Sub-constituencies and Legislative Responsiveness: Evidence from the States

Abstract

This article tests whether theories of congressional behavior that link legislative responsiveness to the preferences of sub-constituencies at the expense of party preferences apply to the state level. Using ten years of state-level data and roll-call data from nearly 4,000 individual votes on E-Verify legislation, I examine the competing influences of party and constituency preferences on legislative behavior. The results confirm that state legislatures/legislators are responsive to sub-constituencies, but find that responsiveness plays out in different ways depending on the level of analysis and the political party and constituents in question. These results have important implications for our understanding of legislative representation: because responsiveness to sub-constituencies can yield policy results that are antithetical to stated party goals, what appears to be collective irresponsibility from a party may actually be individual legislators striving to be responsive to those constituents that they anticipate will hold them accountable.

Keywords

responsiveness state legislatures representation E-Verify

Early studies of public influence on state legislative behavior painted a dismal picture of responsiveness. Not only did there appear to be no link between public opinion and policy output (Dye 1961; Plotnick and Winters 1985), the accepted rationale for this disconnect was that voters were uninformed and apathetic (Treadway 1985). Recent research, however, is more optimistic. After Erikson, Wright, and McIver (1993) demonstrated that states do adopt policy regimes that align with the ideology of their citizens, research has repeatedly shown a strong relationship between public opinion and state policy on salient issues (Brace et al. 2002; Lax and Phillips 2009; Lupia et al. 2010), but not in less salient policy areas (Lax and Phillips 2009, 2012). Most conclude that without a strong position from the general public, legislators shirk constituency preferences and succumb to party-based influences (Jenkins 2010; Kirkland and Harden 2016; Lax and Phillips 2012; Patterson 1996) or overestimate voter positioning on the issue and enact policies that are actually more extreme than public preferences (Lax and Phillips 2012). While both of these accounts are reasonable, I contend that this body of work overlooks an alternative explanation well-established in theories of congressional behavior (Bishin 2000; Clinton 2006; Fenno 1978; Hayes and Bishin 2012; Miller and Stokes 1963): when faced with less salient policy issues, legislators respond to the preferences of sub-constituencies with a vested interest in that policy. Accordingly, this research sets out to test the applicability of congressional-level theories, which link constituency preferences to legislative behavior, at the state level. As the federalist system places state governments in a prime position to tailor their policies to the preferences of their constituents, exploring the relationship between state legislative responsiveness and sub-constituencies is a natural and overdue line of inquiry.

To test the relationship between sub-constituency preferences and legislative responsiveness, I examine variation in state-level adoption of E-Verify laws from 2006 to 2015 and legislator roll-call voting on this measure. E-Verify, a system that mandates employers electronically verify the legal status of prospective employees using federal databases, is an apt case for showing the essential role that sub-constituencies can play in the state legislative process. While E-Verify has received little attention from the general public, putting it neatly into the category of low-profile legislation, it is especially salient to sub-constituencies who are directly impacted by its enforcement: farm owners, Latinos, and Asians. In addition, because Democrats lack a formal position on E-Verify and the official Republican position is distinct from that of the interested sub-constituencies, E-Verify allows us to clearly differentiate between the relative effects of sub-constituency and party-based pressure on legislators.

In short, the findings partially confirm and extend to the state-level early work by Fenno (1978) and Miller and Stokes (1963). State legislatures/legislators are responsive to sub-constituencies, but responsiveness plays out in different ways depending on the level of analysis and the political party. At the state level, E-Verify is kept off the agenda in Republican legislatures where pressure from farm owners is strongest while such legislation passes in states where the electoral power of this sub-constituency is considerably weaker. Yet, even among those states that do pass E-Verify, there is strong evidence to suggest that individual Republican legislators are willing to split with party ranks to protect the interests of their farmer constituents. Interestingly, the presence of a large Latino/Asian population has no statistically significant effect on the probability of a state adopting E-Verify even if that state’s legislature is controlled by the Democratic Party. That said, larger Latino/Asian communities do appear to diminish the probability of an individual Democrat voting “yea” on E-Verify, though the robustness of this effect is called into question as the roll-call margin narrows. Given recent findings by Rogers (2017), which suggest that state legislators are not held accountable for their actions and, thus, need not worry about their constituents’ preferences, these findings suggest a more nuanced and optimistic picture of state politics: parties may be viewed as collectively irresponsible, but only because individual legislators are responsive to certain constituencies—a responsiveness that likely occurs precisely because they anticipate that particular groups will hold them accountable.

Legislative Responsiveness

Some level of legislative responsiveness is key to achieving a democracy. Indeed, political scientists have long found that members of Congress are incentivized to represent their districts’ preferences or face grave electoral consequences (Canes-Wrone, Brady, and Cogan 2002; Jacobson 1993; Mayhew 1974; Nyhan et al. 2012). Yet, the connection between state legislative behavior and constituency preferences has been more tenuous. It was not until the early 1990s, when Erikson, Wright, and McIver (1993) demonstrated that state policies reflect the general ideology of their citizens, that a connection between public preferences and state policy direction was firmly established. Since then, updated models have added the role of organized interests, but still find that public opinion is the strongest predictor of policy direction (Gray et al. 2004). Jacoby and Schneider (2001) further show that public opinion not only influences policy direction but also state policy priorities. Other studies confirm a direct link between public attitudes and the adoption of specific policies, such as gay rights, the death penalty, and abortion policy (Arceneaux 2002; Brace et al. 2002; Gerber 1996; Haider-Markel and Kaufman 2006; Lax and Phillips 2009; Norrander 2000).

Although research has taken significant steps in terms of extending Erickson, Wright, and McIver’s findings to specific attitudes and subsequent policy adoption, these improvements are not without their own limitations. Thus far any connection between public opinion and state policy has been limited to those issues that are especially salient to the general public. For example, Gerber (1996) finds that members of California’s legislature voted in accordance with their district when legislation was highly salient, while those representing districts where the same issues were deemed insignificant to constituents tended to vote against their majority’s preferences. Similarly, Lax and Phillips (2009, 370) conclude simply and emphatically that “higher salience means greater responsiveness.”

But if legislators are acting without regard for public opinion on less visible issues, then what is guiding their behavior? Without taking up this question explicitly, most state-level studies assume that as issue salience diminishes, party pressure takes over to drive legislative decision making (Jenkins 2010; Kirkland and Harden 2016; Lax and Phillips 2012; Patterson 1996). The Downsian logic behind this assumption is clear: if the preferences of the median voter on a particular issue are undefined, then a legislator’s actions are not threatened by the ballot box and they should do their best to curry favor with party leadership (Carson et al. 2010; Cox and McCubbins 2005; Jenkins 2008). Yet, this supposed constituency-party trade-off is rarely straightforward. It could be that what is interpreted as party pressure is actually legislators acting on their personal preferences (Krehbiel 1993) or party pressure masking “an underlying constituency influence” (Kingdon 1968, 6). Lax and Phillips (2012) speculate that any disconnect between state legislative behavior and aggregate preferences might be due to misjudgment of majority preferences or because legislators are appealing to an ideological base rather than the median voter, but they never test this possibility explicitly.

A second problem with the reasoning of existing literature is that it assumes accountability is linked to salience. While studies of Congress demonstrate that legislators can profit or perish from extreme or incongruous roll-call behavior (Canes-Wrone, Brady, and Cogan 2002), evidence of electoral accountability at the state level is more contested (Birkhead 2015; Hogan 2004; Lowry, Alt, and Ferree 1998; Rogers 2017). In the most extensive study to date, Rogers’s (2017) finds a poor connection between roll-call votes and electoral accountability. Yet, like those before him, Rogers’s conclusion is specific to voters collectively failing to hold state legislators accountable for roll-call votes that do not reflect majority opinion. At no point does the author parcel out whether those with a highly vested interest in that policy issue in fact supported their representative’s roll-call vote or their reelection campaign.

At the congressional level, on the contrary, legislators are found to be accountable to sub-constituencies with strong preferences on niche issues (Bishin 2009). According to Fenno (1978), congressional districts are made up of concentric circles of constituents, each with a distinct role to play in helping representatives achieve their goals. Arnold (1990) stresses that knowing the policy preferences and opinion intensity of these groups is crucial to avoid alienating sub-constituencies on key issues, thereby safeguarding reelection prospects. Indeed, the preferences of sub-constituencies have proven to be more reliable at predicting legislative behavior than estimates of average district preferences at the national level (Clinton 2006; Fenno 1978; Goff and Grier 1993; Miller and Stokes 1963) and levels of policy responsiveness and congruence are much stronger when we look at the relationship between legislators and sub-constituencies than legislators and their geographic constituencies (Bishin 2009; Bullock and Brady 1983). Could it be that taking into account sub-constituency preferences will also help us to paint a clearer picture of responsiveness at the state level?

Building from the foundation laid by studies of congressional behavior, I examine whether state-level studies underestimate legislative responsiveness to constituents and overestimate legislator commitment to party. I expect that we will find that the relationship between constituent preferences and state legislative behavior is analogous to the relationship found at the national level. Specifically, we should find that less salient legislation will generate responsiveness from state legislatures/legislators, but that responsiveness will be directed at the preferences of those sub-constituencies with a vested interest in that issue. Of course, we would not expect all interested sub-constituencies to be treated equally by the parties. As partisan polarization has grown among elites, so too has it among the voting public. Thus, if electoral accountability is what drives legislators to be responsive, there is no reason to think that they would go out of their way to cater to groups unlikely to support them in the next election. This means that Democrats and Republicans should be especially sensitive to the preferences of those sub-constituencies that are existing members of their reelection constituencies.

E-Verify in the States

E-Verify is a fitting case for showing the important role that sub-constituencies can play in the state legislative process for two reasons. First, it is a policy issue that lacks general public salience, but is highly significant to those constituents directly affected by its implementation. National opinion polls reveal that public preferences on E-Verify are weak and contingent upon question phrasing (Ekins 2013). Perhaps more importantly, a review of public opinion polls shows that, on average, 10 to 20 percent of respondents simply “don’t know” where they stand on the issue, signifying the lack of public interest and/or knowledge about an employee verification system. Zingher (2014, 107) suggests that because E-Verify is not as “high profile” as other immigrant-related policies, it yields weaker preferences and is less politicized. Yet, even if E-Verify triggers weaker public preferences than other immigration enforcement efforts in general, it is surely important to particular groups of constituents who are directly impacted by its enforcement: specifically, farm owners, Latinos, and Asians.

Opposition to E-Verify from farm owners is motivated by their labor demands, which already suffer under a dysfunctional H2-A temporary worker visa program. Cathleen Enright, Vice President of Governmental Affairs at the Western Growers Association, calls the H2-A program “crushing” because of its inability to supply farm owners with access to a constant and dependable workforce (Enright, as quoted in Caldwell 2011: paragraph 10). The dairy industry, for example, requires a skilled year-round workforce that cannot be supplied by H2-A seasonal workers. Unable to find American workers who are willing to do the job, dairy farmers rely almost entirely on unauthorized immigrants (Wickham 2011).

In light of this, the President of the National Council of Farmer Cooperatives, Chuck Conner, has called for any E-Verify legislation to include an improved H2-A Visa program and a process to adjust the status of existing workers (Conner 2011). This two-pronged approach has become the mantra across agriculture.¹ Indeed, the American Farm Bureau Federation, the nation’s largest farmers’ organization, claims that unless E-Verify also includes a worker program, its implementation will “cripple agriculture production in America” (Farm Bureau 2017).

The Hispanic Federation and the Asian American Justice Center echo agriculture’s position. However, in this case, the concern is that E-Verify will disproportionately affect work-eligible immigrants, legal residents, and naturalized as well as U.S.-born citizens. Both groups cite a Department of Homeland Security-commissioned study (U.S. Congress House Committee on the Judiciary 2013) that found employers pre-screen employees to weed out those who may be more likely to receive a non-confirmation notice and that error rates were twenty times higher for foreign-born workers. Thus, it is not surprising that nearly 72 percent of foreign-born and 35 percent of U.S.-born Latinos report being worried about themselves, a family member, or close friend being deported (Lopez and Minushkin 2008). Similarly, with 50 percent of the Asian American and Pacific Islander community speaking English less than very well, the Asian American Justice Center is particularly concerned with the high rate of misidentifications that are likely to affect citizen and naturalized residents. In 2011, the Latino community undertook an extensive public campaign opposing E-Verify (Khan 2011). Mike Garcia (as quoted in Kahn 2011: paragraph 7), President of Service Employees International Union, insisted that elected leaders would be held accountable by the Latino community: “Let there be no mistake: Latino constituents understand the devastation of this policy. Janitors, security guards, farm workers and others have all seen it in their communities and they refuse to standby.”

If election-minded legislators are responsive to sub-constituencies, then we should find that legislatures/legislators that represent larger farming and Latino/Asian communities are less prone to adopt/favor E-Verify. Yet, as argued above, we are unlikely to see the parties respond to these groups in the same manner. Farmers, for example, support the Republican Party by a margin of 3:1 (Agri-Pulse Communications 2016), while Asians and Latinos support the Democratic Party by a margin of at least 2:1 (Pew Research Center 2015). Thus, despite their shared position on E-Verify, we would expect that Republican and Democratic opposition to E-Verify will be most likely to occur when the farm owner and Latino/Asian populations, respectively, make up the largest shares of their districts.

E-verify is also an appropriate test case because it enables us to differentiate between constituency effects and party pressure (Krehbiel 1993). To date, all but four states have introduced E-Verify legislation, but only twenty-three states adopted policies. Importantly, variation in adoption has not been tied to party or ideology (Newman et al. 2012; Zingher 2014). This is in stark contrast to other state immigration policies, which were established under distinctively partisan circumstances (Creek and Yoder 2012; Monogan 2013; Wong 2012). Although Democrats, on average, have expressed greater opposition to E-Verify, the official party position is unclear. The 2012 Democratic Party Platform (46) expressed a commitment to “hold employers accountable for whom they hire,” but even this vague language was absent from the 2016 platform. Without looming party pressure, Democratic legislators should be especially reactive to the preferences of Latinos and Asian constituents.

The Republican Party, by contrast, has publically expressed its support of E-Verify (Lind, Rankin, and Harris 2016) as a key component of immigration enforcement in the national platform and in thirty-six state party platforms. Yet, there is evidence to suggest that some Republicans are more reluctant to toe the party-line because of what E-Verify would mean for farmers. In North Carolina, for example, Republican legislators representing large agricultural districts helped to override Governor McCrory’s veto of legislation exempting agricultural businesses from E-Verify (Wilson 2013). These lawmakers expressed concern that farmers who have grown to depend on immigrant labor would be irrevocably harmed by E-Verify. In short, Republican legislators are cross-pressured when it comes to E-Verify and must decide if they will continue the refrain of the Republican Party or guard the interests of their farmer constituents.

Data and Method

The first part of the analysis examines what factors increase the probability of a state adopting an E-Verify law between 2006 and 2015 using a discrete event history analysis (DEHA), where the dependent variable is coded 1 if a state adopted E-Verify and 0 otherwise. The DEHA model predicts the occurrence of a particular event by comparing the characteristics of those population members for which the event did occur in a particular time frame with those for which it did not occur in the same time frame (Allison 1982). In this case, our population is the fifty U.S. states, the event is passing an E-Verify law, and the time frame is annual units from 2006 to 2015. States are coded 1 if they pass any E-Verify policy during one of these time frames. As several states have implemented more than one E-Verify policy (gradually widening the scope of employers affected), a state remains in the dataset as long as it is “at-risk” of passing an E-verify law.² Once all employers are covered by a state’s E-Verify policy, that state has no probability of passing an additional E-Verify law and exits the dataset. So, for example, Arizona is coded 0 in 2006, 1 in 2007, and 1 again in 2008, but then exits the dataset as there are no other employers for which it could apply an E-Verify requirement.

The second part of the study evaluates the motivations behind a legislator voting for or against E-Verify using state fixed-effects logistic regression models. The dependent variable is coded 1 if the legislator voted in favor of the E-Verify bill and 0 if they voted against it (legislators who did not vote were excluded from the analysis). This results in twenty-eight bills and nearly 4,000 votes across nineteen state legislatures. It should be noted that the votes analyzed are specific to E-Verify legislation that passed. Ideally, we would also be able to examine roll-call records for states that proposed, but did not pass E-Verify. However, E-Verify has yet to fail due to “no” votes. Instead, its fate has been repeatedly sealed during committee review. Some may worry that only analyzing roll-calls in states where E-Verify passed biases the results in favor of observing more support for E-Verify and potentially more partisan polarization. Yet, while E-Verify may be adopted by states where farming and Latino/Asian constituencies make up a smaller proportion of the population, the size of these sub-constituencies varies markedly at the district level. Consequently, we should still find constituency effects in those districts with the largest farm owner and Latino/Asian populations. If anything, by only having voting data from states that passed E-Verify, the bar is raised in terms of proving that sub-constituency preferences have an effect on legislative behavior.

To determine what influences legislative behavior on less salient issues, I focus on three key independent variables: partisanship, the ideology of the aggregate constituency, and the size of the farm owner and Latino/Asian constituencies. If existing theories linking the motivations behind legislative behavior to issue salience are correct, then partisanship should strongly determine support/opposition to E-Verify. At the state level, partisanship is represented by two dummy variables: Democratic-controlled (43%) and Republican-controlled (44%) legislatures (split legislatures are the omitted baseline category). In addition to partisanship, I include a dummy variable indicating whether the Republican Party’s state platform explicitly supports E-Verify to account for the possibility that the Republican Party may push E-Verify at the national level, but not at the state level. The party affiliation of individual legislators, coded 1 for Republicans and 0 for Democrats,³ was determined from legislator biographical information. Approximately 58 percent of the legislators identify as Republicans and 42 percent identify as Democrats.

Spatial theories predict that legislators are especially sensitive to the preferences of the median voter. While less salient issues offer limited support for this contention, it could be that because E-Verify is part of a policy area that generates strong public reactions—immigration enforcement—we would find legislators react to E-Verify in a way that conforms to aggregate opinion on this broader issue. For example, a Republican legislator may worry about looking “soft” on immigration and losing constituent support to more traditional Republicans in the primaries if they vote against E-Verify. If this is the case, then legislators should be more likely to vote along district ideological lines regardless of the size or intensity of sub-constituency preferences. To account for this, I use Tausanovitch and Warshaw’s (2013) state- and district-level measures of the average level of constituent conservatism/liberalism. Smaller values indicate a more liberal citizenry and larger values more conservative.

To test the theory of sub-constituency pressure, I use the percentage of farm owners and immigrant-derived (Latino and Asian) populations within the state and within each state legislative district. For farm owners, the estimates were obtained from the 2007 and 2012 U.S. Department of Agriculture (USDA) Census of Agriculture. The Census includes farm owner information at the national, state, and county level. To obtain estimates for the district level, I used the U.S. Census State Legislative Districts by Counties relationship files to merge the county-level data with the appropriate legislative district. As districts often serve several partial counties, I divided each county’s total number of farm owners by the number of districts that each county spans. Next, I combined the farm owner sub-totals for each partial county within a district to arrive at a district total. This total was divided by the district’s population to arrive at the percentage of farm owners for each legislative district. While it is certainly true that not every county’s population is divided equally between the various legislative districts that represent it, this should be an adequate approximation of agricultural interests at the district level. Overall, the size of the farm owner constituency ranges from less than 1 to 7 percent at the state level and from less than 1 to 9 percent at the district level. The measure for the Latino and Asian populations combine estimates from the American Community Survey (ACS). This pooled constituency is relatively larger than the agricultural constituency, reaching as much as 53 percent of a state’s population and as much as 95 percent at the district level.

Although I contend that we will find a strong relationship between legislative behavior and the size of the farm owning constituency, it could be that what looks to be a relationship between Republicans and farm owners is actually pressure from agribusiness. After all, agribusiness consistently ranks in the top 10 of the Republican Party’s most generous and reliable contributors. For example, in the 2010, 2012, and 2014 election cycles, agribusiness contributed more than $2 billion to political campaigns and party committees—about 70 percent of which went to the Republican Party and its candidates (Center for Responsive Politics 2016). Still, evidence of a direct link between agriculture contributions and voting is weak at best (Alvarez 2005; Vesenka 1989), whereas research demonstrates a strong relationship between farming constituency pressures and legislative behavior (Bellemare and Carnes 2015). As a result, it is more likely that the size of the farmer community rather than agribusiness contributions will foretell how Republicans respond to E-Verify. Nevertheless, using data from the Money in State Politics database, I include the amount of campaign dollars given by agribusiness to the parties (at the state level) and to the individual representatives (at the district level) for the election year preceding their vote on the E-Verify bill.

To account for the possibility that ideology acts independently of legislator partisanship (Jenkins 2006), I use Shor and McCarty’s (2011) updated ideological scores for state legislatures and legislators. These scores are derived from a combination of legislator responses to the Project Vote Smart National Political Awareness Test and fifteen years of roll-call voting data. Smaller values indicate a more liberal legislator and larger values more conservative.

Latino and female legislators tend to be associated with more liberal voting patterns as well as a proclivity for supporting legislation that advances immigrant interests (Bratton 2006; Bratton and Haynie 1999). To account for this, the percentage of state legislators that are women or Latino were identified using data from the Center for American Women and State Politics and the National Directory of Latino Elected Officials, respectively. At the individual level, each legislator’s gender and ethnicity were identified from legislator biographical information. Consistent with extant studies on state-level immigrant-related policies, I include the population size, the median household income, the unemployment rate, and the percentage of residents that obtained a bachelor’s degree or higher at the state and district level.

Finally, I included measures for candidate incumbency and electoral competitiveness at the legislator level. If, as Bernstein (1989, 100) claims, incumbent legislators can “generally afford to vote for what they think is right,” then incumbents may be more resistant to sub-constituency pressures. Similarly, legislators running unopposed should be less worried about accountability. To determine incumbency and electoral competitiveness (contested election), I used Klarner et al.’s (2011) updated State Legislative Election Returns data series. Election results post 2010 were acquired via the relevant Secretary of State website allowing the author to calculate incumbency and electoral competiveness for each legislator.

State-Level Results

Existing scholarship on state legislative responsiveness would lead us to believe that because E-verify lacks general public salience, its passage/failure should be explained by legislative partisanship. Yet, neither Zingher (2014) or Newman et al. (2012) find partisanship to play an explanatory role. By including the sub-constituency variables⁴ and expanding the range of E-Verify events through 2015,⁵ this analysis seeks to provide much needed clarity in terms of identifying what is responsible for state-level variation in E-Verify adoption. I hypothesize that if, as Fenno suggests at the congressional level, state legislators view their districts as composed of multiple constituencies, then responsive legislators should yield to pressure from sub-constituencies. Specifically, a legislature’s passage of E-Verify should be contingent on the size of those constituencies most affected by its implementation: farm owners and the Latino/Asian communities.

Table 1⁶ begins our investigation. The first model tests the relationship between legislature partisanship and the likelihood of passing an E-Verify policy (controlling for states that passed a prior E-Verify policy and states whose Republican Party platform explicitly supports E-Verify). Confirming previous studies, there is no statistically discernible difference in the probability of Republican- and Democratic-controlled legislatures passing E-Verify legislation. In model 1.2, I add our primary constituency variables—the state percentage of farm owners and Latino and Asian residents—to test the alternative explanation that salient constituencies—not partisanship—determine which state legislatures pursue E-Verify.⁷ This model depicts a strong relationship between constituency size and policy output when considering the effect of farm owners, which holds in model 1.3 when we control for the aggregate ideology of the state and other factors potentially relevant to that state’s decision to adopt E-Verify.

Table 1.

State Adoption of E-Verify.

	(1.1)	(1.2)	(1.3)
Republican control	1.364 (0.797)	1.449* (0.755)	0.567 (0.769)
Democratic control	−0.021 (0.924)	−0.135 (0.911)	0.506 (0.937)
Farm owners		−0.498** (0.157)	−0.865** (0.298)
Latino/Asian		−0.030 (0.0266)	−0.101 (0.064)
Legislature characteristics
Legislature ideology			0.590 (0.336)
% women			−3.639 (4.506)
% Latino			11.07 (9.357)
Agribusiness contributions (logged)			−0.068 (0.062)
Prior policy	1.367** (0.435)	1.231** (0.805)	0.754 (0.732)
Republican platform supports E-Verify	0.127 (0.510)	0.019 (0.484)	−0.024 (0.027)
State characteristics
Average citizen ideology			0.024 (0.027)
Population (logged)			1.003** (0.381)
Median household income (logged)			−0.811*** (0.243)
Unemployment rate			−0.533 (0.357)
Bachelor’s degree (+)			0.046 (0.189)
Constant	−3.617*** (0.932)	−1.808 (0.992)	−1.125 (4.954)
n	330	330	285
R ²	.142	.187	.277

Models use discrete event history models with state-clustered standard errors and time-period fixed-effects.

p < .05. **p < .01. ***p < .001.

Figure 1 illustrates the marginal effects (from model 1.3) of farm owners on the probability of adopting an E-Verify law. As shown, states with the fewest farm owners have about a 16 percent chance of adopting an E-Verify law. This probability drops to less than 1 percent by the time farm owners make up 5 percent of the state’s population. Although the effect of the Latino/Asian community lacks statistical significance in both models 1.2 and 1.3 (p values of .2 and .12, respectively), the relationship is in the expected direction: the probability of adoption decreases (and our confidence in that probability increases) as the Latino/Asian population increases. Indeed, the probability of adoption drops from 22 percent in states with the smallest Latino/Asian populations to less than 1 percent once that population reaches 40 percent of the state population.

Figure 1.

Predicted probability of state adopting an E-Verify law as size of farm owner constituency increases (with 95% confidence intervals).

The null effect from the Latino/Asian population is likely because models 1.2 and 1.3 fail to take into account our expectations built into the second hypothesis: Republicans and Democrats will only be responsive to those sub-constituencies that are likely to support their reelection efforts. In Table 2, we test this expectation formally. As anticipated, we find a strong and statistically significant effect from farm owners on the probability of a Republican legislature passing E-Verify. A Republican legislature with the fewest farm owners has a nearly 40 percent probability of passing an E-Verify law. This probability drops dramatically to 6 percent when the farm owning constituency reaches just 3 percent of the state’s population. Indeed, despite farm owners making up as much as 7 percent of a state’s population, E-Verify has not yet been implemented in any state where the total percentage of farm owners is greater than 3.1 percent. Interestingly, we also discover that a state Republican Party’s official endorsement of E-Verify is negatively associated with the probability of adoption, implying that at least some Republicans are willing to oppose E-Verify even when the state party’s platform explicitly endorses it. I explicitly test the possibility of such a trade-off between party and constituency pressures in Table 5 below.

Table 2.

Effect of Sub-Constituencies on Republican and Democratic Legislatures.

	(2.1)	(2.2)
	Republican control	Democratic control
Farm owners	−1.564*** (0.386)	−0.379 (0.545)
Latino/Asian	−0.178 (0.120)	−0.041 (0.050)
Legislature characteristics
Legislature ideology	0.973* (0.459)	0.547 (0.783)
% women	4.483 (7.733)	−5.543 (6.349)
% Latino	17.60 (18.85)	9.119 (8.654)
Agribusiness contributions (logged)	−0.149 (0.124)	−0.0298 (0.0492)
Prior policy	0.076 (0.742)	3.513 (2.014)
Republican platform supports E-Verify	−1.933* (0.886)	2.464* (1.200)
State characteristics
Average citizen ideology	0.048 (0.041)	−0.011 (0.053)
Population (logged)	0.755 (0.703)	2.949** (1.082)
Median household income (logged)	−0.536 (0.360)	−1.624 (1.637)
Unemployment rate	−0.401 (0.417)	−2.858** (1.021)
Bachelor’s degree (+)	−0.028 (0.232)	−0.140 (0.288)
Constant	−1.125 (4.954)	−11.609 (23.509)
n	102	112
R ²	.322	.325

Models use discrete event history models with state-clustered standard errors and time-period fixed-effects.

p < .05. **p < .01. ***p < .001.

Surprisingly, model 2.2 does not find a statistically significant effect from the Latino/Asian population on Democratic legislatures. Moving from a state with the smallest proportion of Latinos/Asians to one with the largest decreases the probability of a Democratic legislature adopting E-Verify by 8 percentage points—a meager effect when compared with that of farm owners on Republican legislature adoption rates. Although this result was not predicted, it is not unusual given the disparate levels of political organization and resources between farmers, on one hand, and Latinos/Asians, on the other. Farm owners are an established and fully equipped pressure group—what Converse (1964) would call an issue public—prepared to take political action if their preferences are not upheld. Latino and Asian political interests, by contrast, are informally organized which limits their capacity to strategically and effectively influence legislative behavior (Andersen 2006; Hero 2010). Accordingly, it should not surprise us to find that, as direct stakeholders in E-Verify, the articulated economic concerns of farmers have a more pronounced impact on legislative behavior than Latino and Asian opinion toward E-Verify.

Yet, even if the preferences of Latinos and Asians inconsistently influence legislative behavior when compared with their farmer counterparts, we would still anticipate that legislators representing districts with large Latino/Asian populations would strive to uphold their constituents’ preferences. After all, in contrast to their fragmented political organization, Latino and Asian populations tend to be geographically concentrated—far more so than farmers—and thus should hold sway over the few legislators that represent their districts. If this is the case, then the legislator-level analysis should reveal a significant difference in the roll-call behavior of Democratic legislators representing districts with few versus many Latinos and Asians.

Legislator-Level Results

The state-level analysis offered preliminary evidence that sub-constituency pressure affects whether states adopt E-Verify. Yet, there are lingering questions about the interplay between partisanship and constituency preferences. On one hand, cross-pressured Republicans seem willing to break from their party to appease farm owners, but under what roll-call conditions, do such votes occur? Does the size of the Latino/Asian population have any influence on Democrats or is their behavior better explained by non-constituent forces? Table 3 uses state fixed-effects logistic regression models to control for all time-invariant state characteristics that might be influencing the relationship between our variables of interest and the outcome, a legislator’s vote in favor (1) or opposition (0) to E-Verify. Unlike the state-level results, model 3.1 reveals a strong partisan division in legislator roll-call voting with Republican legislators more likely than Democrats to vote in favor of E-Verify.

Table 3.

Legislator Support for E-Verify.

	(3.1)	(3.2)
Republican	3.027*** (0.129)	1.992*** (0.202)
Farm owners		−15.448* (6.639)
Latino/Asian		−0.901 (0.603)
Legislator characteristics
Ideology		0.325** (0.110)
Female		−0.110 (0.131)
Latino		−0.302 (0.366)
Incumbent		−0.167 (0.135)
Contested election		−0.114 (0.136)
Agribusiness contributions (logged)		0.041* (0.020)
District characteristics
Average ideology		0.946*** (0.227)
Population (logged)		0.223 (0.154)
Median household income (logged)		0.222 (0.311)
Unemployment rate		−1.446*** (0.284)
Bachelor’s degree (+)		−0.320*** (0.093)
n	3,918	3,744

Models use logistic regression with state fixed-effects.

p < .05. **p < .01. ***p < .001.

Model 3.2 confirms that the effect of legislator partisanship on support for E-Verify persists even when controlling for a host of other factors. However, it also reveals support for a spatial theory of representation and our alternative hypothesis of sub-constituency influence. In the case of the former, the model indicates that moving from the most liberal to the most conservative district increases a representative’s predicted probability of supporting E-Verify by 14 percentage points. As for sub-constituency preferences, we again find that the effect of farm owners is most potent: moving from a district with the largest to smallest proportion of farm owners decreases the probability of a vote in favor of E-Verify by 9 percentage points.⁸ By contrast, the 5 percentage point drop shifting from the largest to the smallest Latino/Asian district is not statistically significant (p value of .112).

Tables 4 and 5 provide more rigorous tests of the sub-constituency hypothesis by looking at how sub-constituency and party compete to influence legislators. If our expectations are correct, then Republicans that represent larger farming constituencies should be less likely to favor E-Verify than their fellow party members and Democrats representing larger Latino/Asian constituencies should be even more likely to oppose E-Verify than the average Democrat. Based on Table 3, we know that Republicans in these states are exceedingly committed to supporting E-Verify, making the bar for finding a constituency effect quite high. Model 4.1, however, reaffirms that Republicans are cross-pressured when it comes to E-Verify: inclined to support it from a partisan perspective, but compelled to oppose it because of a loyalty to farmers. Indeed, of all the variables tested in this model, only the size of the farm owning constituency and the unemployment rate are capable of altering how Republicans vote on E-Verify.

Table 4.

Effect of Sub-Constituencies on Republican and Democratic Legislators.

	(4.1)	(4.2)
	Republican legislators	Democratic legislators
Farm owners	−21.80* (8.789)	8.927 (10.83)
Latino/Asian	−2.413 (2.123)	−1.500* (0.695)
Legislator characteristics
Ideology	−0.083 (0.224)	0.075 (0.151)
Female	−0.077 (0.302)	−0.211 (0.164)
Latino		−0.408 (0.394)
Incumbent	0.082 (0.234)	−0.151 (0.191)
Contested election	−0.245 (0.256)	−0.083 (0.184)
Agribusiness contributions (logged)	0.022 (0.038)	0.059* (0.027)
District characteristics
Average ideology	0.170 (0.539)	1.683*** (0.300)
Population (logged)	−0.198 (0.284)	0.625** (0.214)
Median household income (logged)	−0.036 (0.635)	0.132 (0.426)
Unemployment rate	−2.749*** (0.705)	−0.894* (0.361)
Bachelor’s degree (+)	−0.044 (0.205)	−0.382** (0.122)
n	1,821	1,474

Models use logistic regression with state fixed-effects.

p < .05. **p < .01. ***p < .001.

Table 5.

Trade-off Between Party and Sub-Constituency Preferences.

	(5.1)	(5.2)
	Republican legislators	Democratic legislators
Farm owners	−23.46* (9.117)	9.538 (11.91)
Latino/Asian	−2.500 (2.172)	−1.722* (0.764)
Vote margin	6.595*** (1.377)	20.59*** (1.951)
Farm Owners × Vote Margin
Latino/Asian × Vote Margin
Legislator characteristics
Ideology	−0.255 (0.230)	0.363* (0.171)
Female	−0.191 (0.306)	−0.099 (0.186)
Latino		−0.565 (0.427)
Incumbent	0.064 (0.240)	−0.224 (0.217)
Contested election	−0.246 (0.264)	0.122 (0.218)
Agribusiness contributions (logged)	0.037 (0.039)	0.094** (0.032)
District characteristics
Average ideology	0.311 (0.553)	1.664*** (0.330)
Population (logged)	−0.773* (0.316)	−0.036 (0.244)
Median household income (logged)	−0.028 (0.651)	0.245 (0.470)
Unemployment rate	−2.941*** (0.728)	−0.868* (0.405)
Bachelor’s degree (+)	0.011 (0.210)	−0.440** (0.136)
n	1,819	1,472

Models use logistic regression with state fixed-effects.

p < .05. **p < .01. ***p < .001.

Interpreting this result alongside model 3.2 suggests that conventional beliefs about the relationship between party pressure and legislative position taking on policies of lower public salience are partially correct, but lack nuance. Partisanship does influence roll-call behavior; however, legislators also react to the well-defined preferences of sub-constituencies—even if doing so means splitting with their party. Figure 2 illustrates the magnitude of this effect for Republicans. Moving from a district where farm owners make up less than 1 percent of the population to a district where farm owners reach their highest level decreases the probability of a Republican “yea” vote by a nearly 30 percentage points.

Figure 2.

Effect of farm owner constituency on republican and democratic support for E-Verify (with 95% confidence intervals).

Democrats, by contrast, are more likely to vote in favor of E-Verify as the size of their farming population increases (though not to a statistically significant degree). Assuming that conservative districts house larger farming populations, model 4.2 helps us to explain this. As the average ideology of their district becomes more conservative, the probability of a Democrat voting for E-Verify increases 25 percentage points. This could indicate that Democrats consider their constituencies in aggregate terms even when faced with less salient policy issues. However, model 4.2 also reveals that the size of the Latino/Asian population has a statistically significant effect on the direction of a Democrat’s vote. Specifically, the likelihood of a “yea” vote drops from 80 percent when representing a district with the smallest proportion of Latino/Asian constituents to less than 60 percent as the proportion of Latinos/Asians grows to 90 percent (Figure 3). This not only confirms that the effect of Latino/Asian preferences on legislative behavior is weaker than that of farmers (recall the null relationship found between Democratic-controlled legislatures and the size of the Latino/Asian community in the state-level models) but also demonstrates that at least some representatives are responsive to Latino/Asian interests.

Figure 3.

Effect of Latino/Asian constituency on republican and democratic support for E-Verify (with 95% confidence intervals).

The results from Table 4 yield encouraging results for the second hypothesis: Republicans are responsive to farm owners and Democrats to Latinos/Asians. Yet, it is possible that what we are perceiving as a trade-off between party and constituency is actually a product of the vote margin. In other words, legislators vote with their constituents only when they can do so without consequence, for example, when the vote is all but assured to go the direction of their party majority’s liking. According to Snyder and Groseclose (2000), party leaders intensify pressure on their members to vote the party-line when the roll-call is expected to be close. This means that for us to judge sub-constituency pressure as having a truly meaningful effect on legislative behavior, we would need to show that legislators are willing to deviate from their party majorities even when the vote margin is close. This is tested in Table 5, which adds “vote margin” to the regression model. Vote margin is operationalized as the percentage of “yea” votes, which varies from 53 to 100 percent. Larger values indicate the bill passed by a wider margin.

Models 5.1 and 5.2 reveal that both Republican and Democratic legislators are less likely to vote for E-Verify when they face a close roll-call vote. The predicted probability of the vote margin’s effect on Republicans and Democrats is captured in Figures 4 and 5, respectively: with 93 percent of votes cast in favor of E-Verify, Republicans have a 98 percent and Democrats a 96 percent predicted probability of voting for E-Verify (all other variables are kept at their mean values). However, when the percentage of “yea” votes is at its minimum (53%), the probability of Republicans and Democrats supporting E-Verify drops to 87 percent and less than 1 percent, respectively. Even without an interaction term, we can already see how including vote margin in the analysis limits the potential effect of sub-constituencies on their legislators’ roll-call behavior.⁹ As the vote margin approaches its maximum (100%), sub-constituencies are precluded from having any effect on Republican and Democratic legislators alike as the probability of a “yea” vote nears 1. Similarly, as the vote margin approaches its minimum (53%), it becomes exceedingly unlikely that the size of the Latino/Asian community will meaningfully influence Democratic behavior as the predicted probability of a Democratic “yea” vote is already less than 1 percent.

Figure 4.

Effect of vote margin on predicted probability of Republican yea vote (with 95% confidence intervals).

Figure 5.

Effect of vote margin on predicted probability of Democratic yea vote (with 95% confidence intervals).

These findings have important implications. They confirm once again that Republicans and Democrats are less likely to support E-Verify as the size of their salient constituencies increase—even if the roll-call vote is narrow. Whereas previous studies have characterized state legislatures as collectively irresponsible (Rogers 2017), these findings suggest that aggregate irresponsiveness may be due to a high level of individual responsiveness. This behavior is exemplified by Republican legislators who, even when faced with a close roll-call, defended the preferences of farm owners despite their party’s official commitment to stricter immigration enforcement, including E-Verify. Yet, the dramatic difference in Democratic support for E-Verify tied to the size of the roll-call margin (Figure 5) indicates that Democrats are unlikely to support E-Verify under heightened party pressure irrespective of the Latino/Asian community’s size. Although party pressure is aligned with the interests of Latinos/Asians in this case, this behavior implies that Democrats may be more willing to cast aside constituent preferences when party pressure demands they do so. Additional tests interacting vote margin with the sub-constituency variables corroborate these results and are discussed in detail in the supplemental appendix.¹⁰

Conclusion

This work contributes to our understanding of legislative behavior by extending national-level theories connecting constituency pressure and legislative responsiveness to the state level. Most state-level research has concluded that a policy’s salience generates a district-party trade-off whereby aggregate preferences inform legislative behavior on salient issues, but party pressure drives behavior on less salient policy. The findings of this research suggest that this trade-off is less straightforward and may depend on the level of analysis and the party/constituents in question.

At the legislature level, E-Verify is kept off the agenda in states where pressure from farm owners is strongest; yet, when we account for party-control of the legislature, we find that the influence of farm owners on policy output is limited to Republican legislatures. The size of the Latino/Asian community, on the contrary, has no effect on the probability of a state adopting E-Verify regardless of partisan control. At the district level, there is some indication that Democrats are responsive to Latino/Asian preferences, but it is unclear whether this influence occurs independent of party pressure as Democrats are prone to voting against E-Verify in general. Among Republicans, there is strong evidence that the preferences of farm owners influence behavior even in contexts where party pressure should be greatest: close roll-call margins. Thus, state legislatures are not collectively irresponsible (Rogers 2017). Responsiveness to sub-constituencies may yield policy results that appear collectively irresponsible, but are actually the result of individual legislators striving to be accountable to particular constituents.

To provide greater cogency to this argument, future studies should seek a better understanding of the “behind the scenes” process that motivates legislative behavior and the coalitions that form between parties and constituents. In the case of the former, topics that require greater scrutiny include examining the strategic decisions of party leaders in terms of scheduling legislation and/or preventing legislation from appearing on the floor, accounting for the possibility of logrolling and vote trading, and achieving an enriched understanding of the innerworkings of the committee process that so often halts legislation from moving forward. It is probable that these aspects of the legislative process—which are not accounted for in the prior analyses—would help us to gain a more comprehensive picture of observed relationships. In terms of the latter, the findings presented imply that certain constituencies may hold more influence over legislators than others. For example, farm owners have a strong effect on state legislative behavior in general while the preferences of Latino and Asian communities only seem to influence the behavior of individual Democrats—and, then, only if the roll-call margin is large (and, presumably, party pressure is weak). Future research would be wise to closely examine the mechanisms that link constituent preferences to legislative responsiveness, paying special attention to how responsiveness varies depending on the group’s level of organization and their stake in the outcome.

Supplemental Material

Supplemental_Appendix – Supplemental material for Sub-constituencies and Legislative Responsiveness: Evidence from the States

Supplemental material, Supplemental_Appendix for Sub-constituencies and Legislative Responsiveness: Evidence from the States by Jillian Jaeger in Political Research Quarterly

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

Supplemental materials for this article are available with the manuscript on the Political Research Quarterly (PRQ) website.

References

Agri-Pulse Communications. 2016. “2016 Producer Survey Results.” January 28. https://www.agri-pulse.com/ext/resources/pdfs/a/g/r/1/6/AgriPulse_Presidential_Survey_Results_1_29_2016.pdf. Accessed January 20, 2018.

Allison

Paul D.

1982. “Discrete-Time Methods for the Analysis of Event Histories.” Sociological Methodology 13:61–98.

Alvarez

José

. 2005. “Sweetening the US Legislature: The Remarkable Success of the Sugar Lobby.” Political Quarterly 76 (1): 92–99.

Andersen

Kristi

. 2006. “In Whose Interest? Political Parties, Context, and the Political Incorporation of Immigrants.” In New Race Politics in America: Understanding Minority and Immigrant Voting, edited by Junn

Jane

Haynie

Kerry

, pp. 17-38. Cambridge: Cambridge University Press.

Arceneaux

Kevin

. 2002. “Direct Democracy and the Link between Public Opinion and State Abortion Policy.” State Politics & Policy Quarterly 2 (4): 372–88.

Arnold

R. Douglas

. 1990. The Logic of Congressional Action. New Hartford: Yale University Press.

Bellemare

Marc F.

Carnes

Nicholas

. 2015. “Why Do Members of Congress Support Agricultural Protection?” Food Policy 50 (January): 20–34.

Bernstein

Robert A.

1989. Elections, Representation, and Congressional Voting Behavior: The Myth of Constituency Control. Englewood Cliffs: Prentice Hall.

Berry

William D.

DeMeritt

Jacqueline H. R.

Esarey

Justin

. 2010. “Testing for Interaction in Binary Logit and Probit Models: Is a Product Term Essential?” American Journal of Political Science 54 (1): 248–66.

10.

Birkhead

Nathaniel A.

2015. “The Role of Ideology in State Legislative Elections.” Legislative Studies Quarterly 40 (1): 55–82.

11.

Bishin

Benjamin G.

2000. “Constituency Influence in Congress: Does Subconstituency Matter?” Legislative Studies Quarterly 25 (3): 389–415. doi:10.2307/440413

12.

Bishin

Benjamin G.

2009. Tyranny of the Minority: The Subconstituency Politics Theory of Representation. Philadelphia: Temple University Press.

13.

Brace

Paul

Sims-Butler

Kellie

Arceneaux

Kevin

Johnson

Martin

. 2002. “Public Opinion in the American States: New Perspectives Using National Survey Data.” American Journal of Political Science 46 (1): 173–89.

14.

Bratton

Kathleen A.

2006. “The Behavior and Success of Latino Legislators: Evidence from the States.” Social Science Quarterly 87 (5): 1136–57.

15.

Bratton

Kathleen A.

Haynie

Kerry L.

1999. “Agenda Setting and Legislative Success in State Legislatures: The Effects of Gender and Race.” The Journal of Politics 61 (3): 658–79.

16.

Bullock

Charles S.

III Brady

D. W.

1983. “Party, Constituency, and Roll-Call Voting in the U. S. Senate.” Legislative Studies Quarterly 8 (1): 29–43.

17.

Caldwell

Alicia A.

2011. “E-Verify Bill Gives Agriculture Extra Time.” The San Diego Union-Tribune. June 14. http://www.sandiegouniontribune.com/sdut-e-verify-bill-gives-agriculture-extra-time-2011jun14-story.html. Accessed February 19, 2017.

18.

Canes-Wrone

Brandice

Brady

David W.

Cogan

John F.

2002. “Out of Step, Out of Office: Electoral Accountability and House Members’ Voting.” American Political Science Review 96 (1): 127–49.

19.

Carson

Jamie

Koger

Gregory

Lebo

Matthew

Young

Everett

. 2010. “The Electoral Costs of Party Loyalty in Congress.” American Journal of Political Science 54 (3): 598–616.

20.

Center for Responsive Politics. 2016. “Agribusiness: Long-term Trends.” Open Secrets. https://www.opensecrets.org/industries/totals.php?ind=A (accessed July 20, 2018).

21.

Clinton

Joshua D.

2006. “Representation in Congress: Constituents and Roll-Calls in the 106th House.” Journal of Politics 68 (2): 397–409.

22.

Conner

Chuck

. 2011. “Statement by National Council of Farmer Cooperatives President Chuck Conner at the Press Conference of Agricultural Leaders Discussing the Impact of Mandatory E-Verify Legislation.” June 16. http://ncfc.org/press-release/statement-by-ncfc-leaders-at-todays-press-conference-of-the-impact-of-e-verify-on-agriculture/. Accessed February 11, 2017.

23.

Converse

Philip E.

1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, edited by Apter

David

, pp. 206-261. New York: The Free Press.

24.

Cox

Gary

McCubbins

Mathew

. 2005. Setting the Agenda: Responsible Party Government in the US House of Representatives. Cambridge: Cambridge University Press.

25.

Creek

Heather M.

Yoder

Stephen

. 2012. “With a Little Help from Our Feds: Understanding State Immigration Enforcement Policy Adoption in American Federalism.” Policy Studies Journal 40 (4): 674–97.

26.

Dye

Thomas R.

1961. “A Comparison of Constituency Influences in the Upper and Lower Chambers of a State Legislature.” The Western Political Quarterly 14 (2): 473–80.

27.

Ekins

Emily.

2013. “Support for Immigration Reform Correlated with Perception of Immigrants’ Economic Impact.” Reason-Rupe Public Opinion Survey. https://reason.com/assets/db/13648288657856.pdf (accessed March 3, 2017).

28.

Erikson

Robert S.

Wright

Gerald C.

McIver

John P.

1993. Statehouse Democracy: Public Opinion and Policy in the American States. Cambridge: Cambridge University Press.

29.

Farm Bureau. 2017. “E-Verify.” http://www.fb.org/issues/immigration-reform/e-verify/. Accessed February 11 2017.

30.

Fenno

Richard F.

1978. Home Style: House Members in Their Districts. New York: Longman.

31.

Gerber

Elisabeth R.

1996. “Legislative Response to the Threat of the Popular Initiative.” American Journal of Political Science 40:99–128.

32.

Goff

Brian L.

Grier

Kevin B.

1993. “On the (Mis)Measurement of Legislator Ideology and Shirking.” Public Choice 76 (1–2): 5–20. doi:10.1007/BF01049340

33.

Gray

Virginia

Lowery

David

Fellowes

Matthew

McAtee

Andrea

. 2004. “Public Opinion, Public Policy, and Organized Interests in the American States.” Political Research Quarterly 57 (3): 411–20.

34.

Haider-Markel

Donald P.

Kaufman

Matthew S.

2006. “Public Opinion and Policy Making in the Culture Wars: Is There a Connection between Opinion and State Policy on Gay and Lesbian Issues?” In Public Opinion in State Politics, edited by Cohen

Jeffrey E.

, 163–82. Stanford: Stanford University Press.

35.

Hayes

Thomas J.

Bishin

Benjamin G.

2012. “Issue Salience, Subconstituency Politics, and Legislative Representation.” Congress & the Presidency 39 (2): 133–59.

36.

Hero

Rodney E.

2010. Latinos and the U.S. Political System: Two-Tiered Pluralism. Philadelphia: Temple University Press.

37.

Hogan

Robert E.

2004. “Challenger Emergence, Incumbent Success, and Electoral Accountability in State Legislative Elections.” The Journal of Politics 66 (November): 1283–303.

38.

Jacobson

Gary C.

1993. “Deficit-Cutting Politics and Congressional Elections.” Political Science Quarterly 108 (October): 375–402.

39.

Jacoby

William G.

Schneider

Saundra K.

2001. “Variability in State Policy Priorities: An Empirical Analysis.” The Journal of Politics 63 (2): 544–68.

40.

Jenkins

Shannon

. 2006. “The Impact of Party and Ideology on Roll-Call Voting in State Legislatures.” Legislative Studies Quarterly 31 (2): 235–57.

41.

Jenkins

Shannon

. 2008. “Party Influence on Roll-Call Voting: A View from the US States.” State Politics & Policy Quarterly 8 (3): 239–62.

42.

Jenkins

Shannon

. 2010. “Examining the Influences over Roll-Call Voting in Multiple Issue Areas: A Comparative U.S. State Analysis.” Journal of Legislative Studies 16 (1): 14–31.

43.

Khan

Mahwish

. 2011. “Spanish-Language Media Ad Campaign Highlights Hypocrisy of E-Verify’s GOP Backers.” America’s Voice. September 28. https://americasvoice.org/blog/five-figure_spanish-language_media_campaign_opposes_mandatory_e-verify/. Accessed February 5, 2018.

44.

Kingdon

John W.

1968. Candidates for Office: Beliefs and Strategies. Ann Arbor: University of Michigan Press.

45.

Kirkland

J. H.

Harden

J. J.

2016. “Representation, Competing Principals, and Waffling on Bills in US Legislatures.” Legislative Studies Quarterly 41:657–86.

46.

Klarner

Carl

Berry

William D.

Carsey

Thomas M.

Jewell

Malcolm

Niemi

Richard

Powell

Lynda

Snyder

James

. 2011. State Legislative Election Returns (1967-2010). Ann Arbor: Inter-university Consortium for Political and Social Research [distributor], 2013-01-11. doi:10.3886/ICPSR34297.v1

47.

Krehbiel

Keith

. 1993. “Where’s the Party?” British Journal of Political Science 23:235–66.

48.

Lax

Jeffrey R.

Phillips

Justin H.

2009. “Gay Rights in the States: Public Opinion and Policy Responsiveness.” American Political Science Review 103 (3): 367–86.

49.

Lax

Jeffrey R.

Phillips

Justin H.

2012. “The Democratic Deficit in the States.” American Journal of Political Science 56 (1): 148–66.

50.

Lind

Nancy S.

Rankin

Erik T.

Harris

Gardenia

. 2016. Today’s Economic Issues: Democrats and Republicans. Santa Barbara: ABC-CLIO.

51.

Lopez

Mark Hugo

Minushkin

Susan

. 2008. “2008 National Survey of Latinos.” Washington, DC: Pew Hispanic Center. September 18.

52.

Lowry

Robert C.

Alt

James E.

Ferree

Karen E.

1998. “Fiscal Policy Outcomes and Electoral Accountability in American States.” American Political Science Review 92 (4): 759–74.

53.

Lupia

Arthur

Krupnikov

Yanna

Seth Levine

Adam

Piston

Spencer

Von Hagen-Jamar

Alexander

. 2010. “Why State Constitutions Differ in Their Treatment of Same-Sex Marriage.” The Journal of Politics 72 (4): 1222–35.

54.

Mayhew

David R.

1974. Congress: The Electoral Connection. 2nd ed. New Haven: Yale University Press.

55.

Miller

Warren E.

Stokes

Donald E.

1963. “Constituency Influence in Congress.” The American Political Science Review 57 (1): 45–56. doi:10.2307/1952717

56.

Monogan

James E.

2013. “The Politics of Immigrant Policy in the 50 US States, 2005-2011.” Journal of Public Policy 33 (1): 35–64.

57.

Newman

Benjamin J.

Christopher D.

Johnston

Strickland

April A.

Citrin

Jack

. 2012. “Immigration Crackdown in the American Workplace: Explaining Variation in E-Verify Policy Adoption across the US States.” State Politics & Policy Quarterly 12 (2): 160–82.

58.

Norrander

Barbara

. 2000. “The Multi-Layered Impact of Public Opinion on Capital Punishment Implementation in the American States.” Political Research Quarterly 53:771–94.

59.

Nyhan

Brendan

McGhee

Eric

Sides

John

Masket

Seth

Greene

Steven

. 2012. “One Vote Out of Step? The Effects of Salient Roll-Call Votes in the 2010 Election.” American Politics Research 40 (September): 844–79.

60.

Patterson

S. C.

1996. “Legislative Politics in the States.” In Politics in the American States, edited by Gray

Jacob

, 6th ed., 159–206. Washington, DC: Congressional Quarterly.

61.

Pew Research Center. 2015. “A Deep Dive into Party Affiliation: Sharp Difference by Race, Gender, Generation, Education.” April 7. http://www.people-press.org/2015/04/07/a-deep-dive-into-party-affiliation/. Accessed January 2, 2018.

62.

Plotnick

Robert D.

Winters

Richard F.

1985. “A Politico-Economic Theory of Income Redistribution.” The American Political Science Review 79 (2): 458–73. doi:10.2307/1956660

63.

Rogers

Steven

. 2017. “Electoral Accountability for State Legislative Roll-Calls and Ideological Representation.” American Political Science Review 111 (3): 555–71.

64.

Shor

Boris

McCarty

Nolan

. 2011. “The Ideological Mapping of American Legislatures.” American Political Science Review 105 (3): 530–51.

65.

Snyder

James M.

Groseclose

Tim

. 2000. “Estimating Party Influence in Congressional Roll-Call Voting.” American Journal of Political Science 44 (2): 193–211.

66.

Tausanovitch

Chris

Warshaw

Christopher

. 2013. “Measuring Constituent Policy Preferences in Congress, State Legislatures, and Cities.” The Journal of Politics 75 (2): 330–42.

67.

Treadway

J. M.

1985. Public Policy Making in the American States. New York: Praeger.

68.

U.S. Congress House Committee on the Judiciary. 2013. How E-Verify Works and How It Benefits American Employers and Workers: Hearing before the Subcommittee on Immigration and Border Security, 113th Cong., 1st sess., February 27.

69.

Vesenka

Mary H.

1989. “Economic Interests and Ideological Conviction.” Journal of Economic Behavior & Organization 12 (2): 259–63.

70.

Wickham

Greg

. 2011. “Statement by Greg Wickham, Chairman of NCFC and CEO of Dairylea Cooperative, Syracuse, N.Y. at the Agriculture Press Conference on the Impact of Mandatory E-Verify Legislation.” June 16. http://ncfc.org/press-release/statement-by-ncfc-leaders-at-todays-press-conference-of-the-impact-of-e-verify-on-agriculture/. Accessed February 11, 2017.

71.

Wilson

Reid.

2013. “North Carolina GOP Split on Immigration.” The Washington Post. September 4, 2013. https://www.washingtonpost.com/blogs/govbeat/wp/2013/09/04/north-carolina-gop-split-on-immigration/?utm_term=.7b54a815257d. Accessed February 25, 2017.

72.

Wong

Tom K.

2012. “287(g) and the Politics of Interior Immigration Control in the United States: Explaining Local Cooperation with Federal Immigration Authorities.” Journal of Ethnic and Migration Studies 38 (5): 737–56. doi:10.1080/1369183X.2012.667983

73.

Zingher

Joshua N.

2014. “The Ideological and Electoral Determinants of Laws Targeting Undocumented Migrants in the US States.” State Politics & Policy Quarterly 14 (1): 90–117.

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

0.98 MB