← The PropTech Desk

How home valuation tools work

The Adjustment Arrives Late

6 min read

Appraisal standards already contain a correction for the time between a comparable sale and the valuation in front of you. It is required, it is enforced, and where the cost of money has just moved it is close to useless. Not because anyone applies it badly. Because of what it is allowed to measure.

Fannie Mae's Selling Guide, section B4-1.3-09, is direct about the obligation. The appraiser "must analyze comparable sales for any changes in market conditions from their contract dates through the effective date to determine whether time adjustments are warranted," and "time adjustments, or the lack thereof, must be supported by evidence." Fannie Mae has since made the posture plainer still: failure to make market-derived adjustments, including time adjustments, is unacceptable.

The evidence that may support one is listed. Home price indices. Statistical analysis. Modelling. Paired sales. Other commonly accepted methods.

Every item on that list is a record of prices that have already moved.

What a rate move does first

A change in mortgage rates does not move prices on the day it happens. It moves borrowing capacity, immediately and exactly, for every buyer holding a monthly payment they can afford rather than a purchase price they have chosen.

The 30-year fixed averaged 6.65% in the week ending 20 August 2026 and 6.95% in the week ending 17 September, with nineteen of those thirty basis points arriving in the final week. The Federal Reserve raised its target range to between 3.75% and 4.00% on 16 September, unanimously.

Take a $416,000 purchase with twenty percent down, a loan of $332,800 over thirty years. At 6.65% the payment is $2,136. Hold that payment and move the rate to 6.95%, and the principal it supports falls to $322,753. The buyer who signed a contract on 20 August was carrying $10,047 more loan than the buyer being priced against that sale today. Three percent of the loan amount, in four weeks, on identical income.

Price evidence of that shift does not exist yet, and cannot. It requires contracts signed after the move, closings thirty to sixty days behind those, and an index published after that. Over the same August to September window the national median list price fell one percent, which is the residue of earlier conditions rather than a reading of this one.

So a time adjustment supported by price movement across that interval lands somewhere around one percent, or is reported as unidentifiable. The capacity gap is three.

The sentence that closes the loop

The guide anticipates this, in a line that reads as housekeeping and is the whole problem:

"Comparable sales with a contract date that is recent in relation to the effective date of the appraisal will likely not have a time adjustment given the inability to identify a change in the market."

That is correct as written. A change cannot be identified in price data that has not been generated. But it means the adjustment is at its weakest precisely where the distortion is freshest, and at its strongest where the distortion is old enough that everyone had already priced it in.

Run the same arithmetic across three vintages and the shape is plain.

Capacity gap by age of comparable

Comparable contractedRate that weekCapacity gap to todayTime adjustment likely
Week ending 10 September 20266.76%$6,377, or 1.92%None. Too recent to identify
Week ending 20 August 20266.65%$10,047, or 3.02%Around one percent at most
September 20256.26%$22,916, or 6.89%Supportable, and measuring price

Rates: Freddie Mac Primary Mortgage Market Survey, weekly 30-year fixed averages. Capacity: monthly payment held constant on a $332,800 loan over thirty years, solved for supported principal. Derived arithmetic, reproducible in a spreadsheet.

Why this is not a data quality problem

Nothing here is wrong. The comparable sale is real, the price is complete, the record is accurate, and the appraiser applying no adjustment to a four week old sale is following the guide correctly. There is no error to find and no party to blame, which is why the gap persists quietly rather than being caught.

It differs from the case where the recorded price itself is incomplete, as when a builder's forward commitment delivers a rate concession that never reaches the settlement statement. That is a hole inside the number. This is a sound number produced by a buyer standing in different conditions.

It also differs from the familiar complaint that automated tools work from stale data. Pulling this week's rate to the basis point does not help, because the comparable sales are old by construction. A comparable has to have happened.

What is missing is a field. The record carries square footage, bedrooms, condition, contract date and price. It carries nothing about the cost of money the buyer faced, and there is nowhere it would go. Recency stands in for it, and recency is a proxy that works in flat conditions and fails in exactly the weeks when a valuation is hardest to defend.

Does an appraisal adjust for mortgage rates changing between the comparable sale and today?

Not directly. The appraiser is required to analyse changes in market conditions between each comparable's contract date and the effective date, and to make a time adjustment where the evidence supports one. That evidence is price based: indices, paired sales, statistical analysis. A rate move shows up in it only after transactions clear at the new borrowing capacity, so the adjustment reflects the consequence rather than the cause, some months later.

What is a market conditions adjustment and what evidence does an appraiser need?

It is an adjustment to a comparable sale for movement in the market between the date that sale went under contract and the effective date of the appraisal. Fannie Mae requires that it be supported by evidence, and that the absence of one be supported too. Acceptable methods include home price indices, statistical analysis, modelling and paired sales, with the reasoning documented in the report.

Why do recent comparable sales usually get no time adjustment?

Because no change can be demonstrated yet. The Selling Guide says comparables with a contract date recent in relation to the effective date will likely carry no time adjustment given the inability to identify a change in the market. That is a statement about available evidence rather than about whether conditions moved, and in a week where rates jump it is the one comparable that has moved most.

What an agent does with this

Date the comparable set by contract date rather than closing date, because contract date is when the buyer's rate was fixed and it is the date the guide itself measures from. A sale that closed last week may have been negotiated in July.

Then decide whether the set spans a period in which the cost of money moved, and say so on the page. A set drawn across a thirty basis point month is not a single body of evidence, and a range that widens for a stated reason is a stronger document than a narrow one that quietly averaged two different markets.

Where a comparable set spans that kind of interval, the dispersion of the adjusted values reflects it, and the confidence assessment should widen with it rather than assert a precision the evidence cannot carry. The reasoning about contract dates belongs in the agent's own commentary, since no data source holds it, and it needs to survive into week six when the appraisal is contested. Keeping it on the deal record rather than in memory is the difference between a position and a recollection.

The standard will catch up. It will catch up the way it always does, through price evidence, once enough transactions have cleared at the lower capacity for an index to register them. By then the houses priced this month will have sold, or not.


This article is general information and analysis, not financial, lending, or appraisal advice. Verify any home value with a licensed professional before acting.

The Independent Agent
Substack | Spotify | CMAflow FAQ | YouTube | Free CMA | Home valuation | Insights | Blog