HOW TO READ ECO I
Three readings
| Reading | It answers | Horizon | Updated | Made from |
|---|---|---|---|---|
| Risk stage, 1 to 7 | How fragile is the system? Would a shock spread? | Months to years | With each official release | Rules from research, replayed since 1990 |
| Market stress | How strained are markets now, and is it getting worse? | Today, one or two days late | Daily | Five measured mechanisms against their own history since 2000 |
| Second opinion, 0 to 10 | What does an independent reader of the news and the data conclude? | Days | Every weekday | An independent reading; its claims are scored publicly |
Risk stage
- 1 Calm: nothing is on.
- 2 Stretched: stocks unusually high, a credit-fuelled price boom, or lenders asking unusually little for risk.
- 3 Watch, 4 Early warning: one, or two or more, early warnings like an inverted yield curve or rising layoffs.
- 5 Cracking, 6 Breaking, 7 Crisis: one, two or three of markets, banks and jobs failing.
The highest layer on sets the stage. Bank and payroll confirmations were added after seeing 2023, the credit condition in October 2026. Missing data never reassures.
Tested since 1990
Months followed by each event within 24 months, with release delays:
| Reading | Recession | S&P 500 fell 20% | OFR stress above 3 |
|---|---|---|---|
| Stage 1–2 | 14% | 10% | 21% |
| Stage 3–4 | 34% | 23% | 37% |
| Stage 5–7 | 15% | 39% | 24% |
| All months | 21% | 21% | 28% |
| Yield-curve rule, when on | 49% | 51% | 44% |
Events warned of; ranking skill (0.5 is chance):
| Warning | Recession | S&P 500 fell 20% | OFR stress above 3 |
|---|---|---|---|
| Stage 3 or higher | 3/3, 0.51 | 3/3, 0.71 | 3/3, 0.50 |
| New York Fed curve, 30%+ | 3/3, 0.87 | 3/3, 0.76 | 2/3, 0.67 |
Stage 3 or higher warned of every event but was on in 64% of months, and 68% to 76% of its warnings were false. For recessions and stress episodes it ranks months no better than chance; the yield curve does far better. Few events; revised data flatter replays.
Market stress (tested)
OFR index parts and Treasury yields; cut-offs from 2000–2012.
| Level | When |
|---|---|
| Normal | Fewer than two above their 90th percentile |
| Guarded | Two or more above the 90th, two days running |
| Elevated | Three above the 95th, or Guarded with funding or credit above the 95th |
| Acute | Funding or credit above the 99th and one more above the 95th, or four above the 95th |
Levels fall only after five days, five points lower. Replayed on first-published data; live since September 2026. Tested on U.S. stock-market falls since 2013, each counted once:
| Fall | Falls | Warned before it began | False alerts a year |
|---|---|---|---|
| 5% within a week | 19 | 2 | 2.5 |
| 10% within a month | 8 | 2 | 2.3 |
| 20% within three months | 3 | 0 | 2.7 |
It mostly warns once a fall is under way: it describes strain, it does not forecast a fall.
Sources
- Fed · Equity-to-net-worth
- OFR · Financial Stress Index
- Fed · SLOOS
- BLS · Payrolls
- FRED · Sahm rule
- BLS · 2025 survey gap
- New York Fed · Yield curve
- Gilchrist & Zakrajšek (2012)
- Greenwood et al. (2022)
- BIS · Total credit
The New York Fed reference rate data is subject to the Terms of Use posted at newyorkfed.org. The New York Fed is not responsible for publication of the reference rate data by Economic Risk Clock, does not sanction or endorse any particular republication, and has no liability for your use. The Secured Overnight Financing Rate (SOFR) Data and Broad General Collateral Rate (BGCR) Data are calculated using data provided under a license granted to the New York Fed by DTCC Solutions LLC (“Solutions”), an affiliate of The Depository Trust & Clearing Corporation. Solutions, its affiliates, and third parties from which they obtained data have no liability for the content of this material.