Chasing every deduction costs more than it returns

There is an entire content economy built on the premise that you are leaving money on the table. It's a good premise, because it's occasionally true and always uncomfortable. Nobody wants to be the person who didn't know.

It is an easy premise to be captured by. It produces running notes on phones headed "possible deductions," forum threads read at midnight, and parking receipts photographed in the rain on the theory that they might, under some interpretation, count.

What ends it is an honest accounting of the whole project. Here is that accounting.

The arithmetic nobody puts in the headline

The thing that breaks the spell is understanding what a deduction is actually worth, which is not what it says on the tin.

A deduction reduces the income you're taxed on. It does not reduce your tax bill by its own amount. So a $200 deductible expense, for someone in a middling tax bracket, is worth somewhere in the region of forty or fifty dollars off the final bill — not two hundred.

This is obvious once stated and it is not obvious to a great many people, because every piece of content on the subject describes the expense, not the benefit. "Don't forget you can deduct X!" is technically accurate and psychologically misleading, because the number attached to X is roughly double to triple what you'll actually feel.

It is easy to spend years optimising a number you are silently multiplying by about two and a half in your head.

What the marginal ones cost

Working with the real numbers, deductions sort into two piles.

The first pile is large and obvious: the things clearly and unambiguously applicable to your situation, each worth a meaningful amount, each requiring approximately no effort to identify because the software asks about them directly. This pile is worth almost all of the money.

The second pile is everything else. Small amounts, ambiguous applicability, requiring records you would not otherwise keep, occasionally requiring you to read a forum argument between two anonymous people about the definition of a word. This pile takes the overwhelming majority of the effort and returns a small fraction of the benefit.

It is worth putting the cost of that second pile in plain terms, because "time spent" undersells it:

• Ambient background anxiety. Once you're in the mode of thinking any expense might be deductible, no purchase is neutral. Every coffee is a small unresolved administrative question. This is a genuinely unpleasant way to move through a year.
• Record-keeping overhead. The marginal deductions are usually the entire reason a filing system is complicated. Removing them simplifies the whole thing enormously.
• Decision distortion. This is the alarming one. People catch themselves making a purchase partly because it might be deductible. Spending a hundred real dollars to save maybe thirty is not a saving. It's a discount on something you weren't going to buy.

The bit where this might be wrong

The other side deserves a fair hearing, because the calculus isn't universal.

If you're self-employed with substantial genuine business expenses, the second pile isn't small — it's your actual cost of doing business, and being rigorous about it is straightforwardly correct. The advice to be thorough is aimed at that person and it's right for them.

Where it goes wrong is that the same advice gets served to everyone, including people whose lives contain almost no legitimately deductible marginal spending, and for whom the entire exercise is a hunt for something that isn't there. It can take years and one honest spreadsheet to notice which group you're in.

The other thing to concede: some of the diligence has value that isn't the deduction. Photographing receipts makes people more aware of what they spend. That's real. But there are direct ways to become more aware of what you spend, and they don't require a tax pretext.

What works better

The alternative is short enough to state completely.

Answer the questions the software asks, honestly and carefully, because the large obvious deductions are the ones it asks about and they're worth nearly all of the money. Once a year, read one current summary from an official source to check nothing significant has changed for your situation. If something genuinely large and unusual happened in the year — a move, a big medical event, a change in how you earn — buy an hour of professional time, because that's exactly the case where an hour pays for itself many times over.

Then stop. Skip the forums. Give up on the theoretical maximum.

The thing worth saying to anyone starting out

The premise of the whole genre is that there's an optimal return sitting just out of reach, and that the gap between your return and the optimal one is a personal failing measured in dollars.

For a small number of people with complicated finances, that's a fair description and the gap is worth closing. For most people it's a treadmill. The last few percent costs a wildly disproportionate share of the effort, and the effort is paid in the most expensive currency there is, which is thinking about it.

Claim the things that apply to you. Keep records that would survive a question. What comes back is a meaningful block of hours each year and a low, persistent hum of worry — which, valued honestly, is usually worth considerably more than the deductions given up.