Should You Sell Your RSUs? A Tax-Aware Decision Framework

The default answer is sell

An RSU vest is a wage event, not an investment decision. Under IRC §83(a), the fair-market value of the shares on the vesting date is ordinary W-2 income in the first year the property is transferable or no longer subject to a substantial risk of forfeiture — the operative test for RSU vesting; the general year-of-inclusion framework in IRC §451 supplies the residual taxable-year rule when no more specific provision applies. After tax is withheld and the vest hits the brokerage, your basis equals vest-day FMV under IRS Publication 525. What happens next is a portfolio decision, not a tax decision masquerading as one.

Selling at or near vest collapses three frictions at once. It removes the single-stock idiosyncratic risk that the position represents — a meaningful issue at FAANG-scale grants where vest tranches routinely run six figures. It clears the withholding gap with cash that exists, instead of cash that has to come from somewhere later. And it forces the rebalance the household would otherwise procrastinate on. Selling is the default because the default reasons against selling tend to be behavioral — anchoring, attribution to the company, the feeling that selling is somehow disloyal — not analytical.

The “would I buy this fresh today?” test

When the vest hits the brokerage, do not ask “should I sell?” Ask: if the cash equivalent of these shares were sitting in my checking account, would I use it to buy this much of my employer’s stock at today’s price? Holding the vested shares is the same dollar exposure as taking the after-tax proceeds and buying the position fresh.

For an engineer at a public tech company who already holds ten or twenty percent of their net worth in the same ticker through prior vests, the honest answer is almost never yes. A rational outside investor with the same balance sheet would not voluntarily buy more concentration; the only reason most households end up in that position is that the equity arrived unbidden through compensation. The fresh-buy frame removes the endowment effect — the cognitive bias that makes already- owned assets feel safer than identical assets you would have to choose to buy — and surfaces the actual risk-return call.

The frame also bounds the conviction argument. If you would buy more at today’s price with outside cash — meaning you have an information edge, a long-horizon thesis, and a household balance sheet that can absorb the concentration — holding may be the right call. That is rare. It is rare enough that the default of selling is the right starting point for most households, and the burden of proof should fall on the hold case rather than the sell case.

Tax cost of selling — the math

Same-day sale. Vest-day FMV is $200,000. Basis equals FMV per Pub 525, so the sale produces approximately zero capital gain — minor wiggle from the price drift between the vest print and the brokerage execution. The only tax bill is the ordinary income on the vest itself, which exists whether you sell or hold. From a tax standpoint, the same-day sale is free.

Sale within one year. The same vest, sold six months later with the stock up 20%, produces $40,000 of short-term capital gain on the $40,000 of appreciation. Short-term gain is taxed at ordinary rates under IRC §1(h)’s rate hierarchy — 32–37% federal at top brackets, plus state, plus 3.8% Net Investment Income Tax under IRC §1411 for households over the statutory MAGI thresholds ($250K MFJ / $200K single, set in 2013 and not indexed for inflation). Net federal-plus-NIIT bill on the appreciation: roughly $14,000 to $16,300, leaving the household with $23,700 to $26,000 of after-tax appreciation gain on top of the original vest.

Sale after one year and one day. The same $40,000 of appreciation, now long-term under §1(h), is taxed at 20% plus 3.8% NIIT at top brackets — roughly $9,500 federal-plus- NIIT. The rate spread between short-term and long-term is approximately 13–17 percentage points federal, which on a $40,000 slice is the difference between keeping $26,000 and keeping $30,500. That is the optionality the “hold for LTCG” case is buying — and the math works only if the appreciation slice is real, the stock holds, and the household can stomach the concentration risk for the holding period. The vest-day ordinary tax does not change in either scenario; it has already been paid.

When holding is rational (rare)

Four cases. Each is narrow and each requires specific conditions the household has to actually satisfy.

Case 1: known lower-bracket year. Sabbatical, pre-retirement gap, voluntary unpaid leave, planned spousal career break. If next year’s ordinary income will run in the 12–22% federal bracket instead of 32–37%, holding RSUs across the year boundary and selling in the lower-bracket year converts short-term-rate exposure to a much smaller bracket. This is the bracket-arbitrage strategy in §5 below.

Case 2: QSBS-eligible founder stock under IRC §1202. Founder or early-employee shares in a qualified small business C-corp can qualify for up to 100% federal capital-gains exclusion on a sale after a five-year hold, capped at the greater of $10M or 10x basis. RSUs at a public mature company are not QSBS-eligible; the case applies only to early-stage founder grants where the stock was issued at a low FMV and the company satisfies the §1202 gross-asset test at issuance. See the QSBS section of the career-arc guide for the full mechanics.

Case 3: charitable intent. Donating long-held appreciated shares directly to a donor-advised fund or qualified public charity deducts at fair-market value and recognizes no capital gain. Holding RSUs more than one year past vest converts the position into a charitable-giving vehicle that delivers more after-tax value to the charity than selling and donating cash. The conviction here is charitable, not investment.

Case 4: small relative position. Vested RSUs worth two percent of a household’s liquid net worth at a household with a clean balance sheet, zero other employer exposure, and a long horizon — the post-tax dollars would buy the same stock anyway under the fresh-buy test. The hold and the sell-then-rebuy are within transaction-cost noise of each other; the household can hold for the long-term-rate optionality with minimal risk-budget impact.

The bracket-arbitrage strategy

Most engineers stay in the 32–37% federal bracket year over year, and within that band the LTCG decision turns on the 20%-vs-37% delta on the appreciation slice. Planned career events change the band materially, and the right move coordinates RSU sales with the bracket profile.

Worked example. A senior engineer at a public tech company vests $300,000 of RSUs in December 2026, plans a six-month unpaid sabbatical in 2027, and returns to W-2 income in 2028. The 2027 calendar year will likely run a partial-year salary totaling under $150,000 of ordinary income, putting the household in the 22–24% federal bracket for that year with no NIIT exposure on capital gains under §1411 thresholds for MFJ. Holding the December 2026 vest into early 2028 — past the one-year mark for LTCG — and selling either in the 2027 sabbatical year (short-term but at a lower ordinary bracket) or in early 2028 after the LTCG clock triggers can save several thousand dollars on the appreciation slice. The trade is the concentration risk for the 12–15 month holding window. For a household whose other capital is already diversified, that risk may be acceptable; for a household whose net worth is already 30%+ in the same ticker, the rate savings rarely justify the variance.

The bracket-arbitrage move only makes sense when the bracket delta is known and material — five percentage points federal or more — and when the appreciation slice is large enough to make the rate spread meaningful relative to single-stock variance. Mechanical “always hold one year for LTCG” advice ignores both inputs and overweights the rate optimization at the expense of risk-budget hygiene.

The same-day sale advantage

The same-day sale at vest is the cleanest possible execution. Vest hits, sell-to-cover withholds at the 22% IRS supplemental rate (statutory authority IRC §3402 and Treas. Reg. §31.3402(g)-1; supplemental wage tables in IRS Publication 15), the residual shares hit the brokerage, you sell the residual at market on the same trading day, and the cash settles in T+1. Tax-wise the sale produces near-zero capital gain because basis equals sale price — the entire transaction is a wage event, not an investment event, and the §1411 NIIT does not apply because there is no investment-income gain.

Two structural benefits follow. First, the same-day sale produces clean dollars to fund the supplemental-withholding gap when the household’s actual marginal bracket exceeds 22%. Most engineers in the 32–37% bracket plus 0.9% Additional Medicare under IRC §3101(b)(2) face a 10–15 percentage-point gap that has to be paid by April. The same-day sale produces the cash to pay it without market-timing risk on the residual position. The full withholding-gap mechanics are covered in the RSU withholding-gap explainer, and the deeper same-day-sale mechanics live in the RSU tax guide.

Second, the same-day sale leaves a clean audit trail. Brokers routinely report RSU sales with $0 cost basis on 1099-B, forcing a Form 8949 adjustment to avoid double-taxing income that has already been ordinary on the W-2. Near-zero gain on a same-day sale simplifies the filing position even if the 1099-B basis is wrong.

When the position is concentrated

RSUs that have piled up across multiple vests at a single employer become a portfolio concentration problem somewhere around 10–15% of liquid net worth. Past that line, the question is no longer “sell this vest or hold this vest” — it is “how do I unwind a concentrated position efficiently.” Embedded gain matters, the LTCG rate matters more, and structural diversification levers (exchange funds under §721 or §351, charitable remainder trusts, direct indexing tax-loss harvesting) become viable.

The full set of strategies for a concentrated position lives in the concentrated-stock guide. Two notes that bear on the sell-vs-hold call. First, the decision is no longer per-vest — it is per-tranche, modeled against the household’s entire embedded-gain stack with basis lots from prior vests at varying purchase prices. Second, the optimal answer often involves selling the highest- basis lots first (lowest embedded gain) while routing low- basis lots through a deferral structure. That sequencing is not visible from a single vest-day’s decision; it requires a position-wide model. The InverseWealth diagnostic surfaces the bracket and withholding-gap math; the strategy layer above it requires real planning work.

Bunching vesting events

Vesting schedules are not negotiable, but the deductible items paired with them are. Bunching — concentrating discretionary charitable gifts, mortgage-interest prepayment, or deferred- comp payouts in alternate years — can shift the marginal bracket on the slice of RSU income that lands in a given calendar year. The pattern works best for households with a large RSU vest cliff (initial-grant 25% one-year cliff, or a refresh that lands all in one quarter) and lumpy discretionary giving they would otherwise smooth.

The mechanics are simple. In the cliff year, donate two or three years’ worth of charitable giving in a single contribution to a donor-advised fund — preferably as appreciated shares to skip the capital-gain recognition. The bunch year itemizes against the larger deduction stack; the off years take the standard deduction. The differential, on a household giving $30K/year, can be $20–30K of taxable income shifted into the lower-bracket year. This is the kind of move that is invisible on any individual vest-day decision and only shows up as savings on a multi-year plan.

Bunching does not change the sell-vs-hold answer on any single vest. It changes the bracket the income lands in, which compounds with the same-day-sale or hold-for-LTCG decision when modeled across years.