Examples of integrating research

To help highlight some of the challenges of combining population-level and individual-level interest group research, and hopefully reveal the benefits integration might yield, I will pick on my own work on group competition and multi-venue lobbying.

Interest group competition

I define interest group competition as the degree to which one organization's policy goal fulfilling its members' collective interest is perceived as coming at the expense of the collective interests of other groups' members (Holyoke 2011). In practice, the level of competition is this degree of interest conflict weighted by the amount of advocacy resources each organized interest brings to a fight (Holyoke 2009). For Gray and Lowery, however, competition primarily occurs between groups with similar interests. They see group competition as not so much about influencing policy as organizational survival, acquiring more resources necessary to survive than other groups. This means competition between interest groups is in the same interest-niche. These conflicting definitions only arise, though, because we consider group competition and how lobbyists make strategic decisions vis-a-vis each other at different levels of analysis, embedded in different populations. I explore competition in the issue- niche, while Gray and Lowery largely focus on the interest-niche.

More precisely, Gray and Lowery argue that competition is a consequence of group executives trying to maintain and grow their organizations. Interest groups compete with each other to define who they represent and thereby attract members. Ultimately their goal is to eliminate the competition by attracting to their membership rolls nearly all of the people who could possibly be represented by their organization. Once it dominates the interest-niche, the group's lobbyist finds his or her credibility and legitimacy in the political arena greatly enhanced because he or she speaks on behalf of an entire segment of the population. But competing with other groups in the same niche is time consuming and expensive, so, drawing on Browne's (1990) work, Gray and Lowery find that in order to realize the political benefits of niche domination, group leaders will often narrow the niches they are trying to dominate to reduce competition. At the same time, this reduces the number of people they can potentially represent and the resources they can amass. Group leaders must therefore be careful, a viable interest-niche can only be so small if it is to sustain the group, but too large means too much competition with other organizations.

I approach competition as a state of conflicting interests; greater differences in interests leading to a greater likelihood that two or more interest groups will fight each other as they lobby for or against proposed policy on the government's agenda. Since I focus on conflict or cooperation between groups with differing interests instead of the need to gain and secure resources, the group population in my work is at the issue-niche level, not the interest-niche level. In my work, lawmakers are not so much an audience to be impressed by whichever group dominates a niche as they are a pressure on lobbyists to overcome differences in interests to form coalitions (or alternatively pressure lobbyists to fight other interest groups). How to integrate my approach to competition at the issue-niche level with Gray and Lowery's at the interest-niche level perhaps begins with this thought: interest groups jockey for preeminence in their interest-niche, but those same groups may come together to stand against their common enemies (groups with much different interests) in the issue-niche.

Think about interest groups representing the banking industry. The American Bankers Association (ABA) represents all banks big and small, Financial Services Roundtable represents the biggest of the big, Independent Community Bankers Association (ICBA) represents small banks, and Mortgage Bankers Association (MBA) only represents banks specializing in home loans. To a large extent the common bonds defining their members' interests, and therefore defining their potential memberships, overlap. They are an interest-niche. And they compete to dominate all of it, or parse it into smaller sub-interest-niches. ICBA makes it an almost daily goal to point out how ABA is controlled by the likes of Citibank and Bank of America with no regard for community banks. MBA argues that banks specializing in mortgage lending also have little in common with the great banks, even though institutions like Citibank do considerable home lending. The Roundtable argues that by making small banks feel included, fighting off encroachments from ICBA and MBA, ABA cannot represent the interests of multi-national banks. ABA just wishes these others would go away, claiming it is perfectly capable of representing everyone and that banks big and small have mostly common interests anyway when it comes to public policy.

These groups compete to dominate the banking interest-niche, just as Gray and Lowery predict, differentiating themselves from each other as they lobby. ICBA seeks exemptions for banks with assets of less than $250 billion from capitalization requirements, thus reducing their regulatory burden and freeing more capital for home and business lending. It also pursues its decades-old effort to strip the credit union industry of its tax exemptions.3 These are not big concerns for larger banks, so ABA's apparent reticence to lobby on them becomes ammunition for ICBA in its efforts to steal away small banks. This fighting to dominate a large, powerful banking interest-niche or parse it into a series of smaller interest-niches is low-level competition. When it comes to actually lobbying issues on the government's agenda pushed by interest groups from other interest-niches, such as efforts by consumer protection interest groups to enact and implement the Dodd-Frank Act regulating all kinds of banking activities, banking interest groups big and small are mostly on the same page. How could they not be?

This is one way Gray and Lowery's work informs and, hopefully, improves my own. Although the dependent variable in my original paper on competition is a dyad capturing how lobbyists for two groups react to each other, it is in the issue-niche context (Holyoke 2009). Issue problems emerging on the government's agenda, and policies proposed to address them, tend to embrace several interest-niches. The issue- niche is therefore at a higher level than, and encapsulates, the interest- niches. The degree of difference between the interests of any pair of groups can be larger at the issue-niche level than at the interest-niche level because groups from different interest-niches get paired together. If the likelihood that competition between groups will erupt into conflict rather than cooperation in the issue-niche is at least partially a function of differences in interests, then conflict is more likely to occur and to be more intense between groups from two different interest-niches than two from the same interest-niche.

This integration suggests that interest groups are less likely to fight (and more likely to cooperate) with other groups in their own interest- niche when there is an issue at the issue-niche level threatening most of them. Interest-niche rivals will stand shoulder to shoulder against common foes from other issue-niches, fighting each other at the issue-niche level. Without a threat from outside the interest-niche, these groups are more likely to fight with each other. My original analysis could be re-done with a new, nominal three-category dependent variable capturing whether two observed group lobbyists chose to cooperate, fight, or ignore each other. One crucial independent variable would be a continuous or ordinal measure of the degree of difference between the two groups' interests, with some cut-point indicating whether or not the two groups are too different to be in the same interest-niche. Another crucial variable would be a binary indicator of whether or not there is a policy at the issue-niche level potentially affecting both lobbyists' members. This threat/no-threat condition would affect expectations of whether the two lobbyists would fight, cooperate, or ignore each other given the degree of differences in interests between them, yielding the hypotheses summarized in Table 5.1. If the issue-niche threat is present, then the analysis uses variables regarding the issue-niche, the interest- niche(s), and characteristics unique to each group in the observed pair in a hierarchical statistical model. If the condition is not present, then only interest-niche and individual group variables are used.

Table 5.1 Lobbyist predicted responses given variation in interests and issues affecting these interests

Degree of interest difference between two observed groups

There is no issue affecting many groups

There is an issue affecting many groups

Small (groups are probably in the same interest-niche)

High (groups are

probably in different interest-niches)

Prob. of fighting is high Prob. of cooperating is low Prob. of ignoring is high

Prob. of fighting is low Prob. of cooperating is low Prob. of ignoring is high

Prob. of fighting is low Prob. of cooperating is high Prob. of ignoring is low

Prob. of fighting is high Prob. of cooperating is low Prob. of ignoring is low

Finally, assuming, as Gray and Lowery do, that a sub-population, like an issue-niche or an interest-niche, is greater than the sum of its parts, at least some variables from these niches must capture the density of group populations at these levels. Presumably these variables would be counts of the number of interest groups in the interest-niche containing one or both of the lobbyists in the observed pair, and then a count of the groups in the issue-niche population if there is an issue on the government agenda affecting the lobbyists in the relevant interest-niches. Employing aggregated data at the sub-population level to estimate a two-lobbyist dyad also means we must be very careful of running afoul of the ecological inference problem.

< Prev   CONTENTS   Source   Next >