S&P 500 Sector Rotation Across the Business Cycle: A Four-Layer Public-Data Analysis

Author: PPBF Research  |  Dataset as of: May 14, 2026  |  Research status: Public-data reconstruction and methodology update

This article is research content only and should not be interpreted as investment advice. It is a public-data reconstruction inspired by the Visual Capitalist/SPDR analysis, but it does not directly replicate the original Conference Board LEI dataset, the SPDR Americas Research return dataset, or a full 1960-present official sector total-return history.

Executive Summary

Sector rotation is one of the most familiar narratives in market research. The basic idea is intuitive: different sectors of the equity market tend to perform differently depending on the stage of the business cycle. Defensive sectors are often expected to hold up better during recessions and slowdowns, while cyclical and growth-sensitive sectors are expected to lead during recoveries and expansions. The question is not whether this story is plausible, but how robust it remains when the cycle definition and the sector-return dataset are changed.

This article presents a four-layer public-data framework for studying S&P 500 sector performance across the business cycle. The project was inspired by Visual Capitalist's well-known presentation of top-performing S&P 500 sectors over the business cycle, which reports an analysis based on the Conference Board Leading Economic Index and SPDR Americas Research covering the period from December 1, 1960 to November 30, 2019.^[1] Our current work should not be read as a one-to-one replication of that study. It is instead a transparent, reproducible public-data reconstruction designed for periodic updates.

The four-layer setup compares two cycle-classification methods and two sector-return datasets. The first cycle method uses NBER fixed windows, anchored to official U.S. recession dates. The second uses a leading-indicator-style classifier based on the OECD Composite Leading Indicator for the United States, distributed through FRED.^[2] The first return dataset uses sector ETF adjusted-close proxies. The second uses public Yahoo Finance S&P 500 sector price-index proxies, which generally extend further back in time than ETFs but are price-return rather than total-return series.

The main conclusion is nuanced but useful. The classic sector-rotation narrative remains visible directionally: defensive sectors such as Consumer Staples, Utilities and Health Care remain relevant in weaker macro phases, while Technology, Consumer Discretionary, Industrials, Financials and other cyclical or growth-sensitive sectors appear more often in stronger phases. However, the identity of the exact winning sector changes across methods. That sensitivity is not a weakness of the analysis — it is the central finding.

The most reliable interpretation is that sector rotation is a macro-aware allocation framework, not a mechanical rule. The framework helps investors ask better questions about sector exposure, but it does not by itself prove that a simple investable rotation strategy will outperform after costs, taxes, turnover and timing risk.

The Four-Layer Research Design

The updated framework separates two methodological questions that were previously bundled together. The first question is whether the definition of the business cycle changes the results. NBER recession dating is authoritative for U.S. contractions, but it is retrospective and does not directly define all recovery, expansion and slowdown regimes used in sector-rotation frameworks.^[3] A leading-indicator model is conceptually closer to the Visual Capitalist/SPDR framing, but the exact Conference Board LEI dataset is not fully replicated here. The OECD/FRED CLI series is therefore used as a public proxy.

The second question is whether the return data change the results. ETF adjusted-close data are practical, investable and easy to refresh, but most sector ETF histories begin only in the late 1990s, with Real Estate much later. Public S&P 500 sector price-index proxies provide longer public histories for most sectors, generally from the early 1990s, but they are price-return series and therefore do not fully reflect dividends.

This design allows the analysis to distinguish cycle-definition effects from return-data effects. If a sector leads across most variants, the evidence is stronger. If the winner changes materially, the result should be treated as conditional rather than universal.

Data Sources and Coverage

The research uses public data sources so that the analysis can be updated periodically. The NBER business-cycle chronology provides the official recession anchor. The OECD Composite Leading Indicator for the United States, accessed via FRED, provides the public leading-indicator proxy. Sector returns are calculated from sector ETF proxies and public Yahoo Finance sector-index proxy tickers.

The public sector-index proxy coverage check found that most sector proxies are available from approximately May 1993. The Energy sector required an alternate Yahoo symbol (^GSPE), while Real Estate begins later and should be interpreted with additional caution. Communication Services and Real Estate also require careful interpretation because sector definitions changed materially over time.

Results: Phase Winners Across the Four Layers

The table below summarizes the winning sector in each business-cycle phase across the four method/data combinations. It is more informative than a single ranking table because it shows which conclusions survive across methods and which ones are sensitive to methodological choice.

Three observations stand out. First, Technology is the most consistent expansion leader, appearing as the winner in three of the four variants. Second, the recession winner shifts from Consumer Staples in the ETF-based variants to Health Care in the sector-index proxy variants, but the broader defensive interpretation remains intact. Third, slowdown leadership is more method-sensitive. Utilities lead in three variants, while Health Care leads in the sector-index/LEI combination.

Recession: The Defensive Message Survives

Under the ETF-based methods, Consumer Staples ranks first during recessions. Under the public sector-index proxy methods, Health Care ranks first. The difference is meaningful, but it does not overturn the defensive interpretation. Consumer Staples and Health Care are both defensive sectors whose earnings streams are generally less tied to discretionary spending, capital expenditure and credit-cycle sensitivity than sectors such as Financials, Industrials or Energy.

The repeated weakness of Financials and Industrials during recessions is economically plausible because these sectors are sensitive to credit conditions, investment cycles and broader business activity. Energy also appears weak in the ETF recession result, reflecting the macro and commodity shocks represented in the modern sample.

Recovery: Cyclical Leadership, but No Single Universal Winner

Recovery results are more sensitive to methodology than recession results. Under the ETF/NBER baseline and sector-index/NBER variant, Real Estate ranks first. Under the ETF/LEI variant, Consumer Discretionary leads. Under the sector-index/LEI variant, Technology leads.

The common thread is not one specific sector, but the preference for cyclical, growth-sensitive and normalization-sensitive exposure. Consumer Discretionary, Financials, Industrials, Materials, Real Estate and Technology all appear in top-three positions depending on the variant. This suggests that recovery leadership is a broad cyclical phenomenon rather than a single-sector rule.

Expansion: Technology Is the Most Persistent Leader

Expansion results show Technology as the strongest repeated winner. Technology leads in the ETF/NBER, sector-index/NBER and sector-index/LEI variants. The ETF/LEI variant is the exception, where Energy ranks first and Technology ranks second. This is a useful caution: even when the same sector-return dataset is used, the cycle definition can shift measured leadership.

The magnitude of expansion returns should not be compared mechanically across methods. NBER fixed-window expansion phases can cover long periods and therefore generate very large compounded returns. The LEI/OECD-CLI classifier creates more frequent regime segmentation, so average phase returns are lower and less dominated by long compounding windows.

Slowdown: Defensive Leadership Persists, with Health Care Gaining Importance

Slowdown results are directionally defensive but not identical across methods. Utilities lead under both ETF variants and under the sector-index/NBER variant. In the sector-index/LEI variant, Health Care leads, followed by Utilities and Technology.

The updated interpretation is therefore more nuanced than the first baseline. Utilities and Consumer Staples remain important, but Health Care becomes more prominent when the analysis combines the leading-indicator phase method with the longer sector-index proxy data.

How This Compares with the Visual Capitalist Reference

The Visual Capitalist article remains the inspiration and reference point, but the current output should not be described as a direct replication. The original article reports a Conference Board LEI-based framework and a 1960-2019 study window.^[1] The current project moves closer to that architecture by adding a public leading-indicator proxy and longer sector-index proxy histories, but two gaps remain.

First, the phase classifier uses the OECD Composite Leading Indicator via FRED, not the official Conference Board LEI with full diffusion detail. Second, the sector-index proxy track uses public price-index series, not official S&P sector total-return histories. Since dividends can materially affect long-horizon sector returns, especially in sectors such as Utilities, Consumer Staples, Financials and Energy, a final institutional-grade reconstruction should ideally use official total-return sector data.

Research and Investor Implications

The four-layer framework is more useful than a single ranking table because it highlights which conclusions are robust and which are conditional. The robust conclusion is that sector leadership changes with the macro regime. Defensive exposure remains most relevant in recessions and slowdowns, while cyclical and growth-sensitive sectors tend to perform better in recoveries and expansions. The conditional conclusion is that the identity of the number-one sector is sensitive to phase classification and return source.

For investors, this means sector rotation should be treated as a macro-aware allocation framework, not as a mechanical trading rule. The analysis does not prove that a simple sector-rotation strategy outperforms after costs, taxes, turnover and timing errors. Any investable strategy would still require explicit rebalancing rules, signal timing, valuation controls, out-of-sample testing and risk management.

For PPBF, the strongest publication framing is to present this as a transparent international update of a familiar sector-rotation narrative. The research intentionally shows multiple methods side by side. That strengthens credibility because readers can see how much of the conclusion survives when both the cycle definition and the return dataset change.

Recommended Next Steps

The current four-layer public-data setup is strong enough for a transparent research article. It is not yet sufficient for a definitive historical reconstruction of the Visual Capitalist/SPDR dataset. The highest-value next improvement would be to obtain official S&P Dow Jones sector total-return histories or another licensed institutional dataset. If those histories become available, the existing pipeline can absorb them as a fifth and more authoritative data layer.

A second possible next step is to turn this article into a simple interactive research tool. The tool could allow readers to switch between ETF/NBER, ETF/LEI, sector-index/NBER and sector-index/LEI views — making the methodological sensitivity visible rather than hidden.

Conclusion

The project has moved from a first reproducible baseline to a more mature public-data research framework. The most important improvement is not only the addition of new data. It is the ability to compare sector leadership across four methodological layers and to separate data-source effects from cycle-definition effects.

The updated findings support the broad sector-rotation framework. Defensive sectors remain important in weaker phases, and cyclical or growth-oriented sectors gain prominence in recovery and expansion phases. At the same time, exact sector rankings are not universal — they depend on how the business cycle is defined and which return series is used.

This four-layer version is therefore suitable as a public research article, provided that the caveats remain clear. Until licensed total-return histories and a fuller Conference Board LEI implementation are added, the article is best described as a transparent, reproducible and internationally publishable public-data reconstruction of S&P 500 sector rotation across the business cycle.

References

1. Visual Capitalist — The Top Performing S&P 500 Sectors Over the Business Cycle
2. The Conference Board — US Leading Indicators
3. FRED — OECD Composite Leading Indicator: Normalised for the United States (USALOLITONOSTSAM)
4. NBER — US Business Cycle Expansions and Contractions
5. Yahoo Finance — Public sector-index proxy data
6. S&P Dow Jones Indices — S&P Sectors