One Employer, Four Exposures: Salary, Unvested Equity, Vested Stock, and Your Home

Consider a mid-career employee at one large company, living in a metro where that company and its industry are a visible share of the high-income job market. Before making a single investment decision, this household holds a claim on the firm’s future in four places: the pay that arrives every two weeks, the unvested grants that will only convert if the employee is still there and the price holds up, the vested shares that have piled up in a brokerage account, and the equity in a house whose value depends on how many people want to live and work nearby.

When a vest lands, the usual question is whether the stock looks attractive. That question skips a prior one. The shares are the fourth dose of an exposure the household already carries in three other places, and one of those three is financed with a mortgage.

So which of the four can you actually change, and what does changing it cost?

Four claims on one employer

The four exposures are not four independent bets. They share a common input, which is the firm’s economic health, and in the fourth case the health of the industry it belongs to. That is what makes the stack different from ordinary single-stock concentration, where a bad outcome costs you money and nothing else.

Benartzi, Thaler, Utkus and Sunstein made the canonical version of this argument about company stock in 401(k) plans, and their framing is exactly right here. Employer stock combines single-security risk with human-capital risk, because an employee’s economic prospects tend to be positively related to the employer’s performance. “In the worst-case scenario, workers can lose their jobs and much of their retirement wealth simultaneously.”⁴ The addition this guide makes is the fourth layer, which their paper does not treat: the house.

Home prices do not track your employer’s share price. The correlation between any one metro’s housing market and any one stock is weak enough to be useless. The mechanism running through this stack is indirect: a shock to a large local employer can move local labor demand, and local labor demand moves housing. The stock price and the house price can respond to the same underlying event without responding to each other.

Human capital is the largest line and the least liquid

For most people well short of retirement, the present value of remaining pay is larger than everything else on the balance sheet combined. It cannot be sold, it cannot be hedged at any reasonable price, and the only way to reduce it is to go work somewhere else. It is the reason the other three layers deserve scrutiny: you are already carrying as much employer exposure as you can carry involuntarily.

Two caveats keep this line honest. It is gross of tax, and it has a matching consumption liability that no balance sheet in this guide shows. Read it as a measure of how much of your future depends on this employer rather than as a balance you could spend. Our guides on the household portfolio and human capital risk for tech workers work through the balance-sheet framing properly, so this one stays short.

Unvested equity is a claim on the price and on still being there

Unvested grants require two things to go right. The share price has to hold up, and you have to still be employed on the vest date. Those two conditions fail together far more often than they fail independently, which is why a scenario that halves the stock and cuts future compensation does more damage to unvested equity than either shock alone. Refresh grants add a third dependency, since the size of next year’s grant is set by a company deciding how much it wants to spend on retention.

The tax mechanics are worth getting right because they are widely garbled. Restricted stock units are not Section 83 property when they are granted, which is why no 83(b) election is available for them. Section 83 applies at settlement, when the stock is actually transferred, and the value transferred is includable in income at that point.¹³ Withholding is a separate question from tax owed. For 2026 the IRS continues to specify a flat 22% on supplemental wages under the optional flat-rate method, rising to 37% on supplemental wages above $1 million in a calendar year, withheld without regard to your Form W-4 and aggregated across businesses under common control.¹² For a high earner, 22% is usually well below the marginal rate, which is what our guide on RSU withholding exists to solve. See also the equity compensation guide and trading future equity for cash.

Vested shares are the layer with a market price

At vest, your basis is set to the amount that was included in your income, and your holding period starts just after the shares substantially vest.¹³ Sell that day and there is almost nothing left to tax as a gain or a loss. The tax event that mattered has already happened, and it happened whether or not you sell. This is what makes layer three different from the other three: it has a price, a sell button, and no exit cost.

Keeping the shares is an allocation decision, and it is worth being precise about which risk that decision takes on. The stock’s systematic exposures, its market beta and whatever size and profitability tilts it carries, are compensated, and you keep all of them by owning a diversified portfolio instead. What is added by concentrating is company-specific risk, which is diversifiable and therefore should not be expected to command a reliable premium. Compensated vs. uncompensated risk works through the distinction.

The distribution of individual-stock outcomes is what makes that distinction expensive to ignore. Bessembinder found that from 1926 to 2016, across roughly 25,300 US firms, only 42.6% of stocks beat one-month Treasury bills over their lifetimes, and that 1,092 firms, about 4% of the total, accounted for all of the roughly $35 trillion of net wealth creation.¹ The global replication found 55.2% of US and 57.4% of non-US stocks underperforming Treasury bills from 1990 to 2020, with the top 2.4% of firms accounting for all $75.7 trillion of net global wealth creation.² A 2026 working paper extending the US sample through 2025 puts 46 firms behind half of $91 trillion of net wealth creation, with nearly 60% of stocks producing wealth reductions relative to Treasury bills, and a cross-stock mean buy-and-hold return above 30,000% against a median of −6.9%.³ None of that says your employer is likely to lose money. It says the distribution is extreme enough that an index owns the rare giant winners by construction and a single position mostly does not. Most stocks lose to T-bills covers the data properly.

There is a well-known figure attached to employer stock specifically: that a dollar of it can be worth less than fifty cents once you adjust for the risk. It is usually attributed to Benartzi and colleagues, but it is Meulbroek’s. She estimated that employees sacrifice an average of 42% of their company stock’s market value by taking on risk that could have been diversified away, and produced the two illustrations that get quoted: 58 cents on the dollar for a worker holding 25% of assets in company stock over ten years, and 33 cents for one holding 50% over fifteen.⁵ Benartzi and colleagues quote her and add that her calculations understate the cost, because human capital is not in them.⁴ Both figures depend heavily on the allocation, the holding period and the stock’s volatility, so they are illustrations rather than a rate anyone should apply to their own position.

Our guide on selling RSUs at vest turns all of this into one question, the cash-bonus test, which remains the fastest way to settle a vest. For an aged position with a large embedded gain the arithmetic changes, and the break-even guide handles that case.

Where a home enters the same shock

A large, high-paying employer does not only employ its own staff. Its employees buy restaurant meals, childcare, dentistry and home renovations. Suppliers and complementary firms cluster nearby. Workers move in. Housing sits inside that same local equilibrium, so a shock to the employer can reach the housing market through a chain: employer health, then local labor demand, then migration, then housing demand, then price.

Blanchard and Katz measured the housing end of that chain directly. Following an adverse state employment shock, relative housing prices decline steadily to a trough of about 2 percent after four to five years, then return to their previous level. The mechanism they document is migration rather than recovery: a state returns to normal after an adverse shock not because employment picks up but because workers leave. A one-worker decline in employment maps to roughly 0.3 more unemployed, 0.05 fewer participants, and 0.65 workers of net out-migration. The unemployment and participation effects fade within five to seven years, while the effect on the employment level is permanent.⁸

Two percent is a small number and it should be read as one. This layer is not going to be what ruins you. It earns its place in the stack for a different reason: it moves at the same time as the other three, and it is the only one you have borrowed money to hold.

Notowidigdo supplies the mechanism in a spatial equilibrium model, built on Roback with a durable, concave housing supply curve, in which housing prices adjust to local labor demand shocks. One finding of his is worth stating carefully, because it is easy to garble: the asymmetry he documents, where positive shocks raise population more than negative shocks reduce it, is absent for wages, housing values and rents. Those respond symmetrically. His point about incidence is the one that matters here, which is that adverse shocks substantially reduce the cost of housing, shifting the burden onto homeowners and landlords.¹⁰

On the employment side, Gathmann, Helm and Schönberg used German administrative data to trace what happens to a region after a mass layoff. Affected regions lose considerably more jobs than the initial layoff, with the additional losses concentrated in firms in the same broad industry, and they estimate a local employment multiplier of about 0.4.¹¹ That is at the low end of the published range, which brings us to the size of the multiplier.

How large the local multiplier is, and the case that it is smaller

Moretti estimated that each additional tradable-sector job in a city generates additional jobs in the local non-tradable sector, with the size depending on the sector and the skill level.⁶ The high-tech figure is the one that gets quoted, and it is the largest in his table.

Osman and Kemeny revisited Moretti’s estimates and reached materially smaller numbers: about half a non-traded job per traded job, less than one per high-tech job, and no evidence that manufacturing has a higher multiplier than the average traded sector.⁷ That is a direct challenge to the 4.9, and it is the reason the calculator below opens on its middle linkage setting rather than its highest. If the multiplier is closer to Osman and Kemeny’s estimate, the housing layer is smaller than the strong setting implies.

Supply elasticity decides how much of a demand shift becomes price

A change in housing demand can show up as more housing or as higher prices, and which one depends on how easy it is to build. Saiz measured this across US metros and found enormous variation driven by geography and land-use regulation together. The population-weighted average metro elasticity is 1.75. In land-constrained large metros it falls below one: Miami at 0.60, Los Angeles at 0.63, San Francisco at 0.66, San Diego at 0.67, Oakland at 0.70, Salt Lake City at 0.75. At the other end, Wichita is 5.45, Fort Wayne 5.36, Indianapolis 4.00.⁹

The same employer shock therefore lands very differently in two metros. Where supply is elastic, the adjustment happens through construction and migration and prices move less. Where it is inelastic, more of it becomes price. This cuts both ways, and it is a large part of why housing in constrained metros appreciated so much in the first place. Our guides on housing affordability for high earners and the fifty-year mortgage both lean on the same research.

If you rent, layer four largely drops out. A renter in an inelastic metro meets the same local shock through rent, which is a cost that falls when the local economy weakens, and holds no levered claim on local property at all.

A mortgage levers the fourth layer

Here the word leverage stops being a metaphor. Home equity is the difference between two numbers, only one of which moves with the housing market. The percentage change in your equity equals the percentage change in the home’s value multiplied by the ratio of home value to equity.

The mortgage does not shrink when the house does. That asymmetry is what converts a modest price move into a large move in the equity you actually own, and it is the only place in the four-layer stack where borrowed money is involved. It also runs in your favor when prices rise, which is the reason people take the loan. Its effect on this stack is to make the household more sensitive to a local downturn than the headline price change suggests, at a moment when three other lines are responding to the same event. Our guides on counting a home in net worth and the house-poor stress test cover the affordability side of the same purchase.

Put your own numbers in. The calculator values the four layers, applies one shared shock, and shows what changes if the vested shares are sold and reinvested.

What selling the vested shares changes

Run the comparison and the striking result is how little moves. Three of the four rows are identical between the household that holds and the household that sold, because selling employer stock does nothing at all to your salary, your unvested grants or your house. The fourth row swaps one concentrated position for a diversified one whose systematic exposures you keep.

The dollar difference between the two households is just the vested position multiplied by the gap between the employer stock’s move and the market’s. For the household the calculator opens on, that is a modest fraction of the total damage the scenario does. Most of the loss lands in the layers that have no market price and no sell button.

That modesty is the argument for selling, not against it. You cannot do much about the three big layers. You can do something about the small one at essentially zero cost, and refusing to is a choice to keep the only piece of employer exposure that was ever optional. Selling is also not a view on the company. If the employer thrives, the household still participates through the pay, the promotions and the grants that keep arriving. Diversifying the vested shares trims the exposure you can control and leaves the rest intact.

When holding the vested shares is defensible

The default is not a prohibition. Several situations make holding reasonable, and a few make it the right answer.

• A small position under a written policy. Deciding in advance to hold a specific percentage of investable assets, with a review date, is a different act from letting quarterly vests accumulate. Deliberate concentration you sized is defensible. Concentration that happened to you is not a thesis.
• You rent, or your metro is not driven by this employer. Layer four largely disappears, which leaves a shorter stack and more room in the risk budget.
• Plan mechanics come first. Insider status, blackout windows and 10b5-1 requirements determine when you can trade at all. ESPP shares often carry holding rules attached to the discount. Sort those out before the portfolio question.
• An aged, low-basis position. Shares held for years carry an embedded gain that selling realizes, which changes the math from the at-vest case. See the break-even guide for what the stock has to beat, and the four things you can do with a capital gain for the exits.
• Shares with special tax treatment. Qualified small business stock under Section 1202, or employer stock inside a 401(k) where net unrealized appreciation applies, involve one-shot decisions that are worth professional advice before you sell anything.

How Summitward helps

The dashboard’s portfolio tab measures what share of your invested assets sits in any one position and what that concentration does to the portfolio’s volatility and drawdown profile. The assets page tracks home equity and mortgage balance alongside the rest of the balance sheet, so the fourth layer is visible instead of implied. If you want to see the concentration before signing up, the free portfolio X-ray tool takes a ticker list and returns the same overlap analysis.

Frequently asked questions

Does selling my vested shares mean I am betting against my employer?

No. Your salary, your promotion prospects and every future grant remain claims on the company’s success, and for most employees those are far larger than the vested shares. Selling reduces the one exposure you chose to keep and leaves the three you did not choose.

My employer is not a tech company. Does the housing layer still apply?

It applies wherever an employer or its industry is a meaningful share of local high-income employment. Moretti’s largest multipliers are in high tech, but he finds a multiplier of roughly 1.6 for manufacturing generally, and Blanchard and Katz measured state-level employment shocks across the whole economy. The layer is proportional to how much of the local job market your employer and its peers account for.

I rent. Which layers apply to me?

The first three. A renter still has human capital, unvested equity and vested stock loading on the same employer, but holds no property and no mortgage, so the fourth layer and its leverage drop out. Rent in a weakening local economy tends to fall, which cushions rather than compounds the other three.

Is the housing effect large enough to worry about on its own?

On its own, probably not. A trough of about 2 percent in relative state house prices, recovering over subsequent years, is not a catastrophe, and national interest rates, demographics and construction costs move housing far more than any single employer does. It is worth attention because it arrives alongside a weaker job market, smaller grants and a lower share price, and because a mortgage magnifies whatever move does occur.

Should I count my home in this at all?

For this exercise, yes, at full value with the mortgage shown separately. A house is exposed to local prices at its whole value no matter how it was financed, and netting the loan against it hides exactly the leverage the exercise is meant to reveal. For net worth reporting the netted figure is the right one, which is a different question covered in counting a home in net worth.

What about the tax bill on selling?

On shares sold at or near vest there usually is not one worth mentioning, because the vest already reset your basis to the amount included in your income. The tax question becomes real for shares held long enough to build a gain, which is a different decision with different math.

Does diversifying into a sector fund solve this?

Only partly. Moving from one employer to a fund holding its closest competitors removes company-specific risk and keeps the industry exposure that also drives your pay and your local labor market. A broad global portfolio is the version that addresses the stack. See the tech-bro portfolio, the case for global diversification and position sizing.

Key takeaways

• Count the exposures before you evaluate the stock. Human capital, unvested equity, vested shares and, for homeowners in an employer-dominated metro, housing wealth can all load on the same firm.
• The housing channel is indirect and modest. Blanchard and Katz measured relative state house prices reaching a trough of about 2 percent four to five years after an adverse state employment shock, with adjustment running through out-migration rather than an employment rebound.
• How much of a demand shift becomes price depends on the metro. Saiz measured supply elasticities from 0.60 in Miami to 5.45 in Wichita, against a population-weighted average of 1.75.
• The local multiplier is contested. Moretti put high tech at 4.9 additional local jobs per traded job; Osman and Kemeny put it below one. Assume the smaller number unless you have local evidence.
• A mortgage is the only real leverage in the stack. On a home worth two and a half times its equity, a 1% price move is a 2.5% equity move, and the loan balance does not fall with the house.
• Vested shares are the cheapest exposure to shed and the only optional one. Sold at vest they carry almost no embedded gain, and selling them leaves your pay, your grants and your house exactly where they were.

Related guides

• Should You Sell RSUs at Vest? for the cash-bonus test and what to do at each vest.
• What your company stock has to beat for the aged, low-basis case where tax deferral is worth something.
• Human capital risk for tech workers for the balance-sheet framing behind layer one.
• The house-poor stress test for whether the mortgage itself is affordable on base pay alone.
• Most stocks lose to T-bills for the skewness of individual-stock outcomes.
• Compensated vs. uncompensated risk for which risks carry an expected premium.
• Your household portfolio for seeing every account and asset as one allocation.
• Concentration risk for measuring concentration and planning a sell-down.
• Counting a home in net worth for how home equity belongs on the balance sheet.

Sources and method

1. Hendrik Bessembinder, Do Stocks Outperform Treasury Bills?, Journal of Financial Economics 129(3), 2018, 440–457. CRSP common stocks 1926–2016, approximately 25,300 firms. 42.6% of stocks beat one-month Treasury bills over matched horizons; 1,092 firms, slightly more than 4%, account for all of the roughly $35 trillion of net lifetime wealth creation measured to December 2016. SSRN
2. Bessembinder, Te-Feng Chen, Goeun Choi and K.C. John Wei, Long-Term Shareholder Returns: Evidence from 64,000 Global Stocks, Financial Analysts Journal 79(3), 2023, 33–63. January 1990 to December 2020. 55.2% of US and 57.4% of non-US stocks underperform one-month US Treasury bills in compound returns; the top 2.4% of firms account for all $75.7 trillion of net global wealth creation. Earlier working-paper title: Do Global Stocks Outperform U.S. Treasury Bills? SSRN
3. Hendrik Bessembinder, One Hundred Years in the U.S. Stock Markets, SSRN working paper posted 19 March 2026. Not yet peer reviewed. 29,754 stocks issued by 29,081 firms, January 1926 to December 2025. $91 trillion of net wealth creation, of which 46 firms account for half and 1,082 firms (3.7%) account for all; nearly 60% of stocks produced wealth reductions relative to Treasury bills; cross-stock mean buy-and-hold return above 30,000% against a median of −6.9%. SSRN
4. Shlomo Benartzi, Richard H. Thaler, Stephen P. Utkus and Cass R. Sunstein, The Law and Economics of Company Stock in 401(k) Plans, Journal of Law and Economics 50(1), 2007, 45–79. Source of the worst-case quotation and of the point that Meulbroek’s certainty-equivalent calculations understate the cost of company stock because human capital is excluded from them. SSRN
5. Lisa Meulbroek, Company Stock in Pension Plans: How Costly Is It?, Journal of Law and Economics 48(2), 2005, 443–474. Employees “sacrifice an average 42% of their company stock’s market value by taking on risk that could otherwise have been diversified away.” The 58 cents and 33 cents figures are hers, not Benartzi et al.’s, and depend on allocation, horizon and volatility. Working paper PDF
6. Enrico Moretti, Local Multipliers, American Economic Review: Papers & Proceedings 100(2), 2010, 373–377. Manufacturing 1.6, skilled tradable 2.5, unskilled tradable 1.0, high tech 4.9. The frequently quoted “five jobs, two professional and three non-professional” formulation comes from The New Geography of Jobs (2012), not from this paper. PDF
7. Taner Osman and Tom Kemeny, Local Job Multipliers Revisited, Journal of Regional Science 62(1), 2022, 150–170. Each new traded-sector job adds about half a non-traded job; each high-tech job adds less than one; manufacturing is no higher than the average traded sector. DOI
8. Olivier J. Blanchard and Lawrence F. Katz, Regional Evolutions, Brookings Papers on Economic Activity 1992(1), 1–75. Relative housing prices decline to a trough of about 2 percent four to five years after an adverse state employment shock, then return to their previous level. A one-worker employment decline maps to roughly +0.3 unemployment, −0.05 participation and 0.65 net out-migration; unemployment and participation effects disappear after five to seven years while the employment level effect persists. PDF
9. Albert Saiz, The Geographic Determinants of Housing Supply, Quarterly Journal of Economics 125(3), 2010, 1253–1296. Population-weighted mean metro elasticity 1.75 (unweighted 2.5); Miami 0.60, Los Angeles–Long Beach 0.63, San Francisco 0.66, San Diego 0.67, Oakland 0.70, Salt Lake City–Ogden 0.75; Wichita 5.45, Fort Wayne 5.36, Indianapolis 4.00, Dayton–Springfield 3.71. QJE
10. Matthew J. Notowidigdo, The Incidence of Local Labor Demand Shocks, Journal of Labor Economics 38(3), 2020, 687–725. Spatial equilibrium model with a durable, concave housing supply curve in which housing prices adjust to local labor demand shocks. The asymmetry between positive and negative shocks applies to population; it is absent for wages, housing values and rental prices, which respond symmetrically. Adverse shocks substantially reduce the cost of housing, shifting incidence onto homeowners and landlords. PDF
11. Christina Gathmann, Ines Helm and Uta Schönberg, Spillover Effects of Mass Layoffs, Journal of the European Economic Association 18(1), 2020, 427–468. German administrative data; affected regions lose many more jobs than the initial layoff, concentrated in firms in the same broad industry, with an estimated local employment multiplier of about 0.4. DOI
12. IRS Publication 15 (Circular E), Employer’s Tax Guide, for use in 2026, section 7. Supplemental wage withholding remains a flat 22% under the optional flat-rate method, and 37% on supplemental wages exceeding $1 million in a calendar year, withheld without regard to Form W-4 and aggregated across businesses under common control. PDF
13. IRS Publication 5992 (6-2024), Equity (Stock)-Based Compensation Audit Techniques Guide, section G.3. Restricted stock units are not Section 83 property at grant, so no 83(b) election is available; Section 83 applies when the stock is actually transferred at settlement, and the value transferred is includable in income. Basis reflects the amount included in income and the holding period begins just after substantial vesting under Treas. Reg. sections 1.83-4(b)(1) and 1.83-4(a). PDF
14. The calculator is reproducible from web/src/lib/employer-exposure-math.ts, which carries its reference values. The home shock is the Blanchard-Katz trough scaled by a linkage dial and by the ratio of Saiz’s population-weighted mean elasticity to the selected metro’s, so a mean-elasticity metro at the moderate setting reproduces the measured trough exactly. The linkage settings are scenario dials rather than estimates for any particular metro or employer, which is why an override is provided. Unvested equity takes the stock move and the pay haircut together. Human capital is a level real annuity, gross of tax, with no offsetting consumption liability. Nothing here is a forecast, and nothing here is tax, legal or investment advice.