A way to read your financial life.
Modern wealth is stacked. Each layer is a claim on the layer beneath it. The base of the stack is what remains when every layer above depends on something else to perform.
The ordering is after John Exeter, a former Vice President of the Federal Reserve Bank of New York. It is not a product pitch and not a price forecast — it is a way to see plainly what each layer of a household’s wealth actually is, and how much of the stack rests on which foundation.
The ordering
A stack of claims, each resting on the one beneath it.
The four tiers
From capital appreciation at the top to base assets at the foundation.
How the layers interact
Correlation, efficiency, and the architectural case for a base layer.
Reading your own statement
How to read it against the portfolio you already have.
Where a reserve sits
The base layer, specifically — and what it does that nothing above it can.
On sizing and allocation
A factual ordering, not a prescription.
Frequently asked
The questions it invites.
A stack of claims, each resting on the one beneath it.
John Exeter proposed this ordering in the 1970s, during the unwind of the Bretton Woods monetary system and the inflation that followed. He had spent his career inside the Federal Reserve and the international banking system and was writing about what, in the end, financial wealth is actually made of.
His observation was simple and structural. The instruments that look the most like money — equities, bonds, derivatives, deposits — are each, on inspection, a claim on something else. A claim on a company. A claim on an issuer. A claim on a bank. A claim on the orderly functioning of a settlement system. The further up the stack you go, the more claims-on-claims are layered underneath the instrument you are holding.
At the base of the stack is the narrowest layer — the assets that are not claims on anyone at all. Historically: physical precious metals, directly held. This is not a claim that someone else must honor. It is a thing.
The practical use of the ordering is not to rank the layers from best to worst. It is to see what each layer depends on, and to make deliberate choices about how wide the base of the household’s stack should be given how much of the rest rests on it.
Read top to bottom.
Four layers, ordered from the widest and most dependent at the top to the narrowest and most elemental at the base. Each is presented the same way: what it is, what it is a claim on, what it does well, and where it fails.
The engine of long-run real returns.
Capital appreciation
Equities, private capital, venture, commodity funds, real-estate vehicles, derivatives
Claims on the future performance of enterprises, operators, and markets. The widest layer of modern wealth — and the primary engine of long-run real returns. It is also the layer that depends on every layer beneath it functioning.
- What it is a claim on
- Corporate performance, operator skill, market valuation, custodial integrity, settlement infrastructure, and the continued function of the legal and regulatory system that enforces ownership.
- What it does well
- Long-run wealth creation above the rate of nominal growth. Compound return on productive capital. The layer that makes portfolios grow.
- Where it fails
- Mark-to-market volatility that can force poorly-timed decisions. Custodian, broker, and settlement risk. Sensitivity to market-structure events. Correlated drawdowns across the equity complex in dislocation.
Predictable cashflow with credit risk.
Income & stability
Government bonds, investment-grade credit, municipal debt, annuity claims
Fixed claims on issuers — sovereigns, corporations, municipalities, insurers. A layer built for predictable cashflow, lower volatility, and duration management. Still a paper claim on a counterparty whose willingness and ability to pay must hold.
- What it is a claim on
- The creditworthiness of the issuer, the stability of the currency of denomination, and the rate environment that reprices the claim day-to-day.
- What it does well
- Predictable income. Lower drawdowns than equities. Duration-matching of liabilities. Ballast during equity dislocation.
- Where it fails
- Purchasing-power erosion under inflation. Credit events — sovereign, corporate, municipal. Rising-rate capital losses. A layer that can underperform in real terms for a generation without ever defaulting.
The operating layer of the household.
Liquidity
Cash, money-market funds, bank deposits, short-dated treasury bills
The operating layer — immediate purchasing power held as claims on financial institutions. Essential for household function, rebalancing, and emergency reserves. Still a claim, subject to the solvency and settlement of the institutions holding it.
- What it is a claim on
- The solvency of the holding institution (bank, fund, custodian), the function of deposit insurance and money-market backstops, and the operational continuity of the payments system.
- What it does well
- Immediate purchasing power. Operational continuity. Optionality to act. Emergency reserve.
- Where it fails
- Real-terms preservation over multi-year horizons. Institutional counterparty risk (bank failures, money-fund breaks). Silent erosion under sustained inflation.
The foundation. Owned outright, not anyone else’s liability.
Base assets
Physical precious metals — directly titled, physically held, documented
The counterparty-risk-free layer. Owned outright, titled directly, held physically. Nothing else in the stack has to perform for this layer to continue to exist. Historically narrow in allocation; structurally the base on which the rest rests.
- What it is a claim on
- Nothing. The metal is the claim and the thing at once. No issuer, no custodial chain of promises, no settlement step outstanding.
- What it does well
- Preservation outside the paper layer. Continuity through counterparty and market-structure dislocation. Multi-generational holdability. A planned exit that does not depend on a single venue or counterparty.
- Where it fails
- Intraday liquidity at institutional scale. Short-term price volatility. No yield, no dividends, no compound cashflow. A layer that is measured in decades, not quarters.
The math of the stack.
Reading the stack as a set of layered claims is not only a structural observation — it has a measurable consequence. Because the top three layers are claims on the continued function of the same counterparty and monetary system, they tend to move together when that system is stressed. Equities, bonds, and cash-like instruments exhibit elevated correlations in exactly the environments that diversification is meant to protect against.
The base layer is structurally different. It is not a claim on the system at all. That independence shows up in the data as persistent low correlation to the paper stack — and low correlation, in portfolio terms, is the mechanism by which a narrow allocation does disproportionate architectural work.
Across long measurement windows, the directly-held base layer’s correlation to broad U.S. equities has sat near zero, and its correlation to investment-grade bonds has been similarly low or mildly negative. This is a repeated finding across data providers, time windows, and methodologies — published by the World Gold Council, academic researchers in portfolio theory, and institutional consultants. The structural reason is visible in the ordering itself: the base layer and the paper layers are not claims on the same counterparty and monetary system.
A small, directly-held allocation of precious metals has repeatedly been shown to move the efficient frontier of a diversified portfolio up and to the left — higher risk-adjusted return at the same risk, or equivalent return at lower risk. Ibbotson Associates’ analysis of 1974–2012 data and the World Gold Council’s successive “Gold as a Strategic Asset” studies have reached this conclusion across decades of out-of-sample updates, using allocations typically in the 2% to 10% range.
The improvement in risk-adjusted return from adding a base-layer allocation is not a bet on metal outperforming. It is the mathematical consequence of low correlation. Across long-horizon Sharpe-ratio studies the lift has been directionally consistent — modest in magnitude, robust to methodology, and present whether the look-back window begins in the 1970s, the 1980s, or the 2000s. The finding that repeats is architectural: a narrow allocation of a non-paper asset quietly widens the base of the stack.
These studies do not forecast future returns. They do not promise that gold will outperform equities, that silver will outperform gold, or that the next decade will resemble the last. They describe what the historical correlation structure has done for portfolio efficiency when a base-layer allocation is held. Everything a household does with that observation belongs in its own brief — not a page on a website.
Correlation and Sharpe-ratio analyses are architectural evidence, not price forecasts. They describe the mathematical behavior of the layers — not the direction of the next quarter, the next year, or the next decade.
Read it against what you already have.
It is only useful when applied to a specific financial life. Three common patterns, seen through the stack.
Equities, bond funds, cash. Reading from the top of the stack: T4, T3, T2. The base layer — T1 — is often absent. The portfolio is diversified across asset classes but held entirely as paper claims on counterparties in the layers above the base.
Public equities, private capital, municipal bonds, cash, and — sometimes — an allocation in gold ETFs or commodities funds. The ETF sleeve lives in T4 alongside other equity-like exposure. The base layer may still be absent in any directly-held, counterparty-risk-free sense.
The same set of layers above, plus a directly titled physical reserve — bar-listed, depository-receipted, insured, documented. T4 through T2 continue to do the jobs they do. T1 does the job none of them can: it sits as the counterparty-risk-free foundation under the rest.
None of these patterns is universally right or wrong. The question the ordering asks is narrower: across the four layers, which ones the household is holding, and under what structure.
A portfolio is what you have exposure to. A reserve is what you own.
The base layer, specifically.
A physical reserve is specifically a T1 instrument. Directly titled precious metals, held at a chosen custody venue, documented end-to-end. The metal is the claim and the thing. No issuer, no trust, no fund, no custodial chain of promises between the owner and the asset.
What the base layer does in the stack is not return maximization — that is the work of T4. It is continuity. It is the layer that keeps functioning when the layers above are disrupted — by counterparty events, by market-structure dislocation, by currency instability, or by the slow erosion that a generation of inflation can produce in paper claims.
The practical consequence is architectural. A household with a deliberate base layer is holding the rest of its stack on a foundation that is not itself a claim on anyone. A household without one is holding the stack entirely as layered claims on counterparties above the base. That is a choice. The ordering makes the choice visible.
Continuity across dislocation.
The layer that continues to function when the layers above are impaired — counterparty events, settlement disruptions, extraordinary monetary policy, and long-cycle currency instability.
Multi-generational holdability.
An asset that can be documented, titled, insured, and transferred across generations without depending on the ongoing function of a fund, a trust, a broker, or a custodian chain that may not outlive the holder.
A planned exit, not an improvised one.
A reserve that cannot be drawn down on a planned timeline is not a reserve. Exit counterparties, spreads, and timelines are set at the start — not improvised at the moment of need.
A foundation, not a substitute.
The reserve does not replace the equity layer, the bond layer, or the liquidity layer. It is the layer those layers rest on.
An ordering, not a prescription.
Specific allocation figures are a question for the brief, written against a specific situation. At the conceptual level, four ideas are worth naming in advance.
A reserve is sized for its job, not the market
The base layer is structural, not tactical. It is not sized against a return target, a price forecast, or a sector call. It is sized against what the household wants the foundation to look like across the multi-decade horizon the reserve is built for.
The historical range is narrow
Across long horizons and across household types, directly-held precious-metals allocations have historically lived in a single-digit to low-double-digit percentage of total assets. The point is not the percentage — the point is that the base does not need to be large to do its job. It needs to be real.
The layer is not a replacement
A reserve is not a substitute for equities, for bonds, or for cash. It is the layer beneath them. Sizing discussions that pit T1 against T4 state the question wrong; the four layers do different jobs.
This is a question for the brief
Specific sizing for your situation belongs in the Physical Reserve Strategy Brief, written against your allocation context, timeline, and custody constraints. It is not a question this page answers.
The questions it invites.
- Is the capital stack an investment strategy?
- No. It is a way of reading how financial wealth is actually structured — as a set of layered claims resting on each other. Exeter proposed it as a way to think about counterparty-risk and monetary stability, not as a trading model. It is useful because it names what the layers actually are; what you do with that reading is a portfolio conversation, not a page on a website.
- Does this mean physical metal is always better than paper claims?
- No. Each layer does a different job. Equities generate long-run real returns; bonds provide predictable cashflow; cash provides immediate function; physical metal provides counterparty-risk-free continuity. A well-built financial life uses all four. Arguing that one layer is universally better than another is the same mistake as arguing that the roof is better than the foundation.
- Isn’t a gold ETF effectively the same as the base layer?
- The ordering answers this directly. A gold ETF is a fractional claim on a trust that holds metal at a custodian. It is structurally an instrument in T4 — a paper claim on a counterparty chain — even though its price is linked to the metal. The ETF is useful for price exposure inside a brokerage account. It is not the base layer in the architectural sense Exeter meant.
- Why precious metals specifically, and not other tangible assets?
- Real estate, collectibles, and private businesses are tangible but not fungible, not globally priced, and not portable in the way base-layer assets need to be. Physical precious metals — gold and silver in particular — have a multi-millennia history of functioning as the narrowest layer of the stack: fungible, portable, globally priced, outside any single issuer, and not a claim on anyone else’s performance.
- What does Hard Asset Reserve actually do within this ordering?
- The firm is a private-client office for building the base layer deliberately — the Physical Reserve Strategy Brief, custody architecture, titling and documentation, acquisition and logistics, and ongoing review. The ordering on this page is conceptual; the brief is where it becomes specific to a household’s actual situation.
The base does not need to be large to do its job. It needs to be real.
The base is missing until it is built.
The ordering on this page is conceptual. The Physical Reserve Strategy Brief is where it becomes specific to your situation — allocation context, custody architecture, and the tradeoffs that apply to the household holding the stack. Every month without the layer is a month your foundation is missing.
The metal is yours — not a fund’s, not a claim on any counterparty.
Reviewed brief delivered in five business days of intake. The engagement structure is named in the brief — you proceed only if both fit your situation.
The Office accepts a small number of new engagements each quarter. Selection is by considered fit, not by pace of inbound.
The capital stack is a conceptual ordering after John Exeter, former Vice President of the Federal Reserve Bank of New York. It is not a performance claim, not a forecast, and not individualized investment advice.