The Pseudo-Paradox of Partisan Mapmaking and Congressional Competition

The Pseudo-Paradox of Partisan Mapmaking and Congressional Competition Nicholas Goedert Visiting Professor Department of Government and Law Lafayette College August 2015 Contact Information: goedertn@lafayette.edu (240) 281-6518 Department of Government and Law Lafayette College Kirby Hall of Civil Rights 005 Easton, PA 18042

The Pseudo-Paradox of Partisan Mapmaking and Congressional Competition ABSTRACT: Why are fewer congressional elections competitive at the district level when the national electoral environment is at its most competitive? This paper explores this pseudoparadox, and argues that the answer can be found in partisan redistricting. Through an analysis of forty years of congressional elections, we find that partisan gerrymanders induce greater competitiveness as national tides increases, largely due to unanticipated consequences of waves adverse to the map-drawing party, particularly in seats held by that party. The phenomenon anecdotally coined by Grofman and Brunell as the dummymander is thus actually quite common and has significant effects on rates of congressional competition nationally. In contrast, bipartisan maps are shown to induce lower competition, while nonpartisan maps induce higher competition, under all electoral conditions and competitiveness measures. But the effects of partisan gerrymanders on competition, while strong, can only be seen in interaction with shortterm national forces. 1

The link between electoral competition and accountability has been a focus of both public and academic debate since the founding of democracy. In the United States in recent years, much of this debate has targeted congressional elections, with good-government groups and the media worrying about the overwhelming number of seemingly entrenched incumbents and scholars of American politics trying to explain why the number of close elections seemed to be low and declining. This paper explores the question of congressional electoral competition through one common explanation, gerrymandering, interacted with a less common one: shortterm national forces. We find that partisan gerrymandering explains both the variation in congressional competition, and the perverse connection between national partisan balance and the prevalence of close elections at the district level. Much of this concern about noncompetitive elections and entrenched incumbency has highlighted the recent period in the late 1990 s and early 2000 s when competition appears historically low. But this period was immediately followed by three consecutive wave elections from 2006 to 2010 where a significant number of incumbents were defeated, and the make-up of the U.S. Congress changed dramatically. And it does not seem like the research has caught up with this era, either by arguing that these elections were a fluke, or that the overall pattern of reduced competition has now been reversed. And among those policies that influence electoral competition, perhaps none is a more important or live issue among state legislatures than gerrymandering. While recently, some states moves toward more nonpartisan procedures or commissions have come through public referenda (in Florida and California for example), bills to reform the process of districting have been introduced in the legislatures of at least thirty states in the past year (Brennan Center 2015). One of the most common arguments from advocates of such legislation has been the potential for more electoral competition under nonpartisan procedures. And in the case of some enacted 2

reforms, it true that balanced districts are even an explicitly stated goal: the Arizona Proposition creating that state s districting commission requires competitive districts are to be favored as long as they comply with other constitutional requirements (Adams 2005, Kang 2004)). In other cases, the promotion of competition is more subtle: the Legislative Services Bureau responsible for districting in Iowa is not permitted to incorporate incumbency or voting data in creating their proposals. Yet past research on the true impact of redistricting reform on competition is conflicting and incomplete. This paper argues that variations in levels of congressional competition can largely be explained by backfiring partisan gerrymanders when tides turn against the mapmaking party. Past research has recognized this phenomenon and even given it a name (the dummymander ) but has mostly only analyzed it anecdotally. We employ a data set of forty years of congressional elections and finds that partisan dummymanders are not only widespread in the face of big swings in national tides, but also responsible for a larger and superficially paradoxical inverse relationship between national and district-level partisan competition. While the electoral effects of bipartisan gerrymanders and nonpartisan commission maps tend to be resilient to waves (with nonpartisan maps generating consistently more close races than bipartisan ones), partisan maps suppress competition when the national electoral environment is closely balanced, but incite it when one party wins a substantial national majority, largely due to members of the map-drawing party facing unexpectedly close races. This paper incorporates the anecdotal evidence from Grofman and Brunell s dummymanders into a more general theory about the interaction of tides and partisan gerrymandering on congressional competition. In doing so, the results in this paper will demonstrate that backfires of partisan gerrymanders are both historically common and largely responsible for national variations in the competitiveness of congressional elections across 3

election cycles. Additionally, incorporating the role of national tides in district-level competition can potentially resolve much of the confusing and conflicting evidence in past literature on the effects of a state s choice of redistricting institutions. The empirical argument of the paper proceeds as follows. First, we show that an often unacknowledged factor, national tides, strongly and negatively correlates with competitive elections ( A). Next, we demonstrate how this interaction of national and district-level competitiveness is explained by focusing on partisan gerrymanders ( B). Thirdly, we show that partisan gerrymanders increase close elections during strong wave cycles because of unanticipated competition specifically in states with maps drawn by a party suffering through adverse tides ( C). Finally, we show that this increase in competition disproportionately occurs in seats already held by the gerrymandering party ( D). By focusing on each step of this argument in sequence, the results establish both the general impact of tides and redistricting on competition, and the specific role of backfiring partisan gerrymandering on this larger phenomenon. This argument is advanced through an analysis of various measures of competition in a data set encompassing the past four decades of congressional elections. Wave Elections & National Tides As this article argues that short-term shifts in national public opinion are crucial to understanding how gerrymandering impacts electoral competition, we first must operationalize this concept. And as we are primarily concerned with the closeness of electoral competition while caring less about the partisanship of the winner, it is mostly more important to measure how strongly the leading party is favored by rather than which party is favored. Throughout the analysis, this is done through the variable National Tides, defined as the national popular vote margin for the winning party. In contrast, the terms waves or wave elections are used 4

casually and without formal definition to refer to election cycles in which one party wins a large majority of the congressional popular vote. However, these concepts are closely connected: what we commonly refer to as wave elections will be those cycles in which National Tides is highest (e.g. 1974 or 2008 on the Democratic side; 1994 or 2010 on the Republican side). And as National Tides increase, we might also expect district-level election results to deviate more from their normal vote of the baseline partisanship of the district. Specifically, National Tides is the absolute value of National GOP Vote Margin, the aggregated Republican margin in the nationwide congressional popular vote in a given cycle. Our key independent variables of interest are the interactions of National Tides with redistricting institutions. For example, National GOP Vote Margin takes a value of -7.9, and National Tides takes a value of 7.9 for all 2006 data points, because Democrats won the national congressional popular vote that year by 7.9%. In a very closely divided election like 2000, National Tides takes a value 0.3. Note that in our forty-year data set, eleven of the 20 elections saw a national popular vote advantage for the Democrats of greater than 5%, while only two elections saw so great a national advantage for Republicans. So our ability to test our predictions with respect to the interaction of Republican waves may in some cases be limited to anecdotal evidence. 1 Previous Research on Districting and Competition Declining competition in U.S. congressional elections has worried scholars since David Mayhew s seminal article The Case of the Vanishing Marginals (1974). Mayhew does not arrive at an explanation for the paucity of close elections and districts that he observes, but the role of redistricting in fostering or suppressing such competition has been controversial throughout subsequent work. Tufte (1973) argues that redistricting was the cause for a reduction in marginal seats during the 1960 s. However, later studies concluded that redistricting, whether 5

partisan or bipartisan, has had little effect on the competitiveness of seats or the advantages of incumbency (Glazer et al. 1987, Ferejohn 1977). Moreover, Gelman and King (1994) find that temporally proximate redistricting, both partisan and bipartisan, leads to an increase in electoral responsiveness in state elections, measured by the slope of the seats/votes curve. And Gopoian and West (1984) suggest that many partisan maps in the 1980 s appear to have reduced the security of their incumbents, from an analysis of vote margins in two electoral cycles. Scholars took up this topic in perhaps the greatest force in the mid-2000 s, when competitive elections were at their lowest ebb. Cox and Katz (2002) argue that the reapportionment revolution in the 1960 s itself has led to decreased competition by reducing the frequency of quality challengers. Research also commonly points to bipartisan legislative agreements as reducing competition and nonpartisan commissions as encouraging it. Carson Crespin, and Williamson (2004 and 2014) find evidence that courts and commissions increase competition in years immediately following redistricting (using data from 1992 and 2002 in their original article, and 1972-2012 in the 2014 update). Lindgren and Southwell (2013) also argue that independent commissions did reduce average margin of victory in U.S. house elections from 2002-2010. But neither of these articles distinguishes between bipartisan and partisan legislative maps. More generally, Cain et al. (2005) find that an overall decline in competitive elections has tracked with an increase in the number of districts drawn by bipartisan agreement. And McDonald (2006) concludes that legislative districting was responsible for decreased competition in the 2000 s, while nonpartisan commissions facilitated more competitive districts. Yet other recent works claim much less of a role for districting in contributing to this trend. In an examination of state legislative elections in 2000 to 2008 (and using the same definition of competitive elections as this paper), Masket et al. (2012) find little impact of redistricting institutions on the likelihood of a close election. And Abramowitz et al. (2006) find 6

that while much of the decline in competition since the 1970 s can be attributed to the increased partisan polarization of districts, this is explained much more by population sorting than intentional districting. And the specific role of partisan districting in inducing or suppressing competition is perhaps even more ambiguous. Hirsch (2003) focuses on Republican-drawn maps as causing historically low competiveness in the 2002 elections. But employing data from the same election cycle, Yoshinaka and Murphy (2011, 2009) find that partisan maps increase competition and the recruitment of quality challengers, particularly in years immediately following redistricting. But Yoshinaka and Murphy constrain their explanation to the deliberately increased difficulty that out-party members face rather than unanticipated close races faced by in-party members in the face of unexpected waves. And as part of series of articles showing the effects of moving voters to unfamiliar districts, Hood and McKee (2008) found that a mid-decade partisan gerrymander in Georgia made two districts in the state more competitive. Yet the fact that partisan gerrymanders do not always turn out as planned for the controlling party has not gone unnoticed by scholars. Grofman and Brunell (2005), in a series of short case studies referred to as dummymanders, show that many maps drawn by Southern Democrats in the 1990 s failed to anticipate trends favoring the Republican party. Conversely, Seabrook (2010) argues that the effects of Republican partisan maps were largely washed out by mid-decade partisan trends in the 2000 s. However, the global effect of dummymanders on competition, and the frequency of the their occurrence, is largely still unexamined beyond anecdotal evidence and narrow case studies. Recent scholarship by McGhee and Stephanopolous (2015) has tested the efficiency of partisan gerrymanders across an array of possibly electoral environments, but not the specific effect of those maps on competition (see also McGhee 2014). 7

Why Partisan Gerrymanders? It is easy to see why scholars have focused on contrasting bipartisan and nonpartisan maps in their study of gerrymandering and competition; the incentives and constraints of these institutions make it obvious what the hypothesized effect on competition should be. In the case of nonpartisan commissions, increased competition is frequently an explicit or implicit goal given their structure and criteria. On the flip side, without the added incentive of maximizing one party s seats, the uniform desire to protect incumbents of both parties in the case of bipartisan maps clearly incentivizes noncompetitive districts. Figure 1 below acutely exemplifies this in the case of the 2002 California map. But the expected effect of partisan maps on competition is less obvious; the desires of a party to maximize seats and simultaneously protect their own incumbents work at crosspurposes, and it is up to the individual map-maker to choose how much emphasis to place on each goal. Cain (1985) shows how partisan mapmakers packed and cracked opposing party members into a small number of safe districts, and this has become the standard strategy for partisan gerrymanders. Yet, a broad array of tactics are available within this strategy. If too many opposing districts are cracked, mapmakers risk spreading their own party s voters too thinly, leaving many of their own incumbents at risk if political tides shift slightly against them. But if partisans fortify all their own districts safely enough to withstand an unlikely adverse wave, they fail to contest seats that could have been won under more typical electoral environments. Scholars have certainly acknowledged the trade-off facing partisan legislators between maximizing expected seats in a close election and reinforcing their seats against adverse waves. Owen and Grofman (1988) formalize this decision, and argue that California Democrats in the 8

1980 s chose to draw a risk-averse map that drew mostly noncompetitive seats. Kang (2004) frames this trade-off in terms of offensive and defensive gerrymanders. Additionally, Niemi and Deegan (1978) demonstrate formally how neutrality and competitiveness may be at odds when the overall vote is not evenly split, even when a fair map is desired. But there has been very little evidence empirically as to how often partisans choose a riskier strategy, or how often such a strategy backfires, beyond the aforementioned case studies. The range of aggressiveness (i.e. willingness to risk their own party s seats to maximize expected seats) that partisan mapmakers chose in the 2000 s is also shown in Figure 1 below, which gives a visual representation of three Republican-controlled gerrymanders (Pennsylvania, Ohio and Florida) and one bipartisan gerrymander (California) used in the 2002 elections, all anecdotal examples of the phenomenon explored more robustly in the results section. The x-axis in these density plot figures represents how heavily Republican a congressional district is (measured by Cook PVI), while the y-axis represents the share of congressional districts in each state with that level of partisanship (PVI s greater than 20 or less than -20 are coded as 20 and -20 respectively). The difference between the partisan and bipartisan maps is immediately clear. The three Republican maps each have peak on the Republican side of zero, along with several heavily Democratic districts (usually majority African-American). In contrast, California has no competitive districts in the middle of the graph, and a clear bimodal distribution of both strongly Republican and strongly Democratic districts. Under this bipartisan map, we would expect very few close elections unless tides in favor of one party or the other were almost historically extreme. [Figure 1 about here] 9

But there is also variation within the Republican maps with respect to their aggressiveness and thus their robustness to withstand changes in political climate. In Pennsylvania, many congressional districts lie between D+2 and R+4; Republicans in Pennsylvania drew several swing districts that they were counting on factors like incumbency and continued close national elections to hold. Ohio is only slightly less aggressive, with its peak around R+2. But the Florida map, a much more risk-averse gerrymander, containing many districts reinforced from mild swings toward the Democratic Party, peaks around R+5. The electoral effect of these different strategies was manifest in 2006 and 2008, as Republicans lost 38% of their seats in Pennsylvania and Ohio, but only 17% of their seats in Florida. With this in mind, this paper advances a more generalized theory of dummymanders, i.e. when we should observe greater competition under partisan gerrymanders, while largely reaffirming past research contrasting competition under bipartisan legislatures and nonpartisan commissions. Some previous literature not distinguish partisan gerrymanders from bipartisan ones, while other works suggest that partisan maps either suppress competition or increase it only in immediately subsequent elections or in seats specifically targeting by the gerrymandering party. Thus, the corpus of literature on the influence of districting on competition, when taken as a whole, looks scattershot and confusing. This article posits that a third variable, short-term national tides, can realign much of this confusion into a consistent theory, observed through analysis of data across many states and many decades. Previous work has suggested this impact only though the anecdote, showing how an individual wave election can overturn the intended results of a partisan gerrymander. This paper argues that unanticipated consequences of tides adverse to the gerrymandering party are actually driving differences in the competitiveness of congressional elections under partisan maps, and indeed driving nationwide differences in competition levels across election cycles. And by fully incorporating national tides and wave 10

elections into an analysis across all states and several decades, many of the apparent conflicts in the previous, more narrowly drawn, studies can be resolved. Hypotheses Developing from the theory advanced in Owen and Grofman (1988) and examples in Figure 1, this paper hypothesizes that a typical strategy employed by partisan mapmakers will be to draw seats that will be largely safe for their own incumbents under neutral electoral conditions (when the national popular vote is close to even), but which will become increasingly competitive as national tides adverse to the gerrymandering party increase. Thus, the overall effect of partisan gerrymanders on competition can only be considered in interaction with national competition. Individual partisan gerrymanders will make varied choices as to how to balance the competing considerations of maximizing expected seats under neutral conditions and protecting incumbents against adverse tides. But despite these diverse choices, all partisan maps will have the potential to become a dummymander when tides reach a certain magnitude. As adverse tides grow, the number of partisan maps that backfire should increase, leading to more competitive elections where mapmakers expected safe sailing for their own party s incumbents. In contrast, we should not observe this interaction of tides and competition when it comes to bipartisan and nonpartisan maps. Therefore, we would expect to see the following interactions between national tides and redistricting institutions: 2 We expect low competitiveness in districts drawn by bipartisan agreement regardless of electoral environment. 11

We expect high competitiveness among districts drawn by nonpartisan commissions when the national electoral environment is close. 3 As national tides increase, competitiveness in these districts may decline or stay steady. When the national electoral environment is close, we expect low competitiveness among districts drawn by legislatures controlled by one party. As national tides increase, we expect competitiveness to increase, particularly when tides run adverse to the gerrymandering party. In cases where we observe high competition during wave elections under partisan gerrymanders, we expect a greater proportion of competitive races to occur in districts held by the party adverse to tides, relative to both other redistricting regimes and other electoral environments. Figure 2a below depicts a schematic of these hypotheses, showing low competition under bipartisan maps, high competition under nonpartisan maps, and competition under partisan maps increasing with adverse tides. 4 Figure 2b depicts 2a folded in on itself at the middle, so that the x-axis represents the magnitude of a wave without regard to direction, and combines the two partisan lines into one. [Figures 2a & 2b about here] The analysis in this paper will first test the theory as presented in Figure 2b (leveraging the size of tides irrespective of direction) in Results B, as this allows us to test on the entire data set of congressional elections using a straightforward two-way interaction. The analysis will then move on to testing the asymmetric prediction for the partisan maps in 2a in Results C, examining the subsets of the elections data for Democratic and Republican gerrymanders. Note that while we have a large enough subsample to test the asymmetric effects of tides on 12

Democratic gerrymanders in an empirically robust way, limitations on the data largely only allow us to see these same effects in Republican gerrymanders anecdotally. Finally, Results D will test the specific piece of the hypothesis that increased competitiveness under partisan gerrymanders during adverse tides will be found in seats drawn to be held by the gerrymandering party. Measures & Controls Our data consists of all congressional elections falling on a national election day from 1972 (following the first national round of post-wesberry redistricting) through 2010. The process by which each state was redistricted at the start of the decade was coded as Democraticcontrolled, Republican-controlled, bipartisan, nonpartisan, or court. 5 Throughout the analysis, the bipartisan maps, under which we hypothesize low levels of competition regardless of tides, serve as the control group. 6 Maps drawn by courts serve as a separate control for which we provide no particular directional predictions. Errors are clustered by district interacted with decade to account for serial autocorrelation within districts. This analysis attempts to isolate the effects of the redistricting institution and not the effects of the specific districts drawn. Therefore, any controls that might be endogenous to the actual districts have been deliberately omitted. This of course includes district-specific demographics and partisanship, but also candidate-specific data like campaign spending and incumbency. Although all of these factors are of course important to the outcome of a congressional race, the controls that are included (statewide presidential vote, region and statewide demographics, and redistricting institution) are causally prior, and thus the exclusion of district-specific factors should not contaminate the coefficients testing the theory. 7 This 13

requirement is necessarily suspended in the final segment of the analysis, which tests the portion of the hypothesis related to the incumbent party holding the seat. Two district-level variables in particular may come to mind as strongly affecting electoral competition: incumbency and district partisan balance. Yet both are also clearly endogenous to our independent variables of interest. In the case of district partisan balance, we would expect more closely divided districts to generate close election more frequently; however, the number of closely divided districts is typically a deliberate choice of the gerrymander, with some institutions consistently drawing more competitive districts than others (as shown in Appendix B). Controlling out balance at the district level may actually remove some of the genuine effects of gerrymandering regime. Similarly, we would expect more close races in the case of open seats. But open seats themselves may partially be a function of both gerrymandering and tides, as members may choose to retire because their district has been drawn to be much more difficult to run in, or the overall political climate is such that they expect adverse partisan tides. Nevertheless, the inclusion of these two factors, both of which strongly correlate with competitiveness, ultimately has little effect on the influence of partisan tides, gerrymandering institutions, or their interactions; results including both of the variables are shown in Appendix Table C9. We should be much less concerned that these variables are causing, or otherwise correlated in a causally prior way, the gerrymandering institution or national-level tides. To get a complete picture of the effects of redistricting on competition, we employ two measures of competitiveness as independent variables: National Tides (described above) and Statewide Competition, a control for the competitiveness of the state as a whole. Statewide Competition is a measure of the overall partisan balance of the state, with lower values indicating swing states and higher values indicating solidly partisan states. It is used as an important control in the analysis, as it is a measure of competitiveness that is not affected by the shape of 14

congressional districts, and thus causally prior to gerrymandering. It is defined as the absolute value of the difference between the average statewide GOP presidential vote margin and the average national GOP vote margin over the previous two elections for a given district in a given year; i.e. a rough measure of a state s partisan extremism in a given period. E.g. for the California 1 st district in 2006, this variable takes a value of 13, because California voted 13% more Democratic than the nation in the 2000 and 2004 presidential elections. The dependent variable in the analysis is Close Race, a dummy variable coded for each individual congressional race in a specific year as 1 if the election was won by less than 10 points. 8 As we are mainly interested in how often congressional elections are competitive, this is the dependent variable throughout most of the analysis. Overall, 14% of congressional elections are Close Races under this definition. While we do not want to control out district-level variables whose correlation to gerrymandering is causally posterior, we do need to be wary of selection effects from statewide variables that may correlate with gerrymandering institutions in a causally prior way. If states with a particular redistricting regime also share some other quality, it is possible that this other quality may be causing what appears to be an effect of that regime. Specifically, three such effects are anticipated, with specific controls employed in various configurations to account for plausible alternate causal stories. First, states that lean heavily in one partisan direction are more likely to have a gerrymandering process controlled by that party, so the Statewide Competition control is employed in all specifications. Second, because small states may be simultaneous less amenable to drastic gerrymanders and more likely to establish nonpartisan commissions, results are run both with and without a control for state size, while additional robustness checks employing a small state dummy variable as a control are included in the supplemental tables. Finally, because the partisanship of southern states during much of this period tended to be much 15

more Democratic at the state legislative level than the presidential level, results are run both including and excluding the Southern region (defined as the former Confederacy). 9 Results A. The Pseudoparadox of Competition Looking at the partisan balance of a district assuming a neutral environment, as is common in the scholarship, tells us little about what will occur when tides swing distinctly in one direction. For it is immediately apparent that national tides have a dramatic and surprising impact on district-level competition. As discussed above, many of the recent claims about declining congressional competition, in both the media and the literature, occurred in the wake of an era of parity at the national level. Between 1996 and 2004, no party won a majority of the national popular vote in congressional elections, nor did any party win that popular vote by more than 5 points, or win more than 54% of congressional seats. This is actually something of a historical anomaly. Of the twenty election cycles from 1972 to 2010, one party failed to win by at least 5-points in the popular vote only seven times, included these five consecutive years. And during a time when the country appeared so evenly divided, it would be intuitive to expect that many individual races would also be close, but less than 11% of races during this era were decided by 10 points or less (our definition of Close Races). But this era of parity was immediately followed by three consecutive wave elections from 2006 through 2010. And despite the national electorate clearly favoring one party in these three years, the number of races that were closely contested rose to 15%. Recently, it seems that close national competition has led to less competitiveness at the local level, and this phenomenon is born out looking further into the past, at least as far back as the equal-population standard has 16

been applied to congressional districts. The top left quadrant of Figure 3 below shows the correlation of the proportions of Close Races in each cycle 1972-2010 with National Tides; as the national popular vote gets closer, the number of competitive races tends to decline. [Figure 3 about here] As a more rigorous test, Table 1 below shows the effect of National Tides on the percentage of Close Races (each data point is an individual race, with races clustered by state crossed with decade) from 1972-2010, when we control for the competitiveness of individual states by including Statewide Competition. In the first column, we see that the competiveness of a state has very little effect on its propensity toward competitive congressional elections. 10 However, the positive coefficient on National Tides (significant at p<.01), indicates that more extreme wave elections do tend to create more close races. The second column excludes races in the South, with no effect on the National Tides coefficient, indicating the phenomenon cannot merely be explained by increased competitiveness of the Republican party in the South in the most recent decades, or the creation of majority-minority districts in those states. The third column addresses the argument that competition has steadily declined over time, by including a Year variable, as well as the argument that competition increases in years immediately following redistricting (e.g. Yoshinaka and Murphy (2011); Hetherington et al. (2003)); the Gerrymander Year dummy variable takes a value of 1 for election years ending in 2. Although the coefficients for these variables take on the expected sign, neither is statistically significant, and neither mediates the strong effect of National Tides. Finally, we see in the Figure 3 that 1974 appears to be an outlier both in terms of wave strength and number of close races. The fourth column of Table 1 shows that the effect of National Tides on Close Races is still significant at p<.05 even when this cycle is excluded. 11 [Table 1 about here] 17

B. Effects of Gerrymandering So what explains this pseudo-paradox that less national competitiveness correlates with greater local competition in congressional races? Far from being a universal phenomenon, the remaining three quadrants of Figure 3 show that the pseudo-paradox appears limited only to states with partisan gerrymanders. When we isolate only the partisan maps, the negative effect of national competitiveness on local competitiveness is strengthened (and significant at p<.02). But the magnitude of National Tides has no effect on competitive elections under bipartisan maps, which follows if these maps drew districts safe enough for both parties to withstand strong tides in either direction. Moreover, the coefficient for the nonpartisan maps is in the opposite direction of partisan maps, although not significant due to the high variance from the small sample size; this would also follow if such maps tended to draw many naturally competitive districts. The conforms exactly to the schematic in Figure 2b: an upward slope in the case of partisan gerrymanders, a flat line with respect to bipartisan maps, and a downward (though insignificant) slope in the case of nonpartisan maps. Yet it is also possible that these observed differences are merely the result of the types of states that tend to adopt these varying institutions. That is, perhaps swing states tend to adopt nonpartisan regimes or extreme states adopt bipartisan regimes. So we also run a probit analysis of Close Race on National Tides and the various gerrymander dummies including the Statewide Competition control. We also assess the slope of the effect of tides by interacting National Tides with the gerrymander dummies. As before, the unit of analysis in Table 2 below is individual congressional races from 1972-2010, clustered by year crossed with decade; the excluded category is bipartisan maps. The key analyses are shown with and without a control for state size (operationalized as number of districts). 18

[Table 2 about here] Figure 4 below shows the probit coefficients from column 1 for the redistricting variables interpreted for neutral state and national electoral conditions (when National Tides are set to 0). [Figure 4 about here] In all specifications, we see that bipartisan maps and partisan maps create fewer close elections than nonpartisan commissions, when controlling for state ideology and national tides, with the effect of nonpartisan maps large enough to be significant despite the small sample size. From the figure, we estimates that while only 11% of elections overall will be close races with neutral electoral and tides conditions under partisan and bipartisan maps, almost 20% of races will be close under nonpartisan commission maps; this result also seems consistent with the analysis in Appendix B that nonpartisan maps draw more competitive districts at the presidential level. But note that in all these specifications, the effect of National Tides is positive and significant, consistent with A. We can test how this effect varies under different gerrymanders in two ways. First, we can include interaction terms nested into the model run on the entire data set. This is shown in columns 4 through 6 of Table 2, which adds the interactions of each regime with tides. Matching the aggregated data in Figure 3, we see that the pseudo-paradox effect of tides is again largely explained by gerrymandering institutions. When applied to nonpartisan gerrymanders, the effect of National Tides is negative and not significant, but it is even more positive and significant when applied to Republican-drawn maps. In the case of Democratic gerrymanders, the coefficient has the expected sign but is not significant, which is understandable given the small number of Republican wave elections in the data set where we would hypothesize a steep slope among Democratic maps. As second way of analyzing this interacted effect would be to measure the effect of tides on the data subsetted by redistricting regime. Table 3 shows the slope coefficient for National 19

Tides when the probit from the first column of Table 1 (predicting close races using only Tides and Statewide Competition IV s) is run only on subsets of districts drawn under a particular institution. Again, the results conform to our predictions. We see a large positive coefficient for Republican and Democratic gerrymanders, suggesting that partisan maps suffer a backlash when tides go against them, leading to many close races as national tides increase. In contrast, bipartisan gerrymanders, with safe districts drawn so as to be resilient to partisan tides, show no significant effect of tides. The difference between these coefficients for partisan and bipartisan maps is significant at p<.02. In support of our hypotheses, competition at the district level increases as tides increase under partisan maps, but not bipartisan or nonpartisan maps. [Table 3 about here] The results shown in Table 3 predict that as National Tides rise from 0 to 10 (where 10 represents a wave similar to 2008), the number of close races in an average state will rise from 9% to 15% under partisan gerrymanders, remain stable at 13% under bipartisan gerrymanders, and fall from 22% to 19% under nonpartisan maps. This change in estimated proportions is shown in Figure 5, almost perfectly matching the predictions in the schematic Figure 2b. [Figure 5 about here] Note: Continuous versus Dichotomous Tides Throughout the analysis in this paper, National Tides is coded as a continuous variable. Yet Figure 1 shows that many partisan maps may have a specific inflection point, a threshold past where we should start to see effects on competition (e.g. Florida at R+5). This would suggest coding tides as a dichotomous dummy variable, and testing various such thresholds. However, Figure 1 also shows that this inflection point, to the extent that it exists in a given state, varies widely across states. Thus, we hypothesize that when results from all states and decades are 20

aggregated, competition in partisan maps will increase steadily as tides increase. But additionally, we have dichotomized tides at several thresholds as a robustness check, with very similar substantive conclusions. Results replicating Table 2 using a threshold of 6.5%, dividing the data set evenly into wave and non-wave elections, are shown in Supplemental Appendix Table C5-C7, with additional results available from the author. C. Differences by Party of Wave and Gerrymander The analysis thus far has shown that electoral competition conforms to the schematic in Figure 2b, showing interaction with tides without regard to direction. But we have yet to test the asymmetric expectations from partisan gerrymanders depicted in Figure 2a, to actually show that increasing competition under them is generated largely from backfires or dummymanders. We hypothesize that we should only observe greater competition under waves adverse to the map drawing party (i.e. Republican maps under Democratic waves such as 1974 and 2008, and Democratic maps under Republican waves like 1994). And specifically, we would hypothesize that the slope of the National Tides coefficient for the subset of Democratic-drawn maps would be greater when Republicans win the national popular vote than when Democrats win (and conversely for Republican maps). Table 4 below offers support for this hypothesis for Democratic maps, with the difference in responsiveness to tides under Democratic and Republican waves for Democratic gerrymanders significant at p<.01. Unfortunately, performing the same analysis for Republican maps would involve drawing conclusions based on a very small number of elections 12 ; the difference in responsiveness by wave direction for Republican maps is not significant. [Table 4 about here] 21

Although the sample size of elections, particularly Republican-wave elections, limits our ability to demonstrate the different responses to waves under Democratic and Republican regimes in a statistically robust way, we can at least observe these differences anecdotally. Table 5 depicts the percentage of close races in states with Democratic maps in the 1990 s and 2000 s, and in states with Republican maps in the 1970 s and 2000 s. 13 Elections with strong opposing tides, where we would hypothesize a greater number of close races, are shown in italics. [Table 5 about here] With one exception, the results conform to our expectations. In both the 1990 s and the 2000 s, Democratic maps saw more close races in the two Republican wave elections (1994 and 2010), than other years during their respective decades. Republican maps also saw the greatest number of close elections in the post-watergate cycle of 1974, and many more close elections in the Democratic waves of 2006 and 2008 than the previous two cycles. Contrary to Yoshinaka and Murphy (2011), partisan gerrymanders do not seem to increase competition in years immediately following redistricting, but rather in years when tides turn against the party in control. The chart also reveals that one cycle in particular deviates from our predictions: 2010, which saw many close races under both Democratic and Republican maps. 2010 is a very unusual election in that it involved a strong Republican wave immediately succeeding a strong Democratic wave; the difference in popular vote between 2008 and 2010 is by far the largest between any two consecutive elections in the data set, and it unsurprisingly resulted in the largest seat turnover. 14 This created many Democratic incumbents facing a Republican wave in Republican-leaning districts drawn by Republican gerrymanders, and these situations tended to generate close elections. So we must be mindful that the factors measured in this paper are of course not the only factors that contribute to competitiveness. For example, 2010 shows an 22

effect of incumbency, though perhaps in the opposite direction we would typically expect: Democratic incumbents were able to keep many races close under tides conditions that would have resulted in easy Republican wins if the seat were open or held by a Republican. Despite this outlier, the asymmetric effects of tides on competition shown in Figure 2a are strongly supported through both statistical and anecdotal evidence in the case of Democratic gerrymanders, and more weakly supported through anecdotal evidence in the case of Republican gerrymanders. 15 D. Differences by Party Holding Seats Crucial to a generalized theory of dummymanders is that gerrymandering parties will see increased competitiveness under adverse tides in the seats that they already hold, as opposed to seats they have ceded to the opposing party. Testing this proposition requires testing a somewhat complicated three-way interaction involving the size of the wave, the party drawing the lines, and the party controlling the seat prior to the election. We have already shown that the number of competitive elections increases in states with partisan gerrymanders as tides adverse to gerrymandering party increase. And it should be obvious that under strong partisan tides, most of the vulnerable seats will be those held by the adverse party, regardless of gerrymandering regime, and thus we should see more competitive elections under Democratic tides in seats held by Republicans (and vice versa). What the generalization of the dummymander hypothesis predicts is that, as Democratic tides increase (e.g.), the share of competitive elections that are contested in Republican-held seats should increase more in states with Republican gerrymanders than in states with other redistricting regimes. Specifically, the interaction of (1) Republican held-seats with (2) Republican gerrymanders AND (3) Republican national vote share should be negative. And we 23

should see a similar effect of Republican waves on Democratic seats in Democratic gerrymanders. Table 6 below tests this interaction, with Close Races as the dependent variable, and Republican Seat a dummy variable taking a 1 if the seat was held be a Republican prior to the election, and a 0 otherwise. As a reminder, GOP Vote Margin is the margin by which Republicans won the national congressional popular vote (negative if Democrats won the popular vote). Unsurprisingly, from the negative coefficient of Republican Seat*GOP Vote Margin in all specifications, Republicans-held seats are more likely to be competitive during Democratic waves, and Democrat-held seats more competitive during Republican waves. However, in column 1, we see that this effect is exaggerated by a significant negative coefficient, as expected, in states gerrymandered by Republicans. We would expect a positive coefficient on the three-way interaction involving Democratic gerrymanders, but this is not significant. [Table 6 about here] However, by including all elections in the data set, we are considerably muddying our hypothesis, in that we would not expect an effect on Republican gerrymanders based on the size of Republican waves, nor an effect on Democratic gerrymandering during Democratic waves; we expect similarly low competition in seats held by the gerrymandering party in both neutral years and under friendly tides. So a better test of the hypothesis is to examine Republican gerrymanders excluding the Republican wave years (where Republicans won the popular vote by more than 5%), and Democratic gerrymanders excluding the Democratic wave years; this is done in columns 2 and 3 respectively. Note that there are far more Democratic wave years to exclude than Republican ones. In column 2, testing Republican gerrymanders, we see an even stronger negative coefficient on the three-way interaction. Under Democratic tides, competition becomes especially concentrated on Republican-held seats when Republicans drew the lines. In column 3, 24

testing Democratic gerrymanders, we observe a coefficient with the expected (positive) sign, of substantive size almost equal to the Republican test. However, because there are very few Republican waves in the data set, the variance on this coefficient is extremely large and thus it is not statistically significant. E. Competition and Partisan Turnover Many may consider electoral competitiveness a political good to be valued on it own; for example, it may enhance citizen engagement with the electoral process. But others may argue that competition is only instrumental to public responsiveness in the form of seat change. And indeed, essential to the idea of the dummymander is that misestimating future political trends causes not just increased competition, but ultimately the loss of seats for the gerrymandering party, potentially also leading to loss of partisan control of Congress. Grofman and Brunell s (2005) original example of the dummymander concerned Democratic seats losses under southern Democratic gerrymanders in the 1994 Republican wave election, while Seabrook (2010) showed how Republicans lost a disproportionate number of seats in states they gerrymandering in the consecutive Democratic waves in 2006 and 2008. But we can also observe the impact of dummymanders on partisan election outcomes in both our four-decade data set as a whole, and in the largest single wave election during that time. In the 1974 election following the Watergate scandal, Democrats disproportionately won seats in where Republicans drew the lines. Among states outside the south, Democrats netted 20 pick-ups and a 68% of the 119 seats in states gerrymandered by Republicans. This compares to netting 8 pick-ups, and 59% of the 81 seats drawn by bipartisan agreement, and 3 pick-ups and 59% of the 39 seats in states drawn by Democrats. 16 25

Additionally, we can analyze the sensitivity of partisan election outcomes to tides and gerrymander in a similar fashion to our analysis of competitiveness. In this case, our controls for national tides and statewide presidential partisanship will be directional, with positive values indicating Republican tides or disposition. The complete results are included in the supplemental tables, but the key outcome is that Republican gerrymanders are significantly more sensitive to Democratic tides than bipartisan or other gerrymanders. (We would also expect Democratic gerrymanders to be more sensitive to Republican tides, but as with much of the analysis above, the very limited number of Republican waves in our data set do not allow us to draw conclusions beyond the anecdotal sort similar to Grofman and Brunell.) Using the coefficients in Appendix Table C10 17 for a swing state (nationally average presidential partisanship) predicts that moving from neutral tides to an strong, 15-point Democratic tide (a slightly bigger wave than 2008 but smaller than 1974), would reduce Republican seats won under Republican gerrymanders by 22% (from 58% to 36%), but only reduce Republican seats under bipartisan maps by 12% (from 52% to 40%), and Democratic maps by 13% (from 38% to 25%). The difference in this slope between Republican and bipartisan maps is significant at p<.05. And note that the number of Republican seats during a 15-point Democratic wave is actually fewer in absolute terms under a Republican gerrymander than a bipartisan map. So backfiring of Republican gerrymanders under Democratic waves not only lead to more close races but ultimately more responsiveness to voters in the form of partisan seat turnover. Discussion & Conclusion The results in this paper provide evidence of both the limitations of partisan gerrymanders and the effectiveness of nonpartisan reforms, results derived even in the absence of 26

information about what lines were actually drawn and how those lines specifically affected the resulting campaigns. As hypothesized, seats gerrymandered by nonpartisan commissions appear significantly more competitive than those drawn by bipartisan legislatures, whether competiveness is measured by district partisan demographics or by vote margins in congressional elections. Additionally, while partisan maps draw many districts that appear noncompetitive at the demographic level, they generate greater competition in actual elections as national tides increase. Further, the pseudo-paradox of greater national competition inducing less local competition, specifically under partisan maps, suggests that partisan mapmakers are drawing lines to create safe seats particularly in anticipation of close national conditions, somewhat oblivious to or discounting of the possibility of uncertain future tides. It is only through accounting for this pseudo-paradox that previous conflicts in the literature can be resolved: works such as Hirsch (2003) and McDonald (2006) find that partisan gerrymandering reduces competition, but draw from elections where tides were low and the national partisan environment evenly balanced, while works drawing from cycles with larger and less stable tides (e.g. Gopoian and West 1984; Hood and McKee 2008) show the opposite. And while Carson et al (2004) and Yoshinaka and Murphy (2011) find increased competition immediately following a redistricting cycle, the effects of gerrymandering are here shown to persist through the decade. Moreover, this article clarifies how we should measure competition. One of the primary arguments that reform advocates advance for nonpartisan competition is the lack of competitive districts under legislative districting. And indeed, both partisan maps and nonpartisan maps draw districts that are less closely balanced ideologically than nonpartisan commission (see Appendix B). But closely balanced districts are not the same as competitive elections. When speaking of the impact of legislative districting on actual election results, partisan maps foster competition and responsiveness to changes in voter preferences. As Grofman and Brunell 27

implied in coining the term dummymander, much of the increased competition may be unintended by the mapmakers, but it is observable not just as anecdote, but as a statistically and substantive significant result across the nationwide election data over the past forty years. Additionally, these results emphasize that not all reform plans will have a similar impact on competitiveness. While real nonpartisan commissions that explicitly strive to create more closely-contested districts (such as exists in Arizona) may indeed enhance the probability of close elections, bipartisan commissions appointed equally of legislators from both parties (such as exists in New Jersey, and has been proposed in Tennessee) may ossify noncompetitive districts to an even greater degree than even the most egregious partisan gerrymander. The impact of national tides on how we view gerrymandering and electoral competition further implicates politics at the state level in two crucial ways. The choice of districting institution is of course the province of the states, and a very live political issue within many state legislatures. While a lack of competitiveness and/or responsiveness is often sited as a drawback of continued legislative gerrymandering, the evidence here shows that this is only conditionally true. And indeed, partisan gerrymanders can foster competition when perhaps it is most important: when the public is speaking with a clear voice through a national wave. Giving advocates and states legislators a fuller view of the actual impact of potential reforms can only improve the future debate over the structure of our government. Secondly, it is also plausible, although untested in this article, that national tides could interact with districting to create competition at the level of state legislative elections. Past work has varied in its treatment of national tides as applicable to state politics (see e.g. Makse 2014 for discussion). Recent research does suggest that state legislative election results consistently follow waves in congressional elections (see e.g. Rogers 2013). But the potential for variation in tides among states leads to murky effects of gerrymandering both in past studies (with Masket et 28

al. 2012 finding no effects in their analysis of state legislative election data during a period with multiple wave elections) and anecdotally (state legislative bodies with Republican gerrymanders in Pennsylvania, Ohio, and Michigan fell to Democrats in 2006 or 2008, but an opposite trend was seen in a Democratic state legislative map in Tennessee). So while more work is needed to clarify the role of tides at various levels in state legislative elections, this evidence broaches the possibility that national tides may also interact with local political decisions and local political outcomes in unexpected ways. And these unexpected outcomes may even debunk the shared conventional wisdom of pundits on the short-term future of the U.S. Congress. While the most recent election results may depict partisan gerrymandering at its most powerful and insidious, a wider historical view presents a somewhat different picture. Thus, these results suggest why Democrats may have reason to be hopeful about their chances to retake control of the House at some point during the next decade, despite the counter-majoritarian outcome in 2012 in which Republicans lost the national popular vote but won the majority of seats. Because the national vote was extremely close, we expect very few close races in states with partisan control of redistricting (mostly Republican control in the current decade). However, these same maps could see a great deal of competition and turnover if Democrats can win a decent popular vote margin in the future, suggested by the fact that President Obama in 2008 won five Pennsylvania districts currently held by Republicans. Gerrymandering has historically worked very well for the controlling party when the national vote was evenly split, but led to greater and often unexpected competitiveness as the national margin increased. There is no reason to expect these same trends will not continue in the foreseeable future. 29

References Abramowitz, Alan I., Brad Alexander, and Matthew Gunning. 2006. "Incumbency, redistricting, and the decline of competition in US House elections." Journal of Politics 68(1): 75-88. Adams, Florence P. 2005. Minorities and Representation in the New Millennium. In Redistrciting in the New Millennium, 155-79. New York: Rowman & Littlefield. Barone, Michael & Richard E. Cohen. The Almanac of American Politics. Washington: National Journal Group, Bi-annual Publication. Brennan Center for Justice. 2015. Redistricting Laws Roundup Available at: https://www.brennancenter.org/redistricting-laws-roundup-2015. Cain, Bruce E., Karin MacDonald & Michael McDonald. 2005 From Equality to Fairness: The Path of Political Reform since Baker v. Carr. In Party Lines, 6-30, Thomas E. Mann & Bruce E. Cain, ed. Washington: Brookings Institution. Cain, Bruce E. 1985. Assessing the Partisan Effects of Redistricting. American Political Science Review 79: 320-33. Carson, James L. & Michael H. Crespin. 2004. The Effect of State Redistricting Methods on Electoral Competition in United States House Races. State Politics and Policy Quarterly 4(4): 455-69. Carson, Jamie L., Michael H. Crespin, and Ryan Dane Williamson. 2014. "Re-evaluating the Effects of Redistricting on Electoral Competition, 1972-2012." State Politics & Policy Quarterly1 14(2): 165-177. Cox, Gary W. & Jonathan Katz. 2002. Elbridge Gerry s Salamander: The Electoral Consequences of the Reapportionment Revolution. New York: Cambridge University. Ferejohn, John A. 1977. On the Decline of Competitiveness of Congressional Elections. American Political Science Review 71: 166-76. Gelman, Andrew & Gary King. 1994. Enhancing Democracy through Legislative Districting. American Political Science Review 88(3): 541-559. Glazer, Amihai, Bernard Grofman & Marc Robbins. 1987. Partisan and Incumbency Effects of 1970s Congressional Redistricting. American Journal of Political Science 31: 680-707. Grofman, Bernard & Thomas L. Brunell. 2005. The Art of the Dummymander: The Impact of Recent Redistrictings on the Partisan Makeup of Southern House Seats. In Redistrciting in the New Millennium,183-200. New York: Rowman & Littlefield. 30

Gopoian, David L. & Darrell M. West. 1984. Trading Security for Seats: Strategic Considerations in the Redistricting Process. The Journal of Politics 46(4): 1080-96. Hirsch, Sam. 2003. The United States House of Unrepresentatives: What Went Wrong in the Latest Round of Congressional Redistricting. Election Law Journal 2(2): 179-216. Hetherington, M. J., Larson, B., & Globetti, S. 2003. The redistricting cycle and strategic candidate decisions in US House races.: Journal of Politics 65(4): 1221-1234. Hood, M. V. & Seth C. McKee. 2008. "Gerrymandering on Georgia's Mind: The Effects of Redistricting on Vote Choice in the 2006 Midterm Election" Social Science Quarterly 89(1): 60-77. Kang, Michael S. 2004 The Bright Side of Partisan Gerrymandering. Cornell Journal of Law and Public Policy 14: 443. Lindgren, Eric & Priscilla Southwell. 2013. The Effect of Redistricting Commissions on Electoral Competitiveness in U.S. House Elections, 2002-2010. Journal of Politics and Law 6(2): 13. Mann, Thomas E., & Bruce E. Cain. 2005. Party Lines: Competition, Cartisanship, and Congressional Redistricting. Washington: Brookings Institution. Makse, Todd. 2014. The Redistricting Cycle, Partisan Tides, and Party Strategy in State Legislative Elections. State Politics and Policy Quarterly 14(3): 342-363. Masket, Seth E., Jonathan Winburn, & Gerald C. Wright. 2012. The gerrymanderers are coming! Legislative redistricting won't affect competition or polarization much, no matter who does it. PS Political Science and Politics 45(1): 39. Mayhew, David R. 1974. Congressional Elections: The Case of the Vanishing Marginals. Polity 6(3): 295-317. McDonald, Michael P. 2006. Drawing the line on district competition. PS: Political Science & Politics 39: 91-94. McDonald, Michael P. 2004. A Comparative Analysis of Redistricting Institutions in the United States, 2001-02. State Politics and Policy Quarterly 4(4): 371-395. McGhee, Eric. 2014. Measuring Partisan Bias in Single-Member District Electoral Systems. Legislative Studies Quarterly 39(1): 55-85. Moxley, Warden. 1973. Congressional Districts in the 1970s. Washington: Congressional Quarterly. 31

Murphy, C. & Antoine Yoshinaka. 2009. Are mapmakers able to target and protect congressional incumbents? The institutional dynamics of electoral competition. American Politics Research, 37(6), 955-982. Niemi, Richard G. & John Deegan, Jr. 1978. A Theory of Political Districting. The American Political Science Review 72(4): 1304-23. Owen, Guillermo, & Bernard Grofman. 1988. "Optimal partisan gerrymandering." Political Geography Quarterly 7(1): 5-22. Rogers, Steven. 2013. Accountability in State Legislatures: How Parties Perform in Office and State Legislative Elections. Disseration chapter; available at: http://www.stevenmrogers.com/dissertation/chapterdrafts/collectiveaccountability/ Rogers-CollectiveAccountability.pdf. Seabrook, Nicholas R. 2010. The Limits of Partisan Gerrymandering: Looking Ahead to the 2010 Congressional Redistricting Cycle. The Forum: 8(2), Article 8. Stephanopoulos, Nicholas O. and Eric M McGhee. 2015. Partisan Gerrymandering and the Efficiency Gap. Forthcoming in University of Chicago Law Review 82. Tufte, Edward R. 1973. The Relationship between Seats and Votes in Two-Party Systems. The American Political Science Review 67(2): 540-54. Yoshinaka, Antoine & Chad Murphy. 2011. The Paradox of Redistricting: How Partisan Mapmakers Foster Competition but Disrupt Representation. Political Research Quarterly 64(2): 435-47. 32

Table 1. Probability of Close Race Controlling for Statewide and National Competitiveness No South No 1974 Pr(Close Race) (1) (2) (3) (4) Statewide Competition -0.003 0.000-0.001-0.003 (0.003) (0.003) (0.003) (0.003) National Tides 0.019*** 0.017*** 0.017*** 0.013** (0.005) (0.006) (0.005) (0.006) Year - - -0.002 - (0.002) Redistricting Year - - 0.004 - (0.041) Constant -1.18-1.16 3.32-1.15 (0.048) (0.055) (4.37) (0.050) Observations 8,700 6,185 8,700 8,265 Notes: Entries are probit coefficients. Standard errors clustered by district interacted with decade. * = p<.10, ** = p<.05, *** = p<.01 33

Table 2. Probability of Close Race controlling for Redistricting Institution and State and National Electoral Trend (Congressional Races 1972-2010) w/cds w/cds No South control No South control Pr(Close Race) (1) (2) (3) (4) (5) (6) Statewide Competition -0.0030 -.00023-0.0047-0.0027 0.00017-0.0045 (0.0031) (0.0034) (0.0031) (0.0032) (0.0034) (0.0031) National Tides 0.020*** 0.019*** 0.020*** 0.015* 0.012 0.016** (0.0045) (0.0054) (0.0045) (0.0083) (0.090) (0.0080) Democratic Gerrymander -0.10** -0.085-0.14*** -0.12-0.122-0.17* (0.052) (0.077) (0.052) (0.090) (0.147) (0.088) Republican Gerrymander -0.016 0.036-0.015-0.20* -0.195-0.21** (0.059) (0.066) (0.058) (0.11) (0.118) (0.10) Court Gerrymander 0.065 0.083 0.091* 0.068 0.090 0.13 (0.050) (0.062) (0.050) (0.083) (0.101) (0.081) Nonpartisan Gerrymander 0.33*** 0.32*** 0.21** 0.41** 0.382** 0.29* (0.098) (0.099) (0.10) (0.17) (0.172) (0.17) Democratic Gerry*Tides - - - 0.0035 0.0065 0.0045 (0.011) (0.017) (0.011) Republican Gerry*Tides - - - 0.025** 0.032** 0.028** (0.012) (0.013) (0.012) Court Gerry*Tides - - - -0.00040-0.0008-0.0057 (0.0099) (0.0118) (0.0098) Nonpartisan Gerry*Tides - - - -0.013-0.011-0.014 (0.021) (0.021) (0.021) CDs - - -.0094*** - - -.0096*** (0.0018) (0.0018) Open - - - - - Constant -1.20*** -1.21*** -1.00*** -1.17*** -1.17*** -0.98*** (0.058) (0.065) (0.067) (0.074) (0.081) (0.079) Observations 8,700 6,300 8,700 8,700 6,300 8,700 Notes: Entries are probit coefficients. Standard errors clustered by district interacted with decade. * = p<.10, ** = p<.05, *** = p<.01 34

Table 3. Coefficient for effect of National Tides on Close Races by redistricting institution (Congressional Races 1972-2010) Institution Coeff. SE n Democratic.017**.008 2663 Republican.041***.010 1627 All Partisan.028***.006 4290 Bipartisan.004.009 2536 Nonpartisan.003.020 385 Overall.019***.005 8700 Table 4. Coefficient for effect of National Tides on Close Races under Democratic Gerrymanders (Congressional Races 1972-2010) Institution Coeff. SE n Democrats win Popular vote.012.009 2041 Republicans win popular vote.089***.022 622 All elections.017***.008 2663 35

Table 5. Percent Close Races by Year and Gerrymandering Party Close Races in Seats Drawn by Democrats National GOP Year Vote Margin Seats % Close 1992-5.3 122 16% 1994 6.9 122 20% 1996-0.3 122 14% 1998 0.9 122 8% 2000 0.3 122 7% 2002 4.6 111 11% 2004 2.6 79 8% 2006-7.9 66 5% 2008-10.5 66 6% 2010 6.6 66 15% Close Races in Seats Drawn by Republicans Nat'l GOP Year Vote Margin Seats % Close 1972-5.3 119 18% 1974-16.6 119 28% 1976-10.8 119 17% 1978-8.7 119 17% 1980-2.7 119 12% 2002 4.6 98 7% 2004 2.6 130 5% 2006-7.9 143 13% 2008-10.5 143 13% 2010 6.6 143 15% 36

Table 6. Close Races by Wave, Gerrymandering, and Party Holding Seat Pr(Close Race) (1) (2) (3) Exclude GOP Exclude Dem All Elections Waves Waves Statewide Competition -0.0040-0.0024-0.011** (0.0033) (0.0033) (0.0046) National GOP Vote Margin 0.025*** -0.0020 0.064*** (0.0041) (0.0049) (0.010) Republican Gerrymander 0.17** 0.17** 0.12 (0.080) (0.087) (0.11) Democratic Gerrymander -0.034-0.015-0.040 (0.064) (0.068) (0.089) Republican Seat -0.0036 0.25*** 0.20** (0.059) (0.071) (0.079) Republican Gerry*Republican Seat -0.65*** -0.70*** -0.63*** (0.14) (0.16) (0.17) Democratic Gerry*Republican Seat -0.20* -0.21* -0.19 (0.11) (0.12) (0.13) Republican Seat*GOP Vote Margin -0.058*** -0.025*** -0.13*** (0.0069) (0.0084) (0.019) Republican Gerry*Republican Seat*GOP Vote Margin -0.038*** -0.044*** -0.0021 (0.012) (0.013) (0.036) Democratic Gerry*Republican Seat*GOP Vote Margin -0.0057-0.0057 0.028 (0.011) (0.013) (0.030) Constant -1.03*** -1.26*** -1.09*** (0.052) (0.058) (0.070) Observations 8,700 7,830 3,915 Notes: Entries are probit coefficients. Standard errors clustered by district interacted with decade. * = p<.10, ** = p<.05, *** = p<.01 37

Figure 1. Partisan Balance of Districts in Four States, 2002 (Density Plot) 38

Figure 2a. Hypothesized Effect of Tides on Close Elections Figure 2b. Hypothesized Effect of Tides without Regard to Direction on Close Elections 39

Figure 3. Close Races by Year and Gerrymandering Institution 40