Municipal institutions and local policy responses to immigrants: policies towards day labourers in California

ABSTRACT

This paper examines the various factors that influence how local elected officials (mayors and council persons) in the United States develop policy responses to immigrant day labourers in their jurisdictions. We argue that in addition to demographic and political factors, the design of municipal political institutions shapes their responses. Inclusive day labourer policies are more likely to occur in mayor–council systems than in council–manager systems. In addition, we argue that it is important to consider how the interaction of factors affects local policy responses. Several factors have been identified as impacting local immigration policies, but in order to understand the widespread variation in local policy responses, it is necessary to examine how these factors interact in particular localities over extended periods of time. Interaction leads to dynamic policies that change over time. Local policy responses are examined quantitatively – using a dataset based on day labourer hiring sites in California and fitting multilevel generalized linear models – and qualitatively – using transcribed city council meetings.

KEYWORDS:

• immigration
• day labour
• policies
• municipalities
• United States
• council
• mayor

INTRODUCTION

For three decades, the number of immigrant day labourers in the United States has increased considerably (Crotty, 2015, 2017; Varsanyi, 2008; Visser et al., 2017). Mayors and city council persons have introduced policies to regulate the activities of day labourers within their jurisdictions. However, these responses have varied widely (Crotty, 2017; Nicholls, 2019; Varsanyi, 2008; Visser et al., 2017). For instance, in Costa Mesa, California, day labourer issues have been a source of contention for years. The city opened a formal day labour centre in the late 1980s accompanied by no-solicitation ordinances and increased police enforcement. For many years, Costa Mesa’s elected officials battled over who should be allowed to use the day labour centre. Several councilmembers believed that the city should not serve day labourers with an unauthorized immigration status. In the late 2000s, when Costa Mesa had a mayor that made immigration enforcement a key point during his term in office, the day labour centre was closed. On the other hand, in Pasadena, California, responses to day labourer issues were more inclusive. The emphasis was on a viable solution to address concerns from neighbours while also providing a space for day labourers to look for work. Checking the immigration status of day labourers was not considered to be a responsibility of the city’s day labour centre.

The examples of Costa Mesa and Pasadena illustrate that two seemingly similar cities, large middle-class suburban cities in Southern California, responded differently to the issue of day labourers. Such differences were expressed in terms of the level and breadth of involvement by different elected officials (council persons and mayor) and preferences for policies. What explains for such varied responses by elected officials to immigrant day labourers?

The literatures on day labourers and local immigration policies leave us somewhat ill-equipped to answer this question. The literature has shown that policies are largely bifurcated between restrictive, such as no-solicitation ordinances and increased police enforcement, and inclusive policies, such as operating a day labourer centre (Crotty, 2017; Nicholls, 2019; Varsanyi, 2008; Visser et al., 2017). Such differences in local policies are commonly explained by demographics, political preferences, history of place, immigration policies in other jurisdictions and saliency of immigration at the federal level (de Graauw & Vermeulen, 2016; Hopkins, 2010; Huang & Liu, 2018; Ramakrishnan & Wong, 2010; Visser & Simpson, 2019; Walker, 2014, 2015; Walker & Leitner, 2011). These accounts have identified key factors that shape responses by elected officials, but they tell us less about how the institutional features of municipal government matter as well.

Scholars studying the representation of minorities and accountability to minority interests have found that characteristics of institutional design can be important influencing factors (Eisinger, 1973; Ihlanfeldt, 2011; Maser, 1985; Trebbi et al., 2008; Trounstine, 2010). This literature suggests that the design of city governing institutions mediates how elected officials respond to the demands and preferences of their constituents. More specifically, the form of government (mayor–council versus council–manager) and form of elections (district based versus at-large) can enhance or dilute minority influence on municipal policies. Thus, as immigrants and day labourers often constitute a minority in local municipalities, it is important to assess how institutional features of municipal government impact local policy responses to immigrants and day labourers.

This paper draws on these various literatures to assess how mayors and council persons develop policies in response to immigrant day labourers. It examines the extent to which institutional design of local government interacts with other factors to shape local policy responses. In doing so, the paper considers the role of institutional design to be part of a complex web of factors that operate between and across multiple scales while affecting local policy responses. The two main design features assessed here are: (1) the form of municipal elections (district based versus at-large) and (2) the form of government (mayor–council versus council–manager). We discuss how these features of local institutional design impact types of responses by elected officials to day labourer issues. We construct and employ a database on day labourer hiring sites in California (1985–2020) to assess local officials’ responses. The dataset provides data on day labourer hiring sites and demographic and political characteristics and allows one to fit statistical models. The paper also draws on the direct transcriptions of council meetings addressing day labourer issues of three case studies to illustrate the complexity of responses.

LOCAL IMMIGRATION POLICIES AND INSTITUTIONAL DESIGN

Day labourer policies

Since the 1990s the number of immigrant day labourers in the United States has increased rapidly (Crotty, 2015; Rodriguez, 2008; Visser et al., 2017). Day labourers are employed in informal labour markets and get paid by day or project. In the United States, day labourers generally try to sell their labour in public spaces, such as street corners, parking lots or vacant lots (Valenzuela & Melendez, 2003). Although the presence of immigrant day labourers in the United States can be traced back centuries, tensions started to emerge in the late 1980s. Traditionally day labourers had worked, mostly out of the public eye, in agriculture and rural areas. However, a substantial growth in residential construction increased the demand for a flexible and low-paid workforce (Parrado & Kandel, 2008), which lured day labourers to urban areas. Day labourers became more visible to the community in these urban areas as they gathered at hiring sites in public spaces (Esbenshade, 2000; Valenzuela, 1999). The increased visibility of a rapid growing immigrant day labour population resulted in tensions in municipalities across the United States (Crotty, 2015; Esbenshade, 2000). Residents, business owners and local officials complained about traffic and public safety issues, deteriorating quality of life, and the presumed unauthorized immigration status of day labourers (Crotty, 2017; Varsanyi, 2008; Wells, 2004).

Local responses to tensions surrounding day labourer issues vary. Restrictive policies are implemented, such as no-solicitation ordinances and increased police enforcement, to repress day labour activities. Other responses focus on integrating day labour activities by installing organized hiring sites. And, finally, some responses entail ignoring the issue in the hope that tensions surrounding day labour activities will diminish over time (Crotty, 2017; Nicholls, 2019; Rodriguez, 2008). Local policy responses to immigration issues in general have increased over the past 30 years, and those responses have shown great variation (Ellis, 2006; Rodriguez, 2008; Varsanyi, 2008; Visser, 2017; Visser & Simpson, 2019; Walker, 2014, 2015; Wells, 2004).

Several studies have examined the factors responsible for the variation in local policy responses to immigration. The following are the most prominent factors responsible for impacting local policy responses: demographics, such as growth of the foreign born or Latino population, homeownership and educational attainment (Huang & Liu, 2018; Visser & Simpson, 2019; Walker & Leitner, 2011; Wong, 2012); political preferences of residents (Huang & Liu, 2018; Ramakrishnan & Wong, 2010; Walker & Leitner, 2011); history of place, such as a lack of experience with ‘others’ can increase the likelihood of a hostile response (de Graauw & Vermeulen, 2016; Hanlon & Vicino, 2015; Lal, 2013; Pastor & Mollenkopf, 2012; Rodriguez, 2008; Vitiello, 2014; Walker & Leitner, 2011); immigration policies in other jurisdictions (Ellis, 2006; Varsanyi, 2008; Visser & Simpson, 2019; Walker, 2015; Wells, 2004); and saliency of immigration at the federal level (Hopkins, 2010).

Institutional design and involvement in day labourer policies

Though immigration scholars have focused on the above factors to explain policy variations, scholars of minority politics in American cities have maintained that institutional design mediates how elected officials respond to the competing preferences of constituents in their cities. The two important characteristics of institutional design are, according to this literature, the form of municipal elections and the form of government.

First, the form of municipal elections refers to the voting system used to elect local representatives. The two most common forms of municipal elections in the United States are at-large and district systems. In an at-large system, representatives are elected by the entire city electorate. In district systems, the city is divided into geographical districts of roughly equal population size and each of these districts elects its own district representative (Trounstine, 2010; Trounstine & Valdini, 2008). Studies have found that at-large systems make it more difficult for minorities to be represented on the city council, while district systems make it more likely that minorities will be represented (Eisinger, 1973; Trebbi et al., 2008; Trounstine, 2010). However, district systems are only beneficial to minorities if the minority constitutes a moderate share of the city population and is geographically concentrated (Trounstine & Valdini, 2008; Collingwood & Long, 2021). In addition to representation, the form of municipal elections also affects the accountability to minorities. An at-large elected representative has fewer incentives to cater to minority interests as this is not considered to be efficient. The representative is elected citywide by the majority, and is thus more likely to respond to the interests of the ‘median voter’ (Eisinger, 1973; Trounstine, 2010). Thus, demographics and political features are important, but the form of municipal elections can either enhance or dilute minority representation and accountably.

Second, the representation of minorities and accountability to minority interests are also affected by the form of government. The two main forms of government are mayor–council systems and council–manager systems. In a mayor–council system, the mayor holds executive power while the council holds legislative power. A council–manager type of government resembles a business structure in which the manager is appointed by the council (Wolman, 1995). Mayor–council and council–manager systems differ with regard to the role of the mayor. In a mayor–council system the mayor enjoys considerable authority, including oftentimes veto powers (Clingermayer & Feiock, 2001). Consequently, a mayor in a mayor–council system is provided with many possibilities to shape local policies. In a council–manager system, a mayor enjoys less authority. It has been argued that reducing the power of a mayor can be a disadvantage for minorities, as mayors can be a source of protection and access for minorities (Maser, 1985). Mayors can be more responsive to minority interests than city managers. The reason for this is that mayors are motivated by political incentives, while city managers focus on management efficiency. Satisfying minority interests is not considered to be particularly efficient for city management (Ihlanfeldt, 2011).

Thus, the form of municipal elections and the form of government impact the representation of minorities and the accountability to minority interests. Minorities fare better in district systems and mayor–council systems, and it has been argued that the effect is cumulative. That is, if a municipality is structured with a district and mayor–council system, minorities are even better served than when only one of these institutional components is present (Eisinger, 1973). When applying these findings to the case of local immigrant policies, it can be expected that immigrants are likely to benefit from mayor–council and district systems as immigrants often constitute a minority. We argue that these findings apply to the case of day labourers as well, even though a large share of day labourers may be ineligible to vote due to their unauthorized immigration status. For instance, providing day labourers with a formalized space to account for their need to search for work may alleviate the concerns of constituents. In addition, local elected officials have to honour the constitutional rights of day labourers as well. Thus, there can be incentives for local elected officials to account for the needs of day labourers.

Day labourer population

When the findings from the immigration policy literature and the literature on minority representation and accountability are combined, it can be argued that local day labourer policies are steered by demographics, political preferences, history of place, policy activities in other jurisdictions, saliency of immigration and the design of municipal government. Based on previous studies, it can, for instance, be expected that a rapid growth of the Latino population increases the likelihood of restrictive day labourer policies, while a predominantly Democratic voting population fosters the chances of inclusive day labourer policies (Huang & Liu, 2018; Ramakrishnan & Wong, 2010; Visser & Simpson, 2019; Walker & Leitner, 2011). With regard to the design of municipal government, it would be expected that day labourers benefit from mayor–council and district systems.

Though these factors and institutions are important, two important issues also impact how elected officials address the day labourer issue. First, day labourers are often impoverished, racially stigmatized and unauthorized, and many seek employment in highly visible public locations. Their precarious and public position makes them into a highly polarized issue, even in municipalities where we would expect to find greater support for immigrants. Polarization increases the likelihood that a municipality responds less welcoming to immigrants (Pastor & Mollenkopf, 2012). Second, when confronting the issue, the policy tools available to local elected officials are constrained. Many local governments have become involved in federal immigration enforcement, but their participation is mostly circumscribed to federal programmes such as Secure Communities. Outside of these federal programmes, they must adhere to the legal principles of federal pre-emption over immigration matters and equal personhood for all persons present in the United States. Consequently, they have limited authority to design and enact policies that explicitly address the immigration status of people within their jurisdictions (Ellis, 2006; Gilbert, 2009; Rodriguez, 2008; Varsanyi, 2011; Walker, 2014; Wells, 2004). This provides supporters of immigrants with a legal space to challenge such policies. Thus, while polarization sustains heated conflict over an extended period of time, constitutional limits on local government restrict the range of policy options available to elected officials.

It is important to acknowledge that the factors that have been identified as impacting local policy responses are part of a complex web of forces that interact between and across multiple scales (Jones-Correa, 2008; McDaniel et al., 2019). Factors combine in highly contingent ways and change over time. The contingent combination of these factors in particular places and changes to some or all factors across time result in policies that are dynamic, changing and not entirely predictable.

Taken together, it can be argued that there is variation in local policy responses to day labourer issues. As discussed above, several factors have been identified as impacting local policy responses to immigration issues, but not much attention has been paid to the role of institutional design. In order to understand local policy responses to immigration issues, it is important to examine the extent to which local elected officials are influenced by the governmental institutions in which they operate. It can be expected that immigrants benefit from mayor–council and district systems. However, when examining the effects of institutional design and other factors on local policy responses, it is essential to remember that these factors are interdependent and dynamic. The continuous interaction of factors between and across multiple levels results in policy outcomes that can change over time. In the case of day labourers, an important mediating factor is the potential of polarization.

METHODS

Two main data sources are used for this paper. The first data source is derived from a dataset on conflict and day labourer hiring sites in California between 1985 and 2020. For all Californian municipalities included in the 2010 decennial census, extensive online searches were conducted to identify conflict and day labourer hiring sites. The sources used were newspaper archives, published studies on day labourers and Google Street View. Google Street View is increasingly used as a data source in the field of urban studies (e.g., Wang & Vermeulen, 2021). It has been used to confirm if day labourer hiring sites named in newspaper articles and studies were still active, as hiring sites are informal and temporal gathering places. In addition, all Californian municipalities were checked via Google Street View for potential hiring sites. These potential hiring sites include home improvement stores (Home Depot, Lowe’s), rental facilities (U-Haul), paint stores, lumber yards, convenience stores (7-Eleven) and locations suggested online. The addresses of these potential sites have been explored using Google Street View, and if day labourers were visible at the addresses, the sites are added to the dataset. Day labourers are classified as men, standing around in small or larger groups, while wearing working clothes and/or backpacks in the vicinity of a potential hiring site. Gatherings of people for other purposes (such as waiting areas for public transport) have been taken into consideration to make sure that those instances were not included. The data have been complemented with demographic, political and policy data. Much of the previous research on day labourer issues has used fieldwork to identify and analyse hiring sites. Though the approach taken in this paper cannot capture the ethnographic depth of a fieldwork approach, it can capture the breadth and diversity of hiring sites in California.

The dataset that has been deployed for the quantitative analyses in this paper is a subset of the dataset on day labourer hiring sites and conflict in California. The subset includes municipalities with a municipal government system that have had at least one hiring site within their jurisdiction. Only municipalities with a municipal government are included as this paper aims to examine the impact of municipal design. The data source used for the statistical models consists of 278 hiring sites in 116 municipalities in California that were active between 1985 and 2015 (Table 1).

Table 1. Overview of the subset employed for the quantitative analyses: municipalities, hiring sites and implemented policies.

Period	Municipalities with hiring sites	Active hiring sites	Municipalities that implemented pro-day labourer policies	Municipalities that implemented anti-day labourer policies
1985–90	21	51	8	7
1991–95	45	73	7	10
1996–2000	53	91	9	12
2001–05	80	148	10	8
2006–10	102	190	15	15
2011–15	88	145	2	3
Total	116	278	39	44

The second data source consists of transcribed recordings of council meetings. Three cities have been selected that provide recordings of their council meetings on day labourer issues. These municipalities are Baldwin Park, Mission Viejo and Pasadena. The recordings that addressed day labourer issues were transcribed for each municipality. The transcribed meetings were then coded to identify actors, motivations of actors to get involved in discussions on day labourer issues and proposed solutions to day labourer issues. The coded data were used to obtain more insight into the involvement of local officials in day labourer issues.

The dataset on day labourer hiring sites is used to perform statistical analyses, while the transcribed council meetings are used to assess how the factors analysed in the statistical analyses interact with one another in dynamic ways to generate varied policy responses.

Multilevel generalized linear model

The dataset on day labourer hiring sites contains repeated measures that are clustered in sites, cities and counties. To account for the repeated measures and clusters, multilevel generalized linear models have been fitted to the data (Finch et al., 2019). Three different models with binary dependent variables have been constructed. In the first model the dependent variable is ‘no policy’ (0) versus ‘some policy’ (1), the dependent variable in the second model is ‘no restrictive day labourer policy’ (0) versus ‘restrictive day labourer policy’ (1), and in the third model the dependent variable is ‘no inclusive day labourer policy’ (0) versus ‘inclusive day labourer policy’ (1). In the second and third models, instances of ‘no policy’ are coded as zero. The included independent variables are the same for the three models.

The first set of independent variables relate to demographic characteristics of the census tracts in which day labourer hiring sites are located. The level of census tract has been chosen because this the most detailed level available for comparison over time of census data. As responses to immigrants do not have to be uniform across a city (McDaniel et al., 2019), studying the census tracts of the hiring sites allows one to account for variation in responses within cities. The per cent Latino residents in a census tract is included to examine if the share of the Latino population affects the response to day labourers, who are presumed to be largely Latino. The growth of the Latino population, and the percentage owner-occupied housing, are included because previous studies have found that an increase of these factors increases the likelihood of restrictive local immigration policies. The per cent with a bachelor’s degree or higher is included because studies have found that a higher level of education increases the likelihood of inclusive immigration policy responses. Finally, the diversity index and change of the diversity index have been included because several scholars argue that unfamiliarity with ‘others’ increases the chances of a hostile response to immigrants. The diversity index measures the racial diversity of a community. It expresses the likelihood that any two persons chosen at random will have different racial backgrounds.¹

The second set of independent variables is measured on the city level. It can be expected that if a city and/or neighbouring municipalities have experienced conflict over day labourers, this impacts local policy responses. To measure conflict the variables ‘conflict’ (0 = no, 1 = yes) and ‘conflict in adjacent municipalities’ (0 = no, 1 = yes) have been included. In order to asses if institutional design affects local policy responses, the following variables have been included: type of elections (0 = at-large, 1 = district), form of government (0 = council–manager, 1 = mayor–council), position of the mayor (0 = elected, 1 = selected), are there Latino council members (0 = no, 1 = yes), and is there a Latino mayor (0 = no, 1 = yes).² Previous studies have found that the political preferences of residents affect policy responses. A variable is included that measures the per cent of the population voting for a Republican governor. Literature on local immigration policies argues that policies in other municipalities can affect local policy responses. A variable is included that measures if bordering municipalities have implemented restrictive day labourer policies (0 = no, 1 = yes), and a variable indicating if inclusive day labourer policies were implemented in adjacent municipalities (0 = no, 1 = yes).

Previous studies have found that policies and developments at other levels of government affect local policy responses to immigration. As day labourers are a local issue, we include anti-immigrant (0 = no, 1 = yes) and pro-immigrant policies (0 = no, 1 = yes) implemented at the county level. In addition to that, saliency of immigration at the federal level has found to be providing frames to local actors to push their immigration policy preferences (Hopkins, 2010). A variable is included that measures the saliency of immigration at the federal level as well as a variable that measures saliency of day labourers at the federal level.

The models have been fitted using R statistical software and the ‘glmer’ function of the ‘lme4’ package. A maximum likelihood approach (adaptive Gauss–Hermite quadrature) was used to fit the models.

RESPONSES TO DAY LABORERS

The focus of the statistical analyses and case studies is on cities with day labourer hiring sites. Local officials have a strong incentive to implement day labourer policies when day labourer activities generate concerns (Crotty, 2017; Rodriguez, 2008). Day labourer hiring sites are an important source of concern. In the period 1985–2020 no cities in California have implemented inclusive day labourer policies prior to having one or multiple hiring sites (Table 2). This can be explained by the fact that inclusive day labourer policies are often highly controversial as these policies are framed by opponents as supporting unauthorized immigration. If informal gatherings of day labourers are not considered to be an issue, local officials have no incentive to touch such a controversial subject.

Table 2. All municipalities in California by types of policy and hiring sites.

	No day labourer policy	Restrictive day labourer policy	Inclusive day labourer policy
No hiring site	1240	104	0
One hiring site	68	32	17
Multiple hiring sites	37	19	32
Total	1345	155	49

However, it is interesting to note that most restrictive day labourer policies are implemented in cities that have no hiring site (Table 2). The majority of policies in these cities without hiring site, curb solicitation of employment on roads, parking lots and/or private property. Further research is necessary to explain why these cities implemented these policies. Possible explanations can be spillover effects from neighbouring municipalities that have experienced conflict over day labourers and/or have implemented restrictive day labourer policies. In response to these developments, officials can decide to implement a restrictive policy to prohibit the move of day labourers to their municipalities. In addition, the establishment of certain businesses that might be attractive as hiring sites – such as large home improvement stores or rental companies – can be an incentive to implement a no-solicitation ordinance to get ahead of any future day labourer issues. It has to be said that restrictive day labourer policies in cities without a hiring site do not go further than no-solicitation ordinances. Restrictive day labourer policies in cities with one or multiple hiring sites are often more extensive. For example, these policies include increased police enforcement, a sit/lie ban, mandatory registration of employers who hire day labourers, no loitering and trespassing ordinances, and immigration enforcement.

Factors affecting local policy responses to day labourers

This section presents the results of the binary multilevel regression models (Table 3). The models include 2597 observations for 278 hiring sites in 116 municipalities in California between 1985 and 2015. A hiring site constitutes an observation only in the year(s) it was determined to be active. The first model explores the implementation of ‘no day labourer policy’ versus ‘some day labourer policy’. Conflict over day labourers in a city and/or in neighbouring municipalities, having a mayor–council system instead of a council–manager system, having a mayor of Latino descent, and saliency of day labourers at the federal level all foster the adoption of day labourer policies. However, having one or more council member(s) of Latino descent and saliency of immigration at the federal level decrease the likelihood of implementing some day labourer policy. Thus, it does not seem like Latino council members feel especially inclined to take up the issue of local day labourer policies. In addition, public debates on day labourers seem to spur the implementation of day labourer policies.

Table 3. Binary multilevel regression models.

	No versus some policy		No restrictive versus restrictive policy		No inclusive versus inclusive policy
Predictors	Coefficient	SE	Coefficient	SE	Coefficient	SE
Intercept	−2.16**	0.76	−3.08*	1.22	−1.69	0.97
1991–95	−0.92**	0.32	−1.01*	0.47	−0.75	0.39
1996–2000	−0.27	0.31	−0.83	0.49	0.14	0.38
2001–05	−2.24***	0.35	−0.97*	0.49	−2.55***	0.45
2006–10	−1.71***	0.37	−1.25*	0.56	−2.13***	0.47
2011–15	−1.94***	0.44	−2.50**	0.77	−2.08***	0.56
Tract level
Latino population	0.46	0.62	−0.91	1.07	1.01	0.76
Growth of the Latino population	−0.05	0.20	−0.08	0.29	0.16	0.26
Per cent owner-occupied housing	−0.08	0.46	0.28	0.83	−0.09	0.56
Per cent bachelor’s degree or higher	0.51	0.87	−2.37	1.71	1.36	1.03
Diversity index	0.40	0.57	−0.71	0.97	1.09	0.68
Change of diversity index	−0.09	0.32	−0.10	0.48	−0.34	0.43
City level
Conflict (yes)	1.68***	0.17	2.08***	0.24	1.18***	0.23
Conflict in an adjacent municipality	0.81***	0.18	0.08	0.28	1.25***	0.23
Type of elections (district)	−0.23	0.41	0.01	0.57	−0.73	0.57
Form of government (mayor–council)	1.13*	0.56	−0.55	0.88	1.79*	0.76
Mayor (selected)	−0.03	0.28	0.01	0.39	−0.05	0.39
Latino council members (yes)	−0.47*	0.23	−0.22	0.31	−0.94**	0.32
Latino mayor (yes)	0.83**	0.27	0.33	0.60	1.22***	0.33
Growth of the Latino population at city level	−0.64	0.46	−0.49	0.64	−1.12	0.62
Per cent voting Republican	0.14	0.97	1.71	1.40	−3.65**	1.41
Adjacent municipality: anti-policy	−0.67**	0.23	0.90*	0.36	−1.31***	0.29
Adjacent municipality: pro-policy	0.17	0.28	−1.04	0.66	0.55	0.31
County policy
County: anti-policy	−0.25	0.20	0.29	0.30	−0.73*	0.28
County: pro-policy	0.26	0.33	−0.24	0.53	0.28	0.41
Federal level
Saliency of immigration	−0.14*	0.06	0.13	0.09	−0.26**	0.08
Saliency of day labourers	0.19***	0.05	−0.00	0.07	0.27***	0.07
Random effects
σ^2	3.29		3.29		3.29
τ00	0.00Site		0.00Site		0.00Site
	0.52Place		1.08Place		0.92Place
	0.00County		0.00County		0.00County
N	278Site		278Site		278Site
	116Place		116Place		116Place
	21County		21County		21County
Observations	2597		2597		2597
Marginal R^2/conditional R^2	0.303/n.a.		0.344/n.a.		0.427/n.a.

Note: *p < 0.05; **p < 0.01; ***p < 0.001.

The second model examines the implementation of restrictive day labourer policies (no versus yes). Conflict over day labourers is an important factor that significantly increases the likelihood of adopting restrictive day labourer policies. In addition, restrictive day labourer policies in neighbouring municipalities also increase the chances of implementing restrictive day labourer policies. The other predictor variables that are included in the model are not significant. Thus, conflict over day labourers and restrictive day labourer policies in adjacent municipalities seem to contribute to the adoption of restrictive day labourer policies.

The third and final model explores the implementation of inclusive day labourer policies (no versus yes). Conflict over day labourers in a city and/or in neighbouring municipalities, having a mayor–council system instead of a council–manager system, having a mayor of Latino descent, and saliency of day labourers all significantly increase the chances that inclusive day labourer policies are implemented. Having one or more council members of Latino descent decreases the likelihood of implementation of inclusive day labourer policies. Contrary to the findings of the literature on minority representation, Latino day labourers do not seem to benefit from the presence of Latino city council members. An explanation for this finding can be the precarious status of day labourers. Day labourers are often stigmatized as unauthorized, Latino immigrant men who search for work in highly visible and public spaces. These characteristics turn day labourers into easy targets of polarization, which increases the chances of a hostile policy response. In addition, the unauthorized immigration status of many day labourers makes the group mostly ineligible to vote and that can make it unprofitable for council members to account for their interests, especially when important constituents (voting homeowners, businesses) are unhappy with the presence of day labourers.

With regard to political preferences, the findings are in line with previous studies that have found that voting Republican decreases the chances of inclusive immigration policies. When the per cent voting for a Republican governor increases, the likelihood of an inclusive day labourer policy decreases. The local policy responses of a city are also influenced by policy making in other cities and at the county level. Having one or more neighbouring municipalities that have implemented restrictive day labourer policies decrease the likelihood of implementation of inclusive day labourer policies. The same holds true at the county level; the presence of restrictive policies at the county level decrease the chances of inclusive day labourer policies at the city level. Finally, saliency of immigration at the federal level decreases the likelihood of implementation of inclusive day labourer policies.

It is interesting to note that the models suggest that in the case of day labourer policies, demographic factors – such as growth of the Latino population, rate of homeownership, and education – do not play a significant role. In addition, the form of municipal elections does not appear as a significant predictor in any of the three models. Thus, for day labourers, district elections do not seem to provide advantages over at-large elections. The most influential predictors of day labourer policies seem to be conflict over day labourers, and policies in other jurisdictions. This finding illustrates the importance of the makeup of the day labourer population. The marginalized status of day labourers makes them the target of local policies that are mainly fuelled by local conflict and neighbouring policy responses.

However, it is important to remember that conflict and policy responses in neighbouring municipalities interact with other factors, which can result in varying policy responses to day labourers. For instance, based on the model on inclusive day labourer policies it can be expected that the effect of conflict is mitigated if a city operates under a mayor–council system, because the likelihood of inclusive day labourer policies increases in mayor–council systems compared with council–manager systems. To illustrate the dynamic character of local policy responses to day labourers the next section discusses three case studies.

Local responses to day laborers in three cases

The cases of Mission Viejo, Pasadena and Baldwin Park illustrate how local responses to day labourers are shaped by a complex web of factors. The three municipalities have different local contexts, such as variation in demographics, politics and day labourer policies. This variation allows to explore how demographic factors, political preferences, history of place, policies in other jurisdictions, saliency of day labourers, and institutional design interactively impact local day labourer policies, resulting in a variety of local policy responses.

Mission Viejo

Mission Viejo is located in Orange County and had a population of 93,305 residents in 2010. The Latino population is rather small when compared with the other case studies, although Mission Viejo experienced the largest growth of the Latino population of all three cases. A large share of Mission Viejo’s residents are homeowners and a fair-sized proportion of the population has a high level of education. The diversity in Mission Viejo increased considerably between 1990 and 2010 (Table 4). Residents of Mission Viejo predominantly voted Republican. The city operates under a council–manager form of government, council members are elected at-large, and the mayor is appointed by the council. Mission Viejo did not have a mayor of Latino descent, but since the 2000s it had a member on the city council with Latino heritage. Together these demographic, political and institutional factors shape an environment in which restrictive day labourer policies can blossom.

Table 4. Demographic characteristics of case studies.

	1990	2000	2010
Mission Viejo
Population	72,820	93,102	93,305
Proportion Latino	13.0%	12.1%	17.0%
Owner-occupied housing	80.0%	81.4%	77.9%
Bachelor’s degree or higher	39.1%	41.2%	43.8%
Diversity index	0.27	0.38	0.49
Pasadena
Population	131,591	133,936	137,122
Proportion Latino	27.3%	33.4%	33.7%
Owner-occupied housing	46.3%	45.8%	45.0%
Bachelor’s degree or higher	36.3%	41.3%	45.7%
Diversity index	0.67	0.69	0.71
Baldwin Park
Population	69,330	75,837	75,390
Proportion Latino	70.8%	78.7%	80.1%
Owner-occupied housing	60.1%	61.0%	60.2%
Bachelor’s degree or higher	9.7%	9.0%	11.6%
Diversity index	0.46	0.41	0.34

However, a growth of the Latino population in a predominantly Republican city that largely consists of homeowners does not in itself result in restrictive day labourer policies. There has to be an incentive to implement day labourer policies. In the case of Mission Viejo there were two important drivers that spurred the city to implement restrictive day labourer policies. The bordering city of Lake Forest implemented a no-trespassing ordinance to address day labourer issues. In response to the ordinance, day labourers moved to a shopping centre area in Mission Viejo. Business owners located in the vicinity of the shopping centre and a group of activists who had pressured the Lake Forest city council to restrict day labourer activities pushed the Mission Viejo city council to address the presence of day labourers in Mission Viejo. The city council of Mission Viejo decided to implement a no-trespassing ordinance resembling the ordinance in Lake Forest. Thus, the implementation of a restrictive day labourer policy in an adjacent municipality and complaints by business owners and activists moved the Mission Viejo city council to implement a restrictive day labourer policy. These findings are in line with the results from the binary multilevel regression model on restrictive day labourer policies.

As the mayor of Mission Viejo is appointed by the city council, this makes the mayor more a part of the council than a detached position. Not surprisingly then, council members and the mayor played similar roles during debates on day labourer issues. Day labourers were framed as a threat to the residents of Mission Viejo by elected officials. For instance, during a council meeting a council member stated the following:

I drive down the street all the time and I know I don’t want to stop in the shopping center, because of the, I just wouldn’t stop anywhere that there were ten or fifteen men standing around talking to each other in any language. As a woman I would not want to get out of the car.³

Another council member stressed the importance of protecting the interests of business owners in Mission Viejo:

I would like to suggest that we have the City Attorney and Police Chief sit down with some of these business owners that have been complaining. Because we already hear that they are incurring real costs because of this issue. So I think we should be concerned and if we want to be a pro-business city we should look out for the interests of businesses.⁴

The council members and mayor of Mission Viejo did not account for the interests of day labourers. Day labourers and their advocates were almost completely absent from the council discussions on day labourer issues. As mentioned previously, demographic and political factors shaped an environment in which restrictive day labourer policies can flourish. Combined with a council–manager system that promotes efficiency, and a mayor appointed by the council, there were few opportunities for day labourers and their advocates to promote the interests of day labourers. However, day labourer advocates were able to have some indirect impact on Mission Viejo’s day labourer policy. Some council members preferred the implementation of a no-solicitation ordinance, but they were limited by legal battles over no-solicitation ordinances in other jurisdictions. According to the City Attorney a no-solicitation ordinance would be unconstitutional: ‘Somebody standing there [sidewalk] and somebody else believing that their presence is menacing, has not been something that courts have been real prone to embrace.’⁵ Thus, although the interests of day labourers were of marginal importance to the city council of Mission Viejo, battles of day labourer advocates in other places had some spillover effect.

The case of Mission Viejo illustrates how the interaction of multiple factors impact local policy responses to day labourer issues. Demographic, political and institutional design characteristics created an environment that promotes the adoption of restrictive day labourer policies. When conflict over day labourers and policies in adjacent municipalities interacted with these factors, Mission Viejo’s city council implemented a restrictive day labourer policy. However, the restrictiveness of Mission Viejo’s policies was somewhat softened by successful legal battles of day labourer advocates in other jurisdictions.

Pasadena

Pasadena had a population of 137,122 residents in 2010 and is located in Los Angeles County. The share of the Latino population rose steadily over the last three decades, resulting in approximately one-third of the population being Latino in 2010. The city has more renters than homeowners, and a sizeable proportion of the residents has a high level of education. The diversity in Pasadena is rather high and remained quite stable (between 0.67 and 0.71). Residents of Pasadena mostly vote for Democratic candidates. The city has a council–manager system, with council members being elected by district. The mayor is elected directly by the residents. Since the mid-1990s, Pasadena had city council members of Latino descent, but the city did not have a Latino mayor in the period 1985–2020. The demographic and political factors in Pasadena created an environment for inclusive day labourer policies, as there were no major demographic changes and residents predominantly voted Democratic. District elections, according to the discussed literature, provide minorities with opportunities for representation and accountability to their interests, however the binary multilevel logistic regression models did not confirm these arguments. The case of Pasadena provides an opportunity to explore how district elections affect day labourer policies.

The hiring sites at which day labourers congregated in Pasadena were largely located in a single and predominantly Latino council district: district five. It is not surprising then that council members representing district five initiated the implementation of day labourer policies in Pasadena. Whereas the first council member to initiate a day labourer policy was white, the council member who succeeded him in 2001 was Latino. In response to complaints by residents and business owners about informal hiring sites, a formal hiring site was created. This policy aimed to address the concerns of residents and business owners, and to provide a formalized space for day labourers to find a job. Thus, as could be expected based on the interaction of demographic, political and institutional factors in Pasadena, the city implemented an inclusive day labourer policy.

Although the Pasadena Job Center alleviated some of the issues associated with day labourers, there were still complaints from residents and business owners. Some day labourers refused to use the centre, and instead continued to solicit work on the streets. In response, the council member representing district five started to promote restrictive day labourer policies – a no-solicitation ordinance and increased police enforcement – to address the concerns of constituents. The council member acknowledged the needs and rights of day labourers, but also emphasized that his constituents should be protected:

We all agree that this is an issue we are all dealing with. We don’t want to limit anyone’s ability to seek employment. To the contrary, we want to provide an adequate location for them to seek employment and we want to help improve people to come hire them. But we have to protect the neighbourhood.⁶

Another council member stressed the importance of a feasible solution:

Because what we are here for really, we are not here to punish anybody. We are not here to do anything other than find a solution to the problem. Which is really you know get rid of the impact on the neighbours and provide the employment opportunity for the people who want to work.⁷

So the interaction of demographic, political and institutional factors provided possibilities for day labourers in Pasadena to push for their interests. In response, the Pasadena city council tried to balance the interests of residents and business owners with those of day labourers. However, when residents and business owners remained unhappy about the day labourer situation, their interests prevailed over the interests of day labourers. While acknowledging the needs and rights of day labourers, restrictive day labourer measures were considered to be a justified solution to appease constituents.

Literature on the representation of minorities and accountability to minority interests maintains that a strong mayor can provide access and protection to minorities. The binary multilevel regression model on inclusive day labourer policies supported this argument, as inclusive day labourer policies are more likely to occur in a mayor–council system. In Pasadena the mayor is elected, which gives the mayor of Pasadena a somewhat more powerful role than the appointed mayor in Mission Viejo. However, Pasadena operates under a council–manager system, which means that the mayor does not has as much power to shape policies as mayors in mayor–council systems have. In Pasadena the mayor took up the role of chairman, guiding the city council discussions on day labourer issues in district five. The mayor did not initiate any day labourer policies. Thus, for day labourers and their advocates the mayor of Pasadena did not provide much access and protection.

The case of Pasadena illustrates that the interaction of different factors results in dynamic policy responses to day labourer issues. Demographic, political, and institutional characteristics shaped a favourable environment for day labourers, which led to accountability to the interests of day labourers. However, the case of Pasadena also reveals how the marginalized status of day labourers affects local policy responses. Elected officials are willing to take the interests of day labourers into account, but not at the expense of the interests of other constituents, in particular homeowners and businesses. When these constituents deem day labourer activities to be a nuisance, their interests seem to outweigh the need of day labourers to find a job. In such instances, even a sympathetic Latino council person representing the district can advocate for policies to criminalize day labourer activities.

Baldwin Park

Baldwin Park is situated in Los Angeles County and had a population of 75,390 residents in 2010. A large majority of the population in Baldwin Park is Latino. The diversity in the city is rather low, and it has decreased over time. A majority of the residents are homeowner and about one-tenth of the population has a high level of education. Residents of Baldwin Park mostly vote Democratic. The mayor and city council are elected at-large. The city operates under a council–manager form of government. Since the 1990s the city council predominantly consists of members of Latino descent, and since 1999 the mayor of Baldwin Park has been of Latino descent as well. Taken together some of these factors suggest that inclusive day labourer policies would be most likely, as residents are familiar with Latinos and mainly vote liberal. However, some other factors indicate that restrictive day labourer policies could thrive as well; a fair share of homeowners, a relatively small share of the population is highly educated, the city functions under a council–manager system and the council is elected at large.

In this context, complaints by business owners and residents urged the Baldwin Park city council to develop day labourer policies. Council discussions on day labourer issues were fiery, and all council members and the mayor had strong personal policy preferences. Some council members and the mayor emphasized the need to protect the safety and quality of life in Baldwin Park. For instance, a council member argues:

I am bringing this to the council so that we can hopefully come up with some sort of solution to this. Now it has gotten to a point that it has become a safety issue for women, and it is a safety issue for any visitor that comes to the city of Baldwin Park in terms of their property broken into.⁸

Another council member states:

I argue that if you work hard to open a business, you should be able to run it without having your business being undermined by the presence of labourers or vendors. This is not a race issue, this is not about criminalizing Latinos. It is about maintaining standards and approve the quality of life for residents and business owners alike.⁹

Together with the mayor, these council members supported a no-solicitation ordinance to address the concerns of residents and business owners.

However, other council members fiercely opposed restrictive day labourer policies. According to one council member a no-solicitation ordinance would criminalize hardworking people:

I voted for it because I am against having this ordinance, because I didn’t believe that it was right. In the way it began to evolve it was to criminalize an individual because he is asking for a job or asking for work.  …  These individuals are not criminals.¹⁰

These council members favoured the creation of a day labourer centre to address the issues brought forward by residents and business owners and accused the supporters of restrictive policies of being anti-Latino. Despite the fierce resistance of these council members, the other council members with support of the mayor implemented a no-solicitation ordinance. Thus, even though Baldwin Park residents and officials were familiar with Latino immigrants and voted mainly liberal, a majority expressed hostility to the presence of day labourers. Being a predominantly Latino city, all elected officials were keen to stress that immigration was not the issue at stake, instead the main issue for those supporting restrictive policies was the threat day labourers posed to the quality of life in Baldwin Park.

Being an all-Latino city council that proclaimed sympathy for immigrant rights, created opportunities for day labourers and their advocates to push the Baldwin Park city council to take the interests of day labourers into account. However, only a minority of the city council members was sympathetic to the advocacy of day labourer representatives and immigrant right organizations. Even though the binary multilevel regression model shows that a Latino mayor increases the likelihood of inclusive day labourer policies, the Latino mayor of Baldwin Park responded quite negatively to the advocacy by immigrant rights organizations. During a council meeting the mayor stated:

One thing I have to say, the most thing that I hate when individuals come here, sent letters, representing an organization. It is easy for them because they are coming from their plush upper-class communities to the city of Baldwin Park telling us what is best for Baldwin Park. When in fact they don’t have those issues, nor would they tolerate those in their communities. So don’t come to Baldwin Park giving us your personal agenda, that is unacceptable.¹¹

Being unable to win a majority on the city council or to get support from the mayor, the Mexican American Legal Defense and Educational Fund sued Baldwin Park over its no-solicitation ordinance. The lawsuit forced the city council to repeal the ordinance. In addition, day labourers organized themselves in order to protect their interests. After being forced to repeal its solicitation ordinance, the Baldwin Park mayor and city council decided to set up a committee that would work together with the representatives of the day labourers to come up a viable solution. This resulted in the creation of a canopy at Home Depot property to provide the day labourers with a space to look for work while also addressing the complaints of residents and business owners.

The case of Baldwin Park illustrates the dynamic character of local policy responses and shows that having Latino representation in municipal government and a Latino mayor do not necessarily benefit day labourers. The interaction of several demographic, political and institutional factors in Baldwin Park creates an environment in which both opponents and supporters of day labourers can push their policy preferences. This results in local battles and policy responses that move between restrictive and inclusive. The emphasis on ‘quality of life’ arguments by supporters of restrictive policies, stresses the importance of the marginalized status of day labourers. Although elected officials claim to be sympathetic to the plight of immigrants, they define the conduct of day labourers as not meeting the standards of Baldwin Park. It has to be said that council members and the mayor of Baldwin Park were more sensitive to the interests of day labourers, than the council and mayor of Mission Viejo, and this sensitivity can be partially attributed to their own ethnicity and the ethnicity of most of their constituents. But just as was the case with Pasadena, the interests of long-time residents and business owners seemed to prevail over the interests of day labourers, in spite of expressions of ethnic solidarity.

CONCLUSIONS

Local policy responses to day labourer issues vary, and the involvement of local government officials seems to be an important piece of the puzzle to understand this variation. Council members and mayors possess the authority to introduce and adopt policies, and their involvement provides an important contribution to the direction of local day labourer policies. This paper shows that local policy responses are impacted by a complex web of factors. Mayors and council members are part of this web of factors in which they exert influence on policy responses. At the same time, council members and mayors are also influenced by the other factors that make up the web.

Demographic and political factors – such as growth of the Latino population, the proportion of homeowners and voter preferences – shape an environment in which restrictive and/or inclusive day labourer policies can thrive. This paper adds to these factors the impact of certain aspects of institutional design. The form of government (mayor–council versus council–manager) affects the implementation of inclusive day labourer policies. In cities that operate under a mayor–council system the likelihood that inclusive day labourer policies are implemented increases when compared with cities with council–manager systems. A reason for this can be the focus on efficiency in council–manager systems, which makes policies aimed at minority interests unprofitable as these do not meet the requirements of efficient management. Thus, day labourers seem to fare better in mayor–council systems.

However, the identified factors do not set the outcome of local policy responses in stone. Local policy responses are dynamic and impacted by the continuous interaction of multiple factors. This can result in policy responses that contrast with expectations based on demographic and political characteristics of a municipality. For instance, the case of Baldwin Park illustrated that an absence of major demographic change, familiarity with immigration and Latino representation do not guarantee the implementation of inclusive day labourer policies. The marginalized position of day labourers turned out to be an important mediating factor, pushing local elected officials in the direction of restrictive day labourer policies.

Our findings are important because they elevate the importance of the dynamic character of local policy responses to immigration. Much of the literature focuses on the impact of demographic, political, and other factors on local immigration policies without taking into consideration the importance of the interaction between these factors. Seemingly similar municipalities can take opposing policy decisions on immigration issues, because the interaction of local factors lead to different outcomes. In addition, even if municipalities implement policies that are in line with expectations based on local characteristics this is not the end of the story. New factors can emerge or the relation between factors can change, which results in dynamic policies that change over time.

Our findings also suggest that it is important to take the role of institutional design into account as a factor shaping immigration policies. In the case of day labourers, the form of government (council–mayor versus council–manager) was found to be significantly affecting local policy responses. No evidence was found that the form of municipal elections (at-large versus district) significantly impacts day labourer policies. Further research is necessary to determine to what extent institutional design affects other immigration policies and other immigrant groups. Especially in the light of the push of advocacy organizations for district elections to enlarge the representation of and accountability to minority interests.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1 The diversity index is based on the USA Today index of ethnic diversity (Meyer & McIntosh, 1992). It takes the proportion of an ethnic group squared, and then sums up all the scores for each ethnic group. The diversity index is 1 – all summed scores. Thus:

Diversity index = 1 - (non-Hispanic White²) + (non-Hispanic Black²) + (non-Hispanic American Indian²) + (non-Hispanic Asian²) + (non-Hispanic other²) + (Hispanic²).

2 The Latino descent of elected officials has been defined using two sources. First, the Censuses of Government of 1987 and 1992 have been used. In these surveys municipalities report the ethnicity of their elected officials. Second, a list of the names of council members and mayors in cities in California has been created from 1995 onwards. To determine the ethnicity of elected officials, the package ‘wru’ was deployed in R. This package allows one to compute the probability that a person is from a given ethnicity by using the surname and county in which the official is elected (Imai & Khanna, 2016).

3 Mission Viejo City Council meeting, 17 March 2008, https://missionviejo.granicus.com/MediaPlayer.php?view_id=12&clip_id=149/.

4 See note 3.

5 Mission Viejo City Council meeting, 5 February 2007, https://missionviejo.granicus.com/MediaPlayer.php?view_id=12&clip_id=38/.

6 Pasadena City Council meeting, 13 September 2004, https://pasadena.granicus.com/MediaPlayer.php?view_id=25&clip_id=40/.

7 See note 6.

8 Baldwin Park City Council meeting, 6 December 2006, http://baldwinpark.granicus.com/MediaPlayer.php?view_id=10&clip_id=625/.

9 Baldwin Park City Council meeting, 21 January 2009, http://baldwinpark.granicus.com/MediaPlayer.php?view_id=10&clip_id=1065/.

10 Baldwin Park City Council meeting, 15 August 2007, http://baldwinpark.granicus.com/MediaPlayer.php?view_id=10&clip_id=789/.

11 Baldwin Park City Council meeting, 2 May 2007, http://baldwinpark.granicus.com/MediaPlayer.php?view_id=10&clip_id=729/.