Why 19% Is Actually Better Than 25% — The Supermodularity Argument for Jewellery Studios

The Gap Is Where the Architecture Lives

In Post 1, I corrected a two-year mathematical error: the peer-reviewed empowerment ceiling is 19.4% variance explained (Seibert et al., 2011), not the 25% I’d been citing. That correction wasn’t a retreat. It was setting up the real argument.

If psychological empowerment alone explained 25% of task performance variance, there would be no room left for synergy. The framework would be a single-lever intervention with a documented ceiling. You’d implement it, capture the 25%, and hit the wall.

But at 19.4%, there’s a gap. The other 80.6% of performance variance is driven by skill, resources, market conditions, organisational systems, random variance, and measurement error. Some of that 80% is noise. Some of it is structural reality you can’t change. But some of it — a meaningful, actionable portion — is addressable through complementary interventions that amplify each other’s effects.

That’s where the Coetzee Convergence Framework (CCF) lives. Not in claiming that any single component is stronger than the literature supports, but in the architectural thesis that empowerment + habit-based routines + blockchain provenance produce supermodular effects that exceed the sum of their individual contributions.

This post is the mathematical and empirical case for that claim.

Additive vs. Supermodular — The Mathematical Difference

In economics, supermodularity is the formal term for what business consultants vaguely call “synergy.” The concept was rigorously defined by Paul Milgrom and John Roberts in a 1990 paper in the American Economic Review titled “The Economics of Modern Manufacturing.”

Their formulation: A function f is supermodular if for all x, x’:

f(x) + f(x’) ≤ f(min(x, x’)) + f(max(x, x’))

In plain language: if you have two activities (say, empowerment and blockchain verification), they are complements when doing both together produces more value than doing each separately and adding up the results.

The calculus-based test (from Topkis, 1998) is even more precise: components are supermodular if the cross-partial derivative is positive:

∂²f/∂x₁∂x₂ > 0

This means: the marginal value of increasing x₁ (empowerment) is higher when x₂ (blockchain) is already at a high level. The presence of one component makes the other more effective. That’s complementarity. That’s what separates a bundled system from a collection of independent tools.

Now let’s apply this to the CCF.

The Arithmetic: Additive vs. Supermodular Scenarios

Additive scenario (no complementarity):

• Psychological Empowerment: 19.4% variance explained (Seibert et al., 2011)
• Habit-Based Routines: ~15-20% variance explained (estimated from Gersick & Hackman, 1990; Winter, 2013)
• Blockchain Provenance: ~8-12% variance explained (estimated from agricultural traceability premium studies, Hobbs, 2004)

Simple addition: 42-51% total variance explained if the three components operate independently with no interaction effects.

That would still be substantial. A jewellery studio capturing 45% of addressable performance variance through three validated interventions would be operating at a level most competitors cannot match. But it’s an additive model — each lever works in isolation, contributing its share with no mutual reinforcement.

Supermodular hypothesis (positive complementarity):

If the three components are architecturally designed to amplify each other — if empowerment increases the marginal value of blockchain verification, if routines free cognitive bandwidth for empowered problem-solving, if blockchain documentation reinforces habit consistency — then the joint effect exceeds simple addition.

The econometric test requires a regression model with interaction terms:

Performance = β₀ + β₁(Empowerment) + β₂(Habits) + β₃(Blockchain) + β₄(Empowerment × Habits × Blockchain) + ε

A significant positive β₄ coefficient proves supermodularity. The CCF’s theoretical prediction: β₄ > 0, producing total variance explained in the range of 55-65% under disciplined implementation.

That’s the hypothesis. It’s testable. It’s falsifiable. And it’s currently unvalidated in the jewellery sector — which is exactly why the CCF represents a research frontier, not a proven intervention.

The Ichniowski Precedent — Empirical Proof from Steel Mills

The strongest empirical precedent for the supermodularity claim comes from an unlikely source: steel finishing lines.

In 1997, Casey Ichniowski, Kathryn Shaw, and Giovanna Prennushi published a study in the American Economic Review analysing 36 steel production lines. They tested whether bundled High-Performance Work Systems (HPWS) — incentive pay + self-directed teams + cross-training + flexible job assignments — outperformed traditional management practices.

The findings were unambiguous:

• Lines using the full HPWS bundle: 7% higher uptime compared to traditional lines
• Lines adopting individual practices in isolation: zero productivity effect
• Partial bundles (2-3 practices): marginal gains, statistically insignificant

The lesson: complementarity is real, but it’s all-or-nothing. Implementing 60% of a complementary system doesn’t give you 60% of the benefit — it gives you close to zero. The practices only work when implemented together, because each one creates the conditions that allow the others to function.

Incentive pay without autonomy breeds resentment. Autonomy without skills produces chaos. Skills without feedback mechanisms decay. The bundle works because each element reinforces the others.

Now, the honest caveat: steel mills are not jewellery ateliers. Ichniowski’s study involved unionised industrial manufacturing with 200+ employees per line. The CCF targets artisan micro-enterprises with 8-25 staff. The work is creative, high-skill, low-volume — fundamentally different from repetitive industrial production.

So Ichniowski isn’t proof that the CCF’s complementarity claim holds in jewellery. It’s precedent that complementarity exists, is empirically detectable, and produces measurable performance gains when the bundle is complete. The hypothesis that similar mechanisms operate in craft sectors is plausible. The validation is pending.

The Three Mutual Reinforcement Mechanisms

The CCF’s supermodularity claim rests on three specific, testable mechanisms by which the pillars amplify each other:

Mechanism 1: Empowerment → Blockchain Value

The theory: Blockchain verification systems depend on input integrity. If the artisan logging a service record or origin certification is incentivised to falsify data, the immutable ledger simply immortalises fraud. Blockchain secures what is written — it doesn’t verify what should have been written.

In a high-trust, psychologically empowered environment (Spreitzer, 1995), the artisan’s intrinsic motivation to document accurately is higher. When competence, autonomy, meaningfulness, and impact are satisfied (the four dimensions of psychological empowerment), the need to game the system diminishes. Trust becomes the norm, not the exception.

The marginal effect: Blockchain’s value as a verification tool increases when empowerment reduces the falsification incentive. The two components are complements: empowerment raises the floor of data quality; blockchain raises the ceiling of auditability.

What’s validated: The empowerment → trust pathway is well-documented (Spreitzer et al., 1997). The blockchain → immutability property is definitional. What’s unvalidated is whether the combination produces measurably higher provenance premiums in jewellery markets than either component alone. That’s Proposition 1 in the CCF research agenda.

Mechanism 2: Habits → Empowerment Capacity

The theory: Daniel Kahneman’s dual-process model (2011) distinguishes between System 1 (fast, automatic, effortless) and System 2 (slow, deliberate, cognitively demanding). High-quality creative work — the kind that justifies a luxury price premium in jewellery — requires System 2 capacity: design iteration, client consultation, quality control, problem-solving under uncertainty.

But System 2 has limited bandwidth. Cognitive resources are finite. Every operational decision that requires deliberate thought — “Where did I file that invoice?” “What’s the next step in this repair protocol?” “Who has authority to approve this?” — depletes the capacity available for creative work.

Habit-based routines automate the operational layer. When the daily service log, the inventory audit, the client follow-up sequence, and the quality checkpoint are embedded as routines (Gersick & Hackman, 1990), they shift from System 2 to System 1. They become automatic, requiring minimal cognitive load.

The marginal effect: By freeing System 2 bandwidth, routines create capacity for empowerment. An artisan with cognitive bandwidth available can exercise autonomy, solve complex problems, and experience the intrinsic satisfaction (competence, impact) that psychological empowerment depends on. An artisan drowning in operational chaos cannot — even if the founder claims to empower them.

What’s validated: Kahneman’s dual-process model is among the most cited frameworks in cognitive psychology. The routines-as-automation mechanism is validated by Winter (2013) and the organisational routines literature. What’s unvalidated is whether habit deployment measurably increases empowerment effectiveness (the moderation hypothesis) in artisan settings. That’s Proposition 2.

Mechanism 3: Blockchain → Habit Consistency

The theory: Habits decay without feedback. Duhigg’s (2012) habit loop — cue, routine, reward — depends on the reward being consistent and immediate. In organisational settings, the “reward” for maintaining a routine is often social (manager approval) or economic (performance bonus). Both are delayed, subjective, and unreliable.

Blockchain documentation changes the feedback structure. Every service record logged, every lifecycle event documented, every artisan signature recorded creates an immutable accountability loop. The ledger is the feedback mechanism. It’s immediate (timestamped on-chain), objective (cryptographically verified), and permanent (cannot be retroactively edited).

The marginal effect: Blockchain reinforces habit consistency by making compliance visible and non-compliance detectable. The routine becomes self-enforcing — not through managerial surveillance, but through the structural property of the verification system itself.

What’s validated: The habit-feedback relationship is core to behavioural psychology (Wood & Neal, 2007; Lally et al., 2010). Blockchain’s immutability and timestamping properties are definitional. What’s unvalidated is whether blockchain-based feedback loops produce measurably higher habit compliance rates than traditional SOPs or manager oversight. That’s Proposition 3.

The convergent property: These three mechanisms create a reinforcement triangle. Empowerment increases blockchain’s marginal value. Habits increase empowerment’s marginal effectiveness. Blockchain increases habits’ marginal consistency. Each component makes the others work better.

That’s the supermodularity claim. It’s theoretically grounded. It’s empirically testable. And if validated, it represents a genuinely novel contribution to the organisational science of craft micro-enterprises.

The Provenance Premium — What Blockchain Actually Contributes

One component of the supermodularity equation deserves specific attention: the blockchain provenance premium. What does verified traceability actually contribute to jewellery business performance?

The empirical benchmark comes from agricultural supply chains. Jill Hobbs’ 2004 study in the British Food Journal analysed traceability systems in high-value food markets (organic beef, specialty coffee, artisan cheese) and found that verified provenance reduces information asymmetry between buyers and sellers.

The documented premium: consumers are willing to pay 15-30% more for goods with verified “credence attributes” — characteristics that cannot be observed even after purchase (ethical sourcing, environmental impact, origin authenticity). Traceability converts credence attributes into verifiable claims, reducing the buyer’s risk premium.

In jewellery, the GIA (Gemological Institute of America) certification system is the existing proof-of-concept. A GIA-certified diamond commands a measurable premium over an uncertified stone of identical physical properties, because the certification solves Akerlof’s (1970) “market for lemons” problem. The buyer can verify quality without expert knowledge.

The CCF’s blockchain provenance system (Diamond Stack Digital Passport) extends this logic across the entire product lifecycle:

• Origin: Mine or lab of manufacture, documented on-chain
• Cutting: Artisan signature, yield data, quality control checkpoints
• Setting: Bench jeweller attribution, materials provenance, assembly timeline
• Service history: Every repair, resize, re-polish, stone replacement logged immutably

The hypothesis: if GIA certification (which verifies a single point in time) commands a premium, then full lifecycle traceability (which verifies the entire chain of custody and service history) should command a larger premium — because it solves a broader information asymmetry problem.

What’s validated: The provenance premium in credence goods markets (Hobbs, 2004). The GIA premium in diamond pricing (industry standard, though peer-reviewed jewellery pricing studies are sparse). What’s unvalidated is whether blockchain-based lifecycle traceability produces measurably higher premiums than GIA certification alone in the luxury jewellery market. That requires price elasticity studies the industry hasn’t published yet.

The conservative estimate for the CCF model: 8-12% variance explained by blockchain provenance — treating it as the smallest component of the three-pillar system. But if the provenance premium hypothesis holds, that figure could be substantially higher.

Counter-Evidence: When Bundling Fails (The Credibility Guardrail)

The case for supermodularity is strong. The counter-evidence is equally important. Because the CCF’s intellectual positioning depends on acknowledging failure modes upfront — not burying them in footnotes or dismissing them as “implementation issues.”

Here are the three documented threats to complementarity in bundled organisational interventions:

Threat 1: The Digital Panopticon Effect

The research: Bendoly, Croson, Goncalves, and Schultz (2006) studied electronic monitoring systems in manufacturing and found that constant surveillance — even when framed as “transparency” or “quality control” — can trigger a reduction in intrinsic motivation. When workers perceive monitoring as a control mechanism rather than a support tool, psychological reactance kicks in. Performance declines.

The CCF risk: Blockchain verification creates a permanent, immutable record of every artisan’s work. In a high-trust environment, this is experienced as professional recognition and IP protection. In a low-trust environment — particularly one transitioning from micromanagement to empowerment — it can be experienced as a “digital panopticon”: the founder’s eye, encoded in software, watching every move.

The mitigation: The CCF’s implementation protocol requires psychological safety first (Edmondson, 1999). The founder’s public commitment to transparency (via CEO Diaries), the explicit framing of blockchain as “artisan IP protection” rather than “owner surveillance,” and the artisan’s control over what they choose to document (within minimum standards) are designed to prevent the panopticon interpretation. But the risk is real, and the mitigation is behavioural, not technical.

The honest assessment: If the founder cannot credibly surrender the “smartest person in the room” identity, the blockchain pillar will undermine the empowerment pillar. The supermodularity collapses into interference — a negative interaction effect where the combination performs worse than the individual components.

Threat 2: Implementation Complexity Overwhelms Small Teams

The research: Ennen and Richter (2010) analysed strategic complementarities in management systems and found that while synergies exist in theory, they are bounded by implementation capacity. Adding a third or fourth component to a bundled system doesn’t just add complexity linearly — it increases it exponentially, because each new component creates interaction effects with all existing components.

A three-component system (empowerment + habits + blockchain) has three pairwise interactions plus one three-way interaction. A five-component system has ten pairwise interactions plus ten three-way interactions plus five four-way interactions plus one five-way interaction. The cognitive load of managing the system grows faster than the performance gains.

The CCF risk: An 8-25 person jewellery studio does not have the management infrastructure of a 200-person steel mill. The owner-manager is often the COO, CFO, and Head of Sales simultaneously. Adding a three-pillar transformation framework on top of daily operations can produce “initiative overload” — where the cost of managing the bundle exceeds the benefit of the synergy.

The mitigation: The CCF’s 12-week deployment protocol is explicitly designed to phase implementation: identity activation first (Weeks 1-4), energy infrastructure second (Weeks 5-8), habit automation third (Weeks 9-12). Blockchain adoption happens after the first cycle is stable, not simultaneously. Sequential deployment reduces cognitive load. But it also extends the timeline to measurable results — which tests the founder’s patience and commitment.

The honest assessment: The CCF is not a quick fix. Studios that attempt to implement all three pillars simultaneously, without external support or phased rollout, are likely to fail. The framework’s complexity is a feature (it addresses multiple performance drivers) and a bug (it demands sustained execution discipline).

Threat 3: The Micromanagement Trap (Empowerment as Myth)

The research: Amy Edmondson’s 1999 study on psychological safety in work teams found that empowerment initiatives fail when the espoused organisational value (autonomy, trust, delegation) contradicts the enacted behaviour (founder intervention, second-guessing, post-hoc criticism). Subordinates learn quickly that “empowerment” is performative — the founder claims to delegate but still controls the outcome.

In that environment, the “habit routines” become rigid scripts that eliminate discretion, and the “blockchain documentation” becomes evidence collected for future reprimand. The bundle doesn’t empower — it bureaucratises. Performance declines.

The CCF risk: Jewellery studio founders are, almost by definition, control-oriented. They built the business through personal technical mastery, client relationships, and curatorial judgement. The inventory is high-value and theft-vulnerable. The reputational risk of a single quality failure is severe. These are rational reasons to maintain tight control.

The CCF asks the founder to surrender that control — not recklessly, but systematically, by encoding their expertise into routines and trusting the team to execute. That is an identity transformation, not just a process change. Many founders cannot make it.

The mitigation: The CCF’s “founder identity reconstruction” intervention (Weeks 1-4 of the deployment protocol) explicitly addresses this. The founder must articulate a new identity: from “master craftsman” to “systems architect,” from “sole expert” to “knowledge curator,” from “quality gatekeeper” to “empowerment designer.” This is therapeutic-grade work, often requiring external coaching or peer accountability.

The honest assessment: If the founder is not ready to change, the CCF will not work. No amount of blockchain sophistication or habit design can compensate for a leader who claims to empower but continues to micromanage. The framework filters for readiness. Studios that aren’t ready should not implement it.

The Realistic Forecast — What Supermodularity Predicts for Your Studio

Given the theoretical grounding, the empirical precedent, and the documented failure modes, what should a jewellery business owner reasonably expect from implementing the CCF?

Here are three scenarios, calibrated to implementation quality:

Scenario 1: Poor Implementation (Negative Interaction Effect)

• Conditions: Founder claims to empower but micromanages. Habit routines imposed without artisan input. Blockchain perceived as surveillance. No psychological safety.
• Predicted outcome: Empowerment collapses to near-zero (artisans learn it’s performative). Habits become rigid bureaucracy (compliance without buy-in). Blockchain triggers reactance (intrinsic motivation declines).
• Total variance explained: 25-35% — worse than the additive prediction, because the components interfere with each other. The negative interaction effect (β₄ < 0) cancels part of the individual contributions.
• Conclusion: Partial implementation of a complementary system is worse than doing nothing. This is the Ichniowski lesson applied to jewellery.

Scenario 2: Disciplined Execution (Modest Synergy)

• Conditions: Founder undergoes identity reconstruction. Habits co-designed with artisans. Blockchain framed as IP protection. Psychological safety established through CEO Diaries. Phased implementation over 12-18 months.
• Predicted outcome: Empowerment reaches ρ = .44 (19.4% variance). Habits contribute ~18% (routines automate operations, freeing cognitive bandwidth). Blockchain contributes ~10% (provenance premium + accountability loop). Modest positive interaction effect (1.2x multiplier on the additive baseline).
• Total variance explained: 55-60% — meaningfully above the 48% additive baseline, representing genuine synergy.
• Conclusion: The base case for a well-executed CCF implementation. This is the realistic target for studios with moderate readiness and disciplined execution.

Scenario 3: Full Transformation (Strong Synergy)

• Conditions: Founder fully surrenders “smartest person” identity. Artisans co-own the habit design and blockchain protocol. High psychological safety (error reporting without punishment). Habit compliance >85%. Full blockchain adoption with client-facing provenance storytelling.
• Predicted outcome: Empowerment reaches ρ = .55 (30% variance — upper range of Seibert’s moderator analysis for high-complexity work). Habits contribute ~22% (routines become cultural norms, not imposed rules). Blockchain contributes ~15% (provenance premium + IP asset creation). Strong positive interaction effect (1.4x multiplier).
• Total variance explained: 65-70% — the optimistic upper bound, achievable but rare.
• Conclusion: The CCF operating at full potential. This is the scenario where the framework becomes genuinely transformative — but it requires near-ideal implementation conditions that most studios cannot sustain.

The honest framing: These are projections based on steel mill precedent (Ichniowski et al., 1997), HPWS meta-analyses (Combs et al., 2006), and psychological empowerment moderator analysis (Seibert et al., 2011). They are not jewellery-specific validated outcomes. The CCF’s empirical contribution will be testing whether these mechanisms hold in artisan micro-enterprises.

What we know: the components are validated. What we’re testing: whether the synergy claim holds in this specific context. That’s the intellectually honest positioning — and it’s a stronger sales argument than “10x guaranteed.”

Why 19% Is Better Than 25%

The correction in Post 1 wasn’t a retreat. It was clearing space for the real claim.

If psychological empowerment alone explained 25% of performance variance, the CCF would be redundant. The framework would be a single-lever intervention dressed up as a system. But at 19.4%, there’s room for architecture. There’s space for complementarity. There’s a gap where supermodular effects can live.

The CCF’s value proposition isn’t that each component is stronger than the literature supports. It’s that the components are designed to amplify each other:

• Empowerment increases blockchain’s marginal value by reducing falsification incentive
• Habits increase empowerment’s marginal effectiveness by freeing cognitive bandwidth
• Blockchain increases habits’ marginal consistency through immutable accountability loops

That’s the architectural advantage. It’s grounded in Milgrom and Roberts’ (1990) theory of supermodularity. It’s precedented by Ichniowski’s (1997) empirical work on bundled HPWS practices. It’s threatened by documented failure modes (digital panopticon, implementation complexity, micromanagement trap) that the CCF explicitly addresses.

The hypothesis is testable. The failure modes are acknowledged. The projections are conservative and conditional. And if validated, the framework represents a genuinely novel contribution to the organisational science of luxury craft micro-enterprises.

That’s harder to promise than “10x Multipliers.” It’s easier to defend. And it’s the only positioning that survives scrutiny from the jewellery business owner who has seen every consultant promise in the book — and learned to demand receipts.

The gap between 19% and 25% is where the CCF lives. Not in inflated individual effect sizes, but in the disciplined architectural claim that well-designed systems outperform the sum of their parts.

That’s the supermodularity argument. That’s why 19% is better than 25%. And that’s the intellectually honest case for why the Coetzee Convergence Framework is worth your attention.

References

Akerlof, G. A. (1970). The market for “lemons”: Quality uncertainty and the market mechanism. Quarterly Journal of Economics, 84(3), 488–500.

Bendoly, E., Croson, R., Goncalves, P., & Schultz, K. (2006). Bodies of knowledge for research in behavioral operations. Production and Operations Management, 15(4), 434–452.

Combs, J., Liu, Y., Hall, A., & Ketchen, D. (2006). How much do high-performance work practices matter? A meta-analysis of their effects on organizational performance. Personnel Psychology, 59(3), 501–528.

Duhigg, C. (2012). The power of habit: Why we do what we do in life and business. Random House.

Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383.

Ennen, E., & Richter, A. (2010). The whole is more than the sum of its parts—Or is it? A review of the empirical literature on complementarities in organizations. Journal of Management, 36(1), 207–233.

Gersick, C. J. G., & Hackman, J. R. (1990). Habitual routines in task-performing groups. Organizational Behavior and Human Decision Processes, 47(1), 65–97.

Hobbs, J. E. (2004). Information asymmetry and the role of traceability systems. Agribusiness, 20(4), 397–415. https://doi.org/10.1002/agr.20020

Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource management practices on productivity: A study of steel finishing lines. American Economic Review, 87(3), 291–313.

Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.

Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009.

Milgrom, P., & Roberts, J. (1990). The economics of modern manufacturing: Technology, strategy, and organization. American Economic Review, 80(3), 511–528.

Seibert, S. E., Wang, G., & Courtright, S. H. (2011). Antecedents and consequences of psychological and team empowerment in organizations: A meta-analytic review. Journal of Applied Psychology, 96(5), 981–1003. https://doi.org/10.1037/a0022676

Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions, measurement, and validation. Academy of Management Journal, 38(5), 1442–1465.

Spreitzer, G. M., Kizilos, M. A., & Nason, S. W. (1997). A dimensional analysis of the relationship between psychological empowerment and effectiveness, satisfaction, and strain. Journal of Management, 23(5), 679–704.

Topkis, D. M. (1998). Supermodularity and complementarity. Princeton University Press.

Winter, S. G. (2013). Habit, deliberation, and action: Strengthening the microfoundations of routines and capabilities. Academy of Management Perspectives, 27(2), 120–137.

Wood, W., & Neal, D. T. (2007). A new look at habits and the habit-goal interface. Psychological Review, 114(4), 843–863.