A worked model with stated assumptions: why a 20% income cut is a 70% savings cut, why spending is the shock absorber, and the input the calculators leave out

If your contract renews next year at 10% less and your mortgage was sized to the old number, your retirement date moves out nine years. Nobody can forecast the cut. Its arithmetic is knowable, and every retirement calculator hides it by treating your income as a constant: type in the number, set a savings rate, pick a return, and the tool tells you a date.
This is an illustration with stated assumptions, not a projection for you. Your own numbers go in the savings rate calculator.
Take an attending who starts saving at 35 with nothing invested. Gross income $400,000, near the 2026 average for U.S. physicians and a little under the specialist average. Assume 30% goes to taxes, leaving $280,000. She saves 20% of gross, $80,000 a year, which is the classical guideline for physicians, and spends the other $200,000. Her target is twenty-five times annual spending, the amount that supports a 4% starting withdrawal, so $5 million in today's dollars. Assume a 4% real return on a diversified portfolio, which is between the long-run historical figure for a stock-heavy mix and the lower forward-looking estimates that major fund companies are publishing in 2026.
On those numbers she reaches the target in 32 years, at 67. At a 3% real return it takes 36 years; at 5%, 30. The date is sensitive to return, but not as sensitive as it is to what comes next.
Now cut her gross income by 10%, 20%, or 30%, permanently, starting at 35, and assume she keeps spending $200,000. Spending is the sticky variable in real households: the mortgage, the school, the lifestyle that was built on the old number. Savings become the residual.
A 10% cut in income is a 35% cut in savings, and it moves retirement out nine years. A 20% cut in income is a 70% cut in savings, and it moves retirement past any plausible working life. A 30% cut ends saving altogether. The relationship is not proportional, because savings are what is left after the sticky part of the budget is paid. A high earner with a high fixed cost base has a savings rate that is far more fragile than the income that feeds it.
The same arithmetic is why a physician earning $400,000 can feel one bad contract away from trouble. The contract hits the residual, and the residual is $80,000.
Run it again, but this time she cuts spending in proportion to the income cut.
The date doesn't move at all. That looks like a trick and isn't. When spending falls, two things fall with it: the target, which is a multiple of spending, and the savings, which are a residual. Because she kept the same savings rate, the ratio between the two held, and the arithmetic doesn't care about the absolute level. Hold the savings rate through the cut and the retirement date holds; the table shows it at 10, 20, and 30%.
Spending is the variable that absorbs an income shock. Watch the savings rate, not the income.
Most income shocks don't arrive at 35. They arrive at 45 or 50, at a contract renewal, a practice sale, a policy change, or a slow reimbursement slide that shows up as more work for the same pay. So run it once more with the shock at 45, after ten years of saving $80,000 a year, which leaves her about $960,000 invested.
Ten years of compounding soften the blow, but the shape is the same. Holding spending through a 20% cut at 45 costs eleven years. Adjusting spending costs none and, because the target falls, brings the date slightly forward. The saved balance helps; it does not rescue a plan whose savings rate went to zero.
All of the above assumes she works until the target is reached. That assumption is the shakiest one in the model, and the available evidence does not pin it down. Among physicians who did leave clinical practice, a 2026 Permanente Journal study found a mean departure age of 48.1, against 57.1 in an earlier cohort; that is a figure for leavers, not a forecast for any individual career. Doximity's 2026 survey found 46% of physicians considering early retirement, which is an intention rather than an outcome. So treat early departure as a stress test, not a base case. If she leaves clinical work at 48 on the base-case plan, she has about $1.3 million invested, which supports roughly $53,000 a year at a 4% withdrawal, against $200,000 of spending. At 55 the figure is about $2.4 million and $95,000 a year.
Career length is the largest single input in physician retirement arithmetic and the one AI pushes in both directions at once. Administrative offload can add years. The productivity treadmill, and the burnout it produces, takes them away. Who keeps the hours an AI scribe saves is a retirement question.
Faced with a cut, a household has three levers. Spend less, which the model shows is the lever that holds the date, and which is limited by fixed obligations and by how much of the budget was built on the old number. Work longer, which the model shows can't close a large gap and which is the lever that AI-driven pace is also pulling against. Or earn from somewhere other than clinical labor, which is the only lever that adds to the numerator rather than shrinking the denominator. That third lever is what the second engine means.
There is a fourth response that isn't a lever: accept a lower target. A 4% withdrawal on $3.5 million is a real retirement, and a household that runs the numbers at 45 can choose it.
The conventional three-to-six-month emergency fund was designed for a general audience with diversified household income and a job market where the next position is a few weeks away. A physician's income runs through one credential, often one employer, and sometimes one market. Credentialing alone can take months. That argues for a larger floor than the generic rule, and the size of it depends on how exposed your specialty is and how concentrated your household's income is, which is what the Specialty Exposure Map is for. Disability insurance answers a different question: it transfers one specific risk, after an elimination period of 90 to 180 days, which is the period the cash is carrying you.
Measure your actual savings rate, not the one you intended. Most physicians I know can name their income to the dollar and their savings rate only to the nearest guess. Run the table above with your own numbers, and run it for the mid-career case, because that is the one that will happen to you. Look at where your specialty sits on the map. And decide, while nothing is wrong, which of the three levers you would pull first, because the one you'll reach for in a bad year is the one you've already thought about.
The model uses a flat 30% effective tax rate, which understates taxes in high-tax states and ignores the fact that a lower income is taxed at a lower marginal rate, so the after-tax cut is a little smaller than shown. It ignores student loans, employer retirement contributions, Social Security, inflation in spending, and the sequence of returns, which matters a great deal in the years around retirement. It uses the 4% rule as a target; recent research puts the safe starting withdrawal anywhere from 3.9% to 4.7% depending on the assumptions. None of this changes the shape of the result, but it does change the dates by more than a little: the return assumption alone moves the base case from 30 years to 36 across a 5% to 3% range, and the omitted items move it further. Treat every age in this piece as the middle of a wide band. This is an illustration, not advice, and your own plan belongs with a fee-only fiduciary planner who can see all of it.
The Income Variable takes this arithmetic and builds the plan around it: how to size a cash reserve off your own exposure rather than a generic range, which layer of a physician's plan is usually the one that's missing, how to keep the principles steady while re-measuring the numbers every year, and what to do in the year the shock actually arrives. The career-stage playbook sequences it for early, mid, and late career. Join the launch list for the book, the savings-rate calculator, and the other free companion tools.
This article is general information and analysis, not individualized medical, financial, investment, tax, or legal advice. See the Disclaimers page for the full statement.
Double board-certified reconstructive surgeon in Austin, Texas. Author of The Income Variable.
About the author
Join the launch list for book updates, new articles, and free companion tools.
By submitting, you agree to our Terms of Use and acknowledge our Privacy Policy.