Household Finance DemoDThe accountability layer for the research pipeline: what the targets rest on, what would change them, and a pre-committed record of each decision — written before outcomes are known, scored after.
The Carter household's investable assets stay pointed at 45% US equity, 20% international equity, 25% bonds, and 10% cash until the down-payment window opens.
Revisit if the house timeline moves inside 18 months, if either earner's job changes, or if reserves fall below the $24,000 floor.
Bank cash plus the money-market reserve stays between the $24,000 floor and the $42,000 ceiling before any contribution goes to equities.
Revisit if monthly essentials move by more than 10% for two straight months, or if the childcare step-down arrives early.
Direct new savings to US and international equity until the mix is within 5 points of target; sell nothing this quarter.
Why: The drift is contribution-fixable at the demo savings pace, and selling would realize gains for no benefit.
What would prove it wrong: If two quarters of contributions cannot close half the US-equity gap, the no-sell rule gets revisited.
Trim the Treasury reserve halfway to target now and the rest after the January review.
Why: Keeps reserves above the floor through the enrollment and tax checkpoints.
What would prove it wrong: If reserves would dip below $24,000 at either stage, the second stage is cancelled.
Leave the Family HSA invested in the Carter Treasury Reserve rather than moving it into the broad market index.
Why: Medical spending in the demo year ran close to the HSA balance, so the reserve's stability mattered more than the index's expected return.
What would prove it wrong: A year of low medical spending with the HSA balance above $10,000 would have argued for investing the excess.
Outcome: Kept in the reserve fund; demo medical spending stayed inside the card-paid budget.
Forecasts extracted from each source, resolved quarterly. Brier score: lower is better calibrated (0.25 = coin flip). Weights move slowly — a source needs ~30 scored forecasts before its weight means much.
| Source | Forecasts scored | Brier | Weight |
|---|---|---|---|
| Fictional market notes | 12 | 0.21 | 0.6 |
| Synthetic rates desk memo | 8 | 0.24 | 0.4 |
| Carter demo research digest | 3 | — | — |