Socialism: give the bank a working legal basis to ask loan applicants' race and sex, since the federal small-business data rule won't reach it, with a separate custodian, deletion and small-number limits

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01QWPs69B8nQZjtPEUXMKzZx

JoshHeaps.Net/Content/Socialism/open-problems.md

@@ -29,7 +29,7 @@ An honest list of what the model hasn't solved yet, most dangerous first.
 
 10. **The pilot's own weak points.** [The pilot](pilot-spec.md) puts the burden of proof on the co-ops, which means roughly a 1-in-10 chance of shutting down a pilot that was actually working. It tests individual co-ops, not the [federation](federation.md) support network Mondragon relied on. Its lotteries test help after the loan and, since the pre-launch lottery was added, help before it ([Help Before the First Loan](before-launch.md)), but only among groups that come forward, not outreach to workers who never think of starting a co-op. The pre-launch help now has a cost ceiling set in advance, $400,000 of total public cost per extra co-op still open at year four, but there's no good benchmark for it, and the lottery is small enough that "not settled" is a likely verdict ([The cost ceiling](before-launch.md#the-cost-ceiling)). Founders now put in a small stake they can pay in unpaid work, but founders who can't afford months of unpaid work still face a barrier. Its comparison group (small businesses with government-backed loans) is biased in directions that partly cancel out, by amounts nobody knows. And market-rate lending based on collateral may mostly reach founders who already have some wealth. Approval rates by race, gender and wealth are published so that shows up, but publishing a gap doesn't close it.
 
-14. **Drift toward the safest co-ops.** *(Partly answered.)* A bank that lends only to co-ops can still lend mostly to the safe ones, like consultancies started by people with savings, as America's co-op bank did. [Who the Bank Lends To](who-gets-loans.md) sets aside $90 million of the loan money for co-ops in lower-paid work, which goes back to the general pool only if the reviewers certify the bank handled those applications fairly. Results are reported for that group and for first-time owners without much wealth, and the verdict has to say which kinds of co-ops it covers. Still open: the set-aside makes money available but can't make groups apply; it can be used mostly on work just under the average wage; first-time owners without wealth get reporting, not money; and in work whose prices are set by Medicaid or contractors, a failure may say more about the price than about co-ops.
+14. **Drift toward the safest co-ops.** *(Partly answered.)* A bank that lends only to co-ops can still lend mostly to the safe ones, like consultancies started by people with savings, as America's co-op bank did. [Who the Bank Lends To](who-gets-loans.md) sets aside $90 million of the loan money for co-ops in lower-paid work, which goes back to the general pool only if the reviewers certify the bank handled those applications fairly. Results are reported for that group and for first-time owners without much wealth, and the verdict has to say which kinds of co-ops it covers. Still open: the set-aside makes money available but can't make groups apply; it can be used mostly on work just under the average wage; first-time owners without wealth get reporting, not money; and in work whose prices are set by Medicaid or contractors, a failure may say more about the price than about co-ops. Even measuring who gets loans needs its own rule: lending rules bar asking applicants' race and sex, and the federal exception for small-business lenders won't reach a bank this small, so the bank's law has to require the question, and a later Congress could drop it ([Asking needs its own legal basis](who-gets-loans.md#treated-the-same)).
 
 15. **Taxes.** *(Partly answered.)* A locked co-op is a corporation that pays tax once on members' profit shares, under the part of the tax code written for co-ops. [How a Locked Co-op Is Taxed](taxes.md) fits the model to it: profit is shared per hour worked, workers who aren't members get a share too, wages reach the going rate before any profit is shared, and charges come out of the current year's share before older, already-taxed balances. Still open: nobody has ruled that a co-op with an outside lock-holder and a reserve that goes to the federation counts as a co-op for tax purposes; whether profit shares owe payroll tax is unsettled; the ramp no longer shields senior members from new hires unless it's paid as wages; workers without work authorization can't be members; the kept reserve pays 21% even though nobody owns it; following the deferred-pay rules for lower-income members' untaxed credits could be read as conceding that the credits are pay; and a member who loses their account when a co-op fails can probably deduct only $3,000 a year of it, after being taxed on it in full, until Congress refunds that tax.
 

JoshHeaps.Net/Content/Socialism/who-gets-loans.md

@@ -78,7 +78,21 @@ The yearly audit of rejected applications ([Making It Last](durability.md#riggin
 - **Are the numbers themselves fair?** It checks each part of the bank's scoring, like collateral, credit history and founders' own money, to see whether it counts against a group more than it predicts repayment. Then it publishes whether a different scoring would predict repayment as well with less of that effect.
 - **Is the bar too high?** If approved co-ops in one group repay better than the rest, the bank was demanding too much of that group. The lottery among borderline applicants ([The Pilot, in Numbers](pilot-spec.md#who-co-ops-are-compared-against)) is the cleanest place to check.
 
-The audit also reports by the founders' gender and race, as the pilot already does, collected separately from the loan file so underwriters don't see it. Answering is optional and changes nothing about the loan. Lending law generally bars a lender from asking about race or sex, with exceptions for data collection that federal rules require or allow, with the same kind of wall between the answers and the loan decision ([12 C.F.R. § 1002.5(a)](https://www.law.cornell.edu/cfr/text/12/1002.5)). The bank's law names which exception it relies on, and if none applies, the independent reviewers ask after the decision is made. Race and sex measure results; they never decide a loan. If the audit finds a scoring rule that works against a group, a replacement is adopted only if it predicts repayment as well or better for all applicants.
+The audit also reports by the founders' gender and race, as the pilot already does, collected separately from the loan file so underwriters don't see it. Answering is optional and changes nothing about the loan.
+
+**Asking needs its own legal basis.** The fair-lending law forbids discriminating, not asking ([15 U.S.C. § 1691](https://www.law.cornell.edu/uscode/text/15/1691)). The ban on asking an applicant's race, color, religion, national origin or sex is in the rules written under it, with a few exceptions ([12 C.F.R. § 1002.5](https://www.law.cornell.edu/cfr/text/12/1002.5)). The law covers government lenders too ([§ 1691a](https://www.law.cornell.edu/uscode/text/15/1691a)), so the bank is bound by both. None of the ready-made exceptions fits it:
+
+- **The federal small-business lending data rule** requires lenders to ask, behind a wall that keeps the answers from anyone deciding on the loan ([§ 1002.108](https://www.ecfr.gov/current/title-12/part-1002/section-1002.108)). Since a May 2026 change, it covers only lenders that make at least 1,000 small-business loans in each of two years, starting in 2028 ([§ 1002.105(b)](https://www.ecfr.gov/current/title-12/part-1002/section-1002.105), [§ 1002.114](https://www.ecfr.gov/current/title-12/part-1002/section-1002.114)). The pilot's bank makes far fewer than 1,000 loans a year. Collecting early is allowed only for lenders that expect to be covered, and continuing is allowed only for lenders that used to report. The rule is also being reconsidered again.
+- **A lender's own fair-lending self-test** can ask, if it tells applicants that answering is optional, why it's asking, and that the law forbids using the answer or a refusal against them ([§ 1002.5(b)(1)](https://www.law.cornell.edu/cfr/text/12/1002.5), [§ 1002.15](https://www.ecfr.gov/current/title-12/part-1002/section-1002.15)). But collection the government requires doesn't count as a voluntary self-test, and the pilot requires these figures.
+- **Asking later, through the reviewers,** doesn't escape the rule: it covers "every aspect of an applicant's dealings with a creditor" about a loan ([§ 1002.2(m)](https://www.ecfr.gov/current/title-12/part-1002/section-1002.2)). And applicants who were turned down rarely answer a later survey, which would lose the very people the audit is for.
+
+So the bank's law uses the route the rules already leave open: information "required by a regulation" to monitor compliance ([§ 1002.5(a)(2)](https://www.law.cornell.edu/cfr/text/12/1002.5)). The law directs the bank's federal regulator to issue that rule, using the categories the small-business data rule uses (ethnicity, race and sex of each principal owner), so the results can be compared with national figures if those are ever collected. The rule copies the self-test notice to applicants and goes further than the data rule's wall. That rule lets a lender skip the wall, with notice, when keeping it isn't feasible, and a bank with a few loan officers would. So the bank never holds the answers at all: a separate custodian collects them under coded IDs and matches them to loan files only after decisions are final. As a fallback, the law authorizes the question directly, "notwithstanding" the rules. A deal with a single agency wouldn't do: it would last only as long as that agency's goodwill. The bank asks nothing about birthplace or language.
+
+- **Kept apart, kept briefly.** They're used only for the audit and the pilot's published results, never shared with law enforcement or immigration authorities, and kept out of public-records requests. Each answer is deleted three years after the audit that used it, the data rule's own retention period ([§ 1002.111](https://www.ecfr.gov/current/title-12/part-1002/section-1002.111)). Results for fewer than ten people aren't published, and refusal rates are.
+- **Publishing has a price, paid on purpose.** Published results lose the protection the self-test rule gives a lender's own findings, and anyone suing the bank can use them. That's the point of an audit the public can check. The test a replacement scoring rule has to pass is published in advance, so turning one down is a rule being followed, not a judgment call.
+- **If Congress won't include it,** estimates from surname and address, the method the Consumer Financial Protection Bureau uses, can stand in for totals. They say nothing about sex, work poorly for small groups and for co-ops with several founders, and can't be checked by the people they describe. A better fallback is matching founders, with their consent, to their own answers on the census, inside the Census Bureau's secure research program, so the bank never holds a race file. The reviewers compute the surname-and-address estimates from the first year anyway, so if a later Congress repeals the question, there's a comparable series to continue.
+
+Race and sex measure results; they never decide a loan. If the audit finds a scoring rule that works against a group, a replacement is adopted only if it predicts repayment as well or better for all applicants. That's the bank's own rule, not a claim about what the courts require, so it holds whatever happens to the law on unintended discrimination. Gaps by race are reported both as they are and after taking founders' wealth into account, so it shows how much of any gap runs through wealth and how much doesn't.
 
 ## What this doesn't solve