Aghion–Bolton Control-Rights Model

The Aghion–Bolton Control-Rights Model is the landmark incomplete-contracts theory of corporate finance, capital structure, and venture capital governance in business and economics theory, developed by economists Philippe Aghion and Patrick Bolton in their 1992 Review of Economic Studies article "An Incomplete Contracts Approach to Financial Contracting". Extending the Grossman–Hart–Moore property rights theory and incomplete contracts theory from the boundaries of the firm to the design of financial securities, Aghion and Bolton demonstrated that corporate financial instruments—such as voting vs. non-voting common equity, convertible preferred stock, and standard debt with bankruptcy liquidation—are defined not merely by how they divide a firm's cash-flow stream, but by how they allocate residual rights of control (corporate governance and voting rights) across contingencies. In a setting where a wealth-constrained entrepreneur raises outside capital K from an investor for a project that generates both verifiable monetary returns r and non-verifiable private benefits of control l to the entrepreneur, future operating decisions a ∈ A cannot be contracted upon ex ante. Aghion and Bolton proved a pecking order of governance: when entrepreneurial control is feasible (satisfying the investor's break-even participation constraint), it implements the first-best efficient action; when entrepreneurial control violates investor individual rationality and unconditional investor control destroys valuable private benefits l, the second-best optimal financial contract is state-contingent control—implemented by standard debt with bankruptcy transfer of control, which leaves the entrepreneur in control during high-cash-flow solvent states and transfers control to the creditor upon default.

From Modigliani–Miller and Agency Theory to Control Rights

Prior to Aghion and Bolton (1992), corporate finance theory relied on two primary paradigms to explain why capital structure matters beyond the frictionless Modigliani–Miller theorem (1958):

1. Principal–Agent Cash-Flow Incentive Models: Jensen and Meckling (1976), Myers (1977), and Innes (1990) analyzed moral hazard, asset substitution, and debt overhang, focusing on how the slope of a manager's monetary compensation schedule affects hidden effort or risk-taking.
2. Asymmetric Information and Costly State Verification: Myers and Majluf (1984) analyzed adverse selection in equity issuance, while Robert Townsend (1979) and Gale and Hellwig (1985) showed in costly state verification that debt minimizes deadweight auditing costs when realized cash flows are privately observed by the borrower.

However, neither classical agency theory nor costly state verification could explain why different financial claims carry distinct corporate governance and control rights. Common equity carries unconditional voting rights to elect the board of directors; venture capital preferred stock carries board representation and state-contingent control covenants; and corporate debt carries zero voting rights in normal operations, yet triggers a transfer of control to creditors the moment the firm defaults on a scheduled debt payment D.

Why Control Allocation Matters Under Incomplete Contracts

Aghion and Bolton (1992) resolved this puzzle by recognizing two structural features of entrepreneurial and corporate projects:

• Contractual Incompleteness Over Future Operating Actions: At the time a startup or corporation raises financing (Date 0), future strategic decisions a ∈ A—such as whether to expand or liquidate a division, replace the founding CEO, accept an acquisition offer, or pivot R&D strategy—are too complex and state-dependent to be specified in an ex-ante contract enforced by a court. Consequently, the initial financial contract must allocate residual control rights: the legal authority to choose action a at Date 1 after the state of nature θ is realized.
• Divergence of Objectives (Monetary Returns vs. Private Benefits of Control): While outside investors care about verifiable monetary cash flows r (dividends, interest, and capital gains), the founding entrepreneur or manager cares both about their monetary payoff and about non-verifiable private benefits of control l—including psychic satisfaction from pursuing a scientific vision, founder reputation, continuation of employment, or avoidance of liquidation. Because l is non-pecuniary and non-verifiable, the entrepreneur and the investor have misaligned preferences over ex-post actions a ∈ A, and wealth constraints prevent the entrepreneur from bribing the investor during ex-post renegotiation.

Players, Wealth Constraints, and Timeline

Consider a two-party contracting problem between a risk-neutral Entrepreneur (E) with zero initial liquid wealth (W₀ = 0) and a risk-neutral Outside Investor (I) operating in a competitive capital market where the required expected return is normalized to 0. The Entrepreneur has an indivisible investment project requiring an upfront capital outlay of K > 0 at Date 0.

The strategic interaction unfolds across four sequential stages:

1. Date 0 (Financial Contracting): E and I sign an incomplete financial contract specifying:
   • A monetary sharing rule granting the Entrepreneur a share t(y) = α y + β of verifiable revenue y (with 0 ≤ α ≤ 1 under limited liability), leaving (1 − α)y − β to the Investor; and
   • A control allocation rule mapping verifiable first-period signals s ∈ {0, 1} (or first-period verifiable cash flow y₁) into a probability π(s) ∈ [0, 1] that the Entrepreneur holds residual control rights over the project action a, with 1 − π(s) being the probability that the Investor holds control.
2. Date 1a (State of Nature and Verifiable Signal Realization): The underlying state of nature θ ∈ Θ = [0, 1] is realized and observed symmetrically by both E and I (though θ is not verifiable by a court). Simultaneously, a verifiable public signal s ∈ {0, 1} (such as first-period cash flow y₁) is realized with conditional distribution Pr(s = 1 | θ) = φ(θ), where φ′(θ) > 0 (monotone likelihood ratio property: higher states θ are more likely to generate high first-period cash flow s = 1).
3. Date 1b (Action Choice and Potential Renegotiation): The party holding residual control rights selects an operating action a ∈ A (e.g., continuation/expansion a_E vs. restructuring/liquidation a_I). Before a is implemented, the non-controlling party may offer to renegotiate the action choice, subject to the Entrepreneur's zero-wealth / limited-liability constraint.
4. Date 2 (Realization of Monetary Returns and Private Benefits): Given state θ and chosen action a, the project yields a verifiable second-period monetary return y ∈ {0, 1} with expected monetary profit r(θ, a) = E[y | θ, a], plus a non-transferable, non-verifiable private benefit l(θ, a) accruing to the Entrepreneur E.

First-Best Social Efficiency Criterion

In every realized state θ ∈ [0, 1], total ex-post social surplus equals expected verifiable monetary profit plus entrepreneurial private benefits: S(θ, a) = r(θ, a) + l(θ, a). Thus, the first-best efficient action rule a^FB(θ) maximizes total surplus in every state θ:

Because private benefits l(θ, a) cannot be transferred to the outside investor, the Investor's Individual Rationality (Break-Even Participation) Constraint requires that the investor's expected share of verifiable monetary returns cover the initial capital outlay K: E_θ[(1 − α)r(θ, a(θ))] ≥ K.

Regime 1: Unconditional Entrepreneurial Control (Voting Equity / Non-Voting Debt)

Suppose the financial contract grants 100% of residual control rights to the Entrepreneur in all states (π ≡ 1), as when the Entrepreneur retains majority voting common stock and issues non-voting equity to the outside investor.

In state θ, if the Entrepreneur receives a monetary equity share α ∈ [0, 1], the Entrepreneur initially prefers the action a_E(θ) maximizing α r(θ, a) + l(θ, a). However, if another action a^FB(θ) generates higher total surplus (r(θ, a^FB) + l(θ, a^FB) > r(θ, a_E) + l(θ, a_E)), the Investor is unconstrained by wealth and can always bribe the Entrepreneur out of the extra monetary gain r(θ, a^FB) − r(θ, a_E) to switch to the first-best action a^FB(θ).

Proposition 1 (Efficiency and Feasibility of Entrepreneurial Control): Because the Investor is not wealth-constrained in ex-post renegotiation, unconditional Entrepreneurial Control always implements the first-best efficient action a^FB(θ) in every state θ. However, when the Entrepreneur holds ex-post bargaining power, the maximum expected monetary payoff the Investor can collect under Entrepreneurial Control is Π_I^E-control = E_θ[r(θ, a_E(θ))]. Therefore, Entrepreneurial Control is feasible if and only if E_θ[r(θ, a_E(θ))] ≥ K. Whenever E_θ[r(θ, a_E(θ))] < K < E_θ[r(θ, a^FB(θ)) + l(θ, a^FB(θ))], the Entrepreneur cannot raise financing K while retaining unconditional control.

Regime 2: Unconditional Investor Control (Non-Voting Founder Shares)

Suppose instead that the financial contract grants 100% of residual control rights to the Outside Investor in all states (π ≡ 0), as when a venture capitalist controls the board of directors outright.

In state θ, the Investor holds residual authority to choose the action a_I(θ) ∈ arg max_a∈A r(θ, a) that maximizes monetary returns r(θ, a), ignoring the Entrepreneur's private benefit l(θ, a). Can the Entrepreneur bribe the Investor in ex-post renegotiation to switch from a_I(θ) to the socially efficient action a^FB(θ)? No: because the Entrepreneur has zero liquid wealth (W₀ = 0) and r(θ, a^FB(θ)) < r(θ, a_I(θ)), the project generates less verifiable monetary cash under a^FB(θ) than under a_I(θ). Thus, the wealth-constrained Entrepreneur cannot pledge enough verifiable cash to compensate the Investor for the lost monetary profit r(θ, a_I) − r(θ, a^FB). Consequently, Unconditional Investor Control maximizes the Investor's monetary recovery E_θ[r(θ, a_I(θ))], relaxing the financing constraint K, but causes ex-post inefficiency by destroying entrepreneurial private benefits l(θ, a^FB).

Comonotonicity vs. Countermonotonicity of Returns and Total Surplus

To bridge the gap between Entrepreneurial Control (which is ex-post efficient but fails to raise K) and Investor Control (which raises K but destroys social surplus in all states), Aghion and Bolton (1992) analyzed State-Contingent Control conditioned on the verifiable first-period performance signal s ∈ {0, 1} (or first-period cash flow y₁ vs. a debt threshold D).

In the canonical Comonotonic Case, both verifiable monetary profit under investor control and total social surplus increase with the underlying state θ, and the conflict of interest between the Entrepreneur and the Investor is concentrated in bad states of nature (low θ):

• Good States (High θ, high cash flow s = 1): The project is performing well; continuation/expansion (a₁) maximizes both monetary returns r(θ, a) and private benefits l(θ, a). Leaving the Entrepreneur in control when s = 1 achieves the first-best outcome without hurting monetary returns.
• Bad States (Low θ, low cash flow / default s = 0): The project is in distress; efficient restructuring or liquidation (a₂) maximizes monetary recovery r(θ, a₂) and total social surplus, whereas the Entrepreneur under E-control would cling to inefficient continuation (a₁) to preserve private benefits l(θ, a₁). Transferring control to the Investor when s = 0 enforces efficient restructuring/liquidation in low states and boosts expected monetary recovery enough to satisfy the Investor's participation constraint K.

Why Standard Debt with Bankruptcy Is the Optimal Financial Contract

Aghion–Bolton (1992) Debt Optimality Theorem: Under the comonotonicity condition where E_θ[r(θ, a_E(θ))] < K ≤ E_θ[r(θ, a_I(θ))], the unique second-best optimal financial contract is a Standard Debt Contract with Bankruptcy Transfer of Control, characterized by a face-value payment D* in Period 1 such that:
1. Solvent Region (y₁ ≥ D*, or s = 1): The Entrepreneur pays the scheduled debt service D* and retains 100% of residual control rights over Period 2 operations (a = a_E(θ)).
2. Default / Bankruptcy Region (y₁ < D*, or s = 0): The Entrepreneur defaults on D*, and residual control rights transfer automatically to the Creditor (Investor) via bankruptcy court or debt covenants, enabling the Creditor to implement the monetary-maximizing restructuring or liquidation action a = a_I(θ).

By choosing the face value D* so that E_θ[φ(θ)r(θ, a_E(θ)) + (1 − φ(θ))r(θ, a_I(θ))] = K, the standard debt contract transfers control to the Investor only in the minimum necessary measure of low-θ states required for the Investor to break even, preserving Entrepreneurial Control across all higher states.

Kaplan and Strömberg's (2003) Empirical Verification in Venture Capital

The sharpest empirical test of the Aghion–Bolton (1992) control-rights model was conducted by Steven N. Kaplan and Per Strömberg in their landmark 2003 Review of Economic Studies study "Financial Contracting Theory Meets the Real World: An Empirical Analysis of Venture Capital Contracts". Analyzing 213 venture capital investments in 119 portfolio companies across 14 VC partnerships, Kaplan and Strömberg documented three regularities that directly confirm Aghion and Bolton's predictions:

1. Separate Allocation of Cash-Flow Rights and Control Rights: Venture capital convertible preferred stock separates cash-flow shares, board voting rights, liquidation rights, and redemption rights.
2. State-Contingent Control Transfers: Control rights are allocated contingently on verifiable financial and non-financial performance milestones (such as EBIT targets, FDA clinical trial approvals, or patent grants).
3. Pecking Order of Control by Performance: If the portfolio company performs poorly (low θ / missed milestones), the venture capitalists obtain full control of the board of directors and voting majority (Investor Control). As company performance improves, the entrepreneur retains or regains more control rights. If the company performs well (high θ), the VCs relinquish their special control covenants and convert into standard common stock at an initial public offering (Entrepreneurial Control).

Corporate Bankruptcy Law: Chapter 7 vs. Chapter 11

Aghion and Bolton's framework also transformed the law-and-economics analysis of corporate bankruptcy codes (such as U.S. Chapter 7 liquidation vs. Chapter 11 reorganization, and Aghion, Hart, and Moore's 1992 debt-for-equity swap mechanism). Rather than viewing bankruptcy merely as an accounting procedure to divide a fixed pie among creditors (as in costly state verification), the Aghion–Bolton model reveals that bankruptcy is the institutional mechanism that executes the state-contingent transfer of corporate control from equityholders/managers to creditors when low cash-flow realizations reveal that current management's continuation strategy is no longer viable.

How does the Aghion–Bolton (1992) debt theory differ from Townsend's (1979) costly state verification?

In Townsend's costly state verification model (1979) and Gale and Hellwig (1985), the project's cash-flow distribution is exogenous (no ex-post operating action a is chosen), and debt is optimal because realized cash flows y are unobservable to the lender unless the lender pays a deadweight auditing cost γ in bankruptcy. In the Aghion–Bolton (1992) model, cash flows y are publicly verifiable at zero auditing cost, but contracts are incomplete over future operating actions a ∈ A. Debt is optimal in Aghion–Bolton because default (y₁ < D) serves as a governance switch that transfers residual voting and control rights from the entrepreneur to the creditor in bad states of nature.

Why can't the Entrepreneur simply bribe the Investor not to liquidate the firm under Investor Control?

When the Investor holds residual control rights, the Investor prefers the action a_I(θ) that maximizes verifiable monetary cash flow r(θ, a_I) over the socially efficient action a^FB(θ) that preserves the Entrepreneur's non-pecuniary private benefit l(θ, a^FB). Because r(θ, a^FB) < r(θ, a_I), switching to a^FB reduces the project's verifiable monetary cash flow, and because the Entrepreneur is wealth-constrained (W₀ = 0), the Entrepreneur has no outside cash with which to compensate the Investor for the monetary loss r(θ, a_I) − r(θ, a^FB).

Why does Entrepreneurial Control always achieve the first-best action when it is feasible?

Unlike the wealth-constrained Entrepreneur, the Outside Investor has deep pockets (unconstrained liquid wealth). If the Entrepreneur holds control and threatens to choose an action a_E(θ) that sacrifices a large amount of verifiable monetary profit r for a small private benefit l, the Investor can credibly offer the Entrepreneur a side payment out of the extra monetary profit r(θ, a^FB) − r(θ, a_E) > l(θ, a_E) − l(θ, a^FB), ensuring that ex-post Coasean bargaining under Entrepreneurial Control always reaches the first-best action a^FB(θ).

1. Aghion, P., & Bolton, P. (1992). "An Incomplete Contracts Approach to Financial Contracting." Review of Economic Studies, 59(3), 473–494. https://doi.org/10.2307/2297860
2. Kaplan, S. N., & Strömberg, P. (2003). "Financial Contracting Theory Meets the Real World: An Empirical Analysis of Venture Capital Contracts." Review of Economic Studies, 70(2), 281–315. https://doi.org/10.1111/1467-937X.00245
3. Hart, O., & Moore, J. (1998). "Default and Renegotiation: A Dynamic Model of Debt." Quarterly Journal of Economics, 113(1), 1–41. https://doi.org/10.1162/003355398555496
4. Dewatripont, M., & Tirole, J. (1994). "A Theory of Debt and Equity: Diversity of Securities and Manager-Shareholder Congruence." Quarterly Journal of Economics, 109(4), 1027–1054. https://doi.org/10.2307/2118355
5. Hart, O. (1995). Firms, Contracts, and Financial Structure. Oxford: Oxford University Press. https://doi.org/10.1093/0198288816.001.0001