Best Cities to Invest in Real Estate: Gross Yield vs Real Return

The lists are not exactly wrong. The cities on them are mostly real places with real economies. The problem is narrower and more expensive than that: the number the lists lead with, gross rental yield, is not a return. It is rent divided by purchase price, before a single expense. By the time you have paid property tax, insurance, vacancy, maintenance, and management, a market advertised at 10% gross can land closer to 5% or 6%. That gap is not a rounding error. It is the difference between a property that pays you and a property you pay for.

This guide does something different. Instead of naming cities and hoping the arithmetic works out, it walks through the arithmetic itself, using data sources that are free, public, and published by agencies rather than by companies selling property. It shows you where the published figures disagree with each other, because they do, sometimes by a factor of three. And it shows you how to run the same screen on any market, including the ones nobody has written a listicle about yet.

Two things worth saying at the top. First, we are not going to tell you which city to buy in. We sell storage units. We have no position on your portfolio, and a storage company handing out market picks would be worth exactly what you paid for it. Second, there is a section near the end that tells you when renting a storage unit is a waste of money for a property investor, including from us. Both of those are deliberate.

Table of Contents

1. Why Most Best Cities Lists Fall Apart at the Spreadsheet
2. Gross Yield Is Not Return: The Arithmetic Every Ranking Skips
3. The Property Tax Number You Are Reading Is the Wrong One
4. Insurance Is the Line Item That Changed the Sun Belt Math
5. What the Published Insurance Averages Actually Disagree About
6. How to Pull Free Government Rent Data Instead of Guessing
7. What Fair Market Rent Does Not Tell You
8. Building the Gross-to-Net Bridge: A Worked Example
9. The Expense Lines First-Time Investors Underestimate Most
10. Vacancy, Turnover, and What It Costs When a Tenant Leaves
11. Population Growth Is Not the Same Thing as Rental Demand
12. Job Concentration Risk: When One Employer Is the Market
13. The Texas Case: No Income Tax, High Property Tax
14. The Midwest Value Markets and Where the Yield Math Holds Up
15. The Sun Belt Reassessment
16. Appreciation Markets Versus Cash Flow Markets
17. Short-Term Rental Regulation Is a Real Underwriting Risk
18. How to Read a Best Cities Ranking in Ten Minutes
19. Errors We Found in the Currently Ranking Guides
20. Where 10 Federal Operates and Where We Do Not
21. What Landlords and Flippers Actually Put in Storage
22. Sizing a Storage Unit for a Rental Portfolio
23. When You Should Not Rent a Storage Unit
24. Frequently Asked Questions
25. The Short Version

Why Most Best Cities Lists Fall Apart at the Spreadsheet

Ranking posts have a structural problem that has nothing to do with the honesty of the people writing them. A national list has to compare markets using numbers that exist for every market. Only a few numbers meet that test: median price, median rent, population change, job growth, and the ratio of rent to price. Everything that actually determines whether a specific property makes money is local to the parcel, and no national dataset carries it.

So the list gets built out of the available numbers, and the available numbers happen to be the ones that flatter every market roughly equally. Rent divided by price looks fine almost everywhere in the affordable half of the country. It is only when you subtract the local costs, which vary enormously and are not in the ranking, that markets separate.

Consider how much variation hides underneath a single national figure. The effective property tax rate on owner-occupied housing runs from roughly 0.3% to roughly 2.2% depending on whose 2026 table you read, which is a spread of more than sevenfold on the same purchase price. Insurance runs from under $1,000 a year to several times that. Neither of those appears in a rent-to-price ratio. Both of them come out of the same rent check.

The second structural problem is incentive. Many of the pages ranking for this query are published by companies that sell property, sell leads, or sell software to investors. That does not make their data false. It does mean the page is built to move you toward a transaction, and the framing tends to favor the number that looks best. A guide that led with net yield after local taxes and insurance would produce a much less exciting list.

The third is staleness disguised as freshness. Ranking posts get date-stamped and lightly refreshed while the substance underneath stays several years old. A post updated last week can still be describing a 2022 insurance market and a 2023 rate environment. The date at the top tells you when someone touched the file, not when the analysis was done.

None of this means you should ignore the lists. They are a reasonable way to generate candidates. They are a bad way to choose one. The rest of this guide is about the part that happens after the list.

Gross Yield Is Not Return: The Arithmetic Every Ranking Skips

Gross rental yield is annual rent divided by purchase price. A $250,000 house renting for $2,000 a month produces $24,000 a year, which is a 9.6% gross yield. That number is easy to compute, easy to compare across markets, and almost entirely disconnected from what you will actually earn.

What gross yield leaves out is every expense. Property tax. Insurance. Vacancy. Repairs and maintenance. Capital replacements like roofs and systems. Property management if you are not self-managing. HOA dues if there are any. Leasing costs when a tenant turns over. And, separately, financing, which is a capital structure question rather than a property question but still determines whether cash actually lands in your account.

The industry term for yield after operating expenses but before financing is the capitalization rate, or cap rate. Net operating income divided by price. Cap rate is the honest comparison figure, and it is dramatically lower than gross yield on the same property.

Here is the useful part. You do not have to take our word for the size of the gap, because the ranking pages sometimes publish both numbers on the same page without connecting them.

The guide currently ranking at the top of this query for us at the time of writing tells readers that Texas submarkets are producing gross rental yields of 10% to 15%, and lists target gross yields of 8% to 9.5% for San Antonio as a cash flow play. Further down the same page, the publisher displays two of its own Texas listings with full figures:

• Converse, Texas: $250,000 purchase price, $2,005 monthly rent. That is a 9.6% gross yield. The same listing block reports a 6.2% cap rate.
• San Antonio, Texas: $250,000 purchase price, $1,875 monthly rent. That is a 9.0% gross yield. The same listing block reports a 5.0% cap rate.

Read those two lines again. On the publisher's own inventory, priced and underwritten by the publisher, a 9.6% gross yield becomes a 6.2% cap rate and a 9.0% gross yield becomes a 5.0% cap rate. The haircut is roughly 3.4 to 4.0 percentage points. The article never explains that, and the reader who anchors on "8% to 9.5% gross" as a target has anchored on a number that the same page quietly reduces by about 40%.

That is not a hidden fact. It is not a controversial interpretation. It is arithmetic sitting on the page, in two places, unreconciled. And it is the single most useful thing you can learn before evaluating any market: the number in the headline and the number in the pro forma are different numbers, and the distance between them is where returns are actually decided.

One clarification, because it matters. Cap rate excludes financing on purpose, so that properties can be compared regardless of how each buyer paid for them. If you are borrowing, your cash-on-cash return, which is annual pre-tax cash flow divided by cash invested, is a different figure again, and at current rates it can be lower than cap rate rather than higher. Leverage cuts both directions.

The Property Tax Number You Are Reading Is the Wrong One

This one costs people real money, and it is almost universally gotten wrong in ranking content.

When a guide cites an effective property tax rate by state, it is nearly always citing the effective rate on owner-occupied housing. That figure comes from Census American Community Survey data on taxes actually paid by homeowners as a share of home value, which is the standard the Tax Foundation and most comparison tables use. It is a legitimate number. It is also the wrong number for a rental property, because owner-occupants receive tax treatment that investors do not.

Homestead exemptions reduce taxable value for an owner who lives in the home. Assessment caps limit how fast the taxable value of an owner-occupied home can rise year over year. Some jurisdictions classify non-owner-occupied residential property differently and assess it at a higher ratio. An investor typically gets none of the exemption and none of the cap.

The gap is not small. One 2026 comparison pairing the Tax Foundation's owner-occupied figures against the Lincoln Institute and Minnesota Center for Fiscal Excellence 50-state study of rental property found that Texas shows roughly 1.24% on owner-occupied housing but roughly 2.04% on an apartment in Houston, and noted the direct cause: rentals cannot claim the school homestead exemption or the 10% appraisal cap that owner-occupants receive. That is an effective rate about 65% higher than the benchmark most ranking posts quote.

The direction is not universal. The same comparison found Illinois running the other way, roughly 1.79% owner-occupied against roughly 1.44% for a Chicago rental, because Cook County assesses residential at a lower ratio than commercial and the classification rules cut differently there. So the correction is not "add 60% everywhere." The correction is "the statewide owner-occupied rate is a screening benchmark, not your bill."

There is a second trap on top of the first. In many jurisdictions, a sale resets the taxable value to the purchase price. If you are underwriting off the seller's current tax bill, and the seller has owned the property for fifteen years under an assessment cap, your first bill after closing can be substantially higher than the number in the listing. This surprises new investors constantly and it is entirely predictable.

What to do instead, in order:

1. Use the statewide effective rate only to screen. It is fine for deciding which five states to look at. It is not fine for underwriting a specific house.
2. Pull the actual parcel record from the county assessor or appraisal district. These are public, free, and searchable by address in most counties.
3. Check what exemptions are currently applied and confirm which of them you lose as a non-occupant.
4. Find out whether the sale triggers reassessment in that jurisdiction, and at what value.
5. Ask a local tax professional or the assessor's office directly before you rely on any of it.

We are not going to give you thresholds, exemption amounts, or a planning strategy here. Property tax law is set at the state and often the county level, exemption values change by legislative session, and a specific number published in a national blog post is a liability rather than a service. Get the parcel record and talk to someone licensed in that jurisdiction.

Insurance Is the Line Item That Changed the Sun Belt Math

For most of the last two decades, insurance was a rounding error in a residential pro forma. You budgeted a number, it went up a little each year, and it never changed a decision. That stopped being true, and a great deal of ranking content has not caught up.

The scale of the move is the part worth internalizing. LendingTree's analysis of Quadrant Information Services rate filings found home insurance rates rose 46.8% cumulatively between 2020 and 2025 nationally. In 2025 alone rates rose 6.0% nationwide, and no state saw a decrease. That is not a regional story. That is the entire national cost base of owning residential property resetting inside five years.

Two findings from that data matter specifically for market selection.

The increases are not where the headlines put them. The same analysis found Colorado led 2025 with an 18.3% single-year increase, followed by Minnesota at 17.0% and Iowa at 14.7%. Meanwhile Florida rose 0.4%, Montana 0.5%, and Texas 0.6%. Over the full 2020 to 2025 window, Colorado rates roughly doubled, up 100.8%, with Iowa at 96.0% and Minnesota at 88.2%. An investor who avoided Florida in 2024 because of insurance headlines and bought in Colorado instead may have moved toward the problem rather than away from it.

The severe convective storm belt now rivals the hurricane coast. The old mental model said coastal wind risk was the expensive risk. In the 2026 LendingTree figures, Oklahoma at $5,298 is the most expensive state in the country, followed by Nebraska at $4,956 and Colorado at $4,310, against a national average of $2,395. Hail and tornado exposure, not hurricanes, drives the top of that list. If your screen assumes inland means cheap, it is running on a model that no longer describes the market.

Then there is the part specific to rentals. A landlord policy is not a homeowners policy, and it does not cost the same. Across multiple 2026 industry sources the consistent finding is that landlord policies run roughly 15% to 25% more than a comparable homeowners policy on the same property, reflecting higher claim frequency, broader liability exposure, and coverage features like loss of rental income that an owner-occupant policy does not carry. Whatever homeowners figure you find for a market, the landlord number is meaningfully above it.

A few structural points that a single annual premium figure hides entirely:

• Wind and hail deductibles are often a percentage of the insured value, not a flat dollar amount. A 2% wind deductible on a $300,000 dwelling limit is $6,000 out of pocket before the policy pays anything. Your annual premium can look reasonable while your exposure in the event that actually happens is very large.
• Flood is a separate policy. Standard property insurance does not cover flood. Flood coverage comes through the National Flood Insurance Program or a private carrier, and it is priced on the parcel's flood risk. Check the FEMA flood map for the specific address, not the city.
• Roof age and roof schedules drive both price and availability. In hail-exposed markets, carriers increasingly write actual cash value rather than replacement cost on older roofs, which shifts a large repair cost onto you at exactly the moment you need the coverage.
• Availability is a real constraint, not just price. In some markets the binding question is not what a policy costs but whether an admitted carrier will write one at all, which pushes you to surplus lines or a state-backed plan of last resort.

The operational takeaway is simple and it is the one thing in this section you should actually act on: get a real quote on the specific address, in writing, before your inspection period closes. Not a state average. Not a per-square-foot estimate. A quote on the parcel, with the deductible structure spelled out.

What the Published Insurance Averages Actually Disagree About

Here is something we did not expect to find, and it changed how we would tell anyone to use published cost data.

We pulled 2026 homeowners insurance averages by state from six separate sources. They do not agree. Not by a little.

For Florida alone, the published 2026 average annual homeowners premium came back as $8,471 from one source, $7,900 from another, $7,136 from a third, and $2,691 from a fourth. That is a spread of more than three to one on the same state in the same year. Worse for anyone trying to rank markets, the sources do not agree on which state is most expensive. Several name Florida. The LendingTree analysis names Oklahoma at $5,298 and places Florida at $2,691, well down the list.

Landlord insurance figures scatter even further. One 2026 source puts the national average landlord premium at roughly $1,516 a year with Louisiana highest at $2,484 and Oregon lowest at $883. Another puts the national average at roughly $3,250 with Florida at $9,875 and Hawaii at $756. A third describes a typical range of $800 to $3,000 for a standard three-bedroom rental, reaching $2,200 to $4,600 and up in catastrophe-exposed states.

None of these sources is obviously lying. They are measuring different things and mostly not saying so. The variables that produce this spread include:

• Dwelling coverage limit. A study priced at a $300,000 dwelling limit and one priced at $500,000 are not comparable, and the limit is frequently in a footnote or absent.
• Data source. Rate filings submitted to regulators, actual bound policies, and online quote requests produce systematically different numbers.
• Whether wind and hurricane coverage is included. In several coastal states, wind is commonly excluded from the base policy and written separately. A "Florida average" that omits the wind policy is describing a fraction of the real cost.
• Whether state-backed insurers of last resort are in the sample. Including or excluding those pools moves a state average substantially.
• Vintage. A figure labeled 2026 may be built on 2024 filings.

So what is actually usable? The relationships, not the levels. Across every source we checked, these held:

1. Landlord policies cost meaningfully more than homeowners policies on the same property, in a consistent 15% to 25% band.
2. The hail and tornado belt has moved to the top of the cost tables alongside or above the hurricane coast.
3. Colorado's multi-year increase is extreme by any measure, roughly a doubling across 2020 to 2025.
4. No state got cheaper in 2025.

Directional findings that hold across independent methodologies are worth something. A specific state average from any single source is worth much less than it appears, and plugging one into a pro forma without a quote on the actual property is how a deal that pencils on a spreadsheet stops penciling at closing.

Apply the same skepticism to every "average" in a ranking post, including median price, median rent, and appreciation forecasts. Ask what was measured, over what period, at what coverage or quality level, and by whom.

How to Pull Free Government Rent Data Instead of Guessing

Almost every rent figure in a ranking post is either a proprietary estimate from a listings platform or an unsourced number. There is a public alternative that most retail investors have never used, and it costs nothing.

The Department of Housing and Urban Development publishes Fair Market Rents every federal fiscal year. FMRs exist for every metropolitan area, every metropolitan subdivision, and every nonmetropolitan county in the United States, broken out by bedroom count. They are published at least 30 days before taking effect and become effective at the start of the federal fiscal year, generally October 1.

FMR is defined as an estimate of the 40th percentile gross rent for standard-quality units in the market. HUD builds it from Census American Community Survey five-year data as a base, then updates it forward. It is used to set payment standards for the Housing Choice Voucher program, which is why it exists at all, but nothing stops an investor from using it as a rent benchmark.

In designated metropolitan areas HUD also publishes Small Area Fair Market Rents, which are calculated at the ZIP code level rather than metro-wide. For a large metro where rents vary dramatically between submarkets, the SAFMR is far more useful than the metro-wide figure, and it is the closest thing to free neighborhood-level rent data from a government source.

To give a sense of what the FY2026 two-bedroom figures look like across markets:

• Harris County, Texas (Houston metro FMR area): roughly $1,573 per month
• Fulton County, Georgia (Atlanta metro FMR area): roughly $1,820 per month
• Maricopa County, Arizona (Phoenix metro): roughly $1,839 per month
• Miami-Dade County, Florida: roughly $2,436 per month
• King County, Washington (Seattle metro): roughly $2,501 per month

Both the metro-wide FMR dataset and the SAFMR lookup are published on HUD's data portal, along with full methodology documentation for each fiscal year. Search for the HUD FMR documentation system and the Small Area FMR lookup by name and you will find the official tools rather than the many paid sites that repackage the same public data behind a subscription.

Two practical uses. First, as a sanity check: if a ranking post claims rents in a market that sit far above the FMR for that county, ask what quality tier they are describing. Second, as a floor: FMR gives you a defensible conservative rent assumption when you are screening dozens of markets and cannot pull comparables for each one.

What Fair Market Rent Does Not Tell You

FMR is genuinely useful and it is also widely misused, including by people who should know better. Four limitations matter enough to change your numbers.

It is a gross rent, and gross means utilities are inside it. HUD's definition of gross rent includes shelter rent plus the cost of tenant-paid utilities, excluding telephone, cable, satellite, and internet. If your tenant pays electricity and gas separately, the rent you can actually charge is FMR minus a utility allowance, not FMR. Treating FMR as collectible rent overstates your income by whatever utilities run in that market, which in a cold-winter or hot-summer market is not trivial.

It is the 40th percentile, not the median and not the market. By construction, 60% of standard-quality units rent for more. FMR describes a modest unit, deliberately, because it exists to set voucher payment standards. Using it as your expected rent on a renovated property will understate your income, sometimes badly.

It excludes new construction and low-quality units. HUD's methodology removes units built in the last two years, already-subsidized units, and substandard units from the calculation. If you are buying newer product, FMR is not describing your property.

It is drawn from recent movers. HUD uses recent-mover rents rather than the rents long-tenured tenants pay, because long-term tenants generally pay below market. That actually makes FMR more useful for an investor than a raw average would be, since you will be signing new leases. It is worth knowing that this is a deliberate methodological choice rather than an accident.

One more thing that trips people up: published FMR is a benchmark, not an approval. Local housing authorities set their own payment standards, commonly in a band around 90% to 110% of FMR, and they apply rent reasonableness tests, utility allowances, and local program rules on top. If your plan involves voucher tenants, the published FMR is the starting point of a conversation with the local authority, not the answer.

Used correctly, FMR is a free, methodologically documented, geographically complete rent floor published by a federal agency. That is a better foundation than an unsourced number in a listicle. It is not a substitute for pulling actual comparable leases in the submarket before you buy.

Building the Gross-to-Net Bridge: A Worked Example

Abstract warnings about expenses do not change behavior. Watching a number fall does. So here is the bridge, run end to end on a single hypothetical property.

The illustration below uses a $250,000 single-family rental at $2,000 a month, which is roughly the profile the ranking guides describe as a Sun Belt cash flow property. Every expense assumption is labeled. These are illustrative figures chosen to demonstrate the method, not market data for any specific city. Your job is to replace each one with a real number for your actual parcel.

Starting point. $2,000 per month is $24,000 a year. Against a $250,000 price that is a 9.6% gross yield, which is the number that would appear in a ranking post.

Now the subtractions.

• Vacancy and credit loss, 8%: minus $1,920. Assumes roughly one month of vacancy per year plus a small allowance for uncollected rent. Effective gross income $22,080.
• Property tax at a 2.0% investor effective rate: minus $5,000. Note this uses an investor rate, not an owner-occupied benchmark, per the earlier section.
• Landlord insurance: minus $2,400. A placeholder in the middle of the published ranges. In a hail or coastal market this could be double.
• Repairs and maintenance, 8% of gross rent: minus $1,920.
• Capital reserves, 8% of gross rent: minus $1,920. Roof, HVAC, water heater, flooring. These are not optional; they are deferred, and deferring them is not the same as avoiding them.
• Property management, 9% of collected rent: minus $1,987. Plus a typical leasing fee amortized across the tenancy, which we are folding into the turnover discussion rather than counting twice here.

Total operating expenses: roughly $13,227. Net operating income: roughly $8,853.

Against the $250,000 price, that is a 3.5% cap rate. The property that advertised at 9.6% gross produced a 3.5% unlevered return before a dollar of debt service.

That result is deliberately harsh, and you should push back on it. Self-manage and you add back roughly $1,987, moving to about 4.3%. Buy in a 1.0% investor tax jurisdiction rather than 2.0% and you add back $2,500, moving to roughly 5.3%. Do both and you are near 6.1%, which lands right about where the competitor's own listings reported. Reduce the capital reserve on a recently renovated house and it improves further.

That is the entire point. Every one of those swings came from a local variable that the ranking post did not contain. The market did not change. The tax jurisdiction, the management decision, and the condition of the building changed, and together they roughly doubled the return.

Run this bridge before you get attached to a market. If you cannot fill in every line with a number you can source, you are not underwriting, you are hoping.

The Expense Lines First-Time Investors Underestimate Most

Some expenses get underestimated more reliably than others. These are the ones that show up repeatedly in the gap between projected and actual performance.

• Capital expenditure reserves. The most commonly omitted line entirely. A roof has a finite life, and so do the furnace, the water heater, the flooring, and the exterior paint. If you owned the property for the full life of those components you would pay for all of them. Setting aside nothing does not mean they are free, it means you will fund them from savings at an unpredictable moment. Reserve against expected replacement cost divided by remaining life, not against a percentage that feels comfortable.
• Turnover cost, as distinct from vacancy. Vacancy is the rent you do not collect. Turnover is what you spend to get the next tenant: paint, cleaning, carpet, lock changes, minor repairs, listing photos, and a leasing fee that is frequently half a month to a full month of rent. These are separate lines and both are real.
• The first-year tax reset. Covered earlier and worth repeating because it recurs. If the jurisdiction reassesses on sale, your year-one tax bill can materially exceed the seller's, and it hits before you have banked any reserves.
• Insurance at renewal, not at binding. The quote you bind at is a snapshot. Given the national trend described earlier, budget for renewal increases rather than treating year-one premium as fixed.
• HOA dues and special assessments. Dues appear in listings. Special assessments do not, and in condo and townhome inventory they can be substantial. Ask for the association's reserve study and recent meeting minutes during diligence.
• Utilities during vacancy. Somebody pays to keep the power and water on between tenants, and it is you. Small line, but it is always there and it is never in the pro forma.
• Eviction and legal. Infrequent but not rare, and the cost varies enormously by jurisdiction because timelines vary enormously by jurisdiction. A market with tenant-favorable procedural timelines carries a different risk profile than one without, and that difference does not show up in a rent-to-price ratio.
• Your own time. Self-managing to add 9% back to your return is a real choice with real economics. It is also a job. Count it honestly, especially if you are underwriting a market several states away.

The pattern across all of these is the same. They are lumpy, they are local, and they are invisible in national comparison data. Which is exactly why national comparison data cannot tell you where to buy.

Vacancy, Turnover, and What It Costs When a Tenant Leaves

Vacancy is usually entered as a percentage because that is convenient. Understanding what actually generates it is more useful than picking a number.

Vacancy has three separate components, and conflating them produces bad estimates. There is frictional vacancy, the days between one tenant leaving and the next moving in, which is a function of how fast the unit leases and how well you manage the overlap. There is turnover downtime, the days the unit is uninhabitable because you are painting and repairing it, which is a function of condition and contractor availability. And there is market vacancy, the risk that the unit simply does not lease at your asking rent because supply exceeded demand, which is the one that actually varies by city.

The Census Bureau publishes rental vacancy rates through its Housing Vacancies and Homeownership survey, quarterly, at national and regional level and annually for the largest metros. That is the public source for the third component. It is worth checking against whatever a ranking post asserts, because "low vacancy" is claimed for nearly every market in nearly every guide, and it cannot be true everywhere at once.

Turnover frequency is the variable most within your control and the one most worth optimizing. Every avoided turnover saves you the downtime, the make-ready cost, and the leasing fee simultaneously. The levers are unglamorous: screen carefully, respond to maintenance requests quickly, and raise rent in moderate annual increments rather than large infrequent jumps that push otherwise good tenants to shop the market.

There is a market-selection implication here that rankings rarely discuss. Markets with a high proportion of short-tenure renters, such as those dominated by transient student populations or short-assignment workforces, generate structurally more turnovers per year than markets with stable family renters, even at identical vacancy rates. Two markets can report the same vacancy percentage and produce very different actual costs, because one is turning units twice as often. When a guide recommends a market specifically for student housing or a rotating workforce, that is a turnover-heavy profile, and the make-ready cost belongs in the model.

Our neighborhood guides get at this at the submarket level in several of the markets we serve. The El Paso neighborhood guide covers how the rotating Fort Bliss population shapes the rental profile there, and the North Myrtle Beach guide covers how seasonal short-term rental demand changes the character of a submarket between July and January.

Population Growth Is Not the Same Thing as Rental Demand

Every ranking post leads with population growth. It is the most intuitive metric in the genre and one of the least reliable, because it answers a question adjacent to the one you are asking.

Rental demand is a function of household formation at your price point, not of headcount. Those come apart in specific, predictable ways.

Growth can be absorbed by new supply. A metro adding 40,000 people while permitting 25,000 new units is a different market from one adding 40,000 people while permitting 4,000. The first has rising population and softening rents. The second has rising population and rising rents. Population growth alone does not distinguish them. Residential building permit data is published by the Census Bureau and is the counterweight you need alongside any growth figure.

Growth can arrive as owners rather than renters. In-migration of established households buying homes adds population and can reduce rental demand at the margin. Growth composition matters more than growth rate.

Growth can land at a price point you are not serving. A metro absorbing high-income tech relocations produces demand for a different product than a metro absorbing entry-level logistics employment. If you are underwriting a modest three-bedroom, you care about the second, and a headline growth number tells you nothing about which one is happening.

Growth can be concentrated in exurbs you are not buying in. Metro-level statistics aggregate a large area. The growth in a metro figure may all be happening thirty miles from the submarket you are looking at.

The corrective is to look at three published series together rather than one. Census population estimates for the direction of travel, Census building permits for supply response, and Census Housing Vacancies data for whether the market is actually tight. All three are free. Two of them are almost never cited in ranking content.

The same discipline applies to job growth claims. A ranking post that says a city has strong job growth is describing net employment change. The Bureau of Labor Statistics publishes that data by metro area with industry breakdown, and the industry breakdown is where the information is. Growth concentrated in one sector is a different risk profile than growth spread across many, which brings us to the next section.

Job Concentration Risk: When One Employer Is the Market

Ranking posts routinely present a large local employer as a reason to buy. A major corporate headquarters, a university, a military installation, a hospital system, a distribution hub. The presence of a large stable employer genuinely is a positive. It is also a concentration, and concentration cuts both ways.

The question to ask is not whether the anchor employer is strong. It is what share of the local employment base it represents, and what happens to your rent roll if that share contracts.

Several of the market types that appear most often in these lists are concentration plays whether or not the guide says so:

• Corporate headquarters towns. When a guide recommends a small metro because a Fortune 500 company is headquartered there, it is recommending a market where one firm's employment decisions substantially set local housing demand. The upside is a high-income, stable tenant base. The downside is correlated: a restructuring hits your occupancy, your rent, and your resale value at the same time.
• University towns. Reliable demand, structurally high turnover, pronounced seasonality, and enrollment trends that are demographic rather than economic. National undergraduate enrollment is sensitive to birth cohort size, and that is a knowable multi-year trend rather than a surprise.
• Military markets. Steady demand and a tenant base that relocates on a schedule, which means predictable turnover. Force structure and basing decisions are policy variables outside the local economy entirely.
• Single-industry regional economies. Energy, ports and logistics, and manufacturing clusters all produce genuine growth and all carry cycle exposure that a five-year appreciation forecast will not capture.

None of these is a reason to avoid a market. Concentration is a risk to price, not a disqualification. A market with a single dominant employer should clear a higher return threshold than a diversified one, because you are being asked to hold more risk. The mistake is not buying in a concentrated market. The mistake is buying in one at a price that assumes it is diversified.

Diversification at the portfolio level is the other answer. Two properties in the same submarket, serving tenants from the same employer, are one bet held twice.

The Texas Case: No Income Tax, High Property Tax

Texas appears near the top of nearly every list, and the reason given is almost always the same: no state income tax, strong job growth, population inflow. All three are accurate. The conclusion frequently drawn from them is not.

Start with the income tax claim, because it is the one most often misapplied to rental property. Texas does not levy a personal state income tax. For a rental property owner, the relevant question is what that actually changes about the property's economics, and the answer is less than the framing implies. Your rental income is subject to federal income tax regardless of which state the property sits in. If you live in a state that taxes income, your resident state generally reaches your income wherever earned, subject to credits for taxes paid elsewhere. The absence of a Texas income tax is a meaningful benefit primarily to people who live in Texas.

Meanwhile the property tax is the expense that hits the rental pro forma directly, every year, and it is high. Texas funds locally in large part through property taxation. And as covered earlier, the investor's effective rate is substantially above the owner-occupied statewide figure that ranking posts cite, because rentals do not receive the school homestead exemption or the appraisal cap. The comparison we found showed roughly 1.24% owner-occupied against roughly 2.04% for a Houston rental.

Run that through the bridge from Section 8 and the effect is not subtle. On a $250,000 property, the difference between a 1.24% assumption and a 2.04% reality is $2,000 a year, which is roughly 8% of gross rent, taken straight off net operating income.

Insurance is the second correction. Texas carries hurricane exposure on the Gulf coast and severe hail exposure across a wide inland band, and it sits high in the 2026 cost tables at roughly $3,969 in the LendingTree analysis. Notably, that same analysis found Texas rates rose only 0.6% in 2025, one of the smallest increases in the country, so the level is high while the recent trend is flat. Both facts are true and a ranking post will typically cite neither.

What this adds up to is not that Texas is a bad market. Texas has genuine structural strengths: real population growth, a diversified large-metro economy in DFW and Houston, and a landlord-lease framework investors generally find workable. It adds up to something narrower and more useful: the case for Texas that ranking posts make is built partly on a tax argument that does not do what they imply, while the expense that most affects a Texas rental is understated by the benchmark they quote. Underwrite it with the investor tax rate and a real insurance quote and it may still clear your threshold. Underwrite it with the owner-occupied rate and a national average premium and you will be disappointed.

Nothing in this section is tax advice, and property tax treatment, exemption eligibility, and appraisal procedure differ by county and change by legislative session. Pull the appraisal district record for the specific parcel and talk to a Texas CPA or property tax consultant before relying on any figure here.

The Midwest Value Markets and Where the Yield Math Holds Up

The 2026 crop of ranking posts has rotated toward Midwestern and Rust Belt markets, and the rotation is not baseless. Lower entry prices mechanically produce higher rent-to-price ratios, and several of these metros have tight supply relative to demand at the affordable end.

The gross-to-net discipline still applies, and it applies differently here than in the Sun Belt.

What tends to work in favor of these markets. Entry prices genuinely are lower, which means a smaller absolute capital commitment and less exposure to a price correction. Insurance in much of the interior Midwest, outside the hail belt, sits below the national average. Rental demand at the affordable end is frequently durable because there is little new construction being delivered at that price point, since it does not pencil for builders.

What tends to work against them, and gets omitted. Several of these states carry high effective property tax rates, and in the highest-rate states the tax line alone can consume a large share of the yield advantage. Housing stock is older on average, which raises both capital expenditure reserves and insurance cost, and in some markets affects insurability outright. Population trends in a number of these metros are flat or declining even where the housing market is currently tight, which is a different long-run appreciation profile than a growing market. And the hail belt runs straight through part of the region, which is why Minnesota and Iowa posted 17.0% and 14.7% single-year insurance increases in 2025 and roughly 88% and 96% cumulative increases across 2020 to 2025.

The honest summary is that Midwest value markets frequently do produce better net yields than Sun Belt growth markets, and frequently produce worse appreciation. That is a real tradeoff and which side of it you want depends on your objective, your holding period, and whether you need current income. It is not a case where one region is simply better.

Be particularly careful with headline appreciation forecasts in this category. Several 2026 ranking posts cite double-digit projected price growth for specific smaller metros. Those are model outputs from forecasting products, they carry wide error bands that the ranking post does not reproduce, and they have historically been revised substantially. Treat a single-year price forecast for a small metro as a weak signal, not a thesis.

The Sun Belt Reassessment

The Sun Belt thesis of the last decade was straightforward: people and jobs are moving south and west, supply cannot keep up, buy ahead of it. It worked for a long time and a great deal of money was made on it.

Several things changed at once, and ranking content has been slow to reprice them.

Supply caught up in specific metros. The building response to the 2021 and 2022 demand surge delivered into 2024 and 2025. In the metros that permitted most aggressively, particularly in multifamily, that delivery has pressured rents. This is exactly the population-versus-permits distinction from Section 11, and it is the clearest current example of why growth alone is not a thesis.

Insurance repriced. Covered at length above. The important nuance is that the repricing is not uniformly where the headlines put it. Florida's 2025 increase was among the smallest in the country at 0.4%, following very large increases in prior years, while Colorado, Minnesota, and Iowa led. The Sun Belt insurance story is largely a story about what already happened rather than what is happening now, and the current increases are concentrated elsewhere.

Property tax followed values up. Assessments in fast-appreciating markets rose with prices. For an owner-occupant, caps and exemptions dampened that. For an investor, they largely did not.

Affordability constrained rent growth. Rent cannot durably outrun local incomes. In metros where prices and rents both rose faster than wages, the ceiling is a real constraint rather than a theoretical one.

What this does not mean is that the Sun Belt is finished. Population inflow to much of the South is real and continuing, the labor markets in the large metros are genuinely diversified, and several of these markets have already absorbed their supply wave. What it means is that the thesis is no longer a free option. In 2018 you could buy a Sun Belt rental somewhat carelessly and be rescued by appreciation. That is a much weaker assumption now, and the difference between a good and a bad purchase has moved back to underwriting.

Appreciation Markets Versus Cash Flow Markets

Ranking posts often present cash flow and appreciation as two categories of market you can simply choose between, as though both were equally available and it were purely a matter of preference. The relationship is more structural than that.

Rent-to-price ratios and expected appreciation tend to move inversely, and the reason is not mysterious. Buyers bid prices up in markets where they expect future growth, which mechanically lowers the yield on current rent. Markets where nobody expects much growth are cheap relative to rent for the same reason. A high current yield is frequently the market pricing in low expected appreciation, and treating it as free money misunderstands what you are being paid for.

This matters for how you read a list that claims a market offers both. Sometimes that is true, and it usually reflects a genuine information or access advantage: a market that is growing but not yet widely noticed, or one where local knowledge lets you buy below market. More often, a guide claiming both is quoting a gross yield for the cash flow claim and a forecast for the appreciation claim, and neither number survives contact with the net-yield bridge.

Some practical framing rather than a recommendation:

• Current income and appreciation are different products with different risk profiles. Appreciation is unrealized until you sell or refinance, and it is the more volatile of the two. Cash flow is realized monthly and is the more predictable.
• Your holding period should drive the weighting. A short holding period makes you dependent on price movement over a window too small to be reliable. A long holding period lets rent growth and amortization do more of the work.
• Leverage amplifies whichever one you get, and both directions. A thin cash flow margin becomes negative quickly when a rate resets or a vacancy runs long.
• Negative cash flow is a position that requires funding. Buying for appreciation while running a monthly deficit means you need external income to hold the position through any downturn. That is a legitimate strategy for someone with the balance sheet to support it and a serious hazard for someone without.

Whether that tradeoff is appropriate for your situation depends on your income, your reserves, your tax position, and your risk tolerance, none of which we know. Those are questions for a financial advisor or CPA who has seen your full picture.

Short-Term Rental Regulation Is a Real Underwriting Risk

A significant share of ranking posts recommend markets specifically for short-term rental potential, usually tourism and hospitality metros. The projected returns for short-term rentals are higher than long-term, sometimes dramatically. So is the risk, and the largest component of that risk is not occupancy.

It is that the use can be regulated out from under you after you buy.

Short-term rental rules are set locally, they have been in active flux across many jurisdictions, and they take a wide variety of forms: outright prohibition in defined zones, primary-residence requirements, annual permit caps, minimum-stay requirements that eliminate the weekend market, registration and licensing regimes, occupancy taxes, and homeowners association restrictions operating independently of municipal law. Any one of them can convert a short-term rental pro forma into a long-term rental pro forma overnight.

The specific hazard is the gap between the two. If you paid a price justified by short-term rental income and the use is restricted, the property has to work as a long-term rental at a price you set assuming it would not have to. In tourism markets, where purchase prices are frequently elevated relative to long-term rents, that gap can be large enough to make the property unviable.

Diligence questions worth answering before you commit, all of which require local sources rather than a national guide:

• What does the current municipal or county ordinance actually permit at this specific address, by zone?
• Is there a permit cap, and is it currently at the cap with a waiting list?
• Are permits attached to the property or to the owner, and do they survive a sale?
• Is there pending legislation or a moratorium under consideration?
• What do the HOA covenants say, and can they be amended by a vote of owners?
• Does the property still work financially as a long-term rental if the answer to any of the above changes?

That last one is the underwriting test, and it is the one that matters. If the deal only works as a short-term rental, you are taking regulatory risk as a primary risk rather than a secondary one, and you should be paid for it.

Local land use and licensing law is genuinely local and it changes. This section describes a category of risk, not the rules in any jurisdiction. Consult a local real estate attorney and the municipality directly before purchasing on a short-term rental thesis.

How to Read a Best Cities Ranking in Ten Minutes

Ranking posts are useful for generating candidates. Here is a fast triage that tells you how much weight to give one before you invest any real time in it.

1. Find out who published it and what they sell. Scroll to the footer and the author bio. A page published by a company that sells turnkey investment property, generates investor leads, or brokers loans is not disqualified, but you now know which direction the framing points. A phone number repeated three times in the body is a tell.
2. Check whether it says gross or net. If every yield figure is gross, or the article never uses the words cap rate or net operating income, it is describing the top line only. Apply the Section 8 bridge mentally and discount accordingly.
3. Look for property tax and insurance. If neither appears as a line item anywhere in the article, the analysis omits the two largest fixed costs of ownership. That is disqualifying for underwriting purposes even if the city selection is reasonable.
4. Trace one attribution. Pick a single specific claim, ideally a number, and see whether the source is named with a date and whether you can find it. Vague attribution ("according to Zillow," "experts say," "reports show") with no link and no vintage is the most common failure mode in this genre.
5. Check the date against the content. Look for something that would have changed recently and see whether the article reflects it. Insurance is the best current test. An article date-stamped this month that never mentions insurance cost is not actually current regardless of its date stamp.
6. Count the near-duplicates. Look at the publisher's own related links. If they have eight nearly identical posts targeting slight variations of the same query, the pages are built for search coverage rather than for a reader, and depth is unlikely.
7. Look for anything that argues against a market. A list of twenty cities in which all twenty are good is not analysis. Genuine market assessment produces reasons not to buy somewhere.

A page that fails most of these is still useful as a source of city names. Take the names, discard the analysis, and run your own.

Errors We Found in the Currently Ranking Guides

We read the pages currently ranking for this query rather than summarizing their summaries. These are the specific problems we found, documented so you can check them yourself.

1. Gross yield presented as return, contradicted on the same page. Covered in Section 2 and worth listing here as the headline finding. The top-ranking guide advertises 10% to 15% gross yields in DFW submarkets and 8% to 9.5% targets in San Antonio, then displays its own two Texas listings showing 9.6% gross converting to a 6.2% cap rate and 9.0% gross converting to a 5.0% cap rate. Both sets of numbers are on the page. Neither is reconciled to the other.

2. "No state income tax" cited as a cash flow advantage. Multiple guides list the absence of a state income tax among the reasons a Texas market produces strong rental cash flow. As covered in Section 13, this conflates the investor's personal residency tax position with the property's operating economics. It is also cited without the offsetting property tax burden, which is the expense that actually hits the rent roll.

3. Owner-occupied property tax rates applied to rental underwriting. Effectively universal across the guides we read. Where property tax appears at all, it is the owner-occupied effective rate. For a Texas rental that understates the burden by a wide margin.

4. A place name that does not exist. The top-ranking guide lists "Hüeysville" among recommended Birmingham, Alabama submarkets, alongside Center Point and Roebuck. There is no Hüeysville in the Birmingham metro. The intended reference is almost certainly Hueytown, a real city in Jefferson County. Small on its own, but it is a production error in the specific list of neighborhoods a reader is being told to buy in.

5. Insurance omitted entirely. The top-ranking guide recommends Miami, Tampa, Orlando, and Jacksonville and does not mention insurance cost anywhere in the article. Given that insurance is among the largest and most volatile expense lines in Florida residential property, and that national rates rose 46.8% cumulatively from 2020 to 2025, a Florida recommendation that omits it is incomplete in a way that directly affects returns.

6. A recycled statistic with no clear meaning. Several guides, across different publishers, repeat that Detroit leads the nation in appreciation with a "26.8% housing market premium." We could not establish what that figure measures, over what period, or from what underlying source. A percentage with no defined denominator, propagating across multiple sites in identical phrasing, is a sign of syndication rather than analysis. Treat it as unverified.

7. Vague attribution throughout. Claims are credited to "Zillow," "Realtor.com," and "experts" without a report name, a date, or a link. In several cases the underlying rankings appear to be from prior years, presented in an article date-stamped this year.

Two things we want to be fair about. First, some of what these guides say is accurate. The observation that Alabama's very low effective property tax rate improves net cash flow is correct, and it is exactly the right kind of reasoning. Second, our objection is not that these publishers sell property. It is that the analytical framing consistently favors the number that supports the transaction, and the reader is not told that a different number exists.

What Landlords and Flippers Actually Put in Storage

Storage shows up in a rental operation in a small number of recurring situations. Knowing which ones apply to your strategy is the difference between a unit that earns its rent and a unit that quietly becomes a monthly subscription to a pile of things you will never use.

Turnover staging. The most common legitimate use. When a tenant leaves and the unit needs paint, flooring, or repairs, the contractors need the space empty. If you own furnished rentals, or you are holding appliances, window treatments, and fixtures between tenancies, they need somewhere to go for two to six weeks. A drive-up unit near the property makes this a two-hour job instead of a logistical problem.

Renovation materials and fixtures. On a value-add purchase, materials frequently arrive before the site is ready and cabinetry, flooring, appliances, and fixtures need a secure interim home. This is also where buying ahead on a discounted lot of materials becomes practical, because you have somewhere to put it.

Tools and equipment between projects. If you are running renovations across multiple properties, the ladders, saws, compressors, and paint equipment have to live somewhere between jobs. A unit is usually cheaper than a garage bay and considerably cheaper than replacing tools stolen from a job site.

Staging inventory. If you flip and stage, staging furniture is a reusable asset that is only deployed part of the time. Storing it between listings is what makes owning it rather than renting it economical.

Abandoned tenant property. This one is legal rather than logistical, and it is important. Many states require a landlord to store a departed tenant's personal property for a defined period, give specified notice, and follow a defined disposal procedure before discarding or selling it. The required period, the notice, and the procedure vary substantially by state. A storage unit is frequently how landlords satisfy the storage obligation. What that obligation actually is in your jurisdiction is a question for a local attorney, not for us. Do not rely on a national article, including this one, for a legal timeline.

Document retention. Leases, applications, inspection records, and receipts accumulate, and the retention periods that matter to you are driven by tax and landlord-tenant law rather than convenience. If you keep paper, it needs somewhere secure and organized.

A note on what a storage unit is not. It is not a workshop and it is not a job site. Facility leases and local codes generally prohibit operating a business inside a unit, running power tools in one, or using it as workspace, and standard facility insurance excludes activity of that kind. We cover the specifics in our guides on whether you can use a storage unit as a workshop and whether storage units have electricity. Plan to load and unload, not to work.

If your storage need is inventory for a product business rather than a property operation, the calculus is different and we have written about it separately in our comparison of a storage unit versus a warehouse and our guide to inventory rotation in storage.

Sizing a Storage Unit for a Rental Portfolio

This is the arithmetic nobody publishes, so here it is. Sizes below are the standard industry footprints.

5x5 (25 square feet). Roughly a large closet. Realistic for document archives, small tools, and a modest set of fixtures. Not enough for furniture. This is the right size if your use case is records retention and a few boxes of spare hardware across a small portfolio.

5x10 (50 square feet). Holds the contents of a studio or small one-bedroom, or a full set of renovation hand tools plus materials for a single bathroom or kitchen. For a landlord with one or two units, this is usually the practical minimum for turnover work.

10x10 (100 square feet). The most common size rented in the industry and the workhorse for small portfolios. Holds the furnished contents of a one to two bedroom unit, or staging furniture for a modest flip, or tools plus materials for a full renovation. If you are running turnovers on three to five doors, start here.

10x15 (150 square feet). Contents of a two to three bedroom home, or staging inventory for multiple listings simultaneously, or a full renovation kit including appliances waiting to be installed.

10x20 (200 square feet). Roughly a one-car garage. Contents of a three to four bedroom home. This is the size that makes sense when you are furnishing multiple units, holding appliance inventory, or running renovations on several properties at once. Drive-up access matters more at this size than at any smaller one, because you will be moving large items frequently.

10x30 (300 square feet). Contents of a large home plus equipment. For most residential investors this is more than the operation requires. It becomes appropriate at portfolio scale, or when you are consolidating storage from several properties into one location.

Three sizing decisions that matter more than the square footage:

• Drive-up versus interior. If you are loading appliances, furniture, or lumber on a schedule, drive-up access saves real labor on every trip. If you are storing documents and fixtures you touch rarely, interior is fine and often cheaper.
• Climate-controlled versus standard. Our climate-controlled units are temperature-regulated, which protects stored items against extreme heat and cold. That is worth paying for on electronics, wood furniture, documents, and anything you intend to reinstall in a finished unit. Standard drive-up units are appropriate for tools, ladders, and durable materials.
• Proximity beats price. A unit fifteen minutes farther from your properties costs you time on every single trip, and turnover work involves many trips. The cheaper unit across town is frequently the more expensive choice.

If you would rather not estimate, our storage size calculator walks through it by contents, and you can browse available units by size directly. All of our rentals are month to month, which for a turnover-driven use case is the feature that actually matters: you take the unit for the six weeks you need it and give it back.

When You Should Not Rent a Storage Unit

We sell storage. We would still rather you did not rent a unit you do not need, because a unit that does not earn its keep is a recurring expense against a rent roll you are trying to optimize. Here are five situations where the answer is no.

1. You own one property and it has a garage. If you have a single rental with a garage or basement you control between tenancies, and your turnover work is measured in days rather than weeks, you do not need a unit. Use the space you already own. Renting storage for a single-property operation with on-site space available is a straightforward waste of roughly $100 to $200 a month.

2. You are storing furniture you will never reinstall. This is the most expensive mistake in the category and it is very common after an eviction or an abandonment. Once your legal obligation to hold a former tenant's property has been satisfied, storing used furniture that you will not put back into a unit costs more than replacing it within a year or two. Run that number before you sign. If the annual storage cost exceeds the replacement cost, you are financing a decision you have already made.

3. Your renovation is fully funded and scheduled. If materials are arriving on a just-in-time schedule to an active job site with secure storage, adding a unit is redundant cost. The unit earns its rent when there is a gap between delivery and installation, not when there is not.

4. You are storing documents you could digitize. Leases, applications, and inspection reports scan. If the only reason you are holding a unit is paper, the scanning cost is one time and the storage cost is monthly. Confirm your retention requirements with your CPA and your attorney first, since some records have format requirements, but for most landlord paperwork this is a straightforwardly better answer.

5. You are between deals with nothing to store. Investors sometimes hold a unit through a gap because giving it up feels like losing capacity. Our leases are month to month specifically so you do not have to do that. Give the unit back and take another one when the next project starts. Paying to reserve optionality you are not using is a cost with no offsetting return.

The single test that covers all five: if you cannot name what goes in the unit and roughly when it comes out, you are not ready to rent one.