Why a Rounded Payment Can Leave a Balance
A level payment is usually calculated from a present-value equation. The exact result may have fractions of a cent, while actual payments are made in cents. Rounding the regular payment therefore changes the amount paid over the life of the loan. Even when the difference in each payment is small, interest causes those differences to accumulate.
The final payment is often adjusted to clear the remaining amount rather than repeating the rounded regular payment. To calculate it correctly, keep track of the direction of the rounding error: a payment rounded down leaves a balance owing, while a payment rounded up can create a small overpayment.
The balance calculation assumes interest is applied at the stated rate to the balance before each payment, and payments are made at the end of each period. Unless stated otherwise, retain full precision in the calculations and round only the payment amounts that are actually made. If a lender rounds interest separately each period, the resulting schedule can differ slightly from this model.
How the Payment-Rounding Error Accumulates
Let \(B_0=L\). After interest is added and the rounded payment is made, the balance follows \(B_k=(1+i)B_{k-1}-P\). The exact payment \(R\) would instead leave a zero balance at the end of period \(n\). Comparing these two schedules isolates the effect of rounding.
Let \(D_k\) be the difference between the balance under the rounded payment and the balance under the exact payment, immediately after payment \(k\). Since both schedules start at the same principal, \(D_0=0\). At each period, the previous difference earns interest, and the rounded schedule pays \(\varepsilon\) more than the exact schedule. Thus \(D_k=(1+i)D_{k-1}-\varepsilon\). Repeated substitution gives the accumulated effect:
The exact-payment schedule has a zero balance after \(n\) payments, so the rounded-payment balance is \(B_n=D_n\). Therefore:
This result gives the balance that would remain if the rounded regular payment were made even in the last period. In practice, the last payment is usually changed. Just before the final payment, the balance has earned one more period of interest; the amount then due is the adjusted final payment.
This identity follows directly from the recurrence: after the hypothetical rounded final payment \(P\), the remaining balance would be \(B_n\). The amount due immediately before that payment must therefore be \(P+B_n\). If the final amount is itself rounded to cents, the paid amount may leave a tiny residual or overpayment. Report that practical consequence rather than implying that a cent-rounded payment always clears the balance exactly.
Use the loan principal, payment-period rate, and number of payments to calculate \(R=L/a_{\overline{n}|}\).
Set \(P\) to the required currency precision and calculate \(\varepsilon=P-R\), keeping its sign.
Use the balance recurrence period by period, or use \(B_n=-\varepsilon s_{\overline{n}|}\) for the hypothetical balance after \(n\) rounded payments. The adjusted final amount is \(P+B_n\).
Round the adjusted amount to cents only after calculating it. If needed, report the small residual caused by this last rounding.
Worked Applications
Worked Example: A Payment Rounded Up
A hypothetical loan of $6,000 is repaid by two annual payments at an effective annual interest rate of 4%. The exact level payment is rounded to the nearest cent for the first payment. Find the signed balance after two payments of that rounded amount, then determine the adjusted final payment to the nearest cent. Assume interest is applied before each end-of-year payment and no intermediate interest is rounded.
Method. First calculate the exact payment, then apply the balance recurrence. This avoids treating the cent-rounded payment as if it were still the exact level payment.
The exact payment is found from the present value of two end-of-year payments:
Rounded to cents, \(P=\$3,181.18\), so \(\varepsilon=3181.18-3181.176470588\ldots=0.003529411\ldots\). The balance after the first payment is:
After the second year’s interest, the amount due is \(3058.82(1.04)=3181.1728\). If the regular payment of $3,181.18 is made again, the resulting balance is \(3181.1728-3181.18=-\$0.0072\). The negative sign indicates a small overpayment. This agrees with the accumulated-error formula:
The exact adjusted final amount is \(P+B_2=3181.18-0.0072=\$3,181.1728\). Rounded to cents, the final payment is $3,181.17. That cent-rounded payment is $0.0028 below the exact amount due, so a residual of $0.0028 remains under the full-precision model. The calculation checks: the exact due amount rounds to $3,181.17, and rounding the regular payment up produced a small overpayment when the regular amount was repeated.
Common mistake avoided: Do not add the magnitude of the negative balance to the final payment. A negative balance means the repeated rounded payment is too large, so the adjusted payment is smaller.
Worked Example: A Payment Rounded Down Over Three Years
A hypothetical $1,500 loan is repaid by three annual end-of-year payments at an effective annual rate of 10%. The exact level payment is rounded to the nearest cent. Calculate the balance after three rounded payments and the adjusted final payment to the nearest cent. Assume full-precision interest calculations.
Method. Use the exact level payment to identify the rounding direction, then verify the ending balance by advancing the schedule one period at a time.
The regular payment is \(P=\$603.17\), so \(\varepsilon=603.17-603.172205438\ldots=-0.002205438\ldots\). The rounded payment is slightly too small. The schedule gives:
Thus the balance after three payments of $603.17 is $0.0073, positive because the regular payment was rounded down. The accumulated-error check gives the same result: \(s_{\overline{3}|}=1+1.10+1.21=3.31\), so \(B_3=-\varepsilon(3.31)=0.0073\).
Before the final payment, the amount due is \(548.343(1.10)=\$603.1773\). The adjusted final amount is also \(P+B_3=603.17+0.0073=\$603.1773\), which rounds to $603.18. Paying $603.18 leaves an overpayment of $0.0027 in the full-precision model. The two calculations of the final amount agree, checking the result.
Common mistake avoided: The positive balance after three hypothetical regular payments is not the final payment itself. The final payment must cover the balance accumulated through the third year, which is \(P+B_3\).
Worked Example: Matching a Monthly Rate to Monthly Payments
A hypothetical $2,500 loan is repaid by two monthly end-of-month payments. The nominal annual interest rate is 12% convertible monthly, so the effective rate per payment period is \(0.12/12=0.01\). The exact level payment is rounded to cents. Find the balance after two rounded payments and the adjusted final payment to the nearest cent. Assume no intermediate rounding of interest.
Method. Convert the nominal rate to the monthly rate before calculating either payment or balance. This avoids using an annual rate with monthly periods.
The regular payment is \(P=\$1,268.78\), which is below the exact payment. After the first month, the balance is \(2500(1.01)-1268.78=2525-1268.78=\$1,256.22\). After the second month’s interest, the amount due is \(1256.22(1.01)=\$1,268.7822\). If $1,268.78 is paid again, the remaining balance is \(1268.7822-1268.78=\$0.0022\).
The formula confirms this: \(\varepsilon=1268.78-1268.781094527\ldots=-0.001094527\ldots\), and \(s_{\overline{2}|}=1+1.01=2.01\), so \(B_2=-\varepsilon(2.01)=\$0.0022\). The adjusted final amount is \(P+B_2=1268.7822\), which rounds to $1,268.78. Paying that cent-rounded amount leaves $0.0022 owing under the model. The balance recurrence and accumulated-error calculation agree.
Common mistake avoided: The nominal rate is not the rate for each monthly period. Using 12% per month would substantially overstate interest; the stipulated monthly rate is 1%.
Interpreting the Result and Avoiding Traps
The signed balance is useful because it preserves the direction of the difference. A positive value is an amount still owed; a negative value indicates that repeated rounded payments would pay more than required. The sign should not be discarded before adjusting the final payment.
| Regular payment rounding | Sign of \(\varepsilon\) | Sign of \(B_n\) | Effect on adjusted final amount |
|---|---|---|---|
| Rounded down | Negative | Positive | Increase the final amount due |
| Rounded up | Positive | Negative | Decrease the final amount due |
| Exact to the stated precision | Zero | Zero | No adjustment under the model |
Three distinctions prevent most errors. First, keep the exact payment \(R\) separate from the rounded payment \(P\); the error is \(P-R\), not \(R-P\). Second, identify whether the question asks for the balance after a hypothetical final regular payment or the amount required just before that payment. Those quantities differ by \(P\). Third, round the adjusted final amount only after calculating it. A result such as $3,181.1728 is the model’s exact amount due; $3,181.17 is the cent-settled payment and may leave a fraction of a cent.
These calculations matter when interpreting an amortization schedule, quoting the final installment, or checking why a displayed balance is not exactly zero. Small differences can also arise if a schedule rounds interest charges each period, uses a different convention for payment timing, or applies a different payment frequency. The formulas here apply to end-of-period payments at a constant effective rate per payment period, with interest calculated at full precision.
Check Your Understanding
Assume end-of-period payments, full-precision interest calculations, and cent-rounded regular payments unless a question states otherwise.
- A hypothetical $8,000 loan has two annual payments at an effective annual rate of 3%. Find the balance after two payments of the cent-rounded exact level payment and the adjusted final payment to cents. (Answer: balance \(-\$0.0067\); final payment $4,180.88.)
- A hypothetical $9,000 loan has two annual payments at an effective annual rate of 7%. Find the signed balance after two rounded regular payments and the adjusted final payment to cents. (Answer: balance \(-\$0.0081\); final payment $4,977.82.)
- A hypothetical $2,000 loan has zero interest and seven equal annual payments. Find the cent-rounded regular payment, the balance after seven such payments, and the adjusted final payment. (Answer: regular payment $285.71; balance $0.03; final payment $285.74.)
- A hypothetical $10,000 loan has two annual payments at an effective annual rate of 6%. Find the balance after two rounded regular payments and the adjusted final payment to cents. (Answer: balance \(-\$0.0022\); final payment $5,454.37.)
- A hypothetical $1,500 loan has three annual payments at an effective annual rate of 10%. If every payment, including the last, is initially set to the rounded regular amount, what is the ending balance? What final payment should be substituted, rounded to cents? (Answer: ending balance $0.0073; substitute a final payment of $603.18.)