[EM] Market-Based multiwinner system II

[Warren Smith]

Here's a different rule set which might be better for the internet
age.   It pretty much depends on modern technology (fortunately, or
not).

W-winner election with C candidates, 0<W<C.   We assume there is a
public "bulletin board" on which the "total" for each candidate is
announced.  These public totals are updated in real time.   Each voter
has 1000 pseudodollars and can allocate them in any way to the C
candidates, and can change her allocation at any time until the
pre-announced final deadline.  When the deadline comes, the W
"richest" candidates win.

This system automatically seems to obey a "proportionality theorem."

It is very simple (provided you are willing to trust the technology).

It also elects parliaments that are "optimal" in the sense that if there is any
reasonably obvious change that enough voters feel would be an
improvement, they can make it happen?

A criticism is: if not-very-good candidate B is currently "rich"
enough to win a seat,
and better candidate C isn't, voters may be afraid to move their
"money" from B to C
because to do so they'd have to pass through an intermediate state
where neither B nor C was winning, but (therefore) ultra-bad candidate
U was winning.   So they'd stay with B
for fear of "wasting their vote."  This -- if it happens -- would be a
vicious cycle causing initially-ahead candidates to stay ahead
regardless of their actual quality, and thus causing initial
perceptions from propaganda and big money to have too much power.

My original rules discussed at
   http://rangevoting.org/MarketBasedVoting.html
seem to be better in that respect (seems to be no such thing as a
"wasted vote"), and do not require high technology.

-- 
Warren D. Smith
http://RangeVoting.org  <-- add your endorsement (by clicking
"endorse" as 1st step)
and
math.temple.edu/~wds/homepage/works.html