Monthly Archives: April 2016

campaign finance

Is Our Campaign Finance System Broken?

A Quick Glance at Campaign Finance

Bernie Sanders has made the campaign finance system a central point of his presidential candidacy.  Anybody who’s listened to the Senator speak has heard him rail against the “corrupt campaign finance system” that allows billionaires to “buy” or otherwise “rig” elections.

But are his accusations true?

A thorough study is very difficult due to few presidential elections in the post Citizens United era — this is just the second.  With little data available, a rigorous quantitative study cannot yet be done.  As such, I leave you with this chart that pulls super PAC data from OpenSecrets.  Flush with cash, these super PACs have often backed losing candidates (which isn’t surprising — all but one candidate ultimately loses).  I’ll let you draw your own conclusions.

CandidateSuper PAC Dollars RaisedIs He/She President?
Mitt Romney$153,741,731No
Scott Walker$24,127,173No
Bobby Jindal$4,472,589No
Lindsey Graham$4,182,034No
Rand Paul$9,909,755No
Carly Fiorina$14,131,949No
Chris Christie$20,276,835No
George Pataki$1,547,674No
Jeb Bush$118,864,488No
Marco Rubio$57,933,414No

bernie sanders wisconsin

Bernie’s Big Wisconsin Night

Our model gave Bernie Sanders a 58 percent change of winning the Wisconsin primary and estimated he would earn 53 percent of the vote, thus carrying 46 delegates.  The actual results? 57-43 in favor of Sanders, leading to his taking 48 delegates (to Hillary Clinton’s 38).  He clearly beat our expectations, though at the same time, a large victory in an 82 percent white state (with a small African American population) is not necessarily shocking.  Regardless, the victory is sure to lead to another influx of hard money into the Sanders campaign coffers and will serve as a springboard into New York and other Northeastern primaries.

That said, Bernie Sander’s performance in Wisconsin still leaves much to be desired: He’s still losing African-Americans by a lot (69-31).  The upcoming states have decent African American populations that can make or break Sanders’ attempts to close the delegate gap.  This problem isn’t new – a failure to connect with African Americans has plagued the Sanders campaign since its inception and is the prime cause for Clinton’s large delegate lead.  Her sweep of Southern states, many by shockingly large numbers, rendered Sanders’ victories in small caucus states and his close upset in Michigan meaningless.  And though Clinton could not best Sanders in Wisconsin, the demographics of the upcoming states and her continued strength with African Americans (though she’s not as strong with minorities as a whole, winning them in Wisconsin 57-43) should portend well for her.

Sanders supporters will naturally counter with two arguments.  The first would be that Sanders is narrowing his gap among African Americans.  Compared to the opening days of the campaign, that is indubitably true.  Sanders lost African Americans in South Carolina by a whopping 72 points, 86-14.  Compare that to Wisconsin and it’s easy to claim that significant inroads have been made.  However, we’re not seeing any current movement.  In Michigan, his most significant upset, Sanders lost African Americans 28-68; similarly, in Illinois a week later, he lost the minority 30-70.  Numbers have hardly changed since the middle of March despite Sanders’ win streak and supposed momentum.  As of yet, he’s simply not changing numbers.

The other argument Sanders supporters would advance revolves around momentum.  Sanders is on a large winning streak and that’s giving his campaign renewed hope and his supporters renewed faith in ultimate victory.  With this momentum has come millions of dollars – $44 million in March alone.  Can Sanders translate momentum into votes?  There lies the make-or-break question.  If Sanders can translate momentum – a seemingly overrated and overplayed buzzword – and his cash windfall into votes and support among minority communities, the upcoming states might become competitive and maybe, just maybe, he could close the delegate gap.

Is Wisconsin a turning point in the race?  Not in itself.  It’s a state Sanders needed to win and did; he achieved the necessary.  Only if Sanders can compete in New York will race be altered.

[Be sure to follow us on Twitter and like us on Facebook for election news and analysis!]

us senate apportionment

Why U.S. Senate Apportionment Matters

This question has its foundation in legal history. During the constitutional convention, James Madison proposed Senate apportionment based on population rather than equality among the several states.  Alexander Hamilton joined Madison in calling for proportional representation, claiming that equal representation despite population inequality “shocks too much the ideas of justice and every human feeling” (The Avalon Project 2008).  This issue incited the most debate in the convention and threatened the creation of the Constitution.  Delegates were deeply divided on the issue and many small states let it be known that they would leave the convention should Madison’s plan be adopted.  That said, Madison and Hamilton had early success – state delegations originally voted 6-5 to implement proportional representation for the Senate. However, small states later reopened the issue and, with the votes of a few medium and large states, successfully adopted equal representation of states in the Senate.  In return, the House would be apportioned based on population.  This, along with other concessions between the two sides, resulted in the Great Compromise that held the convention together.

It is unlikely that the Founding Fathers anticipated the population disparity we see today.  After the convention, the smallest state – Delaware – had a little more than 8 percent of the largest state’s population.  Today Wyoming, the nation’s smallest state, has around 1.5 percent of California’s population.  Figure 1 plots the smallest state’s percentage of the largest state’s population as well as those states’ populations over time (note that the population of the smallest state barely registers on the graph).  The left-hand y-axis denotes state population while the right-hand y-axis shows the total population percentage of the largest and smallest states; the gray line represents the smallest-largest state population percentage, attained simply by dividing the smallest state’s population by that of the largest.

senate apportionment

Figure 1: The blue line denote the largest state’s population (left-hand axis) and the orange lines – barely noticeable – show the population of the smallest state.  The grey line depicts the smallest-largest state ratio.  Numbers derived from the U.S. Census and

Relative Senate representation is inversely related to state population.  As the smallest-largest population percentage decreases, representational inequality increases.  Over time, small states gained power over their larger counterparts as they maintained an equal voice in the Senate as their share of total population and largest-state population percentage shrank.  This can also be seen by examining whether state population percentage equates to its share of Senate seats.  Based on a one person, one vote principle, a state’s share of legislative seats will equal its share of the population.  Clearly, based on Figure 2, the Senate fails to meet such standards.

us senate apportionment one person one vote

Figure 2: Blue lines show the nation’s population share of the largest state and the orange lines, again hardly noticeable, show the population share the nation’s smallest state.  The grey lines picture each state’s Senate seat share.  Numbers again derived from Census data

The blue lines denote the population percentage of the largest state, the orange line the population percentage of the smallest state, and the grey line shows each state’s share of Senate seats.  Under perfect one person, one vote representation, the three lines should be equal.  Obviously, that is not the case.

Population inequality isn’t limited to the two extreme states – the unequal distribution of people across the country results in numerous low population states capable of controlling the Senate.  The smallest possible coalition that could sit 51[1] Senators consists of 26 states[2] comprising 55 million people, or about 17 percent of the total population.  By comparison, California and Texas – the two most populous states – have 64 million people but only four Senators.  A 60 seat supermajority can be attained through the votes of 72 million people, or 22 percent of the population.  While such coalitions are currently unlikely to happen – the likelihood that liberal Vermont elects senators from the same party (and with the same ideology) as deep-red Wyoming in the current political environment approaches zero – this examples demonstrates the extent to which the American senate can subjected to minority rule.  Considering the collapse of the two major parties onto the left-right ideological spectrum (McCarty, Poole, and Rosenthal 2006; Noel 2013), a small percentage of the population can impose their ideological tendencies onto the majority of Americans, either by electing a majority of Senators or enough to sustain procedural roadblocks (i.e., a filibuster).  Though political scientists have studied the impact of institutional design on the distribution of federal funds and representation of parties, they have not fully delved into how the Senate’s apportionment scheme impacts absolute party numbers and ideological composition of the chamber, and ultimately, how this affects policymaking.


In Reynolds v. Sims, Chief Justice Earl Warren found that the principle of one person, one vote serves as a basic democratic tenet.  It implies the inherent equality of all voices in a polity – regardless of class, gender, race, creed, religion, or geographic location, everyone has one equally weighted vote used as the ultimate show of support for a candidate or idea.  Chief Justice Earl Warren described the Court’s foray into the “political thicket” from which it decided that it had jurisdiction over redistricting cases despite their political nature (Baker v. Carr 1964), paving the way for the “one person, one vote” doctrine (Reynolds v. Sims 1964), as his tenure’s most influential decision (Ansolabehere and Snyder 2008).[3]  Yet despite the theoretic and legal justifications for one person, one vote, the American Senate does not – and cannot, per Article V of the Constitution – subscribe to the standard.  Given the vast state population disparities, there is sizable representational (and thus electoral) inequality in the Senate (Lee and Oppenheimer 1999; Griffin 2006).  On a one person, one vote basis, the Senate is the most malapportioned legislative chamber in the world (Lijphart 1984).

This representational inequality has the potential to generate undemocratic policy outcomes.  Overrepresentation of the least-populous states skews distributive outcomes by ensuring that small states receive larger amounts of federal dollars per capita than do large states (Lee 1999, 2000).  Senate apportionment also effects coalition building.  Those seeking to form a minimal winning coalition will seek to “buy” small state support rather than large state support because it is cheaper to earn small state backing.  For instance, a Senator seeking to pass a transportation appropriation will likely turn to small state Senators for support, wooing them with promises of money that, given the state’s small size, amount to little of the bill’s proposed budget.  Large states, having more roads than small states, would require a larger appropriation share, lowering the amount the sponsor could bring to his or her home state, thus making large state Senators unattractive coalition partners (Lee 1999, 2000).  Coalition strategies owing to Senate apportionment benefit small states and hurt large states (Lee 2000).

Following the Baker v. Carr and Reynolds v. Sims Supreme Court decisions ( 369 U.S. 186 (1962) and  377 U.S. 533 (1964), respectively), state legislatures adopted apportionment based on population, generating a natural case study to assess whether equal geographic representation noticeably and significantly impacts public policy.  Prior to the Supreme Court’s decisions, most state legislatures found themselves malapportioned – through deliberate efforts to thwart redistricting, rural counties often held disproportionate seat shares and thus had undue power and influence on state fund transfers.  Many rural counties thus received state funds per person than did urban counties.  When state legislatures moved to representation by population after Baker and Reynolds, state fund transfers largely matched population percentages (Ansolabehere and Snyder 2008).  Urban counties did not dominate the state purse, a fear many state legislators expressed during a House committee hearing regarding a constitutional amendment to allow states to choose how to apportion legislative chambers (Apportionment of State Legislatures: Hearing before the Committee on the Judiciary 1964).  While this does not speak directly to the Senate or Senate vote outcomes, it demonstrates that changing the apportionment scheme has substantial effects on public policy that would likely be borne out in the Senate.

Existing literature pens a persuasive analysis of the normative importance of one person, one vote and the distributional results of the Senate’s failure to adhere to the principle.  However, there has yet to be an extensive study into how Senate apportionment impacts Senate vote outcome and legislation ideology (i.e., content).  Existing research suggests that Senate apportionment does tend to overrepresent minority parties (Lee and Oppenheimer 1999, Griffin 2006), but may not have an effect on ideological representation (Griffin 2006).  The latter point likely needs to be revisited: since Griffin’s study the South and the Northeast have increasingly moved to single-party dominance.[4]  Furthermore, the continued collapse of party lines onto left-right ideology[5] (McCarty, Poole, and Rosenthal 2006; Noel 2013) furthers the likelihood that the over or underrepresentation of a party corresponds to the over/underrepresentation of an ideology.  The degree to which such representational disparities will hopefully be exposed in my research, which seeks to expand on existing studies by delving into Senate design and determining how it impacts vote and policy outcomes.

Analyzing senatorial outcomes and comparing them to the hypothetical Senate with proportional representation allows us to see the impact of rural overrepresentation in votes and legislation ideology. Recognizing the impact of equal representation in the upper chamber of Congress is crucial in assessing whether the Founding Fathers, in their strides to bolster equality in preference, perversely created a system in which the residents of the least-populous states can exert undue impact on national affairs. After all, can America truly be considered democratic if the will of the many is subverted by that of the few?

[1] Well, 52 given that each state elects two Senators

[2] Obviously

[3] To put this in perspective: Chief Justice Warren wrote the majority opinion in Brown v. Board of Education, which found school segregation to be unconstitutional

[4] In the 107th Senate, the last included in Griffin’s analysis, there were nine southern Democrats and six New England Republicans.  Today, those numbers are one and two, respectively.

[5] Across many political areas, including economics, race, and culture.