§0 · What's already been said, and what we're adding

The behavioral economics of 401(k) participation is one of the most heavily-researched areas in retirement finance. Brigitte Madrian and Dennis Shea documented in 2001 that default enrollment dramatically increases plan participation. Richard Thaler and Shlomo Benartzi proposed Save More Tomorrow in 2004, demonstrating that opt-out architecture combined with future-dated commitments substantially raises savings rates. James Choi, David Laibson, Brigitte Madrian, and Andrew Metrick have produced over a decade of work on plan rules, participant decisions, and the path of least resistance in defined contribution pensions. Alicia Munnell at Boston College's Center for Retirement Research has tracked plan design and adequacy systematically. EBRI publishes regular Issue Briefs on participation rates, contribution patterns, and asset allocation. Fidelity, Vanguard, and Empower publish quarterly and annual plan statistics covering tens of millions of participants.

The legal literature is equally developed. ERISA Section 411 specifies the vesting schedules employers may use for matching contributions: immediate vesting, three-year cliff vesting, or two-to-six-year graded vesting. Section 401(a) and 401(k) regulations govern plan structure. The IRS publishes annual updates to contribution limits and forfeiture treatment guidance. The Plan Sponsor Council of America publishes an annual survey of profit-sharing and 401(k) plan design choices.

None of this is news.

What this investigation does — and what is genuinely underdeveloped in the worker-facing literature — is bring the labor-market analytical lens that Investigation #1 (The Payroll Lag) applied to wage timing to the vesting-schedule structure of employer matching. Academic labor economics has treated vesting as a lock-in / switching-cost mechanism since at least the 1980s (the Even–Macpherson body of work on pension vesting and worker mobility; the broader lock-in-contracts literature). The behavioral literature treats forfeitures as a cognitive failure to optimize tenure. The legal literature treats them as legitimate outcomes of contractually-disclosed plan terms. What none of these framings gives the worker is the analytical move this investigation makes: that the same nominal match produces two entirely different economic packages depending on which side of the vesting threshold a given worker sits on, and that the cliff and long-graded schedules ERISA permits create a labor-vesting asymmetry — labor accrues continuously while vesting accrues binarily (cliff) or in annual steps (graded) — such that the worker bears the entire gap between provided labor and captured match. The structural mechanism is present in the academic literature; the two-class articulation and the consumer-facing frame are what this investigation adds.

The Payroll Lag identified the mechanism by which workers extend their employers interest-free loans through pay-in-arrears at HR-set cadence. The 401(k) Match Forfeiture is the same mechanism extended into the deferred-compensation layer of total compensation: workers provide labor continuously under a compensation package that includes the match; the vesting condition decides whether the labor's contribution to the match converts to vested capital; the asymmetric structure means workers in their first years of tenure bear forfeiture risk while veterans bear none, even though the match is sold to all workers identically. The firm prices the expected forfeiture into its labor-cost budget at the cohort level. The worker hears the headline match percentage.

This is The Payroll Lag, Part Two. Same brand argument applied to the same population through a structurally adjacent instrument. The dollar magnitudes scale up substantially when compounded across working lifetimes.

The 401(k) Match Forfeiture

If you work 36 months and 1 day at an employer with a 3-year cliff vesting schedule, you keep 100 percent of your accumulated employer match. If you work 35 months and 29 days at the same employer, you keep zero percent. For a worker earning $100,000 whose employer match runs the typical ~3.4 percent of pay, that 60-day difference in tenure represents roughly $10,000 in vested versus forfeited deferred compensation. Compounded over the 35 years between mid-career and retirement, the forfeited amount becomes approximately $55,000 to $80,000 in retirement capital (at a 5–6% real return; this is the opportunity cost of the forfeiture invested and left untouched — not the reduction in the worker's total projected balance).

That 60-day binary is not an unfortunate edge case. It is the visible surface of a structural feature: a single 401(k) plan produces two entirely different economic packages inside the same office. A worker in the first three years of tenure bears full forfeiture risk on every dollar of match accumulating in their account — and, if the plan uses a cliff schedule, bears it all binarily. A worker past the vesting threshold bears none. Both workers see the same headline match number in the offer letter and the annual benefits summary. Only one of them actually holds it as vested capital that survives a job change. The firm publishes one match rate; the plan operates as two.

The marketing version of an employer's match is the headline percentage: "we offer a 6 percent match." The actuarial version is smaller in two directions at once: the realized match across employers averages closer to 3.4 percent of pay than to the headline ceiling, and then expected forfeiture across the tenure distribution of new hires takes another cut. The firm prices both into its labor-cost budget. The worker does not run the math.

This is Investigation #2 not because the aggregate dollar magnitude is enormous. Annual national forfeitures run into the billions, but no single audited total is published, and the aggregate is more modest than the household-visible categories that dominate personal- finance coverage. The force of this investigation is the per-worker structure just named — the same nominal match producing two entirely different economic packages depending on which side of the vesting threshold the worker sits on, and the labor-vesting asymmetry that produces it.

The legitimate case for vesting — retention of workers the firm has invested training and ramp-up in — is real; §8 gives it the full steelman it deserves. What this investigation argues is not that vesting shouldn't exist. It argues that the specific cliff and long-graded structures ERISA permits do more than solve the retention problem, and that they persist as an equilibrium despite being reformable — a distinction §8 develops.

The mechanism is identical to The Payroll Lag, applied to the deferred-compensation layer rather than the wage timing layer. The firm has the cohort data, the actuarial model, and the labor-cost line item. The worker has the headline number and a hope of staying past the cliff.

§2 · The mechanism

The Employee Retirement Income Security Act of 1974 (ERISA) governs employer-sponsored retirement plans. Section 411 of ERISA defines the vesting requirements employers must satisfy for employer contributions, including matching contributions. Employee deferrals — money the worker contributes from their own paycheck — must be 100 percent immediately vested by statute. Employer contributions, including the match, may be subject to one of three vesting schedules:

Immediate vesting.

The employer match becomes the worker's property at the moment it is deposited. No forfeiture risk. The worker who joins on Monday and leaves on Friday keeps the match the employer deposited that week.

Three-year cliff vesting.

The employer match remains the employer's property — credited to the worker's account but subject to forfeiture — until the worker has completed three years of service. At the three-year anniversary, 100 percent of accumulated match vests in a single binary event. Workers who separate before three years forfeit the entire accumulated match. Workers who separate one day after three years keep all of it.

Two-to-six-year graded vesting.

The match vests in annual increments starting at year two. Plans must vest at least 20 percent per year beginning year two, reaching 100 percent by year six. The most common graded structure — and the most extractive of the legally-permitted graded options — vests 20 percent per year over six years (0% end of year one, 20% end of year two, 40% end of year three, 60% end of year four, 80% end of year five, 100% end of year six). Some plans use accelerated graded schedules (e.g., 33% per year over three years) but these are less common.

A brief technical note on measurement. Under ERISA §411, a "year of service" for vesting purposes is a 12-month plan year in which the worker completes at least 1,000 hours of service — not a calendar year of tenure. For a full-time worker the two track closely, so the month-based diagnostics later in this essay (the 35-month/37-month cliff example, the month-by-month graded table) are analytically sharp; for part-time, seasonal, or interrupted workers, vesting service can lag calendar tenure meaningfully. SECURE 2.0 carved out a partial exception for long-term part-time workers (500-hour year of vesting service, effective plan years starting 2025); the general 1,000-hour rule still governs standard employees. Where this essay says "months," it implicitly assumes a full-time worker whose service-year thresholds fall at calendar-month equivalents.

The PSCA Annual Survey of Profit Sharing and 401(k) Plans documents the distribution of vesting structures across surveyed plans. Per the 68th Annual Survey (2024 plan-year data), immediate vesting is now the single most common choice — 44.1 percent of plans with a match, up from 39.7 percent the year before and trending upward over the past decade. (Safe-harbor 401(k) plans, required by IRS rules to vest the employer contribution immediately, account for much of this.) The remainder split between three-year cliff and two-to-six-year graded schedules; the precise cliff- versus-graded breakdown sits inside the paid PSCA report. Cliff schedules concentrate in the largest plans (employers with 5,000+ participants use three-year cliffs about 23 percent of the time) while six-year graded schedules concentrate in the smallest plans (about 27 percent of sub-50-participant plans). The structure a worker faces depends heavily on the size of the employer.

Where do forfeited match dollars go? Per ERISA, plan administrators may use forfeitures in any of three ways:

1. Reduce future employer contributions. The employer's next quarter's match deposit is reduced by the amount of recent forfeitures. This is the standard treatment — and, since late 2023, the target of a wave of ERISA class actions alleging that using forfeitures to offset employer contributions rather than reallocate to participants breaches the plan sponsor's fiduciary duty. More than thirty coordinated suits have been filed against major plan sponsors (Intuit, HP, Wells Fargo, Mattel, Bank of America, Qualcomm, and others). Federal district courts have mostly dismissed the cases at the motion-to-dismiss stage — twenty-four of twenty-eight recent decisions — leaving the reduce-employer-contributions treatment legally permitted but actively contested. The forfeiture is a direct savings to the employer's labor cost.
2. Pay plan administrative expenses. The employer's responsibility for plan expenses is reduced. Also a direct savings to the employer.
3. Reallocate to remaining active participants. Active participants' accounts receive the forfeited match on a pro-rata basis. The worker who separated forfeited; the workers who stayed received a slight bonus.

In the first two cases — the majority of plans — the forfeiture flows directly to the employer's bottom line. The structural beneficiary of the cliff and long-graded vesting schedules is the firm, not the remaining workforce.

The match formula itself varies widely. Common structures include 50 percent match up to 6 percent of salary (at 6 percent worker contribution, the employer match equals 3 percent of salary); 100 percent match up to 4 percent of salary; and tiered structures (e.g., 100 percent match on the first 3 percent plus 50 percent on the next 2 percent, totaling 4 percent for a 5-percent-contributing worker), particularly common at larger employers.

In practice the realized match is smaller than the headline ceilings suggest. Per the PSCA survey, employers spend on average 3.4 percent of pay on matching contributions, with the average maximum available match at 4.7 percent. For a worker earning $100,000, the typical match is therefore on the order of $3,400 a year — not the $6,000 a "six percent match" implies, and not the $3,000–6,000 range a reading of the formula ceilings would suggest. This investigation uses ~3.4 percent of pay as its base case. Over a career spanning multiple employers the cumulative match still reaches well into the tens of thousands of nominal dollars; the vesting structures determine how much of it the worker actually keeps.

§3 · The instrument

The "free money" framing dominates how employer matching is discussed in financial-advice press, in HR onboarding materials, and in worker-facing benefits communications. "Don't leave free money on the table" is the standard guidance. The framing is correct as far as it goes — the match is real economic value the worker can capture by participating. But the framing is also editorially load-bearing in ways that obscure the underlying instrument.

Refuting the "free money" framing

The employer's match is not, in any rigorous economic sense, free money. It is part of the worker's total compensation package, negotiated implicitly through the labor market and explicitly through offer-letter disclosure. The financial-advice framing focuses the worker on the contingency (you must contribute) and the gift nature (the firm doesn't have to offer it). It does not focus the worker on the structural reality that the match is in the firm's labor-cost budget, factored into recruiting decisions, and used as a comp-package differentiator in tight labor markets.

Three pieces of evidence:

1. The match appears in offer letters and recruiter pitches.

Job descriptions advertise "competitive 401(k) match" as a benefit. Recruiters use match percentages to attract talent. Workers compare match offerings between job options as part of total-compensation calculations. The match is functioning as compensation, not as a gift.

2. The match is in the firm's labor-cost line item.

From the firm's perspective, the match is a planned expense. CFOs build it into compensation budgets. Total-compensation analyses include match dollars. The match is treated as comp on the firm's books.

3. The match substitutes for higher base salary.

If a firm dropped its match tomorrow, base salaries would need to rise to keep the same total-comp position in the labor market — or the firm would lose competitiveness in hiring. This tells you the match is part of comp, not a separate gift on top. Workers accepted lower base salaries because the match was a delivery vehicle for additional compensation.

Total-compensation accounting treats the match as comp. Modern labor economics treats it as comp. The firm's own books treat it as comp. Only the worker-facing framing treats it as "free money." That framing is doing real editorial work for the firm: it positions the match as a contingent gift rather than as deferred wages, which makes the cliff and long-graded vesting structures feel like reasonable conditions on a gift rather than capture mechanisms on earned compensation.

The labor-vesting asymmetry

The analytical move that unlocks the instrument is naming the asymmetry between how labor accrues and how vesting accrues.

Labor accrues continuously. Every hour worked adds to what the firm has received from the worker. Every day of tenure adds an incremental contribution to the worker's vesting period. The labor side is smooth and continuous.

Vesting accrues discretely. Cliff vesting is a binary event at the cliff date. Graded vesting is a step function with annual increments. There is no intermediate state. A worker who has provided 35 months of labor toward a 3-year cliff has provided 97 percent of the labor needed to vest the match — but captures 0 percent of the vested match if they separate one day before the cliff. A worker who provides 36 months and 1 day captures 100 percent.

The diagnostic example. Two workers at the same firm with the same salary and the same accumulated match deposits, separating 60 days apart. The first separates at month 35; the second at month 37. They have provided 35 versus 37 months of labor — a 5.4 percent difference in labor input. They keep 0 versus 100 percent of the match — an infinite difference in vested compensation per unit of labor. The labor difference and the vested-comp difference do not bear any meaningful relationship to each other.

For a worker earning $100,000 whose match runs the typical ~3.4 percent of pay (about $3,400 a year, roughly $10,200 accumulated over three years), the 60-day tenure difference represents approximately $10,000 in vested versus forfeited match, plus the investment returns those deposits would have earned. The labor difference is 60 days — about 5 percent of the three-year vesting period. The vested-compensation difference is not 5 percent; it is total. Zero versus everything. The cliff produces a per-day vesting rate of zero across 35 months and then a single ~$10,000 step on one day. The size of that step bears no relationship to the labor provided on either side of it.

That is the labor-vesting asymmetry made arithmetically explicit. The cliff is not a smooth retention incentive. It is a binary capture function.

Graded vesting is the same asymmetry, smoothed

Graded vesting is meaningfully better than cliff vesting because it removes the all-or-nothing binary at a single date. But the asymmetry doesn't disappear — it smooths out.

Consider a six-year graded plan vesting 20 percent per year starting year two (the most extractive of the ERISA-permitted graded options). For a worker earning $100,000 with the typical ~3.4 percent match (~$3,400 per year in deposits), the year-by-year picture:

Three observations:

Year one is still a cliff. A worker who separates at month 11 forfeits 100 percent of year one's match. The graded structure begins vesting at year two; year one is binary. The "no cliff" framing applied to graded vesting plans is technically misleading — graded plans have a one-year cliff plus stepwise annual thresholds after that.

Peak forfeiture exposure is years 2–4. The $6,120 forfeit at end of year three is the worst single moment. The labor-vesting gap is widest here — the worker has earned half of full vesting tenure (3 of 6 years) but vested only 40 percent of match.

Years 2–6 are linearly underpaying labor. A worker at month 30 (year 2.5) has provided 42 percent of the six-year labor toward full vesting but vested only 20 percent of match. A worker at month 42 (year 3.5) has provided 58 percent of labor and vested 40 percent. The 35-month/37-month binary becomes the cliff-at-year-two binary and the smaller binaries at each subsequent annual threshold.

The two-class system

Vesting is typically presented as a single plan feature — one cliff schedule, one graded schedule, one match — applied uniformly to all participants.

That framing misses the most important structural observation about vesting: a single match plan produces two completely different economic realities depending on which side of the vesting threshold the worker stands on.

A new hire in the first three years of tenure (or first six years under a graded plan) bears all of the vesting risk. Every dollar of match accumulated during this period is subject to forfeiture if the worker separates before the vesting threshold. The worker has switching costs from the firm — they cannot leave for a better opportunity without paying the forfeiture penalty. The match's expected economic value to this worker is the headline match percentage multiplied by the probability they remain employed through vesting, which can be substantially less than the headline.

A veteran past the vesting threshold bears no vesting risk. Past contributions are 100 percent vested. New contributions vest immediately (in cliff plans) or at the highest graded rate (in graded plans). The worker has zero switching cost — they can leave any day and take all accumulated match with them. The match's expected economic value to this worker is the headline match percentage at full value.

Same nominal match. Two different economic packages. Two different career incentives. Two different switching costs.

The firm publishes one headline match rate. It produces two different economic packages.

Portable in penalty, not in benefit

A veteran at Company A who leaves and joins Company B resets to pre-vesting status at Company B. The cliff applies again. The graded schedule starts again. The worker who has spent six years at Company A reaching full vested status starts over at month zero at Company B.

This means a worker who changes jobs frequently is permanently in pre-vesting status. The cliff is portable in penalty (you always bear the vesting risk at your current employer) but not in benefit (you do not carry over vested status when you switch). A career involving five employers means five separate vesting periods, each subjecting the worker to forfeiture risk during the early years of each employment, never reaching the veteran's post-vesting state of zero switching cost.

For workers whose career paths involve typical American mobility — BLS data puts median private-sector tenure around 4 years for ages 25–34, rising to ~10 years for ages 55–64 — the cumulative effect across a career is meaningful. The cliff and long-graded structures systematically extract from anyone whose career involves multiple employers. They systematically benefit anyone whose career is concentrated at a single employer for long enough to cross the vesting threshold.

The firm prices this. ERISA disclosure data plus actuarial assumptions let plan sponsors calculate expected forfeiture rates by tenure cohort. Cliff vesting is specifically chosen — over graded vesting or immediate vesting — because the firm has run the math and knows the expected capture rate. The match line item in the firm's compensation budget is net of expected forfeitures. The worker does not run that math. The worker hears "6 percent match" and thinks "6 percent," not "6 percent × P(staying 3+ years)."

§4 · Where you sit in the distribution

The math is straightforward enough to run on the back of an envelope. Cumulative match is salary times match percentage times years. Vested percentage is a table lookup based on schedule and tenure (see §3's graded- vesting table for the base case). Forfeiture exposure is the accumulated match not yet vested. The compound retirement-balance projection is the forfeiture times (1 + real return) raised to the years remaining until retirement. For the $100,000 base-case worker used throughout this essay, the numbers scale roughly linearly with income — a $150,000 worker sees numbers ~1.5x larger; a $60,000 worker sees them ~0.6x.

More useful than a personal calculator is the population-distribution data that lets workers self- locate against the broader workforce:

• Vesting structure distribution across US 401(k) plans (PSCA 68th Annual Survey, 2024): 44.1 percent immediate vesting; cliff-versus- graded breakdown across the remainder sits inside the paid PSCA report (see §2). Cliff schedules concentrate in the largest plans; six-year graded schedules concentrate in the smallest.
• Median tenure by age cohort (BLS Employee Tenure, January 2024): 2.7 years for ages 25–34; 4.6 years for ages 35–44; 7.0 years for ages 45–54; 9.6 years for ages 55–64 (overall median 3.9 years, the lowest since 2002).
• Industry forfeiture concentration (PSCA, directional): higher in high-turnover sectors (retail, hospitality, food service), lower in stable-tenure sectors (government, healthcare, education).
• Employer resets across a career — the BLS NLSY79 cohort held an average of ~12.7 jobs from ages 18–56 (not 12 employers, and nearly half before age 25), so the number of distinct vesting resets across a career is smaller than the raw job count implies; the gap still accumulates across each genuine employer change.

§5 · Who actually pays

Demographic cohorts

The cliff and long-graded vesting structures systematically extract more from demographic cohorts whose career paths involve more frequent employer changes or shorter expected tenure at each employer.

• Younger workers (ages 25–34) carry median tenure of just 2.7 years per BLS (January 2024) — below the three-year cliff. The median worker in this cohort separates from their average employer before the cliff vests, meaning more than half of new-job experiences for this cohort end in forfeiture. With six-year graded vesting, the median worker captures only about 20 percent of match across their average employer.
• Lower-income workers experience higher involuntary turnover through layoffs, restructuring, and temporary or contract work. They cross fewer vesting thresholds per career than higher-income workers in stable salaried positions.
• Women in childbearing years experience career interruptions and re-entries that fragment tenure across multiple periods. Each re-entry resets vesting at the new employer.
• Career-switchers and mid-career professional pivots reset to pre-vesting status at each switch. The cliff is portable in penalty, not in benefit.
• Workers in high-turnover industries (retail, hospitality, food service, early-stage tech, gig work) face industry baseline turnover that exceeds typical vesting horizons.

Stable mid-career and late-career workers in low-turnover industries (corporate finance, government, established healthcare, established education) bear less vesting risk per unit of nominal match. The cliff structure thus systematically subsidizes stable veterans at the expense of mobile new hires — same nominal benefit, different economic reality, no acknowledgment of the asymmetry in the firm's marketing of the comp package.

Veterans are not villains in this picture. They are not capturing anything from new hires directly. The firm is the structural beneficiary; veterans happen to be on the favorable side of the structure.

Tenure-cohort exposure

The five subscriber-cohort framework used in the Subscription Lag investigation doesn't transfer directly to 401(k) Match Forfeiture, because the relationship to forfeiture is largely structural rather than behavioral. Workers don't usually choose to "Optimize" or "Default" into vesting exposure the way they choose subscription engagement levels. Vesting exposure is dictated by tenure path.

A more useful cohort framework:

Pre-vesting workers

First three years (cliff) or first six years (graded) at their employer. Bear full or partial vesting risk on accumulated match. Have switching costs from the firm. Carry implicit downward pressure on negotiating leverage because leaving means losing the unvested match.

Vested veterans

Past the vesting threshold at their current employer. Past contributions and new contributions vest immediately. Switching costs are zero. Highest negotiating leverage; can leave for a better opportunity without compensation penalty.

Permanent pre-vested

Workers whose career paths involve frequent employer changes (typically 2–3 year tenures with each employer). Reset to pre-vesting status at every new job. Never reach the veteran state at any employer. Bear the most cumulative forfeiture exposure across their careers.

Lifetime non-participants

Workers at employers that do not offer matching, or workers who do not participate in the plan. Forfeit no match because there's no match to capture. Bear a different opportunity cost — the absence of match value entirely — outside this investigation's scope.

The salience problem

Workers do not generally feel the cliff and graded vesting structures as immediately painful in the way that the Payroll Lag's cash-flow timing produces visible monthly stress. The forfeited match is dollars the worker cannot touch for 40+ years anyway. The behavioral economics literature documents heavy discounting of future financial consequences. A ~$10,000 forfeiture at age 30 reads as "abstract money far in the future" to most workers who experience the separation.

The investigation engineers salience through three editorial moves:

1. Compound retirement balance framing. $10,000 forfeited at age 30, compounded at a 5–6 percent real return for the 35 years until age 65, becomes roughly $55,000 to $80,000 in retirement capital. The framing converts "$10,000 you can't touch for decades" into the much larger figure it would have grown into — the cost is not the deposit lost, it is the compounding lost.
2. Multiple-job-change aggregate. Careers involve multiple employer changes, and each genuine change resets vesting. The often-cited "12 jobs" figure (BLS NLSY79: ~12.7 jobs from ages 18–56, nearly half before age 25) overstates the number of employer resets — but even a handful of pre-vesting separations across a career compounds into a retirement-balance gap structurally larger than any single forfeiture event suggests.
3. Reframe in present-tense terms. "Your future self is the worker who pays the bill, and your future self has no standing to negotiate with your current employer." The forfeiture happens to your present self (the deposits show up on your present statement, then disappear). The cost is borne by your future self (the retirement balance that doesn't compound from missing capital). Naming the temporal mismatch sharpens the perceived stakes.

§6 · The scale

Aggregate forfeiture data at the national level is not published as a single audited source. The PSCA Annual Survey, EBRI Issue Briefs, and the major plan administrators (Fidelity, Vanguard, Empower) publish plan-level and cohort-level statistics; the DoL's Form 5500 filings carry the underlying data at plan level. No source rolls them into a single national forfeiture total that can be cited without caveat. What follows is what can be established from public sources plus the primary-source evidence of the 2024 ERISA class-action wave.

Annual aggregate forfeiture of unvested employer matches across US private-sector 401(k) plans runs into the billions of dollars annually. The scale is indirectly attested by the 2024 wave of ERISA class actions: plaintiffs' firms would not organize thirty-plus coordinated cases against major plan sponsors without meaningful per-plan dollar pools at stake. Industry commentary places the aggregate in the low single-digit billions; a defensible national total awaits either the paid PSCA data or a Form 5500 recompute.

Forfeiture destination. The majority of plans use forfeitures in ways that reduce the employer's labor cost — either by reducing future employer contributions (the standard treatment) or by paying plan administrative expenses the sponsor would otherwise fund. A minority reallocate forfeitures to remaining active participants. The precise split is inside the paid PSCA report. The directional claim — that most forfeited match benefits the employer rather than co-workers — is supported by the ERISA class-action wave targeting the reduce-employer-contributions practice specifically.

Aggregate plan asset volume. Per Investment Company Institute (ICI) data, US 401(k) plan assets totaled $8.9 trillion at year-end 2024 across roughly 70 million active participants (and crossed $9.9 trillion by Q1 2026). Ninety-one percent of participants are in plans that offer an employer contribution. Total annual employer matching contributions are not published as a single audited figure; order-of-magnitude estimates run in the tens of billions per year. The forfeited slice — match that fails to vest before participant separation — sits inside that pool but is not separately audited at the national level.

Compound retirement balance impact. The present-value impact of these forfeitures on participants' retirement balances is substantially larger than the nominal annual forfeiture figure, because the cost is the compounding lost, not just the deposit. Compounded at a 5–6 percent real return over affected workers' remaining working lifetimes, the lifetime gap runs to tens of billions of dollars of foregone retirement capital. This is an order-of-magnitude estimate; the input aggregate is not audited at the national level and the tenure-distribution assumptions are drawn from BLS medians per §9. Treat as illustrative, not authoritative.

Per-worker exposure varies dramatically by tenure path. A worker with stable career patterns (one to two employers across a working lifetime, each tenure exceeding the vesting threshold) bears minimal forfeiture exposure. A worker with typical American mobility patterns (overall BLS median tenure of 3.9 years, and an average of ~12.7 jobs over ages 18–56 — not employers, and nearly half before age 25) bears substantial cumulative exposure. The cliff-vesting-with- frequent-employer-changes worker represents the worst-case exposure: every employer change resets to pre-vesting status, with each separation occurring close to the cliff date producing maximum forfeiture per event.

The aggregate figures are modest compared to the household-visible extraction categories the personal- finance press typically covers. But the per-worker exposure for affected workers is meaningful (a typical ~$10,000 cliff forfeiture compounds into roughly $55,000–80,000 in retirement capital), and the structural inequity of the two-class system within identical nominal match plans is the analytical centerpiece — not the headline dollar magnitude.

§7 · The asymmetry

The investigation argues that the 401(k) Match Forfeiture operates with three nested layers of unpriced information, mirroring the structure The Payroll Lag identified in the wage-timing layer.

First, the worker doesn't price the actuarial match value against the headline match value. The headline match is the percentage advertised in offer letters and benefits brochures. The actuarial match value is the headline multiplied by the worker's probability of remaining employed through the vesting threshold. For workers in cohorts with median tenure below the vesting threshold — most younger workers, many lower-income workers, and most workers in high-turnover industries — the actuarial value is meaningfully below the headline. The worker does not run this multiplication. The firm runs it as a standard input to its labor-cost budget.

Second, the worker doesn't price the labor-vesting asymmetry itself. The cliff structure is treated by workers as a "rule of the plan" — something to navigate by trying to stay past the cliff date. It is rarely framed as a binary capture function that bears no relationship to actual labor provided. The 35-month versus 37-month diagnostic is invisible until someone names it. Once named, the asymmetry becomes obvious; before naming, it operates as background noise of "the way 401(k) plans work."

Third, the worker doesn't price the two-class system. Different workers at the same firm bear different vesting risk based on which side of the vesting threshold they sit on. The new hire and the veteran are sold the same "6 percent match" but face very different economic packages. This produces a structural inequity that the firm's marketing of the comp plan does not acknowledge.

The firm prices all three layers. The actuarial value is in the budget. The labor-vesting asymmetry is a chosen feature of plan design (immediate vesting is available; the firm chose cliff or graded for a reason). The two-class system is acknowledged in compensation actuarial models (which distinguish new-hire match cost from veteran match cost). The worker prices none of the three. The match they negotiate over, decide to participate in, and value when comparing job offers is the headline. This is a structural information asymmetry in the formal economic sense (Akerlof, The Market for Lemons, 1970), produced because employers who shroud the actuarial price out-compete employers who reveal it (Gabaix & Laibson, Shrouded Attributes, 2006).

Same nominal benefit. Two completely different economic realities. The gap unpriced at the worker level and explicitly priced at the firm level.

§8 · The case for change

The strongest honest argument for vesting is that it solves a real problem for firms. Employee turnover is expensive. Onboarding, training, and ramp-up to productivity carry real costs — SHRM and BLS data place hiring plus training costs at 50–200 percent of annual salary for skilled roles. Firms that invest in their workers need retention mechanisms to avoid losing the investment to competitors who didn't make the investment. Vesting is one of several retention tools (alongside non-compete agreements, restricted stock units, signing bonuses with clawback) that lets firms invest in workers without bearing the full turnover risk. Without retention mechanisms, the optimal firm strategy is to minimize worker investment — treat workers as fungible, minimize training, never develop talent. That outcome is bad for workers too.

Stable workforces also produce better products. Knowledge transfer, institutional memory, team cohesion, and customer relationships all benefit from longer average tenure. Vesting aligns the firm's investment horizon with the worker's career horizon in ways that benefit both.

These arguments are real. The investigation does not dismiss them.

What the retention steelman does not justify is the cliff structure specifically, or the most extractive of the graded structures. A three-year cliff produces zero vesting for 35 months of labor and 100 percent vesting for 36 months — a binary capture function that bears no meaningful relationship to the labor provided. Graded vesting (20 percent per year over six years) preserves the same retention incentive in the sense that workers still have skin in the game and full vesting still takes years, while producing a smoother capture function that more closely approximates the labor provided. The cliff is the choice the firm makes; the firm can make a different choice and lose almost none of the retention value.

Why hasn't the market fixed this on its own?

If the labor-vesting asymmetry is knowable and the fix is technically simple, why does the equilibrium persist? "Extraction exists" is a diagnosis; "extraction persists in equilibrium despite being reformable" is a diagnosis plus a mechanism. The reform proposals below are more targeted once we know what is actually holding the equilibrium in place.

Part of the answer is that partial arbitrage has already happened. Safe-harbor 401(k) plans require immediate vesting on the safe-harbor employer contribution, and per the PSCA 68th Annual Survey (2024) immediate vesting is now the plurality choice — 44.1 percent of matching plans, rising. Nearly half of the matching-plan market has already priced this and moved.

For the plans that still use cliff or long-graded vesting, four interlocking reasons hold the equilibrium in place:

1. Two-sided information asymmetry.

Workers evaluating offer letters cannot price the actuarial difference between "6 percent match with 3-year cliff" and "6 percent match with immediate vesting." This is the essay's structural thesis. Because workers cannot price the difference, firms that offer better vesting get no competitive credit in the labor market — the reward for reforming isn't there.

2. Selection sorting.

Workers who value vesting sort into safe-harbor plans; workers indifferent to it remain in cliff or graded plans; firms sort by desired retention profile. Two stable equilibria coexist rather than one dominant one. Competitive pressure operates within each equilibrium, not across them.

3. Regulatory anchoring.

ERISA sets the legal maximum vesting periods (three-year cliff or two-to-six-year graded). Plans coordinate to the permitted floor within those maxes — not because the cap is optimal but because plan-design consultants (Aon Hewitt, Mercer, Willis Towers Watson) recommend structures near industry norms. The cap functions as an anchor rather than a ceiling.

4. Historical lock-in.

Plan design decisions live at the intersection of HR, legal, and finance functions. Changing a plan requires coordinating all three simultaneously against a status quo that "works." No single actor has the incentive to bear the coordination cost. Some cliff and long-graded structures persist not because anyone actively benefits from them anymore but because everyone inherited the rule.

Why this matters for the reform proposals that follow: the equilibrium diagnosis makes the reforms targeted rather than a menu. Disclosure (lever #5) targets the information-asymmetry equilibrium — give workers the data to price the difference and competitive pressure does the rest. Mandates (levers #1 and #2) target the regulatory-anchoring and historical-lock-in equilibria — change the floor and plans move without needing consultant permission. Different reforms fit different equilibrium mechanisms. That is what separates the case for reform here from "here is a menu of reforms; pick one."

Congress has already moved on vesting mechanics — twice, with bipartisan support in both cases. The SECURE Act (2019) created the "long-term part-time worker" concept in ERISA, requiring plans to include workers who complete three consecutive years of at least 500 hours of service. The SECURE Act 2.0 (2022) shortened that threshold to two consecutive years, effective plan years beginning after December 31, 2024, and — most relevant to this investigation — established a 500-hour-per-year vesting service credit for those workers, a departure from the standard 1,000-hour year-of-service rule. The reforms did not touch the underlying cliff-versus-graded vesting structure for standard employees; they carved out a narrow protected class (long-term part-timers) and left the general rule intact. What the SECURE Act sequence establishes is not a specific reform template but the political fact that vesting-rule intervention is neither novel nor partisan. Congress has done it. The seven levers below extend the same trajectory to the full-time workforce that bears the bulk of the forfeiture exposure the essay documents.

Seven policy levers, in rough order of political feasibility:

The intellectual move that lands: the cliff and long-graded vesting structures have legitimate operational uses (retention). What they do not justify is the binary capture function or the systematic information asymmetry between new hires and veterans. Reform should preserve the retention value while pricing the worker's labor more honestly. If a firm needs the retention incentive, it can still have it via shorter graded vesting; if it cannot retain workers any other way and needs the cliff capture as a feature, that is information about its retention strategy, not a reason to extract uncompensated labor from workers who leave.

§9 · Methodology + sources

Sources are tiered by verifiability. Tier 1 is regulatory filings and government data. Tier 2 is industry research with attributed methodology. Tier 3 is commentary and aggregator context. All editorial constants used in the essay are documented on the methodology page.