Equity Compensation Design: RSUs, ISOs, NSOs, Vesting, and Refresh Grants

Equity compensation is one of the highest-leverage and least empirically rigorous decisions an early-to-mid-stage company makes. The instrument choice (RSU, ISO, NSO), vesting structure (cliff length, gradient shape, acceleration triggers), refresh cadence, and the interaction with cash compensation collectively determine whether equity attracts and retains the talent the program is intended to retain, or simply creates a compensation overhang that distorts decision-making at vesting milestones.

The empirical literature on equity-as-retention is substantially weaker than the rhetoric implies. Research on broad-based ownership, particularly the work aggregated by the National Center for Employee Ownership, documents modest productivity effects but mixed retention effects, and the academic literature on executive equity (Bebchuk and Fried’s “Pay Without Performance” being the canonical reference) raises serious questions about whether equity-as-incentive aligns with performance the way design intent assumes.

Data Notice: Equity-program magnitudes and vesting conventions cited reflect projections based on published compensation data and academic literature at time of writing. Specific equity grant sizes, strike prices, and vesting conventions vary widely by company stage, industry, and jurisdiction; tax treatment varies by individual circumstance and should be reviewed with qualified tax counsel before relying on planning assumptions.

What equity instruments actually do

The three most common employee-equity instruments differ on mechanics, tax treatment, and behavioral effects:

Restricted stock units (RSUs). A promise to deliver shares (or cash equivalent) on a vesting date, taxed as ordinary income at vest. RSUs have the cleanest economics for the recipient: the value at vest is observable, the tax timing is observable, and there is no exercise decision. RSUs are the dominant instrument at public and late-stage private companies.
Incentive stock options (ISOs). A right to purchase shares at a set strike price for a defined period, with tax-favorable treatment if specific holding-period requirements are met (no ordinary-income tax at exercise if AMT thresholds are not triggered, and long-term capital gains treatment on the spread above strike at qualifying sale). ISOs are limited to ~$100,000 of value vesting per year per recipient and to employees only.
Non-qualified stock options (NSOs). A right to purchase shares at a set strike price, with the spread above strike taxed as ordinary income at exercise. NSOs have no recipient-side ceiling and can be granted to contractors, advisors, and board members.

The instrument choice is partly mechanical (ISO eligibility caps, employee-vs-contractor status) and partly a tradeoff between recipient-side tax efficiency (ISO favors recipient at qualifying sale) and the cleaner observable mechanics of RSUs.

Vesting structures and the cliff debate

The dominant vesting convention in venture-funded companies is a four-year vest with a one-year cliff: 25% vests at the twelve-month anniversary, then monthly or quarterly vesting over the remaining three years. This convention has weak empirical justification and substantial inertia.

The four-year vest reflects the typical fund-investment horizon rather than evidence about retention dynamics. The one-year cliff is intended as a probationary mechanism: an employee who departs in the first year receives nothing, which protects the option-pool from rapid dilution to short-tenure hires. Empirical research on retention dynamics suggests that the cliff effect is real but modest, and that the steepest retention risk is often not at the twelve-month mark but at the four-year vest-completion mark, where the implicit lock-in evaporates.

Several alternative vesting structures have been adopted in practice, with limited published comparative evidence:

Back-weighted vesting (10/20/30/40 across four years rather than 25/25/25/25) increases the implicit retention effect at later years but reduces the early-tenure attraction value.
Front-weighted vesting (40/30/20/10) increases early-tenure attraction at the cost of later-tenure retention; rarely seen.
Five- or six-year vesting schedules at later-stage private companies and some public-company executive grants, intended to extend the retention horizon.
Performance-vested equity (PSUs at public companies, rare at private companies) gating vest on performance metrics rather than time alone.

The conservative default of four-year-with-one-year-cliff remains dominant primarily because it is the convention investors expect, not because comparative-effectiveness research supports it as optimal.

Refresh grants and the cliff problem

The structural problem with the standard four-year vest is that vesting completes at the four-year mark, at which point the marginal retention effect of the original grant has fully amortized. If the company has not granted refresh equity by year three, the four-year vest-completion creates a sharp retention risk: the implicit lock-in disappears, and the employee can depart with full vest realization while collecting a signing equity grant from a competitor.

Refresh-grant programs address this by granting smaller equity tranches at regular cadences (often annually after year two) that maintain a continuous overlap of unvested equity. A well-designed refresh program produces an “equity ladder” where unvested equity is always present, smoothing the cliff effect.

The empirical question of whether refresh grants improve retention beyond a single up-front grant of equivalent total value is largely unanswered in the published literature. Practitioner observation suggests that refresh grants are useful for high-performers (where the marginal-retention effect is concentrated) and less useful for the median employee, but rigorous causal evidence is thin.

For broader treatment of how equity integrates with cash compensation design, see compensation design evidence, which walks through the empirical literature on the five components of compensation.

The behavioral economics of equity

The research on whether equity actually motivates performance is mixed at best. Bebchuk and Fried’s “Pay Without Performance” documents that executive equity grants correlate poorly with performance once granting-decision endogeneity is controlled. Lazear’s work on performance-pay-and-productivity, while more favorable to explicit performance-pay structures, focuses on contexts where output is individually measurable and contracts are short-cycle, which describes few knowledge-work equity programs.

The plausible behavioral mechanisms for equity-as-motivation are (a) ownership-effect on long-term thinking, (b) retention-effect via implicit lock-in, and (c) selection-effect by attracting employees who value the upside profile. The retention-effect has the strongest empirical support; the selection-effect is plausible but under-studied; the ownership-effect on long-term decision quality is the weakest empirically and the most rhetorically emphasized.

The implication for design is that equity programs should be sized to produce the retention effect at intended levels and should not be expected to produce dramatic individual performance effects in roles where the line-of-sight from individual effort to share price is weak.

Acceleration and change-of-control provisions

Acceleration provisions specify what happens to unvested equity at change-of-control (acquisition) or involuntary termination. The two common structures are:

Single-trigger acceleration vests all (or a portion of) unvested equity on change-of-control alone, regardless of employment status post-transaction. Single-trigger is unusual outside senior executive contracts.
Double-trigger acceleration requires both a change-of-control and an involuntary termination (or good-reason resignation) within a defined post-close window. Double-trigger is the more common structure and better aligns with the acquirer’s interest in retaining key talent post-close.

The acceleration negotiation is a higher-stakes element of senior-level offers than is widely recognized; for senior hires being recruited from a position of incumbent equity, the acceleration provision can shift the expected value of the offer by a meaningful percentage.

When equity is the wrong instrument

Equity is structurally better suited to some compensation problems than others. The cases where equity is most useful:

Long-tenure-orientation roles where retention beyond three years is high-value
Roles where company-level outcomes are visibly affected by individual contribution
Recipient demographics where tax-deferred upside value is a meaningful component of utility

The cases where equity is structurally weaker:

Short-tenure-expectation roles or contractor relationships
Recipients whose cash-flow profile cannot accommodate exercise costs (NSOs, ISOs at private companies)
Recipients who are tax-resident in jurisdictions where equity treatment is unfavorable
Mature-company contexts where equity is effectively a cash-substitute and the upside variance has compressed to near zero

For roles that fall into the structurally-weaker categories, loading more compensation into base salary and short-cycle variable pay typically dominates loading into equity.

For closely related coverage of compensation-program design and the integration with selection-method investment, see compensation design evidence, hiring cost economics, and career ladder design.

Takeaway

Equity compensation design has stronger conventional inertia than empirical foundation. The four-year-with-one-year-cliff vesting standard, the choice of RSU vs option instruments, the refresh-grant cadence, and the acceleration structure all matter, but the published research base is thinner than the volume of practitioner advice suggests. The defensible program treats equity as primarily a retention instrument, sizes it to produce the retention effect at intended cohort boundaries, integrates refresh grants to smooth the cliff, and pairs it with cash compensation that meets the attraction binding-constraint independent of equity upside.

Sources

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology. Psychological Bulletin, 124(2), 262-274.
Sackett, P. R., & Lievens, F. (2008). Personnel selection. Annual Review of Psychology, 59, 419-450.
Bebchuk, L. A., & Fried, J. M. (2004). Pay Without Performance: The Unfulfilled Promise of Executive Compensation. Harvard University Press.
National Center for Employee Ownership. (2024). Employee ownership and corporate performance: A research review. NCEO.
Lazear, E. P. (2000). Performance pay and productivity. American Economic Review, 90(5), 1346-1361.
Internal Revenue Service. (2024). Stock options and related tax treatment overview. https://www.irs.gov/
US Securities and Exchange Commission. (2024). Compensation discussion and analysis disclosure requirements. https://www.sec.gov/