Playbook

Reading the macro regime: growth × inflation quadrants

Most large asset moves are not driven by earnings surprises or company-specific news — they are driven by regime shifts in the macro backdrop. Before tilting your factor screen or buying into a bottom-up thesis, it pays to know which quadrant you’re in, what is changing at the margin, and whether the trend filter says you should be leaning in or standing aside. Here is a practical framework for doing that with an agent.

The short version

  • Asset returns are largely regime-driven: classify the growth × inflation quadrant (Dalio’s economic machine) before picking individual names.
  • Watch what is changing at the margin in liquidity (Druckenmiller): the rate-of-change of Fed Funds, M2, and the yield curve often leads asset prices by 12–18 months.
  • COT positioning extremes (Soros / CFTC) flag late-cycle momentum crowding and set up reflexive reversals.
  • The 200-day moving average (Paul Tudor Jones) is the outer loop: if the broad market is below it, reduce gross exposure before applying any other tilt.
2×2
Growth × inflation quadrant grid (Dalio economic machine) — four regimes, each with a distinct historical asset-class ranking.
200-day
Moving-average regime filter (PTJ): below it, reduce gross exposure across all strategies regardless of the quadrant thesis.
COT
CFTC Commitments of Traders — managed-money extremes (>2 SD from trailing mean) mark potential Soros-style reflexive reversal setups.

The growth × inflation quadrant (Dalio’s economic machine)

Ray Dalio’s central insight is that you cannot reliably forecast which quadrant comes next — but you can identify which one you are in right now, and tilt your exposure toward the asset classes that have historically been rewarded in that environment. The 2×2 grid crosses two binary assessments: is growth rising or falling? Is inflation rising or falling?

The signals for each axis are not mysterious. Growth direction is read from nonfarm payrolls trends, unemployment rate direction, real GDP growth, industrial production, and the NBER recession flag. Inflation direction is read from the three-print slope of CPI year-over-year, PCE, and PPI. None of these requires a forecast — just a read of the current trend relative to three to six months ago.

QuadrantConditionsWhat has tended to lead (historically)
Goldilocks Growth rising, inflation falling Equities, investment-grade credit; commodities and gold have tended to lag
Reflation Growth rising, inflation rising Commodities, TIPS, EM equities; nominal long-duration bonds have tended to underperform
Stagflation Growth falling, inflation rising Commodities, gold, TIPS, short-duration assets; equities and long-duration nominal bonds have tended to struggle
Deflation / Bust Growth falling, inflation falling Nominal Treasuries, cash; equities and commodities have tended to be under pressure

One refinement that matters as much as the quadrant label itself: the direction-of-change within the quadrant. “Early Goldilocks” — where growth is accelerating off a low base and inflation is decelerating — has historically been a very different setup from “late Goldilocks rolling toward Reflation,” where growth is still positive but inflation is reaccelerating. The quadrant tells you the current read; the direction-of-change tells you which adjacent quadrant you may be transitioning into and how to position accordingly.

Quadrant failure mode (Dalio’s own warning): in crises, correlations collapse toward 1. Bonds and equities can fall together (as in 2022), and the diversification logic of risk parity breaks precisely when you need it most. The quadrant model is a regime guide, not a drawdown-elimination machine — always monitor whether the usual cross-asset correlations are holding.

Regime analysis also feeds your factor screen. In a Goldilocks environment, quality and growth factors have historically had a tailwind. In Stagflation, energy and materials SICs tend to lead. In a Bust, low-debt, high-cash quality names have historically held up better. See the factor screen playbook for how to tilt those filters based on the regime read.

Liquidity over earnings (Druckenmiller)

Stanley Druckenmiller’s framing is the most practically actionable layer on top of the Dalio quadrant: “Earnings don’t move the market; the Fed does.” The point is not that earnings are irrelevant, but that the rate-of-change of liquidity conditions tends to lead asset prices by 12 to 18 months — long before earnings revisions catch up.

The signals to watch are not the absolute level of rates but the marginal direction: is the Fed Funds rate rising or falling? Is the M2 money supply growth rate reaccelerating or decelerating? Is the yield curve (10Y minus 2Y) inverting further or beginning to un-invert? A curve that is still inverted but beginning to steepen, combined with a reaccelerating M2, is a classic Druckenmiller-style early signal that the marginal buyer is returning to risk assets — well before any earnings confirmation.

The practical implication: position on the change in liquidity, not on current earnings. A stock that looks expensive on today’s earnings multiple may be cheap on the earnings that will be reported 18 months from now, once the liquidity impulse works through. Conversely, a stock that looks cheap while liquidity is tightening can continue to de-rate for a long time before the multiple stops compressing.

Positioning extremes and reflexivity (Soros)

George Soros’s reflexivity framework holds that markets are not self-correcting: the relationship between perceptions and fundamentals forms a feedback loop. Self-reinforcing trends — credit booms, commodity supercycles, currency pegs — ride far longer than fundamentals alone would justify, then break sharply. The distinguishing feature is the asymmetry: booms are slow and long; busts are fast and sharp.

The CFTC Commitments of Traders (COT) report is the most direct window into positioning crowding. The managed-money net position in a commodity or financial futures market shows whether the trend-following community is already crowded in one direction. An extreme net-long position — more than two standard deviations above the trailing mean — in a market that has stopped making new highs is a Soros-style setup: the trend is losing momentum but positioning is still euphoric, which sets up a reflexive reversal.

The COT signal is strongest when combined with a price confirmation: the market fails to make a new high on expanding open interest, or breaks a key support level on volume. On its own, an extreme positioning reading is a caution flag, not a trade entry. The short side of a positioning extreme is the highest-reward setup in this framework, but timing is notoriously difficult — Soros himself was famously early on more than one occasion. Use confirmation before acting.

The 200-day regime filter (Paul Tudor Jones)

Paul Tudor Jones’s most widely cited discipline is also the simplest: defense first. If the broad-market index — in practice, SPY — is trading below its 200-day moving average, reduce gross exposure across all strategies. This is the outer loop that wraps every other signal in the framework.

The logic is not that the 200-day average has any special predictive power on its own. It is that being below it defines a regime where the trend is negative, the distribution of outcomes is skewed to the downside, and the risk-reward of holding full gross exposure is unfavorable regardless of how compelling the individual thesis looks. Below the 200-day: raise cash, tighten stops, do not add new risk. Above it: normal sizing. Jones’s stated requirement was a reward-to-risk ratio of at least 5:1 before entering any trade — and in a below-200-day regime, that threshold is rarely met on the long side.

This filter is also a useful circuit-breaker for the quadrant framework. A Reflation-quadrant commodity tilt looks attractive on paper, but if equities are in a bear trend below the 200-day, the macro backdrop is probably deteriorating faster than the slow-moving FRED data has yet confirmed. The trend filter sees the transition first.

How to read the regime with an agent

When your agent is connected to a markets MCP server, the full regime read runs in a single workflow. You do not write SQL or pull spreadsheets — you describe the question and let the agent call the tools in sequence.

StepToolWhat it returns
Regime snapshot macro_context One-call pull of the key FRED series: DFF, T10Y2Y, CPIAUCSL, UNRATE, NFPAYEMS, M2SL, real GDP, WTI, USD — each with its publication-lag age in days
Trend drill-down get_macro_series Full history for a specific series (e.g. 12 months of CPI) to read the multi-print slope, not just the latest snapshot
Series discovery search_macro / list_macro_categories Find the right series ID by natural language (e.g. “real GDP growth quarterly”) or browse by category (yields, employment, inflation, money, housing, FX)
Positioning extremes top_cot_extremes Top 15 markets by positioning z-score; use report_type="disaggregated" for commodities, "financial" for equity index / rates / FX futures
COT history get_cot_positions Full net-position history for one market (e.g. Gold, WTI, S&P 500 futures) over a user-specified window
200-day filter compute_price_features Returns close_vs_200d for SPY (positive = above; negative = below) and the sma_200d level
Cross-asset check correlation_regime Correlation matrix over a lookback (e.g. SPY, TLT, GLD, DBC, UUP) — a collapse toward 1 is the Dalio crisis-regime warning

A plain-English version you can paste to a connected agent:

“Give me the current macro regime read. Pull the latest values for CPI, UNRATE, NFPAYEMS, DFF, M2, the yield curve, and real GDP growth. Classify the growth-inflation quadrant and identify the direction-of-change. Check whether SPY is above or below its 200-day average. Then show me the top positioning extremes in commodity and financial futures via COT. Finally, give me the Druckenmiller liquidity read: what is changing at the margin in Fed Funds, M2, and the curve?”

That single prompt runs the full regime workflow. The agent calls macro_context for the snapshot, get_macro_series to read the slope on key series, compute_price_features for the PTJ filter, and top_cot_extremes for both disaggregated (commodities) and financial positioning — then synthesizes the four frameworks into a coherent verdict with a playbook tilt.

For deeper integration, a macro watchlist with watch_type='macro' will surface changes in these series automatically as part of your daily scan, without requiring a manual prompt each time.

Important caveats before you act on a regime read: Regimes are clearest in hindsight and only ambiguous in the present. Publication lags are real — CPI and unemployment data can be 3–4 weeks stale; M2 can be 6–8 weeks; the NBER recession flag is often declared months after the fact. Do not overfit: any single data point can be misleading; build your read from multiple series with consistent direction. This is context and risk-framing, not a precision timing system. Combine top-down regime context with bottom-up stock selection — the regime tells you which headwinds and tailwinds to account for, not which individual stocks to buy. Nothing here is financial advice.

What not to conclude

  • Don’t treat the quadrant table as a precise asset-allocation formula. The “historically tended to lead” relationships break in crises (2022 showed bonds and equities can fall together in a way the Dalio model did not predict).
  • Don’t call a regime transition until multiple series confirm it. A single bad CPI print does not make a Reflation; three consecutive prints with a clear slope do.
  • Don’t use USREC=0 as proof of no recession. The NBER declares recessions with a significant lag; use leading indicators (the yield curve, initial claims, purchasing manager surveys) as the early-warning layer.
  • Don’t size into a COT extreme without price confirmation. Positioning can remain extreme for months before it resolves — a reversal setup requires the market to actually start reversing, not just be crowded.
  • Don’t skip the PTJ filter. The 200-day check takes thirty seconds and is the single most reliable way to avoid deploying capital into a structural bear market regardless of how compelling the quadrant thesis looks on paper.

The macro regime framework is most powerful as a risk-framing tool: it tells you which factors and sectors have a structural tailwind, where the crowded trades are, and whether the trend is with you or against you. That context makes every bottom-up signal — an insider cluster buy, a factor screen hit, an activist 13D — more interpretable. A cluster buy in a commodity name during a Reflation regime has more conviction than the same signal against the regime. Read the macro first, then let it inform how much weight to put on the individual signal.

Frequently asked questions

What is a macro regime?

A macro regime is the prevailing combination of growth direction and inflation direction that tends to drive broad asset returns. Dalio’s economic-machine framework defines four regimes from a 2×2 grid: Goldilocks, Reflation, Stagflation, and Deflation/Bust. Each has historically been associated with a different asset-class ranking, though correlations can break down in crises.

What are the growth and inflation quadrants?

The four quadrants are: Goldilocks (rising growth, falling inflation) — historically equities and credit; Reflation (rising growth, rising inflation) — historically commodities, TIPS, EM equities; Stagflation (falling growth, rising inflation) — historically commodities, gold, short-duration; and Deflation/Bust (falling growth, falling inflation) — historically nominal Treasuries and cash. Direction-of-change within a quadrant (early vs. late, transitioning) matters as much as the label itself.

Can you time the market with macro regime analysis?

Regime analysis is a context-setting and risk-framing tool, not a precision timing system. Regimes are clearest in hindsight; the transition points are the hardest to call in real time. Most practitioners use it to tilt factor and sector exposure toward what has historically been rewarded in the current regime, while keeping stock selection as a separate, bottom-up process. The 200-day moving average (PTJ) is the simplest documented overlay: below it, reduce gross exposure across all strategies.

What data shows the current regime?

Key signals: growth (nonfarm payrolls trend, unemployment direction, real GDP, USREC); inflation (CPI YoY slope over the last three prints, PCE, PPI); liquidity (Fed Funds direction, M2 growth rate, yield-curve shape); positioning (CFTC COT extremes in commodity and financial futures). A ClawTerminal agent can pull all of these in a single macro_context call, then drill into any series trend with get_macro_series and check positioning with top_cot_extremes.

Read the regime before you pick the stock

Connect your agent to ClawTerminal and run the full macro-regime workflow — quadrant classification, liquidity read, COT positioning extremes, and the 200-day filter — in plain English. Free closed-beta key.