Trend-and-Cycle Earnings Framework
Forecast market direction by separating structural earnings trends from cycles
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 97%
The framework forecasts equity-market direction by decomposing index earnings into economically coherent stock and sector groups. For each group, the analyst separates the persistent earnings trend from cyclical fluctuations and then identifies the variables that drive those fluctuations. Broad indicators such as ISM manufacturing may explain earnings for traditional companies, while the dollar, specialized industry indicators, or company-specific revenue patterns may matter more elsewhere. The forecasts are then aggregated into an index-level earnings view. The framework must be continually revisited because index composition and business models change; an indicator that once explained the whole market may later describe only part of it. The resulting earnings forecast becomes the foundation for market targets, with positioning and valuation treated as secondary modifiers rather than primary drivers.
Origin
Extracted from Masters in Business, where Binky Chadha described the quantitative frameworks developed by Deutsche Bank's equity-strategy team.
Core principles
- 01Earnings are the primary long-term driver of equity markets.
- 02Different stock groups require different cyclical indicators.
- 03Structural trends and cyclical deviations must be analyzed separately.
- 04Economic indicators can lose relevance when index composition changes.
- 05Sector-specific drivers improve forecasts over one aggregate model.
How to run it
- 1
Create Coherent Market Groups
Divide stocks and sectors according to shared earnings behavior rather than relying solely on official sector labels. Include companies from different sectors together when their revenue streams behave similarly.
Pro tip Use historical earnings correlations to test whether proposed groupings are economically coherent.
Watch out Official classifications can obscure similar business models, as illustrated by Visa and Mastercard behaving like technology growth companies.
- 2
Estimate the Structural Trend
Determine the prevailing long-term earnings path for each group. Treat this as the baseline against which cyclical acceleration or contraction will be measured.
Pro tip Use multiple cycles so a temporary boom does not become the assumed trend.
Watch out A pandemic-era surge or collapse can distort an estimated trend.
- 3
Identify Cyclical Drivers
Select the economic, financial, or industry variables that historically explain deviations from the trend. These may include ISM manufacturing, the US dollar, oil prices, or specialized sector indicators.
Pro tip Favor variables with a plausible causal connection to revenue or margins.
Watch out Correlation without an economic mechanism may fail abruptly.
- 4
Test Relationships by Group
Compare each group's earnings with its proposed cyclical drivers and assess whether the relationship still holds. Do not assume an indicator explains the entire index merely because it did historically.
Pro tip Analyze dominant index constituents separately when their earnings behavior differs from the rest of the market.
Watch out Index-level results can be overwhelmed by a small group of unusually large companies.
- 5
Build the Aggregate Forecast
Forecast earnings for each group using its trend and cyclical drivers, then combine the forecasts according to their contribution to index earnings.
Pro tip Show how much expected growth comes from each group so concentration is visible.
Watch out An attractive aggregate forecast may conceal weakness across most constituents.
- 6
Revisit and Revamp
Regularly revise, rebuild, or discard parts of the framework as market structure changes. Treat model maintenance as part of the process rather than an admission of failure.
Pro tip Investigate structural breaks whenever actual earnings repeatedly diverge from the model.
Watch out Keeping an obsolete relationship can create false confidence and persistent forecast errors.
In the wild
The traditional relationship between ISM manufacturing and S&P 500 earnings appeared to fail for roughly three years. Splitting earnings into megacap growth and technology versus everyone else showed that the relationship still held for the latter group. Exceptional megacap earnings growth was masking the manufacturing-linked weakness across the remainder of the index.
→ The decomposition supported a forecast in which non-megacap earnings could recover as manufacturing improved.
Rather than applying only a broad manufacturing indicator to the materials sector, the framework calls for a specialized chemicals indicator because US materials earnings contain distinct subgroups. This adds an industry-specific cyclical input that a general macro model could miss.
→ The sector forecast becomes more sensitive to its actual earnings drivers.
Common mistakes
Using One Indicator for the Entire Index
A broad variable may fit historical aggregate earnings but stop working when index composition changes. Split the index before concluding that the relationship has disappeared.
Treating Sector Labels as Economic Truth
Companies classified in different sectors can share the same revenue behavior. Grouping only by official labels can weaken the model.
Refusing to Retire a Broken Relationship
Framework components must be revised or discarded when repeated evidence shows that their explanatory power has changed.
Is it for you?
Best for
It is best for equity strategists, asset allocators, and analysts building medium-term earnings forecasts.
Not ideal for
It is not ideal for short-term trading driven primarily by technical signals or event risk.
From the transcript
“So you can think about it as basically a trend and cycle framework for each of the groups.”
“And the question that the trend is, you know, what has basically been prevailing for quite a while, and then the question is what drives…”
“ISM manufacturing is an interesting one because that's historically the one thing that explained SMP five hundred earnings extremely well, and that's kind of like…”
From the episode
Getting Earnings Right with Deutsche Bank's Binky Chadha
Binky Chadha