A guy I know took a consulting gig last November. Nice little project, wrapped before Christmas, paid about eleven thousand dollars. He was pleased with himself.
Then his tax return came back and the eleven thousand had cost him closer to nineteen. Not because of his marginal rate. Because that eleven grand pushed his modified adjusted gross income across three separate lines he did not know existed. He lost part of a credit. He picked up a surtax. And two years later his Medicare premium went up, because the government was still looking at that year.
He did not make a bad decision. He made a blind one. There is a difference, and the difference is a map.
The Tax Code Is a Staircase, Not a Ramp
Everybody understands marginal rates. You earn another dollar, you pay a slightly higher rate on that dollar, life goes on. That part is a ramp. Smooth. Forgiving.
The part almost nobody maps is the staircase sitting on top of the ramp. Scattered through the code are dozens of specific dollar figures where something switches on, switches off, or starts bleeding away. Cross one by a single dollar and you can lose a benefit worth thousands. Not a slightly worse rate. A cliff.
There are two flavors and you need to be able to tell them apart, because they call for completely different responses.
Phase-outs bleed. You cross the first line and a benefit starts shrinking gradually until it is gone at the second line. Annoying, but survivable, and you can plan around the slope. Most credits and deduction limits work this way.
Cliffs guillotine. One dollar over and the whole thing is gone. No slope, no partial credit. The Medicare premium surcharge is the loudest example. Some health insurance subsidy structures behave this way. So do a handful of eligibility tests that are pure yes or no.
Phase-outs you manage. Cliffs you never, ever go near without knowing exactly where you stand. And right now, in the first week of September, you have roughly sixteen weeks of runway to do anything about either one. After December thirty-first the year is a historical document.
Why September Is the Real Deadline
April is when you find out. September is when you can still do something.
By now you have eight months of actual data. You know roughly what your salary looks like, whether the bonus is coming, how the side income is tracking, what the portfolio has thrown off in dividends and realized gains. The year is knowable. It is not yet finished.
That gap between knowable and finished is the entire opportunity. In April you are a historian. In September you are still an architect.
And there is one more wrinkle that makes this year specifically worth your attention. Several of the biggest thresholds do not grade you this year. They grade you later. The Medicare premium surcharge looks back two years, which means the income you are generating right now is being scored for a bill you will not see until 2028. College financial aid formulas use a prior-prior year, which means parents of high school sophomores are already inside the window that determines their aid package. You are being graded on a test you have not realized you are taking.
Step One: Build Your MAGI Baseline
Almost every threshold in the code keys off some version of modified adjusted gross income. Not gross pay. Not take-home. MAGI, and irritatingly, the exact definition shifts slightly depending on which provision is asking.
Start with the workhorse version and refine later. Open a blank sheet and build this stack:
Wages from your latest pay stub, using the year-to-date figure, plus your best estimate of what is left to come including any bonus. Add net self-employment income after expenses. Add interest and dividends, and pull the actual year-to-date numbers from your brokerage rather than guessing, because reinvested dividends still count as income even though you never saw the cash. Add realized capital gains, and note that mutual funds throw off capital gain distributions in December whether you sold anything or not. Add rental income, retirement account distributions, and anything else that lands on the front page of a return.
Then subtract the above-the-line items: deductible retirement contributions, health savings account contributions, the deductible half of self-employment tax, and a handful of others depending on your situation.
What you have now is a working estimate of where you will land. Write it down. This single number is the thing every threshold is measured against, and most people have never once calculated it before the year was over.
Step Two: Pull Only the Lines That Apply to You
There are a lot of thresholds. Almost none of them are yours. The mistake is trying to memorize the whole code instead of building a short personal list.
Work through these filters and keep only what hits.
If you are within a few years of sixty-five, or already there. The Medicare premium surcharge is your primary cliff, and it is a true cliff with multiple tiers. Each tier bumps both your Part B and Part D premiums for a full year, and it grades your income from two years prior. Worth noting: if you have had a genuine life-changing event such as retirement, a spouse's death, or divorce, there is a form that lets you ask the agency to use current income instead of the two-year-old figure. Most people never file it.
If you buy your own health insurance. Marketplace subsidy eligibility is income-tested, and the structure of that test has changed more than once in recent years. This is one where the current-year rules genuinely matter, so verify before you plan around it.
If your household income is roughly in the low-to-mid six figures. The net investment income surtax applies above a fixed threshold that is not indexed for inflation, which means it quietly captures more people every single year. There is a parallel additional Medicare tax on wages above a similar line.
If you are in a low-income year. This is the one people miss because it feels like bad news. There is a band at the bottom of the long-term capital gains structure taxed at zero. A sabbatical year, a startup year, an early retirement year before Social Security switches on, all of these can open a window where you can realize gains and pay nothing on them. That window closes as income rises.
If you contribute to a Roth IRA. Direct contribution eligibility phases out. Above it you are into backdoor territory, which is fine but has its own trap involving pre-tax IRA balances.
If you own a pass-through business. The qualified business income deduction phases out differently depending on whether your business is classified as a specified service trade. If you are a consultant, doctor, lawyer, or financial professional, this line is probably yours and it is steep.
If you have kids approaching college. Aid formulas use a base year that starts earlier than parents expect. Ask which tax year your first application will use, then count backward.
If you are drawing Social Security. The share of your benefit that becomes taxable steps up at two income levels, and those levels have never been indexed. They were set decades ago and have not moved.
Keep the ones that apply. Put a real dollar figure next to each. Verify the current-year number rather than trusting the one you remember, because several of these move annually and a few of them do not.
Step Three: Measure the Gap
Now the arithmetic that makes the whole exercise worth doing. For every threshold on your short list, subtract your projected MAGI from the threshold.
Positive number means headroom. That is space you can deliberately fill with income, and filling headroom on purpose is one of the most underrated moves in personal finance. A Roth conversion into unused headroom converts money at a known rate instead of an unknown future one.
Negative number means you are over. Now the question is how far. Three hundred dollars over a cliff is an emergency worth solving this month. Forty thousand over is a fact you accept and plan around.
Then flag anything within roughly ten thousand dollars of a line in either direction. Those are your live positions. A December mutual fund distribution you did not forecast can move you five thousand dollars without a single decision on your part. If you are sitting three thousand from a cliff in September, you are not safe, you are exposed.
Step Four: Pick Your Levers
You have four ways to move the number, and they work in roughly this order of ease.
Shrink income. Max the pre-tax retirement contributions you have not maxed. If you are self-employed and have not opened a solo plan, there is still time this year and the contribution room is significant. Health savings account contributions come off the top if you are on a qualifying plan. If you control your own billing, an invoice sent in January is January income.
Time income. Ask whether a bonus can be deferred, whether a client payment can land after the new year, whether an equity sale can wait a few weeks. Note the flip side: if you are in a low year with headroom, you want to pull income forward, not push it away.
Offset with losses. Realized losses net against realized gains, and a limited amount of excess loss can offset ordinary income with the remainder carrying forward. Just be careful with the wash sale rule, which reaches across accounts in ways people consistently underestimate.
Deploy deductions. We went deep on charitable timing recently so I will not repeat that ground here, but the principle holds: deductions that reduce adjusted gross income are more powerful than deductions that only reduce taxable income, because the thresholds key off the former.
One caution before you go optimizing. Do not let the tail wag the dog. Turning down eleven thousand dollars of real income to dodge a two thousand dollar surcharge is a bad trade dressed up as a clever one. The map exists so you can see the cost, not so you can avoid every cost.
Making It Run Without You
The failure mode here is not that people cannot do the math. It is that they do it once, in a burst of enthusiasm, and never again.
Turn it into a monthly ping. I run a scenario in Make.com that fires on the first of every month, pulls the current figures from the sheet, and drops a short summary into my inbox showing where I sit against each line. It took about forty minutes to build. It has caught two problems that would have cost real money.
For the reconciliation work itself, the tedious part is reading statements and pulling numbers out of documents that were designed to resist that exact activity. I run those through Galaxy.ai and have it extract the line items into a clean table. Cuts an afternoon down to twenty minutes.
And if you are going to have a conversation with your accountant about any of this, record it. I use Fathom for exactly this reason, because six months later when you are trying to remember whether they said the conversion made sense, the transcript is worth more than your memory.
Run It This Week
Block ninety minutes. Build the MAGI baseline from actual year-to-date numbers, not estimates you feel good about. Pull your personal short list of thresholds. Measure every gap. Flag anything inside ten thousand dollars. Pick your levers and calendar them before Thanksgiving, because December is when the mutual fund distributions land and the good options start closing.
Sixteen weeks. That is the whole window. Most people will spend it not knowing the lines exist, and then find out in April what they cost.
Get the Worksheet
I built The Threshold Map Worksheet to run this exact process. It includes the MAGI baseline builder with every input line labeled and sourced, the threshold filter that narrows the full list down to only the ones matching your situation, the gap calculator, a lever selection matrix showing which moves work for which thresholds, and a December checklist for the distributions and deadlines that ambush people in the last three weeks of the year.
Reply to this email with the word THRESHOLD and I will send it over.
One More Thing
The Grid Inner Circle is open. It is the paid community where we work through this stuff together, share the builds, and pressure test decisions before they get expensive rather than after.
Membership runs 29 dollars a month. There is also a lifetime option at 499 dollars, which is roughly seventeen months of the monthly price and then never again. If you have been reading The Wealth Grid for a while and actually building the systems, the lifetime seat is the obvious math.
Reply with INNER CIRCLE and I will send you the details.
Build the map. The lines are already there whether you can see them or not.
Alex Rivera
Wealth Architect at The Wealth Grid
This is educational content, not tax or investment advice. Threshold figures change annually and several of the provisions mentioned have shifted in recent years. Verify current-year numbers and talk to a qualified professional before acting on any of it.
