401(k) Growth Calculator

The arithmetic

The projection is a single recurrence, applied once per year. Let $B_k$ be the balance at the end of year $k$, $S_k$ the salary earned during that year, $c$ your contribution rate, $\kappa$ the employer's match cap, $m$ the match rate, $r$ the expected gross return and $f$ the annual expense ratio:

$$B_k = B_{k-1}\,(1 + r - f) \;+\; S_k\,c \;+\; S_k\,\min(c,\,\kappa)\,m$$

with salary growing at $g$ per year, so $S_k = S_1\,(1+g)^{k-1}$, and $B_0$ your balance today.

Read the order of operations, because it is the part calculators disagree on. Growth is credited on the opening balance, then the year's contributions land. A dollar you defer in year $k$ earns nothing in year $k$. This is the end-of-year convention, and it is deliberately the conservative one: a plan that deposits money every payroll period would have each dollar invested for an average of roughly half a year longer, which over a long horizon produces a modestly higher number. Beginning-of-year timing sits at the other extreme and overstates it. Reality is in between and closer to the end-of-year figure than most people assume, because early-career contributions are small relative to the balance they eventually compound into.

The four bands in the chart are cumulative and reconcile to the balance exactly. Three of them are simply running totals: your starting balance, the sum of your own deferrals, and the sum of employer match dollars. The fourth, investment growth, is computed as the residual:

$$G_k = B_k - B_0 - \sum_{j\le k} S_j c - \sum_{j \le k} S_j \min(c,\kappa)\,m$$

Computing growth as a residual rather than accumulating it separately is a small implementation choice with a real consequence: the stack always sums to the balance shown at the top, with no rounding drift between the chart and the headline number.

The employer match, modelled the way plan documents write it

Most calculators ask for "employer match" as a single percentage. That is not how plans work, and the simplification quietly changes the answer.

A real match formula has two independent parts. The match rate is how many cents the employer adds per dollar you defer. The cap is the level of your own contribution, expressed as a percentage of pay, above which no further dollars are matched. The classic US formula is 50% up to 6% of pay, which is what this module loads by default. It means:

$$\text{match} = \min(c,\,\kappa) \times m \times S$$

At a 6% contribution: $\min(6\%, 6\%) \times 50\% = 3\%$ of salary. At a 3% contribution: $\min(3\%, 6\%) \times 50\% = 1.5\%$ of salary. At a 15% contribution: still $\min(15\%, 6\%) \times 50\% = 3\%$ of salary, because the cap binds. The match schedule is piecewise linear with a kink at $\kappa$, and it is flat above it.

A calculator that multiplies your whole contribution by the match rate overstates the match for anyone contributing above the cap. One that treats the match as a flat percentage of salary understates it for anyone below. The panel in the module shows the formula evaluated on your own inputs so you can check it against your plan's summary plan description rather than trusting the arithmetic blind.

Two things the match model deliberately leaves out. Vesting determines how much of the employer's money you keep if you leave before a schedule completes, and it can zero out several years of match dollars on a job change. That is a separate mechanism, covered in what does vested mean in a 401(k). True-up provisions, non-elective safe-harbour contributions, and profit-sharing allocations vary too much by plan to model generically.

What the expense ratio actually costs

Fees enter as a straight reduction of the annual return: the projection compounds at $r - f$ instead of $r$. This is an approximation of a fee that in practice accrues daily against assets. A more exact treatment multiplies rather than subtracts, giving $(1+r)(1-f) - 1 = r - f - rf$. The difference is the cross term $rf$, which at a 7% return and a 0.50% expense ratio is 3.5 basis points a year. Compounded over thirty years that is about 1% of the final balance — the subtractive form runs slightly high — which is well inside the noise of every other assumption in the model. The simpler form is used because it makes the mechanism legible: a fee is a permanent haircut on your compounding rate.

The dashed line on the chart runs the identical contribution schedule at $r$ with no fee. The gap is not the fees you paid. It is the fees you paid plus everything those dollars would have earned for the rest of the horizon. That second part is usually several times larger than the first, which is why a difference of a few tenths of a percent in expense ratio shows up as a five-figure or six-figure gap at the end of a long career.

Nominal versus today's dollars

Every figure in the module is nominal except where labelled otherwise. The real value is the standard deflation:

$$B_k^{\,\text{real}} = \frac{B_k}{(1+i)^k}$$

where $i$ is the inflation rate you enter. This is the honest way to read a large number thirty years out. A $2 million balance in 2056 at 2.5% inflation buys roughly what $950,000 buys today. Note that the inflation input does not feed back into salary growth or returns: those are separate inputs you set yourself, and if you want a real-terms projection throughout, enter a real return and a real salary growth rate and set inflation to zero.

What this model does not do

Being explicit about this matters more than the projection itself.

A constant return is not reality. Markets do not deliver 7% every year. They deliver something like +22%, −9%, +14%, −37%, +26%. Compounding a fixed rate produces the right average outcome and the wrong distribution of outcomes. The single number this module returns is closer to a central tendency than to a forecast, and the honest interval around it is wide.

Sequence of returns risk is invisible here. Two portfolios with identical average returns and identical contributions can end at very different balances depending on when the bad years arrive, because the amount of money exposed to each year's return changes over time. The effect is modest during accumulation and severe during withdrawal. This module covers accumulation only.

Taxes on withdrawal are not modelled. A traditional 401(k) balance is pre-tax money. The number shown is not what you get to spend. Roth balances, which are after-tax going in, are not distinguished from traditional balances here, and the two are not comparable dollar for dollar.

Contribution limits are not enforced. The statutory elective deferral limit is exposed as an editable input and used only to raise an informational flag on year one. It is not applied as a cap, because the limit is re-indexed annually and hard-coding a stale figure would be worse than not enforcing one. Age-50 catch-up contributions, the additional catch-up window in the early sixties introduced by SECURE 2.0, and the separate overall limit on combined employee-plus-employer additions are all outside the model. Confirm the figures for your plan year with the IRS.

The career is modelled as a smooth line. No job changes, no gaps, no promotions, no bonuses, no periods where contributions stop. Plan loans, hardship withdrawals, required minimum distributions, and rollovers are absent.

Frequently asked

Why is the number different from my 401(k) provider's calculator?

Almost always because of three things: contribution timing (many providers assume per-payroll contributions, which is more favourable), whether fees are netted from the return at all, and how the employer match is capped. Enter the same inputs here with a 0% expense ratio and you will usually land within a few percent of a provider's figure. If the gap is large, the match formula is the first place to look.

How much of a final 401(k) balance is usually growth rather than contributions?

It depends entirely on the horizon, which is exactly why the decomposition is the interesting output. Over a short window contributions dominate, because there has not been time for compounding to work on them. Over thirty years and more the growth band typically becomes the largest of the four. Run the module at a 10-year horizon and a 35-year horizon with everything else fixed and watch the composition bar re-weight. That comparison is the single most useful thing this tool does.

What return should I put in?

That is a modelling choice, not something a calculator can decide for you, and nothing here is a recommendation. What is useful to know is what the input means: it is a nominal, arithmetic-average annual return net of nothing except the fee you enter separately. Long-run US large-cap equity returns have been roughly 10% nominal before fees over the full postwar sample, bond returns considerably lower, and a blended target-date portfolio sits somewhere between the two and drifts down as it de-risks. Whatever figure you choose, the productive exercise is to run the projection at two or three different rates and look at how wide the spread is, rather than to treat any single run as the answer.

Does contributing above the match cap still make sense?

That is a personal financial question and this is an educational tool, so it is not one we answer. What the module does show is the pure arithmetic: above the cap, each additional percentage point of pay you defer adds contribution dollars and adds zero match dollars. The match panel displays exactly where that kink sits for the formula you entered.

Sources and further reading

• Internal Revenue Service, 401(k) and profit-sharing plan contribution limits — the authoritative figure for elective deferral and catch-up limits, re-indexed every plan year.
• U.S. Department of Labor, EBSA, A Look at 401(k) Plan Fees — the plain-language reference on what plan fees are and where they are disclosed.
• Vanguard, How America Saves — annual survey of participant deferral rates, plan match formulas, and account balances by age.
• W. P. Bengen (1994), Determining Withdrawal Rates Using Historical Data, Journal of Financial Planning — the origin of the sequence-of-returns literature that explains why a single average return is an incomplete description of a multi-decade path.

Educational research only. Not investment advice, and not a projection of your individual outcome.

Methodological sources

• Internal Revenue Service (2026). Retirement Topics — 401(k) and Profit-Sharing Plan Contribution Limits.
• U.S. Department of Labor, Employee Benefits Security Administration (2019). A Look at 401(k) Plan Fees.
• Vanguard Institutional Investor Group (2025). How America Saves.
• William P. Bengen (1994). Determining Withdrawal Rates Using Historical Data.

Educational purposes only. Not investment advice.