Two Economies, One Paycheck: How AI Will Bend the Red Lines

Written By Adrian Maharaj

Written by Adrian Maharaj

(Views mine, not Google’s.)

Summary (for skimmers):
The viral CPI chart compiled by economist Mark J. Perry (from Bureau of Labor Statistics data) is real, but incomplete. Prices fall where distribution is open, supply is modular, and expertise can be encoded. Prices rise where permission, credentials, and local scarcity gate supply. AI is about to pry open the expensive sectors unevenly through new payment models, automation wedges, and audit friendly workflows. Operators who design for the payer, the permit, and the proof will win.

The Chart Everyone’s Sharing and the Question It Can’t Answer

You’ve seen it: TVs, software, and cell service down and to the right; hospital services, college tuition, childcare, and housing rocketing upward. The household whiplash is familiar: My phone gets cheaper every year. Why does daycare cost more than rent?

That picture is a system map, not just a price story.

  • Blue (deflation) = permissionless experimentation + global distribution + silicon scale productivity.

  • Red (inflation) = licensing, credential monopolies, payer bottlenecks, liability, and land or local vetoes places where trust and permission are the product.

Two economies. One paycheck.

(It resurfaced most recently in a viral Marc podcast clip, sparking another wave of shares.)

Where the Meme Misleads (and Why It Still Matters)

  • Hedonic adjustments make “–98% TVs” mean price per quality unit fell, not that OLEDs cost $20. The lesson still holds: when quality leaps faster than cost, the effective price collapses.

  • Baumol’s cost disease isn’t corporate villainy. If a nurse’s productivity can’t compound like a GPU’s, wages must still rise to compete for talent pushing service prices up.

  • Averages hide the squeeze. Wage lines track CPI overall, but median workers buying non-optional red sector services feel poorer even as their apps get cheaper.

The graphic is the scoreboard; governance is the game.

The Greedflation Layer

Beyond regulation and productivity, we need to name the elephant: market power. In concentrated industries, firms raise prices not just to cover rising costs but because they can. Analysts call this “greedflation.” Airlines, hospitals, and even food processors have used informational asymmetries, supply bottlenecks, and consumer inelasticity to push prices above what costs alone would justify.

The result: inflation persistence in red line sectors even when input costs stabilize. AI’s role here will be less about making services cheaper and more about reducing opacity auditable pricing, transparent procurement, and real-time benchmarks that weaken the information advantage incumbents hold.

Shrinkflation—The Invisible Tax

While the Perry chart focuses on official CPI, consumers feel inflation in subtler ways: smaller packages, same price. Shrinkflation has turned 16 ounce bags into 13 ounce bags while the sticker stays constant. It doesn’t show up as a price spike but erodes household budgets just as effectively.

AI supply chain monitoring and real time consumer advocacy platforms could expose shrinkflation in ways regulators have struggled to. But the broader point is clear: not all inflation shows up in the chart lines. Sometimes it hides in the denominator.

Main Street Matters Too

It’s easy to see the Perry chart as an abstract macro story. But for small and mid-sized businesses, these lines show up in the P&L:

  • Healthcare premiums keep climbing faster than revenue, eating into hiring budgets.

  • Childcare costs hit retention employees leave or demand higher wages just to cover family expenses.

  • Housing costs affect labor mobility talent can’t afford to move closer to work.

  • Education & training costs slow upskilling workers can’t easily acquire new skills without debt.

For SMB operators, these “red lines” are not theoretical they’re constraints on growth, margin, and workforce stability.

The opportunity: AI deflation can be leveraged at the SMB level before it reshapes national CPI.

  • AI copilots to cut insurance claims processing or HR paperwork.

  • AI-native training modules that lower employee learning costs.

  • AI-driven scheduling/compliance that lowers admin burden in childcare, healthcare, or local services.

Translation: You don’t need to wait for Washington or Wall Street. Main Street operators who adopt deflationary tools early will see margin gains long before the macro chart bends.

A Systems Lens: The Permission Stack

For any sector, score this Permission Stack from 0–5 each (higher = more locked-in, more inflation-prone):

  1. Licensure & credential monopolies

  2. Payment gatekeepers (insurers, Title IV, subsidies)

  3. Local veto points (zoning, boards)

  4. Liability surface (malpractice, recalls)

  5. Non-substitutability (switching costs in emergencies)

Rule: AI penetrates where the Stack ≤ 10 and there’s a clean data → decision → documentation loop. It stalls where Stack ≥ 15 unless a payer or regulator sponsors the change.

Where AI Bends the Red Lines First

Healthcare – from encounters to continuous care

  • Wedge: ambient documentation, imaging pre reads, payer-aligned triage, continuous monitoring.

  • Prediction: >30% of first-touch care is AI-routed by 2028; cost per resolved episode drops 25–40%.

Education – competency beats credentials

  • Wedge: AI tutors, mastery transcripts, portfolio verification.

  • Prediction: Skills-first hiring pipelines go mainstream; tuition growth flattens in non-elite programs.

Housing – soft costs deflate, dirt doesn’t

  • Wedge: generative BIM, permitting copilots, prefab, automated scheduling.

  • Prediction: Process costs fall 10–20%; total prices flatten only with parallel zoning reform.

Childcare productivity without breaking ratios

  • Wedge: AI scheduling, compliance automation, roster optimization.

  • Prediction: Prices plateau; capacity per site rises through employer sponsored and co-op models.

Food & essentials – logistics eats inflation

  • Wedge: autonomous restock, demand forecasting, robotics.

  • Prediction: Private label share grows; operational excellence beats brand spend.

(See below for a list of data sources to generate predictions.)

The Operator Playbook: Build for Payer, Permit, Proof

  1. Sell to the payer adoption follows the invoice.

  2. Productize outcomes 22 days to permit” beats “AI platform.

  3. Design for audits the audit trail is the product in high-permission zones.

  4. Localize the politics budget for coalition-building, not just sprints.

  5. Instrument the work deflation comes from collapsing swivel chair labor.

The Real Takeaway

Technology doesn’t automatically collapse prices. Distribution permission + trust mechanics + market power decide where deflation shows up.

The next decade rewards operators who route around bottlenecks with payer aligned economics, audit ready systems, and guaranteed outcomes.

If you’re building in these plays payer first healthcare, mastery hiring, permit-through framing construction, employer-sponsored childcare let’s talk. The chart is the headline; the work is bending the lines.

Evidence & Sources

Primary Data Source

  • Mark J. Perry, American Enterprise Institute – original compiler of the CPI chart from BLS data. (Updates and archives here: AEI – Perry CPI Charts)

  • Bureau of Labor Statistics (BLS) – Consumer Price Index (CPI) datasets. (BLS CPI Home)

Methodology References

Recent Economic Context

Prediction & Supporting Evidence

Adrian Maharaj
Previous

Agents Don’t Win. Rails Do.
Next

Incident Command for Revenue