Long-term care capacity cannot be created at the moment someone needs it. A trained workforce takes time to develop. Home-support networks require providers, transport, supervision and sustainable funding. Residential capacity involves buildings, regulation and staff. Rehabilitation, dementia support and community services depend on specialist capability that may be particularly difficult to establish outside major urban centers. For Argentina, the strategic question is therefore not simply how much care is needed today, but what combination of support may be required five, ten or fifteen years from now.
This is the planning challenge examined across the Argentina Aging, Long-Term Care & Community Support Knowledge Hub. Argentina already has an aging population, and the direction of demographic change is clear. Earlier estimates from the Inter-American Development Bank projected the population aged 60 and over to rise from around 15.7% in 2020 to approximately 22% by 2050. Yet population aging alone does not determine future care demand. Functional health, family structures, housing, prevention, workforce availability, economic conditions and policy choices all influence how demographic change translates into services.
Predictive planning should therefore not be confused with predicting which individual will need care. Its stronger use is strategic: testing how populations, demand and capacity could evolve under different assumptions, identifying pressures early enough to respond and making uncertainty visible rather than disguising it behind a single forecast.
For Argentina, this approach could help shift long-term care planning from periodic reaction to a more deliberate assessment of future capacity. But it will only be credible if national projections connect with provincial realities, municipal knowledge and the lived circumstances of older people, disabled people and families.
Demography sets the direction, not the final demand figure
Argentina has substantial demographic evidence with which to begin. INDEC has produced population estimates and projections by sex and age for the country and provinces, while census evidence provides a deeper picture of household composition and population structure. These sources allow planners to examine not simply national aging but how demographic trajectories differ geographically.
The distinction matters because age is an imperfect proxy for care need.
Many people remain independent well into later life. Others experience functional limitations earlier because of disability, chronic illness, socioeconomic disadvantage or environmental barriers. At advanced ages, the probability of needing assistance increases, but the amount and type of support required still varies substantially.
Research on Argentina published by the Inter-American Development Bank estimated that around 23% of people aged over 60 had some degree of difficulty with basic or instrumental activities of daily living in the underlying 2012 evidence. The same analysis emphasized that demographic change would increase the absolute number requiring support even if age-specific dependency rates remained stable.
This makes population needs assessment a better starting point than simple population extrapolation. Planners need to understand the relationship between age, function, health, household circumstances and environment.
A forecast based only on the number of people aged over 65 risks overestimating need among independent older people while simultaneously underestimating intensive support required by smaller groups with substantial disability, dementia or complex health conditions.
The stronger question is therefore: given plausible changes in population and functional need, what forms of support are likely to be required, where, and at what intensity?
Argentina is planning within a fragmented long-term care landscape
Forecasting becomes more complicated because Argentina does not have one unified long-term care system into which projected demand can simply be inserted.
Support is distributed across national programs, provinces, municipalities, PAMI, obras sociales, health services, disability arrangements, private providers, community organizations and households. Families continue to provide a large amount of unpaid assistance. Public and private access varies according to coverage, location, income, eligibility and the availability of services.
This means future capacity cannot be expressed as one national number.
An additional 100,000 people requiring assistance would create very different operational consequences depending on whether support were provided predominantly by families, publicly funded home services, private caregivers, community organizations or residential facilities. The same projected need therefore produces different workforce, funding and infrastructure requirements under different service models.
Predictive planning has to model the architecture of care as well as the volume of demand.
For example, a strategy designed to enable more people to remain at home could reduce some future residential demand while increasing requirements for home-care workers, rehabilitation, transport, respite, primary-care coordination, accessible housing and digital support. It may also shift responsibilities toward municipalities or community organizations unless funding arrangements develop alongside policy.
Forecasting is consequently inseparable from choices about long-term care service models and pathways. A country does not simply discover its future capacity requirement; policy helps create it.
Planning should use ranges rather than a single future
Long-range forecasts are attractive because a precise number appears actionable. Precision, however, can exceed what the evidence justifies.
Argentina can estimate population aging with reasonable confidence over medium-term horizons, but several determinants of care demand are more uncertain. Healthy life expectancy may change. Prevention may delay functional decline. Household sizes may continue to evolve. Migration can alter the distribution of younger and older populations. Economic conditions affect private purchasing and public budgets. Technology may change how some support is delivered.
Forecasts should therefore use scenarios and ranges.
A useful planning model might examine several plausible futures:
- a baseline in which age-specific dependency and current service patterns change relatively little;
- a stronger community-care scenario in which home support, rehabilitation and prevention expand;
- a higher-dependency scenario in which functional need rises faster than expected;
- a reduced-family-capacity scenario in which less unpaid care is available;
- a workforce-constrained scenario in which formal capacity cannot expand at the required rate.
These are not predictions. They are structured ways of asking what the system would need if important assumptions change.
Organizations exploring this type of uncertainty can use the Digital Twin Scenario Modeler to test interactions between demand, workforce, capacity and service stability. For Argentina, any modelling would need locally valid demographic, workforce, service and cost assumptions; the tool is a scenario framework rather than a country-specific forecasting instrument.
Scenario: Córdoba tests what aging means for home-care capacity
A provincial planning team in Córdoba is reviewing how support for older residents might need to change over the next decade. Population projections show continued growth in older age groups, but converting that trajectory directly into residential-bed requirements would assume that today's service pattern remains unchanged.
The province instead develops three scenarios.
The first broadly maintains the existing balance between family care, home support and residential provision. The second assumes stronger investment in home-based assistance, rehabilitation and community services. The third tests what happens if family caregiving capacity falls more quickly as households become smaller and more women remain in paid employment.
Each scenario produces a different operational requirement. Expanded home support reduces some projected pressure on residential capacity but increases demand for workers able to travel between households, supervisors, rehabilitation professionals and coordination with primary care. Reduced family capacity raises formal-service demand in both home and residential settings.
The exercise also identifies geography as a constraint. Capacity that appears sufficient at provincial level is concentrated around larger urban areas, leaving some smaller communities exposed.
Rather than selecting one scenario as the forecast, the province identifies investments that remain useful across all three: better workforce information, stronger home-support infrastructure, improved functional-need assessment and clearer monitoring of geographic access.
Predictive planning has therefore changed the decision. The question is no longer how many beds Córdoba will need on one assumed trajectory, but which capabilities make its care system more resilient across several plausible futures.
Functional need is the bridge between population projections and care demand
Age projections become much more useful when combined with evidence about functional ability.
Long-term care is fundamentally concerned with the support people require to undertake everyday life. Difficulties with bathing, dressing, eating, mobility, shopping, preparing meals, managing medication or using transport provide a closer connection to care demand than chronological age alone.
Argentina has survey evidence on limitations in activities of daily living, although historical measures, methodologies and data availability create uncertainty when projecting forward. This uncertainty should be visible rather than eliminated through modelling assumptions.
Several factors can change future functional need. Better prevention and chronic-disease management may allow people to remain independent for longer. Conversely, longer survival with multiple conditions may increase years lived with substantial support needs. Housing design can either enable independence or turn modest impairment into dependency. Assistive technology and rehabilitation can alter what assistance is required.
A mature forecast therefore separates demographic growth from assumptions about dependency prevalence.
This is particularly important when considering frailty, falls and functional decline. If preventive interventions reduce deterioration, future demand may not follow the same trajectory as population aging. Conversely, failing to invest in prevention can convert manageable functional difficulty into more intensive support needs.
Forecasting should make those policy relationships explicit. Otherwise today's service model is silently built into tomorrow's projection.
Workforce forecasting is where capacity planning becomes operational
A forecast that says Argentina will require more long-term care without translating that requirement into people, skills and working time remains incomplete.
Formal support depends on a diverse workforce: caregivers, nurses, physicians, rehabilitation professionals, social workers, psychologists, community teams, coordinators and managers. Different service models require different combinations of these roles.
The Inter-American Development Bank has developed regional work specifically examining human-resource requirements for older people with care dependency. The principle is important: workforce requirements should be derived from projected need, service intensity and productivity assumptions rather than simply extrapolating current headcount.
For Argentina, workforce forecasting should consider:
- how many people are likely to require different levels of assistance;
- the hours and types of support associated with those needs;
- which tasks require professional or specialist competence;
- travel and scheduling time in home-based services;
- turnover, absence and workforce participation;
- training pipelines and the time required to expand supply;
- regional distribution rather than national totals alone.
The last point is critical. A national surplus in one profession does not resolve a shortage hundreds of kilometers away.
This connects forecasting directly with workforce data and capacity planning. The system needs to understand not only how many workers exist but where they work, what they can do and whether current employment conditions can retain them.
Workforce risk can compound faster than population demand
Long-term care workforce pressure is not simply the mirror image of demographic growth.
Demand can rise gradually while capacity deteriorates quickly if turnover increases, experienced workers leave, recruitment pipelines weaken or particular locations become difficult to staff. A service operating close to minimum viable capacity can become unstable after losing only a small number of employees.
Predictive workforce planning should therefore examine leading indicators rather than wait for service failure.
Vacancy duration, turnover, sickness absence, overtime, use of temporary labor, unfilled shifts, supervisor capacity and training completion can all indicate whether a workforce is becoming less resilient. None predicts failure perfectly, but patterns across several measures can justify earlier management attention.
The Predictive Workforce Risk Module offers organizations a structured way to examine turnover, vacancy, retention and continuity risks. In Argentina, the relevant measures would need to reflect local employment arrangements, provider structures and available data rather than importing assumptions from another labor market.
Predictive analysis should also avoid turning workforce pressure into surveillance of individual employees. The strategic objective is to identify structural instability: inadequate staffing models, difficult geography, weak supervision, poor career progression or unsustainable workload.
Those factors require organizational and system responses. An algorithm cannot recruit, train or retain the workforce on its own.
Scenario: a home-support network sees the shortage before services are cancelled
A network providing home support in Greater Buenos Aires has met its contracted and privately purchased commitments for several years. Headline staffing numbers remain broadly stable, but operational data begins to change.
Turnover among experienced caregivers rises. Vacancies take longer to fill. Supervisors spend more time covering scheduling problems, and workers increasingly travel between distant neighborhoods because recruitment is uneven. Missed visits remain uncommon, so a conventional performance dashboard still appears acceptable.
Capacity modelling shows a different picture. If demand increases by even 8% over the next two years while current retention patterns continue, the service will no longer have sufficient productive hours to maintain its existing coverage. The greatest exposure is in areas where travel already absorbs significant working time.
The organization responds before continuity deteriorates. Recruitment is targeted geographically rather than generically. Scheduling is redesigned to reduce unnecessary travel. Supervisory capacity is protected, and retention information is reviewed by role and location. Leaders also identify which growth commitments should not be accepted until staffing capacity is demonstrated.
At system level, aggregated information reveals that several providers are experiencing similar problems. What initially looked like an organizational recruitment issue begins to appear as a wider labor-market constraint.
The scenario illustrates the practical value of predictive planning. No model can guarantee that a shortage will occur. Its purpose is to expose the conditions under which current capacity becomes unsustainable early enough for decision-makers to act.
Family caregiving is a capacity variable, but not a free resource
Any forecast of Argentina's future long-term care needs that excludes families will substantially misrepresent how support is currently delivered. A forecast that assumes families can indefinitely absorb increasing demand will be equally misleading.
Unpaid support has historically compensated for gaps between formal need and formal provision. Demographic and social change may make that arrangement increasingly difficult to sustain. Smaller families, migration, longer working lives and women's labor-force participation all influence the amount of care households can provide.
The issue is not whether families will remain important. They will. The planning question is what intensity of unpaid support can reasonably be assumed without creating unacceptable consequences for caregivers.
Predictive models should therefore test different family-capacity assumptions. A baseline might assume current patterns continue. An alternative scenario could examine what happens if average unpaid-care availability falls. Another might model stronger formal respite and caregiver support, potentially allowing family care to remain sustainable for longer.
Analysis of caregiver support and navigation becomes part of capacity planning rather than an additional welfare consideration.
This changes the economics. Respite, training and navigation create costs, but they may also protect a major source of support. Conversely, underfunding formal services because families are assumed to fill the gap can shift expenditure and risk into households through lost earnings, exhaustion and reduced wellbeing.
Predictive planning should make that transfer visible.
Residential capacity needs more sophisticated forecasting than bed counts
Residential long-term care remains an important part of Argentina's support landscape, particularly for people with substantial dependency, dementia, complex needs or insufficient support at home. Planning future capacity, however, should not begin and end with the number of available places.
Residential demand is influenced by the strength of alternatives.
If home support, rehabilitation, accessible housing and caregiver assistance remain limited, residential services may receive people whose needs could otherwise have been supported in the community. If those alternatives expand, residential provision may increasingly serve a population with greater complexity.
This has consequences for workforce and quality.
A stable number of residential places does not represent stable capacity if residents require more intensive nursing, dementia support or assistance with everyday activities. Staffing ratios, competencies, building design and clinical support may all need to change even when bed numbers do not.
Forecasting therefore needs to model acuity as well as volume.
This also creates a governance question. Expansion based purely on projected demand can lock a system into institutional patterns that future policy is trying to change. Equally, restricting residential capacity without developing credible community alternatives can leave families carrying needs that cannot safely be sustained at home.
The objective should be a balanced continuum in which residential provision develops alongside home and community-based support, rather than being planned as separate sectors.
Scenario: Mendoza models residential demand and discovers a community-capacity problem
A planning exercise in Mendoza initially projects a significant increase in residential long-term care demand over the following decade. Population aging and expected growth in advanced-age groups appear to justify additional places.
Before using the projection as an infrastructure target, planners examine the assumptions behind it.
Referral information shows that some people enter residential settings after hospital episodes because sufficient support cannot be organized quickly at home. Others move because family caregivers have reached exhaustion. Rehabilitation availability and home-support intensity vary geographically.
The province therefore tests an alternative scenario in which post-hospital rehabilitation, higher-intensity home support and caregiver respite expand gradually.
Residential demand does not disappear. People with severe dementia, substantial functional dependency and complex health needs continue to require specialist provision. However, the projected resident profile changes: fewer people enter primarily because community alternatives are unavailable, while the average support intensity within residential settings increases.
The investment requirement is consequently redistributed. The province still needs residential capacity, but it also needs stronger community services and a more skilled residential workforce.
The scenario prevents a common forecasting error. Historical admission patterns are not treated as an immutable expression of future need. They are partly the result of the service options available at the time.
Funding forecasts need to follow people across the care pathway
Capacity projections eventually become financial questions.
Argentina's fragmented financing arrangements make this particularly complex. Costs may fall across national programs, PAMI, provincial or municipal budgets, obras sociales, households and private purchasers. An intervention funded in one part of the system may generate benefits elsewhere.
A stronger home-support system, for example, requires expenditure before any reduction in hospital use or delayed residential admission becomes visible. Rehabilitation may increase short-term service intensity while improving longer-term independence. Respite may protect family caregiving capacity without generating a straightforward budget saving.
Financial forecasting should therefore examine total system consequences rather than assuming every preventive intervention must pay for itself within the same budget line.
It should also distinguish affordability from value. A service can produce good outcomes and still be difficult to finance at scale. Conversely, a relatively inexpensive service model can be poor value if it creates instability, unmet need or downstream costs.
The connection with budget impact and affordability is therefore central. Forecasts need plausible assumptions about unit costs, wage growth, service intensity, utilization and who ultimately pays.
Where information is uncertain, sensitivity analysis is more useful than false precision. Decision-makers should be able to see which assumptions have the greatest effect on projected expenditure.
Predictive planning can strengthen prevention if it changes investment timing
The strategic value of forecasting is greatest when it allows action before demand becomes unavoidable.
If a province can see that falls-related admissions are likely to rise as its population ages, it can examine prevention and rehabilitation capacity before hospital pressure intensifies. If workforce modelling indicates a future shortage of home-care workers, training and recruitment can begin years earlier. If particular municipalities are likely to experience rapid growth in advanced-age populations, housing and transport planning can respond before accessibility becomes a larger barrier.
This is a different use of prediction from identifying individuals supposedly destined for high-cost care.
Population-level forecasting creates space for preventive value and earlier intervention. It asks which investments might alter future trajectories rather than merely calculating how expensive those trajectories will become.
Evaluation remains essential. A preventive strategy should not be assumed effective because a model predicted benefit. Actual implementation needs to be monitored against outcomes, costs and equity.
Forecasting informs the hypothesis. Real-world evidence determines whether it was right.
Geographic forecasting matters in a country of Argentina's scale
National projections are necessary for national strategy, but they are insufficient for operational planning.
Argentina's provinces differ in age structure, settlement patterns, economic circumstances, service infrastructure and workforce availability. Within provinces, metropolitan areas and rural communities can face entirely different capacity problems.
A projected national workforce requirement may therefore conceal local shortages. Similarly, sufficient residential capacity at provincial level may be geographically inaccessible to families in remote areas.
Predictive planning needs spatial intelligence: where older populations are likely to grow, where services currently operate, how far workers travel, where transport is limited and which communities have the weakest alternatives.
For remote areas, the solution may not be replication of urban service models. Mobile teams, stronger generalist capability, scheduled specialist outreach, transport support and technology-enabled consultation may be more sustainable.
But digital delivery should not be assumed universally accessible. Connectivity, device access, disability, sensory impairment and digital confidence influence who can use it. Future capacity models that count virtual services as equivalent to face-to-face capacity risk overstating availability.
Geographic planning therefore needs to connect physical and digital access. A nominal service is not necessarily usable capacity.
Scenario: northern Argentina tests whether telehealth actually increases capacity
A provincial health authority in northern Argentina is considering greater use of remote specialist consultations for older people with complex conditions in smaller communities. On paper, telehealth appears capable of increasing specialist capacity without locating additional specialists in every area.
The planning model initially counts each remote appointment as additional access.
Local testing reveals a more complicated reality. Some primary-care centers have reliable connectivity and staff who can support remote consultations. Others experience unstable connections. Some older people require assistance using equipment, while people with hearing, cognitive or communication difficulties may need additional support. Remote consultations also generate follow-up tasks for local professionals.
The authority adjusts the capacity model. Telehealth is no longer treated as a simple substitute for physical specialist time. The forecast includes local facilitation, connectivity, equipment, follow-up and the proportion of consultations that still require face-to-face assessment.
Remote care remains valuable, but its realistic capacity contribution is lower than the initial headline figure.
This produces a more credible service plan: investment is directed both to digital infrastructure and to the local workforce needed to make remote care effective.
The wider lesson is important for future long-term care planning. Technology can extend reach, but capacity exists only when the entire delivery pathway works.
Better digital infrastructure can improve forecasting without automating entitlement
Argentina's current digital-health development provides a stronger technical environment for future planning. The Estrategia Nacional de Salud Digital 2025–2030 is advancing through federal collaboration, and the 2026 Encuentro Nacional de Salud Digital included real-time tests of clinical-information exchange, validation of common standards and work toward a federal interoperability roadmap.
The Red Nacional de Salud Digital has long been designed around interoperable nodes rather than requiring every jurisdiction to abandon its own systems. National standards, including FHIR-based exchange and common clinical terminology, support the principle that information generated in different systems can become more usable across institutional boundaries.
These developments primarily concern health information, not a unified national long-term care database. Nevertheless, the infrastructure and governance principles are relevant.
Improved interoperability and data exchange could make forecasting more responsive by reducing the delay between changes in population need and their visibility to planners.
Organizations considering whether their digital infrastructure is capable of supporting more advanced analytics can use the Digital Transformation, AI and Cybersecurity Readiness Assessment to examine information architecture, governance and readiness. It does not establish compliance with Argentine requirements, but it can help structure internal questions about whether the underlying systems are sufficiently mature for increasingly data-intensive planning.
The boundary should remain clear. Better forecasting can inform the amount and location of capacity. It should not automatically decide whether a named individual receives care.
Governance should require models to explain their assumptions
Forecasts can acquire authority simply because they contain numbers. That makes model governance as important as model sophistication.
Decision-makers should be able to understand the assumptions that drive a forecast. What dependency rates were used? What service model was assumed? How much unpaid family care is built in? What workforce productivity is expected? What happens to the result if one of those assumptions changes?
A credible planning model should make at least five things visible:
- the source and age of the underlying data;
- the assumptions connecting population change to service demand;
- the uncertainty or range around key estimates;
- the sensitivity of results to different policy choices;
- the point at which forecasts will be reviewed against actual experience.
This supports trust, transparency and ethical data use. Forecasts should be open to challenge by professionals, policymakers and communities rather than treated as technically unquestionable.
Governance also needs ownership. National authorities may maintain demographic and strategic evidence, provinces may model service requirements, municipalities may hold crucial local knowledge and providers understand operational capacity. Responsibility for updating assumptions and responding when forecasts diverge from reality needs to be explicit.
The Governance Maturity Assessment can help organizations examine whether decision rights, accountability and assurance are sufficiently clear to support this type of planning. In a federal system, mature governance does not require every decision to be centralized; it requires clarity about which level owns which decision.
Forecasts should be continuously compared with what actually happens
Predictive planning is not a report produced once every decade.
Demographic projections may change slowly, but service conditions can change quickly. Economic shocks, migration, provider closures, workforce turnover, new technology or policy reforms can alter capacity assumptions within months.
Forecasting therefore needs a learning cycle.
Projected demand should be compared with actual utilization, unmet need, waiting times, workforce supply and outcomes. Where actual experience differs substantially from the model, planners should investigate why.
The model may be wrong. Alternatively, the service system may be suppressing demand because people cannot gain access. A forecast of 1,000 people needing support cannot be validated simply because only 700 receive it.
Qualitative evidence is therefore important. People using services, families and frontline workers can explain discrepancies that administrative data cannot.
This makes predictive planning part of governance rather than a specialist analytical exercise. The value lies in the repeated cycle of forecast, decision, implementation, observation and revision.
Over time, Argentina could develop stronger local and provincial evidence about which assumptions best predict different forms of long-term care demand. Forecasts should become more useful because the system learns, not because models become progressively more opaque.
Planning to 2040 means building options rather than locking in one model
Argentina's existing official population projections extend to 2040, providing a natural medium-term horizon for thinking about capacity. But the purpose of a 2040 plan should not be to specify every service that will exist in 2040.
The stronger objective is to identify decisions with long lead times and preserve flexibility around those that can adapt later.
Workforce development, accessible housing, digital infrastructure, rehabilitation capability and community-support networks all require sustained investment. Failing to build them early can constrain future choices. Other aspects of service configuration can evolve as evidence becomes clearer.
Planning should also recognize path dependency. If investment is concentrated predominantly in institutional infrastructure, later expansion of community support becomes harder. If the system relies heavily on unpaid care without strengthening family support, future formal demand may emerge abruptly when those arrangements become unsustainable.
The strategic task is therefore to build adaptable capacity.
Argentina does not need certainty about the exact number of home-care hours, residential places or rehabilitation episodes required in 2040. It needs enough intelligence to understand the plausible range, enough workforce strategy to expand supply, enough governance to detect divergence and enough flexibility to shift resources as needs change.
This is the deeper value of predictive planning. It converts uncertainty from a reason for postponing decisions into something that can be managed explicitly.
International learning lies in the planning discipline, not a particular forecasting model
Many countries facing population aging are attempting to forecast long-term care demand. Their institutional mechanisms differ substantially: some operate dedicated long-term care insurance systems, others rely more heavily on taxation, local government, private purchasing or families.
Argentina cannot simply import a demand model built around another country's entitlement structure or workforce.
The transferable lesson lies instead in the planning discipline: separate demographic change from functional need; connect demand to actual service models; model workforce and finance alongside utilization; test several futures; expose assumptions; and update forecasts against real experience.
This approach is especially relevant in fragmented systems. Predictive planning can provide a common strategic language even where delivery responsibility remains distributed.
It can also improve policy debate. Instead of arguing abstractly that aging will create "more demand," governments and communities can examine which forms of demand are likely to grow, which assumptions are driving the estimates and which investments could change the trajectory.
The result is not certainty. It is better preparedness.
Conclusion
Argentina's future long-term care challenge is already visible in its demographic direction, but demographic aging does not dictate one inevitable service future. How much formal care will be required, where it will be needed and what it will cost will depend on functional health, family capacity, workforce supply, prevention, housing, technology and choices about the balance between home, community and residential support.
Predictive planning can help make those choices earlier and more explicit. Its strongest contribution is not identifying a supposedly precise number of people who will require care in a particular year. It is showing how different assumptions produce different demands on workforce, infrastructure and funding, and identifying capabilities that remain valuable across several plausible futures.
For Argentina, that requires national demographic intelligence to connect with provincial planning and municipal knowledge. It requires workforce forecasts that recognize geography and retention, financial models that follow costs across fragmented systems, and scenarios that do not treat unpaid family care as an unlimited resource. Digital interoperability can strengthen the evidence base, but accountability for decisions must remain human and transparent.
The strategic advantage comes from time. Workers can be trained before shortages become severe. Community capacity can be developed before institutional demand rises. Prevention can be strengthened before avoidable dependency becomes embedded. Predictive planning is valuable precisely because it allows Argentina to shape future capacity rather than merely calculate the consequences once demand has arrived.