--- title: "The Capability Trap" subtitle: "When building runway cuts support first" author: Elias Kunnas description: "A last-resort benefit becomes a capability trap when every attempt to build runway—saving, earning, simplifying household risk—reduces support before the new capability is stable." canonical: https://kunnas.com/articles/capability-trap url: https://kunnas.com/articles/capability-trap.md date_published: 2026-01-09 date_modified: 2026-08-31 corpus_frame_url: https://kunnas.com/articles/how-to-read-this.md --- ## How to read this corpus The corpus applies one lens to many domains: what mechanisms produce the outcome? It shares four methodological commitments and one explicit directional commitment. Each linked page argues for its part; the links are derivations and disputes, not evidence inherited by every page. The directional commitment does not by itself settle system boundary, distribution, sacrifice, or institutional authority. 1. **Mechanisms are what act.** Incentive gradients, selection pressures, feedback loops, and capital stocks produce the distribution of outcomes. Intentions, labels, official categories, and stated values are evidence about mechanisms, or are themselves coordination mechanisms. They are not causal substitutes. — [Mechanism Realism](https://kunnas.com/articles/mechanism-realism.md) · [Only Selection](https://kunnas.com/articles/only-selection.md) 2. **The reference telos is sustained flourishing.** The broadest achievable adaptive safety margin over deep time — not the continuity of any incumbent state, coalition, institution, or doctrine. A mechanism's own stated goal can still serve as a local proof obligation — showing that its incentives defeat even the purpose it claims is a bounded finding — but meeting that goal establishes nothing about the margin. — [Flourishing Is Maximum Safety Margin](https://kunnas.com/articles/flourishing-is-maximum-safety-margin.md) 3. **Law, rights, legitimacy, democracy, markets, and sovereignty are mechanisms under evaluation.** They are constraints, carriers, or proxies inside the analysis. None is a terminal value or a boundary of what is real. Treating one as terminal ends the mechanism search before it starts. Evaluation carries current function, replacement cost, path dependence, uncertainty, capture risk, reversibility, and who bears model error into the ledger. — [The Stack](https://kunnas.com/articles/the-stack.md) · [Mechanism Space](https://kunnas.com/articles/mechanism-space.md) 4. **Optimization is a system function.** A civilization has to build, exercise, and revise metamechanisms that search mechanism-space, discard dominated options, install, observe effects, and repair under uncertainty. Not running that loop leaves margin unrealized, and that is itself the failure. No single component — analyst, model, or institution — is presumed to contain a global optimum; the capacity is a property of the system. — [From Telos to Policy](https://kunnas.com/articles/from-telos-to-policy.md) · [The Three-Layer Architecture](https://kunnas.com/articles/three-layer-architecture.md) 5. **Uncertainty is preserved, not spent.** Partial orders, binding constraints, unknowns, and residuals stay explicit. An unmeasured effect is not a favorable default. — [The Compression Paradox](https://kunnas.com/articles/compression-paradox.md) · [Cargo Cult Epistemology](https://kunnas.com/articles/cargo-cult-epistemology.md) *Each essay bears its own evidence. Links carry definitions, derivations, applications, and disputes; they do not transfer proof. Criticism is answered on its substance.* Canonical: --- # The Capability Trap *When building runway cuts support first* Elias Kunnas ## Thesis {#thesis} A last-resort benefit becomes a capability trap when every attempt to build runway—saving, earning, simplifying household risk—reduces support before the new capability is stable. ## Standard objections addressed in this essay - “The Finnish basic-income experiment did not test basic social assistance alone.” — [§II](#ii-the-experiment-that-proved-everything) (Treatment and control groups differed across a wider unemployment-benefit system.) - “Small employment effects do not prove capability was the binding constraint.” — [§II](#ii-the-experiment-that-proved-everything) (Health, demand, eligibility, timing, and treatment design remain rival explanations.) - “A 100% effective marginal tax rate is not universal.” — [§I](#i-the-system-creates-the-behaviors-it-pathologizes) (Rates vary by earnings, housing, taxes, benefits, and reform date.) - “Asset rules and car treatment are more conditional than stated.” — [§I](#i-the-system-creates-the-behaviors-it-pathologizes) (Current legal rules and administrative practice need precise sourcing.) - “Housing First does not identify the best design for general income support.” — [§III](#irony) (It is a capability-platform analogy across different populations and outcomes.) - “Trajectory data does not show the system caused dependency.” — [§V](#v-how-the-system-traps) (Selection, prior conditions, and treatment effects must be separated.) - “‘The government chose incapacity’ imputes intent without evidence.” — [§II](#ii-the-experiment-that-proved-everything), [§III](#irony), [§IV](#iv-the-kuusi-logic)(The mechanism emerges from local objectives and institutional constraints.) --- ## I. The System Creates the Behaviors It Pathologizes {#i-the-system-creates-the-behaviors-it-pathologizes} We have names for the behaviors we see in long-term welfare recipients: learned helplessness, poverty mindset, avoidance, shame spirals around money, paralysis when facing bureaucracy. What if these behaviors are rational responses to the actual rules? *Toimeentulotuki* is a last-resort benefit. Kela counts available income and assets against eligible expenditure. Savings, cash, and readily realizable investments generally affect the calculation. A car does not count as an available asset where it is necessary for work, livelihood, or mobility. These rules do not establish a universal asset threshold or a categorical ban on vehicle ownership. From 1 February 2026, the ordinary €150 monthly earnings disregard was removed for adults; earnings therefore reduce this benefit calculation in full, subject to stated exceptions for carers' fees through 2027. The effective marginal rate on a person's total resources still varies with taxes, housing support, other benefits, and household circumstances. These rules discourage saving, formal earnings, and transparent household arrangements at the benefit margin. The magnitude of each response requires evidence; the rules do not show that every recipient adopts it. **Behavior described as a "poverty mindset" sometimes reflects the benefit system's constraints.** Administrative burden, means testing, and earnings withdrawal add stress and reduce room for experimentation; their causal contribution to shame, paralysis, and avoidance requires direct evidence. ## II. What the Basic Income Experiment Tested {#ii-the-experiment-that-proved-everything} The Finnish Basic Income Experiment (2017–2018) assigned 2,000 unemployed people to an unconditional €560 monthly payment. It tested a wider unemployment-benefit treatment, not basic social assistance alone. The control group remained in the ordinary unemployment-benefit system, so the comparison also reflects differences in eligibility, administration, and benefit design. If earnings withdrawal were the binding constraint for this population, the treatment would predict a larger employment effect. That prediction does not identify the constraint in advance. **Results:** - Employment: +6 days/year (barely significant) - Self-reported wellbeing: higher in the treatment group - Trust: higher in the treatment group **The employment effect was limited in this experiment.** That result does not identify a single binding constraint; health, demand, eligibility, timing, and treatment design remain rival explanations. Capability is one plausible barrier alongside health, skills, local labor demand, eligibility, and timing. The experiment does not establish it as the binding constraint. Asset liquidation reduces financial resilience. Administrative complexity consumes time and attention. Full earnings withdrawal within the benefit calculation weakens the immediate gain from formal earnings. These are candidate mechanisms, not a measured system-wide causal account. The experiment ended when its legislated two-year term expired. Subsequent policy choices require separate evaluation against their stated objectives and observed outcomes. The limited employment result and reported wellbeing differences do not settle whether conditionality is necessary, or which component of the treatment produced each outcome. ## III. The Finnish Irony {#irony} Housing First offers a design contrast, not evidence that general income support faces the same problem or that an actor chose incapacity. **Housing First** (*Asunto ensin*) is Finland's internationally recognized homelessness policy. It provides permanent housing without requiring sobriety or employment first, then wraps support services around the housing. Finnish homelessness has fallen substantially over recent decades. That outcome concerns a distinct population and outcome from general income support. The logic: poverty and homelessness are multi-constraint traps. You cannot address addiction while sleeping on the street. You cannot job-search without a stable address. You cannot build capability without a platform to build from. Housing First provides the platform unconditionally, then builds capability on top of it. The comparison is architectural, not a direct policy ranking. Housing First establishes a stable platform before demanding change; *toimeentulotuki* assesses need and available resources before payment. **The design contrast:** Housing First treats housing as a platform for subsequent support. Whether general income support should use an analogous platform requires separate evidence on costs, incentives, and outcomes. Housing First's reported success does not identify a single cause. Visibility, measurable housing outcomes, a narrower objective, service design, and institutional sponsorship are competing explanations. ### The Deeper Problem: Timing Even Housing First is late intervention. By the time someone is homeless—or on toimeentulotuki—capability has already been destroyed. Heckman's research shows intervention returns decrease with age: ~7-10% annually for early childhood, ~0-2% for adult job training. By age 25, you're rebuilding what should have been built by age 5. (For the full timing analysis and intervention hierarchy, see [Capability Architecture](capability-architecture.md).) Universal services and targeted support address different risks. The trajectory study does not establish where a capability deficit began or whether any particular service failed to prevent it. Housing First demonstrates that Finnish institutions can sustain a distinct homelessness strategy. It does not prove that the same design transfers to general income support. ## IV. The Kuusi Logic {#iv-the-kuusi-logic} How did this institutional configuration emerge? Pekka Kuusi, architect of Finnish social policy in the 1960s, explicitly rejected the term *hyvinvointivaltio* (welfare state). His metaphor for what he wanted to avoid: a "cowhouse" where people are kept alive but not productive. Kuusi's vision: the state invests in people so they return to the labor market. Social support was productivity investment, not charity. "The state supports you so that you may work, not so that you may exist outside the labor market." Section 19 of the Constitution distinguishes indispensable subsistence from a right to basic subsistence. That distinction does not itself prescribe a single means-testing design or establish why present rules took their current form. The 1990s depression cemented this further. "Incentive trap" discourse dominated. *Kestävyysvaje* (fiscal sustainability gap) became permanent anxiety. Activation über alles. Work-oriented social-policy traditions are one influence on current policy. Fiscal constraints, administrative design, and later political choices also shape the present system. ## V. How the System Traps {#v-how-the-system-traps} Finnish longitudinal data (cohorts born 1981-1987, followed ages 19-25) identified six trajectories through social assistance: | Trajectory | % | Description | |----------------|----------|-------------------------------| | No Receipt | 70.1% | Never need assistance | | Transitory | 9.3% | Brief receipt, permanent exit | | Slow Exit | 4.6% | Gradual decrease | | **Occasional** | **8.9%** | **Revolving door** | | **Increase** | **3.2%** | **Getting worse over time** | | **Dependency** | **3.8%** | **Near-permanent receipt** | The study describes different receipt trajectories. It does not show that the support system caused persistence, deterioration, or dependency: selection, prior conditions, labor-market conditions, and treatment effects remain distinct explanations. **Duration dependence:** Longer receipt correlates with lower exit probability. Causal pathways could include selection, skill atrophy, stigma, health, labor demand, and administrative experience; the trajectory data does not separate them. The 2017 Kela reform centralized basic social assistance. The National Audit Office reported that it weakened links between financial assistance and social work. That finding identifies a service-integration problem; it does not establish that the reform was designed to avoid blame. The 2020 National Audit Office report found that the reform severed links between financial aid and social work. Non-take-up and administrative burden warrant measurement; their scale and causes cannot be inferred from that finding alone. ## VI. The 2026 Reform {#vi-the-2026-reform} *Yleistuki* began in May 2026, replacing Kela's labour-market subsidy and basic unemployment allowance. Kela states that assets, a spouse's income, and a cohabitant's income do not affect its amount. *Toimeentulotuki* remains the last-resort benefit, calculated against available income, assets, and eligible expenditure. From February 2026, the *toimeentulotuki* reform removed the ordinary adult earnings disregard, reduced adult basic amounts by 2–3%, and expanded obligations to seek primary benefits and full-time work. The reform's distributional and capability effects are questions for outcome data, not conclusions established at enactment. The resulting system has different eligibility and means-testing architectures across benefits. Whether this creates a capability trap depends on observed transitions, household resources, administrative burden, and employment outcomes. ## VII. Terminal Drift {#vii-terminal-drift} No one decided "let's destroy capability." Political, bureaucratic, fiscal, and cultural pressures each optimize local objectives—visible toughness, audit-proof documentation, this year's spending, engagement, perceived fairness. Each pressure is locally rational. The joint effect can consume the assets, attention, and marginal gains from work that recipients need to build runway—before new capability is stable enough to survive without support. **This is terminal drift:** operational telos diverges from stated telos. The welfare system performs "helping the poor" while rules at the margin penalize saving, earning, and simplifying household risk. The ritual of application, adjudication, and activation can look complete on paper while the measurable runway keeps shrinking. The same pattern that produces [theatrical accountability](theatrical-accountability.md) in professional discipline can produce theatrical welfare in social policy: process that satisfies oversight without changing recurrence. The basic-income experiment supplies one bounded result, not a complete test of capability-oriented policy. A serious assessment asks whether later rules improve transitions, wellbeing, administrative burden, and durable earnings for comparable recipients. ## VIII. Conclusion {#viii-conclusion} Finnish income support combines objectives that can conflict: income security, fiscal restraint, administrative consistency, activation, and public legitimacy. The capability question is whether their joint operation preserves the resources recipients need to exit precarity. Means testing and activation rules create perverse incentives at the margin. Establishing a feedback loop from those incentives to dependency requires causal evidence beyond the existence of persistent receipt. Kuusi's terminology does not settle how to classify the Finnish welfare state. The narrower claim is institutional: a last-resort system structured around means testing and activation faces a recurring tension between control and capability. Non-take-up, persistent receipt, and financial stress each require separate measurement. They reflect some combination of eligibility, information, administrative burden, health, labor-market conditions, household resources, and benefit design. The central test is direct: measure whether benefit design preserves or consumes a recipient's assets, attention, and short-term gains from formal work. --- *Capability & Selection series: **Diagnostic** → [Prescriptive](capability-architecture.md) → [Selection](selection-question.md) → [Institutional](fourth-branch.md)* **Related:** - [Sterile Generativity](sterile-generativity.md) — the cross-domain primitive: welfare output preserved (benefit paid, case closed), generator consumed (recipient's capability). The capability trap is the welfare-system specialization of this pattern. - [Theatrical Accountability](theatrical-accountability.md) — parallel sibling in the institutional-governance domain: form preserved, consequence-producing generator consumed. - [The Physics of Moloch](physics-of-moloch.md) — How selection and proxy drift can settle into stable traps ## Sources and Notes **Finnish Basic Income experiment:** Kangas et al., "The Basic Income Experiment 2017-2018 in Finland," Ministry of Social Affairs and Health (2019). The experiment ran its full legislated two-year course (January 2017 – December 2018). 2,000 unemployed individuals received €560/month unconditionally. Employment: +6 days/year in 2018. Wellbeing: 33% fewer with poor mental health (MHI-5); life satisfaction 7.32 vs 6.76 (p \< 0.01). See also Hämäläinen et al., VATT Institute; Max Planck Institute analysis (2025). **Finnish welfare trajectories:** Salonen & Saikkonen (2018), "Social Assistance Trajectories Among Young Adults in Finland"—longitudinal register study identifying six trajectory types including persistent dependency (~4%). **Current *toimeentulotuki* rules:** Kela, [“Tulot ja varat vaikuttavat toimeentulotukeen”](https://www.kela.fi/toimeentulotuki-tulot-ja-varat) (accessed August 2026). Available income, savings, cash, and readily realizable investments generally affect the calculation. A car is excluded where it is necessary for work, livelihood, or mobility. **2026 changes and earnings treatment:** Kela, [“Muutoksia Kelan etuuksiin vuonna 2026”](https://www.kela.fi/ajankohtaista/muutoksia-kelan-etuuksiin-vuonna-2026) (accessed August 2026). From 1 February 2026, the ordinary adult €150 earnings disregard was removed; earnings affect the *toimeentulotuki* calculation in full, subject to stated exceptions. Kela, [“Miten toimeentulotuen määrä lasketaan?”](https://www.kela.fi/toimeentulotuki-miten-lasketaan) (accessed August 2026), explains that the benefit covers eligible expenditure remaining after income and assets. ***Yleistuki*:** Kela, [“Muutoksia Kelan etuuksiin vuonna 2026”](https://www.kela.fi/ajankohtaista/muutoksia-kelan-etuuksiin-vuonna-2026) (accessed August 2026). The benefit began in May 2026 and replaced Kela's labour-market subsidy and basic unemployment allowance; Kela states that assets and a spouse's or cohabitant's income do not affect its amount. **Early childhood investment returns:** [Heckman, "Skill Formation and the Economics of Investing in Disadvantaged Children,"](https://heckmanequation.org/) *Science* (2006)—7-10% annual return for early intervention vs. ~0-2% for adult job training. Perry Preschool and Abecedarian longitudinal data. **Finnish Housing First:** Pleace et al. (2015), "The Finnish Homelessness Strategy," European Observatory on Homelessness. Homeless population reduced from ~18,000 (1987) to ~3,800 (2024). Y-Foundation model. **Finnish welfare ideology:** Kuusi, Pekka (1961), *60-luvun sosiaalipolitiikka*—founding document emphasizing investment in human capital, later encoded as punitive means-testing. Lutheran work ethic influence: Markkola (2000) on Protestant moral framework in Nordic welfare. **Paradox of Redistribution:** Korpi & Palme (1998), "The Paradox of Redistribution and Strategies of Equality," *American Sociological Review*—universal programs achieve more redistribution than targeted programs.