Twenty Quarters Up: Senior Housing's Supply Cliff and the Wave That Cannot Pay

• 

  Low-rise senior living residences on a quiet residential street

Occupancy has risen for five straight years into the thinnest construction pipeline since 2012. The demographic wave is real and it is enormous. Most of it cannot afford the rent.

The Buckingham sits on a quiet stretch of Houston's River Oaks, 495 units of independent living, assisted living and skilled nursing behind a brick facade that suggests permanence. In November 2025 it filed for Chapter 11 protection. It was the second time in four years.

The first filing, in 2021, was the pandemic's doing, and the reorganization plan that came out of it rested on a reasonable assumption: occupancy would recover as the country reopened, and the community would grow back into its debt.

The recovery came. Nationally, senior housing occupancy has now risen for twenty consecutive quarters. The Buckingham filed anyway.

That is the puzzle worth sitting with, because on the surface American senior housing in 2026 has the most favorable supply-and-demand arithmetic in commercial real estate. Occupancy is at a record. New construction is at a fourteen-year low. The oldest baby boomers turned eighty this year, which means the demographic wave the industry has been promised since the 1990s has finally started to break.

And operators keep failing. Thirteen senior care companies filed Chapter 11 in 2025, up from eleven the year before, and in the first quarter senior care led every other healthcare sector in filings, outpacing pharmaceuticals for the first time since 2021.

The reconciliation is not complicated, but it is uncomfortable. The industry's occupancy record is mostly a supply-side artifact. And the demand wave everyone is counting on is arriving with less money than the sector's pricing assumes.

The number everyone quotes

Start with the good news, because it is genuinely good.

Senior housing occupancy across NIC MAP's thirty-one primary markets reached 89.9 percent in the second quarter of 2026, up four-tenths of a point from the first quarter and marking a twentieth consecutive quarterly gain. Occupied units hit a record 639,650. Independent living reached 91.3 percent and assisted living 88.4 percent, narrowing the gap between the two segments to 2.9 points, the tightest since 2014.

The market-level detail is better still. Fifteen of the thirty-one primary markets were at or above 90 percent occupancy, triple the number just three quarters earlier. San Francisco hit an all-time high of 92.7 percent. Chicago reached 90.7 percent and Kansas City 90.5 percent, both records. Boston led the country at 93.3 percent, with Baltimore near 91.8 percent.

The laggards are instructive in a different way. Miami sat at 86.2 percent, Atlanta at 86.5 percent, San Antonio and Las Vegas at 87.0 percent, all Sun Belt markets that absorbed the heaviest development of the 2015 to 2019 cycle and are still digesting it.

Any way you present it, that is a strong national picture. The question is what produced it.

The decomposition

NIC's own language points at the answer without quite naming it. The organization describes the streak as consecutive quarters in which absorption exceeded inventory growth.

That is true. It is also two variables, and only one of them moved.

Absorption has been steady and healthy, running at a 3.5 to 4.5 percent annualized rate, with first-quarter 2026 absorption up about 8 percent year over year and roughly 3,700 occupied units added in the second quarter. Good numbers. Not extraordinary ones, and not accelerating dramatically.

Inventory is where the story is. Year-over-year inventory growth was 0.4 percent in both the first and second quarters of 2026, the fifth straight quarter below 1.0 percent. Independent living inventory grew 0.5 percent against a historical average near 1.5 percent. Assisted living grew 0.3 percent against a historical average above 3 percent.

Look at it in units and the collapse is unmistakable. Rolling four-quarter net inventory growth fell below 3,000 units as of the first quarter of 2026, the lowest reading in NIC MAP's twenty-year history, against more than 21,000 units at the first-quarter 2020 peak. That is an 85 percent reduction in net new supply.

The forward pipeline confirms it rather than contradicting it. Fewer than 16,000 units were under construction as of the second quarter of 2026, the lowest since 2012 and roughly 2.3 percent of existing inventory. Quarterly construction starts in primary markets fell below 2,000 units in three of the last four quarters, the first time that has happened since 2011; the third quarter of 2025 produced 1,076 starts, the weakest quarterly figure since the second quarter of 2009. Nearly 60 percent of the 140 markets NIC tracks had no active senior housing construction at all. San Jose was the only primary market with more than 10 percent of its inventory under construction.

So: occupancy is at a record because the denominator stopped growing, not because move-ins surged. Both facts are worth celebrating if you already own a stabilized building. Only one of them tells you anything about demand.

The map: where the cliff bites hardest

A national occupancy figure of 89.9 percent is close to useless for underwriting a specific site, because the spread between the strongest and weakest primary markets is more than seven points, and the two ends of that range got there by opposite routes.

Market	Q2 2026 occupancy	What produced it
Boston	93.3%	Entitlement-constrained; almost no new supply in a decade
San Francisco	92.7% (all-time high)	Land cost and approval friction; construction effectively absent
Baltimore	~91.8%	Older stock, limited pipeline, stable senior population
Chicago	90.7% (record)	Slow inventory growth against a large 80+ base
Kansas City	90.5% (record)	Modest development cycle; no overbuild to digest
Las Vegas	87.0%	2015-2019 development wave still absorbing
San Antonio	87.0%	Sun Belt supply cycle; competitive rate environment
Atlanta	86.5%	Heaviest per-capita development of the last cycle
Miami	86.2%	Overbuild plus acute affordability compression

The pattern echoes what is happening in other operator-intensive asset classes. Markets with high land costs, difficult entitlements and slow permitting have effectively had a supply moratorium imposed on them by process, and they are being rewarded for it with occupancy in the low nineties. Markets that were easy to build in during the last cycle are still working through the consequence.

Two details in that table matter more than the ranking itself.

The first is that fifteen of thirty-one primary markets now sit at or above 90 percent occupancy, up from five just three quarters earlier. That is a very fast tightening, and it is not driven by a burst of move-ins. It is what happens when a nearly flat inventory line meets steady absorption for five straight years: markets cross the threshold one after another, on a schedule set by how much surplus they started with.

The second is San Jose, the only primary market with more than 10 percent of its inventory under construction. In a sector averaging 2.3 percent, that is an outlier by a factor of four, and it is happening in one of the most expensive construction markets in the country, which tells you the projects penciling today are the ones with rents high enough to absorb a $364-per-square-foot cost base. That is not a middle-market solution. It is the opposite of one.

For anyone sizing a deal, the practical consequence is that the national numbers should be treated as background, not evidence. A community underwritten to 89.9 percent stabilized occupancy in Miami is underwritten to a figure the metro has not reached. A community underwritten to the same number in Boston is leaving three and a half points of upside on the table, and, more importantly, mispricing the rate growth that comes with a market operating at effective full occupancy.

The Sun Belt markets at the bottom of that table are also where the last cycle's construction lending concentrated, which is worth remembering when the current pipeline eventually turns. The metros easiest to build in are the ones that overbuild, and the correction takes the better part of a decade to clear. Atlanta and Miami are seven years past the peak of their development wave and still trailing the national average by three points.

Why nobody is building

The industry's preferred explanation for the construction drought is capital discipline, that lenders got cautious after the pandemic and have not come back. That is partly right and mostly incomplete. The more precise answer is that the arithmetic stopped working.

CBRE's seventeenth Senior Housing and Care Investor Survey, covering thirty-six projects with data through the second quarter of 2026, puts all-in development cost at $388,830 per revenue unit, or $364 per square foot, up 23.6 percent since 2023. Hard costs account for 72.5 percent of that, soft costs 16.2 percent, site acquisition 8.1 percent (a range of roughly $16,000 to $36,600 per unit), and furniture, fixtures and equipment $11,900 per unit.

One line in that survey deserves more attention than it gets: rentable area has fallen to 55.5 percent of gross building area, down from 59.1 percent in 2023. Buildings are becoming less efficient as care space, commercial kitchens and amenity programming expand. Developers are paying for more square footage per unit of revenue than they were three years ago, on top of paying more per square foot.

In core markets the number runs higher still. Welltower's chief executive, Shankh Mitra, has cited roughly $450,000 per unit all-in.

Against that, the return. CBRE puts return on cost for new development at 8.1 percent, down from 8.2 percent in 2023, while stabilized cap rates have risen to 7 percent from 6 percent over the same period. That is a 110-basis-point development spread, for a project that now takes about 29 months to build, up from 21 months in 2017, before lease-up even begins.

No rational developer takes 110 basis points for four to six years of construction and absorption risk when the same capital buys a stabilized, cash-flowing community at a 7 cap on day one. NIC's head of research, Lisa McCracken, put the consequence plainly: the bottleneck is largely on the capital side, not from lack of demand, and investors favor acquiring existing properties over new construction, which puts pressure on the availability of senior housing for the consumer.

There is a second claim on that capital. Nearly $40 billion of senior housing loans mature between 2026 and 2030, just under $8 billion in 2026 alone, rising to $10 billion to $12 billion annually thereafter. Money that might have funded new communities is going to defend existing basis instead.

What the loan tape shows

Public REIT disclosures and NIC MAP describe the institutional end of this market. The independent operator borrows somewhere else, and that record is public.

Analytics.loan's analysis of SBA 7(a) loan-level data identifies approximately $313.3 million in lending to assisted living operators across 257 loans in 2025, at an average loan size near $1.2 million. For context, the all-program 7(a) average runs closer to $479,000. These are real-estate-inclusive deals, not working-capital lines.

Eighty-five lenders participated, but the book tilts hard toward specialists. Live Oak Banking Company wrote $72.9 million across 42 loans, followed by Pinnacle Bank at $21.8 million and Bank of Hope at $17.6 million. That is a more distributed market than some small-balance property types, though still concentrated enough that a handful of credit committees set the terms for independent operators nationally.

Federal credit is not the constraint here, which is the point. HUD's Section 232 program, mortgage insurance for residential care facilities, had a record fiscal 2025: $8.1 billion in firm commitments, nearly double fiscal 2024's $4.1 billion, with $6 billion closed across 337 loans and the application backlog cut from 130 to 29. The Federal Housing Finance Agency set 2026 multifamily purchase caps at $88 billion each for Fannie Mae and Freddie Mac, up 20.5 percent from 2025, with seniors housing eligible inside that allocation.

So government-supported financing is abundant, cheap by historical standards, and getting more so, and construction is still at a fourteen-year low. That gap is diagnostic. This is not a credit-availability problem. It is a returns problem, and no amount of loan capacity fixes a development spread.

Two notes on the SBA side that matter for anyone underwriting these deals. SOP 50 10 8, effective June 1, 2025, explicitly lists nursing homes including assisted living facilities as limited or special-purpose properties, requiring an appraiser experienced in the property type and, in most start-up, change-of-ownership and construction cases, an independent feasibility study. And as of July 4, 2026, the agency decoupled its programs so a qualified borrower can access up to $5 million under 7(a) and $5 million under 504 for a combined $10 million, the highest in the agency's history.

On credit performance, honesty requires a limit. The SBA does not publish loss experience by industry code. The 7(a) portfolio's annualized default rate reached 4.8 percent in March 2026, the highest since 2013, while the healthcare sector broadly ran 3.6 percent, among the better-performing segments. A senior-care-specific rate has to be constructed from the raw loan file, and the 2019 through 2023 approval cohorts have not finished seasoning. Anyone quoting one today is quoting an incomplete number.

The wave

Now the demand case, which is not in dispute and is genuinely staggering.

The oldest baby boomers, born in 1946, turned eighty in 2026. The 80-and-over population is projected to grow 36.6 percent over the coming decade against roughly 5 percent growth for the population as a whole, more than seven times the national rate. The 85-and-over cohort roughly doubles from about 6.5 million in 2022 to more than 14 million by 2040. By 2050 the United States will have close to 16 million more people over eighty than it has today.

Set that against a pipeline delivering 0.4 percent annual inventory growth and the shortfall math writes itself. NIC estimates a 550,000-unit deficit by 2030 and a $275 billion investment gap, requiring the industry to more than triple its current development pace. Arick Morton, who runs NIC MAP, frames the demand side as necessary rather than discretionary, a reasonable characterization for a product most people enter because they can no longer manage a household alone.

If that were the whole picture, senior housing would be the easiest underwriting decision in American real estate.

The wave cannot pay

It is not the whole picture, and the missing piece is the most important number in this article.

Research from NORC at the University of Chicago, published in Health Affairs and known in the industry as the Forgotten Middle study, projects 15.9 million middle-income Americans aged 75 and over by 2033, an 89 percent increase, roughly 7.5 million more people, from its 2018 base. Of those, 11.5 million, or 72 percent, will have less than $65,000 in combined annual income and annuitized assets. That is below what private assisted living plus associated medical care costs.

Even counting home equity, about 6.1 million of them, 39 percent of the cohort, still cannot cover assisted living.

The definition matters, so it is worth stating. NORC's middle-income band runs roughly $26,500 to $79,000 for ages 75 to 84 in 2018 dollars. These are not poor households. They are households with too much to qualify for Medicaid and too little to buy their way into private-pay senior housing. NORC also projects that 54 percent of them will carry three or more chronic conditions and about 31 percent will have cognitive impairment, meaning they will need the care they cannot purchase.

Caroline Pearson, one of the researchers, described the threshold as roughly $65,000 a year just to get the housing and health care required. Her colleague Dianne Munevar put the fragility more starkly: these seniors are one fall, one hospitalization away from spending down to poverty within a year or two.

Now put the rent next to it. Average asking rent across senior housing exceeded $5,800 a month in the first quarter of 2026, up 4.6 percent year over year, after eight consecutive quarters of growth in the 4.2 to 4.6 percent band and a 6.1 percent peak in mid-2023. JLL's read of the fourth quarter of 2025 is somewhat lower at $5,479, about 28.8 percent above pre-pandemic levels; the two series use different samples and the range is worth stating rather than resolving.

Take the NIC figure. Sixty-nine thousand six hundred dollars a year. That is above the entire income-and-assets threshold that 72 percent of the age-qualified middle-income population is projected to fall below.

The demographic wave and the affordability wall are not two separate stories. They are the same people.

The funding mechanism that broke

There is one more constraint, and it is the least discussed.

Private-pay senior housing entry is usually financed by selling a house. The resident converts a paid-off home into four or five years of monthly rent, and the transaction closes on the timeline of a residential sale.

That market has been shut for four years. Existing-home sales totaled 4.06 million in 2025, the lowest since 1995 and the fourth consecutive annual decline, roughly 34 percent below the 2021 peak and well under the historical norm near 5 million. NIC's Caroline Clapp made the connection directly, noting that slower home sales likely contribute to softening active-adult occupancy because many older adults sell before moving.

The mechanism is worth naming precisely, because it is a demand constraint no supply metric captures. A prospective resident with adequate home equity and an unsellable house is, for underwriting purposes, indistinguishable from one who cannot afford the community at all. The lead exists. The conversion does not.

The margin reset

The final piece of evidence is the one that looks most like recovery and, read carefully, is not.

NIC MAP operating margins surpassed 25 percent in mid-2025, the highest since 2018 and up about 120 basis points year over year. Coverage of that milestone was uniformly positive.

But the pre-pandemic benchmark, from the State of Seniors Housing series, was closer to 28 percent, earned in 2020 at materially lower occupancy than the sector runs today. The industry is generating a thinner margin at 89.9 percent occupancy than it generated at meaningfully lower occupancy six years ago.

Labor is why. Home health and personal care aides are now the single largest occupation in the United States, roughly 4 million workers, at a median around $35,000 a year as of May 2024. Nursing assistants run near $39,610. The broader direct-care workforce numbers about 5.4 million and is projected to add some 772,000 jobs between 2024 and 2034, more than any other sector in the economy. That is not a cyclical wage spike. It is a structural bid for labor that senior housing has to win every year, permanently, against every other employer of the same workers.

And then the divergence, which is the most important operating fact in the sector.

Welltower's senior housing operating portfolio posted a margin above 32 percent in the second quarter of 2026, up 300 basis points year over year and above pre-pandemic levels, with revenue per occupied room up 5.2 percent against expenses per occupied room up just 0.7 percent. Same-store net operating income grew 20.5 percent, the fifteenth consecutive quarter above 20 percent. Mitra's summary of the strategy is not subtle: the company serves the wealthiest age cohort in history.

Brookdale, which serves a broader income band across 541 communities in 41 states, ran consolidated weighted-average occupancy of 82.4 percent in the second quarter of 2026. That is up 230 basis points year over year, and it is also seven and a half points below the national average.

Same asset class, same quarter, same demographic tailwind. Two entirely different businesses.

The counter-case

Several pieces of evidence cut against the reading above, and they deserve equal billing.

Absorption is genuinely positive, not merely a denominator effect. First-quarter 2026 absorption rose about 8 percent year over year, and Brookdale reported its highest monthly net move-ins of 2026 in June. Demand is real and improving.

Welltower argues the affluent cohort can pay and has data behind it: management notes that baby boomer net worth growth has meaningfully outpaced senior housing rent growth, and its highest-occupancy communities, those above 95 percent, generate revenue-per-occupied-room growth above 6 percent. Best-in-class margins have already exceeded pre-pandemic levels, which complicates any blanket claim that margins are permanently impaired.

The supply picture may also be understated. NIC's tracked pipeline excludes conversions, adaptive reuse, and the large and growing unlicensed alternative, home care, small residential care homes, and family arrangements, which absorb an unknown share of the same demand. NIC MAP expanded its coverage to 214 markets in early 2026, which suggests the measured universe has been narrower than the real one.

And capital is voting. Transaction volume reached roughly $27 billion in 2025, up from $17.3 billion in 2024 and the highest since 2015, at an average $182,800 per unit, up 29 percent year over year. Cap rates compressed to about 6.2 percent by year-end, a 210-basis-point spread over the ten-year Treasury against a long-run average near 416. Some 86 percent of institutional investors surveyed said they plan to increase senior housing exposure in 2026.

The honest synthesis is this. The bull case is airtight for owners of existing, stabilized communities in affluent markets, and the numbers coming out of those portfolios are excellent. None of it is an argument that the middle market gets built.

What we are watching

Five measurable things will settle which version of this story turns out to be right.

Quarterly construction starts in primary markets are the first. They have run below 2,000 units in three of the last four quarters. Two consecutive quarters above that line would mean the development spread has reopened and the supply cliff is beginning to resolve.

Existing-home sales are the second. A recovery toward 4.5 to 5 million annually would restart the equity-conversion mechanism that funds private-pay entry, and it would show up in lead-to-move-in conversion before it showed up anywhere else.

Absorption is the third. If net absorption accelerates meaningfully above the 3.5 to 4.5 percent band while inventory growth stays near zero, the denominator argument weakens and demand is genuinely leading.

Middle-market product is the fourth and the most consequential. Somebody has to demonstrate a positive development spread at a rent the sub-$65,000 cohort can actually pay. Until that exists, the 11.5 million figure is a forecast of unmet need, not a market.

The fifth is the Chapter 11 count, thirteen senior care filings in 2025 against eleven in 2024. If that number keeps climbing through a period of record occupancy, the operating reset is deeper than the margin data suggests.

Until those move, the accurate description of American senior housing is a sector with the best headline fundamentals in commercial real estate and a structural problem underneath them. It is nearly full because almost nothing new opened. It is profitable at the top of the income distribution and precarious below it. And the demographic wave it has waited thirty years for is arriving on schedule, in numbers that exceed every projection, carrying roughly two-thirds less money than the pricing assumes.

The Buckingham's second bankruptcy was not a market failure. It was a preview.

Methodology

Figures on SBA lending volume, loan size, lender participation and program terms are Analytics.loan's analysis of the Small Business Administration's 7(a) loan-level FOIA dataset, filtered to the senior-care NAICS codes, covering calendar 2025 approvals. The SBA does not publish loss performance by industry code; a senior-care default rate must be constructed from the raw file, and approval cohorts from 2019 through 2023 have not finished seasoning, so no sector-specific default rate is reported here.

Occupancy, inventory growth, absorption, construction starts and asking rent figures are from NIC and NIC MAP public releases for the third and fourth quarters of 2025 and the first and second quarters of 2026. Readers should note that the consecutive-quarter count advanced between releases, the first quarter of 2026 marked the nineteenth consecutive gain and the second quarter the twentieth, and that some NIC MAP series are available only to subscribers; figures here are drawn from public releases.

Development cost and return-on-cost figures are from CBRE's Senior Housing and Care Investor Survey, data through the second quarter of 2026. Transaction volume, pricing and cap rates are from published brokerage market reporting. Middle-income projections are from NORC at the University of Chicago's Forgotten Middle research, published in Health Affairs, using 2018 Health and Retirement Study data as its base; the 2033 figures are projections, not realized data. Operator results are from Welltower and Brookdale public disclosures. Existing-home sales are from the National Association of Realtors. Wage and employment figures are from the Bureau of Labor Statistics. Where rent series conflict, NIC MAP above $5,800 for the first quarter of 2026 against JLL's $5,479 for the fourth quarter of 2025, the range is stated rather than reconciled.

Key figures for citation

Metric	Value
Senior housing occupancy, Q2 2026	89.9%, 20th consecutive quarterly gain
Occupied units	639,650 (record)
Year-over-year inventory growth	0.4%
Rolling four-quarter net inventory growth	Under 3,000 units, lowest in 20 years of data
Same measure at Q1 2020 peak	Over 21,000 units
Units under construction, Q2 2026	Under 16,000, lowest since 2012
Primary markets with no active construction	Nearly 60% of 140 tracked markets
Development cost per revenue unit	$388,830, up 23.6% since 2023
Return on cost vs stabilized cap rate	8.1% vs 7.0%
Average asking rent, Q1 2026	Over $5,800/month (+4.6% YoY)
Middle-income seniors 75+ by 2033	15.9 million (+89% from 2018)
Share unable to afford private assisted living	11.5 million, 72%
Existing-home sales, 2025	4.06 million, lowest since 1995
Operating margin, mid-2025 vs pre-pandemic	Above 25% vs approximately 28%
SBA 7(a) assisted living lending, 2025	~$313.3m across 257 loans
Senior care Chapter 11 filings, 2025	13 (up from 11 in 2024)
Projected unit shortfall by 2030	550,000 units / $275bn

Frequently asked questions

What is the current senior housing occupancy rate?

Occupancy across the 31 primary markets reached 89.9 percent in the second quarter of 2026, a twentieth consecutive quarterly gain, with occupied units at a record 639,650. Independent living reached 91.3 percent and assisted living 88.4 percent. Fifteen of the 31 primary markets were at or above 90 percent, triple the number three quarters earlier.

Why is senior housing occupancy rising?

Mostly because supply stopped growing rather than because move-ins surged. Rolling four-quarter net inventory growth fell below 3,000 units in early 2026, the lowest in twenty years of data, against more than 21,000 units at the 2020 peak. Absorption held steady at a 3.5 to 4.5 percent annualized rate over the same period. Both variables are healthy; only one of them moved.

Why is senior housing construction so low?

The arithmetic stopped working. All-in development cost reached $388,830 per revenue unit, or $364 per square foot, up 23.6 percent since 2023, while return on cost fell to 8.1 percent and stabilized cap rates rose to 7 percent. A 110 basis point spread does not justify a 29-month build plus multi-year lease-up when the same capital buys a stabilized community at a 7 cap on day one. Nearly $40 billion of senior housing loans also mature between 2026 and 2030.

Can middle-income seniors afford assisted living?

Largely not. Research projects 15.9 million middle-income Americans aged 75 and over by 2033, of whom 11.5 million, or 72 percent, will have less than $65,000 in combined annual income and annuitized assets. Even counting home equity, about 6.1 million cannot cover assisted living. Average asking rent exceeded $5,800 a month in early 2026, which is $69,600 a year, above the entire threshold that most of the cohort falls below.

What are senior housing operating margins?

Operating margins surpassed 25 percent in mid-2025, the highest since 2018. That is below the roughly 28 percent the sector earned before the pandemic at lower occupancy, because the labor cost base reset permanently. Best-in-class REIT portfolios have done considerably better, with one large operator posting a senior housing margin above 32 percent in the second quarter of 2026, up 300 basis points year over year.

How many senior housing units does the US need?

Industry estimates put the shortfall at 550,000 units by 2030 and a $275 billion investment gap, requiring the industry to more than triple its current development pace. The 80 and over population is projected to grow 36.6 percent over the coming decade against roughly 5 percent for the population as a whole, and the 85 and over cohort roughly doubles to more than 14 million by 2040.

Analytics.loan produces lender-grade feasibility and market analysis for SBA, USDA and conventional commercial real estate credit.

Sources:

• U.S. Small Business Administration, 7(a) loan-level FOIA dataset, senior-care NAICS codes, calendar 2025

• U.S. Small Business Administration, SOP 50 10 8, effective June 1, 2025

• National Investment Center for Seniors Housing and Care (NIC) and NIC MAP public releases, occupancy, inventory growth, absorption, construction starts and asking rents

• CBRE, Senior Housing and Care Investor Survey, development cost and return on cost

• NORC at the University of Chicago, "The Forgotten Middle" research, published in Health Affairs

• Welltower Inc. and Brookdale Senior Living Inc., public financial disclosures

• National Association of Realtors, existing-home sales

• U.S. Bureau of Labor Statistics, wage and employment data

• Published brokerage market reporting, senior housing transaction volume, pricing and cap rates