《社会保障概论》课程教学资源（课后阅读资料）APSR_Volume112. Issue 4.Nov 2018_How_Internal_Constraints_Shape_Interest_Group_Activities_Evidence_from_Access-Seeking_PACs

American Political Science Review (2018)112.4,792-808 doi:10.1017/S0003055418000382 American Political Science Association 2018 How Internal Constraints Shape Interest Group Activities:Evidence from Access-Seeking PACs ZHAO LI Stanford Graduate School of Business nterest groups contribute much less to campaigns than legally allowed.Consequently,prevailing the- ories infer these contributions must yield minimal returns.I argue constraints on PAC fundraising may also explain why interest groups give little.I illuminate one such constraint:access-seeking PACs rely on voluntary donations from affiliated individuals (e.g.,employees),and these PACs alienate donors with partisan preferences when giving to the opposite party.First,difference-in-differences analysis of real giving shows donors withhold donations to access-seeking PACs when PACs contribute to out-partisan politicians.Next,an original survey of corporate PAC donors demonstrates they know how their PACs allocate contributions across parties,and replicates the observational study in an experiment.Donors partisanship thus limits access-seeking PACs'fundraising and influence.This provides a new perspective on why there is little interest group money in elections,and has broad implications for how partisan preferences and other internal constraints shape interest group strategy. ne of the most important questions in political there would be much more PAC money in politics than science is whether or how interest groups can observed (Tullock 1972:Milvo,Primo,and Groseclose gain political influence with campaign contri- 2000:Ansolabehere,de Figueiredo,and Snyder 2003). butions.Because it is difficult to assess the causal effects This argument,however,assumes interest group of contributions on policy outcomes,many instead PACs could always give more to campaigns had they look for clues revealed by interest group strategies.A wanted to (except for the legal caps on contributions). striking observation is that interest group PACs spend Few scholars have examined internal factors that could much less on campaign contributions than legally al- constrain PAC contributions,3 and,in particular,no & lowed.For instance,an interest group PAC can give research has explored the fundraising constraints on up to $5,000 per candidate per election.In the 2016 PACs'ability to finance these contributions.The pres- election cycle,the average PACsponsored by a publicly ence of fundraising and other constraints may not only traded company in the U.S.gave $5,000 or more to only shed new light on why there is little interest group seven federal candidates,and the median number is money in elections,but also provide valuable insights one.Since PACs rarely reach these contribution limits into a wide range of questions about interest group influential work in campaign finance argues that PAC strategy and power. contributions must generate minimal returns,or else In this paper,I identify one such constraint:an inter- est group's need to seek access by contributing across Zhao Li is a Ph.D.candidate in political economics,Stanford Gradu- party lines could conflict with the heterogeneous par- ate School of Business(zli19@stanford.edu). tisan preferences of individual group members,whose This paper has benefited from excellent feedback from the editor voluntary donations are critical for PAC fundraising. and three anonymous reviewers In addition,I am grateful for con- On one hand,many PACs focus on gaining access or in- tinued guidance from Adam Bonica.David Broockman.Neil Malho- fluence and,as a result,give predominantly to members tra,and Ken Shotts.I am also thankful for many helpful comments from Michael Barber,Dave Baron,Jon Bendor,Kate Casey,Richard of the majority party (Cox and Magar 1999;Rudolph DiSalvo,Alex Hertel-Fernandez,Keith Krehbiel,Jon Krosnick,Ellie 1999).On the other hand,contributions made by spon- Powell,Dave Primo,Larry Rothenberg.and poster session attendees sored PACs,such as those representing corporations, at the Yale Center for American Politics Summer Conference as well must by law be financed solely with voluntary contri- as the Conference of the Society for Political Methodologists.The butions from a restricted class of individuals,such as survey experiment in this paper was preregistered with Evidence in Governance and Politics (EGAP)(Registration No.20170413AC) managers,shareholders,and employees.These restric- and approved by Stanford University's Institutional Review Board tions continue to hold in the post-Citizens United era (Protocol No.39512).Special thanks to the PhD program at Stan While individual donors share an interest in their or ford Graduate School of Business for providing funding for my ex- ganization's economic success,many also have strong periment:Adam Bonica for sharing version 3 of the Database on Ideology,Money in Politics,and Elections;and George Thampy for and stable partisan preferences.Moreover,these pref- suggestions that enhanced the ecological validity of the experimen- erences are often diverse even within the same orga- tal treatment conditions.All errors are my own.Replication files nization,which could conflict with an interest group's are available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/TITELH. Received:October 30,2017:revised:March 31,2018;accepted:June 3 Much of the existing literature has focused on whether factors 13,2018.First published online:August 3,2018. such as collective action problems among interest groups(see,for example,Grier,Munger,and Roberts 1994;Gray and Lowery 1997 1 Ansolabehere,de Figueiredo,and Snyder(2003)describe some of Hansen,Mitchell,and Drope 2005)lower the investment returns to the common obstacles to causal identification. PAC contributions,thereby leading to low levels of such contribu- 2 This is based on OpenSecret's bulk data (Center for Responsive tions.But there has been much less attention on what could con- Politics 2018).I examined all companies listed in the Compustat strain PAC contributions and other interest group activities,taking North America database that had active PACs in 2016(Compustat their rates of returns as given.I revisit this point in my theoretical North America 2017). discussions. 792

American Political Science Review (2018) 112, 4, 792–808 doi:10.1017/S0003055418000382 © American Political Science Association 2018 How Internal Constraints Shape Interest Group Activities: Evidence from Access-Seeking PACs ZHAO LI Stanford Graduate School of Business I nterest groups contribute much less to campaigns than legally allowed. Consequently, prevailing the￾ories infer these contributions must yield minimal returns. I argue constraints on PAC fundraising may also explain why interest groups give little. I illuminate one such constraint: access-seeking PACs rely on voluntary donations from affiliated individuals (e.g., employees), and these PACs alienate donors with partisan preferences when giving to the opposite party. First, difference-in-differences analysis of real giving shows donors withhold donations to access-seeking PACs when PACs contribute to out-partisan politicians. Next, an original survey of corporate PAC donors demonstrates they know how their PACs allocate contributions across parties, and replicates the observational study in an experiment. Donors’ partisanship thus limits access-seeking PACs’ fundraising and influence. This provides a new perspective on why there is little interest group money in elections, and has broad implications for how partisan preferences and other internal constraints shape interest group strategy. One of the most important questions in political science is whether or how interest groups can gain political influence with campaign contri￾butions.Because it is difficult to assess the causal effects of contributions on policy outcomes,1 many instead look for clues revealed by interest group strategies. A striking observation is that interest group PACs spend much less on campaign contributions than legally al￾lowed. For instance, an interest group PAC can give up to $5,000 per candidate per election. In the 2016 election cycle, the average PAC sponsored by a publicly traded company in the U.S. gave $5,000 or more to only seven federal candidates, and the median number is one.2 Since PACs rarely reach these contribution limits, influential work in campaign finance argues that PAC contributions must generate minimal returns, or else Zhao Li is a Ph.D. candidate in political economics, Stanford Gradu￾ate School of Business (zli19@stanford.edu). This paper has benefited from excellent feedback from the editor and three anonymous reviewers. In addition, I am grateful for con￾tinued guidance from Adam Bonica, David Broockman, Neil Malho￾tra, and Ken Shotts. I am also thankful for many helpful comments from Michael Barber, Dave Baron, Jon Bendor, Kate Casey, Richard DiSalvo, Alex Hertel-Fernandez, Keith Krehbiel, Jon Krosnick, Ellie Powell, Dave Primo, Larry Rothenberg, and poster session attendees at the Yale Center for American Politics Summer Conference as well as the Conference of the Society for Political Methodologists. The survey experiment in this paper was preregistered with Evidence in Governance and Politics (EGAP) (Registration No. 20170413AC) and approved by Stanford University’s Institutional Review Board (Protocol No. 39512). Special thanks to the PhD program at Stan￾ford Graduate School of Business for providing funding for my ex￾periment; Adam Bonica for sharing version 3 of the Database on Ideology, Money in Politics, and Elections; and George Thampy for suggestions that enhanced the ecological validity of the experimen￾tal treatment conditions. All errors are my own. Replication files are available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/T1TELH. Received: October 30, 2017; revised: March 31, 2018; accepted: June 13, 2018. First published online: August 3, 2018. 1 Ansolabehere, de Figueiredo, and Snyder (2003) describe some of the common obstacles to causal identification. 2 This is based on OpenSecret’s bulk data (Center for Responsive Politics 2018). I examined all companies listed in the Compustat North America database that had active PACs in 2016 (Compustat North America 2017). there would be much more PAC money in politics than observed (Tullock 1972; Milyo, Primo, and Groseclose 2000; Ansolabehere, de Figueiredo, and Snyder 2003). This argument, however, assumes interest group PACs could always give more to campaigns had they wanted to (except for the legal caps on contributions). Few scholars have examined internal factors that could constrain PAC contributions,3 and, in particular, no research has explored the fundraising constraints on PACs’ ability to finance these contributions. The pres￾ence of fundraising and other constraints may not only shed new light on why there is little interest group money in elections, but also provide valuable insights into a wide range of questions about interest group strategy and power. In this paper, I identify one such constraint: an inter￾est group’s need to seek access by contributing across party lines could conflict with the heterogeneous par￾tisan preferences of individual group members, whose voluntary donations are critical for PAC fundraising. On one hand,many PACs focus on gaining access or in￾fluence and, as a result, give predominantly to members of the majority party (Cox and Magar 1999; Rudolph 1999). On the other hand, contributions made by spon￾sored PACs, such as those representing corporations, must by law be financed solely with voluntary contri￾butions from a restricted class of individuals, such as managers, shareholders, and employees. These restric￾tions continue to hold in the post-Citizens United era. While individual donors share an interest in their or￾ganization’s economic success, many also have strong and stable partisan preferences. Moreover, these pref￾erences are often diverse even within the same orga￾nization, which could conflict with an interest group’s 3 Much of the existing literature has focused on whether factors such as collective action problems among interest groups (see, for example, Grier, Munger, and Roberts 1994; Gray and Lowery 1997; Hansen, Mitchell, and Drope 2005) lower the investment returns to PAC contributions, thereby leading to low levels of such contribu￾tions. But there has been much less attention on what could con￾strain PAC contributions and other interest group activities, taking their rates of returns as given. I revisit this point in my theoretical discussions. 792 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

How Internal Constraints Shape Interest Group Activities need to seek access no matter which party dominates port for fictitious candidates,respondents report sub- Congress.Hence,insofar as donors withhold donations stantially lower willingness to give to the PAC when to access-seeking PACs when the latter support out- a greater share of those candidates are randomly as- partisan politicians,access-seeking PAC contributions signed to be out-partisans rather than copartisans. will inevitably undermine PAC fundraising and,by ex- Both studies reveal a clear constraint on the tension,its ability to gain policy access.While I focus on fundraising of access-seeking PACs:these PACs alien- partisanship in this paper,my theory is agnostic to ei- ate donors with a partisan preference when contribut- ther partisanship or ideology as the salient donor pref- ing to politicians of the opposite party.I thus provide a erence insofar as they are strongly correlated (Bonica new perspective on why many interest groups give so 2014:Barber.Canes-Wrone.and Thrower 2017). little in PAC contributions:donors'partisan or ideolog- The existing literature has mostly black-boxed PAC ical preferences limit the fundraising of access-seeking fundraising.In particular,because access-seeking PACs PACs,thereby preventing interest groups from reap- are often analyzed as unitary and purely strategic ac- ing the full benefits of PAC contributions,including tors,there has been no systematic examination of how such contributions'important role in facilitating lob- donors'partisan or ideological preferences affect the bying efforts (Langbein 1986;Hall and Wayman 1990; fundraising of these PACs.To this end,I conducted two Powell 2012;Kalla and Broockman 2016).If polar- studies:difference-in-differences analysis of campaign ization continues to rise,the constraints that individ- finance records as well as an original survey of PAC ual donors impose upon interest groups may become donors. increasingly binding,with uncertain normative impli- First,my difference-in-differences analysis of cam- cations on access-seeking interest groups'impact on paign finance records not only reveals that donors democratic representation and accountability. are less likely to give to access-seeking PACs when a Constraints imposed on interest groups by members greater share of PAC contributions go to out-partisan partisan or ideological preferences may also provide (rather than copartisan)candidates,but also demon- valuable insights on other important interest group ac- 4r元 strates the rising trend as well as persistence in such tivities.These constraints may help to explain not only effect on donor behavior.Using panel data (1990- the surprising lack of corporate independent expendi- 2016)of contributions made by access-seeking PACs as tures after Citizens United v.FEC (Bonica 2016a).but well as donations made by individual donors matched also the rise in donations to intermediary organizations to these PACs,I estimate a difference-in-differences that allow interest groups to subsidize major parties model of donors'decisions to give itemized dona with little transparency(Grumbach and Pierson 2017). tions to access-seeking PACs on the extent to which Finally,I shed new light on why there is little interest PAC contributions favor each donor's preferred party. group money in elections by examining how the par- I identify this effect by comparing changes in the rates tisan or ideological preferences of PAC donors limit of giving by donors of different partisan leanings within PACs'ability to fundraise and seek access.This pa- the same PAC.Using this design,I show that,as access- per thus opens a promising research agenda:advanc- seeking PACs increase the share of their contributions ing our understanding of enduring questions on inter- to one party,they lose donors who support the opposite est group strategy and power by exploring new sources party.Additionally,such an effect on donor behavior of constraints on groups (Hacker and Pierson 2002), appears to have grown over time,coinciding with a rise especially those internal to organizations (Dahl 1959; in polarization (McCarty,Poole,and Rosenthal 2016: Cyert and March 1963:Hart 2004).Any theory of in Pew Research Center 2017).Moreover,as the propor- terest groups is incomplete without considering the tion of PAC contributions to one party increases,eligi constraints-partisan or otherwise-imposed by group ble but inactive donors who prefer this party are more members on whom interest groups critically depend. likely to start donating.while active donors who sup- port the opposite party are more likely to permanently stop giving. HOW INTEREST GROUPS USE PAC Next,I conducted my original survey of PAC donors CONTRIBUTIONS TO SEEK ACCESS to unpack the causal mechanism underlying my ob- servational findings discussed above.First,the survey In this section,I outline why many interest groups shows that individuals who donate to access-seeking adopt an access-oriented strategy in campaign giving, PACs are highly knowledgeable of how their PACs and how this could lead their PACs'support for one eys allocate contributions across parties,and they often party over the other to fluctuate over time.Many in- acquire such information through the same processes terest groups use PAC contributions to gain access that employers use to politically mobilize employ- to influential legislators (Denzau and Munger 1986; ees(Hertel-Fernandez 2016),thereby suggesting that Romer and Snyder 1994;Powell and Grimmer 2016). donors possess the information necessary to reward These contributions facilitate lobbying activities by or punish PACs depending on how well PAC con- purchasing precious face-to-face time with legislators tributions align with their partisan preferences.Sec and incentivizing policymakers to exert effort(Lang- ond,I replicate the key finding from my difference- bein 1986;Hall and Wayman 1990;Powell 2012;Kalla in-differences analysis in an experiment in the survey and Broockman 2016).By contributing to legislators on PAC donors.When shown hypothetical PAC solic. who preside over the relevant bureaucratic agencies, itation letters in which a corporate PAC declares sup- interest groups could also obtain favorable regulatory 793

How Internal Constraints Shape Interest Group Activities need to seek access no matter which party dominates Congress. Hence, insofar as donors withhold donations to access-seeking PACs when the latter support out￾partisan politicians, access-seeking PAC contributions will inevitably undermine PAC fundraising and, by ex￾tension,its ability to gain policy access.While I focus on partisanship in this paper, my theory is agnostic to ei￾ther partisanship or ideology as the salient donor pref￾erence insofar as they are strongly correlated (Bonica 2014; Barber, Canes-Wrone, and Thrower 2017). The existing literature has mostly black-boxed PAC fundraising. In particular, because access-seeking PACs are often analyzed as unitary and purely strategic ac￾tors, there has been no systematic examination of how donors’ partisan or ideological preferences affect the fundraising of these PACs. To this end, I conducted two studies: difference-in-differences analysis of campaign finance records as well as an original survey of PAC donors. First, my difference-in-differences analysis of cam￾paign finance records not only reveals that donors are less likely to give to access-seeking PACs when a greater share of PAC contributions go to out-partisan (rather than copartisan) candidates, but also demon￾strates the rising trend as well as persistence in such effect on donor behavior. Using panel data (1990– 2016) of contributions made by access-seeking PACs as well as donations made by individual donors matched to these PACs, I estimate a difference-in-differences model of donors’ decisions to give itemized dona￾tions to access-seeking PACs on the extent to which PAC contributions favor each donor’s preferred party. I identify this effect by comparing changes in the rates of giving by donors of different partisan leanings within the same PAC. Using this design, I show that, as access￾seeking PACs increase the share of their contributions to one party, they lose donors who support the opposite party. Additionally, such an effect on donor behavior appears to have grown over time, coinciding with a rise in polarization (McCarty, Poole, and Rosenthal 2016; Pew Research Center 2017). Moreover, as the propor￾tion of PAC contributions to one party increases, eligi￾ble but inactive donors who prefer this party are more likely to start donating, while active donors who sup￾port the opposite party are more likely to permanently stop giving. Next, I conducted my original survey of PAC donors to unpack the causal mechanism underlying my ob￾servational findings discussed above. First, the survey shows that individuals who donate to access-seeking PACs are highly knowledgeable of how their PACs allocate contributions across parties, and they often acquire such information through the same processes that employers use to politically mobilize employ￾ees (Hertel-Fernandez 2016), thereby suggesting that donors possess the information necessary to reward or punish PACs depending on how well PAC con￾tributions align with their partisan preferences. Sec￾ond, I replicate the key finding from my difference￾in-differences analysis in an experiment in the survey on PAC donors. When shown hypothetical PAC solic￾itation letters in which a corporate PAC declares sup￾port for fictitious candidates, respondents report sub￾stantially lower willingness to give to the PAC when a greater share of those candidates are randomly as￾signed to be out-partisans rather than copartisans. Both studies reveal a clear constraint on the fundraising of access-seeking PACs: these PACs alien￾ate donors with a partisan preference when contribut￾ing to politicians of the opposite party. I thus provide a new perspective on why many interest groups give so little in PAC contributions: donors’ partisan or ideolog￾ical preferences limit the fundraising of access-seeking PACs, thereby preventing interest groups from reap￾ing the full benefits of PAC contributions, including such contributions’ important role in facilitating lob￾bying efforts (Langbein 1986; Hall and Wayman 1990; Powell 2012; Kalla and Broockman 2016). If polar￾ization continues to rise, the constraints that individ￾ual donors impose upon interest groups may become increasingly binding, with uncertain normative impli￾cations on access-seeking interest groups’ impact on democratic representation and accountability. Constraints imposed on interest groups by members’ partisan or ideological preferences may also provide valuable insights on other important interest group ac￾tivities. These constraints may help to explain not only the surprising lack of corporate independent expendi￾tures after Citizens United v. FEC (Bonica 2016a), but also the rise in donations to intermediary organizations that allow interest groups to subsidize major parties with little transparency (Grumbach and Pierson 2017). Finally, I shed new light on why there is little interest group money in elections by examining how the par￾tisan or ideological preferences of PAC donors limit PACs’ ability to fundraise and seek access. This pa￾per thus opens a promising research agenda: advanc￾ing our understanding of enduring questions on inter￾est group strategy and power by exploring new sources of constraints on groups (Hacker and Pierson 2002), especially those internal to organizations (Dahl 1959; Cyert and March 1963; Hart 2004). Any theory of in￾terest groups is incomplete without considering the constraints—partisan or otherwise—imposed by group members on whom interest groups critically depend. HOW INTEREST GROUPS USE PAC CONTRIBUTIONS TO SEEK ACCESS In this section, I outline why many interest groups adopt an access-oriented strategy in campaign giving, and how this could lead their PACs’ support for one party over the other to fluctuate over time. Many in￾terest groups use PAC contributions to gain access to influential legislators (Denzau and Munger 1986; Romer and Snyder 1994; Powell and Grimmer 2016). These contributions facilitate lobbying activities by purchasing precious face-to-face time with legislators and incentivizing policymakers to exert effort (Lang￾bein 1986; Hall and Wayman 1990; Powell 2012; Kalla and Broockman 2016). By contributing to legislators who preside over the relevant bureaucratic agencies, interest groups could also obtain favorable regulatory 793 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

Zhao Li FIGURE 1.Contributions from Access-Seeking PACs Track Seat Shares in House and Senate 70% 60% average percentage of PAC contributions to Republican recipients Republican seat share in the House 50% Republican seat share in the Senate 40% 1990 1994 1998 2002 2006 2010 2014 outcomes (Gordon and Hafer 2005:de Figueiredo and restrictions,individuals affiliated with the parent orga- 4号 Edwards 2007).In short,access-seeking PACs may be nizations.For example,a corporate PAC can ask for able to shape the policymaking process to their ad- donations at any time from only the firm's executive vantage by investing in those with the most legislative and administrative personnel,stockholders,and fami- & clout. lies of the first two.In addition,it may fundraise up to The process of access seeking often requires PACs twice a year from other employees and their families to prioritize giving to members of the majority no mat- who do not fit into the previous categories. ter which party dominates Congress.To target influ- Because of legal restrictions on donations to access- ential legislators,access-seeking PACs focus on candi- seeking PACs,mobilizing a wide range of eligible date traits such as incumbency (Fouirnaies and Hall donors is critical to their fundraising success.The par- 2014),seat safety (Bonica 2013),majority party sta- ent organizations sponsoring PACs must not coerce or tus (Cox and Magar 1999;Rudolph 1999),committee bribe individuals into donating to the PACs.Moreover, membership and position(Grier,Munger,and Roberts each eligible donor can give no more than $5,000 to 1994:Romer and Snyder 1994:Kroszner and Strat- their PAC per calendar year,and a typical donor gives mann 1998;Esterling 2007),party leadership(Heberlig, less than one-twentieth of this threshold.5 To put this in 235.5010 Hetherington,and Larson 2006;Ban,Moskowitz,and perspective,the contribution limits faced by an access- Snyder 2016),and procedural power (Fouirnaies and seeking PAC to candidates and party committees are Hall 2018).Given changes in majority control of between $5,000 and $45,000 per election depending on Congress over time,the legislators most sought after recipient type.So,to be able to contribute to a wide by access-seeking PACs will not always come from range of recipients,access-seeking PACs will have to the same party.As a result,contributions from access- rely on more than their top personnel for funding. seeking PACs are very much attuned to partisan swings However,persuading most prospective donors to in Congress.Figure 1 shows that for all PACs that give need not be easy.Even though strategic PAC con- OpenSecrets code as business PACs(Center for Re- tributions may advance the bottom line of the parent sponsive Politics 2018),the average share of PAC con- organizations (Jayachandran 2006;Akey 2015;though tributions to Republican candidates and party com- see Ansolabehere,Snyder,and Ueda 2004;Fowler, mittees closely tracks Republicans'seat shares in both chambers of Congress. 4 For a complete set of regulations on fundraising by Separate HOW INTEREST GROUP PACS RAISE Segregated Funds,see https://transition.fec.gov/pdf/colagui.pdf as FUNDS well as https://transition.fec.gov/pages/brochures/ssfvnonconnected. pdf.Both accessed on April72018. See "Contribution Limits for 2015-2016 Federal Elections." Before access-seeking PACs contribute to influential https://transition.fec.gov/info/contriblimitschart1516.pdf.Accessed legislators,they must first raise funds through a pro- on April 7 2018. cess that is prone to collective action problems.These For federal election cycles 1990 through 2016,the median amount PACs cannot draw on their parent organizations'trea- of itemized donations conditional on giving such donations is only /:sany sury funds.For example,Walmart is prohibited from $500 per cycle(i.e.,over 2 calendar years).The median amount of donations from all eligible PAC donors,including those who gave writing a check to the PAC it sponsors.Instead,PACs unitemized donations or nothing at all,is likely to be substantially connected to parent organizations can only solicit,with lower. 794

Zhao Li FIGURE 1. Contributions from Access-Seeking PACs Track Seat Shares in House and Senate outcomes (Gordon and Hafer 2005; de Figueiredo and Edwards 2007). In short, access-seeking PACs may be able to shape the policymaking process to their ad￾vantage by investing in those with the most legislative clout. The process of access seeking often requires PACs to prioritize giving to members of the majority no mat￾ter which party dominates Congress. To target influ￾ential legislators, access-seeking PACs focus on candi￾date traits such as incumbency (Fouirnaies and Hall 2014), seat safety (Bonica 2013), majority party sta￾tus (Cox and Magar 1999; Rudolph 1999), committee membership and position (Grier,Munger, and Roberts 1994; Romer and Snyder 1994; Kroszner and Strat￾mann 1998;Esterling 2007), party leadership (Heberlig, Hetherington, and Larson 2006; Ban, Moskowitz, and Snyder 2016), and procedural power (Fouirnaies and Hall 2018). Given changes in majority control of Congress over time, the legislators most sought after by access-seeking PACs will not always come from the same party. As a result, contributions from access￾seeking PACs are very much attuned to partisan swings in Congress. Figure 1 shows that for all PACs that OpenSecrets code as business PACs (Center for Re￾sponsive Politics 2018), the average share of PAC con￾tributions to Republican candidates and party com￾mittees closely tracks Republicans’ seat shares in both chambers of Congress. HOW INTEREST GROUP PACS RAISE FUNDS Before access-seeking PACs contribute to influential legislators, they must first raise funds through a pro￾cess that is prone to collective action problems. These PACs cannot draw on their parent organizations’ trea￾sury funds. For example, Walmart is prohibited from writing a check to the PAC it sponsors. Instead, PACs connected to parent organizations can only solicit, with restrictions, individuals affiliated with the parent orga￾nizations.4 For example, a corporate PAC can ask for donations at any time from only the firm’s executive and administrative personnel, stockholders, and fami￾lies of the first two. In addition, it may fundraise up to twice a year from other employees and their families who do not fit into the previous categories. Because of legal restrictions on donations to access￾seeking PACs, mobilizing a wide range of eligible donors is critical to their fundraising success. The par￾ent organizations sponsoring PACs must not coerce or bribe individuals into donating to the PACs. Moreover, each eligible donor can give no more than $5,000 to their PAC per calendar year,5 and a typical donor gives less than one-twentieth of this threshold.6 To put this in perspective, the contribution limits faced by an access￾seeking PAC to candidates and party committees are between $5,000 and $45,000 per election depending on recipient type. So, to be able to contribute to a wide range of recipients, access-seeking PACs will have to rely on more than their top personnel for funding. However, persuading most prospective donors to give need not be easy. Even though strategic PAC con￾tributions may advance the bottom line of the parent organizations (Jayachandran 2006; Akey 2015; though see Ansolabehere, Snyder, and Ueda 2004; Fowler, 4 For a complete set of regulations on fundraising by Separate Segregated Funds, see https://transition.fec.gov/pdf/colagui.pdf as well as https://transition.fec.gov/pages/brochures/ssfvnonconnected. pdf. Both accessed on April 7, 2018. 5 See “Contribution Limits for 2015-2016 Federal Elections,” https://transition.fec.gov/info/contriblimitschart1516.pdf. Accessed on April 7, 2018. 6 For federal election cycles 1990 through 2016, the median amount of itemized donations conditional on giving such donations is only $500 per cycle (i.e., over 2 calendar years). The median amount of donations from all eligible PAC donors, including those who gave unitemized donations or nothing at all, is likely to be substantially lower. 794 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

How Internal Constraints Shape Interest Group Activities Garro.and Spenkuch 2017).such benefit accrues to or pivotal in individual donors'decisionmaking(Wilson ganizational members regardless of whether they have 1973;Moe 1980).In the context of PAC fundraising,I donated to the PAC,thus producing a collective action argue that one particularly salient purposive incentive problem.While the top personnel might have a suffi- is donors'partisan or ideological preference. ciently large stake in the economic success of the or. Stated formally,my main hypothesis is that donors ganization to find donating worthwhile,each of them with partisan preference should be less likely to donate can give no more than $5,000 per year to their PAC. to their affiliated access-seeking PACs when a greater For other eligible donors,the marginal benefit to their share of PAC contributions go to out-partisan(rather donations might not justify the cost(Olson 1965). than copartisan)politicians.Note that while I focus To tackle this collective action problem,organiza- on partisanship in this paper,it could also be a proxy tions sponsoring access-seeking PACs provide vari- for individuals'ideological preferences (Bonica 2014; ous benefits to induce giving,though it is unclear Barber.Canes-Wrone,and Thrower 2017).In the re- how well these tactics work.Many encourage em- mainder of this section,I argue that this hypothesis, ployees to authorize payroll deductions to maintain if true,implies that partisan preferences of donors can a steady stream of incoming donations,although the significantly constrain PACs'ability to raise funds and Federal Election Commission (hereafter the FEC)re- thereby obtain access.Additionally,I provide three rea- quires such programs to be opt-in rather than opt-out sons for why this hypothesis could be true. (Sabato 1985).Additionally,the parent organizations To see why donors'partisanships could affect the sponsoring these PACs offer various selective benefits strategic success of access-seeking PACs,it is worth em- 元 to donors,from casino nights to golf tournaments to phasizing that the share of PAC contributions to one "high-donor clubs"(Sabato 1985).However,these se- party versus the other may influence not only the rate of lective benefits not only constitute a costly means to returns to PAC contributions.which motivates much of sustain donations (Wilson 1973:Moe 1980).but also the existing literature on PACs'contribution patterns, risk violating campaign finance regulations(Berman but also the amount of fundraising.These components 2014). jointly determine the total returns to PAC contributions Furthermore,even though coercion is expressly for- (in the simplest case,total returns would be the prod- bidden,the parent organizations of access-seeking uct of the two).If PAC donors behave as I hypothesize, PACs often try to cultivate a norm of giving (So- access-seeking PACs may have to strike a compromise rauf 1984).Again,however.its effectiveness on boost- between increasing the per-dollar rate of return to PAC ing PAC fundraising has not been empirically eval- contributions,and raising more donations from eligible uated and might,in fact,be limited.Social pressure, PAC donors.s Consider the case of Google whose PAC especially when coupled with monitoring,could help appears to have a predominantly liberal donor base.? access-seeking organizations overcome collective ac- Under the current government,if Google's PAC prior- tion problems in many realms of political mobiliza- itizes giving to members of the majority (Republican) tion(Hertel-Fernandez 2017).But the general pressure party,it might get more bang for the buck for its PAC to conform might not be sufficient to mobilize many contributions at the expense of fundraising.If Google's prospective PAC donors,since giving to one's PAC en- PAC instead tries to raise more donations by focusing tails a greater personal cost than other forms of politi- on giving to Democratic legislators,it might reduce the 5.501g cal participation that need not conflict with one's parti- effectiveness of its PAC contributions in gaining valu- san or ideological preferences.An employer could,for able access and influencing policymaking. instance,arrange for Democratic-leaning workers to Having discussed why PAC donors'partisan prefer- contact influential legislators on the Democratic side ences might limit a PAC's ability to obtain access and and Republican-leaning workers to Republican legis- gain influence,I now turn to three factors that lend sup- lators.In contrast,since each organization is allowed to port for my hypothesis that PAC donors'willingness sponsor only one PAC,prospective donors cannot give to give decreases in the share of PAC contribution to to access-seeking PACs without compromising their out-partisan politicians.First,many PAC donors have partisanship or ideology at some point.The institu- strong and stable partisan preferences.For example, tional structure of access-seeking PACs could thereby among PAC donors who have given directly to candi- undercut the effectiveness of using social pressure to dates or party committees,77%have only given to one solicit donations party in their lifetime.This is evident in the left panel of eys THEORY:DONORS'PARTISAN PREFERENCES CONSTRAIN PACS'ABILITY tangible costs and benefits.Solidary incentives are intangible costs and benefits of a social nature,for example,from friendship,cama- TO RAISE FUNDS AND GAIN ACCESS raderie,recreational activity,status,social pressure,or a sense of be- longing.Purposive incentives are intangible costs and benefits ulti- When payroll deductions,selective benefits,and social mately grounded on values of a suprapersonal nature,e.g.,notions of pressure are insufficient to induce donations to access right and wrong,moral or religious principles,political ideology,and seeking PACs,other motivations,particularly purpo- notions of fairness and justice." sive benefits(Clark and Wilson 1961),could become Section A.1 of the online appendix illustrates this tradeoff in a sim- ple formal model of PACs'decision problem. 84.3%of Google's PAC donors who have given to candidates or This follows from the typology of incentives proposed in Clark party committees appear Democratic-leaning based on their dona- and Wilson (1961).As Moe (1980)explains:"Material incentives are tion histories. 795

How Internal Constraints Shape Interest Group Activities Garro, and Spenkuch 2017), such benefit accrues to or￾ganizational members regardless of whether they have donated to the PAC, thus producing a collective action problem. While the top personnel might have a suffi￾ciently large stake in the economic success of the or￾ganization to find donating worthwhile, each of them can give no more than $5,000 per year to their PAC. For other eligible donors, the marginal benefit to their donations might not justify the cost (Olson 1965). To tackle this collective action problem, organiza￾tions sponsoring access-seeking PACs provide vari￾ous benefits to induce giving, though it is unclear how well these tactics work. Many encourage em￾ployees to authorize payroll deductions to maintain a steady stream of incoming donations, although the Federal Election Commission (hereafter the FEC) re￾quires such programs to be opt-in rather than opt-out (Sabato 1985). Additionally, the parent organizations sponsoring these PACs offer various selective benefits to donors, from casino nights to golf tournaments to “high-donor clubs” (Sabato 1985). However, these se￾lective benefits not only constitute a costly means to sustain donations (Wilson 1973; Moe 1980), but also risk violating campaign finance regulations (Berman 2014). Furthermore, even though coercion is expressly for￾bidden, the parent organizations of access-seeking PACs often try to cultivate a norm of giving (So￾rauf 1984). Again, however, its effectiveness on boost￾ing PAC fundraising has not been empirically eval￾uated and might, in fact, be limited. Social pressure, especially when coupled with monitoring, could help access-seeking organizations overcome collective ac￾tion problems in many realms of political mobiliza￾tion (Hertel-Fernandez 2017). But the general pressure to conform might not be sufficient to mobilize many prospective PAC donors, since giving to one’s PAC en￾tails a greater personal cost than other forms of politi￾cal participation that need not conflict with one’s parti￾san or ideological preferences. An employer could, for instance, arrange for Democratic-leaning workers to contact influential legislators on the Democratic side, and Republican-leaning workers to Republican legis￾lators. In contrast, since each organization is allowed to sponsor only one PAC, prospective donors cannot give to access-seeking PACs without compromising their partisanship or ideology at some point. The institu￾tional structure of access-seeking PACs could thereby undercut the effectiveness of using social pressure to solicit donations. THEORY: DONORS’ PARTISAN PREFERENCES CONSTRAIN PACS’ ABILITY TO RAISE FUNDS AND GAIN ACCESS When payroll deductions, selective benefits, and social pressure are insufficient to induce donations to access￾seeking PACs, other motivations, particularly purpo￾sive benefits (Clark and Wilson 1961),7 could become 7 This follows from the typology of incentives proposed in Clark and Wilson (1961). As Moe (1980) explains: “Material incentives are pivotal in individual donors’ decisionmaking (Wilson 1973; Moe 1980). In the context of PAC fundraising, I argue that one particularly salient purposive incentive is donors’ partisan or ideological preference. Stated formally, my main hypothesis is that donors with partisan preference should be less likely to donate to their affiliated access-seeking PACs when a greater share of PAC contributions go to out-partisan (rather than copartisan) politicians. Note that while I focus on partisanship in this paper, it could also be a proxy for individuals’ ideological preferences (Bonica 2014; Barber, Canes-Wrone, and Thrower 2017). In the re￾mainder of this section, I argue that this hypothesis, if true, implies that partisan preferences of donors can significantly constrain PACs’ ability to raise funds and thereby obtain access.Additionally, I provide three rea￾sons for why this hypothesis could be true. To see why donors’ partisanships could affect the strategic success of access-seeking PACs,it is worth em￾phasizing that the share of PAC contributions to one party versus the other may influence not only the rate of returns to PAC contributions, which motivates much of the existing literature on PACs’ contribution patterns, but also the amount of fundraising. These components jointly determine the total returnsto PAC contributions (in the simplest case, total returns would be the prod￾uct of the two). If PAC donors behave as I hypothesize, access-seeking PACs may have to strike a compromise between increasing the per-dollar rate of return to PAC contributions, and raising more donations from eligible PAC donors.8 Consider the case of Google whose PAC appears to have a predominantly liberal donor base.9 Under the current government, if Google’s PAC prior￾itizes giving to members of the majority (Republican) party, it might get more bang for the buck for its PAC contributions at the expense of fundraising. If Google’s PAC instead tries to raise more donations by focusing on giving to Democratic legislators, it might reduce the effectiveness of its PAC contributions in gaining valu￾able access and influencing policymaking. Having discussed why PAC donors’ partisan prefer￾ences might limit a PAC’s ability to obtain access and gain influence, I now turn to three factors that lend sup￾port for my hypothesis that PAC donors’ willingness to give decreases in the share of PAC contribution to out-partisan politicians. First, many PAC donors have strong and stable partisan preferences. For example, among PAC donors who have given directly to candi￾dates or party committees, 77% have only given to one party in their lifetime.This is evident in the left panel of tangible costs and benefits. Solidary incentives are intangible costs and benefits of a social nature, for example, from friendship, cama￾raderie, recreational activity, status, social pressure, or a sense of be￾longing. Purposive incentives are intangible costs and benefits ulti￾mately grounded on values of a suprapersonal nature, e.g., notions of right and wrong, moral or religious principles, political ideology, and notions of fairness and justice.” 8 Section A.1 of the online appendix illustrates this tradeoff in a sim￾ple formal model of PACs’ decision problem. 9 84.3% of Google’s PAC donors who have given to candidates or party committees appear Democratic-leaning based on their dona￾tion histories. 795 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

Zhao Li FIGURE 2.Scaled Density Plots of Donors'Inferred Partisan Preferences 1.00 1.00 0.75 0.75 0.50 0.50 0.25 0.25 0.00 0.00 0% 25% 50% 75% 100% 0% 25% 50% 75% 100% percent of individual donor's lifetime percent of Republican-leaning donors within donations to Republican recipients PAC,weighted by number of unique donors Percentile oth 25th 5oth 75th 1ooth Percentile oth 25th 5oth 75th 1ooth Value of%o%98%100%100% Vaue0%45.7%65%82.2%100% 4号 Figure 2,which shows the full distribution of the pro- like the politicians that their PACs support.For many portion of individual PAC donors'lifetime donations access-seeking PACs,the top personnel of the parent to Republican recipients,calculated using OpenSecrets organization,such as the chief executive officer or the & data (Center for Responsive Politics 2018).Interest- director of government affairs,not only appoint mem- ingly,there is considerable diversity in donors'par- bers of the PAC boards but also direct their day-to- tisan preferences even within the same PAC.As the day activities (Sabato 1985;Hart 2001).The rest of right panel of Figure 2 shows,when weighted by the the board typically consists of representatives from dif- number of unique donors in each PAC,the share of ferent departments or regional branches (Sorauf 1984 Republican-leaning donors within a PAC follows a uni- Sabato 1985).The board decides whom the PAC should modal distribution.In other words,no matter which contribute to as well as how to fundraise.The rank- party dominates Congress,there will almost always be and-file members,in contrast,have little say in where many prospective donors to access-seeking PACs who their donations to their PAC would ultimately end up find their PACs'support for the majority party at odds (Sabato 1985;Hart 2004).When prospective donors with their own partisan or ideological preference. disapprove of their PACs'choice of candidates or par- 235.5010 The second condition that may make donors'par- ties,the only effective means of protestation is to with- tisan preferences salient for fundraising is that PAC hold donations. contributions are one of the most divisive and visible forms of interest group participation in politics.Cam- paign contributions directly help to finance the elec- tions of office seekers,who not only bear partisan la- THE ROLE OF DONORS'PARTISAN PREFERENCE:WHAT WE KNOW bels but also push for a broad range of policies that can be polarizing if not controversial (Halper 2014). While the existing literature has not examined whether Moreover,donors can learn about where PAC contri- the partisan preferences of donors affect the fundrais- butions are going in a timely manner.PAC contribu- ing outcomes for access-seeking PACs.it generally sup- tions are disclosed each quarter,and such information ports this plausibility.Grossman and Helpman(1996, is easily obtainable on both the FEC's web site and p.270)speculate that individual donors might ob- www.opensecrets.org.Additionally,my original survey ject to interest group PAC contributions on ideolog- of corporate PAC donors shows that they have access ical grounds in spite of shared material interest,but to numerous other information channels and,as a re- there has been no formal-theoretical prediction of how sult,are highly knowledgeable of how their PACs al- donors'partisan or ideological preferences affect PAC locate contributions across parties.In short,by directly fundraising.Descriptive work by Sorauf (1984)and and publicly inserting themselves into electoral politics, Sabato (1985)on access-seeking PACs includes anec- access-seeking PACs are particularly prone to upset- dotes of PACs'choice of candidates offending donors ting prospective donors on partisan grounds. but says nothing about whether such emotions trans- The third reason why donors'partisan preferences late into donor actions. may matter for the fundraising of access-seeking PACs Among recent work,Lowry (2013)finds that do- is that the exclusive governance structures of these nations to access-seeking PACs (and other sponsored PACs leave little choice but exit for donors who dis PACs),when aggregated at the congressional district 796

Zhao Li FIGURE 2. Scaled Density Plots of Donors’ Inferred Partisan Preferences Figure 2, which shows the full distribution of the pro￾portion of individual PAC donors’ lifetime donations to Republican recipients, calculated using OpenSecrets data (Center for Responsive Politics 2018). Interest￾ingly, there is considerable diversity in donors’ par￾tisan preferences even within the same PAC. As the right panel of Figure 2 shows, when weighted by the number of unique donors in each PAC, the share of Republican-leaning donors within a PAC follows a uni￾modal distribution. In other words, no matter which party dominates Congress, there will almost always be many prospective donors to access-seeking PACs who find their PACs’ support for the majority party at odds with their own partisan or ideological preference. The second condition that may make donors’ par￾tisan preferences salient for fundraising is that PAC contributions are one of the most divisive and visible forms of interest group participation in politics. Cam￾paign contributions directly help to finance the elec￾tions of office seekers, who not only bear partisan la￾bels but also push for a broad range of policies that can be polarizing if not controversial (Halper 2014). Moreover, donors can learn about where PAC contri￾butions are going in a timely manner. PAC contribu￾tions are disclosed each quarter, and such information is easily obtainable on both the FEC’s web site and www.opensecrets.org. Additionally, my original survey of corporate PAC donors shows that they have access to numerous other information channels and, as a re￾sult, are highly knowledgeable of how their PACs al￾locate contributions across parties. In short, by directly and publicly inserting themselves into electoral politics, access-seeking PACs are particularly prone to upset￾ting prospective donors on partisan grounds. The third reason why donors’ partisan preferences may matter for the fundraising of access-seeking PACs is that the exclusive governance structures of these PACs leave little choice but exit for donors who dis￾like the politicians that their PACs support. For many access-seeking PACs, the top personnel of the parent organization, such as the chief executive officer or the director of government affairs, not only appoint mem￾bers of the PAC boards but also direct their day-to￾day activities (Sabato 1985; Hart 2001). The rest of the board typically consists of representatives from dif￾ferent departments or regional branches (Sorauf 1984; Sabato 1985).The board decides whom the PAC should contribute to as well as how to fundraise. The rank￾and-file members, in contrast, have little say in where their donations to their PAC would ultimately end up (Sabato 1985; Hart 2004). When prospective donors disapprove of their PACs’ choice of candidates or par￾ties, the only effective means of protestation is to with￾hold donations. THE ROLE OF DONORS’ PARTISAN PREFERENCE: WHAT WE KNOW While the existing literature has not examined whether the partisan preferences of donors affect the fundrais￾ing outcomes for access-seeking PACs, it generally sup￾ports this plausibility. Grossman and Helpman (1996, p. 270) speculate that individual donors might ob￾ject to interest group PAC contributions on ideolog￾ical grounds in spite of shared material interest, but there has been no formal-theoretical prediction of how donors’ partisan or ideological preferences affect PAC fundraising. Descriptive work by Sorauf (1984) and Sabato (1985) on access-seeking PACs includes anec￾dotes of PACs’ choice of candidates offending donors, but says nothing about whether such emotions trans￾late into donor actions. Among recent work, Lowry (2013) finds that do￾nations to access-seeking PACs (and other sponsored PACs), when aggregated at the congressional district 796 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

How Internal Constraints Shape Interest Group Activities level,are not sensitive to factors such as electoral Responsive Politics 2018).Section A.2.2 of the ap- competition or presidential cycles that would typically pendix contains descriptive summaries of the data affect electorally motivated donations.This suggests donors might not be taking their partisan or ideological preferences into account when giving to access-seeking Empirical Strategy PACs,although Lowry(2013)recognizes the limits to I employ the following difference-in-differences de- causal inference from ecological analyses. Sign:1 Min and You (2015)find that activist sharehold- ers (e.g.,public pension fund and union investors) target corporations whose political activities conflict Giveiit aij +tit +B(Indiv_Ri x PAC_Rit)+Eiit, with the partisan or ideological preferences of these 1) shareholders (e.g.,corporations that predominantly where i denotes an individual donor,j represents a donate to Republican politicians).This lends support PAC,and t indicates a 2-year federal election cycle.The to the conjecture that shareholders and other eligi- outcome variable,Givert,is an indicator that equals 1 if ble PAC donors might also constrain the fundraising and only if donor i gave any itemized donation to PACj of access-seeking PACs due to partisan or ideological during cycle t.2 On the right-hand side of Equation(1) disagreements. are two fixed effects,one for each donor-PAC dyad ( Bonica(2016a)shows that corporate executives,un- and another at the PAC-cycle level (it).In addition,the like the PACs they preside over,behave like ideologues main independent variable of interest is the product rather than investors when donating to candidates on of two variables:Indiv_Ri E[-1/2,1/2],the net share their own.Moreover.there is a remarkable degree of of donor i's lifetime donations to Republican (as op- heterogeneity in political ideologies even on the same posed to Democratic)candidates and party commit- corporate board.The prevalence of bipartisan board- tees,and PAC_Rit E[-1/2,1/2],the net share of PAC rooms,Bonica(2016a)argues,could account for why i's contributions to Republican(as opposed to Demo- 4号元 corporate PACs appear moderate in their contribution cratic)candidates and party committees during cycle patterns,for otherwise a clear partisan or ideological t.Indiv R;is a reliable measure of individual donors bias in PAC giving could antagonize elite donors.That partisan leanings because,as the left panel of Figure 2 being said,we do not know this for sure without a sys- shows,77%of all donors in my sample have given to tematic examination of how donor behavior depends only one party in their entire donation histories (so on the degree to which access-seeking PACs support Indiv R;mostly clusters at either-1/2 or 1/2).Barber. politicians that share the donors'partisan preferences Canes-Wrone,and Thrower(2017)also independently validate my method of classifying donors'partisanship. STUDY 1:DIFFERENCES-IN-DIFFERENCES As usual,I do not preclude the possibility that my mea- ANALYSIS OF CAMPAIGN FINANCE DATA sure of partisanship is a proxy for ideology.Note that I do not separately control for Indiv_Ri or PAC_Rit Using difference-in-differences analysis of campaign fi- because the donor-PAC fixed effect (a)absorbs the nance records,I first verify that patterns of real giving former while the PAC-cycle fixed effect (it)absorbs are consistent with my main hypothesis:access-seeking the latter.I cluster all standard errors at the PAC level 5.501g PACs lose donors with a partisan preference when giv- Since Indiv R;represents donor i's time-invariant ing to politicians of the opposite party.In particular, relative preference for the Republican Party,and such impact on donor behavior appears to have grown PAC_Rit measures PACi's relative support for the Re- in time,coinciding with a rise in polarization(Mc- publican Party in cycle t,their product captures the de- Carty,Poole,and Rosenthal 2016;Pew Research Center gree of alignment between PAC is contributions dur- 2017).Moreover,these effects also appear to be long- ing cycle t and donor i's partisan preference.Therefore, lasting:eligible but inactive donors are more likely to my hypothesis is that B>0 in Specification 1. start giving when the share of PAC contributions to I identify B by comparing within-person changes in their preferred party increases,and active donors are rates of giving by donors of different partisan leanings more likely to permanently stop giving when changes belonging to the same PAC.This becomes clear once in PAC contributions benefit the party they oppose. the donor-PAC fixed effect i and the PAC-cycle fixed I examine PACs that are coded as business PACs by effect rjt are differenced out of Specification 1,resulting OpenSecrets(Center for Responsive Politics 2018).101 in eys then construct a panel of individual-to-PAC as well as PAC-to-candidate/party committee contributions for cycles 1990-2016 using OpenSecrets data(Center for △，Givein-△Giveri=B(Indiv_.R:-Indiv_Rr) ×△PAC_Rit+㎡t, (2) 10 OpenSecrets categorizes PACs as business,labor,ideological,or others.I exclude a small minority of these business PACs that are not sponsored by a corporation,trade organization (such as the Mort- gage Bankers Association),or membership organization (such as the 11 I estimate this model using Stata package reghdfe developed by National Association of Realtors)to focus on access-seeking PACs as /:sony empirically defined in the existing literature(see,for example,Rom Section2 of the onine appendix.I discuss why one cao and Snyder 1994:Cox and Magar 1999:Bonica 2013:Drutman 2015) causally estimate treatment effects on amounts of giving due to post- See Section A.2.1 of the online appendix for details. treatment censoring. 797

How Internal Constraints Shape Interest Group Activities level, are not sensitive to factors such as electoral competition or presidential cycles that would typically affect electorally motivated donations. This suggests donors might not be taking their partisan or ideological preferences into account when giving to access-seeking PACs, although Lowry (2013) recognizes the limits to causal inference from ecological analyses. Min and You (2015) find that activist sharehold￾ers (e.g., public pension fund and union investors) target corporations whose political activities conflict with the partisan or ideological preferences of these shareholders (e.g., corporations that predominantly donate to Republican politicians). This lends support to the conjecture that shareholders and other eligi￾ble PAC donors might also constrain the fundraising of access-seeking PACs due to partisan or ideological disagreements. Bonica (2016a) shows that corporate executives, un￾like the PACs they preside over, behave like ideologues rather than investors when donating to candidates on their own. Moreover, there is a remarkable degree of heterogeneity in political ideologies even on the same corporate board. The prevalence of bipartisan board￾rooms, Bonica (2016a) argues, could account for why corporate PACs appear moderate in their contribution patterns, for otherwise a clear partisan or ideological bias in PAC giving could antagonize elite donors. That being said, we do not know this for sure without a sys￾tematic examination of how donor behavior depends on the degree to which access-seeking PACs support politicians that share the donors’ partisan preferences. STUDY 1: DIFFERENCES-IN-DIFFERENCES ANALYSIS OF CAMPAIGN FINANCE DATA Using difference-in-differences analysis of campaign fi￾nance records, I first verify that patterns of real giving are consistent with my main hypothesis: access-seeking PACs lose donors with a partisan preference when giv￾ing to politicians of the opposite party. In particular, such impact on donor behavior appears to have grown in time, coinciding with a rise in polarization (Mc￾Carty,Poole, and Rosenthal 2016;Pew Research Center 2017). Moreover, these effects also appear to be long￾lasting: eligible but inactive donors are more likely to start giving when the share of PAC contributions to their preferred party increases, and active donors are more likely to permanently stop giving when changes in PAC contributions benefit the party they oppose. I examine PACs that are coded as business PACs by OpenSecrets (Center for Responsive Politics 2018).10 I then construct a panel of individual-to-PAC as well as PAC-to-candidate/party committee contributions for cycles 1990–2016 using OpenSecrets data (Center for 10 OpenSecrets categorizes PACs as business, labor, ideological, or others. I exclude a small minority of these business PACs that are not sponsored by a corporation, trade organization (such as the Mort￾gage Bankers Association), or membership organization (such as the National Association of Realtors) to focus on access-seeking PACs as empirically defined in the existing literature (see, for example,Romer and Snyder 1994; Cox and Magar 1999; Bonica 2013; Drutman 2015). See Section A.2.1 of the online appendix for details. Responsive Politics 2018). Section A.2.2 of the ap￾pendix contains descriptive summaries of the data. Empirical Strategy I employ the following difference-in-differences de￾sign:11 Givei jt = αi j + τjt + β(Indiv_Ri × PAC_Rjt) + i jt, (1) where i denotes an individual donor, j represents a PAC, and t indicates a 2-year federal election cycle. The outcome variable, Giveijt, is an indicator that equals 1 if and only if donori gave any itemized donation to PAC j during cycle t. 12 On the right-hand side of Equation (1) are two fixed effects, one for each donor-PAC dyad (αij) and another at the PAC-cycle level (τ jt). In addition, the main independent variable of interest is the product of two variables: Indiv_Ri ∈ [−1/2, 1/2], the net share of donor i’s lifetime donations to Republican (as op￾posed to Democratic) candidates and party commit￾tees, and PAC_Rjt ∈ [−1/2, 1/2], the net share of PAC j’s contributions to Republican (as opposed to Demo￾cratic) candidates and party committees during cycle t. Indiv_Ri is a reliable measure of individual donors’ partisan leanings because, as the left panel of Figure 2 shows, 77% of all donors in my sample have given to only one party in their entire donation histories (so Indiv_Ri mostly clusters at either −1/2 or 1/2). Barber, Canes-Wrone, and Thrower (2017) also independently validate my method of classifying donors’ partisanship. As usual, I do not preclude the possibility that my mea￾sure of partisanship is a proxy for ideology. Note that I do not separately control for Indiv_Ri or PAC_Rjt because the donor-PAC fixed effect (αij) absorbs the former while the PAC-cycle fixed effect (τ jt) absorbs the latter. I cluster all standard errors at the PAC level. Since Indiv_Ri represents donor i’s time-invariant relative preference for the Republican Party, and PAC_Rjt measures PAC j’s relative support for the Re￾publican Party in cycle t, their product captures the de￾gree of alignment between PAC j’s contributions dur￾ing cycle t and donor i’s partisan preference. Therefore, my hypothesis is that β > 0 in Specification 1. I identify β by comparing within-person changes in rates of giving by donors of different partisan leanings belonging to the same PAC. This becomes clear once the donor-PAC fixed effect αij and the PAC-cycle fixed effect τ jt are differenced out of Specification 1, resulting in tGivei jt − tGivei jt = β(Indiv_Ri − Indiv_Ri ) ×tPAC_Rjt + ξii jt, (2) 11 I estimate this model using Stata package reghdfe developed by Correia (2017). 12 In Section 2.11 of the online appendix, I discuss why one cannot causally estimate treatment effects on amounts of giving due to post￾treatment censoring. 797 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

Zhao Li as PAC-specific time trends14 that could confound the FIGURE 3.Visual Representation of the estimate of B.Therefore,causal identification of the Difference-in-Differences Design treatment effect B in Specification 1 relies on a parallel- trends assumption:that within a given PAC over time observed rates of itemized giving across donors of dif- E[Giveit] E[Giveit R] ferent partisan leanings would have the same counter- factual trends if the allocation of PAC contributions across parties in each election cycle were held constant In Section A.2.5 of the online appendix,I show that estimates obtained from a modified Granger causality test are indeed consistent with the null hypothesis of no differential trends.Moreover,in the online appendix,I E[Giveijt D] show that estimates obtained from Specification 1 are robust to(1)regressing on lagged rather than same- ………………… cycle PAC contribution patterns (Section A.2.6),(2) using a plausible instrument for PAC contribution pat- terns (Section A.2.7),and (3)controlling additionally t t+1 Cycle for a proxy of potential confounding trends(Section PAC Rjt A.2.8).In all three cases,I obtain estimates that are qualitative identical and quantitatively similar to those from Specification 1. 1/2 小。：。。te。。n Because OpenSecrets'data spans 14 election cycles (Center for Responsive Politics 2018),I also explore two types of dynamic patterns.First,I examine whether 4号 the treatment effect,B.has changed over time by al- lowing it to have a linear time trend.Second.I test for t+1 Cycle persistence in the treatment effect on donor behavior. One such persistent effect is entry:eligible but inactive donors might be more likely to start donating to access- seeking PACs when a greater share of PAC contribu- tions support their preferred party.I construct Entryr which equals 1 if donor i's first itemized donation to where donor i and i'both belong to PAC j,and i= PACjoccurred in cycle t;equals 0 for all cycles prior to Areiit-Arer it.For a visual representation of the identi- t;and is coded as NA for all cycles after t.When sub- 是 fication strategy for B in Equation(2),consider a hypo- stituting the outcome variable in Specification 1 with thetical PAC j where,for simplicity,I assume that every PAC donor is either a pure Republican (i.e.,Indiv_Ri= 1/2)or a pure Democrat (i.e.,Indiv_R:=-1/2).Sup- more Republican leaning than the employees,(2)the executives can steer the company's PAC towards giving to Republican candidates, pose PAC j gave only to Democratic recipients dur- and(3)the likelihood of giving to the company's PAC depends solely ing cycle t (i.e.,PAC_Rit=-1/2)and exclusively to on one's income.Under these conditions,the executives will be more Republican recipients in the next cycle (PAC_Rit+1= likely to give to the PAC.At the same time,this PAC's contributions 1/2),as shown in the bottom half of Figure 3.If the will also appear more aligned with the executives'partisan prefer- average rate of Republican donors giving to PAC j ences.The estimate of B would be positive even though partisan changed by y between the two cycles,and the same av- preference,by assumption,plays no role in the decision to give to the company's PAC. erage rate for Democratic donors changed by x,then B 14 Not allowing time trends to vary by PAC could also cause spuri- =y-x is the causal effect of the degree of alignment ous correlations.For example,suppose donors'willingness to give to between PAC contributions and donors'partisan pref- access-seeking PACs depends only on the rates of investment returns to PAC contributions.Also,assume such returns are higher for the erences on rates of itemized donations,as shown in the energy sector when Republicans dominate the House of Represen- top half of Figure 3.Suppose B=0.10,the interpreta- tatives,and that the opposite holds for the technology sector.More- tion would be that as a result of a full-unit increase in over,suppose workers in the energy sector lean Republican while the share of PACcontributions to one party across con- technology workers lean Democratic.Then,if majority control over eys secutive cycles,the change in the absolute probability the House changes from Democratic to Republican across consecu- that donors who prefer said party will donate to the tive cycles t and t+1,two things will happen.First,because access- seeking PACs favor incumbents,PACs representing both sectors will PAC will be 10 percentage points higher than that for give a greater share of their contributions to Republicans in t+1 donors who prefer the opposite party. than int.Second,returns to PAC contributions will rise for the energy The two fixed effects in Specification 1,ai and tit, sector and fall for the technology sector.These factors together could cause donors in the energy sector to give at a higher rate to their control for all time-invariant donor attributes as well PACs in the second cycle than in the first,and the opposite could be true for donors in the technology sector.Without accounting for PAC-specific trends here,comparing within-person changes in rates 13 Time-invariant donor attributes could generate spurious correla. of giving would lead to the false conclusion that donors'willingness tions between the degree to which PACcontributions appear aligned to give depends on the degree of alignment between PAC activities with donors'partisan preferences and donors'rates of itemized giv- and donors'partisan preference (when there is no such effect by as- ing.Consider a hypothetical company where (1)the executives are sumption). 798

Zhao Li FIGURE 3. Visual Representation of the Difference-in-Differences Design where donor i and i both belong to PAC j, and ξii jt = ti jt − ti jt . For a visual representation of the identi￾fication strategy for β in Equation (2), consider a hypo￾thetical PAC j where, for simplicity, I assume that every PAC donor is either a pure Republican (i.e.,Indiv_Ri = 1/2) or a pure Democrat (i.e.,Indiv_Ri = −1/2). Sup￾pose PAC j gave only to Democratic recipients dur￾ing cycle t (i.e., PAC_Rjt = −1/2) and exclusively to Republican recipients in the next cycle (PAC_Rj,t+1 = 1/2), as shown in the bottom half of Figure 3. If the average rate of Republican donors giving to PAC j changed by y between the two cycles, and the same av￾erage rate for Democratic donors changed by x, then β = y − x is the causal effect of the degree of alignment between PAC contributions and donors’ partisan pref￾erences on rates of itemized donations, as shown in the top half of Figure 3. Suppose β = 0.10, the interpreta￾tion would be that as a result of a full-unit increase in the share of PAC contributions to one party across con￾secutive cycles, the change in the absolute probability that donors who prefer said party will donate to the PAC will be 10 percentage points higher than that for donors who prefer the opposite party. The two fixed effects in Specification 1, αij and τ jt, control for all time-invariant donor attributes13 as well 13 Time-invariant donor attributes could generate spurious correla￾tions between the degree to which PAC contributions appear aligned with donors’ partisan preferences and donors’ rates of itemized giv￾ing. Consider a hypothetical company where (1) the executives are as PAC-specific time trends14 that could confound the estimate of β. Therefore, causal identification of the treatment effect β in Specification 1 relies on a parallel￾trends assumption: that within a given PAC over time, observed rates of itemized giving across donors of dif￾ferent partisan leanings would have the same counter￾factual trends if the allocation of PAC contributions across parties in each election cycle were held constant. In Section A.2.5 of the online appendix, I show that estimates obtained from a modified Granger causality test are indeed consistent with the null hypothesis of no differential trends. Moreover, in the online appendix, I show that estimates obtained from Specification 1 are robust to (1) regressing on lagged rather than same￾cycle PAC contribution patterns (Section A.2.6), (2) using a plausible instrument for PAC contribution pat￾terns (Section A.2.7), and (3) controlling additionally for a proxy of potential confounding trends (Section A.2.8). In all three cases, I obtain estimates that are qualitative identical and quantitatively similar to those from Specification 1. Because OpenSecrets’ data spans 14 election cycles (Center for Responsive Politics 2018), I also explore two types of dynamic patterns. First, I examine whether the treatment effect, β, has changed over time by al￾lowing it to have a linear time trend. Second, I test for persistence in the treatment effect on donor behavior. One such persistent effect is entry: eligible but inactive donors might be more likely to start donating to access￾seeking PACs when a greater share of PAC contribu￾tions support their preferred party. I construct Entryijt which equals 1 if donor i’s first itemized donation to PAC j occurred in cycle t; equals 0 for all cycles prior to t; and is coded as NA for all cycles after t. When sub￾stituting the outcome variable in Specification 1 with more Republican leaning than the employees, (2) the executives can steer the company’s PAC towards giving to Republican candidates, and (3) the likelihood of giving to the company’s PAC depends solely on one’s income. Under these conditions, the executives will be more likely to give to the PAC. At the same time, this PAC’s contributions will also appear more aligned with the executives’ partisan prefer￾ences. The estimate of β would be positive even though partisan preference, by assumption, plays no role in the decision to give to the company’s PAC. 14 Not allowing time trends to vary by PAC could also cause spuri￾ous correlations. For example, suppose donors’ willingness to give to access-seeking PACs depends only on the rates of investment returns to PAC contributions. Also, assume such returns are higher for the energy sector when Republicans dominate the House of Represen￾tatives, and that the opposite holds for the technology sector. More￾over, suppose workers in the energy sector lean Republican while technology workers lean Democratic. Then, if majority control over the House changes from Democratic to Republican across consecu￾tive cycles t and t + 1, two things will happen. First, because access￾seeking PACs favor incumbents, PACs representing both sectors will give a greater share of their contributions to Republicans in t + 1 than in t. Second, returns to PAC contributions will rise for the energy sector and fall for the technology sector.These factors together could cause donors in the energy sector to give at a higher rate to their PACs in the second cycle than in the first, and the opposite could be true for donors in the technology sector. Without accounting for PAC-specific trends here, comparing within-person changes in rates of giving would lead to the false conclusion that donors’ willingness to give depends on the degree of alignment between PAC activities and donors’ partisan preference (when there is no such effect by as￾sumption). 798 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

How Internal Constraints Shape Interest Group Activities TABLE 1. Differences-in-differences TABLE 2. Differences-in-Differences Estimates for Giveijt Estimates for Entryijt (1) (2) (1) (2) PAC_Rit x Indiv_Ri 0.0853** 0.0846* PAC_Rit x Indiv_Ri 0.0400** 0.0357* (0.00813) (0.00822) (0.00676) (0.00643) Donor-PAC fixed effect Y Y Donor-PAC fixed effect Y Y PAC-cycle fixed effect Y Y PAC-cycle fixed effect Y Y 2647085 3516248 1568054 2077662 Sample pure partisans all partisans Sample pure partisans all partisans Standard errors are clustered at the PAC level and reported in Standard errors are clustered at the PAC level and reported in parentheses. p<0.05,+p<0.01,**p<0.001 parentheses. p<0.05,#p<0.01,p<0.001 this indicator for donor entries,I expect the estimated opposite effect holds for pure partisan donors who sup- coefficient on Indiv_Rix PAC_Rit to be positive. port the opposite party. At the same time,active donors might be more likely One should note that limitations of campaign fi- to permanently stop donating following a decline in the nance records may considerably attenuate the esti- share of PAC contributions to copartisan candidates mates shown in Table 1.Recall that the outcome vari- To capture such exit effect,I construct Exitiit such that able,Giverjt,equals 0 whenever donor i did not make it equals 0 if donor i gave an itemized donation to PAC any itemized donation to PAC j during cycle t.This 4号元 j in cycle t;equals 1 if donor i permanently stopped could occur either because donor i chose not to do- making such donations to PAC j since cycle t;and is nate,or because donor i was simply ineligible to give coded as NA for all other cycles.Since exits should be (e.g.,donor i had not joined the parent organization of inversely related to donors'willingness to give,I expect PAC j in cycle t or had already left the organization by Indiv_Ri x PAC_Rit to have a negative estimated coef- then).Because the two scenarios are indistinguishable ficient when I substitute the outcome variable in Spec- in campaign finance records,the coefficient estimates ification 1 with this indicator for donor exits. for the interaction term PAC_Rit x Indiv_Ri reported in Table 1 may be biased towards zero by the inclusion Results of donor-cycles during which donors were ineligible to give and hence were not "treated"by changes in the 是 Consistent with my hypothesis,donors are less likely to allocations of PAC contributions across parties. give itemized donations to access-seeking PACs when There is suggestive evidence that the treatment ef- the share of PAC contributions to copartisan recipi- fects shown in Table 1 have risen over time.When I ents declines.Table 1 presents the coefficient estimates allow for a linear time trend in the treatment effect for two versions of Specification 1,where the outcome B in Specification 1,the relationship between donors' variable is Giverit,an indicator for whether donor i gave rates of itemized giving and the share of PAC con- any itemized donation to PACjduring cycle t.The first tributions to donors'copartisan recipients appears to column shows estimate treatment effect on the subset have strengthened over time (statistically significant of PAC donors who are "pure partisans,"i.e..they each at the 10%level).As shown in Section A.2.12 of the have only given to recipients of one party throughout online appendix,the estimated treatment effects on their donation histories.In comparison,the second col- donors'rates of giving have nearly doubled between umn use "all partisans,"i.e.,all donors that have given 1990 and 2016.This may have resulted from a concur- direct donations to recipients of either party. rent rise in polarization at both elite and mass levels In both cases,the estimated treatment effect B is sta- (McCarty,Poole,and Rosenthal 2016;Pew Research tistically significant and positive.Consider this base- Center 2017). line:pure partisan donors give itemized donations to The impact on donor behavior due to changes in access-seeking PACs at an average rate of 0.122 per access-seeking PACs'support for donors'preferred 四 cycle.Using column 1 of Table 1 as reference,the ab- parties is not only immediate,but also persistent.By solute rate of giving for pure partisan donors aligned substituting the outcome variable in Specification 1 with one party declines by 0.00839,or a 6.89%reduc- with Entryit,an indicator for donor entries,I can de- tion from the baseline.for a one-standard deviation de- tect whether donors who had not started giving(item- crease in the residualized share of PAC contributions ized donations)to access-seeking PACs would be more to the same party (19.7 percentage points).15 And the likely to do so when the share of PAC contributions to copartisan recipients rose.As Table 2 shows,this pre- diction is true regardless of whether I restrict to pure 15 Unless noted otherwise.I use to the standard deviation of partisan donors or not.Similarly,replacing the out- (PAC_R-PAC_Rj)across PACs j and cycles t to conduct coun- come variable in Specification 1 with Exiti,an indi- terfactual analyses,following Mummolo and Peterson(2018). cator for donor exits,allows me to examine whether 799

How Internal Constraints Shape Interest Group Activities TABLE 1. Differences-in-differences Estimates for Giveijt (1) (2) PAC_Rjt × Indiv_Ri 0.0853∗∗∗ 0.0846∗∗∗ (0.00813) (0.00822) Donor-PAC fixed effect Y Y PAC-cycle fixed effect Y Y N 2647085 3516248 Sample pure partisans all partisans Standard errors are clustered at the PAC level and reported in parentheses. ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 this indicator for donor entries, I expect the estimated coefficient on Indiv_Ri × PAC_Rjt to be positive. At the same time, active donors might be more likely to permanently stop donating following a decline in the share of PAC contributions to copartisan candidates. To capture such exit effect, I construct Exitijt such that it equals 0 if donor i gave an itemized donation to PAC j in cycle t; equals 1 if donor i permanently stopped making such donations to PAC j since cycle t; and is coded as NA for all other cycles. Since exits should be inversely related to donors’ willingness to give, I expect Indiv_Ri × PAC_Rjt to have a negative estimated coef￾ficient when I substitute the outcome variable in Spec￾ification 1 with this indicator for donor exits. Results Consistent with my hypothesis, donors are less likely to give itemized donations to access-seeking PACs when the share of PAC contributions to copartisan recipi￾ents declines.Table 1 presents the coefficient estimates for two versions of Specification 1, where the outcome variable is Giveijt, an indicator for whether donori gave any itemized donation to PAC j during cycle t. The first column shows estimate treatment effect on the subset of PAC donors who are “pure partisans,” i.e., they each have only given to recipients of one party throughout their donation histories. In comparison, the second col￾umn use “all partisans,” i.e., all donors that have given direct donations to recipients of either party. In both cases, the estimated treatment effect βˆ is sta￾tistically significant and positive. Consider this base￾line: pure partisan donors give itemized donations to access-seeking PACs at an average rate of 0.122 per cycle. Using column 1 of Table 1 as reference, the ab￾solute rate of giving for pure partisan donors aligned with one party declines by 0.00839, or a 6.89% reduc￾tion from the baseline, for a one-standard deviation de￾crease in the residualized share of PAC contributions to the same party (19.7 percentage points).15 And the 15 Unless noted otherwise, I use to the standard deviation of (PAC_Rjt − PAC_Rj) across PACs j and cycles t to conduct coun￾terfactual analyses, following Mummolo and Peterson (2018). TABLE 2. Differences-in-Differences Estimates for Entryijt (1) (2) PAC_Rjt × Indiv_Ri 0.0400∗∗∗ 0.0357∗∗∗ (0.00676) (0.00643) Donor-PAC fixed effect Y Y PAC-cycle fixed effect Y Y N 1568054 2077662 Sample pure partisans all partisans Standard errors are clustered at the PAC level and reported in parentheses. ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 opposite effect holds for pure partisan donors who sup￾port the opposite party. One should note that limitations of campaign fi￾nance records may considerably attenuate the esti￾mates shown in Table 1. Recall that the outcome vari￾able, Giveijt, equals 0 whenever donor i did not make any itemized donation to PAC j during cycle t. This could occur either because donor i chose not to do￾nate, or because donor i was simply ineligible to give (e.g., donor i had not joined the parent organization of PAC j in cycle t or had already left the organization by then). Because the two scenarios are indistinguishable in campaign finance records, the coefficient estimates for the interaction term PAC_Rjt × Indiv_Ri reported in Table 1 may be biased towards zero by the inclusion of donor-cycles during which donors were ineligible to give and hence were not “treated” by changes in the allocations of PAC contributions across parties. There is suggestive evidence that the treatment ef￾fects shown in Table 1 have risen over time. When I allow for a linear time trend in the treatment effect β in Specification 1, the relationship between donors’ rates of itemized giving and the share of PAC con￾tributions to donors’ copartisan recipients appears to have strengthened over time (statistically significant at the 10% level). As shown in Section A.2.12 of the online appendix, the estimated treatment effects on donors’ rates of giving have nearly doubled between 1990 and 2016. This may have resulted from a concur￾rent rise in polarization at both elite and mass levels (McCarty, Poole, and Rosenthal 2016; Pew Research Center 2017). The impact on donor behavior due to changes in access-seeking PACs’ support for donors’ preferred parties is not only immediate, but also persistent. By substituting the outcome variable in Specification 1 with Entryijt, an indicator for donor entries, I can de￾tect whether donors who had not started giving (item￾ized donations) to access-seeking PACs would be more likely to do so when the share of PAC contributions to copartisan recipients rose. As Table 2 shows, this pre￾diction is true regardless of whether I restrict to pure partisan donors or not. Similarly, replacing the out￾come variable in Specification 1 with Exitijt, an indi￾cator for donor exits, allows me to examine whether 799 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

Zhao Li TABLE 3. Differences-in-Differences PAC if it supports a greater proportion of out-partisan Estimates for Exitijt candidates. (1) (2) Sampling Method and Respondent Characteristics PAC_Rit×Indiv_.R -0.0529 -0.0630* (0.0211) (0.0214) In this subsection,I describe my sampling method and Donor-PAC fixed effect Y Y demonstrate that my respondents are highly relevant PAC-cycle fixed effect Y Y for the purpose of this paper.Each donor targeted for N 529383 742791 recruitment received a letter sent to the address listed Sample pure partisans all partisans in their most recent itemized donation recorded in the Database on Ideology,Money in Politics,and Elections Standard errors are clustered at the PAC level and reported in parentheses. (Bonica 2016b),updated through the 2016 election cy- P<0.05,* p<0.01,**p<0.001 cle.The letters provide a link to an online survey,which I describe in detail in Section A.3.2 of the online ap- pendix.These donors were selected from one of two sampling frames. donors who had been giving to access-seeking PACs were less likely to permanently stop doing so when First,I recruited 413 respondents out of 3,500 in- the PAC's support for copartisan candidates grew.As dividual donors (response rate 11.8%)in what I call expected,coefficient estimates from both models in the Compustat sampling frame.These are individual Table 3 are statistically significant and negative. donors randomly selected on two conditions:(1)since The presence of entry and exit effects suggests that the 2010 federal election cycle,each of them has given at least one itemized donation to a PAC sponsored by donors'partisan preferences may have long-lasting im- 4号元 pact on the fundraising of access-seeking PACs.When one of the publicly traded companies in the Compustat North America database (Compustat North America these PACs give a greater share of their contributions 2017),and (2)each has donated exclusively to candi- to one party,they gain new donors who support the party,but they also permanently lose active donors who dates from one of the two major parties.7 & Additionally.for a different study.I also recruited prefer the opposite party.As a result,access-seeking PACs might be wise not to overreact to changing par- 1,347 respondents out of 16,391 individual donors(re- sponse rate 8.22%)in what I call the other sampling tisan power in Congress,since complications to long term fundraising could outweigh the benefits of short- frame.These are randomly selected individual donors who also give exclusively to candidates from one of the term access.Nonetheless,because donor entries and two major parties,though not all of them have experi- exits may be mismeasured due to limitations in cam- paign finance records discussed earlier,the estimated ences with corporate or other access-oriented PACs. treatment effects shown in Tables 2 and 3 should be Among the 1,347 donors from the other sampling frame,198 of them self-reported having had an em- taken as suggestive evidence rather than conclusive findings.16 ployer that sponsors a PAC in a screening question. These 198 individuals,together with the 413 respon- P135501c90 dents from the Compustat sampling frame,form the STUDY 2:ORIGINAL SURVEY OF RECENT subset of respondents that I hereafter refer to as the PAC DONORS known PAC donors.The remaining respondents are collectively called other donors. Having verified that donors are less likely to give to Section A.3.1 of the online appendix describes the access-seeking PACs when a greater share of PAC con- demographic and political characteristics of the re- tributions goes to out-partisan(rather than copartisan) spondents.Here I highlight three features.First,the politicians in my difference-in-differences analysis of known PAC donors are comparable to the population campaign finance records,I now turn to my original of donors to access-seeking PACs.For example,among survey of corporate PAC donors to unpack the under- respondents from the Compustat sampling frame,the lying causal mechanism.First,I show that respondents average itemized donation per cycle since the 2008 who have recently given to PACs sponsored by pub- election cycle is $548 while the median is $217 In com- licly traded companies are well-informed about how parison,the corresponding average for all donors who eys their PACs allocate contributions across parties.Sec- gave to access-seeking PACs since the 2008 election cy- ond,I replicate my observational finding in a preregis- cle is $1,231,and the corresponding median is $500.The tered experiment in the survey:when learning about a known PAC donors thus appear representative of the hypothetical corporate PAC's(randomly assigned)al- rank-and-file donors to access-seeking PACs. location of contributions across parties,donors in my Second,the known PAC donors come from at survey report much lower willingness to donate to said least ten sectors (according to OpenSecret's employ- ment sector classification)that are responsible for the 16 I could not replicate such entry and exit effects in my survey ex. periment,though the survey is not well-suited to study persistence effects in donor behavior.See Section A.3.8 of the online appendix 17 At the beginning of Section A.3.1 of the online appendix.Iexplain for detail. why I restrict my sample to donors who have only given to one party. 800

Zhao Li TABLE 3. Differences-in-Differences Estimates for Exitijt (1) (2) PAC_Rjt × Indiv_Ri –0.0529∗ –0.0630∗∗ (0.0211) (0.0214) Donor-PAC fixed effect Y Y PAC-cycle fixed effect Y Y N 529383 742791 Sample pure partisans all partisans Standard errors are clustered at the PAC level and reported in parentheses. ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 donors who had been giving to access-seeking PACs were less likely to permanently stop doing so when the PAC’s support for copartisan candidates grew. As expected, coefficient estimates from both models in Table 3 are statistically significant and negative. The presence of entry and exit effects suggests that donors’ partisan preferences may have long-lasting im￾pact on the fundraising of access-seeking PACs. When these PACs give a greater share of their contributions to one party, they gain new donors who support the party, but they also permanently lose active donors who prefer the opposite party. As a result, access-seeking PACs might be wise not to overreact to changing par￾tisan power in Congress, since complications to long￾term fundraising could outweigh the benefits of short￾term access. Nonetheless, because donor entries and exits may be mismeasured due to limitations in cam￾paign finance records discussed earlier, the estimated treatment effects shown in Tables 2 and 3 should be taken as suggestive evidence rather than conclusive findings.16 STUDY 2: ORIGINAL SURVEY OF RECENT PAC DONORS Having verified that donors are less likely to give to access-seeking PACs when a greater share of PAC con￾tributions goes to out-partisan (rather than copartisan) politicians in my difference-in-differences analysis of campaign finance records, I now turn to my original survey of corporate PAC donors to unpack the under￾lying causal mechanism. First, I show that respondents who have recently given to PACs sponsored by pub￾licly traded companies are well-informed about how their PACs allocate contributions across parties. Sec￾ond, I replicate my observational finding in a preregis￾tered experiment in the survey: when learning about a hypothetical corporate PAC’s (randomly assigned) al￾location of contributions across parties, donors in my survey report much lower willingness to donate to said 16 I could not replicate such entry and exit effects in my survey ex￾periment, though the survey is not well-suited to study persistence effects in donor behavior. See Section A.3.8 of the online appendix for detail. PAC if it supports a greater proportion of out-partisan candidates. Sampling Method and Respondent Characteristics In this subsection, I describe my sampling method and demonstrate that my respondents are highly relevant for the purpose of this paper. Each donor targeted for recruitment received a letter sent to the address listed in their most recent itemized donation recorded in the Database on Ideology,Money in Politics, and Elections (Bonica 2016b), updated through the 2016 election cy￾cle. The letters provide a link to an online survey, which I describe in detail in Section A.3.2 of the online ap￾pendix. These donors were selected from one of two sampling frames. First, I recruited 413 respondents out of 3,500 in￾dividual donors (response rate 11.8%) in what I call the Compustat sampling frame. These are individual donors randomly selected on two conditions: (1) since the 2010 federal election cycle, each of them has given at least one itemized donation to a PAC sponsored by one of the publicly traded companies in the Compustat North America database (Compustat North America 2017), and (2) each has donated exclusively to candi￾dates from one of the two major parties.17 Additionally, for a different study, I also recruited 1,347 respondents out of 16,391 individual donors (re￾sponse rate 8.22%) in what I call the other sampling frame. These are randomly selected individual donors who also give exclusively to candidates from one of the two major parties, though not all of them have experi￾ences with corporate or other access-oriented PACs. Among the 1,347 donors from the other sampling frame, 198 of them self-reported having had an em￾ployer that sponsors a PAC in a screening question. These 198 individuals, together with the 413 respon￾dents from the Compustat sampling frame, form the subset of respondents that I hereafter refer to as the known PAC donors. The remaining respondents are collectively called other donors. Section A.3.1 of the online appendix describes the demographic and political characteristics of the re￾spondents. Here I highlight three features. First, the known PAC donors are comparable to the population of donors to access-seeking PACs. For example, among respondents from the Compustat sampling frame, the average itemized donation per cycle since the 2008 election cycle is $548 while the median is $217. In com￾parison, the corresponding average for all donors who gave to access-seeking PACs since the 2008 election cy￾cle is $1,231, and the corresponding median is $500.The known PAC donors thus appear representative of the rank-and-file donors to access-seeking PACs. Second, the known PAC donors come from at least ten sectors (according to OpenSecret’s employ￾ment sector classification) that are responsible for the 17 At the beginning of Section A.3.1 of the online appendix, I explain why I restrict my sample to donors who have only given to one party. 800 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382

How Internal Constraints Shape Interest Group Activities FIGURE 4.Jittered Scatter Plot of Guessed versus Actual Percentage of PAC Contributions to Republican Candidates Percent of PAC Contributions to Republican Candidates 100% 75% respondents' 50% guesses 25% 0% 0% 25% 50% 75% 100% 4号元 actual values & overwhelming portion of PAC contributions from from Figure 4,while respondents need not know ex- access-seeking interest groups.In descending order, actly what percentage of their respective PAC's con- they are (1)communications/electronics,(2)finance, tributions went to candidates of each party-a high insurance,and real estate,(3)miscellaneous business, bar for rank-and-file PAC donors-75%of the respon- (4)health,(5)defense,(6)energy and natural re- dents know at least which party's candidates received sources,(7)agribusiness,(8)transportation,(9)con more money from respondents'corresponding PACs 是 struction,and (10)other.In the 2016 election cycle, Moreover,Table 4 shows the distribution of the abso- these ten sectors collectively accounted for 71.5%of lute differences between the guessed versus actual per- all PAC contributions(not just access-seeking PACs). centages,which are fairly small except for some out- Third,both parties are represented in the survey liers.The Pearson correlation between the guesses ver- samples.Among known PAC donors,40.2%give ex- sus the corresponding actual values is 0.297 clusively to Republican candidates while the rest do- To my best knowledge,Figure 4 and Table 4 repre- nate only to Democrats.For the other donors,25.1% sent the first set of evidence that donors are informed are loyal Republican donors while the rest give solely about how their corporate PACs allocate contributions to Democrats. across parties.Section A.3.3 of the online appendix provides detailed arguments for the external validity of this finding.I therefore verify one important precon- Donor Knowledge of PAC Contributions dition for donors'partisan preferences to affect their My survey suggests that donors to access-seeking PACs willingness to donate to access-seeking PACs.More- are highly informed about how their PACs allocate over,my finding contrasts with existing research on contributions across parties.To assess donors'knowl- other corporate stakeholders such as consumers.For edge of such information,I asked each of the 413 example,when presented with multiple-choice ques- respondents from the Compustat sampling frame to tions on basic facts about PAC contributions made by guess the percentage of their PAC's contributions that household brands,consumers'answers were no bet- went to Republican(as opposed to Democratic)can- ter than random chance(Panagopoulos et al.2016).In didates in the 2016 federal election cycle.Figure 4 is a contrast,among the 413 respondents from the Com- jittered scatter plot of the average respondent's guess pustat sampling frame,a majority of their guesses on against the actual percentage of PAC contributions to the percentage of their PAC's contributions to Repub- Republican candidates,along with a fitted Loess curve lican candidates were within 14.1 percentage points of and a dashed 45-degree line for reference.18 As evident the actual values,about only two thirds of a standard deviation of within-PAC changes in the percentage of PAC contributions to Republican recipients in actual The data points are jittered so as to avoid identifying individual campaign finance data(19.7 percentage points).That respondents from their company's donation records. most survey respondents are well informed about how 801

How Internal Constraints Shape Interest Group Activities FIGURE 4. Jittered Scatter Plot of Guessed versus Actual Percentage of PAC Contributions to Republican Candidates overwhelming portion of PAC contributions from access-seeking interest groups. In descending order, they are (1) communications/electronics, (2) finance, insurance, and real estate, (3) miscellaneous business, (4) health, (5) defense, (6) energy and natural re￾sources, (7) agribusiness, (8) transportation, (9) con￾struction, and (10) other. In the 2016 election cycle, these ten sectors collectively accounted for 71.5% of all PAC contributions (not just access-seeking PACs). Third, both parties are represented in the survey samples. Among known PAC donors, 40.2% give ex￾clusively to Republican candidates while the rest do￾nate only to Democrats. For the other donors, 25.1% are loyal Republican donors while the rest give solely to Democrats. Donor Knowledge of PAC Contributions My survey suggests that donors to access-seeking PACs are highly informed about how their PACs allocate contributions across parties. To assess donors’ knowl￾edge of such information, I asked each of the 413 respondents from the Compustat sampling frame to guess the percentage of their PAC’s contributions that went to Republican (as opposed to Democratic) can￾didates in the 2016 federal election cycle. Figure 4 is a jittered scatter plot of the average respondent’s guess against the actual percentage of PAC contributions to Republican candidates, along with a fitted Loess curve and a dashed 45-degree line for reference.18 As evident 18 The data points are jittered so as to avoid identifying individual respondents from their company’s donation records. from Figure 4, while respondents need not know ex￾actly what percentage of their respective PAC’s con￾tributions went to candidates of each party—a high bar for rank-and-file PAC donors—75% of the respon￾dents know at least which party’s candidates received more money from respondents’ corresponding PACs. Moreover, Table 4 shows the distribution of the abso￾lute differences between the guessed versus actual per￾centages, which are fairly small except for some out￾liers. The Pearson correlation between the guesses ver￾sus the corresponding actual values is 0.297. To my best knowledge, Figure 4 and Table 4 repre￾sent the first set of evidence that donors are informed about how their corporate PACs allocate contributions across parties. Section A.3.3 of the online appendix provides detailed arguments for the external validity of this finding. I therefore verify one important precon￾dition for donors’ partisan preferences to affect their willingness to donate to access-seeking PACs. More￾over, my finding contrasts with existing research on other corporate stakeholders such as consumers. For example, when presented with multiple-choice ques￾tions on basic facts about PAC contributions made by household brands, consumers’ answers were no bet￾ter than random chance (Panagopoulos et al. 2016). In contrast, among the 413 respondents from the Com￾pustat sampling frame, a majority of their guesses on the percentage of their PAC’s contributions to Repub￾lican candidates were within 14.1 percentage points of the actual values, about only two thirds of a standard deviation of within-PAC changes in the percentage of PAC contributions to Republican recipients in actual campaign finance data (19.7 percentage points). That most survey respondents are well informed about how 801 Downloaded from https://www.cambridge.org/core. Shanghai JiaoTong University, on 26 Oct 2018 at 03:53:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0003055418000382