Start with a fixed balance
A fictional annual charge of 0.6% on a constant $2,500 balance is $15: $2,500 multiplied by 0.006. Multiplying by 0.6 instead would produce $1,500, a hundredfold error. Write the decimal conversion visibly until the relationship is familiar.
The word annual describes the rate's time basis. If a simplified arrangement charges one quarter of that annual amount each quarter on the same unchanged balance, each payment is $3.75. Do not apply the full annual rate four times unless the actual terms explicitly describe a different charge.
A changing balance needs another assumption
Suppose the account holds $2,500 for the first half of a year and $3,500 for the second half. Under a simple time-weighted illustration, the average balance is $3,000 and the annual charge at 0.6% is $18. This assumes equal half-year periods and ignores daily fluctuations.
An actual provider may use daily or month-end values, so this estimate should not be presented as an exact invoice. Check the contractual calculation base, including whether uninvested cash and particular products count. Two identical headline rates can apply to different amounts.
Separate a rate change from a cost change
If the rate falls from 0.6% to 0.4%, it falls by 0.2 percentage points, or one third of the original rate. On the constant $2,500 example, the charge falls from $15 to $10, a $5 difference. Keeping the units explicit avoids confusing percentage points with money.
Minimums, caps and other fees can alter the final bill. A hypothetical $12 annual minimum would make the second charge $12 rather than $10. The percentage calculation remains useful, but it is not necessarily the amount payable.
Use a worksheet with rate, decimal equivalent, balance basis, period and adjustments. Label estimates as estimates and compare them with the later statement. This makes a small arithmetic exercise into a method for checking whether the provider applied the terms expected, without claiming that a lower percentage alone identifies the better account.
