Alice gives $100/month to the Democrats. Bob gives $100/month to the Republicans. They're just burning money fighting each other. They could both agree to stop and donate to a worthy cause like the Against Malaria Foundation instead. Causeum does this automatically—you just state how much you value different causes, and the platform routes your donation where it does the most good.
Set your monthly donation. Pick an amount. Your card is charged at the start of each cycle. If payment fails, you simply sit that round out.
Rate every cause. Score each cause from $0.00 to $1.00—how much a dollar of funding there is worth to you. You can update these anytime.
The optimizer does the rest. Everyone's payments go into escrow. The optimizer allocates the entire pool across causes to maximize total value—finding mutual compromises where donors on opposite sides both come out ahead. Money goes out to causes. Done. Each round is a fresh optimization.
You get a receipt showing exactly where your money went and how much more effective it was than donating directly.
Every trade is mutual. Both sides give something up. Both come out ahead.
Grace donates $100/month. Her top cause is Emily's List ($1.00). Sam donates $100/month. His top cause is Club for Growth ($1.00). They're on opposite sides. But they both moderately like VoteVets.
| Emily's List | Club for Growth | VoteVets | |
|---|---|---|---|
| Grace | $1.00 | $0.10 | $0.85 |
| Sam | $0.10 | $1.00 | $0.90 |
If both donate directly, Grace sends $100 to EL, Sam sends $100 to CFG. They cancel each other out. Pure waste.
The optimizer routes both to VoteVets instead. Both sacrifice. Both gain more than they lose.
Grace's math: she gave up $0.15/dollar sending her $100 to VV instead of EL. Sacrifice: $15. But Sam's $100 also arrived at VV—a cause she values at $0.85. Gain: $85. Net: +$70.
Sam's math: he gave up $0.10/dollar sending his $100 to VV instead of CFG. Sacrifice: $10. But Grace's $100 also arrived at VV—a cause he values at $0.90. Gain: $90. Net: +$80.
There's nothing to reverse because the money is already spent. Each round is a fresh pool. No positions, no liens, no lockups. Just optimally allocated donations.
The trade works because both sides value receiving money at VoteVets. But there's a bonus: $100 that would have gone to Club for Growth didn't. And $100 that would have gone to Emily's List didn't. Neither person targeted the other's cause. The defunding is a side effect of both people converging on common ground.
The trade is justified entirely by the subsidy—receiving the other person's money at a cause you value. Defunding what you hate is a bonus you don't have to factor in.
We simulated 50,000 donors with 12 causes and realistic partisan starting preferences—40% of each side's money in their party flagship (DNC or RNC), the rest in aligned causes. The optimizer ran exhaustive bilateral trades using only the subsidy criterion: does receiving the other person's money at the compromise cause outweigh my sacrifice?
89% of donors found a mutually beneficial trade. 89% of all money was redirected. Subsidy surplus averaged $0.90 per dollar donated—meaning every dollar your trading partner put toward the compromise cause was worth $0.90 to you, while your sacrifice averaged $0.10.
Where money flowed: the DNC collapsed from $9.3M to $945K. The RNC from $9.3M to $931K. Emily's List from $3.4M to $1.4M. NRA from $3.5M to $1.2M. The money went to causes both sides value: VoteVets surged from $3.5M to $22.8M. Conservation Voters grew to $5.9M. Americans for Prosperity to $4.6M. Against Malaria Foundation went from $0 to $608K. GiveWell from $0 to $587K.
The surplus is real and concrete. When Grace's $100 arrives at VoteVets, Sam values that at $90 (he values VV at $0.90). Sam sacrificed $10 to be in VV instead of CFG. So Sam's net surplus is $80—he received $90 of value for a $10 sacrifice. That's real willingness-to-pay. Across 24,862 bilateral trades and 50,000 donors, the total subsidy surplus was $41.3 million from a $46.1 million pool.
Causeum takes a small percentage of the subsidy surplus it creates for you. If the optimizer can't find you a beneficial trade, your donation goes straight to your top cause and you pay nothing.
You donate $100 to the DNC. The optimizer finds a trade: you and an RNC donor both move to VoteVets. You sacrificed $15 (VV is $0.85 to you vs DNC at $1.00). But the RNC donor's $100 arrived at VV—a cause you value at $0.85. Subsidy gain: $85. Net surplus: $70. Causeum takes a small share of that $70.
If no trade improves your situation, your $100 goes to the DNC unchanged. No fee.
In our 50,000-donor simulation, the average trader's subsidy surplus was $831 on a $927 donation (90%). At 5% of surplus, the fee is ~$42 per trader—about 4.5% of the donation. The trader keeps 95% of the value the system created for them. At that rate, total monthly revenue from 50,000 donors would be $2.1M.
Every receipt shows two things: the subsidy surplus (why the trade was made) and the defunding effect (an informational bonus showing what else changed). The trade is justified entirely by the subsidy. The defunding is a side effect you can see but that doesn't drive the optimizer.
Default: $100 to Emily's List (your top cause, $1.00)
Optimized: $100 to VoteVets ($0.85 to you)
Your sacrifice: −$15.00 ($100 × (1.00 − 0.85))
Subsidy gain: +$85.00 (their $100 arrived at VV, which you value at $0.85)
Net subsidy surplus: +$70.00 ← this is why the trade happened
Defunding effect: −$10.00 (their $100 left CFG, which you valued at $0.10—you lost that incidental value)
Combined surplus: +$60.00
The optimizer couldn't find a mutually beneficial trade for your donation this round. Your $100 went directly to Emily's List, exactly as if you'd donated on your own.
Fee: $0.00
A note on negative valuations: with positive valuations only ($0.00–$1.00), the defunding effect is always negative or zero—you lose whatever incidental value the other person's money was generating at their old cause. An advanced option may in the future allow negative valuations, which would flip the defunding line positive (expressing that funding a cause actively hurts you). This creates larger trades but introduces a cobra risk: someone could fake an extreme starting position to attract trade partners who overpay. Both components are always shown separately so you can evaluate the trade for yourself.
There's nothing to extract. The money is spent every round. There are no positions to manipulate, no stock to resell, no funds to withdraw. Donations go out to causes the moment the optimizer finishes.
You can only do better, never worse. The optimizer only moves your donation if the new allocation is more valuable to you than your default. If no improvement exists, your money goes straight to your top cause.
Faking doesn't help. If you misstate your valuations, the optimizer routes your money to causes you don't actually like. You're the only one who suffers.
Failed payments are excluded. If your card doesn't clear, you're simply not in that round's pool. No one else is affected.
The British colonial government in India once offered a bounty for dead cobras. Locals started breeding cobras to collect the bounty. When the government cancelled the program, the breeders released their stock. More cobras than ever.
If you pay people to move money away from a cause, you incentivize people to put money into that cause to collect the payment.
You can never pay people to stop funding something. You can only give them a reason to move somewhere they like better.
Causeum avoids this because trades are justified entirely by the subsidy—the value of receiving money at a compromise cause you like. The optimizer doesn't reward you for leaving a hated cause; it rewards you for arriving at a valued one. A faker who claims an extreme cause as their default gains nothing—their trading partner's surplus comes from receiving the faker's money at the compromise, not from defunding the fake cause.
A future advanced mode may allow negative valuations, where the defunding component actively drives trades. Participants who opt in would accept the cobra risk for potentially larger surplus.
The optimizer only moves your money if the result is more valuable to you. If no improvement exists, your donation goes to your top cause unchanged.
Money is spent every round. No positions, no stock, no funds on deposit. There's nothing for a bad actor to manipulate.
Causeum takes a small share of the bilateral surplus it creates for you. No trade found? No fee. Your donation goes to your top cause unchanged.
All donations in a round are collected, optimized, and disbursed together. No partial fills.
Misstating your values means the optimizer sends your money somewhere you don't actually want. Same dynamics as any real market.
More donors means more ideological diversity, more common ground, and bigger gains. The optimizer finds improvements that grow with the square of the user count.
You donate monthly. You rate every cause. The optimizer pools everyone's donations and finds mutual compromises where donors on opposite sides both come out ahead. Your sacrifice is small—a slightly less preferred cause. Your gain is large—the other person's money also arrives at a cause you value. Causeum takes a small share of the surplus from each trade. No trade found? No fee. Donations to 501(c)(3) charities are tax-deductible.
Causeum routes donations to designated political organizations (527) and tax-exempt charities (501(c)(3)). Donations to 501(c)(3) organizations may be tax-deductible.
20 donors, 6 causes, one optimization round. Real numbers.
Six causes: Emily's List (EL), Club for Growth (CFG), VoteVets (VV), Americans for Prosperity (AFP), League of Conservation Voters (LCV), Against Malaria Foundation (AMF). Twenty donors, each with a monthly donation and stated valuations. If they donated directly, each person's money would go to their top cause.
| Donor | Default | $/mo | EL | CFG | VV | AFP | LCV | AMF |
|---|---|---|---|---|---|---|---|---|
| Grace | EL | $350 | 1.00 | 0.08 | 0.85 | 0.05 | 0.80 | 0.40 |
| Alice | LCV | $500 | 0.60 | 0.05 | 0.78 | 0.04 | 1.00 | 0.30 |
| Nora | VV | $400 | 0.82 | 0.06 | 1.00 | 0.05 | 0.76 | 0.35 |
| Eve | EL | $250 | 0.75 | 0.04 | 0.80 | 0.03 | 0.90 | 0.25 |
| Rosa | LCV | $500 | 0.70 | 0.03 | 0.68 | 0.03 | 1.00 | 0.50 |
| Quinn | VV | $150 | 0.85 | 0.06 | 0.72 | 0.05 | 0.70 | 0.20 |
| Kim | VV | $200 | 0.80 | 0.07 | 1.00 | 0.06 | 0.82 | 0.30 |
| Sam | CFG | $300 | 0.08 | 0.90 | 0.70 | 0.75 | 0.10 | 0.15 |
| Frank | CFG | $600 | 0.05 | 1.00 | 0.06 | 0.85 | 0.04 | 0.10 |
| Jake | AFP | $500 | 0.06 | 0.82 | 0.08 | 1.00 | 0.05 | 0.12 |
| Dan | AFP | $400 | 0.07 | 0.80 | 0.09 | 0.95 | 0.06 | 0.15 |
| Hank | CFG | $450 | 0.04 | 1.00 | 0.07 | 0.72 | 0.03 | 0.08 |
| Pat | AFP | $200 | 0.05 | 0.70 | 0.10 | 0.92 | 0.04 | 0.18 |
| Tom | CFG | $400 | 0.04 | 1.00 | 0.06 | 0.92 | 0.05 | 0.08 |
| Maya | AMF | $350 | 0.55 | 0.50 | 0.70 | 0.45 | 0.60 | 1.00 |
| Leo | AMF | $300 | 0.65 | 0.40 | 0.75 | 0.35 | 0.70 | 0.80 |
| Ivy | VV | $250 | 0.50 | 0.45 | 1.00 | 0.40 | 0.55 | 0.85 |
| Omar | AMF | $350 | 0.30 | 0.65 | 0.55 | 0.60 | 0.25 | 0.90 |
| Zara | LCV | $300 | 0.60 | 0.15 | 0.65 | 0.10 | 1.00 | 0.45 |
| Ben | VV | $200 | 0.20 | 0.55 | 0.80 | 0.60 | 0.15 | 0.40 |
If everyone donated directly: EL $600, CFG $1,750, VV $1,200, AFP $1,100, LCV $1,300, AMF $1,000.
The optimizer scans every pair of donors and every possible compromise cause. For each combination, it checks: does each side value receiving the other's money at the compromise cause more than their sacrifice? It executes the best trade, then scans again. Repeat until no more beneficial trades exist.
Grace and Sam find common ground at VoteVets. Grace's $350 goes to VV instead of EL. Sam's $300 goes to VV instead of CFG. Grace's math: sacrifice = $350 × (1.00 − 0.85) = $52.50. Subsidy gain = $300 × 0.85 = $255 (Sam's money arriving at VV). Net: +$202.50. Sam's math: sacrifice = $300 × (0.90 − 0.70) = $60. Subsidy gain = $350 × 0.70 = $245. Net: +$185. Both come out ahead.
Hank and Maya find common ground at Club for Growth. Maya's default is AMF ($1.00) and she values CFG at $0.50—a big sacrifice of $175. But she receives Hank's $450 at CFG ($0.50 to her) = $225 of subsidy gain. Net: +$50. Hank sacrifices $450 × (1.00 − 1.00) = $0 (he's already in CFG) and receives Maya's $350 at CFG ($1.00 to him) = $350. Net: +$350.
With 50,000 donors, 12 causes, and realistic partisan preferences, using the subsidy-only trade criterion:
89% of donors traded. 89% of all money redirected. The DNC collapsed from $9.3M to $945K. The RNC from $9.3M to $931K. VoteVets surged from $3.5M to $22.8M. Conservation Voters grew to $5.9M. AFP to $4.6M. Against Malaria went from $0 to $608K. GiveWell from $0 to $587K.
$41.3 million in subsidy surplus from a $46.1 million pool ($0.90 per dollar donated). That's real willingness-to-pay—each dollar of surplus represents the value a trader placed on receiving their partner's money at the compromise cause, minus their own sacrifice.
Average trader surplus: $831 on a $927 donation (90%). Left-leaning: 99% traded. Right-leaning: 100%. Centrists: 100%.
The optimizer searches for mutually beneficial trades—pairs or groups where all participants gain. It can be implemented as a linear program or bilateral matching algorithm. Runs on every donation cycle. With 50 causes and 10,000 donors, solvable in milliseconds.
Multi-way matching. The optimizer finds n-way cycles (Shapley-Scarf Top Trading Cycles) where no pairwise trade exists. A→B, B→C, C→A—all atomic.
Honest pricing is approximately optimal. Misstating your values sends your money where you don't want it. Same dynamics as any pay-as-bid market.
Tax deductions. Each donor receives a receipt showing exactly which causes received their money and how much. Donations to 501(c)(3) organizations are tax-deductible in the normal way.