[EM] [CES #8844] Vague impossibility conjecture for monetizing voting systems

[Abd ul-Rahman Lomax]

At 11:27 AM 6/19/2013, Warren D Smith wrote:
>  CONJECTURE that no good monetization scheme can exist for
>any reasonable voting system in large elections...
>
>I have stated such a conjecture and sketched a proof (which is not a
>real proof) of this impossibility conjecture here:
>     http://rangevoting.org/MonetizedRV.html#monetfails

Vague is an understatement. Start with "good" being undefined. On the 
page, it becomes "reasonable," which merely papers over the lack of 
precision, and precision is essential to proofs. (However, Warren 
does, above, notice that it is not a "real proof." It's really just a 
collection of anti-monetization arguments.)

The core argument is that, allegedly, the cost of voting will be so 
high that most people won't vote. That is *not* an argument against 
the system. People who don't vote don't participate, have no cost, 
and have no benefit. However, the assumption that "poor people" would 
not vote is just that, an assumption. Poll taxes, as Warren has 
pointed out, are unconstitutional in the U.S., in general.

However, it is possible to have monetized voting without poll taxes. 
I proposed an approach, which would involve allowing every voter a 
default, base sum that they can spend as they choose, on their votes. 
This sum might be small, say $1, but it would allow these people to 
vote. Adding weight to the vote by adding more money is really no 
different, in substance, from spending money on campaigns, simply more direct.

Now, such a system might not be accepted, and may not even be worth 
proposing, for that reason, but that is *not* an impossibility 
argument, it's a logistical, practical argument.

The default amount allowed citizens would be adjustable. It might be 
more than $1. The vote might include a designation of a *specific 
charity* to which a refund would be donated, if refunds were below 
some value. Otherwise the "poor voter," might get a check. Notice: if 
they vote in such a way as to get a check, they are voting *against 
their own interest*. Rather, they get a check, sensibly, if they vote 
according to their own personal value in the election, and they get 
the compensation if the result is different. It works.

In *some contexts*, this kind of voting might make sense. In some 
contexts, there is no restriction on poll taxes. A charity might run 
a poll as to how to distribute donations, among legitimate charitable 
purposes, consistent with the purpose and management of the charity, 
and the poll might be weighted by donation. The basic "donation" 
could be the membership dues for the organization. That gives the 
voter the right to vote. And then the voter can offer conditional 
donations for specific causes. If the voter "loses," the voter might 
get an additional amount to be used for voting in future elections; 
essentially, they would not have "spent" their donated funds. So 
this, then, shades into the time-cumulative methods, being more 
specific with easy bookkeeping. It's possible to design the system to 
be workable and practical.