Rethinking Electoral Competition_ Legislative Balance_ and by maqboolshahin

VIEWS: 10 PAGES: 24

									Rethinking Electoral Competition, Legislative Balance, and American State Welfare Policy
Belinda C. Davis Department of Political Science Michigan State University davisbe@msu.edu

Doo-Rae Kim Department of Political Science Michigan State University drkim@msu.edu

Abstract Recent research in state politics has examined the manner in which political variables function in the policy making process, how they interact with one another, and what their effects on policy outcomes are. We add to this body of literature by distinguishing between factors that determine a state's location in the policy liberalism space and factors that affect the consistency of a state's policy outcome. We propose that partisan balance and elite liberalism determine a state’s policy liberalism by its role in fixing a state’s position in the policy space, while electoral competition, legislative professionalism, and citizen ideology function as mediators in that they affect the consistency with which a state takes its position in the policy space. We use the heteroscedastic normal regression model to empirically test hypotheses about which political variables contribute to position and to consistency. Analyses on welfare efforts in the American States between 1973 and 1992 show us that elite liberalism, rather than partisan balance, is confirmed as a determinant of a state’s welfare efforts. Moreover electoral competition, professionalism, and citizen ideology are confirmed as mediators.

Prepared for presentation at the annual meeting of the Southern Political Science Association, New Orleans, LA, January 8-10, 2004.

Introduction
For more than four decades, scholars of state politics have examined the role of political variables in the determination of policy choices or expenditures. In much of the early debate, there was considerable disagreement over whether or not political variables mattered at all in the determination of state policy (Dawson and Robinson 1963; Dye 1966; Hofferbert 1966; Carmines 1974; Fry and Winters 1970). However, methodological advancements that enabled more rigorous empirical testing of theories confirmed the role of political variables in the policy making process (Hanson 1983; Plotnick and Winters 1985; Wright, Erickson, and McIver 1987). More recent research has focused on the manner in which political variables function in the policy making process, how they interact with one another, and what their effects on policy outcomes are (Dye 1984, Barrilleaux, Holbrook, and Langer 2002). We hope to add to this body of literature on two fronts- theoretical and methodological. First, we distinguish between factors that determine a state’s location in the policy liberalism space and factors that affect the consistency of a state’s policy outcome. In other words, we believe that some political variables function as determinants of policy while others serve as mediators. Democratic Party strength and the ideology of a state’s elites are viewed as determinants while electoral competition, professionalism of a state’s legislature, and citizen ideology are seen as mediators. Second, we utilize Maximum Likelihood Heteroscedastic Normal Regression (Franklin 1992), a relatively new addition to political science literature, to test our theory.

1

Political Process Variables
The role of political process variables in the state policy making process has a long and rich history. Over the past forty years, we have seen researchers question whether or not politics matter at all in the process, eventually conclude that they do, and finally turn to the puzzle of how they affect the process and interact with one another. The most well-developed theoretical debate surrounding the role of political process variables has centered on inter-party competition and party control. We use it as an example of the current debate over political process variables. There are three views on the effects of partisan balance and inter-party competitio n on state policy making which have primarily been tested using state welfare policy. The first perspective contends that partisan balance in state legislatures determines the partisan leaning of welfare policy outcomes in the states: the stronger a liberal party is in a state, the more generous welfare policy outcomes are (Fry and Winters 1970, Winters 1976, Plotnick and Winters 1985, Garand 1985, Alt and Lowry 1994). This expectation depends on the assumption that state policymakers may try to produce policy outcomes that can maximize supports of preferred clientele groups. That is, outcomes of policymaking reflect the extent to which state elected officials in the state government reward or penalize certain social groups. The second perspective argues that rather than partisan balance, inter-party competition determines the ideological leaning of policy outcomes. This follows from Key’s (1949) observation that policies promoting the have-nots’ interests were pursued in those states in which there existed regular competitions between two cohesive and enduring factions in the South. Following this notion, many researchers contend that

2

close inter-party competition leads both liberal and conservative parties to provide the have- nots with more generous policy benefits (Plotnick and Winters 1985). The third perspective focuses on the joint effects of inter-party competition and partisan balance on state policy outcomes. For instance, Dye (1984:1100) put it:

[Previous studies] treated competition as a dir ect determinant of public policy. Measures of party competition were simply included as independent variables in regression equations, implying an additive function. [….] But these tests employing additive models do not tell us how political system characteristics might facilitate relationships between citizens’ demands and public policies.

In particular, inter-party competition is considered as a mediating factor that facilitates the translation of partisan balance into policy outcomes. Dye (1984) found that Democratic Party strength tends to increase welfare expenditures only in those states where party systems were competitive. Recently, Barrilleaux and others (Barrilleaux 2000; Barrilleaux, Holbrook, and Langer 2002) provide some theoretical rationales for Dye’s finding of a conditional effect. Barrilleaux (2000) contends that the translation of Democratic Party strength into liberal policy outcomes was conditional on changing GOP electoral strength: “rising GOP strength in the states should force Democratic legislatures to adopt more liberal public policies” (66). Barrilleaux, Holbrook, and Langer (2002) extend this logic further. In counterpoint to the median voter theory, they argue that political parties and candidates are likely to move toward the extremes to obtain the support of their “core” constituents when they encounter serious electoral threats. Therefore, legislators elected in those districts where serious electoral threats exist will take clearer party positions: Democratic legislators show stronger support for more

3

liberal policies, and Republican legislators show stronger opposition against liberal policies. In sum, research on the effects of electoral competition and party strength on state policymaking has developed from simple additive models that emphasize the independent effects of these two variables to joint effect models that focus on the mediating role played by electoral competition between party strength and policy outcomes.

Mediators Versus Determinants
The differentiated roles of political variables as determinants and mediators ha ve been discussed in the existing literature. For instance, Godwin and Shepard (1976) argued that we should treat political process variables as “modifying variables which either facilitate or inhibit the accurate translation of demands into outputs, i.e. mediational” (1129). That is, they maintained that all political process variables be limited to mediatorial roles. We differ from their argument by contending that some political variables affect consistency while others affect direction. Our goal is to more appropriately capture the dynamic process through which political process characteristics affect policy outcomes by considering not only factors that determine a state’s location in the policy liberalism space (aka determinants) but also other factors that affect the consistency of the states’ policy outcomes (aka mediators). (Figure 1 About Here) Figure 1 illustrates this dynamic process wherein a state’s policy position on the liberalconservative continuum are captured by µ’s and the consistency of the state’s positiontaking is represented by σ ’s. As the state’s consistency increases its position will be

4

more clearly represented on the policy space. Consider three hypothetical states, A, B, and C. State A’s position in the policy liberalism space is µA, which is more conservative than the other two states B and C that lie at µB=C. Even though states B and C lie at the same position in the policy space, the consistency differs from each other. State C shows a greater consistency ( σ C ) than does state B ( σ A= B ). Therefore, we need to identify factors that determine the directional change in the policy position from µA to

µB=C but also another set of factors that regulate the change in the consistency from σ A= B
to σ C . Given the rich history of the debate surrounding electoral competition and party strength, it seems that they are the place to start in this discussion, especially in light of the recent work by Barrilleaux et al (2002). They conclude that their results show “electoral competition to be an important linkage mechanism that aids in translating liberal party success into liberal policy outcomes” (2002: 424). In other words, electoral competition forces political parties to enact policies that are consistent with the beliefs of core constituents. In the face of competition, Democrats are moved toward more liberal policies while Republicans move toward more conservative policies. Party strength serves as a determinant of public policy while electoral competition mediates or affects consistency. In light of this, we offer two hypotheses.

H1 : Democratic Party strength in a state will increase state policy liberalism. H2 : Electoral competition will increase the consistency in state policy outcomes.

5

The degree of professionalism in a state’s legislature is another area in which a greater understanding of its function in policy outcomes is needed. Professionalism, as defined by (Mooney 1994), is “the enhancement of the capacity of the legislature to perform its role in the policy making process with an expertise, seriousness, and effort comparable to that of other actors.” The components of professionalism indices can take into account one or many of the following factors: number of committees, size of staff, legislator compensation, length of session, incumbency rate, and operating expenditures. What is the role of professionalism in the policy making process? One view suggests that professionalism serves to buffer legislators from outside forces when they are facing reelection (Berry, Berkman, and Schneiderman 2000; Carey, Niemi, and Powell 2000). Berry et al (2000) conclude that as legislatures professionalize, “elections are less likely to be affected by political and economic conditions outside incumbents’ control” (971). These findings are in line with much of the Congressional literature on professionalization and institutional boundaries: professionalism enables legislators to be autonomous in initiating policy and seeking information (Hibbing 1999; Montcrie f and Thompson 1992; Rosenthal 1998; Squire 1992). As a measure of institutional strength, it also can be argued that professionalism reflects the demands of citizens for greater intervention in the economy (Brace 1993). But others might argue that professionalism reflects a legislature staffed with individuals with greater policy knowledge who are biased toward program expansion (Weber and Brace 1999; Barrilleaux et al 2003). These views are consistent with Carmines (1974) argument that legislative professionalism and its increases in policy knowledge and greater resources will enable legislators to translate party preferences into policy.

6

Regardless of the view (buffer versus preference into policy), both see professionalism as a mediating factor but in very different fashions. If it is true that professionalism serves as a buffer that lessens the need for catering to constituent demands, then professionalism should decrease the consistency of policy outcomes. If it is true that professionalism with its increases in knowledge and resources leads to the translation of partisan interests into policy, then professionalism should increase the consistency of policy outcomes. We are more persuaded by the buffering argument and propose the following hypothesis.

H3 : Professionalism will decrease the consistency in state policy outcomes.

The liberal/conservative cleavage in state politics has played a key role in determining policy outcomes. This should come as no surprise given that democratic theory maintains that there should be congruence between the preferences of citizens, the ideology of elected officials, and government policies (Berry et al 1998). There are two ideological concepts that are of particular importance: elite ideology and citizen ideology. Elite ideology generally refers to the placement of elected officials on a liberalconservative continuum while citizen ideology generally references an attempt to place the preferences of citizens on that same continuum. Work by Erikson, Wright, and McIver (1993) supports the idea that state politics do what democratic theory proposes it should do; state politics translate citizens’ ideological preferences into policy outcomes. Their analysis of elites highlights, “the policy tension between the centrifugal push by party activists toward their ‘extreme’

7

ideologies and the centripetal pull toward the center by the more moderate electorate” (1993:345). This seems to imply that elite ideology determines a state’s position on a liberal-conservative policy continuum while citizen ideology determines consistency. Since party activists are always located at the extremes, it is movement along the liberalconservative continuum by citizens that should produce change. For instance, when a state’s citizens are relatively homogenous in their ideological preferences, skewed to extreme liberalism or conservativism, it seems likely that it is harder for elected officials to justify actions that deviate from constituent ideological preferences, especially in light of the fact that party activists and the mass electorate have preferences that are more closely aligned. Most scholars include citizen ideology in models as a direct determinant of policy (Hill and Leighley 1992; Hill, Leighley, and Hinton-Andersson 1995; Berry, Fording, and Hanson 2003; Tweedie 1994). In light of the argument drawn from Erickson, Wright and McIver’s (1993) research, this does not seem theoretically sound. We believe citizen ideology works to control the consistency of policy outcomes while elite ideology determines the position of a state on the liberal-conservative continuum of policy outcomes.

H4 : As the ideology of a state’s elites moves toward liberalism, state policy liberalism will increase. H5 : The ideology of a state’s citizenry will increase the consisitency in state policy outcomes when it is skewed towards extreme liberalism or conservativism.

8

Model
The differentiated roles of political variables in the state policymaking process have been well recognized in the existing literature. However, we believe that the conventional way of testing, which relies on single-equation regression modes, is not satisfactory. As illustrated by figure 1, it is difficult to examine effects of determinants and mediators on state policy outcomes by using single equation models. In fact, many researchers have pointed out the limitations of single equation models. For instance, Godwin and Shepard (1976) decried that the conventional model, which included all independent variables in single equation, was not useful in examining the role of political process variables as mediators between socio-economic factors and policy outcomes. In a similar vein, Dye (1984) also criticized single equation models because they made it impossible to discern the effect of determinants (e.g., liberal party strength) from the effects of facilitators (e.g., inter-party competition). More recently, Barrilleaux, Holbrook, and Langer (2002) added an interaction term into the single equation model to find the effect of partisan balance that is conditional on electoral competition. We propose to use heteroscedastic normal regression (Franklin 1992) in order to examine both direct effects of determinants and facilitating effects of mediators. The main advantage of using this model in the analysis of state policymaking is that we can utilize two separate equations, one for the determinant effect on the directional change (level) in state policy outcomes and one for the facilitating effects on the consistency. The basic structure of this model is Yit ~ N ( µ it , σ 2 ), it
µ it = X it β , σ 2 = exp( Zit r ), it

where i = state, and t = year. The most noticeable differences of this model from other

9

single equation models is that not only the mean but also the variance of the distribution of Yit can be estimated by two different sets of variables, Xit and Zit. The former is a set of variables that determine the level of the state’s welfare efforts and the latter is another set of variables that determine the variance of the predicted policy outcomes. In other words, Xit includes determinants such as Democratic Party strength and state elite ideology while Zit includes mediators such as electoral competition, legislative professionalism, and citizens’ ideological homogeneity. In general, we expect factors that are thought to increase the consistency should decrease the variance. We can estimate parameters, β (direct effects) and γ (facilitating effects) through the following likelihood function: L = ∏∏
i =1 t =1 N T

 1  (Y − X it β ) 2 exp −  it  2π exp( Z it γ )  2  exp( Z it γ )  1

  .   

Measurement
Dependent variable. We use state welfare efforts to examine policy liberalism. Welfare efforts are measured using per capita indigenous state expenditures for AFDC, general assistance, and income maintenance in 82-84 constant dollars. 1 Determinants. Democratic Party strength is measured by the percent of Democratic seats in the state legislature in the previous year. We expect this variable to have a positive impact on the level of welfare effort. Elite ideology is measured using the index score of Berry, Ringquist, Fording, and Hanson (1998). We expect welfare effort to be positively related to elite liberalism in the previous year. In addition, we control for other variables that have been considered in the past research. Per capita income is expected to have a positive effect on the level of welfare efforts since, without the fiscal
1

We use state welfare effort data collected by Barrilleaux, Holbrook, and Langer (2002). We would like to thank them for sharing their data with us.

10

capability, a state government cannot provide generous welfare benefits even if the state is willing to do so (Tweedie 1994). Unemployment rate is expected to increase welfare efforts since bad economic situations increase the level of demand for welfare benefits, and state governments will show a degree of responsiveness to the increase in social demand. National government transfers to state welfare programs may help state provide more generous welfare benefits. Neighbor states’ welfare expenditures can be thought to have either positive effects following the diffusion effects between neighbor states, which have been identified in the policy innovation literature (Berry and Berry 1999) or negative effects following the logic of race to the bottom (Peterson and Rom 1990). We expect a positive effect mainly based on findings of Barrilleaux, Holbrook, and Langer (2002), which showed that increases in welfare efforts among a state’s neighboring states led to corresponding increases in the state. South and year are also controlled for. Mediators. Electoral competition is measured by a modified Holbrook-Van Dunk index. We expect that electoral competition will decrease the variance, thereby enhancing the consistency of welfare policy. Legislative professionalism is measured by per legislator operating budget in constant 82-84 dollars in the previous year. We expect a positive impact on the variance. Citizens’ ideological homogeneity is measured using a folded state citizen ideology (Berry et al 1998) scores so that extreme liberal or conservative can take greater values. This variable is expected to have a negative effect on the variance. We also control for per capita income. Since wealthier states may have a wider range of choices in terms of the level of welfare efforts while poorer states may be restrained to a lower level of welfare benefits due to their limited fiscal capability, we

11

expect per capita income will increase the variance. South and year dummy variables are also included.

Findings
(Table 1 about here) Table 1 reports the empirical results. First, consider the effects of Democratic Party strength and electoral competition. We proposed that while Democratic Party strength determines the directional change in states’ welfare efforts, electoral competition should facilitate the translation of partisan strength into policy outcomes. Our analysis shows that there is no significant linear effect of Democratic Party strength on states’ welfare efforts. We also estimated a model with a squared term to check for nonlinear effects by Democratic Party strength but failed to find statistically significant results. 2 We expected that a higher degree of electoral competition would lead state legislators to take clearer partisan positions when they make policy decisions. This hypothesis was supported by our analysis. Electoral competition turns out to have a negative impact on the variance, which indicates that this variable increases the consistency of the state policymaking process. Second, we proposed that legislative professionalism would decrease the consistency of state welfare policymaking since legislative professionalization may create institutional environments wherein state legislators can make individualistic policy decisions that may decrease their conformity to constituents’ partisan interests. This hypothesis was also confirmed by the empirical results. Legislative professionalism does have a positive effect on the variance.
2

The estimated marginal effect of Democratic strength was 0.16 -0.002*Democratic Party strength as shown in Appendix A. The likelihood ratio test shows that the estimated effect is not significant (chisquare(2) =3.08 and p = 0.2145).

12

Third, we expected that state elite ideology would determine the partisan leaning of policy outcomes while citizens’ ideological homogeneity would reinforce this relationship. State elite liberalism has a significant positive effect on state welfare efforts in our analysis. The effect of ideological homogeneity of citizens on the consistency was only marginally significant in two-tailed test at .1 level. But, since we expected a negative effect on the variance, we conducted one-tailed test and found a significant effect at .05 level. Lastly, there are some other factors that affected the state welfare policymaking process. The national government’s financial supports led state governments to provide more generous welfare benefits. Neighboring states’ welfare expenditures has a positive effect on state welfare efforts, but the effect is marginally significant at .1 level in onetailed test. Bad economic situations such as unemployment, which may increase the demand for greater welfare benefits, also increased state welfare efforts. Per capita income did not exert significant influence on the directional change of welfare policy outcomes, but it decreased the consistency of state welfare policymaking.

Conclusion
At various times over the years, scholars have discussed the idea that all political process variables are not created equal when it comes to their role in policy making process. The idea that some political variables serve as mediators and others as determinants is not a new one. The real question is why this way of thinking about political process variables has not been more prevalent in state policy making research. We believe that there are two reasons for this.

13

First, scholars have approached this process in a piecemeal fashion that is driven by their individual research interests. For instance, it is only natural that Erikson, Wright, and McIver’s research focuses on the role of ideology given their research interest in public opinion. The same can be said of Barrilleaux and his interest in political parties. We hope this paper will serve as a theoretical umbrella under which these individual theories can be unified in ma nner that will result in a greater contribution to future research. Second, the statistical tools needed to approach the problem correctly have not always been available. Heteroscedastic normal regression was not introduced into political science literature until 1992 (Franklin 1992), and we believe that this is its first application to state policymaking. We consider this model to be the most appropriate statistical tool for our research and for others such as Barrilleaux, Holbrook, and Langer’s (2002) most recent work. This model allows us to think about states’ and their policy outcomes in a more comprehensive manner. Take California and Michigan as examples. Each state is located in nearly the same position in the welfare policy space (estimated

µ’s are 158 and 180 respectively) but their level of consistency varies dramatically
(estimated σ ’s are 177 and 63 respectively). California and Michigan have provided welfare benefits at a similar level, but the latter showed a greater level of consistency than the former. The heteroscedastic normal regression model allows us to empirically test hypotheses about which political variables contribute to position and to consistency. The most obvious direction for future research is to test this theory using other dependent variables. Welfare policy is certainly one of the most widely used vehicles for testing theories of the state policy making process. However, Jacoby and Schneider

14

(2001) maintain that there is more to state policy than welfare and that state program expenditures as a whole can be used to create a measure of state policy priorities. The next step is to test this theory of mediators and determinants using this dependent variable and others. What conclusions can be drawn from this research? Godwin and Shepard (1976) were on the right track when they argued that variables should be classified into mediators and determinants. However, we believe that our research suggests that they were incorrect to lump all political processes into the mediator camp. Confirming our expectations, electoral competition, professionalism, and citizen ideology function as mediators in that they affect the consistency with which states take their positions in the policy space. Elite ideology, on the other hand, is confirmed as a determinant by its role in fixing a state’s position in that same policy space.

15

Figure 1. State Policymaking Dynamics: Location and Consistency

σC

σ A=B

Conservative µ A Policy Liberalism Space

µB=C

Liberal

Three Hypothetical States State State A State B State C Policy Liberalism Consistency µA µB=C µB=C

σ A= B σ A= B
σC

16

Table 1. Welfare Efforts in the American States: 1973-1992 Variable (1) Direction Equation Democratic party strength Elite liberalism National transfers to state Neighbors' welfare spending Per capita personal income Unemployment South Lagged welfare efforts Constant Variance Equation Electoral competition Legislative professionalism Citizen liberalism South Constant Number of observations Wald chi-square
(1) (2)

Coefficient

Std. Error

Prob. z Hypothesis Two-tail One-tail

-0.003 0.019 0.027 0.007 0.00006 0.154 0.570 1.069 -1.008 -0.011 0.001 -0.009 0.00014 -1.261 3.076 1000 53512

0.016 0.010 0.016 0.005 0.00012 0.057 0.483 0.008 1.938 0.004 0.000 0.006 0.00003 0.191 0.456

0.84 0.05 0.08 0.17 0.64 0.01 0.24 0.00 0.60 0.01 0.01 0.10 0.00 0.00 0.00

0.42 0.02 0.04 0.08 0.32 0.00

+ + + + + +

0.00 0.01 0.05 0.00

Per capita personal income

+ +

p<.01

Dummy variables for years (1974-1992) were included in both equations but not reported. (2) Folded citizen ideology score = |citizen ideology score – 50|.

17

Appendix A. Variable (1) Direction Equation Democratic party strength Democratic party strength Elite liberalism National transfers to state Neighbors’ welfare spending Per capita personal income Unemployment South Lagged welfare efforts Constant Variance Equation Electoral competition Legislative professionalism Citizen ideology South Constant Number of observations Wald chi-square
(1) (2)

Coefficient β, γ

Std. Error

Prob. z Hypothesis Two-tail One-tail 0.09 0.08 0.15 0.04 0.14 0.93 0.01 0.38 0.00 0.10 0.01 0.01 0.14 0.00 0.00 0.00 0.01 0.01 0.07 0.00 0.07 0.02 0.07 0.46 0.00 + + + + +

0.155 -0.001 0.015 0.032 0.007 0.00001 0.149 0.432 1.069 -5.044 -0.011 0.0010 -0.008 0.0001 -1.272 3.040 1000 53797

0.092 0.0007 0.010 0.016 0.005 0.0001 0.057 0.488 0.008 3.027 0.004 0.0004 0.006 0.00003 0.191 0.458

2

Per capita personal income

+ +

p<.01

Dummy variables for years (1974-1992) were included in both equations but not reported. (2) Folded citizen ideology score = |citizen ideology score – 50|.

18

Appendix B: Descriptive Statistics Variable Welfare efforts Lagged welfare efforts Democratic party strength Democratic party strength2 Electoral competition Legislative Professionalism Elite ideology Citizen ideology (Folded) National transfers to state Neighbors’ welfare spending Per capita income Unemployment South Obs 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 Mean 81.12 70.81 58.55 3602.27 50.36 116.95 50.85 13.58 41.40 67.82 12029.61 8.03 0.22 Std. Dev. 43.01 37.69 13.19 1616.07 17.79 174.39 21.52 9.70 14.91 33.75 2035.13 3.25 N/A Min 21.12 18.74 25.89 670.32 3.63 2.41 0.00 0.00 4.02 10.78 7511.96 0.25 0.00 Max 237.67 206.45 94.39 8909.10 87.90 1860.12 95.04 43.91 150.91 254.44 19796.77 34.31 1.00

19

References
Alt, James, and Robert Lowry. 1994. “Divided Government, Fiscal Institutions, and Budget Deficits: Evidence from the States.” American Political Science Review 88: 811-828. Barrilleaux, Charles, Thomas Holbrook, and Laura Langer. 2002. “Electoral Competition, Legislative Balance, and American State Welfare Policy.” American Journal of Political Science 46: 413-427. Barrilleaux, Charles. 1997. “A Test of the Independent Influences of Electoral Competition and Party Strength in a Model of State Policy-Making.” American Journal of Political Science 41: 1462-1466. Berry, Frances S., and William D. Berry. 1990. “State Lottery Adoptions as Policy Innovations: An Event History Analysis.” American Political Science Review 84: 395-416. Berry, William D., Evan J. Ringquist, Richard Fording, and Russell Hanson. 1998. "Measuring Citizen and Government Ideology in the American States, 1960-93." American Journal of Political Science 42:327-348. Berry, William D., Michael B. Berkman, and Stuart Schneiderman. 2000. "Legislative Professionalism and Incumbent Reelection," American Political Science Review 94: 859-874. Berry, William, Richard Fording and Russell Hanson. 2003. “Reassessing the ‘Race to the Bottom’ in State Welfare Policy: Resolving the Conflict Between IndividualLevel and Aggregate Research.” Journal of Politics 65: 327-349. Brown, Robert D. 1995. “Party Cleavages and Welfare Effort in the American States.” American Political Science Review 89: 23-33. Carey, John M., Richard G. Niemi, and Lynda W. Powell. 2000. Term Limits in State Legislatures. Ann Arbor: University of Michigan Press.

20

Carmines, Edward G. 1974. “The Mediating Influence of State Legislatures on the Linkage Between Interparty Competition and Welfare Policies.” American Political Science Review 68: 1118-1124. Cnudde, Charles F., and Donald J. McCrone. 1969. “Party Competition and Welfare Policies in the American States.” American Political Science Review 63: 858-866. Dawson, Richard E., and James A. Robinson. 1963. “Interparty Competition, Economic Variables, and Welfare Policy in the American States.” Journal of Politics 25: 265-289. Dye, Thomas R. 1984. “Party and Policy in the States.” Journal of Politics 46: 10971116. Dye, Thomas R. 1966. Politics, Economics, and the Public. Chicago: Rand McNally. Franklin, Charles H. 1992. “Eschewing Obfuscation? Campaigns and the Perception of Senate Incumbents.” American Political Science Review 85: 1193-1214. Fry, Brian B., and Richard Winters. 1970. “The Politics of Redistribution.” American Political Science Review 64: 508-522. Garand, James C. 1985. “Partisan Change and Shifting Expenditure Priorities in the American States.” American Politics Quarterly 13: 355-392. Godwin, R. Kenneth, and W. Bruce Shepard. 1976. “Political Processes and Public Expenditures: A Re-examination Based on Theories of Representative Government.” American Political Science Review 70: 1127-1135. Hanson, Russell. 1983. “The ‘Content’ of Welfare Policy: The States’ Aid to Families with Dependent Children.” Journal of Politics 45: 771-783. Hibbing, John R. 1999. "Legislative Careers: Why and How We Study Them." Legislative Studies Quarterly 24: 149-71. Hill, Kim Quaile, and Jan Leighley. 1992. “The Policy Consequences of Class Bias in State Electorates.” American Journal of Political Science 36: 351-366.

21

Hill, Kim Quaile, Jan E. Leighley, and Angela Hinton-Anderson. 1995. “Lower class Mobilization and Policy Linkage in the U.S. States.” American Journal of Political Science 39: 75-86 Hofferbert, Richard. 1966. “The Relation between Public Policy and Some Structural and Environmental Variables.” American Political Science Review 60: 73-82. Jennings, Edward T. 1979. “Competition, Constituencies, and Welfare Policies in American States.” American Political Science Review 73: 414-429. Jocoby, William G., and Saundra K. Schneider. 2001. “Variability in State Policy Priorities: An Empirical Analysis.” Journal of Politics 63: 544-568. Key, V. O., Jr. 1949. Southern Politics in State and Nation. New York: Knopf. Moncrief, Gary F., and Joel Thompson. 1992. Changing Patterns in State Legislative Careers. Ann Arbor: University of Michigan Press. Mooney, Christopher. 1994. “Measuring U.S. State Legislative Professionalism: An Evaluation of Five Indices.” State and Local Government Review 26: 70-78. Peterson, Paul, and Mark Rom. 1990. Welfare Magnets. Washington, D.C.: Brookings Institution. Plotnick, Robert D., and Richard Winters. 1985. “A Politico- Economic Theory of Income Redistribution.” American Political Science Review 79: 458-473. Rosenthal, Alan. 1998. The Decline of Representative Democracy: Process, Participation and Power in State Legislatures. Washington, DC: Congressional Quarterly Press Squire, Peverill. 1992. “The Theory of Legislative Institutionalization and the California Assembly. ” Journal of Politics 54: 1026-1054. Tweedie, Jack. 1994. “Resources Rather than Needs: A State-Centered Model of Welfare Policy-Making.” American Journal of Political Science 38: 651-672. Weber, Ronald E., and Paul Brace. 1999. American State and Local Politics: Directions for the 21st Century. New York: Chatham House Publishers.

22

Winters, Richard. 1976. “Party Control and Policy Change.” American Journal of Political Science 20: 597-636. Wright, Gerald, Robert Erikson, and John McIver. 1987. “Public Opinion and Policy Liberalism in the American States.” American Journal of Political Science 30: 980-1001.

23


								
To top