Know your magnitude

As I have noted previously, the Democratic Party in California will allocate its delegates to the presidential nominating convention by a two-tier semi-PR system. There are 81 statewide delegates that will be allocated proportionally to those candidates with more than 15% of the votes in the state. Then there are 241 that will be allocated in congressional districts, also ‘proportionally’ to those with more than 15% of the vote in the given district.

The congressional districts have a ‘magnitude’ (number of delegates to allocate) that is based on Democratic turnout in recent elections. These magnitudes range from 3 to 6. The average magnitude is about 4.5. A very important consideration is whether your magnitude is even or odd. In many of the even-magnitude districts, Obama and Clinton could win equal numbers of delegates, even if one leads by a wide margin in the votes cast in that district. On the other hand, where the magnitude is an odd number, a very small margin of one candidate over the other in votes will result in an additional delegate.

Equal delegates in even-magnitude districts are by no means guaranteed, however, as Tom Round noted in a comment to the earlier thread:

Someone at Armbinder made the point that, to win 3 of 4 or 4 of 6, you need 60% or 57% of the viable votes, which may fall far below the total votes if one or two runners-up just fail to crack the 15% barrier.

Although Edwards (and various other contenders) dropped out, they remain on the ballot. Some of them may still get significant votes. Maybe not 15% needed to win delegates, but a few candidates combining for 10-12% might affect the allocation. I do not know the precise formula being used, so I can’t say precisely how. But as Tom notes, a 51-33-10-6 split in votes would go 3-1-0-0 under many PR formulas. However, because the California allocation appears to throw out the votes of those under 15% before allocation begins, such a vote split would probably result in 3-3 2-2 even split of the delegates, despite one candidate being quite far ahead in votes.

There are a lot of even-magnitude districts.
Here are the magnitudes by the number of districts with that magnitude:

    3 seats: 2
    4 seats: 26
    5 seats: 19
    6 seats: 6

That’s a lot of districts with the potential for even division if one candidate does not win by a landslide (or if significant votes for the erstwhile contenders do not impact the distribution). In fact, that is 58% of all district delegates allocated in even-magnitude districts, and thus likely to split evenly.

In the extended branch, I will post the actual magnitudes by district. See the PDF posted by the LA Times for the map.

1 5
2 4
3 4
4 5
5 5
6 6
7 5
8 6
9 6
10 5
11 4
12 6
13 5
14 6
15 5
16 4
17 5
18 4
19 4
20 3
21 4
22 4
23 5
24 5
25 4
26 4
27 5
28 5
29 5
30 6
31 4
32 4
33 5
34 4
35 5
36 5
37 5
38 4
39 4
40 4
41 4
42 4
43 4
44 4
45 4
46 4
47 3
48 4
49 4
50 5
51 4
52 4
53 5

0 thoughts on “Know your magnitude

  1. [MSS, although extremely flattered to be cited, I’m also a bit confused because my 51-33-10-6% example involves a 4-seater going 3-1, but then your next sentence refers to a 3-3 split. Pls delete this comment if I’ve missed the point! Thx, TR]

  2. Pingback: PoliBlog

  3. All this even-or-odd-numbers-per-district matters only because, to my surprise, the 81 statewide delegates are all allocated based only on the statewide popular vote — a parallel basis — not calculated on a top-up or compensatory basis.

    And yet I keep reading that the Democratic Party elects delegates on a proportional basis. That’s semi-proportional, parallel. I can’t call it MMM because the districts aren’t single-seat districts, but it’s a two-tier proportional system with parallel tiers, which is less than fully proportional.

    Excuse my Canadian ignorance, but did the Democrats get it wrong everywhere, or just in California?

  4. But with 81 seats at once, the plurality winner statewide may get only half a dozen more delegates than the runner-up – a 10% lead in the (viable) votes would mean only a 36-45 or 37-44 lead, and that could be cancelled out if the runner-up wins a lot of 3-seaters 2-1 and “loses” a lot of 4-seaters 2-2.

    Who’d’a thought (inadvertant) Tullymandering would show up in California?

  5. Wilf, it works that way in all the states. I agree that it’s ludicrous to call the primaries “proportional”–especially given that one fifth of the delegates are “superdelegates”, who aren’t determined by voting among the public but rather by their position within the Democratic party.

  6. > ‘… it’s ludicrous to call the primaries “proportional”…’

    Concurring. We really need some brief and snappy term that means “runners-up get some positions even if short of their proportional share”. [*]

    Enid Lakeman, IIRC, classified Limited Vote (SNTV and MNTV) and Cumulative (including Borda and other points systems) as “semi-proportional”, which is even more likely to have the press putting the mandatory word “complex” before “system of…”

    [*] With an implied understanding that parties and groups under some very low threshold – not over 5% – can’t complain if they miss out: otherwise the threshold would fluctuate too widely with the size of the body being elected (eg, a mere 0.151% would “deserve” proportionate representation in Germany, but you’d need a whopping 1.538% to do so in Malta).

  7. “Parallel semi-PR” is exactly what I called the California rules some months ago, when first noting them. I believe the rules are basically the same on the Democratic side in all states.

    Republicans have more variety, from statewide bloc plurality to congressional-district bloc plurality, to various “proportional” systems.

    There is substantial malapportionment in both parties (where district allocation is used), though I do not know if all of the Republican primaries that use district multi-seat plurality have equal numbers of delegates in each district (as has California).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.