Rethinking Effective Exit from Employment A Conditional Framework for Objective-Record Portability, Power-Produced Transition Burdens, and Residual Responsibility

Posted on 2026-09-04 In Philosophy , Social Science Views: Disqus: Word count in article: 12k Reading time ≈ 44 mins.

Transcript

Abstract

An employee can possess a legal and contractual power to resign while facing a transition whose practical conditions were materially shaped by the employment relation. Income loss, status change, firm-specific skill, geography, and ordinary search frictions may explain part of the burden. A narrower mechanism arises when a current or former employer controls authoritative evidence of employment or contribution, restricts its portable use, or deploys relation-derived informational power against successor access. This position paper develops a conditional framework for that mechanism.

The paper connects exit and voice, power-dependence, workplace authority, relational autonomy, non-domination, vulnerability, responsibility, labor market power, signaling, and job-reference research. It uses current Japanese Labor Standards Act Article 22 as its primary doctrinal illustration. That provision requires a requested certificate concerning enumerated employment facts, restricts unrequested content, and addresses specified secret communications intended to impede employment. ILO Recommendation No. 166 and British reference guidance provide bounded comparisons. Data-portability and verifiable-credential standards contribute design analogies with explicit limits.

The formal model represents effective exit as a typed, threshold-relative profile. It defines a causal contrast between observed conduct and a coherent ordinary-transition regime; constructs a portability set constrained by truth, attribution, verification, request, minimization, privacy, trade-secret, and third-party conditions; and models successor access as a path through an evidentiary graph. Propositions establish the separation between formal and effective exit, between objective record and favorable evaluation, and between relational termination and the persistence of relation-produced state. A three-valued bridge connects causal findings to defeasible residual responsibility through control, induced reliance, foreseeability, materiality, asymmetry, feasible mitigation, proportionality, and protected interests.

The framework supports a narrow institutional priority: accurate, request-responsive, contestable, and appropriately minimized objective records. Wider transition support requires stronger premises. Prevalence, causal effect, legal scope, remedy efficacy, strategic response, distributional impact, and cross-jurisdictional portability remain open.

Keywords: employment exit; labor mobility; objective records; employment references; portability; transition burden; residual responsibility; generative justice

Discussion Paper Note

This paper is a preliminary discussion paper intended to share an evolving idea and invite further dialogue, criticism, revision, and independent development.

The author claims responsibility for the definitions, formal constructions, taxonomy, arguments, selection of material, and conclusions presented here. Similar or related ideas may have appeared in other intellectual, cultural, or disciplinary traditions. Any legal rights retained in this work are intended to support attribution, responsible use, and protection against exploitative or harmful appropriation, while preserving legitimate inquiry, criticism, revision, and further development.

The arguments should be understood as provisional and historically situated. Readers are encouraged to question, revise, extend, reinterpret, or independently develop the ideas presented here. Acknowledgment of this paper as one point of encounter is appreciated where appropriate, while epistemic ownership over independently developed ideas remains with their contributors.

Responsible Use and Rights Reservation

The author encourages good-faith discussion, criticism, independent development, and responsible use of the knowledge presented in this work. Responsibility for lawful and ethically appropriate use remains with each user.

The author expressly reserves all rights and remedies available under applicable law with respect to unlawful conduct, harmful or abusive exploitation, improper commercial appropriation, infringement of applicable intellectual-property or other legal rights, and conduct contrary to applicable national, regional, or international law.

This reservation preserves the ability to respond to misuse and harmful appropriation. Legitimate academic inquiry, criticism, independent reasoning, and further development remain encouraged.

Notices

Status. This is a working draft circulated for discussion. Its definitions, propositions, examples, and section numbering remain open to revision.

Licence. This work is made available under a Creative Commons Attribution–NonCommercial 4.0 International Licence (CC BY-NC 4.0), subject to the rights reservation stated on the preceding page.

Statement on the use of language models. Drafting, literature search, structural review, formal reconstruction, and argumentative criticism were conducted in dialogue with ChatGPT (OpenAI). The author bears responsibility for the claims, definitions, formal constructions, selection of material, argument, and position taken. References included in the bibliography were checked against publisher, journal, university, governmental, intergovernmental, standards-body, or official legal records during preparation.

Companion papers. P034 develops bounded reciprocity, obligation self-extension, and longitudinal consent within scholarship-linked health service. The present paper owns effective employment exit, objective-record portability, and residual transition responsibility. Earlier papers develop recursive generative power, institutional dependency, and recognition bottlenecks at more general levels.

Suggested citation. Huang, W. Rethinking Effective Exit from Employment: A Conditional Framework for Objective-Record Portability, Power-Produced Transition Burdens, and Residual Responsibility. Working draft.

Introduction

This section establishes the employment-exit problem, its institutional importance, the paper’s conditional position, contribution, method, and sequence. The argument begins with the familiar act of resignation and then tracks the records and relations through which past work becomes accessible in a successor setting.

Resignation has a clear legal and contractual form. A worker gives notice or otherwise ends an employment relation. That event matters greatly. It ends specified powers and duties and can restore time, movement, and occupational choice. The capacity to leave also interacts with internal voice, loyalty, and organizational performance in Hirschman’s classic analysis (Hirschman 1970). Yet the formal event forms only part of the transition. A worker may require income, status continuity, transportation, safety, networks, proof of experience, and access to successor opportunities. Some conditions arise from the wider economy. Others were created, improved, concentrated, or damaged through the employment relation itself.

Consider a worker who can resign today. The former employer remains the only authoritative issuer of the worker’s dates, role, responsibilities, or project contributions. A successor institution uses those facts as an eligibility or credibility gate. The former employer delays the record, supplies a covertly obstructive signal, or offers only an opaque aggregate that erases the worker’s attributable contribution. The worker retains legal departure while losing a practical route through which earlier activity can count in a later relation. The example combines three distinct matters: an exit profile, an evidentiary object, and the causal history of a transition burden.

The paper asks:

Under which causal, institutional, and normative conditions does formal resignation leave effective employment exit impaired, and when does relation-derived control over transition conditions support a proportionate post-relation responsibility for portable objective records or related safeguards?

The preliminary answer is conditional. Legal permission is one coordinate of effective exit. A supported impairment requires a declared set of case-relevant dimensions and thresholds. A power-produced transition burden requires a coherent comparison regime that preserves ordinary transition conditions while disabling the focal employer-controlled pathway. The causal finding then enters a normative bridge. Control, induced reliance, foreseeability, materiality, asymmetric repair capacity, feasible mitigation, proportionality, and protected interests determine whether a residual responsibility receives support.

Objective-record portability is the paper’s narrow institutional priority. It concerns truthful, attributable, verifiable, request-responsive records of employment and contribution that can be used in a successor relation. A favorable recommendation is a discretionary evaluation with a different logical and institutional type. Organizational property, trade secrets, colleague or client data, and uncertain collective attribution impose further boundaries. These distinctions preserve the legitimate interests of former employers, coworkers, clients, and successor institutions.

The contribution has five parts. First, the paper defines effective employment exit as a multidimensional and threshold-relative profile. Second, it replaces an intuitive attribution of difficult transitions with a causal contrast and explicit identification duties. Third, it distinguishes objective records from evaluative references and constructs a constrained portability set. Fourth, it represents successor access through an opportunity graph, allowing evidentiary bottlenecks to become visible. Fifth, it states a three-valued normative bridge for residual responsibility and derives a bounded safeguard portfolio.

The method combines conceptual reconstruction, institutional comparison, formal modeling, countermodel construction, and empirical research design. Section 2 defines the institutional transition. Section 3 reconstructs the conceptual and political-economic antecedents. Section 4 analyzes legal and technical fragments. Section 5 states the paper’s core definitions. Sections 6–8 develop the formal framework. Section 9 supplies derivations and limits. Sections 10–12 address institutions, cases, and research design. The final sections state objections, future directions, and the conditional conclusion.

Employment Exit as an Institutional Transition

This section specifies the paper’s empirical object. It separates termination, transition, and successor access; identifies ordinary and relation-derived sources of burden; and explains why employment records receive focused attention.

Employment is both an exchange and an institution. It coordinates labor, income, authority, training, status, and social relations over time. Some forms of human capital travel readily across firms. Other knowledge and credibility remain tied to tasks, technologies, clients, or reputational networks. Search frictions, job differentiation, and employer concentration can produce labor market power even where legal mobility is available (Bhaskar, Manning, and To 2002; Berger, Herkenhoff, and Mongey 2022). OECD comparative work also treats job-to-job mobility as a significant mechanism of labor reallocation and wage growth (Organisation for Economic Co-operation and Development 2025).

The transition begins before the last day of work and can continue after it. The worker gives notice, completes a handover, requests records, seeks future work, represents earlier experience, and enters a new institutional setting. The employer protects operations, clients, coworkers, confidential information, and truthful communication. A successor employer evaluates uncertain claims and bears risks from false or materially incomplete information. The analysis therefore concerns a multi-party transition whose structure exceeds a bilateral demand for career assistance.

Table [tab:transition] decomposes this transition. The table has an organizational role: it prevents the final resignation event from standing in for the full sequence and locates the record bottleneck among several possible mechanisms.

@p0.16YYY@ Phase & Worker interest & Institutional control point & Principal counterinterest
Pre-notice & Information, advice, feasible timing & Scheduling, status, monitoring, internal mobility & Operational planning and confidentiality
Notice & Clear termination path and safe handover & Notice acceptance, work allocation, access & Continuity, client, and coworker interests
Record request & Accurate selected facts & Issuance, field choice, evidence, timing & Accuracy, privacy, collective attribution, cost
Search & Usable proof and accessible channels & Reference, credential, network, status, reputation & Honest evaluation and successor risk
Successor assessment & Fair interpretation of past work & Schema, verifier trust, translation, licensing & Role-specific competence and safety
Post-entry & Correction and durable provenance & Revocation, update, audit trail & Data minimization and finality

Three causal configurations deserve separation. In an ordinary transition, the worker bears search time, geographic constraints, incomplete information, and other costs present under a fair completion regime. In a joint specialization configuration, both parties knowingly invest in relation-specific skills or trust; later portability is limited by the nature of the activity. In a power-produced burden configuration, the employer uses or retains relation-derived control in a way that materially worsens the worker’s transition relative to a coherent completion regime. Shared temporal origin supplies insufficient grounds for collapsing the three.

Records deserve focused attention for four reasons. They concern a bounded object that institutions can often preserve and issue. They mediate information between a former and successor relation. Their accuracy and scope can receive procedural review. They also reveal a crucial distinction between proof and praise. Spence’s signaling model established the importance of observable signals under hiring uncertainty (Spence 1973). Häfner and Taylor show within a formal labor-market model that references can affect hiring, incentives, welfare, and worker welfare (Häfner and Taylor 2023). Neither source creates a moral entitlement. Together they explain why control over credible evidence can affect future opportunity.

Philosophical and Political-Economic Foundations

This section assigns distinct roles to six bodies of thought. The objective is conceptual inheritance with controlled transfer: exit and voice locate the organizational response; power-dependence and labor economics explain alternatives; workplace authority and non-domination identify power; relational autonomy locates agency conditions; vulnerability and social connection frame responsibility.

Exit, voice, and institutional membership

This subsection positions exit within organizational analysis. Hirschman’s framework distinguishes departure from attempts to repair an organization through voice, with loyalty shaping their timing and interaction (Hirschman 1970). That inheritance resists two reductions. A worker’s ability to resign carries genuine value. An institution may also improve when workers possess protected voice, information, and contestation. Effective exit and meaningful voice can function as complements, especially where an internal complaint could otherwise trigger later reputational costs.

The present framework adds a temporal extension. Exit is represented as a transition into later life and work. The extension creates a measurable question: which conditions of successor access remain controlled by the former relation? It also creates a limit. A wide conception needs a stopping rule that separates organizational contribution from every disappointment after departure.

Dependence, labor-market power, and workplace authority

This subsection supplies the descriptive account of power. Emerson locates relational power in dependence, where valued outcomes and available alternatives structure the relation (Emerson 1962). Modern labor economics identifies search friction, job differentiation, and concentration as mechanisms that can give employers labor-market power (Bhaskar, Manning, and To 2002; Berger, Herkenhoff, and Mongey 2022). Anderson emphasizes the governing authority exercised within employment and the practical cost of worker exit (Anderson 2017).

These accounts operate at different levels. Emerson supplies a relational schema. Labor economics supplies market mechanisms and estimable consequences. Anderson supplies a political-philosophical interpretation of workplace authority. The paper draws from each while preserving the distinction between market power, organizational authority, arbitrary power, and a specific record-control mechanism.

Autonomy, domination, and transition conditions

This subsection locates the normative significance of practical alternatives. Relational-autonomy scholarship treats agency as socially and institutionally conditioned (Mackenzie and Stoljar 2000). Lovett analyzes domination in terms of dependence within a relation of arbitrary power (Lovett 2010). An effective-exit profile can inform both traditions by representing conditions of action and contestability. It remains an analytic input. A low component contributes evidence; judgments of impaired autonomy or domination require their own theories and contextual facts.

The temporal point is central. A worker’s alternatives at departure can be partly generated by decisions made during the relation. Training can expand them. Specialization can narrow some and deepen others. Employer control of records can affect which earlier achievements become legible elsewhere. The history of the alternative set therefore matters alongside its current size.

This subsection frames the post-relation question. Goodin grounds responsibilities in vulnerability extending beyond explicit promises and contracts (Goodin 1985). Young’s social-connection model concerns responsibility for structural injustice among agents whose participation helps produce structural processes (Young 2011). These theories support attention to persistent effects and distributed bearers. They also create risks of overextension if causal contribution, control, institutional role, remedy feasibility, and proportionality remain unspecified.

The paper therefore uses a typed bridge. Bilateral causal contribution is one factor. Market structure, licensing, social insurance, worker organizations, successor practices, and public credential systems may carry responsibility as well. Residual responsibility can be divided, attenuated, or transferred to a better-positioned institution. Its duration follows the mechanism and remedy and loses force as relevant causal contribution and control attenuate.

Table [tab:inheritance] records the theoretical division of labor. It also marks the paper’s original additions so that established positions and repository proposals remain distinct.

@p0.17YYY@ Tradition & Inherited insight & Paper-specific use & Transfer limit
Exit and voice & Departure and internal influence are distinct responses & Exit becomes a longitudinal transition & The extension is a repository proposal
Power-dependence & Alternatives structure relational power & Relation-derived control enters a causal contrast & Dependence carries no automatic verdict
Labor-market power & Friction and concentration affect mobility and wages & Market burden becomes a rival or interacting pathway & Aggregate power differs from record control
Workplace authority & Employers exercise governing authority & Post-relation informational control receives scrutiny & Jurisdictional and institutional variation remains
Relational autonomy & Agency depends on social conditions & Feasible transition conditions enter the audit & A vector remains one input to wider autonomy assessment
Non-domination & Arbitrary power within dependence has moral importance & Contestation and review enter safeguards & Domination requires further criteria
Vulnerability & Exposure can ground responsibilities & Repair capacity and induced reliance enter the bridge & Vulnerability alone leaves allocation open
Social connection & Structural processes support forward-looking responsibility & Responsibility can be distributed across institutions & Bilateral liability remains a separate inquiry
Generative justice & Future value circulation and generator participation matter & Past contribution is linked to future accessibility & The formal model is the author’s construction

Generative justice supplies the wider programme orientation. Eglash develops bottom-up circulation of unalienated value and generator participation in production (Eglash 2016); later work connects that orientation to decolonial transition and computational reparations (Eglash et al. 2024). The present paper asks how a worker’s past activity can remain truthfully accessible within future relations. The definitions, equations, and normative bridge developed here are the author’s responsibility and remain distinct from established generative-justice work.

Legal and Technical Fragments of Portability

This section identifies existing institutional fragments that discipline the proposal. It first examines Japanese employment certificates, then compares an ILO standard and British guidance, and finally evaluates data and credential standards as design analogies. The method is bounded comparison among distinct legal and technical objects, with unification left as a separate project.

Japanese retirement certificates

This subsection presents the primary doctrinal anchor. Article 22(1) of Japan’s Labor Standards Act provides that, upon a worker’s request at retirement, an employer shall promptly deliver a certificate concerning requested statutory fields: period of employment, kind of occupation, position in the enterprise, wages, or reason for retirement, including the dismissal reason when applicable. Article 22(3) restricts inclusion of fields the worker did not request. Article 22(4) addresses specified communications and secret signs made through prearrangement with a third party and with the purpose of impeding the worker’s employment (Government of Japan 2025). Current MHLW explanatory material reiterates the request-responsive certificate obligation (Ministry of Health, Labour and Welfare, Japan 2025).

The provision contributes three structural insights. A worker can request an authoritative record after the relation ends. The record has typed fields and a selection rule. Obstructive inter-institutional signaling is legally salient under specified elements. Article 22 also limits the paper’s claim. Its listed fields form a narrower object than a complete contribution passport. The manuscript adds the effective-exit vector, causal contrast, and residual- responsibility test as original proposals. Japanese specialist review, case law, administrative interpretation, limitations periods, remedies, and interactions with other law remain necessary before doctrinal publication.

International and British reference structures

This subsection supplies two bounded comparisons. Paragraph 17 of ILO Recommendation No. 166 states that a terminated worker should be entitled on request to a certificate specifying engagement and termination dates and work types; an evaluation may be included or separately issued at the worker’s request. Paragraph 16 addresses reasonable paid time during notice for seeking other employment (International Labour Organization 1982). The recommendation principally addresses employer-initiated termination and operates through national implementation. Its importance here lies in the structural separation of service facts, evaluation, and job-search transition.

Acas guidance for Great Britain distinguishes basic factual references from detailed references. When a reference is given, it should be fair and accurate; opinion should have an evidentiary basis, and irrelevant or discriminatory information faces restrictions (Advisory, Conciliation and Arbitration Service 2024). The guidance also recognizes provider choice over detail in many settings. It therefore supports accuracy and type separation while warning against an assumed general duty to provide praise or extensive evaluation.

Constrained data and credential analogies

This subsection tests implementation analogies. GDPR Article 20 provides a right, under specified conditions, to receive certain personal data in a structured, commonly used, machine-readable format and transmit them without hindrance. The right preserves the rights and freedoms of others (European Parliament and Council of the European Union 2016). The analogy contributes format, transfer, and third-party constraints. Its legal object and conditions differ from an employment record of contribution.

The W3C Verifiable Credentials Data Model v2.0 separates issuer, holder, and verifier roles and supports tamper-evident, machine-verifiable credentials. The standard explicitly distinguishes verification from the truth of encoded claims and identifies privacy risks (World Wide Web Consortium 2025). The NISO CRediT taxonomy offers a controlled vocabulary for fourteen types of contribution to research outputs (National Information Standards Organization 2022). These standards show that provenance, typed contribution, and portable presentation are technically representable. They leave fairness, evidentiary adequacy, correction, acceptance, and entitlement to institutional governance.

Japanese trade-secret law supplies a further boundary. The Unfair Competition Prevention Act defines protected trade secrets and regulates specified acquisition, use, and disclosure (Government of Japan 2023). A portable record must therefore distinguish a person’s attributable role from protected technical or business information. Privacy, client duties, collective work, security, and sectoral safety add parallel constraints.

Table [tab:fragments] summarizes the resulting institutional map. Its purpose is to show partial precedents and gaps, with each row preserving its own jurisdiction and object.

@p0.18YYY@ Fragment & Supported structure & Preserved boundary & Remaining gap
Japan LSA art. 22 & Prompt, requested, typed retirement certificate & Enumerated statutory fields and specified anti-obstruction rule & Rich contribution, interoperability, and full remedial analysis
ILO R166 & Service certificate separated from requested evaluation & Employer-initiated termination context and national implementation & Universal enforceable right and causal responsibility
Acas guidance & Factual and detailed references; fairness and accuracy & Provider discretion, relevance, discrimination, evidence & Cross-border or universal portability
GDPR art. 20 & Structured machine-readable transfer under conditions & Defined personal data scope and rights of others & Employment-contribution entitlement
W3C credentials & Issuer, holder, verifier, provenance, integrity & Verification differs from truth; privacy risks & Fair schema, correction, acceptance, governance
NISO CRediT & Typed vocabulary for differentiated contribution & Research-output domain and role-level description & Employment-wide attribution and quality proof
Trade-secret protection & Legitimate exclusion of protected business information & Statutory definitions and elements & Practical minimization protocol

Normative Architecture

This section states the paper’s core concepts before formalization. The definitions separate permission, practical viability, causal production, portable evidence, and responsibility. Each concept remains revisable and requires a declared employment context.

Definition 1 (Formal employment exit). Formal employment exit is the legally and contractually recognized capacity to terminate an employment relation through a valid resignation, expiry, agreement, or other applicable mode. The present paper focuses on worker-initiated resignation and uses other modes only for institutional comparison.

Formal exit records a juridical power and event. It can be strongly protected or materially restricted depending on the jurisdiction and contract. The definition carries independent normative importance and forms one component of the wider profile below.

Definition 2 (Effective employment exit). A worker has effective employment exit relative to a declared target at time $t$ when formal departure is available and every case-relevant transition dimension reaches its stated viability condition. Candidate dimensions include material continuity, institutional status, relational access, portable records, successor opportunity, and safety.

The target matters. Effective exit into immediate retirement, a similar job, a new licensed profession, self-employment, or a period of care has different requirements. The definition protects multidimensional assessment and avoids a universal resource bundle.

Definition 3 (Power-produced transition burden). A transition burden is power-produced relative to a focal actor and mechanism when relation-derived control causes a material increase in that burden against a coherent ordinary-transition comparison. The comparison preserves the completed employment relation, legitimate protected interests, and relevant background conditions while disabling the focal burden-producing conduct.

The term identifies a causal property. Wrongfulness, domination, liability, and remedy require further premises. Deliberate retaliation can satisfy additional legal or moral elements; intention is an optional additional feature outside the causal definition.

Definition 4 (Objective-record portability). Objective-record portability is the practically usable transfer of truthful, attributable, verifiable, request-responsive, and appropriately minimized employment or contribution claims into a successor setting, subject to privacy, confidentiality, trade-secret, collective-contribution, intellectual-property, security, and third-party constraints.

Practical usability includes intelligible fields, timely issuance, durable provenance, feasible correction, and recognition by a relevant verifier. A paper certificate can satisfy these conditions. Digital form can improve some properties while introducing surveillance, correlation, exclusion, and security risks.

Definition 5 (Evaluative reference). An evaluative reference expresses an assessor’s judgment concerning quality, character, suitability, potential, or future performance. Its legitimacy depends on truthfulness, evidentiary basis, relevance, fairness, authorization, and applicable law. Its type remains distinct from an objective record.

Definition 6 (Residual transition responsibility). Residual transition responsibility is a defeasible, bounded responsibility whose potential content persists after termination because a party materially contributed to, controls, or is especially positioned to mitigate a continuing transition burden. Its bearer, content, strength, duration, and allocation are determined through explicit causal and normative premises.

Table [tab:definitions] displays the boundaries among these objects. The table prevents a movement from low effective exit directly to an entitlement to favorable praise or indefinite support.

@p0.20YYY@ Object & Primary output & Required support & Separate conclusion
Formal exit & Valid termination path & Law, contract, institutional fact & Practical viability
Effective exit & Threshold-relative transition profile & Declared target, dimensions, measurements & Causal attribution or injustice
Power-produced burden & Conditional causal contrast & Coherent regimes and identification & Wrongfulness or liability
Objective-record portability & Eligible and usable record set & Truth, attribution, verification, request, constraints & Praise, ownership, or broad disclosure
Evaluative reference & Evidenced assessor judgment & Competence, relevance, fairness, authorization & Objective-record content
Residual responsibility & Typed bearer-content-duration profile & Causal and normative bridge & Unlimited transition insurance

Effective-Exit State Space

This section formalizes effective exit and power-produced transition burden. It introduces domains, thresholds, policy regimes, potential outcomes, and identification conditions in that order. Aggregation remains optional and requires an independent justification.

Symbols and typed domains

This subsection types the construction. A product space preserves dimensions that carry different units and normative meanings. Table [tab:symbols] provides the symbol register used throughout the remaining formal sections.

@p0.16p0.25Y@ Symbol & Domain & Interpretation
$t\in{0,\ldots,T,T+1,\ldots}$ & discrete time & Employment, notice, termination, and follow-up points
$A,B,J$ & actor sets & Employer, worker, and successor institutions
$\mathbf X_B(t)$ & product partially ordered space $\mathcal X$ & Worker effective-exit profile
$\mathbf v_B(z,t)$ & $\mathcal X$ & Viability profile for target transition $z$
$\pi^O,\pi^C$ & conduct or policy regimes & Observed and coherent completion regimes
$\mathbf c_B(t;\pi)$ & product cost space $\mathcal C$ & Transition burdens under regime $\pi$
$P_A^R(t)$ & typed control space $\mathcal P$ & Employer control derived from the relation
$H_t$ & declared history space & Past treatment, outcomes, covariates, and latent variables
$\mathcal R$ & claim universe & Candidate employment and contribution claims
$\mathcal R_B^{\rm port}$ & subset of $\mathcal R$ & Eligible portable objective claims
$G_t=(V,E_t)$ & directed attributed graph & Paths from past contribution to successor opportunity
$\Omega_B(t)$ & opportunity set or typed profile & Successor opportunities accessible to the worker
$\mathbb V_3={1,0,?}$ & audit codomain & Supported, unsupported, or unresolved
$\boldsymbol\rho_{A,t}$ & responsibility product space & Bearer, content, strength, duration, and allocation

The order relation $\succeq$ on $\mathcal X$ is componentwise after every coordinate has a declared direction. Each coordinate can be binary, ordinal, interval-valued, set-valued, or probabilistic. A question mark belongs to the epistemic audit; the physical state uses its declared coordinate domain.

Exit profile and viability condition

This subsection defines the multidimensional state. For worker $B$, let

$$\mathbf X_B(t;z)=
\left\langle
x_B^{\rm legal},
x_B^{\rm material},
x_B^{\rm status},
x_B^{\rm relational},
x_B^{\rm record},
x_B^{\rm opportunity},
x_B^{\rm safety}
\right\rangle_{t;z}.
\label{eq:exit-profile}$$

The effective-exit indicator is

$$E_B(t;z)=
\mathbf 1!\left{\mathbf X_B(t;z)\succeq
\mathbf v_B(t;z)\right}.
\label{eq:exit-indicator}$$

The indicator reports a threshold decision after the underlying vector has been retained. Reporting only $E_B$ would erase which dimension failed, the severity of the shortfall, and the remedy relevant to it. A robust version can require the threshold across a declared scenario set $\mathcal S$:

$$E_B^{\rm rob}(t;z)=
\min_{s\in\mathcal S}
\mathbf 1!\left{\mathbf X_B(t;z,s)\succeq
\mathbf v_B(t;z,s)\right}.
\label{eq:robust-exit}$$

Proposition 7 (Formal and effective exit separation). For a target $z$ with at least one required nonlegal coordinate, satisfaction of the legal coordinate is insufficient for $E_B(t;z)=1$.

Proof. Let $x_B^{\rm legal}\succeq v_B^{\rm legal}$. Select a required coordinate $k\neq\mathrm{legal}$ and let $x_B^k\prec v_B^k$. Componentwise threshold satisfaction in equation [eq:exit-indicator] fails, so $E_B(t;z)=0$. ◻

The proposition is logical and modest. It establishes the possibility of a profile with legal permission and practical impairment. Evidence must determine which coordinates are required and whether a real case falls below them.

Transition burden and causal contrast

This subsection formalizes attribution. Define the burden profile

$$\mathbf c_B(t;\pi)=
\left\langle
c_B^{\rm income},c_B^{\rm housing},c_B^{\rm status},
c_B^{\rm search},c_B^{\rm record},c_B^{\rm network},
c_B^{\rm safety}
\right\rangle_{t;\pi}.
\label{eq:burden-profile}$$

The observed regime $\pi^O$ includes the focal conduct. The coherent completion regime $\pi^C$ preserves the worker’s completed employment, ordinary market conditions, legitimate confidentiality and safety constraints, and a fair handover while disabling the specified burden-producing use of relation-derived control. Conditional on history $H_t=h$, define

$$\boldsymbol\Delta_{\rm PB}(h)=
\mathbb E!\left[
\mathbf c_B(T+1;\pi^O)-\mathbf c_B(T+1;\pi^C)
\mid H_T=h
\right].
\label{eq:power-burden}$$

The contrast is componentwise. A scalar summary $w^\top\boldsymbol\Delta_{\rm PB}$ is admissible only after the application defends the weights, comparability, and distributional meaning. The hypothesized record pathway is

$$P_A^R(T)
\longrightarrow
q_B^{\rm record}(T+1)\downarrow
\longrightarrow
\Pr!\left(\Omega_B(T+1)\ni z\right)\downarrow
\longrightarrow
\mathbf c_B(T+1)\uparrow .
\label{eq:record-pathway}$$

The arrows express a causal hypothesis whose support requires evidence. Market scarcity, worker preference, occupation-specific licensing, health, family constraints, and common causes can affect both record requests and later outcomes.

Assumption 8 (Coherent completion regime). The comparison $\pi^C$ is institutionally possible, preserves legitimate completion and protected interests, and differs from $\pi^O$ only through declared interventions needed to disable the focal pathway.

Assumption 9 (Longitudinal identification). Consistency, temporal order, positivity for the relevant histories, adequate measurement, and exchangeability conditional on a justified history set hold for the target contrast, or a valid alternative identification strategy is available.

Hernán and Robins provide the methodological background for potential outcomes, longitudinal treatments, time-varying confounding, and target-trial discipline (Hernán and Robins 2020). The present variables and regimes remain paper-specific. Where assumptions 8 or 9 lack support, equation [eq:power-burden] is an estimand with unresolved identification.

Record Portability and Successor Opportunity

This section constructs the portable record as a constrained informational object and then models its role in successor access. The order is deliberate: eligibility for disclosure precedes any claim about opportunity effects.

Claim eligibility and record construction

This subsection defines a candidate claim and its audit. For each $r\in\mathcal R$, record

$$\begin{split}
\mathbf a(r)=\langle{}
&a^{\rm true},a^{\rm attributable},a^{\rm verifiable},a^{\rm requested},
a^{\rm relevant},\
&a^{\rm minimized},a^{\rm privacy},a^{\rm secret},
a^{\rm collective},a^{\rm third\ party}
\rangle_r\in\mathbb V_3^{10}.
\end{split}
\label{eq:claim-audit}$$

For the first six coordinates, value $1$ means the positive condition is supported. For the final four, value $1$ means the applicable protective constraint has been satisfied. Define

$$\mathcal R_B^{\rm port}(t)=
\left{r\in\mathcal R:
\mathbf a(r)=\mathbf 1,
\ \operatorname{issuer}(r)\in\mathcal I_t,
\ \operatorname{format}(r)\in\mathcal F_t
\right}.
\label{eq:portable-set}$$

$\mathcal I_t$ is the accepted issuer set and $\mathcal F_t$ the usable format set for the target. A record can be truthful and still fail because attribution is disputed, verification is weak, worker request is absent, disclosure would reveal a trade secret, or the successor rejects the issuer or format.

Let $\mathcal R_A^{\rm eval}$ contain assessor judgments. The type signature is

$$\mathcal R_B^{\rm port}\subseteq\mathcal R^{\rm factual},
\qquad
\mathcal R_A^{\rm eval}\subseteq\mathcal R^{\rm evaluative},
\qquad
\mathcal R^{\rm factual}\neq\mathcal R^{\rm evaluative}.
\label{eq:record-types}$$

Proposition 10 (Objective record and favorable evaluation separation). Eligibility of $r$ for $\mathcal R_B^{\rm port}$ entails neither membership of a favorable evaluation in $\mathcal R_A^{\rm eval}$ nor an obligation to produce such an evaluation.

Proof. Equation [eq:portable-set] ranges over factual claims satisfying the claim audit. A favorable evaluation belongs to the separate evaluative type in equation [eq:record-types]. No inclusion map or normative rule from the first type to the second has been introduced. The entailments therefore fail within the model. ◻

The proposition gives formal expression to a distinction visible in Japanese Article 22, ILO Recommendation No. 166, and British practice. It also establishes a design principle: a factual certificate can receive standardized governance while evaluative speech remains separately regulated.

Record bundles and provenance

This subsection defines a portable bundle. Let

$$\mathsf{RB}_B(t)=
\left\langle
B,\ I,\mathcal R_B^{\rm port}(t),\tau,\sigma,
\mathcal E,\mathcal K,\mathcal H^{\rm corr}
\right\rangle ,
\label{eq:record-bundle}$$

where $I$ is the issuer, $\tau$ the issue and validity metadata, $\sigma$ an integrity or signature mechanism, $\mathcal E$ the evidence basis, $\mathcal K$ the disclosure constraints, and $\mathcal H^{\rm corr}$ the correction and revision history. A compliant digital object can implement some of these fields using the W3C credential architecture (World Wide Web Consortium 2025). Certification of truth or fairness remains a separate institutional task. Rules must determine evidence, challenge, correction, and issuer accountability.

Contribution typing is especially difficult. A role taxonomy similar to CRediT can distinguish conceptualization, investigation, software, supervision, or other functions within its research domain (National Information Standards Organization 2022). Employment requires sector-specific vocabularies, and a role label leaves effort, quality, ownership, and individual causal contribution open. The bundle should therefore preserve provenance and uncertainty and avoid an inflated authorship certificate.

Successor opportunity graph

This subsection represents the route from past activity to future access. Let $G_t=(V,E_t)$ be a directed graph with node types

$$V=V^{\rm activity}\cup V^{\rm evidence}\cup V^{\rm issuer}
\cup V^{\rm verifier}\cup V^{\rm eligibility}\cup V^{\rm opportunity}.
\label{eq:graph-nodes}$$

An edge $e=(u,v)$ has availability $m_e(t)\in{0,1,?}$, reliability $p_e(t)\in[0,1]$, latency $\ell_e(t)\geq0$, and controller set $K_e(t)$. For a path $p$ from an activity node $g$ to opportunity $\omega$, define the path reliability

$$R_p(t)=prod_{e\in p:,m_e(t)=1}p_e(t),
\qquad
L_p(t)=\sum_{e\in p}\ell_e(t),
\label{eq:path-quality}$$

with unresolved availability reported separately. Opportunity accessibility is

$$A_B(\omega,t)=
\max_{p\in\mathcal P_B(g,\omega;t)}
\left{R_p(t):L_p(t)\leq \bar L_\omega\right}.
\label{eq:opportunity-access}$$

The model makes bottlenecks explicit. If every admissible path from a worker’s past contribution to a target opportunity passes through one employer-controlled issuer or evaluator, control is concentrated even when alternative job adverts exist. If an independent credential, public certificate, portfolio, or multiple verifiers creates another path, concentration falls. Referral research shows that network distribution can affect mobility and inequality (Bolte, Immorlica, and Jackson 2026); the graph translates that general concern into a case-audit object without assuming employer control.

Definition 11 (Evidentiary cut set). For worker $B$ and target $\omega$, an evidentiary cut set is a set of evidence, issuer, or verifier nodes whose removal intersects every admissible path from the relevant activity to $\omega$. A minimal cut set is concentrated when one actor controls all of its nodes.

Proposition 12 (Independent-path improvement). Adding an admissible path $p’$ with reliability $R_{p’}$ and latency within the target limit weakly increases $A_B(\omega,t)$, provided existing paths and their qualities remain unchanged.

Proof. Equation [eq:opportunity-access] takes the maximum over admissible paths. Adding $p’$ expands that set. A maximum over a superset is at least the previous maximum. ◻

The result supports plural verification paths. It remains silent about cost, privacy, strategic inflation, and the successor’s acceptance policy. Those conditions enter institutional evaluation.

Residual Responsibility after Termination

This section represents persistence, normative bridging, allocation, and attenuation. The central discipline is separation: a continuing effect is a descriptive state; responsibility requires additional premises; a supported responsibility then requires a typed bearer and remedy.

Post-termination state persistence

This subsection distinguishes relational status from its effects. Let $R_{AB}(t)\in{0,1}$ indicate the formal employment relation, with $R_{AB}(T)=1$ and $R_{AB}(T+1)=0$. Let the transition state be

$$\mathbf L_B(t)=\left\langle
\mathbf X_B(t),\mathsf{RB}_B(t),G_t,\Omega_B(t),
\mathbf c_B(t),P_A^R(t)
\right\rangle .
\label{eq:transition-state}$$

The dynamics are

$$\mathbf L_B(t+1)=
\Phi_t!\left(
\mathbf L_B(t),R_{AB}(t),U_A(t),U_B(t),U_J(t),W(t)
\right),
\label{eq:state-dynamics}$$

where $U_A,U_B,U_J$ represent actions of the former employer, worker, and successor institutions, and $W$ represents background conditions. The update map permits memory, durable credentials, learned skill, reputation, status, network change, and resource depletion.

Proposition 13 (Termination and state persistence separation). The condition $R_{AB}(T+1)=0$ entails neither $\mathbf L_B(T+1)=\mathbf L_B^{\rm baseline}(T+1)$ nor $\boldsymbol\Delta_{\rm PB}=\mathbf 0$.

Proof. Equation [eq:state-dynamics] permits $\mathbf L_B(T+1)$ to depend on $\mathbf L_B(T)$ and on actions taken at or before $T$. Choose a component with persistence, such as an unissued record or durable skill, and a transition map that carries it into $T+1$. Then relation status is zero while the state differs from a comparison baseline. Equation [eq:power-burden] can consequently remain nonzero. ◻

The proposition establishes only logical separation. It leaves the duration, attribution, and normative relevance of the persistent state open.

Three-valued normative bridge

This subsection defines the move from causal profile to candidate responsibility. Let

$$\begin{split}
\widehat{\mathbf B}_{A\to B,T}=\langle{}
&\widehat b^{\rm causal},
\widehat b^{\rm control},
\widehat b^{\rm reliance},
\widehat b^{\rm foreseeable},
\widehat b^{\rm material},
\widehat b^{\rm asymmetric},\
&\widehat b^{\rm feasible},
\widehat b^{\rm proportionate},
\widehat b^{\rm accurate},
\widehat b^{\rm protected},
\widehat b^{\rm allocated}
\rangle\in\mathbb V_3^{11}.
\end{split}
\label{eq:bridge-profile}$$

The causal coordinate refers to a specified burden and mechanism. Control asks whether the actor can prevent or repair it. Reliance records whether the relation reasonably induced the worker to place important career evidence or status under institutional control. Foreseeability concerns the mechanism and risk; precise outcome prediction remains unnecessary. Materiality and asymmetry address impact and comparative capacity. Feasibility and proportionality concern the proposed measure. Accuracy and protected interests preserve successor, coworker, client, privacy, and proprietary concerns. Allocation identifies other responsible institutions.

Define a responsibility report

$$\boldsymbol\rho_{A,t}=
\left\langle
\rho^{\rm bearer},\rho^{\rm object},\rho^{\rm content},
\rho^{\rm strength},\rho^{\rm duration},\rho^{\rm review},
\rho^{\rm allocation}
\right\rangle_t .
\label{eq:responsibility-report}$$

The paper adopts the defeasible bridge

$$\left[
\widehat{\mathbf B}{A\to B,T}=\mathbf 1
\ \wedge
\boldsymbol\Delta{\rm PB}\succcurlyeq\boldsymbol\delta_{\rm material}
\right]
\Rightarrow_{\rm def}
\boldsymbol\rho_{A,T+1}\succ\mathbf 0.
\label{eq:normative-bridge}$$

$\Rightarrow_{\rm def}$ marks a defeasible normative inference whose warrant comes from substantive premises and remains independent of algebra. A zero or unresolved coordinate requires either rejection, qualification, further evidence, or a narrower remedy. Goodin’s vulnerability account and Young’s social-connection model help motivate wider responsibility beyond a completed exchange (Goodin 1985; Young 2011). Equation [eq:normative-bridge] is the author’s proposed restriction on that move.

Attenuation and responsibility allocation

This subsection models temporal and institutional limits. For an initially supported responsibility, let its strength at lag $\tau$ be bounded by

$$0\leq \rho_A^{\rm strength}(T+\tau)
\leq
\rho_A^{\rm strength}(T+1)
,\alpha_A(\tau)
,\kappa_A(\tau),
\label{eq:attenuation}$$

where $\alpha_A(\tau)\in[0,1]$ is the remaining marginal causal contribution and $\kappa_A(\tau)\in[0,1]$ is continuing control or remedy capacity. These terms can decline at different rates. Deliberate maintenance of a bottleneck can keep $\kappa_A$ high; an independently corrected public record can reduce both.

Let $\mathcal D={A,S,U,J,B}$ denote former employer, state, worker organization, successor institution, and worker. For remedy $s$, an allocation vector

$$\boldsymbol\lambda_s=
\langle\lambda_{A,s},\lambda_{S,s},\lambda_{U,s},
\lambda_{J,s},\lambda_{B,s}\rangle,
\qquad
\lambda_{d,s}\geq0,
\quad
\sum_{d\in\mathcal D}\lambda_{d,s}=1
\label{eq:allocation}$$

reports the proposed distribution of implementation burdens. The vector is a governance object distinct from a formula for desert. It forces an account of who holds records, who can verify, who can adjudicate disputes, who finances the system, and who bears privacy or error risk.

Formal Consequences and Identification Limits

This section consolidates the model’s derivations. It states identified sets, observational equivalence, safeguard monotonicity, and the boundary between a causal finding and a normative conclusion. Each result follows from the stated definitions or assumptions; empirical truth remains open.

Identified sets and unresolved status

This subsection gives the reporting rule for incomplete causal knowledge. Let $\mathcal M(h)$ be the set of causal models compatible with the observed data, institutional knowledge, and maintained assumptions at history $h$. Define

$$\mathcal I_{\rm PB}(h)=
\left{
\boldsymbol\Delta_{\rm PB}^{m}(h):m\in\mathcal M(h)
\right}.
\label{eq:identified-set}$$

For a burden direction cone $\mathcal C_+$ and materiality region $\mathcal D_{\rm mat}$, the audit is

$$\widehat{\mathsf{PB}}(h)=
\begin{cases}
1, & \mathcal I_{\rm PB}(h)\subseteq
\mathcal C_+\cap\mathcal D_{\rm mat},\
0, & \mathcal I_{\rm PB}(h)\cap
(\mathcal C_+\cap\mathcal D_{\rm mat})=\varnothing,\
?, & \text{otherwise.}
\end{cases}
\label{eq:burden-audit}$$

The unresolved value prevents a point estimate from representing unsupported precision. It also preserves useful partial results. Investigators can report which burden coordinates have a sign, bound, or sensitivity threshold and which remain dependent on unverified assumptions.

Snapshot observational equivalence

This subsection demonstrates a core identification difficulty. A record delay and a difficult transition observed at one time can arise from ordinary administration, protected-interest review, worker selection, external market change, or strategic obstruction.

Proposition 14 (Snapshot equivalence of transition histories). There exist an ordinary-completion model $m_C$ and a power-produced-burden model $m_P$ that generate the same observed termination date, record-request date, record-issuance date, and successor outcome while producing different values of $\boldsymbol\Delta_{\rm PB}$.

Proof. Let the observed variables be $(T,Q,D,Y)$: termination, request, delay, and successor outcome. In $m_C$, set $D$ through an exogenous verification backlog $U$ and set $Y$ through labor-market scarcity $W$, with the employer-control pathway disabled under both $\pi^O$ and $\pi^C$. Then $\boldsymbol\Delta_{\rm PB}=\mathbf 0$. In $m_P$, let employer action $P_A^R$ produce the same observed $D$ and affect $Y$ through the record pathway. Select the distributions of $(U,W)$ in $m_C$ and latent worker quality and search conditions in $m_P$ so that both models induce the same observed joint distribution of $(T,Q,D,Y)$. Under $m_P$, the intervention that disables $P_A^R$ changes the counterfactual burden, yielding $\boldsymbol\Delta_{\rm PB}\neq\mathbf 0$. The snapshot distribution therefore fails to distinguish the histories. ◻

The result supports event histories, process evidence, policy variation, and mechanism-sensitive measurement. Intention, communications, queue position, field complexity, review grounds, correction histories, successor practices, and labor-market changes may all matter.

Safeguard monotonicity

This subsection identifies a restricted model consequence. Let safeguard $s$ transform the exit profile through $\Psi_s:\mathcal X\to\mathcal X$.

Assumption 15 (Protected coordinate improvement). For the declared target, $\Psi_s(\mathbf X)\succeq\mathbf X$; at least one required coordinate improves materially; the viability threshold remains fixed; and the intervention preserves the protected-interest constraints in equation [eq:claim-audit].

Proposition 16 (Weak effective-exit improvement). Under assumption 15, $E_B(\Psi_s(\mathbf X);z)\geq E_B(\mathbf X;z)$.

Proof. If $E_B(\mathbf X;z)=1$, then $\mathbf X\succeq\mathbf v_B$. Transitivity and $\Psi_s(\mathbf X)\succeq\mathbf X$ give $\Psi_s(\mathbf X)\succeq\mathbf v_B$, so the indicator remains one. If the initial indicator is zero, the transformed indicator belongs to ${0,1}$ and is therefore weakly greater. The protected-interest clause restricts the class of admissible transformations. ◻

Real safeguards still must satisfy the assumption. A credential system can improve record access while increasing privacy risk or decreasing accessibility. A favorable average can hide losses for a vulnerable subgroup. Empirical evaluation must inspect every declared coordinate and population.

Causal and normative separation

This subsection states the argument’s principal stopping rule.

Proposition 17 (Causal finding and responsibility separation). A supported value of $\widehat{\mathsf{PB}}(h)=1$ is insufficient for the responsibility inference in equation [eq:normative-bridge] whenever at least one indispensable bridge premise remains unresolved or unsupported.

Proof. Equation [eq:normative-bridge] has a conjunction containing the full bridge profile and the causal materiality condition. Support for the latter supplies only one conjunct. Classical conjunction introduction requires every premise; the defeasible inference additionally remains open to countervailing conditions. The responsibility conclusion therefore lacks support when an indispensable bridge coordinate is $0$ or $?$, absent a separately defended narrower rule. ◻

This result guards against moralization by equation. It also preserves a practical route forward. A narrower safeguard may have a stronger feasibility, accuracy, and proportionality profile than an expansive transition benefit.

Institutional Safeguards

This section translates the formal objects into a layered institutional portfolio. The sequence follows the mechanism: preserve records, support request and correction, provide plural verification, constrain obstruction, and allocate transition burdens. Each measure retains counterinterests and an evidentiary test.

The narrowest record schema contains identity matching, employment dates, role or occupation, position, and worker-requested fields authorized by governing law or policy. A richer contribution schema can add task type, responsibility, training, license-relevant experience, or contribution roles only where evidence and attribution support them. Every field should carry provenance, issuer, issue time, applicable period, evidentiary basis, confidentiality class, and correction status.

Request-responsive design limits information excess. The worker selects among eligible fields, receives the record in a usable format, and can challenge factual error or omitted requested fields. A correction process should preserve version history and protect the interests of coworkers, clients, and successor institutions. Disputed material can receive a typed status instead of silent deletion or unqualified assertion.

Plural verification reduces concentrated control. Public employment certificates, professional bodies, payroll or social-insurance records, worker-held credentials, independent portfolio verification, and sectoral registries can provide different evidentiary paths. Each path has costs and risks. A public system can reduce employer leverage while expanding state surveillance. A private platform can improve convenience while creating a new bottleneck. Equation [eq:opportunity-access] therefore requires controller and acceptance audits alongside path count.

Table [tab:safeguards] maps mechanisms to candidate measures and safeguards for protected interests. The table organizes a research agenda and supplies no presumption that every listed measure should be adopted.

@p0.18YYY@ Focal mechanism & Candidate measure & Operation evidence & Protected-interest control
Missing basic record & Request-responsive service certificate & Issue time, requested fields, completion rate & Field limits and identity security
Factual error & Notice, evidence access, correction and appeal & Error, correction, reversal, delay & Audit trail and frivolous-claim procedure
Opaque evaluation & Separate factual and evaluative channels & Label accuracy and evidence basis & Evaluator freedom and successor safety
Single-issuer bottleneck & Independent or public verification path & Path acceptance and reliability & Cost, fraud, and governance review
Collective contribution & Typed roles with uncertainty and co-attribution & Contributor confirmation and dispute rates & Coworker credit and project integrity
Confidential work & Minimized role-level representation & Disclosure review and breach monitoring & Privacy, client, security, and trade-secret controls
Covert obstruction & Logged authorized communication and remedy channel & Complaints, findings, recurrence & Due process and legitimate warning
Material transition gap & Targeted notice, record, advice, or training support & Use, outcomes, subgroup distribution & Proportionality and shared financing
Digital credential lock-in & Open format, export, plural wallets, continuity plan & Interoperability, revocation, access & Correlation, surveillance, and exclusion controls

The responsibility allocation in equation [eq:allocation] supports institutional differentiation. Employers are often best positioned to attest their own records. Governments can define minimum fields and adjudication. Worker organizations can advise and aggregate complaints. Successor employers can accept plural forms of proof and specify job-relevant requirements. Workers can request fields, protect credentials, and avoid misleading presentation. Allocation must track capacity and mechanism while preserving a public floor for workers whose former employer has disappeared or refuses cooperation.

Digital implementation requires special caution. W3C credentials can make issuer claims tamper-evident and portable (World Wide Web Consortium 2025); claim validation remains separate. Minimization, selective disclosure, expiry, revocation, offline access, accessibility, multilingual presentation, key recovery, and institutional continuity belong to the design. A paper alternative can remain essential.

Claim 18 (Objective-record priority). Where the causal and normative bridge supports intervention, an accurate, request-responsive, contestable, and constrained objective record is generally a more limited initial measure than compelled favorable evaluation or broad career support. The claim is a defeasible institutional priority whose force depends on domain evidence, cost, and protected interests.

Employment Configurations

This section tests the framework against six stylized configurations. The cases separate causal pathways and demonstrate that difficult departure, dependence, and employer control can receive different audit outcomes.

Table [tab:cases] presents the comparison. The examples are constructed countermodels and supply no findings about any named employer or jurisdiction.

@p0.18YYY@ Configuration & Transition source & Record condition & Audit direction
Ordinary search friction & Labor-market scarcity and search time & Basic facts issued promptly & Effective exit may be low; employer-produced burden unsupported
Joint specialization & Informed investment in firm-specific practice & Limits arise from activity type & Dependency exists; attribution and remedy remain contextual
Basic-record withholding & Employer controls unique authoritative dates and role & Requested eligible facts delayed or refused & Candidate power-produced burden; objective-record remedy salient
Covert obstruction & Relation-derived communication impedes successor access & Formal certificate coexists with hidden signal & Causal, intent, authorization, and due-process inquiry required
Collective project & Individual contribution is interdependent and disputed & Rich attribution risks coworker erasure & Basic record supported; rich contribution remains unresolved
Safety-sensitive role & Successor requires material performance information & Accurate adverse evaluation has evidence & Objective record and fair warning can coexist
Employer closure & Issuer disappears after termination & Records become institutionally orphaned & Public or third-party continuity mechanism may carry responsibility

Ordinary friction and joint specialization

This subsection establishes the main countermodels. Legal exit may coexist with high costs because the labor market is thin, the worker prefers a narrow region, or a skill is valuable mainly within one organization. Labor-market power can also arise from market concentration and search friction (Bhaskar, Manning, and To 2002; Berger, Herkenhoff, and Mongey 2022). These conditions can support public labor-market policy. They leave the bilateral $\boldsymbol\Delta_{\rm PB}$ at zero when the former employer followed the coherent completion regime.

Joint specialization creates a harder case. Employer and worker may knowingly invest in a valuable relation-specific practice. Departure then destroys part of a genuine joint surplus. The resulting dependence can be legitimate, regrettable, or unjust depending on authorization, distribution, alternatives, governance, and conduct. Objective records should describe transferable roles accurately while acknowledging the limits of individual attribution.

Withholding and covert obstruction

This subsection represents the focal cases. Withholding occurs when an employer controls an authoritative eligible record and delays or refuses it relative to a coherent completion regime. Covert obstruction adds a communication or signal intended or foreseeably likely to impair successor access. The second case can present stronger concerns, yet it also creates heavier evidentiary and due-process requirements.

Japanese Article 22 provides a concrete legal fragment: a worker can request specified facts, unrequested fields face restriction, and paragraph 4 addresses specified secret communications intended to impede later employment (Government of Japan 2025). The manuscript makes no inference about liability in a particular case. It uses the provision to show that factual attestation, worker-directed scope, and obstruction can be legally differentiated.

Collective work and safety-sensitive evaluation

This subsection preserves third-party and public interests. Many achievements are collective. An individual record can overstate causal contribution, reveal colleague performance, or appropriate shared output. The portable set in equation [eq:portable-set] therefore admits uncertainty, co-attribution, role-level description, and exclusion.

Some successor roles implicate substantial safety, fiduciary, professional, or client interests. A former employer can possess relevant evidence of serious misconduct or deficient competence. A system that provides only flattering or sterile records can shift risk to others. The proper distinction lies between fair, relevant, evidenced, and authorized evaluation and strategic obstruction. Acas guidance similarly emphasizes accuracy, fairness, relevance, and evidence for opinion (Advisory, Conciliation and Arbitration Service 2024).

Empirical Research Programme

This section converts the framework into a research programme. It specifies units, measures, designs, and adverse-effect monitoring. The initial empirical goal is mechanism identification within one jurisdiction and sector, followed by cautious comparison.

The unit of analysis is a worker–employer–successor transition episode. The episode begins before notice and continues through a declared follow-up period. Administrative records should be linked, with lawful safeguards, to worker reports, record requests, employer processes, successor decisions, market conditions, and dispute outcomes. Qualitative process tracing is essential for interpreting coded events and discovering mechanisms absent from the initial schema.

Table [tab:research] states a minimum design ledger. Its role is to prevent outcomes from appearing without the mechanism, counterfactual, distribution, and protected-interest data needed for interpretation.

@p0.16YYY@ Layer & Administrative measure & Participant or institutional evidence & Principal threat
Formal exit & Notice, contract, termination date, disputes & Legal understanding and perceived feasibility & Misclassification and jurisdictional variation
Material transition & Income, benefits, housing, search duration & Household constraints and acceptable targets & Unobserved preference and macroeconomic change
Record process & Request, fields, evidence, issue time, correction & Usability, omission, reason for request & Selection into requesting
Control pathway & Decision authority, communication, bottleneck structure & Perceived threat, advice, alternative issuers & Hidden communication and common causes
Successor access & Applications, interviews, offers, wages, role quality & Employer interpretation and rejection reason & Selective application and missing outcomes
Protected interests & Privacy, trade-secret, safety, coworker disputes & Affected-party assessments & Underreported harms
Safeguard use & Access, timing, appeal, verification path & Accessibility, trust, burden, strategic response & Implementation fidelity
Distribution & Outcomes by status and vulnerability dimensions & Participatory interpretation & Small cells and privacy risk
Long-term trajectory & Retention, mobility, correction, credential continuity & Reinterpretation and new goals & Attrition and changing targets

Several research designs are plausible. A phased introduction of a standardized certificate or correction process can support difference-in-differences or stepped evaluation when timing, trends, spillovers, and implementation permit. Randomized reminders concerning an existing right can measure access while preserving entitlement across groups. Queue discontinuities or administrative processing reforms can create quasi-experimental variation. A target-trial emulation can compare defined request and issuance strategies if longitudinal confounding is adequately measured (Hernán and Robins 2020).

Record portability also requires mechanism-specific outcomes. Kässi and Lehdonvirta found that voluntary digital skill certification on one online labor platform reduced employer uncertainty and raised the value of obtained projects for certified workers (Kässi and Lehdonvirta 2019). That study supplies evidence that standardized signals can matter. Its platform, selection, and certification context limits transfer to ordinary employment.

The evaluation should measure adverse effects. Employers may shift from logged written records to informal calls. Workers may optimize visible fields. Credentials may become de facto requirements and burden people with incomplete histories. Rich data may expose protected information or enable automated exclusion. Standardization may erase locally meaningful work. Successor firms may over-trust issuer prestige. These outcomes belong in the primary evaluation; appendix-only treatment would understate their importance.

Open Problem 19 (Causal identification of record control). Develop and validate a design that separates record withholding or obstructive communication from ordinary verification time, worker selection, employer risk, external labor-market conditions, and time-varying health or household constraints.

Open Problem 20 (Portable contribution validity). Construct sector-specific contribution fields with adequate truth, inter-rater reliability, collective attribution, predictive validity, correction, privacy, and resistance to strategic gaming.

Objections and Scope Conditions

This section subjects the position to its strongest contractual, epistemic, expressive, institutional, and temporal objections. The replies narrow the proposal and identify unresolved work.

Completed exchange and career insurance

This subsection addresses the baseline objection. Employment ordinarily exchanges labor for compensation under bounded terms. After valid resignation, the former employer should remain free from an open-ended duty to secure the worker’s career. This concern has considerable force. The paper’s response has three limits. It focuses on relation-produced or relation-controlled burdens. It requires materiality and feasible proportional mitigation. It gives priority to a truthful objective record over income replacement, placement, or praise.

A broader transition responsibility can arise in restructuring, dismissal, status sponsorship, or deliberately induced reliance. Those domains need separate arguments. The present framework supplies variables and gates without prejudging them.

Expression, truth, and successor risk

This subsection addresses references as speech and judgment. A manager can have legitimate reasons to withhold endorsement. A successor has legitimate reasons to receive relevant safety or performance information. Compelled optimism would corrupt both. Proposition 10 therefore carries substantive importance. A factual record and an evaluative reference occupy different channels. Evaluation should remain fair, evidenced, relevant, and appropriately authorized; objective attestation should remain accurate, selected, and contestable.

The separation still leaves borderline claims. “Led a team of twelve” can be a verifiable role fact. “Led effectively” adds evaluation. “Produced the model” may conceal collective authorship. A sectoral schema should classify these cases and preserve dispute status.

Privacy, secrets, and collective production

This subsection addresses disclosure constraints. Rich employment records can reveal client identity, medical information, security practice, proprietary methods, strategy, or coworker performance. Japanese trade-secret law and data-protection regimes impose real boundaries (Government of Japan 2023; European Parliament and Council of the European Union 2016). The portability set incorporates minimization and protected-interest constraints before a claim becomes eligible.

Protective constraints can themselves become pretexts for erasure. A procedure should identify the protected interest, explain the minimum necessary restriction where lawful, and offer a more abstract attestation when possible. Independent review can distinguish a genuine secret from a blanket label.

Causal ambiguity and strategic response

This subsection addresses inference. Exit is selected, dynamic, and embedded in labor-market change. Workers who anticipate trouble may request more records; employers facing complex cases may take longer; successor outcomes affect later reports. Proposition 14 proves that a narrow observational signature can fit opposing histories. Longitudinal methods improve discipline while leaving unmeasured confounding and interference possible.

Strategic response can also defeat the remedy. Firms can provide boilerplate, shift to unofficial signals, avoid formal training, or encode hierarchy through issuer prestige. Workers can inflate roles or select only favorable fields. Successor firms can make a formerly optional credential mandatory. Pilot design must therefore measure equilibrium response and distribution alongside issuance rates.

Duration and distributed responsibility

This subsection addresses the temporal boundary. Persistent effects could appear to support indefinite obligation. Equation [eq:attenuation] restricts responsibility through remaining marginal contribution and control. A former employer that issued an accurate record and corrected its error may have little continuing role. A public institution can become the proper archive after firm dissolution. A successor can bear responsibility for accepting equivalent proof. Worker organizations can provide advice and contestation.

The allocation remains normative and institutional. Young’s social-connection model supports attention to distributed forward-looking responsibility (Young 2011); the paper’s vector prevents that insight from becoming an undifferentiated collective duty.

Legal diversity and publication status

This subsection states the doctrinal limit. Article 22 provides a valuable Japanese anchor, and ILO Recommendation No. 166 supplies an international standard with a service-certificate distinction (Government of Japan 2025; International Labour Organization 1982). Legal rights, worker coverage, remedies, evidentiary rules, data duties, and limitation periods vary. The manuscript is a philosophical and formal position paper with a bounded doctrinal illustration. Jurisdiction-specific conclusions require specialist research.

Future Directions

This section converts the framework’s limitations into a sequenced research agenda. The sequence begins with Japanese doctrine and one-sector evidence, then expands to comparison, technology, and adjacent relations.

First, the Japanese legal pass should examine authoritative commentary, administrative guidance, enforcement, case law, worker coverage, Article 22(4) elements, evidentiary access, remedies, and interactions with privacy, trade secrets, defamation, discrimination, and sector regulation. Worker-side, employer-side, regulator, union, and specialist perspectives should be included.

Second, one occupational sector should receive a longitudinal study. A suitable sector would combine meaningful record dependence, repeated transitions, observable contribution fields, and manageable safety concerns. Participatory design should establish the target transition and viability thresholds before quantitative estimation.

Third, comparative law should study employment certificates and references in jurisdictions with materially different institutions. The comparison should code factual fields, worker selection, evaluation, correction, timing, enforcement, blacklisting protections, digital format, and acceptance. A ranking of jurisdictions would be premature before institutional context and implementation evidence are understood.

Fourth, a credential pilot can test the record-bundle architecture in equation [eq:record-bundle]. Open formats, selective disclosure, public or plural issuers, offline access, correction, revocation, and long-term continuity are essential. Technical verification should be evaluated separately from truth, fairness, and opportunity effects.

Fifth, adjacent papers can examine platform deactivation, migrant status tied to sponsors, academic supervision and authorship, non-compete clauses, professional licensing, volunteer service, and organizational closure. These domains change the bearer of power and the content of the portable object. They therefore require domain-specific specialization; automatic inheritance of the present conclusions would exceed the argument.

Sixth, the wider generative-justice programme should examine when portability supports diversity and when it imposes a dominant ontology. A standardized record can make contribution legible across settings. It can also erase collective, tacit, relational, culturally situated, or emergent forms of value. Plural schemas and translation procedures may be required.

Open Problem 21 (Residual-responsibility threshold). Defend a context-sensitive rule for the materiality, causal contribution, continuing control, induced reliance, and proportionality needed to support a post-relation responsibility, together with principled attenuation and allocation across institutions.

Open Problem 22 (Portability and ontological pluralism). Develop a record system that transfers usable evidence across institutions while preserving collective attribution, uncertainty, local meaning, and forms of contribution that resist a single standardized vocabulary.

Conclusion

This section consolidates the conditional position and its stopping rules. The paper begins with the value of formal resignation and extends analysis only as far as a specified employment transition and its evidence pathways.

Formal exit is real and important. Effective employment exit adds a target- relative profile of material, status, relational, record, opportunity, and safety conditions. A low profile can arise from ordinary friction, joint specialization, market power, public institutions, personal circumstances, or relation-derived conduct. Attribution therefore requires a coherent completion regime and an identified causal contrast.

Objective-record portability names a bounded institutional object. It concerns truthful, attributable, verifiable, requested, relevant, minimized, and usable claims. Favorable evaluation has a different type. Privacy, trade secrets, collective contribution, security, and third-party interests constrain both content and format. Japanese Article 22, ILO Recommendation No. 166, British reference guidance, data portability, and credential standards each contribute one part of this architecture while leaving the paper’s wider proposal open.

Termination ends a formal relation while some state changes persist. A power-produced burden consequently can continue after the last day of work. Residual responsibility follows only through additional premises concerning control, reliance, foreseeability, materiality, asymmetry, feasible mitigation, proportionality, accuracy, protected interests, and allocation. Its content can be narrow, shared, declining, and reviewable.

The immediate research priority is a bounded Japanese doctrinal review and a one-sector longitudinal study of record requests, issuance, correction, successor interpretation, and transition outcomes. The immediate institutional priority is more modest: test whether accurate and contestable objective records can widen evidentiary paths without creating new surveillance, standardization, or exclusion burdens. The resulting framework remains revisable and conditional on evidence.

Acknowledgments

The author thanks Ron Eglash for dialogue that helped motivate the broader research programme on generative justice and relational value circulation. The present definitions, formal constructions, distinctions, arguments, and conclusions are the author’s responsibility. The author also acknowledges the verified scholarship and institutional materials on exit and voice, power-dependence, workplace authority, relational autonomy, domination, vulnerability, responsibility, labor mobility, employment references, labor law, data portability, and verifiable credentials that make the present proposal and its limits intelligible.

Advisory, Conciliation and Arbitration Service. 2024. “What Employers Can Say in a Reference.” https://www.acas.org.uk/providing-a-job-reference/what-employers-can-say-in-a-reference.

Anderson, Elizabeth. 2017. Private Government: How Employers Rule Our Lives (and Why We Don’t Talk about It). Princeton, NJ: Princeton University Press. https://doi.org/10.1515/9781400887781.

Berger, David, Kyle Herkenhoff, and Simon Mongey. 2022. “Labor Market Power.” American Economic Review 112 (4): 1147–93. https://doi.org/10.1257/aer.20191521.

Bhaskar, V., Alan Manning, and Ted To. 2002. “Oligopsony and Monopsonistic Competition in Labor Markets.” Journal of Economic Perspectives 16 (2): 155–74. https://doi.org/10.1257/0895330027300.

Bolte, Lukas, Nicole Immorlica, and Matthew O. Jackson. 2026. “The Role of Referrals in Immobility, Inequality, and Inefficiency in Labor Markets.” Journal of Labor Economics 44 (2). https://doi.org/10.1086/733048.

Eglash, Ron. 2016. “Of Marx and Makers: An Historical Perspective on Generative Justice.” Teknokultura. Revista de Cultura Digital y Movimientos Sociales 13 (1): 245–69. https://doi.org/10.5209/rev_TK.2016.v13.n1.52096.

Eglash, Ron, Kwame P. Robinson, Audrey Bennett, Lionel Robert, and Mathew Garvin. 2024. “Computational Reparations as Generative Justice: Decolonial Transitions to Unalienated Circular Value Flow.” Big Data & Society 11 (1). https://doi.org/10.1177/20539517231221732.

Emerson, Richard M. 1962. “Power-Dependence Relations.” American Sociological Review 27 (1): 31–41. https://doi.org/10.2307/2089716.

European Parliament and Council of the European Union. 2016. “Regulation (EU) 2016/679, Article 20: Right to Data Portability.” Official Journal of the European Union, L 119. https://eur-lex.europa.eu/legal-content/EN-FR/TXT/?uri=CELEX:32016R0679.

Goodin, Robert E. 1985. Protecting the Vulnerable: A Re-Analysis of Our Social Responsibilities. Chicago: University of Chicago Press. https://press.uchicago.edu/ucp/books/book/chicago/P/bo5974942.html.

Government of Japan. 2023. “Unfair Competition Prevention Act, Act No. 47 of 1993, Version of Act No. 51 of 2023.” Japanese Law Translation Database System. https://www.japaneselawtranslation.go.jp/en/laws/view/4709/en.

———. 2025. “Labor Standards Act, Act No. 49 of 1947, Article 22.” e-Gov Laws and Regulations Search, version in force 1 June 2025. https://laws.e-gov.go.jp/document?law_unique_id=322AC0000000049.

Häfner, Samuel, and Curtis R. Taylor. 2023. “Working for References.” American Economic Journal: Microeconomics 15 (3): 33–77. https://doi.org/10.1257/mic.20210299.

Hernán, Miguel A., and James M. Robins. 2020. Causal Inference: What If. Boca Raton, FL: Chapman & Hall/CRC. https://www.hsph.harvard.edu/miguel-hernan/wp-content/uploads/sites/1268/2024/04/hernanrobins_WhatIf_26apr24.pdf.

Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. https://wellcomecollection.org/works/kfqxpurh.

International Labour Organization. 1982. “Termination of Employment Recommendation, 1982 (No. 166).” NORMLEX. https://normlex.ilo.org/dyn/nrmlx_en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:R166.

Kässi, Otto, and Vili Lehdonvirta. 2019. “Do Digital Skill Certificates Help New Workers Enter the Market? Evidence from an Online Labour Platform.” OECD Social, Employment and Migration Working Papers 225. Paris: OECD Publishing. https://doi.org/10.1787/3388385e-en.

Lovett, Frank. 2010. A General Theory of Domination and Justice. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199579419.001.0001.

Mackenzie, Catriona, and Natalie Stoljar, eds. 2000. Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self. New York: Oxford University Press. https://doi.org/10.1093/oso/9780195123333.001.0001.

Ministry of Health, Labour and Welfare, Japan. 2025. “Explanations of Labor-Related Laws.” Official English explanatory guide. https://hi-elcc.mhlw.go.jp/en/pdf/Explanations_of_labor_related_laws.pdf.

National Information Standards Organization. 2022. “CRediT, Contributor Roles Taxonomy.” National Information Standards Organization. https://www.niso.org/publications/z39104-2022-credit.

Organisation for Economic Co-operation and Development. 2025. “Reviving Growth in a Time of Workforce Ageing: The Role of Job Mobility.” In OECD Employment Outlook 2025. Paris: OECD Publishing. https://doi.org/10.1787/194a947b-en.

Spence, Michael. 1973. “Job Market Signaling.” The Quarterly Journal of Economics 87 (3): 355–74. https://doi.org/10.2307/1882010.

World Wide Web Consortium. 2025. “Verifiable Credentials Data Model V2.0.” World Wide Web Consortium. https://www.w3.org/TR/vc-data-model-2.0/.

Young, Iris Marion. 2011. Responsibility for Justice. New York: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195392388.001.0001.