Q3 Concentration: Equity HHI Hits 4,150 — Mortgage Takes 70% of Household Debt

Our 2026 concentration print answered the first distribution question: how thick is the top across net worth, equities, deposits, revolving balances, and card issuers — top-1 wealth near 30.5%, top-1 equities near 54%, revolving inverted so the bottom half holds about 30% of card debt. The Q3 theme update then refreshed the tape: NY Fed 2026Q2 household debt at $18.926T, July personal saving at 3.1%, ICI money-market funds revised to $8.02T. This Q3 concentration lens asks the sharper follow-up desks actually trade: how concentrated is the system when you score HHI across ledgers, split the liability stack by product, and watch the vintage slope of the tip?

The interactive dashboard above is built as a Q3 concentration lens. Toggle HHI map, Wealth ladder, Debt products, and Vintage slope. The punchline is deliberately three-sided. On people-side equities, analytical HHI prints near 4,150 with top-1 about 54% and top-10 about 87%. On liability products, mortgage alone is 70.1% of the $18.926T NY Fed stock — top-3 products (mortgage + auto + student) clear about 87.9%, for a product HHI near 5,080. On card issuers, top-1 / top-3 purchase volume sits near 22% / 52% (firm HHI ~1,962) — concentrated relative to an equal split, but mild next to equity ownership. Liquid parking still favors deposits (~65% of the deposits+MMF pair) even with MMF at $8.02T.

The Q3 concentration scoreboard

Read the table as a family of market shares, not one number. Equity HHI tells you who captures market beta. Debt-product HHI tells you how mortgage-dominated the liability ledger is — a concentration story that is orthogonal to who holds revolving stress. Issuer HHI tells you which brands intermediate spend. Analysts who quote only aggregate net worth understate the equity tip; analysts who quote only card delinquency understate that 70% of household debt is still a housing product with a different stress clock.

HHI map: equities and mortgages dominate different axes

Filter the dashboard to HHI map. Ranking analytical HHI on stated bucket shares puts debt products first (~5,080) and equities second (~4,150), then revolving, wealth, deposits, and issuers. That ordering is the Q3 contribution relative to the prior concentration print: we now score how concentrated each lens is on a common 0–10,000 index, not only top-1 / top-10 percentages.

The scatter panel plots each lens’s top-1 share against its thick top. Equities sit upper-right on the people cloud. Debt products sit even further right on the tip axis because mortgage alone is 70%. Revolving sits low on top-1 (about 5%) even though its HHI is mid-range — the mass lives in the 50th–90th and bottom-50 bands. Issuers form a firm-side cluster near 22% / 52%. Same theme, three geometries.

Toggle the asset-lens bars to Top 1%, Top 10%, or Bottom 50%. Sorting by bottom-50 flips the story: revolving and real estate look “democratic”; equities do not. The save-vs-borrow grouped bars make the mirror explicit — top 1% holds about 54% of equities and only about 5% of revolving; bottom 50% holds about 1% of equities and about 30% of revolving.

Wealth ladder: thick top, living middle, thin bottom

Switch to Wealth ladder. The Lorenz-style curve for net worth rises to about 30.5% at the top 1%, 67.5% at the top 10%, and 97.5% once the upper half is included — leaving the bottom 50% with roughly 2.5% of net worth, or about $4T against a ~$169T aggregate. Toggle the curve to Debt products and the shape changes: top-1 (mortgage) already clears 70%, top-3 clears ~88%. That is a steeper tip than the wealth Lorenz — and it answers a different question (product mix, not people ranks).

The ranked wealth bars in share or $ trillions make the same people-side point. The top 1% holds on the order of $52T; the 90th–99th another ~$63T; the 50th–90th about $51T; the bottom half a thin residual. Housing equity and retirement accounts thicken the middle; private business and publicly traded equities thicken the tip. Pair this with July’s 3.1% personal saving rate from the Q3 update: a thin flow does not rebuild the bottom half’s stock on any policy-relevant clock when assets are already this skewed.

Debt products: mortgage is the liability top-1

Open Debt products. On NY Fed 2026Q2 disclosed stocks, mortgage is $13.268T (70.1%), auto $1.702T (9.0%), student $1.661T (8.8%), cards $1.281T (6.8%), HELOC $0.458T (2.4%), other $0.556T (2.9%). Top-3 products clear about 87.9%. Product HHI near 5,080 is the loudest single concentration meter in this Q3 lens — louder than people-side equity HHI — because a single product dominates the stack.

That matters for stress narratives. Card 90+ and early transitions live in a 6.8% sleeve of total household debt. Student 90+ stock near 10.6% lives in an 8.8% sleeve. Mortgage serious transitions near 1.6% live in the 70% sleeve. Our delinquency split already showed product clocks disagree; the concentration lens shows why aggregate “household debt stress” can look calm while revolving and student prints scream — the calm product is also the giant.

The liquid-parking donut keeps cash allocation honest: deposits still about 65% of the deposits+MMF pair against ICI $8.02T MMF. Our money market funds vs deposits piece tracked the product split; here the point is that even after the hiking-cycle MMF boom, the household liquid pair is not MMF-dominated. Issuer purchase-volume remains a separate firm-side story: top-3 near 52%, HHI ~1,962 — moderate next to either equity ownership or mortgage’s product share.

Vintage slope: the tip is sticky, not exploding

Toggle Vintage slope. DFA-consistent rounds from 2019 through the 2025–26 window put top-1 wealth near the high-20s to low-30s and top-1 equity ownership near the low- to mid-50s. The Q3 print does not show a sudden breakout in tip shares; it shows a sticky thick tip that coexists with tape moves in debt stock, saving flow, and MMF AUM. Toggle among top-1 equity, top-1 wealth, and top-10 wealth to see that equity remains the steeper people-side series across vintages.

That is the methodological lesson. Concentration can be “high and stable” while aggregates still move. $18.926T of household debt (+$132B QoQ) and $8.02T of MMF are tape facts; 54% top-1 equity and 70% mortgage-of-debt are distribution facts. Desks that only refresh the tape miss the distribution; desks that only cite the tip miss the quarterly delta.

Who is exposed — and what would change the story

Exposed: households in the bottom half and lower middle whose balance sheets are revolving-heavy and equity-light when the APR−funds gap stays near 17 pp; regional banks and fintechs competing for purchase volume against a top-5 that already clears ~71%; mortgage-centric portfolios if serious transitions keep drifting up from 1.6% even while revolving stress steals headlines; desks that underwrite “households are fine” from Z.1 net worth alone while the bottom 50% holds ~2.5% of that stock.

Relative winners under current rules: top-percentile owners of corporate equities and private business; deposit-rich households capturing elevated cash yields without carrying revolving stress; large card issuers with scale in purchase volume; mortgage holders with locked-in rates whose primary residence still participates in the less-skewed housing ledger — even as mortgage product concentration dominates the liability stack.

What would change the story: a sustained multi-year rise in bottom-50 wealth share above the low-single-digit band; equity ownership diffusion that breaks the ~54% top-1 / ~87% top-10 prints; a material shift in debt-product mix away from mortgage’s ~70% without a matching stress spike; or a collapse of revolving balances in the middle and bottom that would also show up as healing in card 90+ transitions. None of those dominate the 2025–2026 DFA-style map and 2026Q2 product split summarised here.

Caveats and methodology

• HHI values are analytical indexes on stated bucket shares (0–10,000), not Fed supervisory HHI cells. They are comparable within this dashboard, not against banking-market HHI thresholds without translation.
• DFA percentiles are wealth ranks, not income ranks. A high-income renter can sit below a lower-income homeowner on net worth.
• Debt-product shares are disclosed NY Fed 2026Q2 stocks divided by $18.926T aggregateproduct concentration, not people concentration.
• Revolving percentile shares remain estimated to sum to aggregate revolving and to reflect the qualitative SCF/DFA pattern; treat as order-of-magnitude.
• Issuer purchase-volume ranks follow public Nilson-style tallies and can shift with co-brand mix and network definition.
• MMF AUM ($8.02T) is the ICI week-ended Aug 19 revised print; deposits (~$14.95T) are theme-carried. Fund-level institutional/retail mix is not the same as household DFA deposit ownership.
• Vintage slope points are DFA-consistent rounds for visualizationnot a substitute for the full interactive release tables.
• Aggregate net worth (~$169T) is carried theme contextthe concentration story is about shares and HHI, not a new Z.1 reprint.

The shareable takeaway

Consumer finance in Q3 is concentrated at the top — but which top depends on the meter. On equities, people-side HHI near 4,150 with top-1 about 54%. On debt products, mortgage alone is 70.1% of $18.93T household debt (product HHI ~5,080). On net worth, top 1% holds about 30.5% and the bottom half about 2.5%. On card issuers, top-3 purchase volume sits near 52%. On liquid parking, deposits still dominate the pair even with MMF at $8.02T. Households are not one balance sheet. They are a stack of distributions — people, products, and firms — and stress, savings, and market beta live in different parts of that stack.

Related reading: 2026 concentration print, Q3 theme update, research roll-up, and household delinquency split.