Conventional macro research focuses on tactical forecasting and asset selection. This series explores an inverse question: what does an empirically robust, risk-managed equilibrium portfolio reveal about the prevailing macroeconomic regime?
Our foundational premise is that systematic portfolios engineered for downside containment generate valuable structural signals. When an allocation framework is constrained to compound capital with minimal peak-to-trough drawdowns, its weekly equilibrium weights reflect the marginal compensation for risk across asset classes. Interpreted sequentially, these weights provide an objective reading of what the market is rewarding and what it is penalizing.
Over an eight-year out-of-sample period (2018–2026), a long-only allocation across twelve liquid ETFs delivered 7.6% annualized net return with a maximum peak-to-trough drawdown of −6.5% (Calmar ratio 1.16, True Sharpe 1.06). Across the trailing three-year cycle (2023–2026), compounding efficiency accelerated to a 10.8% annualized return with maximum drawdown contained at −2.6%, producing an institutional Calmar ratio of 4.22 and a True Sharpe ratio of 1.57 (evaluated above the 4.39% cash carry rate; Raw Sharpe 2.66). The strategy's quantitative mandate is asymmetric capital preservation, structural compounding efficiency, and strict downside containment across divergent macroeconomic regimes. That empirical stability serves as the calibration benchmark validating the portfolio's weights as an objective macroeconomic gauge.
The dynamic investment strategy operates by systematically recalibrating capital distribution as cross-asset interdependencies, term premia, and liquidity regimes evolve. By continuously reconciling marginal risk contributions across equities, fixed income, real assets, and cash collateral, the framework adapts its posture to prevailing market physics, providing an observable macroeconomic reading of which structural risk factors are being rewarded over time.
This introductory paper establishes the core quantitative framework and equilibrium philosophy. Going forward, the series will utilize the weekly-rebalanced equilibrium portfolio through monthly publications—supplemented by event-driven monitors when structural shifts occur—to observe and diagnose evolving cross-asset market conditions, demonstrating how dynamic allocations serve as an objective empirical barometer of market structure.
To evaluate the strategy properly, one must define its quantitative mandate: delivering steady compound growth while strictly bounding maximum drawdowns and preserving capital across adverse market cycles. Over the 2018–2026 out-of-sample period, the strategy generated an annualized net return of 7.6%, compared with 16.1% for the S&P 500 and 10.1% for a standard 60/40 benchmark, prioritizing path stability and capital preservation across distinct macro regimes.
What the strategy achieved was consistent capital compounding with a fraction of the downside volatility and drawdown severity. Navigating the March 2020 liquidity shock, the 2022 cross-asset correlation breakdown (wherein simultaneous stock and bond declines dismantled traditional 60/40 diversification), and the recent steepening sell-off across long-duration Treasuries, the portfolio's maximum peak-to-trough drawdown was limited to −6.5%. By contrast, the 60/40 portfolio experienced a peak drawdown of −21.0%, while the S&P 500 fell −33.7%.
Robust downside-adjusted stability serves as the foundational empirical prerequisite validating the portfolio's weights as an informative macroeconomic equilibrium signal.
In traditional macroeconomic management, portfolio construction is the terminus of the analytical process: an investment committee forms qualitative views on growth, policy rates, inflation, and sovereign risk, and asset weights represent the tactical execution of those views. We propose inverting this paradigm. A systematic portfolio governed by rigorous downside-containment constraints is itself a rich source of macroeconomic intelligence—and arguably less susceptible to cognitive biases than discretionary forecasting.
The quantitative mechanism is straightforward. At each weekly rebalancing interval, the model evaluates twelve liquid assets spanning global equities, sovereign rates, currency, precious metals, broad commodities, and cash equivalents. It identifies the optimal allocation that maximizes net risk-adjusted compounding while strictly containing downside risk. The resulting vector of weights represents a dynamic equilibrium: a state wherein each asset's marginal contribution to risk is precisely justified by its marginal expected contribution to portfolio stability. Crucially, this equilibrium endogenously reflects prevailing cross-asset interdependencies, volatility clustering, and dynamic market physics.
Hence the macroeconomic barometer. During liquidity contractions or risk-off episodes, equilibrium weights shift toward cash instruments and the US dollar. During supply-side commodity dislocations, energy equities and raw materials expand their footprint. When the term premium widens, duration-sensitive bonds are systematically de-allocated. The evolving asset weights thus constitute an empirical, real-time map of which risks the market is compensating, and which it is penalizing.
Evaluating return relative to maximum drawdown (the Calmar ratio) captures path-dependent capital preservation. Institutional allocators recognize that risk is fundamentally asymmetric: terminal wealth destruction, drawdown depth, and the duration of capital impairment. By systematically optimizing for downside containment, the framework identifies early inflection points where cross-asset interdependencies shift and risk clusters form, yielding the empirical intelligence embedded in the equilibrium weights.
Over the eight-year period, the model's equilibrium allocations exhibit four distinct structural macro patterns:
The quantitative framework adheres to strict principles of model parsimony. In portfolio optimization, unconstrained complexity frequently introduces parameter instability and overfits noise. Four core architectural constraints govern the model.
Twelve highly liquid US-listed ETFs, selected ex-ante to represent distinct and orthogonal macroeconomic drivers[2]: domestic and global growth, value versus growth factors, non-cyclical earnings, commodity inflation, sovereign duration, currency reserve flows, and risk-free collateral. The universe consists exclusively of highly liquid, transparently traded instruments, ensuring frictionless institutional execution, direct price discovery, and custodian-agnostic implementation without synthetic leverage.
| Ticker | What it holds | The macro force it represents |
|---|---|---|
| SPY | S&P 500 | US economic growth |
| VLU | US value-tilted equities | Growth, with a valuation discipline |
| XLV | US healthcare equities | Defensive, non-cyclical earnings |
| XLE | US energy equities | Oil prices and supply shocks |
| EEM | Emerging-market equities | Global growth outside the US |
| MCHI | Chinese equities | China's economic cycle |
| IEF | 7–10 year US Treasuries | Medium-term interest rates |
| TLT | 20+ year US Treasuries | Term premium and long-run fiscal credibility |
| UUP | US dollar index | Global demand for dollar liquidity |
| GLD | Physical gold | A non-fiat, zero-duration hedge |
| DBC | Broad commodities | Real-economy inflation |
| SHV | Short-term Treasuries (0–1 year) | Cash: the option to wait |
The portfolio operates under a fully funded, long-only mandate (weights strictly sum to 100% on the simplex) with zero synthetic leverage and zero short positions. Single-asset concentration bounds prevent idiosyncratic factor dominance. While this aligns directly with institutional investment mandates, it also serves a foundational methodological purpose: a long-only portfolio provides an unambiguous statement of which risk factors offer positive marginal value. Short-term Treasuries (SHV) function as the unconstrained safe asset, enabling the model to de-risk entirely into risk-free collateral when market conditions fail to offer adequate risk premia.
The allocation engine executes a systematic weekly optimization and rebalancing step every Tuesday based on Monday closing prices, eliminating weekend gap noise and holiday settlement anomalies. Between weekly optimization steps, portfolio weights drift passively with market prices. To reflect the economic and operational reality of live portfolio management—and ensure true institutional investability—the optimization incorporates an explicit turnover penalty: rebalancing trades are executed only when the expected marginal improvement in risk-adjusted portfolio utility clears a quantitative threshold covering transaction costs and execution slippage. When the projected efficiency gain falls below this threshold, the prevailing allocation is retained. All reported performance and attribution metrics are net of a conservative one-way transaction cost of 6 basis points (0.06%) per trade leg.
The learning process produces a fully traceable, auditable, and explainable trail of current market conditions. By establishing equilibrium allocations across twelve liquid asset classes, the framework aggregates the collective state of the market into an empirical multi-asset state variable.
The architecture structures its pool of 60 candidate investment models as deterministic-yet-adaptive allocator agents:
Deterministic Execution Policy. Each of the 60 allocator agents operates under a distinct, explicit mathematical constitution θ = (τ, μ*, wmax, κ) defining its memory horizon (τ), minimum expected return target (μ*), concentration ceiling (wmax), and turnover penalty coefficient (κ). The expected return target is evaluated against trailing trend and momentum dynamics over horizon τ, ensuring that capital allocations satisfy a baseline expected drift while strictly containing cross-asset interdependencies. For any observed state of market interdependencies, the agent's target allocation is the unique global optimum of a constrained convex quadratic program. Every position is algebraically auditable through its marginal contribution to total portfolio risk, providing complete causal traceability from asset-level interdependencies to capital weights.
Dynamic Structural Adaptability. While an agent's constitutional rules are fixed, its internal state dynamically reflects the evolution of market physics. As cross-asset interdependencies tighten, correlation regimes invert, or liquidity costs shift, the agent automatically recalibrates its capital distribution to maintain risk equilibrium.
Walk-Forward Agent Selection (1 of 60). Learning and selection of the allocator agents operate via a continuous, out-of-sample walk-forward protocol that consists in activating a single agent among the 60 candidates. At regular calibration intervals, a supervisory audit evaluates all 60 candidate allocator agents across rolling historical windows, measuring how effectively each constitutional rule-set preserved capital and contained drawdowns across trailing market conditions. The single allocator agent demonstrating superior path stability and drawdown containment across the prevailing market regime is activated to govern portfolio capital for the subsequent period.
The Portfolio as an Empirical Macroeconomic State Variable. The resulting allocation vector represents an endogenous equilibrium point where cross-asset interdependencies, term premia, and liquidity conditions are reconciled. Because every step of the selection protocol and optimization policy forms a transparent mathematical trail, the evolving portfolio weights serve as an observable, aggregate state variable characterizing the multi-asset macro environment.
In risk-budgeted portfolio management, nominal capital allocation diverges sharply from marginal risk contribution. A large allocation to cash or short bills exerts negligible volatility drag, whereas a modest allocation to volatile cyclical commodities or energy equities can dominate total portfolio variance. Under Euler risk decomposition[3] as of the October 5, 2026 rebalancing, this divergence is especially pronounced.
Short-term Treasuries, the US dollar, and intermediate Treasuries hold 64.3% of capital but account for only 10.2% of total portfolio volatility.
Equities and broad commodities hold 35.7% of capital and generate 89.8% of total portfolio risk, driven primarily by energy and healthcare equities.
The portfolio's current posture conveys three unambiguous macroeconomic signals:
First, liquidity and short-duration yields remain highly attractive. A 31.4% allocation sits in short-term Treasuries (SHV, initiated at the September 29 rebalancing), harvesting an unencumbered risk-free cash yield with zero duration risk while awaiting clearer risk-premia opportunities. Second, sovereign duration is being penalized across the term structure. In constant-maturity fixed income ETFs, continuous contract roll under upward yield drift generates capital losses that current coupons fail to offset[4]. This dynamic accounts for intermediate Treasuries' (IEF) negative total return year-to-date and the complete divestment of long-dated paper (TLT) as 30-year yields reached 5.62% amid heavy sovereign and corporate debt issuance pressure[5]. The strategy retains a modest 9.1% allocation to intermediate Treasuries solely for correlation diversification while strictly bounding duration risk. Third, rewarded risk premia are concentrated in real asset inflation hedges and non-cyclical cash flows: energy equities (XLE), broad commodities (DBC), and defensive healthcare (XLV).
Across this eight-week sequence, the portfolio traversed four transient cross-asset equilibrium phases:
Value equity and dollar liquidity. A third of portfolio capital in the US dollar (UUP) and 20% in value equities (VLU), balanced with healthcare and emerging market exposures.
Initial inflation hedging. Tactical entries into gold and broad commodities emerge as geopolitical friction and cross-asset correlation risks rise.
Energy dislocation response. Following supply disruptions lifting crude prices above $90/bbl, energy equities (XLE) and broad commodities (DBC) expanded; at peak exposure, commodities represented 40.7% of total portfolio risk.
De-risking into cash equivalents. With the US 30-year yield breaching 5.62% alongside heavy corporate debt issuance, capital shifted to 31.4% short Treasuries (SHV), and gold was fully liquidated.
The full divestment from gold (GLD) warrants explicit commentary. Rising real yields alongside dollar strength sharply elevated the opportunity cost of holding non-yielding precious metals, whereas energy equities and commodities provided tangible inflation-hedging properties combined with positive carry. Crucially, this reflects a tactical re-optimization over the model's weekly rebalancing horizon, adapting to real interest rate and currency dynamics. As demonstrated in the multi-horizon attribution below, gold has historically been one of the highest net contributors to the strategy's overall Calmar ratio.
To isolate each asset's contribution to risk-adjusted compounding without ordering bias, we employ the Shapley value decomposition[6] from cooperative game theory. By evaluating an asset's marginal contribution across all 4,096 possible portfolio sub-coalitions (212), the Shapley formulation provides an axiomatic, path-independent attribution of net Calmar ratio across asset classes.
Over the full eight-year sample, structural compounding efficiency is anchored by three foundational assets: short-term Treasuries (SHV), gold (GLD), and the US dollar (UUP). Cash equivalents provide essential downside convexity by eliminating drawdowns during market liquidations. Gold and the dollar provide counter-cyclical crisis alpha, consistently appreciating during equity market dislocations. In 2026, energy equities (XLE) and emerging market equities (EEM) have generated substantial positive attribution, rewarding the strategy for dynamically capturing the commodity repricing. Conversely, fixed-rate sovereign duration (TLT, IEF) has exerted a persistent negative drag across recent windows, reflecting the structural bear market in bonds.
The primary takeaway is that the assets driving portfolio efficiency rotate systematically with macro regimes. The optimization engine systematically identifies these regime shifts, providing allocators with an objective map of structural market transitions.
Across the full eight-year sample, the trailing three-year tightening cycle, and year-to-date 2026, the performance profile demonstrates structural asymmetry: moderate upside participation relative to equities during momentum expansions, substantially contained peak drawdowns during corrections, and a consistently superior risk-to-drawdown ratio.
| Net of transaction friction | Annualized Return | Annualized Volatility | Maximum Drawdown | Calmar Ratio |
|---|---|---|---|---|
| July 2018 – October 2026 | ||||
| This strategy | 7.6% | 4.7% | −6.5% | 1.16 |
| 60/40 portfolio | 10.1% | 11.6% | −21.0% | 0.48 |
| S&P 500 | 16.1% | 19.1% | −33.7% | 0.48 |
| Last three years (Sept. 2023 – Sept. 2026) | ||||
| This strategy | 10.8% | 4.0% | −2.6% | 4.22 |
| 60/40 portfolio | 14.6% | 9.7% | −10.6% | 1.37 |
| S&P 500 | 22.2% | 15.3% | −18.8% | 1.19 |
| 2026 to date (annualized) | ||||
| This strategy | 14.3% | 5.2% | −1.7% | 8.58 |
| 60/40 portfolio | 7.9% | 9.0% | −6.0% | 1.31 |
| S&P 500 | 17.9% | 13.2% | −8.9% | 2.02 |
Several methodological parameters characterize the strategy's empirical profile. The framework prioritizes capital preservation, volatility containment, and strict drawdown bounds, producing strong compounding convexity across turbulent regimes alongside controlled participation during equity expansions. All reported metrics represent an out-of-sample walk-forward simulation across historical market data, incorporating realistic 6 bps execution friction. In addition, annualized 2026 metrics reflect a trailing nine-month evaluation window, providing empirical evidence of regime adaptability while operating within the broader eight-year historical cycle.
With the quantitative foundation and equilibrium framework established, subsequent publications in this research series—released on a monthly cadence by default, supplemented by event-driven monitors when major market dislocations or regime inflections occur—will leverage the portfolio's weekly equilibrium weights to observe and diagnose cross-asset market conditions: tracking regime inflection points, decoding yield curve repricing, and continuously benchmarking model signals against subsequent realized market outcomes. A macroeconomic barometer is valuable only if its regime signals are continuously benchmarked against evolving market realities.