techdocs/cpsmar14.pdf) (accessed October 19, 2015)

61.

Western

Bruce

. 2002. The impact of incarceration on wage mobility and inequality. American Sociological Review 67:526–46.

62.

Western

Bruce

. 2006. Punishment and inequality in America. New York: Russell Sage Foundation.

63.

Witko

Christopher

Newmark

Adam J.

. 2010. The strange disappearance of investment in human and physical capital in the U.S. states. The Journal of Public Administration Research and Theory 20:215–32.

64.

Cowell

F. A.

1995. Measuring inequality. 2nd ed. New York: Simon and Schuster International, Harvester Wheatsheaf/Prentice Hall.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.32 MB