Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Manifund's approach doc #129

Open
wants to merge 17 commits into
base: main
Choose a base branch
from
Open

Manifund's approach doc #129

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
b11cf58
writing
Mar 13, 2024
dd4bb61
Merge remote-tracking branch 'origin/main' into article
Mar 28, 2024
bad70af
add Articles section
Mar 28, 2024
bd6f239
move articles into their own directory
Mar 28, 2024
deceac2
--amend
Mar 28, 2024
063e30d
no icons for nested tab articles; wrap titles
Mar 28, 2024
f9d50e9
copy everything over from the notion
Mar 29, 2024
a8d9d36
sub-nested and default-open children
Mar 29, 2024
d69e4b0
sidebar scroll
Mar 29, 2024
9f0161d
all pages
Mar 29, 2024
3a8d3be
titles, formatting, uppercasing, etc
Mar 30, 2024
f615285
all sidebar titles
Mar 30, 2024
af2343f
generic article page component
Mar 30, 2024
26aa7f1
titles; how to get involved page
Mar 30, 2024
b1906c8
convert everything to use ArticlePage component
Mar 30, 2024
567dd01
previous button; added explanations to 4 and 5
Apr 2, 2024
d8e1af8
explanations of programs
Apr 2, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
55 changes: 55 additions & 0 deletions app/articles/approach/1/page.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,55 @@
import ArticlePage from '@/components/article-page'
import { Fragment } from 'react'


export default function ApproachPage1() {
const content = (
<Fragment>
<p>
Manifund’s sister company, Manifold, runs experiments with
prediction markets. Because Manifold runs on play money, and because
users get to participate on the market-creation side as well as the
prediction side, it’s arguably the best playground in the world for
experimenting rapidly with financial mechanisms. Since official
currency isn’t involved, both Manifold and its users (and the two in
combination) can run experiments rapid-fire without jumping through
many hoops. For instance, Manifold has tried various types of
automatic market-making algorithms, mechanisms for multiple-choice
questions, short selling, limit orders, contests, leagues, even a
dating app spinoff. And users have made bots, non-question questions
like “[thing] stock (never resolves)”, markets that serve as bounties to
incentivize particular events to happen, questions with meta
resolution criteria like “Will the total volume of YES trading on
this question be more than x mana?”.
</p>
<p>
Manifund is different, because lots of real money is moving through our
programs. But we’re still small and fast-moving,
and our ethos is similar: we aim to iterate quickly on
small- and medium-scale experiments testing out the best ways
to fund charity and public goods.
</p>
<p>
Effective altruism originally asked the question, “What properties make
a charitable project useful, efficient, and valuable?” Manifund is an
attempt to extend that question in two ways:
</p>
<ol>
<li>
What kinds of systems incentivize projects that have those properties?
</li>
<li>
How can we use improvements to the grantee experience to make funding
programs more effective?
</li>
</ol>
</Fragment>
)
return <ArticlePage
articleTitle="A Framework for Funding Experiments"
pageTitle="Economic Experimentation"
nextLink="/articles/approach/2"
nextLinkText="Next: Some programs we've run"
content={content}
/>
}
65 changes: 65 additions & 0 deletions app/articles/approach/2/page.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,65 @@
import ArticlePage from '@/components/article-page'
import { Fragment } from 'react'
import Link from 'next/link'

export default function ApproachPage2() {
const content = (
<Fragment>
In 2023, Manifund ran a few different granting programs.
<ul>
<li>
<strong><Link href="/about/regranting">Regranting</Link></strong>: Donors delegate grantmaking budgets to regrantors, who
allocate the funds to the most promising projects they come across.
</li>
<li>
<strong><Link href="/about/impact-certificates">Impact markets</Link></strong>: Donors offer retroactive prize funding to
successful and impactful projects, and project founders sell shares in these prizes to fund their projects.
</li>
<li>
<strong><Link href="/about/open-call">Open call with assurance-contract donations</Link></strong>: Manifund runs an open call for projects,
which operates like Kickstarter for charity: if a project receives donation offers totaling at least its minimum funding goal,
then the donations kick in and it receives funding.
</li>
</ul>
<p>
(You can read more about these programs in our <Link href="https://www.lesswrong.com/posts/cxjRjs5BWw5pBHZiA/manifund-2023-in-review#2023_Programs">2023 retrospective</Link>.)
</p>
<p>
Is there anything general we can say about these kinds of experiments?
</p>
<p>
Take the system of impact certificates, for example. There are a couple
of differences from how charity is traditionally funded. The donations
come after the fact, for one thing, rather than in advance. And
there’s a middleman involved, who doesn’t even necessarily have
to have altruistic motives.
</p>
<p>
Theoretically, this leads to neat results. There’s no need to learn
new ways to evaluate charities, or to make people more altruistic. The
addition of an intermediate market makes it more efficient to figure
out the most efficient allocation of funds by the donor’s own
metrics.
</p>
<p>
But also, it’s easy to feel like the mechanism just came out of nowhere. Each individual step of this
process is perfectly understandable if you understand the analogous
element of private markets. But altogether, it’s a lot of new moving
parts to conceptualize at once, and it’s not obvious how to come up
with similar ideas or how to evaluate them.
</p>
<p>
So we’ve come up with a first pass at a general framework for
thinking about the design of novel funding mechanisms.
</p>
</Fragment>
)
return <ArticlePage
articleTitle="A Framework for Funding Experiments"
pageTitle="Some Programs We've Run"
prevLink="/articles/approach/1"
nextLink="/articles/approach/3"
nextLinkText="Next: The six steps of funding"
content={content}
/>
}
83 changes: 83 additions & 0 deletions app/articles/approach/3/page.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
import ArticlePage from '@/components/article-page'
import { Fragment } from 'react'

export default function ApproachPage3() {
const content = (
<Fragment>
<h2>1. Predict</h2>
<p>
<strong>What does it entail?</strong> Make a good guess about which
projects will be likely to achieve their aims, what impacts those
achievements will have, and which teams can achieve the best
results for the lowest cost.
</p>
<p>
<strong>What are the desirables?</strong> Predict accurately and at
low cost.
</p>
<h2>2. Front</h2>
<p>
<strong>What does it entail?</strong> Supply the immediate
funding for the costs a team will incur in the process of
doing their project.
</p>
<p>
<strong>What are the desirables?</strong> Predict accurately and at
low cost.
</p>
<h2>3. Execute</h2>
<p>
<strong>What does it entail?</strong> Carry out the project.
</p>
<p>
<strong>What are the desirables?</strong> Do something people
actually want. Do it well. Do it inexpensively. Don’t run off
with the money.
</p>
<h2>4. Evaluate</h2>
<p>
<strong>What does it entail?</strong> Determine how valuable the
project was.
</p>
<p>
<strong>What are the desirables?</strong> Figure out both how successful
the project was and how much impact that success had. Make the targets clear
so founders know what to aim for, but avoid using measures that are too easy
to game. Account for
externalities – for example, if the project was moderately beneficial
to the group that commissioned it, but also benefited lots of
other people as well, then we’d like that to be factored in.
</p>
<h2>5. Pay out</h2>
<p>
<strong>What does it entail?</strong> Allocate money as a function
of the evaluation of the project’s impact. In some cases, like
traditional grantmaking, the fronting is also the payout (although
to some extent this is a question of definitions; you could also
think of traditional grantmaking as just skipping the fronting
step).
</p>
<p>
<strong>What are the desirables?</strong> Incentivize the execution
of projects with the most impact.
</p>
<h2>6. Benefit</h2>
<p>
<strong>What does it entail?</strong> Enjoy the benefits
of whatever it was that got funded.
</p>
<p>
<strong>What are the desirables?</strong> Be helped by the project.
Do not be harmed by the project.
</p>
</Fragment>
)
return <ArticlePage
articleTitle="A Framework for Funding Experiments"
pageTitle="The Six Steps of Funding"
prevLink="/articles/approach/2"
nextLink="/articles/approach/4"
nextLinkText="Next: Some examples"
content={content}
/>
}
186 changes: 186 additions & 0 deletions app/articles/approach/4/page.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,186 @@
import ArticlePage from '@/components/article-page'
import { Fragment } from 'react'

export default function ApproachPage4() {
const content = (
<Fragment>
<p>
Let’s apply this framework to some existing fundraising models, as well as some more experimental ones.
</p>
<h3>The startup model</h3>
A startup founder raises money to run an early-stage company by selling equity in the company’s stock. <br/>
<p>
<strong>Predict:</strong> investors
</p>
<p>
<strong>Front:</strong> investors
</p>
<p>
<strong>Execute:</strong> founder
</p>
<p>
<strong>Evaluate:</strong> consumers
</p>
<p>
<strong>Pay out:</strong> consumers
</p>
<p>
<strong>Benefit:</strong> founder (as profit), investors (as profit), consumers (as consumer surplus)
</p>
<h3>The bootstrapping model</h3>
A founder starts a company with their own money, rather than selling equity early. <br/>
<p>
<strong>Predict:</strong> founder
</p>
<p>
<strong>Front:</strong> founder
</p>
<p>
<strong>Execute:</strong> founder
</p>
<p>
<strong>Evaluate:</strong> consumers
</p>
<p>
<strong>Pay out:</strong> consumers
</p>
<p>
<strong>Benefit:</strong> founder (as profit), consumers (as consumer surplus)
</p>
<h3>The government model</h3>
The government taxes citizens and uses the proceeds to pay for public-benefit projects. <br/>
<p>
<strong>Predict:</strong> elected officials or bureaucrats
</p>
<p>
<strong>Front:</strong> citizens (as taxpayers)
</p>
<p>
<strong>Execute:</strong> contractors
</p>
<p>
<strong>Evaluate:</strong> citizens (indirectly, as voters)
</p>
<p>
<strong>Pay out:</strong> citizens (as taxpayers)
</p>
<p>
<strong>Benefit:</strong> citizens, contractors (as profit), officials (as goodwill for
subsequent elections)
</p>
<h3>The wealthy philanthropist model</h3>
A wealthy person or family donates large sums of money to charity. <br/>
<p>
<strong>Predict:</strong> donor
</p>
<p>
<strong>Front:</strong> donor
</p>
<p>
<strong>Execute:</strong> donor
</p>
<p>
<strong>Evaluate:</strong> donor
</p>
<p>
<strong>Pay out:</strong> donor
</p>
<p>
<strong>Benefit:</strong> charity beneficiaries
</p>
<h3>The classical effective altruism model</h3>
Altruistically-minded people try to donate to charities that will have the most marginal impact. <br/>
<p>
<strong>Predict:</strong> donor, with an Our World In Data tab open
</p>
<p>
<strong>Front:</strong> donor
</p>
<p>
<strong>Execute:</strong> donor
</p>
<p>
<strong>Evaluate:</strong> donor
</p>
<p>
<strong>Pay out:</strong> donor
</p>
<p>
<strong>Benefit:</strong> charity beneficiaries
</p>
<h3>The charity evaluator model</h3>
An organization assesses the effectiveness of various charities, and then donors either donate to those charities
directly, or donate to the organization which then funds its chosen charities. <br/>
<p>
<strong>Predict:</strong> charity evaluator
</p>
<p>
<strong>Front:</strong> donors
</p>
<p>
<strong>Execute:</strong> founders
</p>
<p>
<strong>Evaluate:</strong> charity evaluator
</p>
<p>
<strong>Pay out:</strong> donors
</p>
<p>
<strong>Benefit:</strong> charity beneficiaries
</p>
<h3>The impact certificate model</h3>
Donors offer to retroactively award prizes for successful charitable projects.
The founder of a such a project funds their efforts by selling equity in whatever prize money the project later receives. <br/>
<p>
<strong>Predict:</strong> investors
</p>
<p>
<strong>Front:</strong> investors
</p>
<p>
<strong>Execute:</strong> founder
</p>
<p>
<strong>Evaluate:</strong> donor
</p>
<p>
<strong>Pay out:</strong> donor
</p>
<p>
<strong>Benefit:</strong> founder (as profit), investors (as profit), charity beneficiaries
</p>
<h3>The quadratic funding model</h3>
Philanthropic sponsors fund a matching pool, whose funds are used to match donations to public-goods
projects in a way that incentivizes small donors to donate to the projects that most benefit other donors
too. <br/>
<p>

<strong>Predict:</strong> contributors
</p>
<p>
<strong>Front:</strong> contributors and matching-pool sponsors
</p>
<p>
<strong>Execute:</strong> founder
</p>
<p>
<strong>Evaluate:</strong> contributors
</p>
<p>
<strong>Pay out:</strong> contributors and matching-pool sponsors
</p>
<p>
<strong>Benefit:</strong> contributors
</p>
</Fragment>
)
return <ArticlePage
articleTitle="A Framework for Funding Experiments"
pageTitle="Some Examples"
prevLink="/articles/approach/3"
nextLink="/articles/approach/5"
nextLinkText="Next: Observations about these examples"
content={content}
/>
}
Loading