![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
How does the coordination strategy of interest groups change during a crisis? Shocks reduce an economy's resource pool and increase the competition for what is available. Interest groups participate in the policy-making process by lobbying legislators. In times of crisis, we argue, interest groups lobbying Congress coordinate in cohesive industry-wide communities led by key actors. Rather than lobbying for narrow policy privileges, interest groups seek to support legislation that is most beneficial to their community. To study the cooperative behaviour of interest groups we build an original network dataset based on committee participation in the Ecuadorian Congress between 1996 and 2015. We present evidence of increasingly homophilic industry networks in times of crisis, with umbrella organisations taking the role of hubs. We find that ‘lone wolf’ strategies, prevalent during an economic expansion, are less prevalent during a crisis.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Analizing several European countries, Hanegraaff and Pritoni (Citation2019) show that interest groups might also join coalitions as a mechanism of organisational survival.
2 According to Box-Steffensmeier et al. (Citation2019), strength is not always related to numbers, even though it can be, especially at the early stages of the policy-creation process.
3 While not directly studying group membership, Klüver (Citation2013) shows that in the European context lobbying coalitions are formed by members of similar industries. Likewise, Kingstone (Citation1999/2010) and Murillo (Citation2001) present a similar pattern for business and labour interest in Latin America, respectively. In the empirical section, we show that this is also the case for the Ecuadorian case.
4 See Keller (Citation2018), Woll (Citation2012), and Quaglia (Citation2010) for the effect of the 2008 financial crisis on European interest group strategies; Eilstrup-Sangiovanni (Citation2019) for the effect resource changes have on transnational advocacy group strategies; and Mahoney (2007) for the effect contextual factors have on interest group success and strategies in the U.S. and Europe.
5 Interest groups will have individual identities linked to the members they represent. In addition to these identities, there are shared identities among interest groups. For example, two interest groups representing exporters will share the identity (and, thus, interests and preferences) of the broader exporter community. Members of a community with a shared identities are the ‘in-group’; interest groups identifying with ‘other’ communities would be part of the ‘out-group’.
6 Legislators have limited time and resources. Hall and Deardorff (Citation2006) argue that lobbying works as a form of ‘legislative subsidy’, where interest groups provide information and labour to resource-constrained legislators. While we do not address this argument directly, our conclusions are not at odds with this view. However, we focus on the competition for resources.
7 Lowering taxes is also a mechanism to assign resources to firms. Furthermore, tax codes are often complex and nuanced, allowing for taxes, as well as subsidies, to be highly specific (Kim, Citation2017).
8 There are important differences/similarities between interest groups cosigning amicus briefs and, for example, interest groups cosigning letters sent directly to committee chairs debating a bill (one of the strategies used in this research to identify coordination). One similarity, for example, is the purposive and coordinated action entailed in cosigning both documents. We expand on the underlying decision of interest groups signing letters in the empirical section.
9 The financial crisis was endogenous, while the decision to dollarise the economy was not endogenous. The decision to dollarise the economy amid a financial crisis answer, not in small part, to the political clout of import-oriented firms who would benefit from a change in currency. However, the push to dollarise the economy was supported by a large part of the export-oriented sector, as they welcomed the currency stability that would come with the adoption of the dollar.
10 World Bank Databank: https://data.worldbank.org/indicator/NY.GDP.MKTP.CD?locations=EC. Accessed May 2020.
11 In Ecuador, the law stipulates that interest groups can formally lobby for or against a bill only once the bill has reached a committee. Each committee in the Ecuadorian Congress is then required to produce a report with two key components. One is a detailed explanation of all changes made to the bill initiative during committee debates. The other is a list of all interest groups that participated in meetings where the bill was discussed. We extract interest group names from each report to build a data set of interest group participation. These reports are held at the Archivo-Biblioteca of the Ecuadorian Congress. We thank the staff at the Archivo-Biblioteca of the Ecuadorian Congress for providing the raw versions of these reports.
12 Similar to other legislatures, in the Ecuadorian Congress less than 60% of bills reach committee (or are debated in committee). The allocation of bills to committees is a political rather than a technical process. Furthermore, despite the legal obligation for all bills to be assigned to a committee, committee chairs have the discretionary power to ‘cajonear’ – put in a drawer – a bill and not debate it at all.
13 We use the term ‘industry’ as employed by the Standard Industrial Classification (SIC), who include, for example, membership organisations such as political and religious groups. Furthermore, given the Ecuadorian context, the particular case studied in this paper, we add ‘ethnic associations’ to the categories. Thus, ‘industry’ does not necessarily mean a business-oriented organisation or lobbying group.
14 The SIC for Industry Groups for Membership Associations also includes political organisations such as Political Action Committees (PAC). In Ecuador, however, these types of organisations are prohibited by law.
15 We used the Fruchterman and Reingold algorithm to plot the networks as it places connected nodes closer to each other.
16 Box-Steffensmeier and Christenson (Citation2014) coined the term ‘lone wolves’ to refer to interest groups who pursue this individual lobbying strategy.
17 More disperse interest group activity makes the evidence we present in favour of our theoretical argument stronger. If anything, having more groups lobbying more bills makes it more difficult to find higher density levels in the network.
18 The Guayaquil-based Chambers suggested headquarters of the Council be in Guayaquil, while the Quito-based Chambers argued for a restructuring of the customs authority, which included its move to the capital, Quito.
19 This was in part a consequence on the economic conditions that forced the state to push fiscal and market reforms. The second time around, after the crisis, the bill reached committee with more political support and urgency, with IGs lobbying together to obtain maximum payoffs in a context of low state resources.
20 Interview with Patricio Alarcón, Quito, October 10, 2019.
21 This is not to say that the competition of resources was the only reason interest groups change their strategy. Roberto Aspiazu, the president of the Cámara de Empresarios, suggested that their strategy was also dependent on the power of the executive and the relation between the executive and the legislature. However, these effects are unlikely to be captured by other data. For one, there was no change in executive or legislative cohort between 2011 and 2015, the period where the the oil crisis occurred (and the Cámara de Empresarios appears as a node in our network). This also remains constant before and after the financial crisis of 1998.
22 The degree centrality was estimated from the sub-network of interest groups that were present before and after each crisis. As robustness checks, we estimate the degree centrality of each group from the entire network and rank each group accordingly (see Appendix 1). The results remain unchanged.
23 The E/I Index is estimated by , where EL is the number external links and IL is the number internal links.
24 Professional membership organizations only participated individually before the crisis.
25 A similar characteristic is observed by Box-Steffensmeier and Christenson (Citation2015) in U.S. interest groups networks.
26 Freeman's (1979) general formula for centralisation is , where
is the max centrality of network D.
27 Community detection methods (e.g. leading-eigenvector, walktrap) often rely on some heuristic to the maximisation of modularity within a network. Modularity captures the extent to which communities in a network have denser or more sparse intra- and extra-community ties (Newman, Citation2006; Newman & Girvan, Citation2004). There are no substantive differences, whether statistical or theoretical, in the results of different community detection methods. However, for our network, the partitions obtained from the leading-eigenvector community detection algorithm yields the highest modularity values.
28 We conduct further robustness checks in Appendix 2. We evaluate interest groups entering and exiting the network, as well as a comparison of the behaviour of only surviving interest groups. The results hold for different specifications.
Additional information
Notes on contributors
Joan C. Timoneda
Joan C. Timoneda is assistant professor of political science at Purdue University. His research focuses on democracy and democratization, authoritarianism, and big data methods.
Sebastián Vallejo Vera
Sebastián Vallejo Vera is a Post-Doctoral Fellow at the Department of Political Science, University of Houston, USA. His research focuses on legislative politics, gender, and racial identity in Latin America.