This section moves from the abstract definition of a macro beta to the concrete macro variables that actually drive equity returns. Where the parent node "What a Macro Beta / Elasticity Is" establishes the general machinery — regress an asset's return on a factor, read the slope, watch it drift — this section asks the practical question: which macro factors matter, and how does a stock's sensitivity to each one arise, behave, and break down? Six factors get their own node here because they are the macro variables most commonly priced into equities and most often invoked on trading desks: interest-rate duration, the dollar (DXY), oil and commodities, inflation and breakevens, credit spreads, and the 10-year / real yield. The core tension that runs through every one of them is the same: each sensitivity is real and economically grounded at the cross-sectional level (which names are exposed, and which way), yet noisy, regime-dependent, and frequently sign-unstable at the time-series level (how much a stock will actually move next time the factor moves).

The factors in this section

Each child node defines one factor's exposure — its mechanism, how it is measured, who is exposed and in which direction, the honest evidence, and the #1 misuse. In brief:

• Equity Duration (Interest-Rate Beta) — the bond-duration concept applied to stocks as long-dated cash-flow claims. Long-duration (high-growth, low-payout, high-multiple) names should fall harder when long rates rise. The cross-sectional measure (Dechow, Sloan & Soliman implied duration) is academically robust; the real-time return-vs-rate beta is weak and has inverted across decades.
• Sensitivity to the Dollar (DXY Beta) — how a stock moves with the U.S. dollar, bundling translation, competitiveness, commodity-pricing, and risk-flow channels into one fragile coefficient. Best-evidenced as a priced factor for EM/global equities (BIS); regime-conditional and unreliable as a U.S.-equity timing tool (the dollar-smile problem).
• Sensitivity to Oil & Commodities — revenue-side beneficiaries (producers, with operating leverage) versus cost-side victims (airlines, chemicals). Structural and tradable at the single-stock level; at the index level the sign depends on whether the oil move is supply- or demand-driven (Kilian).
• Sensitivity to Inflation & Breakevens — response to expected inflation read off TIPS breakevens. The uncomfortable foundational fact (Fama & Schwert, 1977): aggregate stocks are a perverse inflation hedge. The sector/value tilts are better supported but lean heavily on the energy-led 2021–23 episode.
• Sensitivity to Credit Spreads / Risk — equity as the residual, most-levered claim repricing alongside corporate credit. The robust signal is the Gilchrist-Zakrajšek excess bond premium; the popular bps-threshold "recession triggers" are commercially published and small-sample.
• Sensitivity to the 10-Year & Real Yields — the discount-rate channel, with the real (TIPS) yield as the cleaner valuation input. Direction is universally taught; magnitude is weak and sign-unstable (Morningstar reports only about a −0.33 correlation between tech stocks and the 10-year yield over roughly the prior 15 years — a modest inverse relationship, not the tight one folklore assumes; AQR shows the stock-bond correlation flips with the macro regime).

A reader who has internalized all six will notice the same skeleton repeats: a clean theoretical mechanism, a structurally reliable cross-sectional ranking, and an unreliable time-series point estimate.

The cross-cutting tensions (why these belong together)

Three problems recur across every factor and are the reason this is a section, not six unrelated entries:

1. The factors are not independent. Oil, the dollar, inflation, rates, and credit move together and are often driven by a common third force (growth, policy, risk sentiment). A rise in the 10-year can be an inflation story, a real-rate story, or a growth story — with opposite equity implications. This collinearity means a single-factor beta absorbs exposure that really belongs to a correlated neighbor, and a multivariate fit can produce unstable, hard-to-interpret loadings. The fix — orthogonalizing the factors so each loading reflects a distinct economic channel — is handled in the sibling "Estimating Sensitivities" branch, not here; this section deliberately defines each factor in isolation so the channels are clear before they are disentangled.

2. "Why is the factor moving?" usually dominates "how exposed is the stock?" This is the single most important practical lesson in the section. The same factor move can be bullish or bearish depending on its driver: a demand-driven oil rally is pro-cyclical, a supply-shock spike is a tax on consumers; a growth-driven yield rise can lift stocks and rates together, an inflation-driven one compresses valuations. A naive beta that ignores the shock decomposition (Kilian for oil, the real/breakeven split for yields, the EBP/default split for spreads) conflates very different states.

3. Sign and magnitude are regime-dependent and unstable out of sample. This is the honest, evidence-backed caveat that every child repeats and that the academic literature confirms broadly: standard time-series macro betas (interest-rate, term-spread, credit-spread, inflation) carry over poorly out of sample, which is precisely why Esakia & Goltz (Financial Analysts Journal 2022) propose a firm-level measurement approach to improve out-of-sample robustness over conventional regression betas. Separately, the stock-bond correlation itself inverted from negative (2000s–2010s) to positive (2021–22), per AQR. Backward-looking betas mislead at exactly the regime turns that matter most.

When this section matters — and when it doesn't

These sensitivities matter most when a macro shock is the dominant return driver: policy pivots, inflation surprises, oil shocks, credit-stress events, sharp dollar moves. In those windows the cross-sectional map (which names face a headwind vs a tailwind) is genuinely decision-useful. They matter least for individual stocks in calm regimes, where idiosyncratic news (earnings, guidance, M&A) dwarfs every macro beta, and they are actively dangerous when a stale beta is treated as a stable, tradable constant.

Sources

• Esakia, M. & Goltz, F., "Targeting Macroeconomic Exposures in Equity Portfolios: A Firm-Level Measurement Approach for Out-of-Sample Robustness," Financial Analysts Journal Vol. 79, No. 1 (2022; print 2023; Graham & Dodd Top Award) — documents that standard time-series macro betas (rates, term spread, credit spread, inflation) are not robust out of sample and proposes a firm-level method that improves out-of-sample exposure stability. https://rpc.cfainstitute.org/research/financial-analysts-journal/2022/targeting-macroeconomic-exposures-in-equity-portfolios
• The six child nodes in this folder carry the primary, factor-specific sourcing (Dechow-Sloan-Soliman; Bruno-Shim-Shin / BIS; Kilian & Park; Fama & Schwert; Gilchrist & Zakrajšek; Morningstar; AQR). See each node's Sources section.
• IFM Investors, "Macro-factors revisited" and SimCorp, "When macro factors speak" — practitioner framing of the standard growth/inflation/rates/credit/commodities factor set and the need to limit redundant, multicollinear variables. https://www.ifminvestors.com/news-and-insights/thought-leadership/macro-factors-revisited-an-evolving-approach-to-portfolio-resilience/

Dispute flags inherited from the children: every factor in this set has a contested time-series component — the cross-sectional rankings are robust, the dynamic betas are not. The most overstated folklore in the section is the precise, mechanical application of any single beta ("yields up = short tech", "600 bps = sell", "stocks hedge inflation"); each child documents why the measured evidence is weaker and more regime-conditional than the folklore implies.