Are the GOP Delegate Allocation Rules Actually Proportional?

That word does not mean what you think it means (at least if you work in the mass media).

Well, the answer to that question is:  no (despite the repeated usage of the term in the press).

For the allocations to be truly proportional, each candidate would have to win roughly the same percentage of delegates that they won in votes in a given state.  This is not the case:  even in the so-called “proportional” states what we are really see is one big plurality contest with a number of nested plurality contests within the big one (i.e., the candidate with the most votes in a given contest, whether the big one or the nested smaller ones, wins the delegates at stake).

To wit:  in Michigan (which was constantly called “proportional”) the vote totals were as follows (according to the AP via Google):

Romney  41.1%

Santorum 37.9%

Paul 11.6%

Gingrich 6.5%

Other 2.9%

However, of the 30 delegates, Romney won 16 (53.3%) and Santorum won 14 (46.7%)–with others, even double-digit finisher Paul getting nothing.  Now, this is more proportional than had the system been winner take all, as in that case 41.1% would have won 100% of the delegates.  However, had the contest been truly proportional, the outcome would have looked like this:

Romney  12.3 (40.0%)

Santorum 11.4 (36.7%)

Paul 4 (13.3%)

Gingrich 2 (6.7%)

Other 1 (3.3%)

This is what a truly proportional outcome would have looked like.  Keep all of this is mind as you listen to the news folks carry on about the “proportional” nature of the system.

In short:  a voting system is proportional insofar as the percentage of votes won roughly translates in the same percentage of seats (or, in this case, delegates) won.  As those percentages deviate, the more disproportionality there is in the given electoral system.

If the GOP really wants a long, drawn-out contest (and a greater chance of a brokered convention) then truly proportional outcomes would help foster that.

(For the sake of the example, I counted “other” as one candidate.  The candidates with smaller vote totals gain due to rounding—since delegates have to be allocated in wholes.)

More on this from Matthew Shugart.

FILED UNDER: 2012 Election, Political Theory, US Politics, , , ,
Steven L. Taylor
About Steven L. Taylor
Steven L. Taylor is a Professor of Political Science and a College of Arts and Sciences Dean. His main areas of expertise include parties, elections, and the institutional design of democracies. His most recent book is the co-authored A Different Democracy: American Government in a 31-Country Perspective. He earned his Ph.D. from the University of Texas and his BA from the University of California, Irvine. He has been blogging since 2003 (originally at the now defunct Poliblog). Follow Steven on Twitter


  1. I always hated word math problems I`l;l take your word for it.

  2. James Joyner says:

    I gather that they allocated some proportionally, some to the winner of each congressional district, and some bonus to the overall winner. Very odd.

    I don’t understand why they thought extending the process and being more like the Democrats was a good idea. I do get that they were trying to force states to go later, punishing them by diluting their vote if they went early. But that was just doomed to fail.

  3. @James Joyner: There are some variation state-to-state (I think that in GA, for example, the district level delegate go to a candidate who wins 50%+ of the vote, but is split between first and second otherwise). I don’t think any state truly allocates proportionally, but I could be mistaken,

    The whole process is a mess, to be honest (but that problem goes deeper than allocation rules).

    Exactly what the RNC thought they were created and why is baffling to be sure.

  4. In general, the press simply uses the word “proportional” to mean anything other than a winner-take-all contest.

  5. rodney dill says:

    Michigan had the potential to be even farther out of whack than that. It’s winner take all by congressional district. (2 per district) with 2 at large which went to the winner over all. You could potentially win the popular vote, but lose by delegates.

  6. PD Shaw says:

    Michigan’s distribution is:

    28 congressional district delegates (2 per each of the 14 districts): allocated winner-take-all based on the vote in the congressional district.

    2 at-large delegates: allocated winner-take-all.

    0 automatic delegates: Penalized states lose their automatic delegates.


    The odd thing here is that each district is treated equally, regardless of whether the district leans +30 R or +30 D. I think if each district were weighed in proportion to its historic [Republican] vote totals, you would probably proximate a proportionate voting system, with some advancement of the state policy that the candidate have broad support across the state.

  7. @PD Shaw:

    The odd thing here is that each district is treated equally, regardless of whether the district leans +30 R or +30 D. I think if each district were weighed in proportion to its historic [Republican] vote totals, you would probably proximate a proportionate voting system, with some advancement of the state policy that the candidate have broad support across the state.

    Perhaps. But that seems a long way to go when you could just allocate proportionally at the state level.

  8. PD Shaw says:

    @Steven L. Taylor: The state party has an interest in selecting a candidate that does well in the state and in the districts, so I don’t believe some weighing by district is bad politics.

  9. Dave Anderson says:

    @James Joyner: I can see the idea of stretching out the process having some appeal in 2010 and early 2011 when the schedule was being made. It would give the country exposure to several high profile and competent Republican candidates and lead most news reports with Republicans arguing on Republican terms in a favorable light. That would lead to increased interest in the race and increased interest in the eventual GOP nominee. At the same time, the multiple well financed, well disciplined, well run campaigns would lay out a massive campaign structure in numerous swing states that could be readily converted into both general election campaign structure and down ballot campaign support. That is basically what happened in 2008 for the Democrats as there were two candidates that were led by knowledged candidates who were well financed, well run, well disciplined, and acceptable to 90% of the party.

    And then most of the plausible candidates who could match these desired qualities decided to sit this round out. And of the two plausible heavyweights, one is a Massachusetts technocrat who no one really likes all that much, and the other was Perry who failed the competent or knowledgeable part. So we now have one plausiblely conventional nominee (Romney) dealing poorly with a narrow cast ideological point making candidate (Santorum) and someone on an unconventional book/gig recruiting tour (Gingrich).

  10. Peter Frey says:

    It would seem that the new delegate allocation rules are likely to produce a major battle at convention time. The new rules clearly state that delegate allocation in the early primaries has to be proportional to the primary or caucus vote. Individual states are not complying with this requirement. Will the RNC enforce the new rules?

    If delegate allocation were proportional, the current delegate count would be 39% Romney, 26% Santorum, 18% Paul and 15% Gingrich. These numbers are dramatically different from the numbers currently being reported by the media.

  11. Mary says:

    @Steven L. Taylor: The rules were changed because of McCains nomination in 08, after competition in only 4 states. All the states also wanted to have more influence and recognition in the primaries, so they changed the rules. Add in the new Super Pac funds and you have candidates dragging the process on even longer.
    They never looked at the big picture, only from a state to state level.