Purchasing Managers Indexes and business survey indicators
Purchasing Managers Indexes (PMIs) are monthly surveys of corporate purchasing managers asking about new orders, output, employment, supplier delivery times, and inventories. Respondents report whether each factor is better, the same, or worse compared to the prior month. The resulting diffusion index reads above 50 when more respondents report improvement than deterioration, and below 50 when the reverse is true. PMIs lead actual industrial output and GDP by approximately 2-6 months because purchasing decisions precede production, which precedes revenue, which precedes GDP measurement.
The ISM Manufacturing PMI is the most widely tracked single PMI for US investors. An ISM reading above 55 signals strong expansion; below 50 signals contraction; and a sustained break below 48 has historically coincided with recession in the manufacturing sector within 2-3 quarters. The ISM Services PMI covers the larger services sector and tends to be more stable than manufacturing, making the manufacturing index a more sensitive leading indicator despite manufacturing representing a smaller share of US GDP. The Global Manufacturing PMI (J.P. Morgan/S&P Global) aggregates PMI data from dozens of countries, providing a leading signal for global trade and earnings for internationally exposed companies.
New orders subcomponents of PMIs are considered the most forward-looking component within the PMI itself. The ISM New Orders minus Inventories spread is a particularly clean signal: when new orders exceed inventories, businesses need to ramp production to meet demand (positive for manufacturing output); when inventories exceed orders, businesses can fulfill demand from stock without increasing production (negative for manufacturing output). This spread leads the overall ISM by approximately 1-2 months, making it a useful intra-PMI leading signal.
Business confidence surveys (Conference Board CEO Confidence, NFIB Small Business Optimism, the ifo Business Climate Index in Germany) provide qualitative leading signals about hiring plans, capital expenditure intentions, and pricing expectations. These are not as mechanically quantifiable as PMIs but add valuable color about the direction of business decision-making before the decisions show up in hard economic data. Small business optimism is particularly valuable for predicting labor market conditions because small businesses account for a disproportionate share of US employment growth.
Yield curve, jobless claims, and financial market indicators
The yield curve spread (10-year minus 2-year Treasury yield) is one of the most reliable leading indicators of recession in the United States. An inverted yield curve (where short-term rates exceed long-term rates) has preceded every US recession since the 1970s, typically by 12-24 months. The inversion reflects the bond market's expectation that the current high short-term rates (set by the Federal Reserve to control inflation) will need to be cut in the future as economic growth slows. Investors watch both the level of the spread (how deeply inverted) and the duration of the inversion (how long it has remained inverted) as indicators of recession probability.
Initial jobless claims are weekly filings for unemployment insurance from newly unemployed workers. Because they are reported weekly (versus monthly for most other leading indicators) and measure a concrete administrative action rather than a survey response, they are among the most timely leading signals available. Rising jobless claims indicate labor market softening before it appears in the monthly payroll or unemployment rate data. A sustained move above 300,000 weekly claims (adjusted for seasonal factors) has historically been an early recession warning. Claims below 250,000 indicate a tight labor market.
Credit spreads (the yield premium that investment-grade and high-yield corporate bonds pay over equivalent-maturity Treasuries) are forward-looking financial market indicators that lead economic activity. Widening credit spreads indicate rising default expectations and reduced appetite for credit risk, which historically precedes economic slowdown as businesses find it more expensive and difficult to borrow. High-yield spreads are the most sensitive because high-yield borrowers are more leveraged and more exposed to economic downturns. A move from normal levels (300-400 basis points over Treasuries for high-yield) to stressed levels (600+ basis points) has historically preceded recession by 3-12 months.
Equity market returns themselves lead economic activity, though with significant noise. The stock market is considered a leading indicator because equity prices are claims on future earnings, which depend on future economic conditions. The S&P 500 typically peaks 6-12 months before a recession and troughs 6-12 months before recovery begins. The challenge is that the stock market also generates many false signals (corrections that do not precede recessions). More reliable are earnings estimate revision trends (falling forward earnings estimates from analysts lead falling actual earnings by 2-4 months) and credit-equity risk signals (simultaneous widening in credit spreads and equity market weakness confirms rather than just suggests economic deterioration).
Building a composite leading indicator framework
No single leading indicator is reliably predictive across all economic cycles; different cycles are led by different components. Building a composite framework that weights multiple indicators reduces the false signal rate and provides a more robust forward view. The Conference Board Leading Economic Index is the most widely referenced composite, combining 10 components: building permits, ISM new orders, stock prices, credit conditions, consumer expectations, manufacturing hours, jobless claims, leading credit index components, the yield curve, and capital goods orders. A sustained decline of 4-5 points over 6 months has historically preceded recessions.
Investors can build a simplified version of a composite leading indicator using 4-5 high-signal components: the ISM New Orders minus Inventories spread (manufacturing leading signal), initial jobless claims 4-week moving average (labor market), the yield curve (2Y-10Y spread, credit cycle), high-yield credit spreads (risk appetite and default expectations), and the 6-month rate of change in commodity prices (inflation and demand signals combined). Monitoring the direction and rate of change across all five simultaneously, looking for confirmation across multiple signals, reduces the risk of acting on a single-indicator false positive.
Leading indicators fail most conspicuously in two scenarios. First, when the cycle is driven by financial system stress rather than traditional demand dynamics (the 2008-09 global financial crisis): credit spreads and financial conditions indicators led well but traditional survey-based PMIs and the yield curve were slower to signal the severity. Second, when exogenous shocks (pandemics, geopolitical supply disruptions) override gradual cycle dynamics: no leading indicator framework predicted the March 2020 economic collapse more than a few days before it occurred. These limitations argue for combining leading economic indicators with a monitoring framework for tail risks rather than relying on them as exclusive predictors of economic turning points.
Frequently asked questions
What are leading economic indicators?
Leading economic indicators predict future economic activity before it appears in lagging data like GDP or the unemployment rate. They lead because they measure decisions or conditions that precede actual economic output: business surveys measure purchase orders before production; initial jobless claims count new unemployment filings before layoffs show in payroll data; the yield curve reflects market expectations about future interest rates before the Fed changes them. The Conference Board Leading Economic Index aggregates 10 such indicators into one composite signal.
Does an inverted yield curve always predict recession?
An inverted yield curve (short-term rates higher than long-term rates, typically measured as the 10Y minus 2Y Treasury spread) has preceded every US recession since the 1970s, but with variable lead times of 12-24 months. It does not predict the timing or severity of the recession, and the economy can continue growing for over a year after inversion before contraction begins. The signal is reliable as a directional indicator of elevated recession risk, but it is not a precise timing tool and has generated false signals outside the US in some other countries.
What does the ISM Manufacturing PMI measure?
The ISM Manufacturing PMI is a monthly diffusion index based on surveys of US corporate purchasing managers in manufacturing industries. A reading above 50 means more respondents reported improvement than deterioration in manufacturing activity (new orders, output, employment, deliveries). Below 50 means the reverse. Readings above 55 signal strong expansion; below 50 signals contraction; sustained readings below 48 have historically coincided with manufacturing recession. The new orders component leads the overall index by 1-2 months.
Why do credit spreads lead the economy?
Credit spreads (the yield premium corporate bonds pay over equivalent Treasuries) reflect the market's assessment of default risk and risk appetite. When spreads widen, borrowing becomes more expensive for businesses, reducing investment, hiring, and output. High-yield spreads are the most sensitive because high-yield borrowers are more leveraged. Widening spreads also reflect reduced bank willingness to lend, tightening financial conditions across the economy. These financial tightening effects take 3-12 months to flow through to economic output, making credit spreads one of the most useful leading indicators of economic slowdown.
How far ahead do leading economic indicators predict?
Most leading economic indicators lead economic turning points by 3-12 months. Business surveys (PMIs) lead by 2-6 months; initial jobless claims lead by 1-3 months (they are more timely but shorter lead); the yield curve leads by 12-24 months; credit spreads lead by 3-12 months. Composite indexes (Conference Board LEI) aim to balance short and long lead times across their components, providing an average lead of about 6-9 months before recession starts and about 3-6 months before recovery begins.