The second core mechanism is competition . Because federal units compete for mobile businesses and residents, sub-national governments can be pressured to adopt attractive policies pursued in other units—or risk losing tax revenue and economic activity.The policy diffusion perspective has been highly fruitful. It has shed light on the degree to which policy decisions by governments are interdependent and explored several compelling mechanisms that drive this interdependence. But, I argue, this perspective is incomplete. It fails to fully account for the role of interest groups in the policy process—and how prior policy decisions across the federal system shape interest group politics. Policy diffusion scholarship focuses primarily on re-election motivated lawmakers who learn and compete because they, broadly speaking, want to produce good policy outcomes for their constituents. Yet, we know that much more goes into policy decisions besides lawmakers seeking good policy. Significant policy reforms usually represent just the final outcomes at the tail end of hard-fought political battles—which generally continue post-enactment in the implementation phase. These battles can draw a diverse array of interest groups like businesses, unions, and citizens groups, as well as government bureaucrats. Painting a complete picture of policy interdependence in American federalism therefore requires considering how prior policy decisions adopted across the federal system construct and empower political actors engaging across the federal system. That is the approach taken in this dissertation. I focus primarily on organized economic interests,indoor vertical garden system whose engagement is among the strongest drivers of policy decisions in American politics broadly , and who are particularly important in considering policy reforms that affect sectors of the economy.

Literature studying “policy feedback” has demonstrated the powerful ways in which the public policy landscape affects the representation of organized economic interests in the political system . For instance, the public policy landscape shapes the types of firms that grow and prosper—and as a result, which have the capacity to influence politics . Similarly, public policies like collective bargaining rules affect the ability of unions to grow and maintain membership, which in turn influences their political sway . Shifts 1 In the international sphere, Elkins and Simmons similarly categorize diffusion as either “adaptation to altered conditions” and learning. in policy, therefore, can affect the power of different organized economic interests in the political system. In a federal system of government in which states have significant authority and interest groups are active at multiple sites and levels of government, I argue that policy feedback effects on interest group politics can also generate powerful policy inter dependencies—in some cases driving the spread of policies across jurisdictions. More specifically, state-level reforms can increase the political power of interest group coalitions supporting the geographic and jurisdictional expansion of those reforms. These dynamics can play out, first, horizontally across the states. State-level reforms that benefit existing organized interests, or give rise to new ones, also tend to strengthen them politically. The groups that benefit from particular state-level reforms are likely to also benefit from the propagation of those reforms to other states. Thus, these groups might apply newfound strength to propagate reforms horizontally through lobbying and other political activities. The political implications of state-level reforms are not restricted to other states. The groups that benefit, and are politically strengthened, by a state-level reform might also leverage newfound strength to advocate for aligned reforms at the federal level. The geographic structure of representation in Congress provides a key avenue for this type of vertical, state-national feedback. Members of Congress represent geographically demarcated districts that are embedded in state policy landscapes.

Shifts to those landscapes precipitated by state policy reforms can in turn affect the political pressures that members face. More specifically, to the extent that state policy reforms influence state political economies, this can affect the ability of organized economic interests to engage in politics and make demands on their representatives. Finally, reforms achieved at the state level can affect the national interest group politics by drawing new actors into pro-reform coalitions. This dynamic is particularly relevant in considering the engagement of philanthropists, a growing topic of study in political science . State policy experiments can provide a proof-of-concept of the legitimacy of some set of reforms, and thus draw philanthropic investment. Once invested, foundations might use their financial resources to fund advocacy groups working to propagate those new policies. In the empirical portion of the dissertation, I apply this new theoretical perspective on policy interdependence in American federalism to three policy cases: rooftop solar policy, marijuana policy, and charter school policy. These are each areas in which state governments have taken the lead on driving forward policy reforms with major implications for sectors of the economy, and where, as I show, state government action has had implications for the interest group politics in the broader federal system.Even more so than in the case of rooftop solar, state actions have precipitated a major shift in marijuana policy over the past 20 years. Since California pioneered legalization of marijuana for medical use in 1996, 32 other states and Washington D.C. have followed suit. As of 2020, 15 states had also legalized marijuana for recreational use. This represents a profound shift from the policy regime associated with the War on Drugs that was initiated in the 1970’s. And, like in the case of rooftop solar, these policy shifts have also engendered shifts in the interest group politics. In particular, the advent of adult-use legalization, pioneered by Colorado and Washington in 2012, has driven rapid growth in the marijuana industry from just 3.5 billion dollars of revenue in 2014 to over 13.5 billion dollars of revenue in 2019.

This has led the industry to develop a greater political presence, both in the states and at the federal level. The costs from federal prohibition have led the industry, unlike in the case of rooftop solar, to focus to a greater extent on federal policy than propagating reforms across the states. Federal lobbying from marijuana industry rose from just $45,000 in 2012 to $6 million in 2019. And members of Congress representing legalizing states have, I show, become critical allies in efforts to liberalize federal marijuana policy and resolve costly state-federal legal tension. Take Cory Gardner , for instance. There is little in Gardner’s record prior to 2012 that would indicate he would become an important marijuana proponent. Yet, during his tenure in the Senate , Gardner became a central figure in federal marijuana policy. In 2018, Gardner vowed to block judicial nominees in the Senate until he received a commitment that the federal government would not prosecute marijuana industry . In the 116th Congress, Gardner sponsored core marijuana-related legislation including the SAFE Banking Act and the STATES Act. It is no coincidence that Gardner represents the state of Colorado,clone rack which has one of the strongest marijuana industries in the country. Indeed, interview evidence suggests that the sway of marijuana industry and marijuana voters in Jared Polis’s successful 2018 bid for governor was a major reason why Gardner, who anticipated a tough re-election in 2020 , made marijuana such a priority. To test whether the relationship between state-level legalization and representation in Congress generalizes, I leverage exogenous variation in likelihood of legalization generated by variation across the states in ballot initiative rules. This exogenous variation is necessary due to the inferential challenges in estimating the effects of state policy on national representation. Broadly speaking, to the extent that state policy decisions and representation in Congress are both shaped by factors like a state’s overall ideology, I would expect a correlation between state policy and national representation without any causal relationship. Variation in the availability of citizen initiatives across the states helps to overcome this causal identification problem in the case of marijuana policy. A number of states adopted procedures allowing citizens to enact statutes or constitutional amendments directly through statewide ballot initiatives in the Progressive era of the early 20th century. In the current era, ballot initiatives have been a critical tool for marijuana policy reform. The ability to bypass state legislatures is important because, as one advocate told me, citizens tend to be much more liberal on marijuana issues than their representatives in state legislatures. As a result, legalization efforts have been concentrated in states that allow ballot initiatives, and whether states allow initiatives strongly predicts legalization both for medical and recreational use. At the same time, whether states allow initiatives is not correlated with other factors generally associated with congressional behavior such as measures of ideology. And more importantly, whether states allow initiatives is not associated with member behavior on marijuana issues prior to the wave of state legalization initiated by California in 1996. This suggests that availability of the initiative is a valid instrument for estimating the effect of state legalization on national representation in the contemporary period.

I study the 116th Congress, which, as one journalist put it, was “the first Congress in history where, going into it, it seem[ed] that broad marijuana reforms [were] actually achievable” . Broadly speaking, I find evidence that state legalization affected national representation. Members of Congress representing legalizing states were more likely to sponsor or co-sponsor key pro-marijuana pieces of legislation. They were also more likely to cast certain pro-marijuana roll-call votes. Bringing quantitative evidence and elite interviews together to investigate mechanisms, I find the most support for the role of growing industry influence in legalizing states, but also find some support for the role of the initiative vote in signaling constituent preferences. I find little support for the potential alternative hypothesis that effects were driven by positive shifts to public favorability wrought by legalization. Like in the other two cases, state policy decisions regarding charter schools have driven major shifts to a sector of the economy and society: K-12 education. Charter schools, independent but publicly funded, have grown steadily since the early 2000’s. As Finn, Manno, and Wright write: “Aside, perhaps, from mayoral control, chartering is by far the most significant manifestation of structural and governance innovation in public education…” . In 1999, there were just 507 charter schools operating. By 2017, nearly 7000 charter schools were enrolling over 3 million students—about 7 percent of overall public K-12 enrollment. Charter schools owe their existence to the adoption of “charter laws” across 40 states between 1991 and 2003, which allowed new schools to form apart from the traditional district structure. Unlike in the case of marijuana policy, but like the rooftop solar case, charter school growth presents an existential threat to powerful organized economic interests—teachers unions. Charter schools generally have much lower rates of unionization than traditional public schools. The charter sector’s growth, despite opposition from unions and other incumbent education interests, is notable. It has depended in part, I argue, on the development of a nationally-scoped network of pro-charter advocacy groups—which have on several occasions gone toe-to-toe with powerful teachers unions. In the paper, I examine the role of prior state policy decisions in seeding this pro-charter interest group network. A key difference between the charter school policy case and the other cases studied is in the types of organized interests driving the sustainability and spread of reforms. In rooftop solar policy and marijuana policy, the story is relatively straightforward: state policy decisions gave rise to new industries that leveraged their economic growth to develop greater political influence. While charter growth precipitated by state policies has similarly generated new political interests in the form of large charter networks like Success Academy and KIPP, the political power of these organizations is highly limited. Since charter schools are mostly non-profits with limited revenue streams generally funneled into operations, the political activity of these charter networks has been modest compared to large marijuana and rooftop solar firms. Given that limitation, the financial backing of philanthropists like the Gates Foundation and the Walton Family Foundation has been crucial to building the pro-charter advocacy network. But foundations’ investments in charter advocacy did not arise in a vacuum. Drawing on elite interviews, I show that state policy decisions in the 90’s promoting charter growth in leading states like Minnesota and California were instrumental to generating support from philanthropists and building the pro-charter group coalition.