How to Analyze Executive Compensation vs. Delivery

When you analyze executive compensation vs. delivery for U.S. public companies, the evidence-backed verdict is this: pay and performance are directionally correlated over multi-year horizons, but the magnitude and timing of that correlation are frequently distorted by measurement choices, one-time events, and disclosure gaps that make annual comparisons unreliable. The SEC's Pay Versus Performance (PVP) rule, which mandates Compensation Actually Paid (CAP) disclosure, provides the most accessible public-records basis for this analysis, and Pay Governance finds CAP correlates meaningfully with total shareholder return (TSR) percentiles. The Harvard Law School Forum on Corporate Governance reinforces this, documenting practitioner methods that demonstrate alignment while flagging where PVP's annual framing can mislead.

Two conditions determine whether a pay-for-performance reading is defensible:

• When alignment is reliable: Multi-year, equity-linked compensation tracked against three-to-five-year TSR, CEO-tenure aggregates, or Realizable Pay (RP) constructs that strip out prior-grant noise. These windows reduce the distortion introduced by in-flight award valuations and market timing.
• When alignment is misleading: Single-year CAP comparisons that embed prior-grant effects, large severance or change-in-control payments, and one-time restructuring charges that inflate or deflate the CAP figure without reflecting the CEO's current-period delivery.

Key Takeaways

Multi-year, equity-linked pay tracked against peer-relative TSR is the most defensible basis for assessing executive compensation alignment with shareholder delivery; annual CAP comparisons alone are insufficient and frequently misleading.

Table of Contents

• What the empirical evidence shows about pay-for-performance
• Core metrics every analyst must understand
• How design choices change what pay-for-performance analysis concludes
• What SEC PVP disclosure provides and where it falls short
• When ESG metrics in pay plans signal delivery and when they obscure it
• A reproducible analyst checklist for testing pay vs. delivery
• Common red flags and the major criticisms of pay-for-performance analysis
• How Lacunaindex's execution-score approach augments pay-for-performance testing
• What boards and investors should do with this analysis
• Statistical and practical limitations analysts must disclose
• Final verdict and a prioritized action list for this quarter
• Lacunaindex helps analysts scale pay-vs-delivery analysis
• Why the forensic, multi-evidence approach is the only defensible standard
• Sources

What the empirical evidence shows about pay-for-performance

The governance literature has converged on a qualified consensus: pay and performance are correlated in direction, but the strength and stability of that correlation depend heavily on which pay metric is used and over what window. Pay Governance's analysis of S&P companies finds that CAP tracks TSR percentile ranks more closely than the Summary Compensation Table (SCT) does, because CAP marks equity to market rather than recording grant-date fair value. That is a genuine improvement. The problem is that CAP still carries the valuation of awards granted in prior years, so a CEO who received a large option grant three years ago will show elevated CAP in a bull market even if current-year operational delivery is mediocre.

The Harvard Law School Forum on Corporate Governance documents how practitioners are using PVP disclosures in 2026-era analyses, noting that the rule made pay-for-performance comparisons more accessible to investors but did not resolve the peer-group selection and one-time-event treatment questions that drive the most contested conclusions. Analysts who rely solely on the mandated PVP table without reading the CD&A footnotes and Outstanding Equity table will miss the structural factors that explain CAP's year-to-year swings.

The CEO Pay-vs-Delivery Scorecard, a reproducible audit of S&P-100 CEOs, places CAP alongside board payout targets and peer-relative TSR, and records documented gaps without issuing value judgments. Its cases show that headline pay and delivered shareholder return can diverge substantially even when a company's absolute TSR is positive, because the relevant benchmark is peer-relative performance, not absolute gain.

The contested zone in the literature is whether RP, which aggregates unvested equity at current prices over a defined window, produces a more stable and interpretable alignment signal than CAP. Pay Governance's VP182 Viewpoint documents specific distortions in CAP related to prior-grant effects and recommends RP as a complementary method, particularly for companies with complex equity structures or significant option overhang.

Core metrics every analyst must understand

Precise definitions matter here because the three primary pay metrics — SCT, CAP, and RP — measure fundamentally different things, and conflating them produces false alignment conclusions.

Summary Compensation Table (SCT) records the grant-date fair value of equity awards in the year they are granted, regardless of whether those awards vest or what they are worth at vesting. SCT is the longest-running, most comparable series across companies and years, but it systematically overstates pay in bear markets (grants that expire worthless still appear at full grant-date value) and understates it in bull markets (grants that appreciate far beyond their grant-date value are not adjusted upward).

Compensation Actually Paid (CAP), mandated under SEC Item 402(v), replaces grant-date accounting with a calendar-year mark-to-market of all outstanding equity. The SEC's own explanatory language is explicit: CAP is a point-in-time accounting construct intended for directional comparison with TSR, and it does not equal realized cash pay. CAP includes the change in value of unvested PSUs, RSUs, and options from the prior year-end to the current year-end, using Monte Carlo or expected-value models for market-condition awards. That modeling sensitivity adds volatility that has nothing to do with the CEO's operational decisions.

Realizable Pay (RP) is a practitioner construct, not an SEC-mandated figure. It aggregates the current value of all equity granted during a defined window (typically three to five years) at the current stock price, plus cash actually received. Because RP is not standardized, analysts must document their own construction rules, but it tends to produce a more stable alignment signal than annual CAP because it averages out the prior-grant noise over the measurement window.

Pro Tip: Never read CAP as the amount the CEO took home in cash. A CEO whose unvested PSUs appreciated by $20 million in a strong market year will show elevated CAP even if no equity vested and no cash was paid. Always cross-reference the Outstanding Equity table and the vesting schedule in the proxy footnotes before drawing a realized-pay conclusion.

How design choices change what pay-for-performance analysis concludes

The single largest source of disagreement in pay-for-performance research is not the data — it is the methodological choices analysts make before they run a single calculation. Four design dimensions move conclusions more than any other factor.

Measurement horizon is the most consequential. Annual CAP comparisons capture market noise more than executive decisions. A three-to-five-year RP window, or a full-tenure aggregate that covers the CEO's entire service period, filters out the cyclical distortions that dominate single-year readings. Pay Governance practitioners caution that PVP's annual CAP focus risks a short-term view, and that RP's multi-year aggregation better matches the time horizon over which executive incentives are designed to operate.

Attribution of one-time events is the second major design choice. M&A transactions, divestitures, and large restructuring charges can inflate or deflate both the pay figure (through accelerated vesting or severance) and the performance figure (through one-time gains or impairments). The defensible default is to flag these events explicitly, run the analysis with and without them, and disclose both results. Analysts who absorb one-time events into the baseline without flagging them produce conclusions that cannot be replicated or challenged.

Peer group selection determines what "alignment" means in relative terms. Index-based peers (S&P 500 constituents in the same GICS sector) are replicable and transparent but may not reflect the actual labor market for a given CEO. Custom peers, which companies construct themselves and disclose in the CD&A, are more economically meaningful but introduce selection bias — companies tend to choose peers that make their pay look competitive. The recommended default for investor analysis is to use the company's own disclosed peer group as the primary benchmark, then run a sensitivity check against an index-based peer group to test whether the alignment conclusion is robust to peer selection.

• Use the company's disclosed CD&A peer group as the primary benchmark.
• Run a secondary check against the GICS-sector index peer group.
• Document any peer group changes year-over-year, which can signal retroactive optimization.
• For equity-heavy sectors (technology, financials), apply sector-aware CAP-to-reported-pay ratio baselines, since practitioner analyses show tech-sector CAP-to-reported-pay ratios significantly exceed 1.0 in strong market years while energy sector ratios can fall below 1.0.
• Exclude peers that were acquired or went private during the measurement window, and document the exclusion.

Sample selection and survivorship complete the design framework. Analyses that include only companies with complete multi-year data will systematically exclude firms that underwent CEO transitions, went private, or were acquired, which biases the sample toward stable, larger-cap companies with generally stronger governance. Analysts should document their inclusion and exclusion criteria before running screens, not after.

What SEC PVP disclosure provides and where it falls short

The SEC's Pay Versus Performance rule, effective for fiscal years ending on or after December 15, 2022, requires companies to disclose CAP alongside TSR, peer-group TSR, net income, and a company-selected financial performance measure in the DEF 14A. This was a material improvement over relying solely on SCT, because it forces a market-adjusted pay figure into the public record and creates a standardized comparison point across companies.

The practical limits, however, are significant. The SEC's own Item 402(v) guidance specifies that CAP uses point-in-time valuations for unvested equity and Monte Carlo or expected-value models for market-condition PSUs, which introduces modeling sensitivity that varies by company and auditor. Two companies with identical equity structures can report materially different CAP figures depending on their option-pricing assumptions. That is not a disclosure failure — it is an inherent feature of marking complex equity instruments to market — but it means analysts cannot treat CAP as a clean, apples-to-apples comparison without reading the footnotes.

The CD&A (Compensation Discussion and Analysis) section of the proxy contains the qualitative context that the PVP table cannot provide: the rationale for peer group selection, the weighting of financial versus non-financial metrics, the treatment of one-time items, and the board's own narrative on how pay was calibrated to performance. Analysts who skip the CD&A and read only the PVP table will miss the structural factors that explain the numbers. The Outstanding Equity table and the footnotes to the SCT are the two additional locations where analysts can reconstruct a more accurate RP figure by identifying grant dates, vesting schedules, and the current-price value of unvested awards.

Hire and termination events create the most acute distortions in CAP. A CEO who joins mid-year with a large sign-on grant, or who departs with an accelerated vesting provision, will produce a CAP figure that reflects those structural payments rather than the company's operational delivery. These events must be identified and treated separately in any alignment analysis.

When ESG metrics in pay plans signal delivery and when they obscure it

The adoption of non-financial metrics in executive incentive plans has grown substantially across U.S. public companies, but the ECGI working paper on ESG-linked executive compensation finds the evidence on whether these metrics drive financial delivery is mixed. The critical analytical distinction is between ESG metrics that are operationally anchored — tied to measurable unit economics, safety records, or customer outcomes — and those that are aspirational, defined loosely enough that target achievement is nearly guaranteed regardless of underlying performance.

Red flags that suggest ESG metrics are symbolic rather than operational:

• No disclosed baseline or prior-year comparison for the metric.
• Target set at or below recent historical performance, making achievement near-certain.
• Metric defined in terms of process completion ("publish a report," "conduct a review") rather than outcome ("reduce Scope 1 emissions by X% per unit of revenue").
• Weighting below 10% of total incentive opportunity, which limits financial consequence even if the metric is missed.
• Metric that cannot be independently verified from public disclosures or third-party audits.
• Sudden introduction of a new ESG metric in a year when financial performance metrics would have produced below-target payouts, without a multi-year commitment to the metric.

A poor metric: "Advance diversity and inclusion initiatives," with no baseline, no target, and no measurement methodology disclosed. The former belongs in an alignment score; the latter should be flagged as a disclosure opacity signal. For a deeper evaluation of tools that assess ESG-linked corporate metrics, the methodology considerations extend well beyond the incentive plan itself.

A reproducible analyst checklist for testing pay vs. delivery

A defensible pay-for-performance analysis follows a structured sequence. The steps below constitute a replicable workflow that produces results analysts can document, challenge, and update as new filings become available.

1. Collect the primary data fields. Pull the DEF 14A PVP table (CAP, SCT, TSR, peer TSR, net income, company-selected measure) for each year of the CEO's tenure. Collect the Outstanding Equity table and the SCT footnotes for grant dates, vesting schedules, and in-flight PSU assumptions. Automated parsing of DEF 14A PVP tables via XBRL-enabled APIs can return these fields programmatically, which reduces transcription error and enables multi-company screens.

2. Identify and flag structural events. Before any calculation, mark years containing CEO transitions, M&A transactions, divestitures, large restructurings, or sign-on/severance payments. These events require separate treatment and should not be absorbed into the baseline without disclosure.

3. Construct the multi-year pay series. Aggregate CAP across the full CEO tenure to produce a tenure-level pay total. Separately, reconstruct RP by summing the current-price value of all equity granted during the tenure window, using the Outstanding Equity table and current stock price. Document the construction rules.

4. Construct the performance series. Compile TSR and peer-relative TSR for the same window. Add the company-selected financial measure from the PVP table and, where available, the board's disclosed payout-vs-target percentage from the CD&A.

5. Apply the alignment scorecard. The table below shows the calculation logic for a composite alignment score.

6. Run sensitivity checks. Rerun the scorecard using the index-based peer group instead of the company's disclosed peers. Rerun excluding flagged structural-event years. If the alignment conclusion changes materially, document the sensitivity and report both results.

7. Document filters and publish the raw data. Save the raw proxy filings, the data extraction log, and the filter decisions. Reproducibility requires that another analyst can replicate the score from the same public sources. The systematic accountability scoring framework developed for institutional investors formalizes this documentation discipline.

Pro Tip: Always record the pre-analysis baseline before applying any filters. Document the full universe of companies or years considered, then the filters applied, then the final sample. An undisclosed filter that removes inconvenient outliers is the most common source of non-reproducible pay-for-performance conclusions.

Common red flags and the major criticisms of pay-for-performance analysis

Misalignment between executive pay and shareholder delivery tends to cluster around a small set of structural patterns. Analysts who screen for these patterns first will identify the highest-priority cases for deeper investigation.

Red flags indicating potential misalignment:

• Pay percentile materially exceeds TSR percentile relative to the disclosed peer group, sustained over two or more years.
• Incentive payouts at or above target in years when peer-relative TSR is below median.
• Large severance or change-in-control payments that represent a material fraction of total tenure-level pay, effectively guaranteeing a floor on realized compensation regardless of delivery.
• CAP figures that are dominated by prior-grant effects, meaning the year-over-year change in CAP is driven primarily by appreciation of awards granted before the current performance period rather than current-period decisions.
• Peer group composition that changed between the grant year and the measurement year in a direction that lowered the performance hurdle.
• Non-financial metrics introduced at low weighting in years when financial metrics would have triggered below-target payouts.

The major academic and policy criticisms of pay-for-performance analysis are distinct from these operational red flags. SCT misuse remains the most persistent: boards and proxy advisors who use SCT as the primary pay figure in alignment analyses are comparing a grant-date accounting number to a market-return outcome, which is a category error. The short-term focus of annual comparisons is the third major criticism, and it is the one most directly addressed by multi-year RP aggregation.

Consider an anonymized archetype: a large-cap technology company whose CEO received a substantial equity grant at the start of a three-year bull market. Annual CAP figures for each of those three years showed strong positive correlation with TSR, producing an apparent alignment signal. When the analysis was rerun using RP over the full tenure, stripping out the prior-grant appreciation that predated the CEO's tenure, the alignment signal weakened considerably. The headline conclusion and the forensic conclusion pointed in opposite directions.

How Lacunaindex's execution-score approach augments pay-for-performance testing

Standard pay-for-performance analysis, even when executed rigorously, operates on a single dimension: financial and market-return metrics disclosed in the proxy. It does not capture the gap between what management communicated to investors and what the company actually delivered operationally. Lacunaindex addresses that gap through a forensic execution-score methodology that parses proxy tables, earnings call transcripts, press releases, and SEC filings into a normalized event record, then scores the degree to which stated commitments were met.

The workflow begins with public-records inputs: DEF 14A proxy statements (PVP table, SCT, CD&A), earnings call transcripts, 8-K press releases, and annual reports. These are parsed into discrete commitment events — a guidance range, a strategic target, a capital allocation pledge — each tagged with a source, a date, and a measurable outcome criterion. The execution score is computed as the ratio of commitments met to commitments made, adjusted for materiality weighting and disclosure opacity signals.

When combined with CAP-vs-TSR alignment data, the execution score adds a second axis: not just whether pay tracked market returns, but whether the CEO delivered on the specific operational and strategic commitments made to investors. A company can show strong CAP-TSR correlation in a bull market while simultaneously accumulating a large aspiration-to-execution gap in its operational disclosures. That combination — apparent pay-for-performance alignment on the financial axis, material delivery shortfall on the operational axis — is the archetype Lacunaindex classifies as "borrowed" credibility.

The scope is limited to public records. Lacunaindex does not access management guidance provided in private meetings, non-public financial projections, or internal operating data. For analysts who require that layer, the execution score is a complement to, not a substitute for, direct management engagement. Guidance on incorporating execution scores into equity research workflows covers the practical integration steps in detail.

What boards and investors should do with this analysis

Pay-for-performance analysis is only as useful as the governance actions it informs. The following recommendations are prioritized by their expected impact on alignment quality.

For compensation committees:

• Adopt multi-year RP as the primary internal alignment metric, supplementing the mandated CAP disclosure rather than treating CAP as sufficient.
• Set performance targets before the fiscal year begins and disclose the target-setting rationale in the CD&A, including the relationship between the target and the prior year's actual result.
• Limit non-financial metric weighting to categories where the causal link to unit economics is documented and the measurement methodology is independently verifiable.
• Implement clawback provisions that cover not only financial restatements but also material misrepresentation of operational delivery, consistent with the SEC's 2023 clawback rule requirements.
• Review peer group composition annually and disclose the criteria for peer inclusion and exclusion, including any changes from the prior year.

For investors and proxy voters:

• Build a pay-for-performance screen that uses multi-year peer-relative TSR as the performance benchmark, not absolute TSR, and that flags companies where pay percentile exceeds TSR percentile by more than one quartile over a three-year window.
• Use alignment screen results to prioritize engagement targets: companies with persistent misalignment warrant direct compensation committee engagement before the proxy vote.
• Vote against say-on-pay proposals at companies that show two or more consecutive years of pay-percentile-exceeds-TSR-percentile misalignment without a credible board explanation in the CD&A.
• Demand disclosure of the payout-vs-target percentage for each incentive plan metric, not just the aggregate payout. Aggregate payout figures can mask below-target financial performance offset by above-target non-financial metric achievement.

Questions to put to compensation committees directly:

• What is the three-year peer-relative TSR percentile, and how does it compare to the CEO's three-year pay percentile?
• How were performance targets set relative to the prior year's actual results, and what was the historical achievement rate for each metric?
• What portion of CAP in the most recent year reflects appreciation of awards granted before the current performance period?

Statistical and practical limitations analysts must disclose

No public-records pay-for-performance analysis is free of measurement error, and analysts who present results without disclosing their limitations undermine the reproducibility and credibility of their work.

Key limitations and mitigation approaches:

• CAP measurement error: CAP's reliance on Monte Carlo and expected-value models for PSUs introduces valuation sensitivity that varies by company and auditor. Mitigation: run sensitivity checks using a range of equity-valuation assumptions and report the range of resulting alignment scores.
• Survivorship bias: Multi-year samples that include only companies with complete data exclude CEO transitions, acquisitions, and delistings, biasing toward stable large-cap firms. Mitigation: document the full initial universe and the exclusion criteria before filtering.
• Fiscal year-end misalignment: Companies with non-December fiscal year-ends will have CAP figures that span different calendar periods than December-year-end peers, complicating cross-company comparisons. Mitigation: align all TSR calculations to the company's fiscal year-end, not the calendar year.
• Option valuation assumptions: Black-Scholes or binomial model inputs (volatility, dividend yield, risk-free rate) vary across companies and years, producing non-comparable option fair values in the SCT and CAP. Mitigation: where option grants are material, reconstruct fair values using standardized inputs and disclose the reconstruction methodology.
• Peer group noisiness: Custom peer groups change year-over-year and may be selected to minimize apparent pay-for-performance gaps. Mitigation: run parallel analyses using both the company's disclosed peers and an index-based peer group, and report both results.

Presenting uncertainty in published research notes requires stating the range of conclusions that result from different methodological choices, not just the central estimate. A conclusion that holds across multiple horizon choices, peer group specifications, and event-treatment rules is materially more defensible than one that depends on a single set of assumptions.

Final verdict and a prioritized action list for this quarter

Executive compensation is directionally aligned with shareholder delivery for most U.S. public companies when measured over multi-year horizons with equity-linked pay, but that alignment is frequently overstated in annual CAP comparisons that embed prior-grant effects, market cycles, and one-time structural payments. The SEC's PVP disclosure improved the analytical baseline, but it did not resolve the peer-group, horizon, and attribution questions that determine whether a pay-for-performance conclusion is defensible. Analysts who treat the PVP table as a complete answer rather than a starting point will systematically misread both alignment and misalignment.

Prioritized actions for analysts this quarter:

1. Pull DEF 14A PVP tables and Outstanding Equity tables for your coverage universe. Use XBRL-enabled data sources to automate extraction and reduce transcription error.
2. Construct a three-to-five-year RP series for each CEO in your coverage, using the Outstanding Equity table and current stock price. Record the construction rules before you run the numbers.
3. Build a baseline alignment scorecard using the four-dimension template above (pay-TSR direction, pay percentile vs. TSR percentile, payout vs. target, non-financial metric quality). Score each company and flag those where pay percentile exceeds TSR percentile by more than one quartile.
4. Identify structural events (M&A, CEO transitions, large severance) in the measurement window and run the scorecard with and without those years. Document both results.
5. Save all raw proxy filings, extraction logs, and filter decisions in a version-controlled repository. Publish your filter criteria alongside your conclusions so the analysis can be replicated and challenged.

Lacunaindex helps analysts scale pay-vs-delivery analysis

Forensic pay-for-performance analysis at the scale most institutional investors require — hundreds of companies, multiple years, consistent methodology — is not feasible through manual proxy review alone. Lacunaindex provides a subscription-based forensic analytics platform that systematizes the workflow described in this guide: parsing proxy statements, earnings transcripts, and SEC filings into normalized commitment records, computing execution scores and narrative scores, and classifying companies into delivery archetypes based on the gap between stated and delivered outcomes.

Sector benchmarks, which are publicly accessible at Lacunaindex's benchmark pages, provide the peer-distribution context analysts need to interpret individual company execution scores without a full subscription. For institutional users who require company-level forensic reports, audit-traceable scoring, and sector comparisons, the full platform is available by subscription. The scope is limited to public records; Lacunaindex does not access private management data or non-public financial projections, and its outputs are designed to complement, not replace, direct management engagement. To understand how execution scores integrate into equity research coverage, the User Guide provides a step-by-step walkthrough of report interpretation and scoring methodology. Institutional teams can request a demo to evaluate the platform against their specific coverage and governance workflows.

Why the forensic, multi-evidence approach is the only defensible standard

The governance industry's default framing — that pay-for-performance analysis is primarily a matter of reading the PVP table and checking whether CAP moved in the same direction as TSR — understates the analytical problem by a significant margin. Annual CAP correlation with TSR is a necessary condition for alignment, not a sufficient one. A CEO who benefits from a broad market rally, a favorable prior-grant vintage, and a carefully selected peer group can show strong apparent alignment on every mandated disclosure metric while simultaneously failing to deliver on the specific operational commitments that justified the pay package in the first place.

The multi-evidence approach — combining CAP-TSR analysis with RP aggregation, CD&A forensics, non-financial metric quality assessment, and execution-score measurement of the aspiration-to-execution gap — closes that analytical gap. It is more demanding, but it is the only method that produces conclusions robust enough to support engagement, proxy voting, and capital allocation decisions. Analysts who adopt the reproducible checklist in this guide and supplement it with execution-score data will find that the most consequential misalignment cases are rarely the ones that show up in the headline PVP table.

Sources

The sources below are the primary references for the methodology, regulatory text, and empirical evidence covered in this guide.

• Pay for Performance Mandated SEC Proxy Disclosures - Role of PVP and CAP | Pay Governance LLC
• Measuring CEO Pay-for-Performance: Demonstrating alignment with shareholder outcomes
• R2.htm
• Executive Compensation API: CEO Pay vs Performance — StockFit Developer Blog | StockFit API

Recommended

• User Guide — How to read a Lacuna report
• Lacuna Index — We read between the lines
• Leadership Accountability: Scoring Executive Presence and Insider Buying Together — Lacuna Journal