The Dashboard

Financial stress is the distance between normal and panic. In a healthy economy, banks lend freely, companies borrow cheaply, volatility stays low, and the labor market absorbs workers faster than it sheds them. Stress is what happens when any of those assumptions breaks. A bank refuses to roll over a loan. A bond market freezes. Investors dump risky assets for Treasuries. Employers start filing layoff notices. Each of these actions leaves a footprint in the data — a spread widening, a volatility spike, a claims surge — and if you know where to look, you can see trouble coming before it arrives.

The challenge is that no single indicator tells the whole story. Credit spreads can widen without a recession following (they did in 2015–2016 during the oil price crash). The yield curve can invert without an immediate recession (it inverted in 2022 and the recession still hasn’t arrived by April 2026). The VIX can spike on a single geopolitical event and mean-revert within weeks. Each indicator captures one dimension of stress — credit risk, market fear, monetary tightness, labor weakness — and each has blind spots that the others cover.

That’s why professionals don’t watch one indicator. They watch a dashboard. The seven indicators in this series were chosen because they span the full anatomy of financial stress: the bond market (high-yield spreads), the equity market (VIX), the rate market (yield curve and mortgage rates), the labor market (initial claims and the Sahm Rule), and the overall financial system (the Chicago Fed’s National Financial Conditions Index). Together, they’ve predicted every recession since the 1970s and flagged every financial crisis since the data began. Not always perfectly, not always in the same order, but always with enough lead time to matter.

The purpose of this series is not to predict the next recession. It’s to build literacy. To teach you what each indicator measures, why it matters, what levels signal trouble, and how to read them in real time. By the end of ten episodes, you’ll be able to glance at a dashboard of seven numbers and know — without commentary, without pundits, without headlines — whether the financial system is healthy, strained, or about to break.

Before we dive into each indicator over the coming episodes, here’s the full dashboard as it stands in April 2026. Each row shows the indicator, its current reading, where that reading falls in the historical distribution, and a signal color: green for calm, yellow for elevated, red for crisis-level. The thresholds aren’t arbitrary — they’re calibrated to the levels that historically preceded recessions and market dislocations.

Six of seven signals are green. The only yellow is the 30-year mortgage rate, which at 6.37% sits above the post-2000 average of about 5.1% but well below the historical median of 7.24% (which includes the double-digit rates of the 1980s and early 1990s). Elevated mortgage rates create slow-burn stress for housing affordability, but they don’t signal acute financial danger.

The standout readings are the high-yield credit spread and initial jobless claims, both sitting near their historical floors. A 2.90% HY spread means investors are demanding barely any premium over Treasuries to hold junk bonds — a sign of extreme confidence in corporate creditworthiness. And 219,000 initial claims per week means fewer Americans are being laid off than at almost any point in the 58-year history of the data. These are not the readings of an economy in trouble.

The high-yield credit spread (BAMLH0A0HYM2) measures the extra yield investors demand to hold bonds issued by companies rated below investment grade — the “junk bonds” that financed the LBO boom of the 1980s, the telecom bubble of the late 1990s, and the energy boom of the 2010s. When times are good and defaults are rare, the spread is tight — typically between 3% and 4%. When fear sets in, it widens violently. During the Lehman Brothers bankruptcy in September 2008, the spread went from 8% to 17% in two months. During the COVID crash of March 2020, it spiked from 4% to nearly 11% in three weeks. The spread is a real-time measure of how much risk the market perceives in the corporate sector, and it has a nearly perfect record of spiking before or during every recession since the data began in 1996.

The VIX (VIXCLS), formally the CBOE Volatility Index, measures the market’s expectation of 30-day volatility in the S&P 500, derived from options prices. It’s often called the “fear gauge,” though fear is only half the story — VIX also captures uncertainty, hedging demand, and market structure effects. A VIX of 12–15 indicates complacency. A VIX of 20–25 indicates normal market churn. A VIX above 30 signals genuine distress, and above 50, outright panic. The all-time intraday peak was 82.7 on March 16, 2020 — the Monday after President Trump declared a national emergency — surpassing even the 2008 financial crisis peak of 80.9. The VIX’s key property is mean-reversion: spikes are violent but temporary, and the index almost always reverts to the 15–20 range within weeks to months.

The yield curve (T10Y2Y) is the difference between the 10-year Treasury yield and the 2-year Treasury yield. In normal times, longer-dated bonds yield more than shorter-dated ones — investors demand compensation for locking up their money. When the curve “inverts” (turns negative), it means short-term rates exceed long-term rates, which historically signals that bond traders expect the Fed to cut rates in the future because the economy will weaken. An inverted yield curve has preceded every U.S. recession since the 1970s, with a lead time of 6 to 24 months. The curve inverted in July 2022 and stayed negative through late 2024 — the longest continuous inversion on record — before normalizing in early 2025. As of April 2026, the curve sits at +0.51%, positive but below its historical average of +0.85%.

The Sahm Rule (SAHMREALTIME) was created by economist Claudia Sahm in 2019 as a simple, reliable recession indicator. The rule triggers when the three-month moving average of the national unemployment rate rises 0.50 percentage points or more above its low over the previous twelve months. It has a perfect track record: every time it triggered since 1960, the economy was either already in recession or about to enter one. The rule briefly crossed 0.50 in mid-2024, sparking recession fears — but the trigger was driven by labor force growth (immigration) rather than job losses, and the economy kept expanding. It currently reads 0.20, well below the threshold, though as we’ll explore in Episode 5, the 2024 episode raised real questions about whether the rule’s mechanics can misfire in an economy with unusual labor force dynamics.

Initial jobless claims (ICSA) measures the number of Americans filing new applications for unemployment insurance each week. It’s the timeliest labor market indicator available — released every Thursday with just a one-week lag — and it captures the speed of deterioration in real time. In a healthy labor market, claims run between 200,000 and 250,000 per week. During the 2008 recession, they climbed to 665,000. During COVID, they exploded to 6.1 million in a single week — a number so large that many state unemployment offices couldn’t process the filings. The current reading of 219,000 is unremarkable in the best way: it tells you that layoffs are historically low and the labor market is functioning normally.

The National Financial Conditions Index (NFCI) is published weekly by the Chicago Fed and aggregates 105 measures of financial activity into a single number. It captures conditions across money markets, debt markets, equity markets, and the banking system. A reading of zero means conditions are at their historical average. Negative values mean conditions are “loose” — credit is flowing, markets are open, and risk appetite is high. Positive values mean conditions are “tight.” During the 2008 crisis, the NFCI hit 3.06 — financial conditions tighter than at any point since the Volcker era. During COVID, despite the market panic, the NFCI peaked at a relatively modest 0.60 because the Fed’s immediate intervention prevented a true credit freeze. The current reading of −0.43 indicates moderately loose conditions — the financial plumbing is working.

The 30-year fixed mortgage rate (MORTGAGE30US) is the price that average Americans pay to finance the largest purchase of their lives. It’s not a stress indicator in the traditional sense — it doesn’t spike during panics the way credit spreads or VIX do. Instead, it measures a different kind of stress: the slow, persistent squeeze on household balance sheets when the cost of shelter rises. In 1981, the 30-year rate peaked at 18.63% — a monthly payment on a $300,000 house would have been $4,664. By January 2021, it had fallen to 2.65% — the same house would cost $1,208 per month. It currently sits at 6.37%, roughly where it was in 2002 and 2007. Not crisis-level, but a world away from the 3% rates that defined the post-COVID housing boom.

The chart below plots two of the most important stress indicators — the high-yield credit spread and the NFCI — on a shared timeline from 1997 to 2026. The gray columns mark major stress events. What jumps out immediately is that the two indicators move in near-lockstep during crises but can diverge during calmer periods. Credit spreads widened dramatically in 2001–2002 (the dot-com bust and Enron/WorldCom fraud), but the NFCI stayed relatively contained because the banking system was healthy. In 2008, both exploded simultaneously — the banking system was the crisis. In 2020, the HY spread spiked to 10.87% but the NFCI barely crossed zero because the Fed intervened within days to backstop corporate credit markets.

The pattern that emerges across thirty years is consistent: credit spreads are the first mover. They widened in late 1998 (the LTCM/Russia crisis), through 2000–2002 (the dot-com bust), through 2007–2009 (the GFC), and in March 2020 (COVID). In every case, spreads moved before the official recession dating. The NFCI, because it aggregates a broader set of financial variables, tends to confirm rather than lead — but when it confirms, the signal is extremely reliable. A sustained NFCI reading above zero has preceded every recession since 1971.

The current moment is characterized by an unusual combination: credit spreads are historically tight, financial conditions are loose, the yield curve has de-inverted, and the labor market is strong. The only pressure point is mortgage rates, which remain elevated enough to freeze housing mobility but not enough to cause systemic stress. As of April 2026, the dashboard is about as green as it gets. But dashboards change. The GFC stress signals went from green to red in approximately 15 months (June 2007 to November 2008). COVID went from green to red in three weeks.

To understand what stress looks like when it arrives, consider how the seven indicators behaved during the four major stress events of the last three decades. The table below captures the peak (worst) reading for each indicator during each crisis. The shading darkens with severity.

The GFC and COVID each set records in different categories. The 2008 crisis produced the worst credit conditions (21.82% HY spread, NFCI at 3.06) because the banking system itself was failing. COVID produced the worst labor market shock (6.1 million initial claims, Sahm Rule at 9.50) and the highest VIX reading in history (82.7) because the shutdown was instant and total. But COVID’s financial conditions stress was muted — the NFCI peaked at just 0.60, one-fifth of the GFC level — because the Fed immediately deployed every tool it had learned to build over the previous decade.

The dot-com recession of 2001 was the mildest of the three by most measures, yet it produced credit spread readings (10.25%) that would be considered catastrophic today. The distinction matters: a 10% HY spread in 2001 reflected Enron, WorldCom, and telecom overbuild — specific sectors in specific trouble. A 10% spread in 2008 reflected systemic counterparty risk — banks doubting whether other banks would survive the week. The same number can mean very different things depending on whether the stress is concentrated or systemic.

What the four-crisis comparison reveals is that financial stress has multiple anatomies. Some crises are credit crises (2008). Some are volatility crises (COVID). Some are slow-burn deteriorations that show up first in employment data (2001). The dashboard captures all of these because the indicators cover different channels. A dashboard with all green lights doesn’t mean nothing can go wrong. It means the specific failure modes that these instruments detect are not currently active. That’s valuable information — but it’s not a guarantee.

The percentile chart tells the story in a single visual. High-yield spreads are in their 8th percentile — only 8% of historical readings have been lower. Initial claims are in their 10th percentile. The NFCI sits near its median. The mortgage rate, at the 36th percentile, reflects the fact that the full history includes the double-digit rates of the 1980s; relative to the post-2000 era, today’s 6.37% is actually above the 80th percentile. The yield curve’s 38th percentile reflects a positive but below-average spread — normal enough, but flat enough that a modest shift in rate expectations could push it back toward zero.

The Sahm Rule’s 66th percentile might seem high, but the interpretation is different for this indicator. Most of its historical readings cluster near zero (no recession signal), so a reading of 0.20 — comfortably below the 0.50 trigger — still ranks above the median simply because the median itself is 0.07. The Sahm Rule is a binary alarm, not a gradient. Below 0.50, it’s off. Above 0.50, it’s on. At 0.20, it’s off.

The GFC chart above illustrates what stress actually looks like as it develops. In June 2007, the HY spread was 2.98% and the VIX was 18.9 — numbers nearly identical to today’s readings. By August 2007, after two Bear Stearns hedge funds collapsed, the VIX had spiked to 30.8 and spreads had widened to 4.59%. The first warning shot. Then a partial recovery through early 2008, lulling many into complacency. Then Bear Stearns’s forced sale to JPMorgan in March 2008 pushed spreads to 8.62%. Another partial recovery. And finally, the cascade: Lehman Brothers’ bankruptcy on September 15, 2008, sent spreads from 11% to 22% and the VIX from 47 to 81 in just ten weeks.

The lesson is not that the GFC was inevitable from the first warning shot. It’s that the dashboard gave 15 months of increasingly urgent signals before the systemic collapse. Investors who were watching spreads and the NFCI in the summer of 2007 had over a year to reduce risk. Most didn’t, because the economy was still growing, unemployment was still low, and the stock market was still rising. The dashboard is only useful if you’re willing to act on it before the headlines confirm what the numbers already show.