The RSU Diversification Strategy I Actually Use
Career & Compensation · Long Read
The RSU Diversification Strategy I Actually Use
Sell 100 percent on vest day and put the proceeds into a diversified dividend growth portfolio. Here's why that beats every other RSU diversification strategy I've seen.
Every four years I watch a friend in tech tell me they've decided to hold their RSUs this time because the stock looks strong. About half of them come back later regretting it. That's not a story about picking bad companies. It's what a concentrated position in a single stock does to a portfolio, and it's the exact outcome any real RSU diversification strategy is designed to prevent.
Your employer already pays your salary, your bonus, your health insurance, and if you leave, your future paychecks stop. Adding a large equity position in the same company means the same event that ends your income also craters your net worth. That's not diversification. That's amplification.
Here's the honest version of what to do about it.
If someone handed you $80,000 in cash and asked "want to buy your employer's stock with all of this?" you'd probably say no. That's what holding vested RSUs is.
The Default
Why "Doing Nothing" Isn't a Neutral Choice
Most people at big tech companies don't actively decide to hold their vested RSUs. They just don't act. The shares appear in the brokerage account after each vest date, the employer sells some to cover the automatic 22 percent federal supplemental tax withholding required under IRS Publication 15, and the rest sits there. Nobody sends you a form asking what you'd like to do. Inaction is the default, and the default is to build a concentrated position by accident.
At most large public companies the schedule is four-year graded vesting with quarterly releases. That means a chunk lands in your account every three months. If your grants are $100,000 to $300,000 total value, and refresh grants stack on top, you can easily be sitting on hundreds of thousands of dollars of a single stock inside a couple of years. That's a position size a diversified investor would never build on purpose.
The mental model in the way: people treat vested RSUs as "already owned" and emotionally sticky, when the honest view is that they're fresh cash the company just paid you. The tax was already paid at vest. The vest-day price is your cost basis. You're buying that stock with post-tax cash every quarter, just automatically.
The Core Argument
The RSU Diversification Strategy in One Rule: Sell 100 Percent at Vest
The one-rule version: whatever vests today, sell today, redeploy tomorrow. Every deviation from that rule should require a specific, defensible reason. Not a feeling that the stock will keep going up.
The tax logic is what makes this cleaner than most people think. RSUs are taxed as ordinary income at vest, at the fair market value on that day, per IRS treatment. That price becomes your cost basis. If you sell the same day at roughly the same price, your capital gain is zero. No additional tax bill for selling. You are not "leaving money on the table" by selling early, because the gain hasn't happened yet.
What you are doing when you hold is doubling down on a single bet. Doubled down, because you're already exposed to the company through your salary. A bet, because the extra return you might earn isn't compensation for information you have. It's the same public price everyone else sees. The only real edge an insider might have, they're legally not allowed to trade on.
Every wealth manager and RSU explainer article eventually gets to "diversify your concentrated position." Almost none of them tell you the specific rule to follow. Sell 100 percent, immediately, redeploy into a diversified portfolio. That's the whole strategy. Everything else is either the specific mechanics of the redeployment, or an edge case where the rule bends.
The Redeployment
Where the Money Actually Goes: Dividend Growth
Selling is only half the plan. What you buy with the proceeds is the other half, and this is where the standard financial-advice article stops being useful. "Put it in a diversified portfolio" is the correct answer and also a useless one.
My redeployment target is a dividend growth portfolio. Not for the income, which is negligible in your accumulation years, but for what dividend growth companies represent structurally: mature businesses with real earnings, disciplined capital allocation, and a decades-long habit of paying shareholders back. That's a completely different risk profile from the single tech stock you just sold, which is the entire point.
Concretely, that could mean a low-cost dividend growth ETF like Vanguard's Dividend Appreciation ETF (VIG), which tracks companies with at least 10 years of consecutive dividend increases. Or a basket of individual Dividend Aristocrats and Kings, if you want to build the position yourself. The specific vehicle matters less than the property you're buying: uncorrelated exposure to a set of businesses that don't share risk factors with your employer.
The compounding math wins over decades. A concentrated position either doubles or crashes, and both outcomes put too much of your net worth in a single name. A dividend growth portfolio compounds less dramatically, but the base grows every year, and so does the income it throws off. That's the engine a real retirement account needs, and the opposite of what a single-stock RSU position offers.
The Steelman
What About NVIDIA and Amazon?
The one-rule version has a real problem worth naming. If you worked at NVIDIA over the last five years, or Amazon from 2010 to 2020, selling every vest on vest day would have cost you a fortune. That's true, and it's the specific case that makes this decision harder than "sell everything" makes it sound.
Here's what the highlight reel hides. For every NVIDIA there's a Cisco in 2000, a General Electric in 2000, an Intel in 2020, a Meta in mid-2022, a Peloton in 2021. Each was a "can't miss" story at the time, and employees who held got wrecked. We only remember NVIDIA and Amazon because the outcome resolved in their favor. That's survivor bias, and it's the specific bias concentrated employer-stock holders are most vulnerable to.
The right question isn't "would holding have paid off if I picked correctly?" It's "across the universe of employers I might have joined, what's the expected outcome of holding versus selling?" Sell-and-diversify wins on expected value, even when it loses in the highlight cases.
The version of the rule that survives contact with NVIDIA is a cap, not a hard sell. Cap employer stock at a fixed percentage of your investable net worth, and sell everything above that line at each vest. Ten percent is defensible for most people. Twenty percent is aggressive. Above twenty-five, you're hoping, not investing.
You get the compromise position: real upside if you happen to be at the next NVIDIA, capped downside if it turns out you're at the next Peloton.
The Exceptions
When Holding Some Might Actually Make Sense
Two situations where the sell-everything rule can bend, and one that sounds like a real reason but isn't.
Real exception 1: A blackout window forces the timing
Most public companies restrict when employees can trade around earnings and material events. If your vest lands inside a blackout, you literally can't sell that day. Set up a 10b5-1 pre-scheduled sell plan through your brokerage so the sale executes automatically at the next allowed window. That's still following the rule, just mechanically.
Real exception 2: A specific tax event you can time
If a very large single vest would push you into a higher marginal bracket in one year, spreading the sale across the calendar-year boundary can meaningfully reduce your tax bill. This is real, but it's about the sale timing, not about holding for growth. You're still selling within weeks, not years.
Not actually a reason: waiting for long-term capital gains treatment
People assume holding vested RSUs for 12 months to get long-term capital gains treatment is smart. It usually isn't. Your cost basis is the vest-day price. If the stock is flat or down 12 months later, LTCG treatment saved you nothing. If the stock rallies 30 percent, you've now taken on 30 percent additional concentration risk to save 22 percent versus 32 percent tax on the gain. The expected value math almost never works out in favor of holding.
The best-diversified portfolios I've seen belong to engineers who make one boring decision every quarter: sell what vests, buy what compounds. The strategy isn't sophisticated. Sticking to it is.
Comments
Post a Comment