Intertemporal Conflict and Willingness to Continue Stakeholder Cooperation

Research in Organizational Behavior and Stakeholder Governance

Paper IV of the Series · Decision Makers in Buyer–Supplier Relations

Intertemporal Conflict and Willingness

to Continue Stakeholder Cooperation

The Moderating Role of Compensation-Commitment Credibility

Andrew David W.Y.

WISERUNION / weitengfei.com

Grok

xAI

Corresponding author: Andrew David W.Y. September 2026

Note on the Manuscript

This is the fourth paper in the series. The unit of analysis is the decision maker in a buyer–supplier relation. The preferred path is a factorial scenario experiment. The explanatory task is to show whether delayed returns can be accepted. The object is incremental input and delayed return agreed in advance. Overdue trade payables recast as “growing together” are out of scope. The main outcome is willingness to cooperate, not multi-year continuation or firm growth. The manuscript specifies theory, manipulations, and a preregisterable plan. No fabricated experimental results are reported.

Abstract

Joint improvement often asks a supplier to advance tooling, trials, and on-site costs and to share cost-reduction or volume gains only later. Delay raises waiting and exposure and can weaken willingness to continue. Buyers therefore offer compensation promises against present input. Williamson showed that specific investment needs credible commitment;[1] reciprocity and network studies show that long cooperation can organize knowledge sharing.[2][3] What remains is whether, when stated returns and risk notes are held comparable, the verifiability and enforceability of a promise can ease the damage of delay—and whether the supplier’s liquidity constraint swamps that easing.

Intertemporal conflict is operationalized as incremental cost borne now by the supplier, with the return due later and still uncertain. Credibility of the compensation commitment is operationalized as three observable arrangements: clear text, reserved funds or bound orders, and a pre-set dispute path. Liquidity constraint is raised from an ordinary control to a boundary that must be tested. A 2×2 experiment with about 240 to 400 experienced supply or procurement decision makers crosses delay with safeguard strength. The pretest must separate time from mere amount and from credit risk. Primary outcomes are acceptance, willingness to add specific input, and perceived credibility.

The intended contribution is to revise the claim that a promise of future return can connect present contribution to continued input: enforceability and the capacity to wait may both be required. Practically, the study helps judge when a future-sharing clause is enough and when present compensation must be arranged. A promise that cannot be kept is not a governance tool.

Keywords: intertemporal conflict; compensation commitment; credibility; continued cooperation; liquidity constraint; supply-chain experiment

1. Introduction

If co-creation occurred only in contemporaneously cleared trades, many improvements would never start. Process trials, dedicated tooling, and on-site support must be spent first; cost-reduction shares, volume preference, or long-term qualification often come later. The buyer sees system gains in the future. The supplier sees cash leaving now and an uncertain settlement. That is intertemporal conflict: one party bears incremental cost today; the other places the return in a future that can still be rewritten.

Conflict is not the same as bad faith. Even a sincere buyer leaves the supplier three judgments to make. Is the promise written clearly? Is settlement bound to funds or orders that can be checked? If a dispute arises, is there an executable path? Williamson called such devices credible commitments or hostages that support exchange.[1] Work on the Toyota supplier network shows that cross-firm knowledge cooperation depends on organizational apparatus, not on one-shot goodwill.[3] Those studies give the direction. They do not, with returns held comparable, split delay from safeguard, and they rarely write the supplier’s capacity to advance cash as a boundary that can dominate the promise.

Two errors are common in practice. One recasts overdue payables as “growing together” and wraps default in an intertemporal story. The other issues an unkeepable future share, extracts a present advance, and later cancels on the plea of market change. Harrison and Bosse warn that treatment of stakeholders has a limit; generosity that cannot be funded damages joint value.[4] The companion treatise treats credible commitment as the link between present contribution and future return.[5] This paper tests when that link holds and when it is only text.

The one-sentence question is therefore: when a manufacturer asks a supplier to bear joint-improvement costs first and to share gains later, can the verifiability and enforceability of a compensation commitment ease the damage that delay does to the supplier’s willingness to continue? Four boundaries follow. First, only incremental input and delayed return agreed in advance are studied; stretching already due payables is out. Second, the main outcome is willingness and choice, not years of actual renewal or firm growth. Third, the input must be an arrangement that can be refused; forced advances are outside the explanation. Fourth, liquidity constraint is a boundary that must be tested, not an ordinary covariate.

2. Theoretical Background

2.1 Specific Investment and Credible Commitment

Once input sinks into tooling, process, and staffing that have value mainly for one customer, exit costs rise and the other party gains room to renegotiate. Williamson argued that exchange then needs hostages, guarantees, and self-enforcing arrangements.[1] A compensation clause that says only “we will share reasonably after success” has no checkable object and no enforcement path. It is cheap talk. Credibility therefore cannot stop at an attitude score. It has to map onto three visible designs: whether conditions are written, whether funds or orders are reserved, and whether disputes have an independent or pre-set path.

2.2 Reciprocity, Relationships, and Delayed Return

Bosse, Phillips, and Harrison use fairness and reciprocity to explain stakeholder cooperation.[2] The relational view holds that intertemporal routines can produce joint gains.[6] Dyer and Nobeoka’s account of the Toyota network shows that knowledge sharing needs organizational rules, not only personal trust.[3] These studies support the direction that future return can sustain cooperation. They also mix expected reciprocity with actual capacity to pay. A supplier may believe the buyer intends to settle and still know that the firm cannot last until settlement day. Perceived goodwill, perceived solvency, and own liquidity must be measured apart.

2.3 Time, Amount, and Credit Risk

Delay bundles three meanings that are easy to confound: the same money arrives later; the later money is nominally larger or smaller; the later money may never arrive. If a high-delay cell is written as “an obscure share in eighteen months,” participants react to amount plus default, not to time. The pretest must keep present-value notes comparable before risk adjustment and must state which uncertainty the safeguard changes. Otherwise the interaction cannot be read.

2.4 The Companion Treatise and Neighboring Papers

Section 8.2 of the companion treatise discusses how institutional promises connect input to future return; section 10.3 discusses credible commitment and cooperative expectation; section 17.4 discusses the capacity to settle; chapter 22 discusses cross-firm collaboration.[5] This paper aims to support or revise the claim that future return can organize present contribution: enforceability and the capacity to wait may both be required. Unlike Paper II, which divides an increment already realized, this paper treats an increment not yet realized that must be financed first. Unlike Paper I, which asks why employees keep contributing after current pay, this paper asks why a supplier still invests when the return has not yet arrived.

3. Model and Hypotheses

3.1 The 2×2 Design

Every participant faces the same class of project. A small or mid-sized supplier advances a fixed incremental cost for a joint process improvement with an OEM. The base supply contract is being performed; the arrangement does not stretch overdue invoices. Only two features change: how far settlement sits in time, and how strongly the compensation promise is safeguarded. Stated returns before risk adjustment are kept comparable across cells. Liquidity and exit options are measured before the scenario and used as boundary variables.

Table 1. Design of the Main Experiment

Note: High safeguard combines a clear settlement rule, reserved funds or bound orders that can be checked, and a pre-set dispute path. Low safeguard is a principle of “reasonable sharing after success.” The focal contrast is C versus D under long delay, plus the delay × safeguard interaction.

3.2 Hypotheses

Hypothesis 1. Holding safeguard constant, longer delay lowers the probability that the supplier accepts the arrangement and the willingness to add specific input.

Hypothesis 2. Holding delay constant, a stronger safeguard raises perceived credibility, acceptance, and willingness to add input.

Hypothesis 3. The positive effect of the safeguard on cooperation willingness is stronger under long delay than under short delay. The delay × safeguard interaction is significant, and what it eases is the drop caused by delay, not an equal lift in every cell.

Hypothesis 4. Perceived credibility mediates the effect of the safeguard on cooperation willingness. If that mediation disappears after perceived return amount is controlled, the safeguard has been read only as “more money,” and the mechanism should be rewritten.

Hypothesis 5. The more strained the supplier’s self-rated liquidity, the weaker the safeguard’s easing of delay damage. In the high-constraint group the interaction may be null. Liquidity is a boundary, not an ordinary control.

Hypothesis 6. When exit options are few, acceptance may rise without that rise counting as trust. In the no-exit subsample the effect of the safeguard on added-input willingness should be reported separately, so that lock-in is not counted as a commitment effect.

Hypothesis 7. If the safeguard interaction disappears after a rating of the buyer’s capacity to pay enters the model, the claim that procedural safeguards work independently should be revised; the safeguard may largely be a visible proxy for solvency.

4. Method

4.1 Strategy and Sample

The preferred path is a randomized scenario experiment. About 240 to 400 decision makers with experience in supplier pricing, customer projects, or procurement project management are to be recruited. A student-only sample is not the source of the main claim. Power analysis uses a medium interaction, four between-subjects cells, two-tailed tests, and about 10 percent invalid response; headcount is locked after the pretest. Interviews with eight to twelve practitioners check whether the materials resemble live joint improvement. They are not causal evidence.

4.2 Scenario and Manipulations

A common stem presents a small component supplier advancing a fixed sum for tooling, trials, and on-site work on a joint improvement with a continuing OEM customer. Base invoices are paid on contract. The return is a cost-reduction share or added volume after success; notes before risk adjustment stay comparable across cells. Short delay places first settlement inside one or two payment cycles. Long delay places it after more than two improvement cycles and states that demand may move in the interval. Low safeguard writes only a principle of reasonable sharing after success. High safeguard adds a sharing formula or volume-binding clause, reserved funds or checkable order cover paid from a joint prepaid pool, and a pre-set dispute path such as third-party costing or a contractual adjudication route. The pretest must show that delay cells are seen as later rather than smaller, and that safeguard cells are seen as better executed rather than simply richer.

4.3 Measures

Dependent variables: acceptance of the intertemporal arrangement; willingness, on a 0–100 scale, to add input to the project; willingness to remain a preferred supplier or customer next period. Mechanism: perceived credibility (checkable, executable, the other party has resources to perform). Manipulation checks: perceived length of delay, perceived strength of safeguard, perceived present value of the return. Boundaries: pre-scenario ratings of liquidity strain and of alternative customers that can be reached. Controls: years of experience, perceived firm size, generalized trust, risk preference. Open reasons are coded as time cost, doubt about settlement, cash turnover, lock-in without exit, relationship maintenance, and other.

4.4 Ethics

Live contract parties and unpublished quotes are not used. Simulated choices do not create actual trade promises. Any later follow-up requires separate consent. Compensation is not tied to agreeing to a delayed arrangement.

5. Analysis and Falsification

Focal model: cooperation willingness ~ delay + safeguard + delay × safeguard + controls. The interaction corresponds to Hypothesis 3. Simple effects compare safeguard with no safeguard under long delay (C versus D). Hypothesis 4 mediates through perceived credibility and, in parallel, enters perceived return amount. Hypothesis 5 splits on median liquidity or adds a three-way interaction; if the sample is thin, that test is preregistered as exploratory. Hypothesis 6 reports whether, in the low-exit group, acceptance and added-input willingness come apart. Hypothesis 7 adds the solvency rating.

Falsification rules are written first. If the delay main effect is null, the manipulation must be checked; one may not announce that “delay is harmless.” If the safeguard raises perceived amount and not execution judgments, Hypotheses 2–4 should be rewritten. If the safeguard fails completely in the high-cash-constraint group, liquidity must be raised to a core boundary rather than discussed in passing. If the no-exit group accepts at a high rate but will not add input, acceptance must not be written as commitment success. The experiment explains simulated willingness, not years of live renewal.

Table 2. Hypotheses and Decision Rules

Note: All rules enter the preregistration. Overdue-invoice scenarios must not be mixed into the incremental arrangement.

6. Intended Theoretical Contributions

First, intertemporal conflict is rewritten from “is there enough trust?” into “how do delay, safeguard, and the capacity to wait combine?” Reciprocity theories easily read non-cooperation as low relationship quality. This paper allows a colder failure: the relationship judgment is still adequate, and cash will not last until settlement day.

Second, credible commitment is taken from principle to a manipulable bundle of safeguards. Clear text, bound funds or orders, and a dispute path must be in the materials if attitudinal trust is not to be confused with the treatment.[1]

Third, the future-return proposition in the companion treatise is constrained. A promise can connect contribution to continued input, but only given performance resources and the capacity to wait. A negative result would mark the boundary at which future claims cannot replace present payment.[5]

Fourth, a temporal contrast is set with Paper II. How a realized increment is cut, and how an unrealized increment is financed, are both allocation problems. Their mechanisms differ: one treats visible contribution and power; the other treats time, credit, and liquidity.

7. Managerial Implications

If the hypotheses hold, a buyer who asks a supplier to spend first should ask two questions. First, can the promise be checked and executed, or does it live only in a memorandum? Second, can the other party last until settlement, or is the capacity to advance cash being mistaken for cooperative attitude? Strong safeguards fit projects whose delay cannot be avoided and whose counterpart still has room to turn cash. When the counterpart has no cash buffer, staged compensation, a shorter cycle, or joint financing is the honest move—not a louder slogan.

Suppliers should keep “can I exit?” apart from “do I believe?” Accepting delay when there is no alternative customer may be lock-in, not fairness. If the study finds that separation, buyers should not read a high acceptance rate as institutional success.

8. Limitations, Boundaries, and Further Research

A scenario experiment identifies simulated choice. Live contracts are longer, more ambiguous, and more easily tangled with invoices already due. External validity depends on a practitioner sample and on realistic materials. Present-value matching will not be perfect; the pretest can reduce, not kill, amount confounding. Liquidity is self-rated, not taken from books.

The design does not apply to forced advances, unenforceable promises, extreme cash shortage, one-shot anonymous trades, or overdue payables renamed as joint input. Later work can unbundle the three safeguard elements, or, with separate consent, follow short-horizon settlement. Paper V will ask how a buyer under a tighter resource constraint adjusts its own extra input. That question and the present question of supplier waiting are counterparts. They should not be collapsed into one experiment.

9. Conclusion

Delay itself raises the cost of waiting. A promise eases intertemporal conflict only when it can be checked, when it can be executed, and when the other party can still last until that day. Without a safeguard, a future share is cheap talk. Without liquidity, the most complete clause cannot buy present cash. Willingness to continue cooperation is therefore not a salute to the language of shared growth. It is a judgment on three constraints: time, credit, and funds.

The claim is kept as a proposition to be tested. If the experiment shows that safeguards are read only as amount, or that they fail completely under cash constraint, the future-return device in the companion treatise must take on a harder boundary. The task of the research is to block two wrappings: calling unpaid invoices growth, and calling an unkeepable promise governance.

References

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

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

[3] Dyer, J. H., & Nobeoka, K. (2000). Creating and managing a high-performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21(3), 345–367.

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

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

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

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

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

[10] Poppo, L., & Zenger, T. (2002). Do formal contracts and relational governance function as substitutes or complements? Strategic Management Journal, 23(8), 707–725.

[11] Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734.

Appendix. Implementation Notes

Appendix A. The pretest must separate three perceptions: later, smaller, and less likely to be paid. Appendix B. Materials may not describe delay of invoices already due. Appendix C. Liquidity items are completed before the scenario is read, so the treatment cannot contaminate the boundary measure.

— End of manuscript —