The government picks its battles – ceding ground with interest groups in exchange for wins on unrelated issues

Alastair Langtry, PhD economics researcher at the University of Cambridge, “Inside the West Wing: Lobbying as a contest,” Journal of Public Economics: Volume 231, March 2024, 105068 https://www.sciencedirect.com/science/article/pii/S0047272724000045

Governments make many different policy decisions. In doing so, they often face pressures from different special interest groups who each care about one particular policy. These groups lobby for their special interest, which is often not aligned with overall welfare. Faced with many groups lobbying for many different policies, a government may not have the resources to get its way on every policy decision. So in practice, a government may ‘pick its battles’ — specifically, seek to do deals with some special interest groups to avoid a fight. However, existing models of lobbying struggle to engage with this intuition because they do not allow the government to do deals to avoid fights. We assume the government has limited resources. This gives rise to the possibility that some special interest groups get their way. Crucially, we also allow the government to cede a policy decision to a special interest group, and for special interest groups to make transfers to the government. Our key contribution is to show that this happens, and to explain why: in equilibrium, the government will not try to fight all the special interest groups but will pick its battles. Specifically, it either engages in a contest and wins with certainty, or it does a deal where it takes a contribution in exchange for ceding the contest to the special interest group. Note that there are three key components of a deal in our model: (1) a special interest group gives some resources, (2) which the government uses either directly or indirectly to win other contests, (3) in exchange for ceding the contest the special interest group cares about. As a motivating example, consider a deal done between the Obama administration and the pharmaceutical lobby to secure passage of the Affordable Care Act (‘Obamacare’). In it, the Obama administration dropped provisions allowing for the re-importation of prescription medicines at lower prices (Baker, 2012), and the pharmaceutical industry agreed to provide $80 billion to help fund expanded insurance coverage (Baker, 2012) and spent millions of dollars on pro-Obamacare advertising campaigns (Murray, 2009). In a complex bill like Obamacare, with many provisions each affecting different interest groups, we view each provision as its own policy. So the Obama administration ceded a contest with the pharmaceutical industry in order to win others (and secure passage of Obamacare in some form). But to tie this properly to our intuition, it must be that: (1) the Obama administration took (or was able to control) some specific resources from the pharmaceutical lobby and (2) either used them directly to help win different legislative battles, or used them to free up other resources that could then be used to win legislative battles. Finally, (3) this must be in exchange for the Obama administration ceding some contest with the pharmaceutical lobby. While we might expect that these deals between special interest groups and the government are typically kept behind closed doors, a leaked cache of private emails from the Obama administration provide direct evidence regarding the advertising spending. First, they show that PhRMA (the pharmaceutical industry association) was willing to spend $150 million on pro-Obamacare advertising through two groups it created, called Health Economy Now and Americans for Stable Quality Care (The Committee on Energy and Commerce, 2012). So in this case, the ‘contribution’ the special interest group made was in the form of advertising spending. They also show that this spending was directed and controlled by the Obama administration. This was through meetings and emails between Obama administration officials and representatives of the pharmaceutical lobby to agree the content and placement of the adverts. One leaked email puts this particularly bluntly: ‘‘Rahm [Emmanuel, Chief of Staff to President Barack Obama] asked for Harry and Louise ads thru third party. We’ve already contacted the agent’’ (email from Bryant Hall, a lobbyist for PhRMA to Jeffrey Forbes, a political strategist, dated 7 July 2009). ‘Harry and Louise’ were a fictional couple from a series of adverts attacking then-President Bill Clinton’s proposed healthcare reforms in 1993 and 1994 (Clinton Digital Library, 2023). These ads were seen as playing a critical role in the failure of the Clinton healthcare reforms (West et al., 1996; Brodie, 2001), giving Harry and Louise a special connection to healthcare reform. The adverts aired one week after the email (The Committee on Energy and Commerce, 2012). Second, they suggest this spending was aimed at securing passage of Obamacare. While direct evidence in the leaked emails is lacking, it is implicit throughout that the purpose of running pro-Obamacare ads was to help pass Obamacare. The specific request for ‘‘Harry and Louise ads’’ is also suggestive, given their particular connection to healthcare reform. Finally, the emails show that this was in exchange for controls on pharmaceutical drug prices not becoming law. An email from Democrat strategist Steve McMahon to PhRMA lobbyist Bryant Hall (dated 3 June 2009, The Committee on Energy and Commerce (2012)) acknowledged, ‘‘I told them there were bright lines, but PhRMA was serious about reform. We agreed that we would talk on a regular basis about what they wanted and what we might be able to do. I also told them if the lines were violated the money would go elsewhere and ‘that wouldn’t be good’’’. This understanding of a quid pro quo is critical to our model. Note that our motivating example looks at a case where the government uses resources from a deal directly to fight other interest groups. It is easiest to see the mechanism in action in this direct case. But the model also allows indirect use of the resources — the government can use them to free up political capital from something unrelated, and use that other political capital in contests. We also generate precise predictions of which groups a government cedes contests to. If the special interest groups’ preferred policies are not too harmful, then the government concedes all contests. It prefers to acquire as much political capital as possible by doing deals, as we assume that having political capital brings some benefit to the government. On the other hand, if special interest groups’ preferred policies are sufficiently harmful, then the government cedes exactly the number of contests needed to allow it to win all remaining contests with certainty. And it cedes to interest groups whose preferred policy outcome is least harmful to society. In stark contrast, if the government cannot do deals, it adopts a ‘scattergun’ approach and spreads its resources across all possible contests, in proportion to the importance of each contest. The model explains why even governments that appear weak may not lose many legislative fights. It is because a weak government chooses not to engage in many fights. This highlights how the success a government has depends on the ambition of its agenda — i.e. how many fights it picks. Further, it shows that the ability to make deals dramatically reduces the amount of lobbying special interest groups do directly. However, special interest groups make contributions to the government — which increases the amount of political capital the government spends on contests. Unsurprisingly, a government that starts with less political capital takes more contributions from special interest groups. But it also spends less on contests. Related Literature. This paper relates to two strands of literature; lobbying and conflict. Our key contribution is to allow the government to make deals with special interest groups; ceding on some policies in exchange for resources that can then be used to help secure other policies the government wants. This adds to the lobbying literature by providing a stylised way of capturing the back-room deals that governments might do with special interest groups. Most of the existing literature on lobbying uses either information or agency issues as the core friction (see Bombardini and Trebbi (2020) for a recent review). In the former case, largely following Crawford and Sobel (1982), special interest groups have better information than the government and affect what the government wants to do through strategic communication. See for example, Esteban and Ray (2000), Bennedsen and Feldmann (2002) and Roberti (2014). The latter, largely following Grossman and Helpman (1994), models agency issues as the government caring both about social welfare and about direct payoffs to itself. Special interest groups influence what the government wants to do by offering money. Included in this second strand is work that models interactions between different interest groups as a conflict (see for example Esteban and Ray (1999) and Kang (2016)). As in our paper, contest success functions are used to model the conflict, but the government is not a party in the contest. A notable exception is Franke and Öztürk (2015). While theirs is a more general study of conflict on a network (including our set-up where one central agent engages in contests with many peripheral agents), it can cover an application to lobbying. However, they have no notion of conceding contests. In the conflict literature, the ability to cede contests in exchange for a transfer of resources represents a novel addition to Colonel Blotto games (an important class of games in the conflict literature).2 We find this has a dramatic impact on equilibrium outcomes. The government moves from spreading its resources across all contests to fighting only where it will win for certain. This suggests the ability to cede contests has potentially important implications for battles on multiple fronts beyond this application to lobbying.