Investment Categories in a High Debt-to-GDP World: A Fiscal Dominance Framework

If you turn on CNBC or open any financial website, you’ll be able to see how each sector in the stock market is performing. There’s eleven of them and they include: technology, healthcare, energy, etc. These sectors have been around for decades, and they organize stocks based on what they sell.

However, I believe there is a different way to organize stocks into categories that takes into account the fiscal situation of the US, inflation, M2 money supply growth, and interest rates.

Productivity Premium

Productivity premium companies create goods and services that help the rest of the economy become more efficient. These companies provide the foundation for other businesses to become more productive.

Currently, there are two tiers of productivity premium companies. They are chips and hardware companies and cloud and software infrastructure.

Tier 1A: Chips and Hardware (Physical Foundation)

Companies here design and manufacture the chips, processors, and semiconductor equipment that every digital system runs on. Everything relies on there being enough powerful hardware.

• Chip designers and manufacturers: NVIDIA, Broadcom, AMD, Intel, Micron,
• Semiconductor equipment: Lam Research, Applied Materials, KLA (these companies make the machines that make the chips)
• Physical infrastructure: Corning (fiber optics), Amphenol (connectors and cables)

Tier 1B: Cloud and Software Infrastructure

This tier sits on top of the chip layer and organizes raw computing power into tools that businesses can actually use. Without Tier 1A, Tier 1B has nothing to run on. Without Tier 1B, businesses have no way to access and deploy the computing power Tier 1A produces.

• Cloud platforms: Microsoft (Azure), Amazon (AWS), Alphabet (Google Cloud).
• Networking infrastructure: Cisco, Arista (the companies that physically connect everything together)
• Enterprise software and databases: Oracle, IBM, Dell, Salesforce

Liquidity Monetization

Companies in this category don’t really produce anything in the traditional sense. Instead, the position themselves inside the flow of money and collect a cut every time it exchanges hands, grows in value, or is managed by someone else.

As I mentioned in my fiscal dominance thesis, I believe interest rates will be suppressed (cheaper borrowing) and inflation and M2 money supply will increase even more.

Pure Volume Risk (lowest exposure)

These companies have almost no downside beyond transaction volume slowing down. Their revenue dips when people buy less, but nothing blows up on their balance sheet. No loans going bad or assets getting written down.

• Visa and Mastercard: Process transactions and collect fees. No lending, no credit risk, no principal at risk under any circumstances.
• S&P Global: Every bond issued needs a rating. Every index product uses their benchmarks. Revenue grows automatically as debt issuance grows.
• American Express: Similar to Visa and Mastercard on the transaction processing side, though they also have a lending component that adds some credit exposure.

Asset Price Risk (moderate exposure)

These companies manage other people’s money and charge a percentage of whatever it’s worth. When markets fall and asset prices drop, their fee base shrinks. But the losses belong to their clients, not to them. Their own balance sheet stays intact.

• BlackRock: Manages roughly $10 trillion in assets. As M2 (the broad money supply) expands and asset prices inflate in nominal terms, that number grows and so do BlackRock’s fees. A prolonged market crash shrinks their revenue but doesn’t destroy their balance sheet.
• Morgan Stanley and Goldman Sachs: Trading fees, advisory fees, and asset management revenue all scale with market activity and asset values.
• Charles Schwab and Interactive Brokers: Retail brokerage revenue grows with more accounts and higher asset values. Worth noting that Schwab also holds client cash in bonds, which created balance sheet stress in 2023 when rates rose rapidly.

Credit Risk (highest exposure)

These companies benefit from the same monetary tailwinds as the groups above, but they actually hold the loans. When borrowers can’t repay, it’s not just slower revenue. Assets on their own books get written off. That’s a fundamentally different kind of risk.

• JPMorgan, Bank of America, Wells Fargo, Citigroup: Traditional banks that hold consumer and commercial loans on their balance sheets. The 2008 financial crisis showed exactly what happens to this group when credit defaults spike.
• Blackstone (private credit division): Blackstone has moved heavily into private credit (direct lending to mid-market companies that can’t access public debt markets). That puts them squarely in credit risk territory for a growing portion of their business.

Government Spending

These companies derive a significant chunk of their revenue directly from the federal government. In a world where government spending keeps growing, that’s about as reliable a revenue stream as you can find.

The category breaks into three main groups.

Defense and Military

Lockheed Martin gets roughly 97% of its revenue from government contracts, almost entirely from the Department of Defense. RTX (formerly Raytheon) is similar at around 60%. These companies don’t really have a market in the traditional sense. You and I aren’t going to Lockheed Martin to purchase weapon systems.

Congress is their customer. Economic cycles barely touch them. Defense budgets are politically protected even in downturns. No modern politician is going to limit defense spending.

Think of it like this. The United States is the most powerful country on earth. They are going to do whatever it takes to keep their power. This means more spending on military and defense.

Examples: Lockheed Martin, RTX, Boeing, GE Aerospace, Honeywell

Medicare and Medicaid

This is the biggest category by dollar volume. UnitedHealth gets roughly 75% of its revenue from administering government health programs through Medicare Advantage (a version of Medicare run by private insurers) and Medicaid. CVS is similarly exposed through pharmacy benefits and insurance administration.

Then every major pharmaceutical company on the list gets a large portion of their drug revenue reimbursed through government health programs. Pfizer, Merck, AbbVie, Eli Lilly, Gilead, and others depend on Medicare Part D (the prescription drug program) and Medicaid for a significant share of what they actually collect. Medical device companies like Stryker, Abbott, and Intuitive Surgical flow through the same reimbursement system.

However, government pricing power cuts both ways. The same federal programs that guarantee revenue can also set the prices. The Biden administration’s $35 insulin price cap directly hit pharmaceutical revenue, which is something that can never happen to Visa or NVIDIA.

Government Technology

This group is growing fast and largely invisible to most investors. Palantir gets over half its revenue from intelligence community and military contracts. But Microsoft, Amazon, Oracle, and IBM are all competing for federal cloud infrastructure contracts worth hundreds of billions over coming decades. Once the government migrates its systems to your platform, it’s almost impossible to move them. That makes government tech revenue extraordinarily sticky.

The Real Economy

Real economy companies are the ones whose performance tracks how consumers are doing. These are companies in the consumer discretionary and consumer staples sectors. However, I like to split them into three categories: Needs, Wants, and Hybrids.

Needs

Companies in the Needs category sell products that we all need in order to survive in the modern world. Walmart, Costco, and Procter and Gamble fall here because people need groceries, cleaning products, and household staples whether the market is up or down.

Tobacco companies like Philip Morris represent an extreme version of this. Addiction essentially removes the consumer’s ability to cut back, making demand almost completely inelastic (unresponsive to price or economic conditions). Even in a severe recession, Philip Morris’s revenue barely moves.

Wants

These companies do well when the consumer has extra money to spend. When people need to tighten their budgets, these companies are on the chopping block.

Starbucks is a clear example. A $7 latte is an easy cut when money gets tight. It’s cheaper to buy coffee at the grocery store and brew it at home. Netflix, Walt Disney, Tesla, and Booking Holdings all sit here. These companies track consumer confidence closely, and their revenue tends to contract faster than the broader economy when households feel financially squeezed.

Also, there’s something interesting to note. Because we live in a K-shaped economy, the Wants companies might continue to perform well because the wealthy are able to spend enough to prop these companies up. So while the average American might cut back at Starbucks, the wealthy will continue getting their lattes. The question is what happens if/even the wealthy cut their spending?

Hybrids

These companies serve both a need version and a want version of the same business, depending on how much money consumers have.

Home Depot and Lowe’s are the textbook example (and the reason this category exists as its own group). When the economy is good, people have enough money to remodel their kitchens and bathrooms. That’s the want version. When money is tight, they still have to fix the broken water heater or patch the leaking roof. That’s the need version. The want side of the business contracts but the need side provides a floor.

In an environment where real wages (your paycheck’s actual purchasing power after inflation) are being squeezed, understanding whether a company serves needs or wants becomes more predictive than traditional financial metrics.

My analysis of wages versus inflation since 1972 shows how that squeeze has played out for American households over the past five decades.

Real/Scarce Assets

This category is about preservation. When the money supply keeps expanding and the purchasing power of a dollar keeps declining over time, some assets hold their value because they’re tied to something genuinely scarce (something you can’t just print more of).

There are two distinct types of scarcity that matter here.

Physical Scarcity

These are finite resources that the economy structurally depends on. Oil companies like Exxon, Chevron, and ConocoPhillips aren’t selling a consumer preference. They’re extracting something that factories, vehicles, planes, and power plants require regardless of what anyone prefers. Newmont mines gold, which has served as a monetary metal (a store of value tied to no government’s promise) for thousands of years across dozens of different economic regimes. Southern Copper produces copper, which is increasingly critical for electrification and infrastructure.

Productive land is another scarce asset. Companies like Welltower (healthcare properties, senior living) and Prologis (logistics and industrial warehouses) belong here.

The key dynamic with real assets is depletion and scarcity. Every barrel of oil pumped is one less in the ground. That creates a long-term floor on value that paper assets simply don’t have. You can print more dollars. You can’t print more copper or more land.

Attention Scarcity

This one requires a slightly different mental model. Human attention is also finite and scarce. There are only so many waking hours in a day. Companies that have built dominant infrastructure to capture those hours at scale are sitting on a resource that has real economic value.

Netflix, Walt Disney, Meta, and Alphabet (through YouTube and Google) have each built systems that capture enormous shares of human attention. These are infrastructure assets that compound over time and get harder to displace the longer they hold the user.

Unlike physical resources, attention scarcity renews every 24 hours. But the infrastructure built to capture it compounds over time the same way a physical asset does. A new oil company can drill a new well. A new streaming service has to somehow convince people to change their evening habits away from something they’ve been doing for years.

Bitcoin

Bitcoin deserves a mention here with an honest caveat. It was designed with mathematical scarcity built in (only 21 million will ever exist). Whether that scarcity translates into a reliable long-term store of value the way gold has is still genuinely unknown. The volatility is real. The regulatory uncertainty is real. But the scarcity argument is mathematically sound, and more institutions are treating it as a partial hedge against currency debasement (when a government inflates its currency, reducing its purchasing power). It belongs in the conversation with appropriate uncertainty attached.

For historical data on how gold and real assets have performed relative to equities and inflation, my S&P 500 sector performance versus gold and inflation is worth reading alongside this piece.

How to Use This Framework

A few things to keep in mind when applying these categories to your own thinking.

• Most companies fit cleanly in one category. But some span multiple. Amazon is productivity premium through AWS, liquidity monetization through its advertising and marketplace fees, and real economy through its retail operations. That overlap is part of why it’s so dominant.

• Sector labels and these categories often disagree. NVIDIA and Netflix are both ‘tech’ in traditional sector terms. But NVIDIA is productivity premium and Netflix is attention scarcity. They respond to inflation, recession, and government policy in completely different ways.

• The real economy category is your read on Main Street. When needs companies start struggling (not just wants), the financial stress on ordinary Americans is deep.

• Risk exposure matters as much as upside potential. All liquidity monetizers benefit from M2 expansion, but pure toll booths like Visa and credit-risk banks like JPMorgan carry very different downside scenarios. Understanding the difference changes how you think about what you own.

The Bottom Line

Sector labels were built for a simpler economy. They’re still useful as shorthand, but they don’t tell you what you actually need to know when you’re trying to understand how a company makes money and how it’ll hold up under different economic conditions.

The five categories do that job better. They force you to ask the right questions about any company before deciding it belongs in your portfolio. What creates or captures the value here? What does it depend on? What breaks it?

Those questions are worth a lot more than knowing which sector a company gets assigned to.

Frequently Asked Questions

Isn’t this just another way of saying sectors?

No. Sectors group companies by what they sell. These categories group them by how they make money and what conditions that depends on. Two companies in the same sector can be in completely different categories. Visa and JPMorgan are both ‘financials’ but they’re in different sub-tiers of Category 2 and carry completely different risks.

What if a company fits in more than one category?

That’s fine and fairly common. Amazon, Berkshire Hathaway, and a few other large conglomerates genuinely span multiple categories. The goal isn’t to force every company into a single box. It’s to understand which value streams are driving the business and what risks each one carries.

Can a company move between categories over time?

Yes, though it’s rare for core business models to shift dramatically. What changes more often is the balance between categories for diversified companies. Amazon’s revenue mix between AWS, advertising, and retail has shifted significantly over time, changing which categories are doing the most work in the business.

Does this framework apply to international stocks?

The same logic applies, but the specific dynamics differ. The categories are about economic function, not geography. That said, government spending categories are highly country-specific. Who the government is, what it spends on, and how much it dominates the economy varies enormously from country to country.