Free China Movement

Evidence · Human Rights · Democratic Future

Take Action

English 简体中文 繁體中文 Français Deutsch Español

🔎

Property Restitution and Its Limits: Housing, Pensions and Confiscated Assets

October 2026

Download: PDF · Word (DOCX)

Series: Transitional Justice, Economic & Social Transition — research papers accompanying the Transitional Justice, Economic & Social Transition working blueprint. Paper T2.

1. The question

Over seven decades the party-state on the mainland confiscated land, houses, businesses, savings and religious property. It also gave housing and enhanced pensions to those who served it. A democratic successor will face three demands at once. It must return what was taken and stop rewarding what was abusive. And it must do both without evicting today’s ordinary occupants or bankrupting the treasury. This paper asks how far restitution should go and what replaces it where it stops.

2. The hub position

The Transitional Justice, Economic and Social Transition hub (https://freechina.live/governance-public-policy/third-republic-governance-blueprints/transitional-justice-economic-social-transition/) treats four tracks separately: truth-finding, individual responsibility, fitness for office, and remedy. It rejects both collective guilt and blanket impunity. It insists that payments, contracts and financial settlement keep working during transition. It lists the legal basis for compensation, and fiscal ceilings on it, as open questions. It also says that ROC (Taiwan) and foreign examples are comparative evidence, not mainland data.

The legislation already exists in model form. The Model Transitional Justice and Victim Reparation Act v1.2 (https://freechina.live/governance-public-policy/third-republic-governance-blueprints/en-transitional-justice-victim-reparation-act-draft/) covers unlawful confiscation (Art. 5). Where feasible it puts restitution first, including the return of property, without overriding third parties’ lawful rights (Art. 12). It sets public compensation benchmarks that include property loss (Art. 13). It bars open-ended fiscal commitments that cannot be enforced (Art. 22). It sets no amounts, deadlines or pension rules and leaves these to a future legislature (Art. 28). This paper proposes policy to fill those gaps. It does not amend the Act.

3. Evidence and comparative options

International standard. In 2005 the UN Sub-Commission approved the Principles on Housing and Property Restitution. They make restitution the preferred remedy. Compensation is reserved for cases where an independent body finds that restoration is factually impossible, or where the claimant chooses it freely (OHCHR handbook, https://www.ohchr.org/sites/default/files/Documents/Publications/pinheiro_principles.pdf). They were written for refugees and displaced persons. A domestic programme can take guidance from them but is not bound by them.

Germany: restitution first, with firm exceptions. The 1990 Property Act returns assets taken by covered measures, on application, unless the Act excludes this (§ 3, https://www.gesetze-im-internet.de/vermg/__3.html). Return is excluded where the nature of the asset makes it impossible. It is also excluded where a natural person, religious body or non-profit foundation acquired the asset in good faith after 8 May 1945. An acquisition is presumed to be in bad faith if it broke the law of the time, came from corruption or abuse of position, or exploited coercion (§ 4, https://www.gesetze-im-internet.de/vermg/__4.html). Land and buildings are not returned in three cases: they were dedicated to public use, they were built into large housing estates, or they became part of an enterprise that cannot release them without serious harm. Claimants may choose compensation instead (§§ 5 and 8, https://www.gesetze-im-internet.de/vermg/BJNR211590990.html). The Act also excludes expropriations made under Soviet occupation authority in 1945–1949 (§ 1(8)(a), same URL).

That exclusion was challenged in court. On 23 April 1991 the Federal Constitutional Court rejected the complaints, noting that the Soviet Union had made irreversibility a condition of unification (https://germanhistorydocs.org/en/a-new-germany-1990-2023/the-federal-constitutional-court-rules-on-expropriations-april-23-1991). The Court also held that the equality guarantee obliges the legislature to create some arrangement compensating those excluded (BVerfGE 84, 90, headnote 4, https://www.servat.unibe.ch/DFR/bv084090.html). In Von Maltzan and Others v. Germany (2005), the European Court of Human Rights Grand Chamber found the related applications inadmissible. It held that claimants had no protected expectation beyond what the restitution and compensation statutes gave them. It stressed that a state has a wide margin when it repairs harms for which it was not responsible (paras 77, 87–92, 110–111, https://lagen.nu/dom/echr/001-68660).

Czech Republic: eligibility rules and the limits of good faith. In Gratzinger and Gratzingerova the European Court held that “the hope that a long-extinguished property right may be revived” is not a possession (para 69, https://lagen.nu/dom/echr/001-22710). Restitution rights therefore exist only where a statute creates them. Yet the legislature’s choices remain open to review. In Des Fours Walderode (2001), the UN Human Rights Committee found that a continuous-citizenship requirement, imposed retroactively in 1996, discriminated between equal victims of confiscation (https://hrlibrary.umn.edu/undocs/747-1997.html). In Pincová and Pinc, a couple who had bought their home in good faith decades earlier were ordered to hand it back and were repaid only the 1967 price. The Court found that this placed an excessive individual burden on them. It warned that repairing old injuries must not create disproportionate new wrongs (paras 58–64, https://lagen.nu/dom/echr/001-60726). Church property was settled later through a mixed package: the churches accepted the return of 56 percent of the confiscated property plus 59 billion crowns of compensation paid over 30 years (https://english.radio.cz/node/8561319).

Poland: systemic claims and fiscal realism. About 80,000 Bug River claimants held entitlements that successive laws had hollowed out. In Broniowski v. Poland (2004), the Court’s first pilot judgment, the Court found a systemic violation. Poland then let claimants choose between land auctions and cash from a special fund. By July 2009 it had made 19,444 payments worth about €206.5 million (Council of Europe, https://www.coe.int/web/impact-convention-human-rights/-/tens-of-thousands-of-people-compensated-over-world-war-ii-property-claims). In Hutten-Czapska v. Poland (2006), the Court found that long-standing rent control violated landlords’ property rights. It closed that pilot procedure in 2011 after Poland adopted a compensation scheme (https://www.justiceinitiative.org/voices/case-watch-speeding-human-rights-justice-europe). An undelivered promise of restitution creates new liability.

Pensions earned in the service of repression. A 2009 Polish law cut the pension accrual rate for pre-1990 Security Service work from 2.6 to 0.7 percent per year. The Polish Constitutional Court upheld it in 2010. It noted that the reduced pensions stayed above the statutory minimum and still averaged well above pensions in the general scheme. This reasoning is set out in the European Court’s 2013 decision in Cichopek and Others v. Poland, which involved 1,627 applicants (paras 6, 95–104, https://app.lexploria.com/document/564357-d).

ROC (Taiwan): party assets. The Act Governing the Settlement of Ill-gotten Properties by Political Parties and Their Affiliate Organizations (2016) shows one way to deal with a ruling party’s wealth. Assets acquired since 15 August 1945 are presumed ill-gotten unless they came from dues, donations or election subsidies (Art. 5). Presumed assets are frozen from the day the Act is promulgated (Art. 9). After public hearings, a committee may order them transferred to the state, to local bodies or to their original owners (Arts. 6, 14). Leases and mortgages held by bona fide third parties survive the transfer (Art. 7). Where return is impossible, compensation is paid instead (Art. 25) (Ministry of Justice, https://law.moj.gov.tw/ENG/LawClass/LawAll.aspx?pcode=A0030286). In Interpretation No. 793 (28 August 2020), the Justices upheld the Act. They held that it neither dissolves a party nor strips it of the assets it needs to operate. They also held that reliance formed under authoritarian rule deserves no protection (https://cons.judicial.gov.tw/docdata.aspx?fid=100&id=310974; https://taipeitimes.com/News/front/archives/2020/08/29/2003742462).

Options. These examples suggest three models:

  • (a) restitution in kind as the rule, as in Germany;
  • (b) compensation only, on a statutory scale;
  • (c) a mixed model: some property returned in kind, compensation for the rest, and recovery of abusively acquired holdings, as in the Czech church settlement and the ROC (Taiwan) party-assets regime.

The evidence favours (c). Restitution in kind matters most for homes, religious sites and small businesses. But collectivisation, urbanisation and repeated redevelopment on the mainland mean that most physical assets cannot be returned without displacing the people now living in or using them.

4. Recommended design

The following policies would implement Arts. 5, 12, 13 and 22 of the Reparation Act.

  1. Freeze first. On entry into force, freeze assets presumed to have been acquired by the former ruling party or its affiliates from sources other than dues, donations and lawful subsidies. Ordinary operations would need approval from an independent body.
  2. One claims window. Publish a single window of at least 24 months for lodging claims, with free assistance and access to state records. Impose no residence or continuous-citizenship requirement.
  3. Limited restitution in kind. Return property in kind only in three cases: owner-occupied homes that still stand and are vacant or state-held; religious and community buildings; and small enterprises that can still be identified as going concerns.
  4. Protection for present occupants. Good-faith occupants and buyers keep possession. Presume bad faith only for acquisitions through corruption, abuse of office or coercion. Where an occupant must leave, the state bears the cost and first provides comparable housing.
  5. Compensation scale. Where return is excluded, pay compensation on a published scale with a cap, where the rate falls as claims grow larger. Full market value is not required. Pay small claims first, in cash. Pay larger claims partly in transferable vouchers or land rights, on the model of the Bug River auctions.
  6. The statute as the only source of rights. State expressly that restitution rights arise only under this law. This keeps the scope of claims in the legislature’s hands rather than leaving it to open-ended litigation.
  7. Recovery of party and affiliate assets. Send abusively acquired assets to a restitution fund rather than the general budget, after public hearings and with judicial review. Exempt assets a party needs for its ordinary lawful work.
  8. Privileged pensions. Reduce to the general-scheme level any pension increments earned for service in organs that the Truth Commission finds committed grave violations. Never reduce a pension below the subsistence minimum. Base each reduction on individual review, with a right of appeal.
  9. Fiscal rule. Cap annual restitution spending as a share of revenue, pay in instalments, and require independent audit, in line with Art. 22.

5. Open questions and risks

  • Cut-off date. Options include 1949, earlier civil-war seizures, or the cases dated in Art. 6. Each includes or excludes very large groups.
  • Rural land. Collective ownership of land and household contract rights make returning farmland difficult. Land-use reform may serve better than restitution.
  • Urban redevelopment. Many original homes no longer exist. The state may have to choose between compensation and new rights in redevelopment projects.
  • Claimants abroad. Excluding them invites discrimination findings like Des Fours Walderode.
  • Fiscal capacity. Sizing the fund needs the inventory of public debt and assets addressed in paper T4.
  • Litigation. Clear statutory limits reduce this risk, but courts will still test proportionality in individual hardship cases.

6. Cross-links

In this series: Reparation Before Rhetoric · Lustration, Amnesty and Truth · Paying for Transition.

Working paper v1.0, published October 2026 as part of the Transitional Justice, Economic & Social Transition blueprint program. This paper will be updated as the evidence base develops.