Presentation is loading. Please wait.

Presentation is loading. Please wait.


Similar presentations

Presentation on theme: "MULTI-CANDIDATE PRESIDENTIAL ELECTIONS And the Contingent Procedure."— Presentation transcript:


2 Duverger’s Law and the Contingent Procedure Recall that Duverger’s Law says that single- winner elections typically will have just two “serious candidates.” –Most configurations of thee or more “serious” candidates are “out of equilibrium.” –Let C 1,C 2, and C 3 be the three candidates, labelled in order of their (expected, probably based on polls) first preference support). Some voters (probably supporters of C 3 ) have an incentive to vote strategically (to avoid “wasting” their vote). At least one candidate (probably C3 but perhaps C2 as well) has an incentive to make a “strategic exit.” –If both C 2 and C 3, a “candidate’s dilemma” results, as each is a (potential) spoiler to the other.

3 Duverger’s Law and the Contingent Procedure (cont.) However, there are special “non-Duvergerian equilibria” involving three or more candidates with significant support in the final vote. –C1 is an (apparent) majority winner. –C1, C2, and C3 appear to be in essentially a three- way tie (an unstable equilibrium). –C2 and C3 appear to be essentially tied. –C3’s supporters are essentially indifferent between C1 and C2, perhaps because of electoral convergence, or because C3 is a centrist relative to C1 and C2 –Anderson (1980), Perot (1992 and 1996) –C1 is a centrist relative to C2 and C3 –Congress Party in India

4 Multi-Candidate Presidential Elections In any event, there have been about a dozen multi- candidate Presidential elections in the U.S. history. Many (but not all) produced “minority Presidents.” –Other “minority Presidents” have resulted from close elections in conjunction with very minor candidates. Other than 1824, none produced an Electoral College deadlock but several certainly threatened to do so. –Apart from an EV tie, an EC deadlock requires a third candidate to actually win electoral votes i.e., to win a plurality of popular votes in one or more states, which in turn requires the minor candidate to win at least 1/3 of the three-candidate popular vote in that state. –This in turn is much more likely if the third candidate is sectional candidate.

5 1860 Lincoln [Rep.] 39.8%180 Douglas [N. Dem.] 29.5%12 Breckinridge [S. Dem.] 18.1%72 Bell [Const. Union] 12.6%39

6 1892 Cleveland [Dem.] 46%227 Harrison [Rep.] 43%145 Weaver [Populist] 8.5%22

7 1912 Wilson [Dem.]42%435 Taft [Rep.] 23%8 T. Roosevelt [Prog.] 27%88 Debs [Socialist] 6%0

8 1924 Coolidge [Rep.] 54%382 Davis [Dem.] 29%136 LaFollette [Prog.] 17%13

9 1948 Truman [Dem.] 49.5%303 Dewey [Rep.] 45%189 Thurmond [States’ Rights] 2.4%39 H. Wallace [Prog.] 2.4%0

10 1960 Kennedy [Dem.] 49.7%303 Nixon [Rep.]49.6%219 Unpledged Electors 0.2% [MS] [+ AL] 15

11 1968 Nixon [Rep.] 43.4%301 Humphrey [Dem.] 42.7%191 G. Wallace [AIP] 13.5%46

12 1980 Reagan [Rep.] 51%489 Carter [Dem.] 41%49 Anderson [Ind.] 7%0

13 1992 Clinton [Dem.] 43%370 G.H.W. Bush [Rep.] 37%168 Perot [Ind.] 19%0

14 1996 Clinton [Dem.] 49%379 Dole [Rep.] 41%159 Perot [Reform] 8%0

15 Uniform National Swing and Electoral College Deadlock While none of these elections produced an EV deadlock triggering the House contingent procedure, several of them would have if the vote for the two major-party candidates had been (even) closer. We can get some sense of the likelihood of such a deadlock by using (a variant of) the uniform national swing analysis previously used to measure the width of the “wrong winner” interval. This entails holding the popular vote of the third candidate constant, while examining the effect of a uniform national swing between the Democratic and Republican candidates.

16 Uniform National Swing and Electoral College Deadlock (cont.) In this analysis, there are two critical threshold for the third candidate –If the third candidate receives more than 50% of the popular vote in a state, he wins that state’s electoral votes regardless of the vote division between the major-party candidates in that state. –If the third candidate, receives less than 1/3 of the popular vote in a state, he loses that state’s electoral votes regardless of the vote division between the major-party candidates in that state. –If the third candidate receives between 1/3 and 1/2 of the popular vote in states, he win the state’s electoral votes or not depending on how evenly the major party candidates split the remainder of the vote.

17 1972:Possible Electoral College Deadlock Without a Third Candidate





22 Electoral College Deadlock In every Presidential election since 1824, once it was known who had won the electoral votes of each state, one candi- date has always had the required majority of electoral votes. If and when it happens that no candidate wins the required majority [at present 270] of electoral votes, there will be a (pros- pective) Electoral College deadlock that must be solved by the contingent procedure

23 EC Deadlock: Candidate Scenarios Apart from the possibility of a 269-269 electoral vote tie, a necessary condition for an Electoral College deadlock is the presence of a third candidate who wins some electoral votes. –This in turn requires that the third candidate wins at least 1/3 of the (three-candidate) popular vote in at least one state. The more lopsided the contest between the two leading candidates, the better the third candidate must do to produce an Electoral College deadlock. –Over the course of the 1968 campaign, polls showed both that Wallace’s support was diminishing; but also the Nixon-Humphrey race was tightening up, so the likelihood of an Electoral College deadlock actually increased.

24 Candidate Scenarios (cont.) There are two distinctive types of third candidates who might deadlock the Electoral College. A factional candidate (probably to the right or left of both major candidates and probably with a sectional base) –who appeals to a section of the electorate with some special grievance, and –whose goal is not to be elected President but to attract attention and try to deadlock the Electoral College and thereby become a “kingmaker.” Such a candidate would run far behind both major-party candidates in popular votes, but with geographically concentrated support would win some electoral votes. –Examples: maybe Weaver (Populist, 1892) maybe LaFollette (Progressive, 1924) certainly Thurmond (States’ Right, 1948) certainly unpledged elector slates (1960) certainly George Wallace (American Independent, 1968)

25 Candidate Scenarios (cont.) A national candidate (probably a centrist running as an independent or on some kind of “national unity” ticket), who appeals to a broad segment of the electorate unhappy with both major parties, and whose goal is to be elected President. Such a candidate might run more or less even with both major-party candidates in popular votes, and the distribution of electoral votes among the three candidates would be highly unpredictable. –Examples: certainly T. Roosevelt (Progressive, 1912) possibly LaFollette (Progressive, 1924) Anderson (Independent, 1980), if he had run much stronger Perot (Independent, 1992), if he had run somewhat stronger Perot (Reform, 1996, if he had run much stronger

26 Electoral College Deadlock (cont.) Even as the identity of the new President remains unknown, the partisan makeup of the new House will be known. –If one party controls 26 or more state delega- tions, the outcome of a House election may be fairly predictable. However, there may be party members in the House who might be willing (or even anxious) to vote against their party’s President candidate. –If neither party controls 26 state delegations, there is the prospect of House deadlock on early ballots. [See below on the 1825 House Rules]

27 Electoral College Deadlock (cont.) Two distinct periods of uncertainty, conflict, and bargaining will occur: –between the selection of electors [on Presidential election day] and the casting of electoral votes [in mid-December]; and –immediately following the counting of electoral votes before the joint session of Congress [on January 6]. Each period corresponds to two distinct modes of breaking the deadlock: –Bargaining over the casting of electoral votes. –Bargaining in the House of Represenatives Plus possible House-Senate interactions

28 Electoral College Deadlock (cont.) However, the outcome of the first stage will depend on expectation concerning the prospective outcome of the second stage. –On one hand, the outcome of the second stage may be quite predictable. –On the other hand, the outcome of the second stage may be highly uncertain. Presumably neither type of third candidate (in contrast to both major-party candidates) would have any automatic base of support in the House. –However, a national candidate might ultimately win in the House (as a compromise candidate), while –a factional candidate presumably could not win in the House.

29 Before the Casting of Electoral Votes A factional candidate may try to strike a bargain with one of the two major candidates, promising that candidate electoral votes in return for concessions in policy and/or personnel. –A factional candidate has such bargaining leverage before electoral votes are cast but loses it entirely afterwards. Other things equal, the factional candidate would expect to make a deal with the more ideologically proximate major-party candidate. But other things probably would not be equal, since partisan makeup of the new House would be known. –The outcome of a House election might be predictable, and –the major-party candidate expected to lose a House election would have more of an incentive to deal with the factional candidate. –It would be generally recognized that the factional candidate could not win in the House.

30 George Wallace in 1968 George Wallace is 1968 is a clear example of a factional candidate. –Wallace extracted from each of his elector candidates a pledge to vote for Wallace or whomever he designated, so as to enhance Wallace’s bargaining power prior to the casting of electoral votes (in the event Wallace was able to deadlock the Electoral College). –Nixon was the ideologically more proximate major party candidate, presumably more willing to make concessions on civil rights. –In the House, Democrats controlled 26 state delegations, Republicans 19, and 5 were equally divided. However, the 26 Democratic delegations included all Southern and Border state delegation except VA [divided]. Some Southern and Border State Democrats were sympathetic to Wallace, more had constituents sympathetic to Wallace, and many probably preferred Nixon to Humphrey for President. So the outcome of a House election would have been quite uncertain (though it was clear Wallace could not have won in the House) The “unpledged elector” gambit in 1960 was similarly motivated.

31 Before the Casting of Electoral Votes (cont.) A national and centrist third candidate might seek to bargain with the major-party candidates, in the same manner as a factional candidate. –Such a candidate could more readily play the two other candidates off against each other. But, especially if he ran ahead of one or both major-party candidates, the national third candidate might try to bargain the other way, i.e., –offer them promises in return for their support. However, the regular party candidates could not offer electoral votes the way Wallace and other factional candidates might. –Democratic and Republican electors are unconditionally pledged to support their party nominees and are bound to do so by party rules (and, in some states, by law as well). So an Electoral College deadlock involving a national third candidate probably would not be resolved prior to the casting of electoral votes. An Electoral College deadlock resulting from a 269-269 tie almost certainly would not be broken prior to the casting of electoral votes.

32 After the Casting of Electoral Votes If no deal is made prior to the casting of electoral votes, it will be evident that the House contingent procedure [unused since 1824] will come into play. –However, nothing official can happen until electoral votes are counted on January 6. –However, there surely would be much preliminary wheeling and dealing, focused primarily on the [newly elected] House or Representatives that will select the President from the top three candidates [12 th Amend- ment]. –An electoral college deadlock for President presumably implies also a deadlock for Vice President, and the [newly elected] Senate will select the Vice President from among the top two candidates [12 th Amendment].

33 The House Contingent Procedure If the counting of electoral votes confirms an electoral vote deadlock, the 12 th Amendment requires the House “immediately” to begin balloting for President (from among the top three candidates [or between the two tied candidates]). –Note that the House cannot elect an appealing compromise candidate (e.g.,a “Colin Powell” type) other than the top three. Unless it choose to change them, the House will follow the rules it drew up in 1825 to resolve the electoral college deadlock of 1824.

34 The 1825 House Rules A quorum consisting of at least one member from 2/3 of the state delegations is required. Election requires support by a majority of all state delegations, i.e., 26 votes. –Though it has electoral votes, DC has no representation in the contingent procedure. Prior to each House ballot, formal balloting takes place within each state delegation. –A Presidential candidate receives the vote of a state if and only if the candidate is supported by a majority of the state delegation. –If no candidate receives such support within the delegation, the state casts a “divided vote” and effectively abstains. Balloting continues until a President is elected. A motion for to adjourn temporarily must be supported by a majority of state delegations.

35 House Contingent Procedure (cont.) Since the House contingent procedure has never come into play since the country has had an established party system, no one has much idea as to how members of the House would decide how to vote. House members might use any the following considerations to decide how to vote. –Vote for the Presidential candidate of my party [probably most likely]. If all followed this norm, the outcome would be predictable on the basis of the previous Congressional election. Except that some delegations might be evenly split between the parties. –Vote for the Presidential candidate who carried my state [and won its electoral votes]. –Vote for the Presidential candidate who carried my district. –Vote consistent with a plurality of first preferences in my delegation, i.e., don’t let my state be “divided.” –Vote on the basis of my personal preferences, ideology, conscience, etc. –Use my vote as a bargaining chip. –Work with others to try work out a reasonable compromise.

36 House Contingent Procedure (cont.) It is also unclear how the three Presidential candidates would behave at this stage. –Would they try appeal to House members on the basis of party, ideology, promises of policy or personnel, etc.? –Might a candidate withdraw in favor of another? –Would they try to strike a deal among themselves? At this stage, a national (vs. factional) might play a pivotal role. –If this third candidate were him/herself an appealing compromise (a “Colin Powell” type), his/her election might be the way to break a House deadlock. How would the House election of a President and the Senate election of a Vice President interact? –At the outset, the Senate would probably wait to see what happens in the House.

37 House Deadlock The House might easily deadlock for multiple ballots (like the House that decided the 1800 election). –If some House members vote for the third candidate, their delegation may remain “divided.” –Even if only two candidates are supported within a delegation, even-number sized delegations may remain “divided.” –Even if no state delegations are internally divided, the number of states is even, so the House as a whole may be divided 25-25. Moreover, strategic absence by a minority of members appropriately distributed over state delegations could block a quorum. In sum, the House contingent procedure is not a strong simple game, i.e., there are blocking coalitions.

38 The Senate Contingent Procedure The Senate’s role in electing the Vice President may seem to be of distinctly subsidiary importance. But the Senate has two advantages over the House, which may give it a critical role in the election of a President. –If the House remains deadlocked as of January 20, “the Vice President elect shall act as President until a President shall have qualified” [20 th Amendment]. –The Senate is unlikely to deadlock, since: the 12 th amendment allows the Senate to chose only between the top two candidates for Vice President; voting is by individuals, not delegations; support of a simple majority is sufficient for election; and while the number of Senators is always even, the [outgoing] Vice President can break a tie (presumably). However, a 2/3 quorum is required. –In sum, the Senate contingent procedure is a strong simple game, i.e., there are no blocking coalitions (apart from the quorum requirement) and is unlikely to require more than one ballot.

39 House and Senate Interaction The Senate would probably hold off electing a Vice President, on the grounds the grounds that they should [if possible] elect the running mate of the Presidential candidate elected by the House. –But note that it may not be able to do this. But if the House deadlocks indefinitely, the Senate [which can’t deadlock] can name the new Vice President who would act as President as of January 20. Since the Senate can’t deadlock, it may be clear who the Senate would elect in the event the House fails to elect. And this very prospect might break the House deadlock.

Download ppt "MULTI-CANDIDATE PRESIDENTIAL ELECTIONS And the Contingent Procedure."

Similar presentations

Ads by Google