[EM] Majority-Choice Approval

Tue May 7 15:35:42 PDT 2002

Majority-Choice Approval.

	BACKGROUND

Forest Simmons has proposed a method which might be called ‘Three-Level 
Majority-Choice Approval’ - or simply ‘Majority-Choice Approval’.  In this 
method, the voter has three possible choices for marking each candidate:  as 
‘favored’ (two checks), as an ‘accepted’ compromise (one check), or as 
‘disapproved’ (blank, no mark).  An ‘affirmative’ mark, i.e. either 
‘favored’ or ‘accepted’, signifies ‘approved’.

As I understand it, the proposal now includes a welcome clarification by 
Alex Small, namely that (as in usual Approval and more generally in Cardinal 
Ratings methods) a voter may mark any candidate independently of other 
candidates: there is no a priori limit on the number of candidates that may 
be marked into any one of the three categories.

This independence of marking choice is just what is needed to avoid problems 
of ‘clones’ and ‘spoilage’.  Such independence is lacking in forced-ranking 
methods such IRV and Borda, and in some other constrained methods such as 
usual Lone-Mark plurality and its Cumulative generalization.  As a result, 
all these noted methods allow clone and spoilage problems.

Forest earlier proposed and I warmly supported Five-Slot Approval - which 
has a natural version for three rather than five ‘slots’ or levels, namely 
with the same marking choices as in Majority-Choice Approval.  In Three-Slot 
Approval, however, the difference between marks ‘favored’ and ‘accepted’ is 
expressive only, and has no direct instrumental effect on winning the 
election.

In Majority-Choice Approval, if at least one candidate is marked ‘favored’ 
by at least 50% of all voters, then the winner is a candidate with the 
highest number of favored marks.  Otherwise, the winner is a candidate with 
the highest number of approved (i.e., favored or accepted) marks.

	DISCUSSION

Demorep has urged that in general an electoral winner be certified only if a 
majority of voters have in effect voted ‘Yes’ for the winner:  otherwise, 
the position should be filled by the legislature or some other designated 
body.  One can argue pro or con, but it is worth noting that ordinary 
Approval and both the three-level versions all allow a direct way to 
implement the proposal.  Namely - except maybe for some metaphysicians - it 
is reasonable to presume that a mark to approve (i.e., to deem as favored or 
as accepted) amounts to a yes vote, and conversely that a ‘yes’ vote surely 
means at least that the candidate is accepted.

Thus, Majority-Choice Approval will elect a candidate who is favored by a 
majority, and the method could further be used a la Demorep, so as to reject 
all candidates in case none gains majority approval (favored or accepted 
status).

Majority-Choice Approval does have a slight drawback: ‘inconsistency’.  In 
its procedure for deciding a winner, the method is a hybrid: it works one 
way under one condition and another way under another condition.  As for 
almost all such hybrids, the method is inconsistent, in the sense that a 
candidate A may win all precincts but not the entire electorate.

Here this inconsistency can occur if A wins some precincts on account of 
being majority favorite; but wins other precincts, which lack majority 
favorites, on account of being most approved.

For instance, consider an electorate of two five-voter precincts, and a 
contest among five candidates A-E.  Each marked ballot favors exactly one 
candidate X and accepts exactly one other candidate Y - symbolized below by 
the format XY.

Ballots in precinct #1 are:    AB, AB, AB, CB, DB.
Ballots in precinct #2 are:    AB, BA, BA, CA, DE.

A wins precinct #1 as the majority choice and precinct #2 as the most 
approved; but B wins the entire electorate as the most approved.

For Majority-Choice Approval (unlike some other methods) such inconsistency 
is easy to accept.  The reason is simple: we prefer a majority favorite, 
which we may in fact happen to get in some precincts but do not necessarily 
expect to get overall.

	SUITABILITY  FOR  REFORM

Let’s keep in view the Bottom Line for single-winner election reform.  The 
key reform problem is that Lone-Mark plurality tragically makes distinct but 
legitimate voter objectives into mutual spoilers:  voters cannot both 
effectively support more than one favorite, or support both a favorite and a 
lesser-evil compromise insurance candidate.

Ordinary Approval and Three-Slot Approval and Majority-Choice Approval all 
assuredly solve the basic spoilage problem, whereas the far more complex IRV 
sometimes does not.  But, as Forest observed lately (Friday May 3, “Re: [EM] 
Majority Winners and 3-Level Approval”):

	“Three levels is just enough for Favorite, Compromise, and Disapproved, the 
minimum required for solving the spoiler problem without erasing the 
distinction between Favorite and Compromise.  This turns out to be an 
important psychological distinction, the main reason that most IRVists 
believe that IRV solves the spoiler problem better than Approval does.”

Majority-Choice Approval not only truly solves the spoilage problem in a way 
that incorporates the three-level distinction, but it also solves the quite 
different ‘majority-rule’ problem - the other big preoccupation of IRVites, 
who however often confuse it with spoilage.

Namely, in cases where a majority favorite does exist, Majority-Choice 
Approval enables majority rule.

Even better, with Majority-Choice Approval:  unless your greater-evil 
candidate really seems close to being a majority choice, you have no 
incentive to mark your lesser-evil compromise as ‘favored’.   For those who 
still worry about this, a simple variant of Majority-Choice Approval would 
suffice to remove such incentive: namely, define the winning majority 
percentage (for a favorite) to be something larger than 50, say 55.

	OPERATIVE  CONCLUSIONS

I cannot conceal my delight with Forest’s proposal!

The ‘inconsistency’ is a small price to pay for directly confronting and 
solving the main IRVist hangups.  At the same time, we have here a method 
which (more than IRV, or quite unlike it) is readily understood, is readily 
implemented, is monotone, and allows ready precinct-by-precinct tabulation 
and aggregation.

This method is also fully compatible with ordinary Approval.  It is simply a 
version which in return for a bit (literally!) more complexity offers 
notable extra capability.  As a result, advocacies both of  Majority-Choice 
Approval and of ordinary Approval can be harmonized.

To aid the requisite positive spin for public presentation, I have here used 
the name ‘Majority-Choice Approval’.  Maybe an arguably suitable and anyhow 
glitzier title exists and should be used (e.g. maybe simply ‘Majority 
choice’ or ‘Guaranteed majority choice’?).

Joe Weinstein
Long Beach CA USA

_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com

----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em