Why Appraisals Rose Faster Than Home Prices
Knox County's average appraisal increased ~60% since 2022, even though home prices rose far less. There's a reason why.
Key Points
Knox County home appraised values increased about 60% on average since the 2022 reassessment, while market home values increased only 20–30%.
Reassessments estimate current market value, not cumulative appreciation since the last reassessment.
Large appraisal increases often reflect both market appreciation and corrections to previously undervalued properties.
The key question is not how much your appraisal increased, but whether it accurately reflects what your home could sell for today.
Use my free Property Tax Change Calculator to estimate how reassessment will affect your annual tax bill based on projected tax rates for the city and county.
Knox County’s 2026 countywide reassessment is underway. While the final numbers won’t be official until the process concludes, preliminary estimates show that assessed residential property values increased by roughly 60% on average since the last countywide reassessment in 2022.
For many homeowners, the dramatic increases reflected in their reassessment notices was difficult to reconcile with current housing market trends. And without additional context, that reaction is entirely understandable. Depending on the source, home prices in Knox County increased roughly 20% to 30% since 2022—substantial growth, but nowhere near 60%.
So what explains the gap? The short answer is that how much your property assessment changes is only loosely related to the home price growth figures cited in the news.
Unlike a home value index, a county property assessment is not a measurement of how much your home’s value changed since the last reappraisal. Rather, it is the county’s best estimate of what your property is worth today. That may sound like the same thing—but, in practice, it’s not.
Because reassessments are designed to estimate a property’s current market value, changes in appraisals often reflect more than just market appreciation. They can also reflect corrections to prior assessments that understated or overstated a property’s value. As a result, an appraisal can increase much faster than home prices in the broader market without necessarily indicating that the home appreciated by the same amount.
That is particularly relevant this time around because the last reassessment occurred during one of the most turbulent and unusual housing markets since the 1980s. Home prices accelerated rapidly during and after the pandemic, and in many cases assessments struggled to keep pace with a market that was changing fast. But to fully understand the discrepancies, it’s helpful to better understand the differences between assessments and the home price figures cited in the news.
The Difference Between Measuring a Market and Valuing a Home
Most figures cited in the news are sourced from various home price indices, which are designed to measure how home values are changing across an entire market, not the value of any specific property. To better understand the differences, it is helpful to know how some of the most common home price indices are constructed and what they’re measuring:
Median Sale Price Index: The most basic way to measure changes in home prices is to compare how the median—or middle—sale price of homes changes over time. It’s easy to understand, and offers the most durable measure of home prices because it is based on actual sales. But using the median sale price alone has important limitations. It reflects only homes that actually sold during a given period, not the entire housing stock, so it is comparing different homes from one period to the next. As a result, changes in median sale prices can reflect shifts in the types of homes being sold, not just changes in market values.
Repeat-Sale Price Index: This is the methodology used for many of the most widely cited home price indices, including the influential Case-Shiller Index. Rather than comparing different homes, a repeat-sale index tracks the same property over time. For example, if a home sold for $200,000 in 2018 and then sold again for $300,000 in 2025, the index would use that change to estimate appreciation. By comparing a property to itself, repeat-sale indices eliminate distortions caused by changes in the mix of homes being sold. When aggregated across thousands of transactions, it provides a credible—though still imperfect—measure of how home values have changed across an entire market. It is still, however, limited in that it only includes homes that have sold multiple times, meaning large portions of the housing stock are excluded.
All-Transactions Price Index: This type of index builds on the repeat-sales framework by incorporating refinancing appraisals in addition to home sales. By adding these extra observations, it captures a larger share of the housing stock and generally provides a more comprehensive measure of market-wide appreciation, particularly in areas with relatively few home sales. However, it still excludes homes that have not been recently sold or appraised, leaving a significant portion of the housing stock unobserved.
In theory, an all-transactions price index is the most comparable to the change in appraised values. But again, a price index is designed to measure price appreciation at market-level. County appraisals, by contrast, attempt to estimate the value of every property—including homes that haven’t been bought or sold in a very long time—without the benefit of an on-site inspection and a full professional appraisal. That’s a much harder task.
The limits of automated valuation
Because county property assessors can’t conduct individualized appraisals for every parcel, they must rely on automated valuation models—or AVMs—to estimate property values at scale. AVMs estimate a property’s value primarily by analyzing recent sales of comparable homes nearby. They are, in essence, algorithms that include thousands of data points: what have similar homes sold for lately, and what does that imply about this one? When there are plenty of recent sales of genuinely comparable properties, AVMs can perform reasonably well.
But an algorithm or statistical model is only as good as the data fed into it. That’s the core problem. AVMs lean heavily on whatever recent sales data is available—and in neighborhoods or more rural areas with low turnover, that data can be thin, dated, or drawn from properties that aren’t truly comparable. In such areas, a home that hasn’t sold in 15 years leaves almost no direct footprint for an AVM to work from. The model then fills that gap by reaching for the nearest proxies it can find, which may be homes of different sizes, conditions, or locations. The further the algorithm has to reach, the less reliable the estimate.
The result is that AVM-based appraisals are imprecise in ways that are not entirely random. They do not simply miss high and low in equal measure. Instead, they struggle most with the types of properties that are hardest to model: older homes, rural properties, custom-built homes, and neighborhoods with relatively few recent sales. Homes with unusual characteristics, significant renovations, or non-standard layouts are similarly difficult for an algorithm to value because there may be few truly comparable sales available to serve as benchmarks.
The same challenge exists at the top end of the market. Luxury homes tend to be more unique, sell less frequently, and exhibit greater variation in pricing than entry-level or mid-market homes. A difference in finishes, lot characteristics, architectural design, or buyer preferences can translate into hundreds of thousands of dollars in value. As a result, high-end properties are often among the most likely to be overvalued or undervalued by automated models.
On the other hand, AVMs generally perform best in neighborhoods where homes are relatively similar and transactions occur frequently, producing estimates that are at or near market value. New subdivisions with consistent floor plans, lot sizes, and construction quality provide abundant comparable sales data, allowing the model to estimate values with relative precision. Popular, up-and-coming neighborhoods with high sales volumes similarly provide AVMs a steady stream of current market information to work from.
Zillow’s AVM Lesson
No case illustrates the limitations of AVMs more vividly than Zillow’s brief, expensive experiment in home flipping. From 2018 to 2021, Zillow ran a business called Zillow Offers, in which the company used its proprietary AVM—the same technology behind the infamous “Zestimate”—to make instant cash offers on homes, renovate them, and resell them for a profit. The premise was that Zillow’s algorithm was good enough to reliably value individual homes at scale. It wasn’t. In late 2021, Zillow announced it was shutting the business down after the algorithm systematically overbid on homes, leaving the company holding thousands of properties worth less than it paid. The total loss exceeded $500 million. Zillow, a company whose entire identity is built around home valuation technology, couldn’t make AVMs work well enough to flip houses profitably—even with access to the most sophisticated AVM in the world and more data than any county assessor’s office could dream of.
The risk of overvaluation for Zillow’s home flipping business was uniquely high. Its business model required determining both an acceptable initial purchase price and post-renovation resell value while leaving enough margin between the two to generate a profit. That’s not the case for AVM-based appraisals. Unlike Zillow, property assessors are not trying to generate a profit from individual properties. Their objective is simply to estimate market value as accurately as possible. In practice, assessment models often produce valuations that modestly trail actual sale prices, reflecting both the limitations of AVMs and the desire to avoid widespread overvaluation.
None of this means AVMs are ineffective. In fact, modern AVMs are remarkably accurate when viewed across thousands of properties. But aggregate accuracy does not translate to accuracy for every individual parcel. A model can do an excellent job estimating values across an entire county while still producing meaningful errors for specific homes.
A Realistic Example
Imagine you live in a 50-home subdivision and have owned your home since 2011. Between 2011 and 2022, only five homes in the neighborhood sold. Most were either significantly larger or smaller than your home, weren’t renovated like other homes, or sat on different types of lots. With so few truly comparable sales available, the county’s valuation model estimated your home’s value at $200,000, even though you could probably sell it for $330,000.
Then, between 2022 and 2026, another 15 homes in the subdivision sell, including five that are nearly identical to yours in size, age, and condition. Those homes sell for around $370,000. Armed with substantially more—and higher quality—sales data, the AVM is now able to estimate your home’s value much more accurately and appraises it at $370,000.
In this case, the home’s appraisal increased by 85% between reassessments yet the home’s actual market value increased by only about 12% over the same period. The dramatic increase in the assessment is not because the property assessor—or the AVM—thinks your home appreciated 85% in last four years. Rather, it reflects the fact that the previous appraisal understated your home’s value due to weak comparable-sales data.
Does the assessment reflect current market value?
The fact that appraisals increased faster than home prices is not, by itself, evidence that the property assessor got it wrong. A large increase often reflects both real market appreciation and the correction of a prior appraisal that was already below market value. Still, automated models are no infallible. They can mischaracterize a property’s condition, square footage, or comparable sales. Homeowners in lower-turnover neighborhoods, or with properties that don’t fit neatly into standard categories, have particularly good reason to scrutinize their assessments carefully.
If your appraisal feels wrong, the relevant question is not whether it increased more than the market, but whether the county’s estimate of your home’s current market value is reasonably accurate. You can assess this by reviewing recent sales of comparable properties or, better yet, asking a Realtor to conduct a market analysis. If similar homes are consistently selling for less than your appraised value, you may have grounds for an appeal.
The appeals process exists precisely for this reason. The key question is whether the county’s estimate reflects what your home would reasonably sell for today. If it doesn’t, there’s a process in place to rectify the error.
Property Tax Change Calculator
Because of Tennessee’s truth-in-taxation law, a higher assessment does not automatically translate into a proportional increase in property taxes. Preliminary estimates indicate that residential assessments increased by about 60% on average during the 2026 reassessment cycle. As a result, homeowners whose assessments increased by less than 60% will see their tax bill decline, while those whose assessments increased by more than 60% will see their tax bill increase.
To estimate how the reassessment may affect your property specifically, use the property tax change calculator linked below.
Other Thoughts On Reappraisal
You Probably Aren’t Paying Taxes On Unrealized Home Equity: According to a report the National Association of Realtors®, 12.8% of homeowners in Knoxville—or 34,723 households—have accrued home equity that exceeds the federal capital gains exclusion of $250,000 for individuals and $500,000 for couples, and thousands more have substantial equity but not enough to exceed the threshold. That raises an interesting question: are you being taxed on unrealized, illiquid home equity? For the vast majority of homeowners, the answer is probably not.
That’s because, in Tennessee, residential property is assessed at just 25% of its appraised value. In other words, homeowners are taxed on only a fraction of their property’s market value. For example, if you purchased a home for $250,000 in 2015 and it is now worth $500,000, your assessed value—the portion of the property subject to taxation—is only $125,000. Despite having accrued $250,000 in equity, you are still being taxed on only half of what you originally paid for the home.
There are, of course, some exceptions. A small share of residents those who have owned their home for a very long time—most of which own their home outright and do not have a mortgage—have assessed values that exceeds their original purchase price. In those cases, at least a portion of their property taxes is effectively based on unrealized equity. If Tennessee were to adopt a law that forbid assessed values from exceeding the original purchase—a change that would likely require a state constitutional amendment—these homeowners would pay next to nothing in property taxes despite utilizing the same publicly funded infrastructure and services as everyone else.
And this does not even begin to address the more fundamental question of whether taxing unrealized home equity is inherently unfair. A home’s value, and the equity it accrues, is not determined solely by the structure and land it sits on; its value also reflects, at least in part, the quality of the roads, schools, stormwater systems, parks, public safety services, and other public investments that surround it—most of which are funded primarily through property taxes. If a portion of one’s home equity, even if it is unrealized, is derived directly from public investment, is it necessarily unfair to tax it?
Ultimately, I do not find this question particularly interesting because most homeowners are not actually paying taxes on unrealized home equity in any meaningful sense.
Property Taxes Are Regressive. But the Alternatives Are Usually Worse.
One of the most common criticisms of property taxes is that they are regressive, meaning its a tax that burdens lower-income households more than higher-income ones. A retiree living on a fixed income and a high-income professional may own homes of similar value and therefore face similar tax bills despite having vastly different economic profiles and abilities to pay. That concern is legitimate and real.
The challenge, however, is that local governments still need revenue to fund the infrastructure and services that make communities desirable places to live, and that ultimately keep home prices stable. Roads must be maintained. Schools must be staffed. Parks must be maintained. Police officers must be paid. The question is not whether these services should be funded, but how. And when grapple with that question honestly, it becomes clear that there aren’t really any other more appealing alternatives.
Income taxes are more progressive, but Tennessee prohibits local governments from levying an income tax—and for good reason. Sales taxes are already a major source of local revenue, but they are even more regressive because lower-income households spend a larger share of their income on taxable goods and services. They are also capped by the state and, without imposing one of the highest sales tax rates in the nation, could never generate enough revenue to replace what is generated by property taxes. User fees can help pay for specific services, but they are modest and generate comparatively limited revenue. Local governments can also increase reliance on state and federal aid, but doing so reduces local control and leaves localities vulnerable to budget decisions made elsewhere.
Property taxes, for all their flaws, offer several advantages. They are relatively stable, difficult to avoid, and closely tied to the local assets that benefit from public investment. A new road, better schools, improved public safety, or well-maintained parks generally make a community more desirable, which in turn tends to increase nearby property values. In that sense, property taxes function, at least in part, as a mechanism for recapturing some of the value that public investment helps create.
Ultimately, the question is less about whether property taxes are flawed—they are—and more about whether there is a better way to generate the revenue necessary to sustain the public services and infrastructure that underpin both quality of life and property values. To advocate for curtailing or eliminating property taxes (which is a real discussion in places like Florida) also requires a realistic discussion of what would replace them. And the reality is that there is no clearly superior alternative.
There are many other engaging questions about property taxes and the best way to fund local government. But I’ll save those for another day and leave it here.