Resource allocation: dividing a fixed amount across items

Resource allocation is the division of a fixed amount, such as 100 tokens or a real budget, across a set of items. Unlike a ranking, an allocation expresses relative size: giving housing 40 tokens and transport 10 says something a first and second place cannot. In the My Vote Forward demo, allocation applies to a defined fictional pool of new funding, and zero is an explicit choice, not missing data.

Definition

A resource allocation task gives a person a fixed total and a list of items and asks how much of the total each item should receive. The total constraint is the defining feature: adding to one item means taking from another, so the answer reveals tradeoffs. The site glossary’s working definition adds two rules used throughout My Vote Forward: allocation is kept separate from priority, and zero is an explicit choice (see the short entry in the glossary).

An allocation can be hypothetical (a survey exercise with tokens) or binding (a public body actually spending money according to the result). Those are different acts with different stakes, and a result from one should not be described as the other.

Origin and history

The best-known binding use is participatory budgeting, which the Participatory Budgeting Project describes as “a democratic process in which community members decide how to spend part of a public budget,” and which it says “started in Porto Alegre, Brazil, in 1989, as an anti-poverty measure”.⁠¹ In New York City, the City Council reports that “in 2011, four New York City Council Members” launched a process in which residents help decide how to spend capital funds, for projects that “cost at least $50,000”.⁠²

Economists have also designed allocation mechanisms meant to capture how strongly people care. Lalley and Weyl’s quadratic voting proposal (2018) has individuals spend “voice credits” on votes at a quadratic cost; the authors argue it can “correct the failure of existing democracies to incorporate intensity of preference”.⁠³ That claim is contested and the authors themselves call the mechanism “still-nascent”.⁠³

How resource allocation is measured and used

In survey or civic settings, allocation usually works through a few design choices that change results:

• The total. 100 tokens invites percentages; a dollar figure invites people to think about real costs.
• The list. Items left off cannot receive anything, so the list is itself a decision. See Question design.
• The default. Whether the screen starts at zero for every item or with an even split affects answers. My Vote Forward’s rule is that zero is an explicit choice.
• The scope. An allocation of “new funding” is different from an allocation of an entire existing budget, most of which is committed by law or contract.

When allocations are combined, the average share per item is the usual summary, but averages can hide that some people give an item nothing while others give it a great deal. Reporting the distribution, not just the mean, is the honest form.

Common confusions and neighboring terms

• Allocation vs. ranking. A ranking is ordinal; an allocation is closer to a ratio measure, because it records size within a fixed total.
• Allocation vs. priority. A person may consider public safety the most important subject yet allocate little new money to it because they think it is already well funded. Importance and funding need are different questions. See Tradeoffs.
• Allocation vs. effectiveness assessment. Deciding how much should go to a subject is not the same as judging whether existing spending in it works.
• Hypothetical vs. binding. A token exercise is a statement of preference; participatory budgeting is a decision with real money.

How My Vote Forward uses the term

My Vote Forward treats allocation as a separate measurement from priority. In the demo, people divide a defined fictional pool of new funding, and the result is shown as simulated. The framework does not present any allocation as a budget recommendation to a real government. The reasoning for separating importance, funding, and effectiveness is set out in Tradeoffs, and the measurement issues in Measurement.

Unresolved questions

• Should allocation exercises use real budget figures from a named jurisdiction, at the cost of complexity, or simple tokens, at the cost of realism?
• How should committed spending (debt service, mandated programs) be shown so people do not allocate money that is not actually available?
• Do intensity mechanisms such as quadratic voting improve the signal, or do they mainly reward people who understand the mechanism?