Proportional Allocation (PA) rules let geographic and other groups within a jurisdiction fund their own projects without new layers of taxes and administration. This makes (hidden) empires less profitable.
A majority that becomes a minority can continue funding some of its priorities -- so project funding rises and falls smoothly.
When the majority controls all spending, their last allocation adds little to their happiness. After they spend $900... on their favorite projects, the last $100... funds a low priority. But that same $100... could fund the top priority for a large minority -- making the minority much happier.
In economic terms, distributing a small amount of spending power can increase the utility value purchased, and distribute opportunities and incentives as well. In political terms, an equitable distribution promotes legitimacy and compliance, social peace and co-operation.
Where there can be only 1 winning proposal, it should be required to get majority support. Therefore power sharing must not apply to policy decisions or essential operating funds. But fair-share funding for large minorities can be applied to discretionary funds. In fact, most winning proposals will get contributions from less than half the reps, showing that we are all, at times, members of minorities.
These voting rules challenge market economics as a means of allocating scarce resources and asks, "Who should design society, the voters or the rich?" Fair, efficient funding rules may increase respect for public funding. They may re-balance our private affluence and public squalor. Or they may spend public funds on private goods, "pork" in other words.
A proposal that stretches policies or funding rules must be subject to a majority veto that keeps it off the ballot. A number of co-sponsors also may be required, and the items with the most co-sponsors might be listed at the top of the ballot.
A low turnout for voting necessitates using a majority rule such as MVP or Condorcet Series rather than a proportional rule.
[ Most public goods should not have to win their funding through a minority funding process. To do so increases the temptation and reward for free riding and all that goes with that: bluffing, cynicism, and the under funding of those shared items. Likely free-ride items could have their funding set by a Median Voter Process. Items that get nothing then go on the Minority Funding Process or STA ballots. (The next sections explain these sophisticated rules.) The size of the discretionary fund is also set by MVP. Thus majority voting funds basic needs while proportional rules decide only which extras to buy. Free rides are not a great concern because no one gets free basics while saving money for extras. But then the proportional voting does not equalize power much.
A proportional allocation system is unlikely to give too much power to minorities. All policies are enacted by a majority rule, Condorcet's. That means agencies spend their budgets supporting central policies. All departmental budgets are set by majority rules: MVP and H-Z. Only a small part of the budget is discretionary and available for proportional allocations by all reps. Finally, most PA funds will be allocated by the majority, if it is cohesive.
Limited as it is, the right to proportional allocation is a major expansion in the concept of democracy -- similar to the earlier expansions in the right to vote and the right to representation. I hope your organization will have the courage to be among the first to give some power to its minority groups.
Here is a funding analogy like the STV analogy. This simplistic Share Allocation Rule (SAR) gives a share of the discretionary budget to each rep. A rep gives her share to her favorite projects. Each item's contributions are totaled. Reps then move their shares to adjust the total for each project. That also avoids any need for the rule to transfer excess funds. Allowing partial funding of an approved project to accumulate for a few years lets a small group of reps fund a large project. It also avoids the need to eliminate the weakest items and transfer funds until all items are either fully funded or eliminated.
SAR has serious flaws: 1) It allows public funds to be spent without effective discussion or agreement. Like a market economy, it meets only the weakest definition of "collective judgment". 2) It ties up some funds with little or no benefit for years. 3) It gives strong rewards and clear opportunities to try for free rides on popular projects and all that goes with that: bluffing, cynicism, and the under funding of those shared items. 4) It may spend public funds on private goods and so confuse the electorate's concept of public goods and weaken support for collective action. STV forms the basis of a rule that can avoid or minimize these problems.
.
2) Voting is simply ranking top preferences. Like a shopper with a limited budget, an STA voter gives top rank to the best deal.
3) Both rules give each ballot a weight, a limited share of power. A ballot's weight is used up as it helps elect candidates or fund projects. A voter's STA weight is money, his share of the budget.
Weight = Budget / Voters
4) For STV the quota needed to win a seat is a large share of the ballots cast. For STA an item's cost is also its quota; the amount of money it needs to win may be small. Cost quotas may vary widely.
5) Under STV a voter usually helps elect just 1 rep. Under STA a low-cost item might cost its supporters only a little of their weights. So a voter might help fund several items.
6) Winning 2 STV seats is always better than winning 1. But buying a second swimming pool might be a waste of money. Sometimes STA must avoid funding two of a kind.
7) Free rides are more likely in council voting than in elections. A Condorcet step and some options below discourage free rides.
8) The budgets are continuous variables and the voting rule should let a rep effect each amount.
Because of the differences above, electoral STV must be adapted for allocations. STV has been criticized because the sequence of wins and eliminations may change who wins. And STV would eliminate a candidate with few firsts but many seconds and thirds. Here are some options to avoid those problems.
1) Each voter will rank most of the proposals and gives them "$ votes". A $ vote is the amount the voter thinks the item should get if it is funded. There is no limit on a ballot's total $ votes. That is all the voter will need to do.
2) Each item's sponsor suggests a funding level for it. The organization's by-laws specify a range of allowable $ votes such as 25% to 300% of the suggested amount. The sponsor may set a narrower range or a fixed price.
3) The by-laws also set a Minimum number of Offers (MinO) required for a proposal to win funding. This replaces the quota needed to win an STV seat.
4) A ballot "offers" and "reserves" money for as many of its top preferences as it can afford. The money a ballot offers to an item is
$ vote / MinO. The money offered is not "contributed" until the item gets at least MinO offers.
The money a ballot actually contributes to a winning item is $ vote / actual number of contributors. That amount is subtracted from the ballot's money supply or weight. Each ballot normally starts with a weight = budget / number of ballots.
[ A partial offer is made when a ballot has offered or contributed some weight and the amount it has left is less than $ vote / MinO. This fraction of a vote helps the item reach MinO. Weight left / ($ vote / MinO) ]
5) A winner's budget is the average $ vote given by its contributors. That is also the total of their contributions.
[ A big contributor might say with resentment "I paid double for an under-funded failure." -- and assert she should not have to pay more than others who will enjoy the same public good. But her $ vote had the intended effect, raising the item's budget. And all contributors pay more than non-contributors. An item's funding could equal the median $ vote and every contribution would equal that median $ vote / the number of contributors. Low contributors would, of course, object. This option could be invoked when more than [15]% but less than 50% of the item's $ votes are near an extreme. It is not compatible with option 4. ]
6) Conditional items let voters enter a rank for "A if B has been funded" to avoid buying 2 of a kind when 1 will do. STA is very good at organizing reps (or their ballots) into interest groups so they are not divided and defeated by 2 of a kind. At least 1 will be funded if interest is strong. But STA is weak at preventing project duplication. Any group improves its chances for efficient funding by agreeing on 1 proposal before voting.
7) Don't eliminate the weakest item; just suspend or by-pass those preferences. When an item is suspended, its $ votes are still offered but not reserved. That means each ballot must offer those $ to some lower preference item(s). If an item with suspended votes gets more offers later, it may yet get the MinO needed to win funding.
Suspending offers leaves a ballot's weight pending for more items than it can afford. So after the first suspension, the tally can fund new winners only 1 at a time. Otherwise a ballot might spend its last weight on both reserved and (several) suspended items.
Ties are common when selecting the weakest item to suspend. There are several ways to break ties. The quickest uses the ranks of items in a Condorcet Series tallied before the STA tally. Suspend whichever tied item was least popular according to Condorcet's rule. This finds an item unlikely to get more offers. Here is a more accurate way to find the item least likely to get more offers. Give each ballot 10% more weight for a tie-breaker vote: 10% times weight not yet contributed. If that fails to break the tie, a larger percentage can be used.
Ties are also common among winners, when 2 or more reach MinO after suspension of another item. Again there are several ways to break ties. The best method funds the 1 which had the most initial offers. It has few offers transferred down from suspended items. If some items are still tied, use the ranks of items from the Condorcet Series. Fund whichever tied item was most popular according to Condorcet's rule. Very few ties will need to be broken randomly.
Those are the most important features of STA. Let's pause and look at what they do. The MinO quota actually combines 2 quotas: a minimum number of ballots and the item's cost. The item must fill both quotas. The quota of ballots must be filled to prove the item's breadth of support. The philosophy is that a number of people is required to qualify an item as a public good. It implies that sharing an item increases its value; not merely an additive increase but with synergy. Perhaps the act of cooperation is a public good in itself. The monetary quota must be filled to proven the intensity of support.
MinO sets no upper limit on the cost of projects. An item can win more than initial ballot weight times MinO. There is no limit on the number of voters who may contribute to any item.
Ballot weights assure each a fair and limited share of power. MinO assures breadth of support while $ votes assure intensity. We need a way to assure a high utility value per dollar -- something governments and other large bureaucracies often do poorly. STA options addressing this are explored along with other funding rules in funding.htm. .
Some of the comics fund is awarded by a central Condorcet Series Allocation (CSA). The comics are ranked as first Condorcet winner, second, third and so on. The CSA part of the budget should be enough to fund several of the top comics. A winner is funded with its median $ vote but voters lose no STA weight. The rest of the budget then goes to the STA tally.
Funding Condorcet's winners before STA reduces the free-rider incentive. For example, Voter: "I know Calvin and Hobbes. is going to win. Why should I waste part of my weight supporting it?" Pollster: "If it is sure to win, it probably will during CSA. Supporters' STA shares won't be reduced. You probably cannot predict later winners of STA voting." So a voter only needs to think about ranking his true preferences. Then Political Sim or other software easily handles the tally, including any STA options.
Each of these rules asks voters to give first preference to the most cost-effective option, 2nd to the 2nd most cost-effective and so on. So one ballot serves CSA and STA tallies. .
How many people should be required to back a project paid for with public funds? In setting MinO, consider the shares of votes for current interest groups. MinO can be equal for all items, or a proportion of their (suggested or current-average) costs so expensive items require more contributors, or a constant plus a proportion.
[ Splitting the discretionary budget into pieces for different rules limits the size of the largest purchase possible. Suppose the by-laws call for CSA with 20% of the budget, then STA33 with 60% and finally STA11 with the last 20%. The budget is $100,000. CSA gets $20,000. STA33 gets $60,000 -- but it would take 100% of the reps each casting $ votes of $60,000 to give it all to 1 item. So the realistic cost limit is much smaller.
The tally software could create 5 or 10 sets of winners. Each set would use a different ratio of CSA to STA and size(s) of MinO. Voters would have the final choice in a Condorcet tally selecting the most popular set.
The ratio of central to distributed goods should not be hidden in the voting rule(s) nor rigidly set in advance because that would sometimes prevent finding the most popular set of winners. But democratic voting has often failed to protect minority rights. So for minorities to have a right to power, that rights must be protected in the by-laws or constitution. This means setting a minimum on the distributed or STA share of discretionary funds. It also means setting a maximum on MinO because a MinO of 50% would change STA to a majority rule.
[ A good research project would investigate voters' preferences for politically central versus distributed projects. The spatial distributions studied by geographers may provide a useful analogy. ]
Twenty people picking 5 pizzas from 30 on a menu could use another rule like STA. The Joint Allocation Rule adapts STA for collectively buying personal goods in bulk lots.
.
Return to budgeting departments.
© 1997 Robert Loring Reprints are permitted.