In a 10-year personal finance projection, most people treat the mortgage as background noise. The balance drifts down by some percentage each year, the interest goes off to the lender, and the rest of the model focuses on the asset side. That is almost right, but the small wrongness adds up over a decade, and the place it goes wrong is the shape of the amortization curve.
A standard fixed-rate mortgage does not pay down its principal evenly. The first years are almost entirely interest, the later years are almost entirely principal. A projection that linearly amortizes the balance underestimates net worth in the early years and overestimates it in the later years, often by amounts that matter when the planning question is "can I retire in 2036."
This piece walks through why the shape matters, what it looks like in numbers, and how to incorporate it without writing a finance library.
Photo by Jakub Zerdzicki on Pexels
What the curve actually looks like
A 30-year fixed mortgage at 6.5 percent, on a $400,000 starting balance, pays out roughly like this in year one: about $2,000 of the $2,530 monthly payment is interest, and $530 is principal. By year 15, that ratio has flipped. By year 25, almost the entire payment is principal.
Over 10 years, the balance does not drop by 10 of 30 (one-third). It drops by closer to 17 percent of the original principal, because the early years are interest-heavy. The remaining balance at year 10 on the example above is roughly $332,000, not $266,000. That is a $66,000 difference, and in a net worth projection, $66,000 of underestimated debt is $66,000 of overstated net worth.
The math is not exotic. The Consumer Financial Protection Bureau publishes example amortization tables and the standard formula. A lender's amortization schedule, available from your loan servicer, shows the exact numbers for your specific loan.
Why the back-loaded shape happens
Each month the lender computes interest on the remaining balance. When the balance is high, the interest portion of the fixed payment is also high. As the balance drops, less of the payment goes to interest, and more goes to principal. The math compounds in the opposite direction from asset growth: instead of accelerating outward, the principal paydown accelerates inward.
The Wikipedia article on amortization schedules lays out the formal version. The Federal Reserve consumer credit data is a useful reality check on how household balances typically behave over the lifetime of a mortgage.
The structural reason matters because it means there is no simplification of the curve that gets close to the right answer. Linear approximations always understate the early balance and overstate the late balance. Polynomial approximations are not much better. The only way to get the right number for any given year is to use the actual amortization schedule.
The contribution feedback loop
Here is the part of the story that makes amortization shape matter more than people expect. As the mortgage payment shifts from interest to principal, the principal portion of the payment is functionally a savings contribution. The dollars are not in a brokerage account, but they are accreting as home equity rather than going to the lender as interest.
A projection that ignores the shape of amortization is also ignoring the gradual ramp-up in implicit savings. By year 7 or 8 of a mortgage, the household is "saving" a meaningful amount through principal payments that they were not saving in year 1, even though the payment amount is identical.
This is also why mortgage prepayment math is harder than it looks. Paying extra principal in year 1 has a large effect on lifetime interest. The same dollar in year 25 saves much less. The hub article on 10-year net worth projection walks through how this interacts with the per-bucket return model.
The reverse is also true. A cash-out refinance or a HELOC draw resets the amortization curve back toward the early-years shape, where most of the payment is interest. Households that periodically refinance their primary residence are also resetting the amortization curve back to high-interest mode, which has implications for the long-term net worth projection that the household often does not fully model.
How to incorporate the real schedule
The simplest version is to pull the actual amortization schedule from your loan servicer and copy the year-end balances into the projection. Most servicers publish a downloadable schedule in the loan documents portal. Twelve numbers (year-end balances for years 0 through 10) is enough. Each year, the projection subtracts that year's principal balance from total assets.
If a downloadable schedule is not available, the standard amortization formula reproduces the same numbers exactly. Spreadsheet functions like PMT, IPMT, and PPMT in Excel or Google Sheets compute the interest and principal portions of each payment. The Bureau of Labor Statistics Consumer Expenditure Survey can provide a sanity check on what total household debt service typically looks like as a share of income, if you want to compare the projection against population norms.
The same approach works for student loans and car loans, both of which also follow amortization curves (though car loans are short enough that the linear approximation is close to right for projection purposes). Student loan refinances often change the rate without changing the term, which means the curve shape changes too, and the projection should be re-run from the new amortization schedule.
What about variable-rate mortgages
Adjustable-rate mortgages introduce a second variable: the interest rate itself can change at scheduled reset points. A projection with an ARM should run two scenarios: rate stays at its current level for the projection window, and rate moves to its cap. The gap between the two is the model's exposure to rate risk. Most fixed-rate projections do not need this, but it is worth knowing the test exists.
The free net worth tracker by EvvyTools at evvytools.com lets you record mortgage balance, rate, and term as separate inputs, which makes it easier to update the projection each quarter as the actual balance drifts down on its real schedule. The tools directory carries the other personal finance calculators in the same series, and the EvvyTools blog walks through how the amortization model fits into the broader 10-year projection workflow.
A short checklist
- Pull the actual amortization schedule from the loan servicer, or generate one with PMT and IPMT functions.
- Use year-end balances for years 0 through 10 in the projection, not a linear approximation.
- Recognize that principal paydown in later years is functionally a savings contribution.
- For adjustable-rate mortgages, run a rate-cap scenario alongside the base case.
- Update the actual balance every quarter, not every January.
A 10-year projection that gets amortization shape right will produce a slightly different number than the linear version. The number is more honest, and the planning decisions that rest on it are more defensible. Over a 30-year mortgage, the cumulative effect of getting this shape right (and using it to time prepayments correctly) can shift net worth by tens of thousands of dollars in either direction. It is not the largest variable in a projection, but it is one of the easier ones to model accurately, and ignoring it gives up easy precision for no gain.
