Ask a client what their depreciation expense was last year and they will read a number off the QuickBooks P&L. Ask their tax return the same question and you will get a different number. Both are correct.
That gap is not an error. It is the most reliable book to tax difference on a business return, it turns up on nearly every 1065, 1120 and 1120S, and it belongs on Schedule M-1 as a reconciling item rather than being quietly fixed in the client's books.
Here is why the two numbers differ, which one goes where, and the specific shortcut that turns a routine difference into an out of balance balance sheet.
The short answer
The books use whatever method the client's bookkeeper set up. Straight line over an estimated useful life, usually. The return uses MACRS, plus Section 179 and bonus depreciation where elected.
Those two systems produce different expense in every year of an asset's life, not just the first. The total over the full life is identical, which is the part people find reassuring and also the reason the difference eventually reverses.
The tax number goes on the return. The book number stays in the books. The difference goes on Schedule M-1. You do not journal the tax number into the client's general ledger.
Why they differ, mechanically
Three things drive the gap.
Method. Book depreciation is usually straight line. MACRS is a declining balance system for most asset classes, so it front loads the deduction. A 50,000 machine on a five year book life is 10,000 a year. Under MACRS five year property, using the half year convention, it runs 10,000 then 16,000 then 9,600, 5,760, 5,760 and 2,880. Same 50,000 in total. Nothing alike year to year.
Life. The bookkeeper picks a life that reflects how long the asset will really be useful. The IRS assigns a recovery period from a table. A computer might be four years in the books and five under MACRS. A commercial building might be 40 years in the books and 39 under MACRS, which sounds close until you multiply it by the basis.
Elections. Section 179 and bonus depreciation can expense most or all of an asset in the year it goes into service. There is no book equivalent. A client who bought 80,000 of equipment and elected 179 on all of it has 80,000 of tax depreciation and maybe 11,000 of book depreciation that year. One election, a 69,000 M-1 adjustment.
Which direction the adjustment goes
This is where people slow down, so be concrete about it.
Schedule M-1 starts at net income per books and works toward taxable income. The only question is whether the return needs more income than the books show, or less.
Tax depreciation larger than book. The return takes a bigger deduction, so taxable income is lower than book income. The adjustment reduces income, landing in the M-1 section for deductions on the return not charged against book income.
Book depreciation larger than tax. This happens in the later years of an asset's life, once the front loading has reversed. The books took the bigger deduction, so taxable income is higher than book income. The adjustment increases income, in the section for expenses recorded on the books not deducted on the return.
A firm with a stable fleet of assets bought across many years will have both directions running at once and net to something small. A firm that bought heavily last year will show one large adjustment in one direction. Neither is suspicious by itself.
Accumulated depreciation on Schedule L stays at book
Here is the part that quietly breaks balance sheets.
Schedule L is a book balance sheet. The accumulated depreciation line ties to the accumulated depreciation account in the client's general ledger. It does not tie to the accumulated total your tax depreciation schedule has been carrying.
The asset module is running its own accumulated figure, and after several years of MACRS and 179 elections that figure will be far larger than the book one. Pull it onto Schedule L because it happened to be the number in front of you and the asset side of the balance sheet drops by the difference. Schedule L then fails to balance by exactly that amount.
So when a balance sheet is out by an oddly specific number that happens to equal a depreciation difference, this is nearly always the reason. If you are chasing one right now, we wrote a separate piece on how to find a Schedule L difference.
The shortcut worth naming
The tempting move is to book tax depreciation straight into the client's general ledger so everything agrees and no M-1 adjustment is needed. One number, no reconciliation.
Do not.
The client's financial statements now report MACRS depreciation, which is not what their bank, their bonding company, or a future buyer expects to see. Fixed asset book value becomes a tax construct instead of an estimate of remaining useful value. And the M-1 stops documenting anything, because you deleted the difference it exists to explain.
The reconciliation is not busywork. It is the record of a real difference between two systems that are supposed to differ. Remove it and the next preparer cannot tell whether the number was reconciled or plugged.
There is a narrower version of the same problem: posting the current year depreciation entry twice, once from the depreciation schedule and once from the client's own books, because the bookkeeper had already been accruing it monthly. That doubles book depreciation, moves net income per books, and throws the M-1 off by the duplicated amount. It is easy to miss because both entries look reasonable in isolation.
Where this actually goes wrong
Almost never in the arithmetic. The arithmetic is a subtraction.
It goes wrong in the handoff. The tax figure comes from a fixed asset schedule that lives outside the trial balance, often inside the tax software's asset module, sometimes in a spreadsheet nobody has opened since last April. The book figure comes from the general ledger. Someone has to hold both at once, compute the difference, decide the direction, and put it on the correct M-1 line.
That handoff is usually a person retyping numbers between two windows at the end of a long day. Which is exactly where 16,000 becomes 1,600 and neither system objects, because both numbers are internally consistent. The return balances. It is simply wrong.
Where Ledger IQ fits
Ledger IQ keeps both systems visible at once instead of asking you to hold them in your head.
You upload the client's trial balance, which carries book depreciation and book accumulated depreciation exactly as the general ledger has them. Those map to the Schedule L lines at book, which is where they belong.
The book to tax difference goes in as a tax journal entry. TJEs adjust the tax balance column of the Working Trial Balance without touching the exported book figures, so the workpaper shows net income per books, the depreciation difference, and the resulting taxable income as three separate visible numbers instead of one plugged total. That column is the M-1 in workpaper form.
Because the tax software computes depreciation from its own asset module, Ledger IQ marks depreciation and fixed asset lines as not exported on purpose. They appear in your workpaper so it ties and so the difference is documented, and they stay out of the import file so they cannot fight the asset module or double up.
The tie out view closes the loop. Once the return is populated, key the as filed amounts back in and Ledger IQ flags any line that disagrees with the workpaper. A depreciation figure transposed on its way into the asset module shows up as a difference on the accumulated depreciation line, before the return goes out rather than after a notice.
Ledger IQ supports Forms 1065, 1120S and 1120, with exports for Drake, Lacerte and UltraTax CS. It is free during early access, so you can try it at portal.ledgeriq.ai on a live client without a credit card.