Common-Signal Economic Dashboard
Combine noisy indicators to isolate the economy's shared momentum
- Difficulty
- Advanced
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 90%
The Common-Signal Economic Dashboard treats economic monitoring as a portfolio problem. Analysts collect indicators across employment, production, demand, and financial conditions because each series mixes useful information with idiosyncratic noise. They normalize the inputs, measure their momentum, and extract the small number of common components moving across the economy. This avoids anchoring on GDP, payrolls, manufacturing surveys, or any other statistic whose usefulness can rise and fall. Job growth remains especially meaningful because hiring and layoffs represent consequential business decisions, while the yield curve is a strong recession indicator. Even so, neither should stand alone. The framework produces a broader estimate of economic direction and improves resilience when a formerly reliable indicator temporarily breaks down.
Origin
Extracted from Masters in Business. Davis described Vanguard's long-running use of big-data techniques to combine economic statistics and isolate their common signals.
Core principles
- 01Every indicator contains both signal and noise.
- 02Indicator usefulness changes over time.
- 03Shared movement is more reliable than one celebrated statistic.
- 04Employment decisions carry substantial economic information.
- 05The yield curve is useful but should remain part of a broader portfolio.
How to run it
- 1
Build the indicator portfolio
Collect economic series from labor, production, consumption, business activity, credit, and markets. Preserve indicators with partial information instead of searching for one perfect measure.
Pro tip Include data with different publication schedules and economic mechanisms.
Watch out Avoid filling the dashboard with multiple versions of the same underlying signal.
- 2
Normalize the inputs
Transform unlike series into comparable measures of direction and momentum. Account for whether higher values indicate strength or weakness.
Pro tip Use changes or standardized deviations when raw units differ substantially.
Watch out Poor normalization can make volatile series dominate the aggregate.
- 3
Extract common components
Identify the two or three shared signals explaining movement across the dataset. Interpret them as broad trend and momentum rather than literal economic variables.
Pro tip Test whether the components remain stable when individual indicators are removed.
Watch out A component that depends on one series is not genuinely common.
- 4
Examine disagreements
Compare prominent indicators such as job growth, GDP, and the yield curve with the aggregate. Investigate whether divergence reflects noise, lag, or a changing relationship.
Pro tip Give extra attention to employment decisions and the cash-to-ten-year yield curve.
Watch out Do not override the aggregate merely because a popular release attracts stronger headlines.
- 5
Update the economic assessment
Use changes in the common signals to revise the direction and strength of the outlook. Retain uncertainty and avoid translating the result into a guaranteed recession or expansion call.
Pro tip Monitor whether weakness broadens across categories before escalating conviction.
Watch out No combination of indicators eliminates forecast error.
In the wild
GDP accelerates because inventories build while labor and demand indicators soften. Rather than interpreting the headline as stronger growth, the dashboard compares momentum across the full indicator portfolio and may conclude that the common signal is weakening.
→ The analyst avoids mistaking an accounting-driven GDP increase for broad economic acceleration.
The cash-to-ten-year curve inverts while employment remains firm. The curve raises recession risk, but the analyst checks whether credit, production, demand, and labor signals are also deteriorating before making a high-conviction call.
→ A powerful indicator informs the assessment without becoming the entire forecasting system.
Common mistakes
Selecting one favorite indicator
Every indicator contains noise and can pass through periods when its predictive value weakens.
Overweighting GDP
GDP contains accounting effects such as inventory changes that can obscure the underlying direction of activity.
Ignoring breadth
A signal concentrated in one category is weaker than a change confirmed across labor, demand, production, and financial conditions.
Is it for you?
Best for
It is best for economists, investors, and business planners monitoring broad momentum or recession risk.
Not ideal for
It is not ideal when a decision depends on one narrowly defined operational measure rather than the overall economy.
From the transcript
“They all have some signal in it and noise.”
“Because every indicator goes through fits and starts of saying being very valuable and then not being valuable.”
“We look at all of them and an extract from that what are the common two or three signals that you could say this is…”
From the episode
Vanguard's Joe Davis Discusses Global Economics (Podcast)