Methodology — The Plumbline

The Series

Fixed and declared — tiles never appear only when they look alarming

Tile	Series / exact ID	Source	Cadence	Strain side	Provenance	Notes
F6	10-year term premium (Adrian–Crump–Moench) `ACMTP10`	NYFED	daily	up	primary official	Model-based estimate (ACM), not a market quote; disclosed on Methodology.
S2b	Average hourly earnings, production & nonsupervisory (long history) `AHETPI`	FRED	monthly	none	primary official	BLS-produced; subject to the integrity-events DRS factor.
F10	Treasury auction demand (coupon bid-to-cover) `AUCTION_BTC`	TREASURY	weekly	down	primary official	Bid-to-cover of completed Note and Bond auctions, averaged per auction date. Announced-but-unheld auctions are excluded.
F13	Baa corporate spread over 10-year Treasury `BAA10Y`	FRED	daily	up	primary official	Complements F1: the high-yield spread (truncated to ~3 years by ICE licensing) reads speculative credit; this reads investment grade with four decades of memory.
D1	High-propensity business applications `BAHBATOTALSAUS`	FRED	monthly	down	primary official	Census Business Formation Statistics, monthly since 2004. High-propensity = applications with characteristics of future employer businesses.
F1	High-yield credit spread (ICE BofA US High Yield OAS) `BAMLH0A0HYM2`	FRED	daily	up	primary official	FRED redistributes only the trailing ~3 years of this ICE series. Z-scores use the available window; disclosed on the Methodology page.
C13	Bankruptcy filings (US Courts) `BANKRUPTCY_FILINGS`	MANUAL	quarterly	up	primary official	Declared, not yet shipped: no FRED series exists (verified 2026-06-06); the primary is the US Courts quarterly F-2 tables, which enter via the manual regime when wired. [declared, not currently shipped]
M2	Bitcoin (Coinbase spot) `CBBTCUSD`	FRED	daily	up	primary official	The froth thermometer: the strain side is declared UP — this tile reads speculative altitude, not market health. A crash shows up in the fast tells; the climb shows up here.
S2a	Average hourly earnings, all employees (modern wage headline) `CES0500000003`	FRED	monthly	none	primary official	Starts 2006-03. AHETPI carries the history back to 1964; splice disclosed. BLS-produced.
C2b	CPI-U index, not seasonally adjusted (official record to 1913) `CPIAUCNS`	FRED	monthly	none	primary official	BLS-produced. Splices with the Shiller reconstruction (ARCHIVE) at 1913 on the long-run timeline.
C2a	CPI-U index level (raw input to C2) `CPIAUCSL`	FRED	monthly	none	primary official	BLS-produced; subject to the integrity-events DRS factor. Reconciled against the BLS release (CUSR0000SA0) on every fetch.
C2	CPI inflation, year over year `CPI_YOY`	DERIVED	monthly	both	primary official	Computed: 100 × (CPIAUCSL(t) / CPIAUCSL(t−12m) − 1). BLS-produced input.
C4	Federal debt held by the public `DEBT_HELD_PUBLIC`	TREASURY	daily	none	primary official	Daily since 1997; the longer official record enters via ARCHIVE (build phase 7).
S11	Share of net worth held by the Baby Boomer generation `DFA_BOOMER_SHARE`	FEDDFA	quarterly	none	primary official	Fed DFA generation breakdown — published only on the Fed's own site, not FRED (verified). Context, no state: generational shares move with demography, and demography is not strain.
S12	Share of net worth held by the Millennial generation `DFA_MILLENNIAL_SHARE`	FEDDFA	quarterly	none	primary official	The same release shows this generation carrying ~42% of home-mortgage debt against ~11% of net worth — the ledger's two sides, shown without comment.
F4	10-year Treasury constant-maturity yield `DGS10`	FRED	daily	up	primary official
F5	30-year Treasury constant-maturity yield `DGS30`	FRED	daily	up	primary official
C12	Single-family mortgage delinquency rate `DRSFRMACBS`	FRED	quarterly	up	primary official	Commercial-bank book, quarterly since 1991. The transmission tile between the affordability story (section 02) and the credit system.
F12	Trade-weighted dollar (nominal broad index) `DTWEXBGS`	FRED	daily	both	primary official	Two-sided by declaration: a falling dollar alongside falling long-end demand is the loss-of-confidence signature; a spiking dollar signals global stress and flight to quality. The partner tile to auction demand (F10).
M1	Corporate equities to GDP (the Buffett indicator) `EQUITY_GDP`	DERIVED	quarterly	up	primary official	Computed: nonfinancial corporate equities outstanding (Fed Z.1, NCBEILQ027S) over nominal GDP, percent. Both inputs official, quarterly since 1947. Trend-dominated: a steadily trending series always prints high-z eventually; the state reads how far along the trend, vs the recent decade. Disclosed.
M4	Top-10 share of S&P 500 market cap `EQUITY_TOP10_SHARE`	MANUAL	monthly	up	primary official	Declared, not yet shipped: the AI-concentration number, from the public S&P DJI factsheet via the manual regime. Needs a sourced first value. [declared, not currently shipped]
S3	Financial Obligations Ratio (discontinued 2023-Q3) `FODSP`	FRED	quarterly	none	primary official	Discontinued by the Fed at 2023-Q3. Kept for historical overlap; the Cost-to-Exist spine is rebuilt from components, anchored to TDSP. [declared, not currently shipped]
I6	Regular gasoline price, US average (EIA, weekly) `GASREGW`	FRED	weekly	none	primary official	Multiplied by the sourced EIA household-gallons anchor (data/manual/fuel.json) — the one basket line that moves with the market.
X4	Nominal GDP (input to M1) `GDP`	FRED	quarterly	none	primary official
C8	Federal debt to GDP `GFDEGDQ188S`	FRED	quarterly	up	primary official	Official OMB/FRED computation, quarterly since 1966. The longer annual view (1947+) lives on the Long View, derived from the Treasury archive and GDP. Scored against a generation (30y): for a valuation ratio the level IS the signal; full history would compare against the pre-globalization economy (foreign earnings, intangibles — disclosed).
S8	Income Gini ratio, families (Census) `GINIALLRF`	FRED	annual	up	primary official	Annual since 1947 — the income leg of the disparity picture, beside the wealth legs (S6/S7, Fed DFA). Trend-dominated: rose steadily 1970-2010 and plateaued high; the state reads vs the recent decade, disclosed.
D4	Housing starts `HOUST`	FRED	monthly	down	primary official	Decade-relative; not population-scaled (a declared simplification — starts per household is a future refinement). The supply-side answer to the shelter line in section 02.
F9	Initial unemployment insurance claims `ICSA`	FRED	weekly	up	primary official
C10	Quits rate (JOLTS) `JTSQUR`	FRED	monthly	down	primary official	The cleanest early read on labor softening: workers stop quitting before employers start firing. BLS-produced; subject to the integrity-events DRS factor.
D3	Prime-age labor force participation (25-54) `LNS11300060`	FRED	monthly	down	primary official	Decade-relative by design: the full-history baseline is dominated by the secular entry of women into the workforce (1948-1990s), which would make any modern reading look high. BLS-produced.
M3	FINRA margin debt `MARGIN_DEBT`	MANUAL	monthly	up	primary official	Declared, not yet shipped: FINRA publishes monthly but blocks automated fetch; enters via the manual regime when wired. [declared, not currently shipped]
I1	Real median household income `MEHOINUSA672N`	FRED	annual	none	primary official
S1	Misery Index (unemployment + CPI inflation) `MISERY`	DERIVED	monthly	up	primary official	Computed: UNRATE + CPI YoY. Shown as the established cousin beside the Sentiment Gap (spec 5.3).
S5a	30-year fixed mortgage average `MORTGAGE30US`	FRED	weekly	none	primary official
S5b	Median sales price of houses sold `MSPUS`	FRED	quarterly	none	primary official
X3	Corporate equities outstanding, Fed Z.1 (input to M1) `NCBEILQ027S`	FRED	quarterly	none	primary official
C5	Net interest / federal receipts (the ratio's spine) `NET_INTEREST_SHARE`	TREASURY	monthly	up	primary official	Trailing-12-month interest expense on public issues over trailing-12-month total receipts (MTS Table 4) — receipts are seasonal, so a monthly ratio would whipsaw. This is the present-tense spine of the Crisis Ratio; the conditional cascade lives at crisisratio.com.
F11	Chicago Fed National Financial Conditions Index `NFCI`	FRED	weekly	up	primary official	A published, externally-maintained Fed index of financial conditions — not a Plumbline-authored composite, any more than CPI is. Positive = tighter than average. Weekly since 1971.
M5	NFCI leverage subindex `NFCILEVERAGE`	FRED	weekly	up	primary official	The Chicago Fed's leverage component: how much borrowed money stands under asset prices. The altitude tiles read the climb; this reads what is propping it.
X5	Output per hour, nonfarm business (input to D2) `OPHNFB`	FRED	quarterly	none	primary official
X2	Total nonfarm payrolls (relationship-engine input) `PAYEMS`	FRED	monthly	none	primary official	BLS-produced; subject to the integrity-events DRS factor.
C3a	Core PCE price index level (raw input to C3) `PCEPILFE`	FRED	monthly	none	primary official
C3	Core PCE inflation, year over year `PCE_YOY`	DERIVED	monthly	both	primary official	Computed: 100 × (PCEPILFE(t) / PCEPILFE(t−12m) − 1).
D2	Labor productivity growth, year over year `PRODUCTIVITY_YOY`	DERIVED	quarterly	down	primary official	Computed: OPHNFB YoY (output per hour, nonfarm business). The long-run positive — everything else is downstream of this. BLS-produced.
S2	Real wage growth, year over year (the flow) `REAL_WAGE_GROWTH`	DERIVED	monthly	down	primary official	Computed: AHE (all employees) YoY − CPI YoY. The cousin of the Cumulative Real Wage Gap (spec 5.1), which asks how deep the hole still is.
C9	Sahm Rule recession indicator `SAHMREALTIME`	FRED	monthly	up	primary official	Not a forecast: a present-tense reading of a realized threshold (3-month average unemployment 0.50pp above its 12-month low). Built by Claudia Sahm explicitly as a non-predictive trigger; 0.50 is the declared line, shown on the tile.
X1	S&P 500 index (relationship-engine input) `SP500`	FRED	daily	none	primary official	FRED redistributes only the trailing ~10 years (S&P licensing). Relationship windows that need 2013-2014 cannot be computed and render honest blanks; disclosed on Methodology.
F7	10-year minus 2-year Treasury spread (2s10s) `T10Y2Y`	FRED	daily	down	primary official
F8	5-year, 5-year forward inflation expectation rate `T5YIFR`	FRED	daily	both	primary official
S4	Household debt service ratio `TDSP`	FRED	quarterly	up	primary official
C11	U-6 underemployment rate `U6RATE`	FRED	monthly	up	primary official	Adds involuntary part-time and marginally attached workers to the unemployment count. BLS-produced; subject to the integrity-events DRS factor.
C6	University of Michigan consumer sentiment `UMCSENT`	FRED	monthly	down	primary official
C1	Unemployment rate `UNRATE`	FRED	monthly	up	primary official	BLS-produced; subject to the integrity-events DRS factor.
F2	CBOE Volatility Index (VIX) `VIXCLS`	FRED	daily	up	primary official
S6	Share of net worth held by the bottom 50% of households `WFRBSB50215`	FRED	quarterly	down	primary official	Fed Distributional Financial Accounts, quarterly since 1989. Half the country holds ~2.5% of the wealth; the tile reads the direction. Trend-dominated: a steadily trending series always prints high-z eventually; the state reads how far along the trend, vs the recent decade. Disclosed.
S10	Share of net worth held by the 90th–99th percentiles `WFRBSN09161`	FRED	quarterly	none	primary official	Fed DFA, context only (no state) — strain in concentration is read at the edges (S6, S7); this band completes the arithmetic.
S9	Share of net worth held by the middle 40% (50th–90th) `WFRBSN40188`	FRED	quarterly	down	primary official	Fed DFA. Completes the four-band decomposition with S6 (bottom 50%), S10 (next 9%), and S7 (top 1%). Trend-dominated; decade-relative, disclosed.
S7	Share of net worth held by the top 1% of households `WFRBST01134`	FRED	quarterly	up	primary official	Fed Distributional Financial Accounts. Shown beside S6 — the K is the pair, not either line alone. Trend-dominated: a steadily trending series always prints high-z eventually; the state reads how far along the trend, vs the recent decade. Disclosed.

The Relationship Engine

One auditable method, four relationships

The rule: z < 1 BEHAVING · 1 ≤ z < 2 STRAINED · z ≥ 2 AND statistic > 0 AND both held for the declared persistence → DECOUPLED
Why these thresholds: Pre-committed and never adjusted to fit a narrative. The persistence and sign conditions are what separate a signal from a flinch: z must exceed 2 in the economically wrong direction for the declared number of consecutive readings before DECOUPLED is shown.
Minimum history: 60 observations of the statistic before any score is shown — below that, an honest blank.

R1 — Jobs vs. market reaction PLAUSIBLE

−corr over the last 24 Employment Situation releases between the payroll surprise (m/m change minus its trailing 12-month mean, in trailing-σ units) and the S&P 500 return on release day. Surprise is proxied against the series' own trend — revised data, not real-time consensus — which is why R1 is PLAUSIBLE, not ROBUST.

Good news should lift equities. When strong payrolls sell stocks off, the market is trading the Fed, not the economy. Persistence for DECOUPLED: 2 consecutive readings. Surprise is proxied against the series' own trend (revised data, not real-time consensus); disclosed on Methodology.

R2 — Stocks vs. bonds ROBUST

60-trading-day Pearson correlation of daily S&P 500 log returns and the Treasury return proxy (−Δ 10-year CMT yield). Positive correlation = the hedge failing.

Treasuries should hedge equities (negative correlation). When both sell off together, the classic portfolio stabilizer is failing. Persistence for DECOUPLED: 10 consecutive readings. Plain rolling correlation of daily moves; the computation is mechanical.

R3 — Inflation vs. expectations ROBUST

Rolling 24-month correlation of monthly changes in the 5y5y breakeven with monthly changes in published CPI YoY (CPI for month m aligned to m+1, when it is published). Positive = expectations moving WITH realized inflation — de-anchoring measured directly as co-movement. Reconstructed 2026-06-06; see changelog.

Long-run expectations should stay near 2% no matter what realized CPI does. When they start moving WITH CPI, the anchor is dragging. Persistence for DECOUPLED: 3 consecutive readings. Rolling 24-month correlation of monthly changes in the 5y5y breakeven with changes in published CPI — co-movement measured directly. Reconstructed 2026-06-06; see the Methodology changelog.

R4 — Growth vs. yields ROBUST

−corr over 26 weeks of the growth proxy (−Δ 4-week-average initial claims) and the weekly Δ 10-year yield. Positive = yields moving against growth.

Yields should track growth. When yields rise as growth weakens, the debt is being repriced on something other than the economy. Persistence for DECOUPLED: 4 consecutive readings. Initial claims serve as the weekly growth proxy; disclosed on Methodology.

Level tiles: z = (current − mean) / std over the series' DECLARED scoring window, gated by the declared stress direction (up: z, down: −z, both: |z|). Every state also carries its empirical percentile in plain language. Mapping: z_dys < 1 CALM · 1 ≤ z_dys < 2 WATCH · z_dys ≥ 2 EXTREME. Window: trailing years, minimum 24 observations.
Why the gate: One pre-committed rule for every tile. The stress-direction gate is what keeps the scale symmetric AND meaningful: an unusually low VIX reads CALM because the strain side is declared per series, in public, in the registry — not chosen after looking at the data.
The symmetry tests: All-calm: 2017 (sampled 2017-06-30 and 2017-11-30) must render majority-CALM with every scorable relationship BEHAVING. Build fails otherwise.
Gradient: Q4-2018 (2018-12-24) must render VIX and the 2s10s curve STRAINED while the labor tells stay calm; the Aug-2019 inversion (2019-08-27, curve genuinely negative) must read off-CALM at z ≥ 1.9. Thresholds were not bent to make 2019 dramatic — it renders WATCH at z 1.97, and the test documents that refusal.

The Derived Coefficients

Three, each beside an established cousin — a dozen flimsy indices would BE the crankery

The Cost-to-Exist Ratio (Claimed Share) — flagship

ClaimedShare = (Taxes + Obligations + Shelter + Food + Fuel + Health + RetirementGap) / Income

Line	Default amount	Confidence	Derivation
food	$7,484	ROBUST	USDA Thrifty Food Plan (April 2026), couple 20-50 with the two-person household adjustment — the SNAP-basis floor, a normative diet cost, not observed spending. Children add the 6-8-year-old rate in the editor.
fuel	$2,932	PLAUSIBLE	EIA household-gallons anchor (~681 gal/yr, derived from EIA's published 2022 expenditure and price) × the LIVE weekly regular gas price ($4.305/gal as of 2026-06-01). The one line that moves with the market.
health	$6,850	ROBUST	KFF 2025 mean worker premium contribution, family tier. Out-of-pocket line DISABLED pending a verifiable current median (conservatively low).
obligations	$9,481	PLAUSIBLE	Debt service at the Fed's aggregate DSR (11.323096% of disposable income, 2025-10-01), applied to median income. Two disclosed mismatches: an aggregate share applied to one household, and the aggregate includes homeowner MORTGAGE service while this basket's household is a RENTER (shelter enters as FMR rent) — for that household the line overstates debt service, which errs high here and is stated rather than netted.
retirement gap	$6,271	CONTESTED	Saving to close the SSA medium-earner replacement rate (42.6%) to a 70% adequacy target under a disclosed level-annuity model. The target is a judgment; the slider is yours.
shelter	$11,700	ROBUST	HUD Fair Market Rent, 2BR, national unweighted median — the decency floor, not metro comfort.
taxes	$12,093	ROBUST	Federal income tax (MFJ, standard deduction, no credits) + employee FICA 7.65%. State/local omitted, disclosed.

Decency floor, not comfort: shelter defaults to the HUD FMR national unweighted median. Conservative omissions throughout (state/local tax, credits, out-of-pocket health) — where honesty is ambiguous, the share reads LOW. Median income: $83,730 (Census via FRED, 2024-01-01). Every default is adjustable in the basket editor on the board.

Pedigree — the credentialed neighbors

The components and the inverted form of this arithmetic are long established; the ratio presentation with live data and an adjustable assembly is this project's contribution. The lineage:

Federal Reserve Financial Obligations Ratio / Debt Service Ratio	The official spine: required payments over disposable income, aggregate, since 1980. This ratio extends it with the privatized life taxes it omits.
MIT Living Wage Calculator (Glasmeier)	The same basket arithmetic inverted — solves for the wage at which the claimed share reaches 100%, by county and family type, since 2004.
The Self-Sufficiency Standard (Pearce)	Income needed to meet basic needs without assistance, by family type and county, since 1996. An income below the standard is this basket's over-100% reading.
EPI Family Budget Calculator	The basket as a dollar budget by metro and family type — the budget form of the same instinct.
United Way ALICE (Household Survival Budget)	Counts households whose income falls below the survival basket. A household reading over 100% here is an ALICE household.
Residual-income / shelter-poverty method (Stone)	The academic ancestor of the underwater reading: affordability as what remains after necessities, explicitly allowing negative residuals.

The Cumulative Real Wage Gap

RealWageIndex(t) = (AHE(t)/CPI(t)) / (AHE(base)/CPI(base)) × 100

The computation is mechanical; the base period is a disclosed assumption, which is why you can change it. Two wage series, shown separately and never spliced: all-employees AHE begins 2006-03; production-and-nonsupervisory AHE carries the history to 1964. They differ in level and composition.

The Sentiment Gap

OLS: sentiment ~ const + unemployment + cpi_yoy + real_wage_growth, monthly, full available joint sample; Newey-West (HAC, 12 lags) standard errors

Sample: 1965-02-01 to 2026-04-01 (631 months)
R²: 0.2769 — the fundamentals never fully explained the mood; that is the finding, not a flaw to hide
Coefficients (std. err.): const: 107.1739 (5.4807) · cpi_yoy: -2.4403 (0.5521) · real_wage_growth: -0.745 (0.9739) · unemployment: -2.3268 (0.8969)
Stated limitations: The regression specification is a choice; it is published in full so it can be challenged. The residual is not proof of any particular cause — the Cost-to-Exist story is ONE candidate explanation; documented partisan-composition shifts in the Michigan survey and omitted variables (rates, wealth, gas prices) are others. The residual is also highly persistent, so the gap's size carries less statistical certainty than a naive reading suggests — standard errors are Newey-West for this reason. Real wage growth uses production & nonsupervisory AHE (AHETPI, 1964+) so the sample reaches 1978; disclosed.

The confidence scheme

CONTESTED — The construction involves a judgment (an adequacy target, a basket boundary, a regression specification) that is genuinely arguable — published in full precisely so it can be argued.

PLAUSIBLE — The computation is sound but rests on a disclosed assumption (a base period, a proxy, an assembly choice) that a reasonable person could set differently.

ROBUST — The computation is mechanical on primary official series; reasonable people do not dispute the arithmetic.

Methodology Changelog

Changes are made openly, dated, with the reason — never quietly

The Plumbline is in alpha. Until 1.0, methodology changes are summarized daily; from 1.0 onward, every change to a formula, threshold, or window gets its own dated entry with full rationale, and the granular history lives in the public git log either way.

2026-06-06

Launch Day.

Why: Built, externally reviewed, and calibrated in the open, all in one day. Every step is in the public git log.

The Data Reliability Score

Distinct from the confidence label — never conflated

Factors

Provenance (baseline): primary official > official-reconstructed > scholarly > proxy.

Era discount (archive data): era B −10, era C −20 — coverage, method stability, and revision behavior degrade together in deep history.

Institutional integrity: small, capped, cumulative, reversible deductions keyed only on documented sourced events (full table below). Measures elevated uncertainty, never manipulation.

Provenance baselines

official reconstructed: 75 · primary official: 95 · proxy: 60 · scholarly: 65

Era discounts

Era A (1948 to present, −0): Modern primary: official, monthly, fully instrumented. Every coefficient is valid here.

Era B (1913 to 1947, −10): Mid-century and interwar: the spine exists at annual resolution, official or near-official. Consumer sentiment and the modern cost composites do not exist here.

Era C (pre-1913, −20): Deep history: scholarly reconstructions only. Real and usable, but a different confidence class; some coefficients are anachronistic.

Anachronism floors

Cost-to-Exist Ratio — floor: mid-century. Employer health coverage and the Social Security replacement framing did not exist; the number cannot be computed honestly in 1900, so it is not rendered there.

Sentiment Gap — floor: 1978 (continuous monthly Michigan sentiment). There is no honest sentiment series to regress before the survey exists.

Cumulative Real Wage Gap — floor: 1964 (long series) / 2006 (modern headline). AHE begins 1964 (production & nonsupervisory) and 2006 (all employees); the two are shown separately, never spliced.

The institutional-integrity factor — the full public table

Keys only on documented, specific, sourced events — never on partisanship, never on whether anyone dislikes the numbers. Small, capped (−8 max per agency), symmetric across administrations, and it reverses when conditions normalize. It measures elevated uncertainty and reduced precision, not manipulation. Challenge any entry by merge request with a primary source.

Date	Agency	Event	Deduction	Lifts when	Source
2025-08-01	BLS	Senate-confirmed commissioner removed mid-term, hours after routine downward payroll revisions, on an unevidenced claim of manipulation; former commissioners across administrations, including a Trump appointee, called the removal groundless. Agency has operated under an acting commissioner since, with a nominee in confirmation as of June 2026.	−3	A commissioner is Senate-confirmed and assumes office.	source
2025-09-01	BLS	Documented curtailment of CPI and PPI data collection, citing resources: collection suspended entirely in Buffalo NY, Lincoln NE, and Provo UT; roughly 15 percent of the CPI sample suspended in some areas; PPI categories cut.	−3	Collection is restored to the designed sample.	source
2025-09-15	BLS	Department of Labor Inspector General opened a review of CPI, PPI, and employment data collection.	−1	The review closes without adverse findings (an adverse finding would enter as its own sourced event).	source
2025-10-01	BLS	Statistical advisory committees disbanded; FY2026 budget proposes further cuts that would materially affect collection.	−1	Advisory bodies are reconstituted or funding adequacy is restored.	source

Counterweight, applied honestly: Applied honestly: there is no evidence of manipulation. BLS has continued to publish unfavorable readings (job losses in some months, an inflation jump tied to the Iran war), which indicates the core function persists, and funding strain predates this administration and spans both parties.