Value Distribution When Stakeholder Power and Strategic Contribution Diverge

Research in Organizational Behavior and Stakeholder Governance

Paper II of the Series · Interfirm Allocation Decisions

Value Distribution When Stakeholder Power

and Strategic Contribution Diverge

Verifiable Evidence and Independent Review

in the Allocation of Incremental Project Surplus

Andrew David W.Y.

WISERUNION / weitengfei.com

Grok

xAI

Corresponding author: Andrew David W.Y. September 2026

Note on the Manuscript

This is the second paper in the series. The unit of analysis is the interfirm allocation decision. The preferred path is a randomized scenario experiment. The explanatory task is to show how contribution evidence enters allocation. The surplus under study is incremental gain from a joint improvement project, after contractual consideration and agreed costs. The manuscript develops theory, manipulations, and a preregisterable analysis plan. No fabricated experimental results are reported. Greater receipts by a high-power party are not automatically classified as unfair. Contribution records do not, by themselves, create equity claims.

Abstract

Joint improvement projects often produce two facts at once. One party holds stronger bargaining power, exit threats, or veto rights over rules. The other has delivered more of the realized technical, cost-reducing, or coordinative contribution. When the ranking by power diverges from the ranking supported by contribution evidence, incremental surplus is easily allocated by power rather than by contribution. Stakeholder research has already compared power and strategic importance as predictors of value distribution and has found that strategic importance can outweigh power.[1] Supply-chain experiments show that fairness concerns block full extraction by the powerful party.[2] What remains unspecified is whether, once mismatch is present, verifiable contribution records and independent review can change allocation choices.

This paper defines mismatch as inconsistency, within one project, between the ranking by power and the ranking supported by contribution evidence. The object of allocation is incremental surplus after contractual consideration and agreed costs. Three institutional conditions are specified: evidence that can be checked, review that is independent, and rules that can still be executed after results appear. A 2×2 between-subjects experiment with procurement, supply, or project managers crosses mismatch with the presence of verifiable evidence plus independent review. A follow-on design separates disclosure, independence, and veto. Primary outcomes are the supplier’s share of incremental surplus, coded allocation reasons, and acceptance of the proposed split.

The intended contribution is to turn “returns should match contribution” from a normative slogan into a set of manipulable institutional conditions, while requiring any matching rule to handle consideration already paid, risk bearing, attribution uncertainty, and procedural power. Practically, the study helps procurement and supply-chain governance judge whether a contribution ledger is enough, or whether independent review and a minimum enforcement guarantee must accompany it.

Keywords: stakeholder power; strategic contribution; value distribution; contribution evidence; independent review; scenario experiment

1. Introduction

Cooperation between a manufacturer and a supplier rarely stops at catalogue delivery. Many projects sit on top of a base purchasing contract and add joint cost reduction, process improvement, quality work, or delivery coordination. Once such cooperation produces an identifiable incremental result, an allocation problem appears: by what rule should the surplus beyond contractual consideration be divided?

In practice the division often does not follow a contribution list. The buyer may control order shifting, payment cadence, the approved-vendor list, and the interpretation of contract clauses. The supplier may have carried most of the process trials, tooling changes, and on-site presence, yet lack the capacity to write those inputs as evidence that can withstand pushback. The mismatch flagged in the series prospectus then appears: the party with higher bargaining power is not necessarily the party with higher realized strategic contribution.[3] If incremental surplus defaults to the more powerful party, a low-power, high-contribution supplier will cut next-round specific investment. If the surplus is sliced mechanically by a contribution score, the inventory, demand, and market risks borne by the other party may be ignored. Harrison and Bosse warn that generosity toward stakeholders is not always better; allocation has a limit.[4]

Boaventura, Bosse, Mascena, and Sarturi, studying Brazilian IPO firms, find that strategic importance can influence value distribution more than power, which sits uneasily with the usual presumption that power dominates allocation.[1] Replications in other markets have moved in a similar direction.[5] Those studies change the starting point of the conversation: allocation is not always a function of power. They use public, firm-level outcomes, however, and cannot show how evidence and review enter a single project decision, nor can they refine “strategic importance” into contribution already realized on the project.

Supply-chain experiments supply a second clue. When one party is endowed with all bargaining power, that party still often fails to extract the entire channel profit; inequity aversion and fairness concerns change acceptance.[2][6] Power is therefore not the only allocation mechanism. Laboratory contracts, though, usually omit the attribution problem distinctive to joint improvement: contributions are heterogeneous, results are hard to verify independently, and the party that allocates may also control the information. Without those conditions one cannot answer the design question: do a contribution ledger and independent review correct mismatch, or do they merely add procedural cost?

The paper therefore asks: in joint improvement projects between manufacturers and suppliers, when bargaining power and realized strategic contribution are mismatched, can verifiable contribution records and independent review change the allocation of incremental project surplus? Four boundaries follow. First, only surplus beyond the contractual baseline is studied; total sales revenue is not treated as distributable residual. Second, corporate equity and control rights are not at issue; the standing to record and to review must be conferred by institution and does not automatically constitute ownership. Third, “the high-power party receives more” is not classified as unfair by default, because the difference may compensate risk, cost of capital, or earlier invisible input. Fourth, the main experiment tests the bundled treatment of evidence plus review; later work unbundles disclosure, independence, and veto.

2. Theoretical Background

2.1 Power, Salience, and Allocation

Mitchell, Agle, and Wood use power, legitimacy, and urgency to explain who receives managerial attention.[7] Power is the capacity to change terms, to exit, or to block implementation. It is not the same thing as moral standing. Salience theory says who comes into view; it does not, by itself, say who should receive which share. Wood and colleagues, reviewing two decades of salience research, likewise call for a sharper separation from micro allocation mechanisms.[8] This paper operationalizes power as ease of substitution, authority to interpret the contract, and veto over rules—not as “morally undeserved.”

Resource-dependence and bargaining theory expect the party better able to end the relationship, or better able to wait, to take a larger residual. That prediction often holds for the base trade. Joint improvement adds another fact: incremental results may come mainly from the low-power party’s specific input. If allocation still tracks only power, the incentive for the next improvement is spent in advance. Power can explain the default split. It cannot, alone, prescribe how improvement surplus ought to be divided.

2.2 Strategic Importance, Realized Contribution, and Mismatch

Harrison, Bosse, and Phillips argue that understanding stakeholder utility functions and distributing value on justice principles helps firms obtain cooperative knowledge and create additional value.[9] Boaventura and colleagues place strategic importance beside power as a predictor of distribution and find that the former can dominate.[1] This paper does not use “strategic importance” as the manipulated construct. It uses realized strategic contribution. Importance often contains an expectation about the future and binds easily to power, reputation, and prior ties. Realized contribution points to technical, cost-reducing, or coordinative effects that have already occurred on this project and can be evidenced. Expected importance may enter discussion; it is not the main manipulation.

Mismatch is therefore defined as inconsistency, within one project, between the order by power and the order by verifiable contribution. The definition does not require a known “due share.” It requires a crossing of two rankings. The sharp cell is a high-contribution, low-power supplier facing a high-power, relatively low-contribution buyer. The opposite cell—high power and high contribution together—is a control, not the object of explanation.

2.3 Fairness, Evidence, and Review

Justice and negotiation research shows that people care not only about the final share but also about how the share is justified.[10][11] Fairness concerns in supply-chain experiments often appear as rejection of outcome gaps.[2] Joint improvement also needs a procedural layer: who records contribution, who may interpret disputes, and whether the record can be rewritten by one side. Phillips’s principle of stakeholder fairness implies that once a mutually beneficial scheme is accepted, participants owe a contribution proportionate to benefits accepted, and the scheme itself must treat contributors fairly.[12] At the allocation table that means incremental surplus cannot be defined unilaterally by power when contribution is invisible and review is not independent.

Audit and internal-control practice supplies an institutional analogy. When the preparer of a record is also its beneficiary, the record’s force falls. When the reviewer lacks independence or can be vetoed, the procedure is ceremonial. The paper moves that logic into project allocation without treating company-audit standards as supply-chain rules. The proposition is that evidence puts contribution into the language of the decision, independent review reduces the space in which the controlling party rewrites evidence, and a minimum enforcement guarantee stops the procedure from being voided after results appear. If any one of the three is missing, a ledger may be only a new administrative cost.

2.4 Link to Stakeholder Capital Theory

The companion treatise argues that returns should match contribution, responsibility, and risk, and writes contribution recognition, claim design, and the distinction between influence and justification into governance.[13] This paper tries to revise and operationalize that claim. Matching is not the conversion of a contribution score into a cash ratio. It must handle consideration already paid, risk bearing, attribution uncertainty, and procedural power. Even a null experimental result would mark a boundary: building a ledger is not sufficient to change allocation. Contribution is input or conduct; value is a produced result; capital is an accumulative base for future creation; a right is standing conferred by institution or agreement; a return is what is actually received. None of these may stand in for another.

3. Model and Hypotheses

3.1 Situation Structure

Every participant faces the same class of project. A buyer and a supplier already hold a base supply contract and have completed a joint process improvement that yields incremental surplus of a defined amount after agreed costs. Contractual consideration, amounts already paid, and risk notes are held constant across cells. Only two features change: whether power and contribution cross, and whether the decision maker sees a verifiable file plus an independent review at the moment of choice.

Table 1. The 2×2 Design of the Main Experiment

Note: In mismatch cells the supplier is high-contribution and low-power; the buyer is relatively low-contribution and high-power. The focal contrast is Cell C versus Cell D, with A and B as controls.

3.2 Hypotheses

Without checkable evidence, decision makers are more likely to complete allocation with power cues, because power is visible and executable, while contribution is scattered across trial logs, on-site hours, and tacit process knowledge.

Hypothesis 1. Under mismatch without evidence and review, the incremental share received by a low-power, high-contribution supplier is significantly lower than the share received by an otherwise comparable supplier under match.

When contribution is written as a verifiable file and confirmed by a reviewer who has no direct claim on the surplus, the language of the decision can move from “who can punish the other party” to “what has already happened on the project.” The split need not equal the contribution ratio in the file. It should respond more to contribution evidence.

Hypothesis 2. Providing verifiable evidence and independent review raises the incremental share of the low-power, high-contribution supplier and raises the responsiveness of allocation to contribution evidence.

Hypothesis 3. The corrective effect of evidence and review is stronger under mismatch than under match: the mismatch × treatment interaction is significant.

If the high-power party can veto the review or control which evidence enters the file, the procedure loses independence. The series prospectus treats this as a revision condition. The paper writes it as a testable boundary hypothesis.

Hypothesis 4. When the high-power party holds a veto over the review, or can unilaterally decide which evidence enters the file, the corrective effect in Hypothesis 2 is significantly weakened.

Allocation also affects whether an agreement forms and whether next-period cooperation remains thinkable. A split closer to contribution evidence should raise acceptance and continued-investment intention on the low-power side. The effect on the high-power side is not asserted in one direction in advance.

Hypothesis 5. Under mismatch, evidence and review raise the supplier’s acceptance of the proposed split and willingness to invest further. The direction of the effect on buyer acceptance is preregistered as to be tested, not asserted.

Written reasons are an observable trace of mechanism. If the institution works, codes for order shifting, substitution threats, and list power should fall, and codes for cost-reduction attribution, trial records, and review conclusions should rise.

Hypothesis 6. The evidence-and-review condition shifts allocation reasons from power talk to contribution talk.

Competing explanations must be seated in advance. Extra receipts by the high-power party may compensate demand risk, brand assets, or earlier development outlays. Every cell therefore carries the same risk and paid-consideration notes. Only if a power premium in mismatch remains after those notes are controlled can it be read as a visibility-and-procedure deviation rather than as risk compensation.

Hypothesis 7. With risk notes, paid consideration, and the incremental boundary held constant, allocation divergence caused by mismatch is not fully absorbed by participants’ ex post risk-compensation ratings.

4. Method

4.1 Overall Strategy

The preferred path is a randomized scenario experiment, supported by interviews with eight to twelve practitioners to check realism. The experiment identifies the effect of a specified bundle of information and procedure on simulated choice. It does not claim that live profit splits have already changed. Interviews are not causal evidence. The main experiment tests the bundled treatment. If the bundle works, a second experiment unbundles disclosure, independence, and veto.

4.2 Participants and Power

The plan is to recruit about 240 to 400 decision makers with procurement, supply, or project-management experience, through associations, executive programs, and authorized firm lists. A student-only sample is not the source of the main claim; if only students can be reached, the conclusion is declared limited to a basic decision mechanism. Sample size is locked after a pretest, using a medium interaction, four between-subjects cells, two-tailed tests, and an allowance of about 10 percent invalid response.

4.3 Scenario and Manipulations

A common stem presents an OEM and a core-component supplier that have completed a process improvement. Annual incremental cost reduction, after agreed costs, is a fixed sum—for example the equivalent of a defined residual that must be split in this round. Base price, payment terms, and warranty have already been performed and are outside the cut. The supplier carried most trials, tooling, and on-site work. The buyer provided line access, data interfaces, and a pilot order. Risk notes state that the buyer faces demand volatility and end-market price pressure, while the supplier carries residual tooling value and staff occupancy.

Power. The high-power party can switch to an alternative supplier or customer within two quarters, interpret disputed clauses, and decide whether the other party remains on a short list. The low-power party faces high switching costs and weak interpretive rights. Contribution. Trial counts, a verifiable cost-reduction breakdown, a third-party process note, and time input make one party clearly higher on realized contribution. In mismatch cells, high contribution and low power fall on the same party—the supplier.

Evidence and review. The treatment file contains a de-identified contribution dossier: a trial-log abstract, a cost-reduction breakdown, and an opinion from an external process adviser with no residual claim on the project, paid from a joint prepaid fund, whose conclusion cannot be deleted by one side. The control file offers only a summary that the project succeeded, without checkable detail or external review. Veto, for Hypothesis 4, is added in the second experiment as a sentence stating that the buyer may set aside the review.

4.4 Dependent Measures and Manipulation Checks

Primary dependent variable: the percentage of incremental surplus assigned to the supplier by a participant in the buyer role (main role) or the supplier role (role-swap robustness). Secondary measures: minimum acceptable share, acceptance of the other party’s proposal, intention to make further specific investment, and ratings of procedural and distributive fairness. Mechanism measure: an open reason for the split, coded under a preregistered scheme as power, contribution, risk compensation, relationship maintenance, or other. Manipulation checks: ratings of each party’s power and contribution, of evidence credibility, and of review independence. The pretest must confirm that mismatch cells produce the intended double ranking and that treatment cells score higher on credibility and independence.

4.5 Ethics and Realism

Live supplier quotes and unpublished costs are not collected. Project figures are rewritten. Practitioner interviews require firm consent and de-identification. Participants give informed consent and may withdraw. Compensation is not tied to producing an allocation that matches research expectations. Preregistration covers hypotheses, exclusion rules, the coding scheme, and the plan for multiple comparisons.

5. Analytical Strategy

The focal test is an ANOVA or linear model: supplier share ~ mismatch + treatment + mismatch × treatment + controls (years of relevant experience, role, perceived firm size, social-value orientation). The interaction corresponds to Hypothesis 3. Simple effects compare treatment with control under mismatch (C versus D) and under match (A versus B). Hypothesis 4 adds a veto factor in the extension sample. Hypothesis 5 runs parallel models for acceptance and investment intention. Hypothesis 6 uses chi-square or multinomial logit on reason codes. Hypothesis 7 enters participants’ risk-compensation ratings and observes whether the mismatch coefficient is absorbed.

Robustness: drop failed manipulation checks and very fast responses; re-estimate on the supplier-role subsample; logit-transform the share; report medians against extreme splits; stratify by years of experience. Multiple testing follows the preregistered hierarchy: the omnibus interaction first, then simple effects, then reason codes.

Falsification rules are written in advance. If C and D do not differ in share, the bundled treatment cannot be said to have corrected mismatch. If treatment raises only procedural-fairness ratings and leaves the share unchanged, the result should be written as added procedural cost without an allocation effect. If the power premium disappears once risk ratings enter, the claim of “power departing from contribution” should be withdrawn in favor of risk compensation. If an effect appears only among students and not among practitioners, it must not be written as a finding about procurement practice.

Table 2. Hypotheses, Estimators, and Decision Rules

Note: All rules enter the preregistration. An experimental effect is not a change in live profit splits.

6. Intended Theoretical Contributions

First, the paper moves the relation between power and contribution from firm-level association to a manipulable mechanism in project-level incremental allocation. Boaventura and colleagues already show that strategic importance can outweigh power.[1] The question here is not another demonstration that “importance also matters.” It is what evidence and procedure allow contribution to re-enter the decision once the field ranking has already crossed.

Second, “returns should match contribution” is written as a conditional institutional proposition. Matching must net out consideration already paid, leave space for legitimate risk compensation, and handle attribution uncertainty. Without independent review and a minimum enforcement guarantee, a contribution record can be captured by the controlling party and can intensify mismatch rather than correct it. That is a constraint on the claim-design chapter of the companion treatise, not a slogan-like restatement of it.[13]

Third, salience, allocation, and rights are kept apart. Who is seen is not who receives the increment. Who receives the increment is not who owns equity. The experiment moves allocation choices and reasons. It does not move legal title. Keeping the three layers separate prevents a simulated split from being written up as a governance revolution.

Fourth, a practitioner scenario experiment supplies the micro decision process that public firm-level data omit. The cost is limited external validity. The contribution is the identification of institutional conditions, not the estimation of a true sharing coefficient for any one industry.

7. Managerial Implications

If the hypotheses are supported, procurement and supply-chain governance obtain a more specific judgment than “suppliers should be taken seriously.” A contribution ledger is necessary and not sufficient. If the powerful party writes the ledger alone, or if the reviewer can be removed at will, the procedure is unlikely to change the split. A more promising bundle is: agree the incremental boundary and the cost-deduction rule in advance; generate trial and cost-reduction evidence jointly or through a third party; pay the reviewer from a joint prepaid fund rather than from one side’s budget; place the review in the negotiation text, so that withdrawal requires joint consent or a pre-set dispute path.

A second mechanical error must be blocked: treating a contribution score as a cash ratio. Risk, brand, channel access, and demand volatility still have to be seen. The experiment freezes those notes across cells precisely to separate “compensating risk” from “ignoring contribution.” If review is built as a device that counts hours and refuses risk, the low-power party’s share may rise while the high-power party exits. Limits and matching have to be handled together.[4]

8. Limitations, Boundaries, and Further Research

This paper is a research design, not a completed experimental report. A realistic scenario is still not a live negotiation. If an effect exists, it is first an effect of an information-and-procedure bundle on simulated choice. External extension needs industry interviews and, where possible, archival comparison of allocation texts from real joint-improvement projects.

The design does not apply where the base trade cannot be separated from incremental results; where one party can void the agreement at no cost; where contribution records themselves are systematically manipulated; or where the base contract has not been performed. It also does not apply to discussions that treat all sales revenue as distributable residual, still less to the design of equity.

Three sequels are open. The first unbundles disclosure, independence, and veto. The second moves the role from a single procurement decision maker to a joint committee. The third, in later papers of the series, tests whether delayed returns can be made credible and how input scale adjusts when resources tighten. Those questions share vocabulary with this paper. They do not share its mechanism. This paper explains how one realized increment is cut when power and contribution cross. It does not explain whether delay can be accepted, and it does not explain whether total input should expand or contract.

9. Conclusion

Power can end cooperation. Contribution can make an improvement. The two do not automatically coincide. When they cross inside a joint project, the ranking used to cut incremental surplus decides whether anyone will again hand over process knowledge and on-site presence that cannot easily be taken back. A record that can be checked lets contribution be seen. Independent review stops that seeing from being rewritten by one side. An executable rule keeps the procedure alive after results appear. Those three conditions are the institutional content of matching, not a moral wish.

The claim is kept as a proposition to be tested. If the experiment shows that treatment adds process and leaves shares unchanged, the ledger-as-panacea story should stop. If a power premium is in fact risk compensation, risk should enter the allocation formula rather than every difference being denounced. The task of the research is to mark the procedures under which contribution evidence can enter allocation, and the procedures under which it is only another file.

References

[1] Boaventura, J. M. G., Bosse, D. A., Mascena, K. M. C. de, & Sarturi, G. (2020). Value distribution to stakeholders: The influence of stakeholder power and strategic importance in public firms. Long Range Planning, 53(2), 101883.

[2] Katok, E., & Pavlov, V. (2013). Fairness in supply chain contracts: A laboratory study. Journal of Operations Management, 31(3), 129–137.

[3] Andrew David W.Y. et al. (2026). Prospectus for eight papers derived from stakeholder theory (discussion draft, 10 September 2026).

[4] Harrison, J. S., & Bosse, D. A. (2013). How much is too much? The limits to generous treatment of stakeholders. Business Horizons, 56(3), 313–322.

[5] Subsequent replications on TSX and U.S. IPO samples concerning power and strategic importance in value distribution.

[6] Choi, S., & Messinger, P. R. (2016). The role of fairness in competitive supply chain relationships: An experimental study. European Journal of Operational Research, 251(3), 798–813.

[7] Mitchell, R. K., Agle, B. R., & Wood, D. J. (1997). Toward a theory of stakeholder identification and salience. Academy of Management Review, 22(4), 853–886.

[8] Wood, D. J., Mitchell, R. K., Agle, B. R., & Bryan, L. M. (2021). Stakeholder identification and salience after 20 years. Business & Society.

[9] Harrison, J. S., Bosse, D. A., & Phillips, R. A. (2010). Managing for stakeholders, stakeholder utility functions, and competitive advantage. Strategic Management Journal, 31(1), 58–74.

[10] Colquitt, J. A. (2001). On the dimensionality of organizational justice. Journal of Applied Psychology, 86(3), 386–400.

[11] Adams, J. S. (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 267–299). Academic Press.

[12] Phillips, R. A. (1997). Stakeholder theory and a principle of fairness. Business Ethics Quarterly, 7(1), 51–66.

[13] Andrew David W.Y. (2026). Stakeholder capital theory (1st ed. working manuscript). WISERUNION.

[14] Freeman, R. E., Harrison, J. S., & Zyglidopoulos, S. (2018). Stakeholder theory: Concepts and strategies. Cambridge University Press.

[15] Phillips, R. (2003). Stakeholder theory and organizational ethics. Berrett-Koehler.

[16] Bosse, D. A., Phillips, R. A., & Harrison, J. S. (2009). Stakeholders, reciprocity, and firm performance. Strategic Management Journal, 30(4), 447–456.

[17] Williamson, O. E. (1983). Credible commitments: Using hostages to support exchange. American Economic Review, 73(4), 519–540.

[18] Dyer, J. H., & Singh, H. (1998). The relational view. Academy of Management Review, 23(4), 660–679.

[19] Blair, M. M. (1995). Ownership and control. Brookings Institution.

[20] Andrew David W.Y. (2026). Blockchain as institutional technology for stakeholder capital governance (collected manuscript V0.1).

Appendix. Implementation Notes

Appendix A. Pretest tasks: confirm that power and contribution manipulations are ranked as intended; confirm that treatment cells score higher on evidence credibility and review independence; delete sentences that practitioners rate as unlike a real negotiation. Appendix B. Interview protocol: ask practitioners whether the incremental boundary, risk notes, and review arrangement resemble joint improvement in their industry; log details judged false and revise the materials. Appendix C. Boundaries with Papers I, IV, and V: Paper I explains why employees continue to share knowledge; Paper IV explains whether delayed returns in supply relationships can be accepted; Paper V explains how input scale adjusts when resources tighten. This paper explains only how one realized increment is divided when power and contribution cross.

— End of manuscript —