An eighty-two-year-old widow begins transferring large sums to a charming new acquaintance. Her daughter is alarmed, and her bank is uneasy. The widow insists the money is hers to give. Every legal system must decide what happens next, and they decide very differently.

The underlying tension is old. Protection and autonomy pull in opposite directions, and older adults occupy difficult ground. They are adults with full legal standing, yet some face cognitive decline, isolation, or dependence on the very people who exploit them. The World Health Organization estimates that roughly one in six people over sixty experiences some form of abuse in community settings, much of it financial and most of it unreported.

Legal systems must answer three questions. How do we decide whether someone can make their own choices? What should replace their authority if they cannot? And who is obliged to act when abuse is suspected? Comparing the answers across jurisdictions reveals more than technical variation. It exposes competing assumptions about what dignity in old age actually requires.

Capacity Assessment Standards

Legal systems have historically used three approaches to capacity. The status approach treats a diagnosis, such as dementia, as sufficient to remove decision-making authority. The outcome approach judges capacity by whether a decision seems sensible. The functional approach asks whether a person can make this particular decision at this particular time. The older civil law doctrine of interdiction and early American incompetency proceedings leaned heavily on status. Most modern reforms have moved toward function.

The Mental Capacity Act 2005 of England and Wales is often cited as the clearest functional model. It begins with a presumption of capacity and requires that all practicable steps be taken to help a person decide before any finding of incapacity. A person lacks capacity only if they cannot understand, retain, use or weigh the relevant information, or communicate a decision. Crucially, the Act states that making an unwise decision does not, by itself, prove incapacity.

American states vary considerably, though most statutes now incorporate functional language, and the 2017 Uniform Guardianship, Conservatorship, and Other Protective Arrangements Act pushes further in that direction. Clinicians often draw on the four-ability model associated with Appelbaum and Grisso: understanding, appreciation, reasoning, and expressing a choice. Yet research has repeatedly found that guardianship petitions are granted at high rates after brief hearings, suggesting a gap between statutory standards and courtroom practice.

The threshold matters as much as the test. Who carries the burden of proof? Many American states require clear and convincing evidence. Is the finding medical or legal? In Hart's terms, capacity rules are secondary rules that determine whose choices the legal system recognises as valid. For the widow, a functional test asks whether she understands what she is giving away, not whether a judge would give it away.

Takeaway

A capacity test is really a rule about whose choices the law will honour. The more a system lets the wisdom of a decision stand in for the ability to make it, the closer it drifts toward judging people rather than their capacities.

Guardianship Alternatives

Full or plenary guardianship transfers nearly all decision-making authority to another person. The guardian may decide where the individual lives, what medical treatment they receive, and how their money is spent. Critics have described it as a form of civil death, because the person often loses the right to contract, marry, or vote. Decisions are typically made according to best interests, or in some American states, substituted judgment about what the person would have chosen.

Many civil law jurisdictions reformed this model by grading it. France distinguishes sauvegarde de justice, curatelle, and tutelle, increasing in restrictiveness. Japan's 2000 adult guardianship reform created three tiers matched to different degrees of impairment. Germany replaced adult guardianship in 1992 with Betreuung, a custodianship limited to specific tasks, and its 2023 reform placed the person's own wishes at the centre of the custodian's duties.

A more radical shift followed Article 12 of the UN Convention on the Rights of Persons with Disabilities, which the treaty's committee interpreted in 2014 as requiring supported rather than substitute decision-making. Ireland's Assisted Decision-Making (Capacity) Act 2015, commenced in 2023, abolished the wards of court system in favour of decision-making assistants, co-decision-makers, and representatives. Texas enacted a statute recognising supported decision-making agreements in 2015, and many states followed. British Columbia's Representation Agreement Act of 1996 anticipated these ideas, and Peru's 2018 reform went further than most by eliminating legal incapacity for adults with disabilities.

Support models face genuine difficulties. The line between support and influence is thin, and the supporter may be the very person exploiting the older adult. Advanced dementia strains the premise that a person's will can always be discerned. Lasting powers of attorney and similar planning tools preserve autonomy but also serve as vehicles for financial abuse when oversight is weak.

Takeaway

Supported decision-making relocates rather than removes risk: it shifts the danger from a state-appointed guardian overriding someone to a trusted supporter quietly steering them. Good design means watching whoever holds the pen.

Mandatory Reporting and Investigation

In the United States, every state operates an Adult Protective Services system, and most impose a legal duty on specified professionals, such as healthcare workers and social workers, to report suspected abuse. Some states extend that duty to any person. Financial institutions have become key actors: the federal Senior Safe Act of 2018 grants immunity to trained employees who report suspected exploitation, and several states permit banks to delay suspicious transactions.

England takes a different path. There is no general mandatory reporting duty for elder abuse. Instead, section 42 of the Care Act 2014 requires local authorities to make enquiries when an adult with care and support needs is experiencing or at risk of abuse and cannot protect themselves. A proposed power of entry was consulted on and not adopted. Scotland's Adult Support and Protection Act 2007 goes further, allowing assessment, removal, and banning orders, but generally requires the adult's consent unless undue pressure is suspected.

Mandatory reporting is contested. Critics argue it treats older adults like children, whose abuse is subject to similar regimes, and may discourage people from seeking help if disclosure triggers investigations they did not want. Evidence on whether reporting mandates improve outcomes remains limited. Australia illustrates a middle position: its Serious Incident Response Scheme, introduced in 2021, mandates reporting in residential aged care, while community settings rely largely on voluntary pathways.

Across systems, consent marks the hardest boundary. In most American states, a capable adult may refuse protective services even after abuse is substantiated. The design questions are consistent: who must report, what triggers investigation, what powers investigators hold, and what happens when the victim says no.

Takeaway

Reporting duties determine who notices abuse, but consent rules determine what can be done about it. A system that excels at detection yet has no answer for a capable victim who refuses help has solved only half the problem.

Seen together, these three questions form a single design problem. A strict capacity test, a graduated range of alternatives, and a consent-respecting investigation process express one philosophy. Diagnosis-based incapacity, plenary guardianship, and broad intervention powers express another.

The comparative evidence suggests that no system has eliminated the tradeoff. Germany and Ireland reduced overreach but must monitor custodians and supporters. American reporting regimes widened detection but still struggle when victims decline help. England's functional test protects unwise choices while leaving hard cases to judgment.

Perhaps the most useful principle is proportionality: each intrusion on autonomy should be the least restrictive response to a specific, demonstrated need. The widow's case is a test not of her judgment, but of ours.