What is the difference between leading, coincident, and lagging indicators, and why do analysts focus mostly on leading ones?

Short answer: the three types differ by WHEN they turn relative to the business cycle. Leading indicators turn BEFORE the cycle (typically 6-9 months ahead), coincident indicators turn WITH the cycle, and lagging indicators turn AFTER. Analysts focus on leading indicators because they have predictive value — they tell you what is likely COMING. Coincident and lagging indicators confirm where you are or were, which is much less useful for investment decisions.

The timing distinction

Loading diagram...

Examples

Type	Typical examples
Leading	Yield curve slope, new orders for capital goods, building permits, share prices, average weekly hours in manufacturing, OECD composite LEI
Coincident	Industrial production, real personal income, payroll employment, retail sales
Lagging	Unemployment duration, ratio of consumer credit to income, change in CPI for services, average prime lending rate

Why leading indicators dominate

For an investor, the question is "what is the economy going to do NEXT" — not "what did it just do." Leading indicators answer the predictive question. The CFA curriculum highlights that most analysts focus primarily on leading indicators because:

1. Forecasting investment returns requires anticipating the next phase of the cycle
2. Asset allocation shifts must happen BEFORE the change is obvious — once everyone agrees the economy is slowing, equity prices have already fallen
3. Monetary policy moves before recessions begin — being early on the policy turn is worth more than being early on the GDP turn

The OECD composite LEI

The OECD constructs a composite leading indicator for each country/region from 5-9 variables. The composite is designed so its peaks and troughs occur 6-9 months before the corresponding GDP peaks and troughs, with reasonable consistency.

The diffusion index — combining multiple LEIs

When you track 10 individual leading indicators, you can compute a diffusion index:

$$\text{Diffusion} = \frac{\text{Number pointing up}}{\text{Total}}$$

If 7 of 10 point up, the diffusion index is 70\%. The interpretation: most leading indicators agree the economy is accelerating. If only 3 of 10 point up, the indicators disagree about direction, but the majority signal slowing.

The look-ahead bias trap

A major weakness of LEI methodology: the entire historical series is revised each month as new data arrives. The "latest" historical LEI series will appear to have predicted past cycles better than the LEI did IN REAL TIME. This is called look-ahead bias — the indicator looks better in backtests than it performs in live forecasting.

Implication: when evaluating an LEI track record, ask "how did this perform on the CURRENT vintage of data, vintages frozen at each historical point in time?" not "how does the latest vintage backtest?"

Why all three types matter for confirmation

Loading diagram...

Coincident and lagging indicators are not USELESS — they help validate or invalidate leading-indicator signals. The full picture comes from triangulating all three.

For more on forecasting approaches see our Module 1.05 article.