Accounting's Missing Welfare Line

Accounting records the manager-investor deal and leaves system costs outside the payoff

The paper models corporate reporting as a bilateral game, G, between managers and capital providers. Because system welfare is not a payoff to either party, the game can certify gains that degrade capital-market information, creditor protection, public finance, and ecological or social systems.

• The missing party is System C: market information infrastructure, non-contracting creditors, tax authorities, communities, and long-term ecological and social systems
• The observed failure pattern is the Hollow Win, coded (0,1,1): system degraded, manager gains, capital provider gains
• The defect is structural, not a claim that accountants or boards are unusually unethical or incompetent

ASC 606 had full procedure and zero welfare analysis

The paper opens with FASB's May 2014 vote to adopt ASU 2014-09, better known as ASC 606. The five-step revenue recognition model came with a 205-paragraph Basis for Conclusions, but the decision record never asked whether the standard would raise or lower welfare destruction.

• FASB adopted ASC 606 by a seven-to-zero vote
• The Basis for Conclusions discusses implementation cost, comparability, decision usefulness, and preparer burden
• The word welfare appears zero times in the 205 paragraphs

The bilateral game optimizes A and B before C is even visible

In G, Party A is managers and Party B is shareholders or creditors. The payoff space is Π_AB, the contracting surplus between those parties. System welfare, W, is excluded because dispersed third parties are not part of the contract.

• A1: Π_AB defines the payoff space; W is not contractible inside the A-B relationship
• A2: managers choose accounting methods using bonus compensation, debt-covenant slack, and political-cost visibility
• A5: the Win-Win-Win outcome is structurally invisible inside G

The paper names eight reporting outcomes, but Hollow Win dominates the shareholder-primacy game

The three binary dimensions are C, A, and B. Shareholder primacy optimizes the A-B subspace, so cases that look successful inside the reporting system can become Hollow Wins once C is included.

• Win-Win-Win (1,1,1): post-SOX U.S. regime with strong enforcement in the paper's taxonomy
• Hollow Win (0,1,1): pre-SOX earnings management, where managers and shareholders gain while creditors and retail investors bear undisclosed risk
• Sustainable Win-Lose (1,1,0): strong conservatism, where system and shareholders are protected but managers lose timing discretion

Bonus plans, covenants, and political costs create the welfare leak

The paper applies Watts and Zimmerman-style incentives to explain why welfare-destructive accounting choices persist. The same choices that raise bilateral surplus can also raise third-party cost.

• Bonus-plan channel: income-increasing methods can raise managerial compensation while diluting shareholders
• Debt-covenant channel: income-increasing methods preserve covenant slack while transferring credit risk to non-contracting creditors
• Political-cost channel: income-decreasing methods lower regulatory visibility while reducing public information quality

Enron and SOX repaired information asymmetry without changing the game

The paper uses Enron to show how severe the failure mode can become. Sarbanes-Oxley tightened fraud detection, audit independence, and CEO/CFO certification, but those reforms still operated mainly within the firm-capital-provider relationship.

• Enron and Arthur Andersen exposed the catastrophic potential of accounting that certifies the A-B game while system costs accumulate
• SOX addressed information asymmetry within the game through fraud detection, audit independence, and executive certification
• The paper's claim is that even GAAP-correct numbers can omit costs borne by third parties

Option non-expensing kept dilution out of earnings until the bubble made forbearance costly

The employee stock option case shows the same bilateral logic across decades. Firms resisted expensing because it would lower reported earnings and the apparent surplus shared by managers and shareholders.

• APB Opinion No. 25 dates to 1972
• FASB's 1993 exposure draft did not end resistance to expensing
• Mandatory expensing arrived in 2004 with FAS 123(R), after the tech bubble raised the political cost of continued non-expensing
• The omitted C effects were distorted capital allocation, concealed dilution for retail investors, and weaker comparability

Environmental-liability accounting lets cleanup costs wait for a probable litigation trigger

The paper treats environmental liabilities as another Hollow Win mechanism. Firms with contaminated sites can delay recognition under SFAS No. 5 until litigation becomes probable, while affected communities, future purchasers, and regulators carry the cost of the concealed obligation.

• The recognition threshold can be shifted by legal strategy, settlement timing, and jurisdictional variation
• The firm and shareholders gain from deferred recognition of cleanup costs
• System C absorbs delayed information, local environmental risk, and regulatory burden

The paper bounds accounting welfare destruction at 0.50 to 2.50 per revenue dollar before systemic risk

The β_W estimate uses U.S. accounting and assurance industry annual revenue of 80 billion as the denominator. The conservative welfare-cost channels produce a lower bound of 40 billion and an upper bound of $200 billion.

• Lower bound: 40B welfare cost divided by 80B revenue gives β_W = 0.5
• Upper bound: 200B welfare cost divided by 80B revenue gives β_W = 2.5
• With systemic-risk channels included, 500B to 2T over a decadal window raises β_W to 2.5-10
• F4 would require an independently audited estimate below 0.1 across all five channels

More disclosure alone leaves the manager objective unchanged

The Welfare-Instrument Ineffectiveness Theorem says disclosure does not change the accounting method set, M, or the manager objective function. If bonus, covenant, and political-cost incentives remain intact, m* remains intact.

• Disclosure adds information about choices already made
• It does not alter bonus compensation, covenant slack, or political-cost visibility by itself
• A disclosure-only ESG regime would falsify the theorem only if it reduced the welfare-destructive region without enforcement or objective-function changes

Field 17 adds system welfare to the governance record and changes the objective function

The paper's proposed reform is mandatory Decision Accounting Field 17, SYSTEM_WELFARE . This transforms the payoff from Π_AB to Π_AB minus δ times ΔW_S, making welfare cost part of the manager's decision calculus.

• In G', managers maximize Π_AB(m; c) - δ·ΔW_S(m; c)
• For δ > 0, welfare cost becomes payoff-relevant
• For δ = 1, the paper calls the result the welfare-complete equilibrium
• δ must exceed Π_AB(m*) divided by ΔW_S(m*) inside the welfare-destructive region Ω

The reform needs coordinated enforcement, not a single new report

The paper's final mechanism is multi-agent activation. Auditors, regulators, plaintiff bar, creditors, and tax authorities must make welfare-destructive choices costly enough that the Hollow Win no longer pays.

• 5 uses k ≥ k* simultaneous enforcement agents against the welfare-destructive accounting choice
• CSRD and Nordic stewardship models are presented as real-world proof of concept for welfare-weighted governance
• The target outcome is Win-Win-Win (1,1,1): managers, capital providers, and System C all gain