The Loyalty Coefficient

Recovered Historical Material

The Loyalty Coefficient

Technical Brief

“The system does not control people. It manages them. The distinction is important, legally and psychologically. The managed don’t feel managed. They feel supported.”— Nexus People Analytics internal documentation

The Number You’ve Never Seen

Every Big Three employee has a number they’ve never seen.

• Management, not control: Control implies someone making decisions for you. Management implies an environment shaped to make certain decisions easier and others harder. The managed don’t feel managed. They feel supported. No court has ever ruled that “environmental optimization” constitutes coercion.

• The observer effect: A metric that changes when observed cannot be shared with its subjects. Every employee lives inside a number that would alter their behavior if they knew it existed. The darkness is not a bug — it is the system’s operating requirement.

• The notebook problem: What happens when someone reverse-engineers their own score? The system assumes darkness is permanent. It has no contingency for an employee who arrives at the number independently — because the model says an 88 doesn’t do that.

Good Fortune’s → /world/corporations/good-fortune

Dr. Priya Achebe’s → /world/characters/dr-priya-achebe

The Loyalty Coefficient is a 0–100 scale maintained by Nexus People Analytics and licensed to Ironclad and Helix. It quantifies how captured you are. Inputs: compensation data, benefit utilization, social graph density within the organization, family dependency metrics, external opportunity assessment, and behavioral telemetry from neural interface monitoring. A zero means you’re already gone. A hundred means you’ll die here.

The language in Nexus’s internal documentation for scores above 90 is “integrated human capital.” That phrase passed legal review on the first attempt. It has never been challenged by HR.

Employees who score below 40 receive retention interventions. Employees above 90 receive nothing. They don’t need anything. That’s what 90 means.

The Coefficient is never shared with employees. This is not a policy born of caution — it is structural necessity. An employee who learns they score 91 might resent it. An employee who learns they score 38 might feel permission. The metric functions only in the dark, which is why Nexus People Analytics occupies a windowless floor in Tower 7 and why its org chart has never appeared in any employee-facing directory. The team’s own Loyalty Coefficients average 94. They find this amusing in a way they can’t share with anyone.

Six primary input vectors feed the score: compensation relative to external market, benefit utilization depth, social graph density within the organization, family dependency load carried through corporate infrastructure, external opportunity assessment scraped from market data, and behavioral telemetry drawn from neural interface monitoring. Each vector is weighted dynamically. A single parent whose children attend a corporate school weights differently than a researcher with three competing offers.

Ironclad averages 81 — highest of the Big Three. When your employer controls your housing block, your transit line, and the air processing unit in your ceiling, the Coefficient is less a measurement than a description of geography. You don’t leave Ironclad. You’d have to move.

Nexus averages 72. Computational skills are portable in theory. The Coefficient accounts for this by weighting Nexus’s social graph density higher — the average Nexus employee has 73% of their non-family social connections inside the organization. Leaving Nexus doesn’t mean changing jobs. It means finding new friends.

Helix averages 63, a number that produces persistent irritation in Helix’s executive suite. Biotech researchers maintain external professional networks that the Coefficient reads as flight risk. Helix has responded by increasing benefit complexity: augmentation maintenance plans requiring Helix-certified facilities, pharmaceutical subscriptions with eighteen-month weaning protocols, genetic monitoring tied to proprietary baselines. The score has risen 4 points in two years. The researchers have not noticed any changes to their benefits package. They’ve noticed it’s harder to imagine leaving.

Their manager receives a briefing packet — framed as “engagement optimization insights” — containing suggested conversation topics, recommended project assignments, and a list of the employee’s most-used benefits ranked by replacement difficulty. The manager is not told the score exists. The manager believes they are receiving standard HR analytics. The analytics happen to suggest, with unusual specificity, that this particular employee might benefit from a team dinner, a visible project, and a reminder about their children’s tuition subsidy.

The Golden Handcuffs integrate directly: each benefit, each dependency, each social connection the employee maintains adds weight to the score. A retention intervention for a score-38 employee might include an unsolicited upgrade to their housing tier — a gift that feels like recognition and functions like a deeper anchor. The employee tells their partner about the upgrade. The partner adjusts expectations. The Coefficient ticks up 3 points. The intervention is logged as successful.

Nexus People Analytics tracks intervention-to-retention conversion at 74.2%. The 25.8% who leave anyway are flagged in a secondary database that Good Fortune’s hiring algorithms can query. Their next employer will know, without knowing how they know, that this person has a pattern.

The Big Three signed voluntary agreements to provide financial access, workplace stability, and community infrastructure. Workers opted in for housing subsidies, healthcare coverage, and social belonging. An entire labor economy whose housing, healthcare, augmentation maintenance, and social graph are now routed through the same entities measuring whether they’re too expensive to retain.

Good Fortune’s Consumer Mirror

Financial anxiety triggers a loan consolidation notification. Social isolation triggers a Fortune-sponsored event invitation. Early signs of questioning the relationship — search patterns for competitor rates, lingering on debt-reduction content — trigger a targeted package about “the value of financial partnership” that arrives with the warmth of a friend checking in and the precision of an actuarial table.

The Index’s most classified application is Sufficiency Threshold monitoring: real-time measurement of whether a district’s population is being provided enough to prevent organized resistance but not enough to produce demands for more. When a Dregs district drops below 62, Wholesome Basic increases caloric variety by 3% and Relief adjusts content streams toward community and gratitude narratives. When it rises above 78, variety contracts and content shifts to individual achievement stories.

The band between 62 and 78 is where a population stays quiet, stays fed, and stays exactly where the quarterly projections need it. Internal documentation calls this range the “civic equilibrium window.” The documentation does not note who decided where the window sits. The window has not moved since 2179.

One Nexus employee — documented in Dr. Priya Achebe’s clinical files — scores 88. His housing is Nexus-subsidized. His social graph is 91% internal. His daughter’s school is a Nexus educational partner. His augmentation maintenance requires Nexus-certified facilities. His pension vests in four years.

He has a notebook. Written in his own handwriting: a list of everything that ties him to Nexus, organized by category, annotated with estimated replacement costs. He does not know the Loyalty Coefficient exists. He is reverse-engineering it from the inside, arriving at approximately the same number through intuition and a growing sense that the comfort he feels is load-bearing.

The Coefficient would score the notebook itself as a risk indicator — evidence of analytical detachment from institutional identity. If People Analytics could see it, his score would drop 6 points and an intervention would trigger within the week. They cannot see it. He writes in it with a pen, on paper, in a drawer with no sensor.

The drawer has no sensor because Nexus does not monitor physical desk storage. This is listed in the employee privacy charter as a protected space. The charter was last updated in 2176. That update added fourteen new categories of neural telemetry to the monitoring framework and removed none. The drawer remains unmonitored. The notebook remains unread. The Coefficient remains 88 — correct about everything except the one thing that matters.

• The thermostat: Sufficiency Threshold monitoring doesn’t suppress dissent. It calibrates the precise amount of comfort needed to prevent dissatisfaction from becoming collective. Too little and people organize. Too much and they start wanting more.

There is a score above 100. Internal documentation references “Coefficient overflow” events — employees whose dependency load exceeds the scale’s upper bound. These individuals cannot leave even if they want to, even if they are fired. Their lives are so thoroughly integrated into corporate infrastructure that separation would require dismantling their identity.

People Analytics has no protocol for these cases. The overflow column in the database is formatted to two decimal places but has never, officially, been displayed on any dashboard. Three Ironclad employees currently overflow. Their names are held in a file requiring two-person authorization to access. The file is titled “Permanent Assets.”