Seminar Session 2 Records | The Attribution Principles of State Compensation Liability: Moving from Fault Liability to Strict Liability

On August 26, 2023, in response to the Executive Yuan's comprehensive revision of the State Compensation Act in 2021—expanding it from 17 articles in the current law to 43 articles—and observing the civil version of the Draft State Compensation Act formulated by lawyers and legal practitioners at the end of 2022, the Judicial Reform Foundation, the Administrative Law Committee of the Taiwan Bar Association, the Constitutional Reform Committee of Taiwan Bar Association, the Judicial Reform Committee of Taiwan Bar Association, and the Taiwan Law Association jointly held this seminar.

In the second session, led by moderator Professor Hung Chia-yin, and inspired by the insights of speaker Professor Lin Chia-ho as well as three panelists—Attorney Chou Yu-hsiu, Attorney Wang Chen-huan, and Professor Wang Yun-ju—along with active discussions from participating guests, a dialogue was initiated on the "attribution principles of state compensation liability."

1. Problem Awareness

   Professor Lin pointed out that state compensation for types of duties performed by public officials currently still takes "unlawfulness" and "intent or negligence" as elements, making claims often very difficult due to attribution requirements. Using the Taichung MRT accident as an introduction, he discussed whether the Taichung City Government, as the entity with a supervisory duty over the project, could potentially establish state compensation liability.

2. Third-Party Connection as an Element for Establishing State Responsibility

   1. The concrete contents and applications of the protective norm theory and third-party connection are not entirely identical and should be distinguished. The protective norm theory concerns primary rights protection, whereas state compensation, acting as secondary rights protection, should not need to consider the protective norm theory (though third-party connection must still be considered); however, domestic discussions seem to conflate the two concepts.

   2. Supreme Court 72 Tai-Shang-Tzu No. 704 Precedent highlights that: "For the general public, public official duties that merely confer reflective benefits do not constitute public law rights, and no claim for state compensation may be asserted." Obviously, it is almost unimaginable what aspect of daily life is unrelated to state tasks; therefore, it is inevitably necessary to have a theory to reasonably narrow the scope of state compensation liability.

   3. The most important and typical coordinating mechanism is social law: transforming private law risks into social law, namely handling through state benefit measures the risks that should otherwise be borne by private third parties.

   4. It is often observed that reliance protection is used to weaken the application of state compensation, which is indeed necessary, because although administrative agencies have administrative tasks, they have very limited understanding and operational capacity regarding certain activities and domains. However, in judicial application (taking German practice as an example), problems of inconsistent standards can often be observed: in the same specific domain, state compensation seemingly cannot be established in an individual case, yet is suddenly established in some other cases. The scope of the "foreseeability" and "controllability" of hazards by administrative agencies is extremely unstable in application, and judges frequently shoot the arrow first and draw the target afterwards (e.g., if they believe state compensation should not be established in a case, they will state that even if third-party connection is established in the case, the agency still lacks foreseeability or controllability regarding the occurrence of the hazard; conversely, if they believe state compensation should be established, they will downplay adverse objective evidence and determine that, based on objective evidence, the agency possessed foreseeability and controllability). Therefore, beyond individual justice, how to construct application stability through theory to ensure legal certainty is of great importance.

   5. Frequently discussed fields include the Architecture Act, and common controversial situations involve government planning, licensing, and supervision (especially financial supervision).

3. Thought Pathways for Establishing State Compensation Liability

   Preventing the occurrence of hazards is a typical administrative task (broadly defined as police objectives), forming the so-called duty of service to protect third parties, whereupon a third party has the right to demand police intervention (primary rights protection), subsequently establishing state compensation liability (secondary rights protection).

   However, theoretically, all state tasks could satisfy the requirements for police intervention, forming subjective public rights and subsequently establishing state compensation obligations—would this be too sweeping? Perhaps this is precisely why fault liability must be used to avoid fully adopting strict liability. Yet, between fault liability and strict liability, whether there is the possibility of an intermediate option is precisely the question Professor Lin Chia-ho aims to discuss in this paper.

4. Introducing the Private Law Concept of Strict Liability into the Public Law Domain

   1. The Meaning of Strict Liability in the Private Law Domain

      Creating and establishing a hazard despite knowing the risk (because it brings benefits), the derived duty of hazard control means that whether a duty of care was violated at the exact moment the hazard went out of control becomes unimportant (replacing subjective attribution with the occurrence of objective hazard). This was first discussed in the field of consumer protection law.

   2. Should the Private Law Concept of Strict Liability Be Introduced into the Public Law Domain?

      • (1) Arguments from opponents

        In the private law domain, those who create hazards enjoy the benefits, whereas public law does not; because the benefits of hazards created by the state do not accrue to the state, but are enjoyed by all citizens. Especially nowadays, where administrative benefits outweigh administrative interference, strict liability should be restricted to specific domains.

      • (2) Arguments from supporters

        Ernst Forsthoff, the father of German administrative benefits, suggested that victims are no closer to the source of damage than the state. Therefore, from the perspective of distributive justice, strict liability should be considered. Nevertheless, he still acknowledged that even if strict liability is introduced into the state responsibility system, a stable and controllable boundary must still be established, such as dividing individual domains, particularly specific fields where hazards directly materialize (rather than merely causing indirect effects).

      • (3) Two strategies frequently introduced by administrative jurisprudence to solve this problem:

        1. Introduce strict liability, but limit it to specific fields, such as public hospitals, as is done in France.
        2. Incorporating the consideration of strict liability into a tiered status system, meaning the fault liability system continues to be maintained, but the determination of attribution is objectified (introducing more objective risk considerations, such as the source of hazard, control, and the sharing of benefits and burdens) ➝ restricting and loosening the elements of attribution.

5. Legislative Suggestions

   1. Screen out specific fields and directly establish strict liability.
   2. Maintain the principle of fault liability, but introduce risk allocation theory to objectify subjective attribution.
   3. From a dege ferenda perspective, it is suggested to insert Paragraph 3 into Article 3 of the civil version: "In the circumstances of Paragraph 1, after comprehensively considering all conditions such as the formation of the hazard, risk assumption, attribution of benefits, professional and factual controllability, how the infringement occurred, and the specific circumstances of the occurrence of the infringement, if it is deemed that the infringement of people's freedom or rights still constitutes state compensation liability even if it is unrelated to the violation of official duties."

1. Thought directions for this issue:
   • (1) Should the elements of unlawfulness and illegality be maintained?
   • (2) Should the fault element be maintained?
   • (3) Extension issue: The Police Use of Weapons Act.

2. J.Y. Interpretation No. 469 has already set forth the state's obligation of hazard prevention or hazard management, and state compensation is the compensation liability arising after the breach of the duty of hazard prevention and mitigation. Therefore, when discussing state compensation, the premise is that the state has breached its risk prevention obligation.

3. If a systematic interpretation is required to handle state responsibility issues, the typical classification must first explore and distinguish whether it is an unlawful or lawful act by the state. If it is unlawful, state compensation legal principles apply, handled in a manner similar to torts under the Civil Code; if it is lawful, the legal principles of loss compensation apply (the current law lacks general provisions).

4. Such a classification method may consume excessive procedural costs at the frontend, making it impossible to determine subsequent relief methods; at the backend, it will affect the amount of compensation (indemnity), as only compensation strictly follows the principle of fully repairing damages, whereas indemnification does not necessarily require complete repair of damages.

5. J.Y. Interpretation No. 670 points out that the essence of the Wrongful Detention Compensation Act is actually indemnification, which also prompts reflection on whether today's practice of sharply distinguishing between compensation and indemnification is appropriate.

6. What J.Y. Interpretation No. 469 discusses is not merely repairing damages, but also how to establish mechanisms to prevent damages from occurring.

7. Possible approaches, as suggested by the speaker, are firstly to establish general risk allocation norms in the State Compensation Act, or secondly to formulate individualized norms for risk allocation in special laws.

8. Regulatory framework of the former Police Use of Weapons Act:
   • (1) Paragraph 1 of Article 11: Indemnification for the lawful use of firearms.
   • (2) Paragraph 2 of Article 11: Compensation for unlawful causation of death or injury.
   • (3) Therefore, Paragraph 2 actually does not require determining elements of intent or negligence; as long as there is unlawful causation of death or injury, compensation liability is constituted.

9. The November 2023 amendment to the Police Use of Weapons Act fully reverted to the State Compensation Act, thereby applying the intent and negligence principles of the State Compensation Act. This caused the past strict liability to regress back to the fault principle, stepping back from strict liability to general civil law damages. Such regression has deviated from the initial intent of the Police Use of Weapons Act as a manifestation of strict liability.

10. The issue of compensation amounts after introducing strict liability can be controlled and risks externalized through provisions on compensation caps and insurance. The fixed-amount solatium provisions in the enforcement rules of the former Police Use of Weapons Act serve as an example.

11. Conclusion: Incorporating strict liability into state compensation helps perfect the state responsibility system, but the ultimate goal remains achieving the objective of hazard prevention through the establishment of strict liability.

1. The core of the issue discussed today is: who should bear the residual risks after the state has borne them to the best of its ability.

2. Current practice considers the essence of state compensation to be a tort, encompassing fault liability, strict liability, and intermediate liability (shifting the burden of proof). If subjective elements are required, difficulties in proof naturally arise, prompting introspection.

3. The legislative reasoning of Article 2, Paragraph 2 of the State Compensation Act already points out that the occurrence of certain damages falls under the scope of social relief—namely the scope of social law mentioned by the speaker. If the legislature fails to actively legislate in this regard, victims will have nowhere to seek compensation.

4. Regarding the aforementioned difficulties in proof of burden, judicial practice has developed the concept of "presuming fault from unlawfulness" (Supreme Court 99 Tai-Shang-Tzu No. 836 Civil Judgment), and even "viewing unlawfulness as fault." Regarding subjective elements, this is indeed gradually becoming unfavorable to agencies.

5. However, the shifting of the burden of proof still requires manifest unfairness as an element. Therefore, whether people must first prove manifest unfairness before having the opportunity to apply the aforementioned effect of shifting the burden of proof remains questionable, which is why there are suggestions to directly delete the subjective elements. Although in the panelist's experience handling cases, such doubts do not exist, and judges have currently not been seen demanding people to prove manifest unfairness.

6. This civil version legislatively solidifies judicial progress. Beyond that, ideally, the speaker's suggestions are fully agreed with, but on the execution level, one might encounter more difficulties in balancing interests, questions of whether state finances can bear the effects after the amendment, legislative techniques, and whether operational capacities can shoulder it.

1. The issue currently discussed—whether to progress from fault liability in civil torts to strict liability or intermediate liability—was first discussed in Germany regarding environmental law, namely the German Environmental Liability Act, which involves concepts such as strict liability, liability insurance, and fixed-amount compensation. These should be complete supporting measures that must be in place when demanding the responsible party to bear strict liability for damages.

2. Characteristics of environmental damage: non-one-time damage. Therefore, in 2007, the EU required clearly listing hazardous operators, presuming fault, and demanding not only the repair of environmental damage but also the prevention and mitigation of environmental damage. This seems to echo the speaker's proposal for special norms in specific fields distinct from civil torts.

3. Complementary measures such as compensation, indemnification, and insurance should be considered and comprehensively weighed together when thinking about risk burden and allocation.

4. The panelist mentioned earlier that there is already a trend in practice to presume fault from unlawful administrative acts, but how should situations be handled where state acts are completely devoid of any illegality? For example, facility licensing is lawful, yet still causes damage (perhaps because the lawful licensing standards were too low, or due to subsequent unlawful acts by the license holder). Is it possible to consider legislatively incorporating specific facilities into strict liability? It is worth considering, but whether legislative techniques can accommodate it must also be weighed.

5. Even if, as suggested by speaker Professor Lin, unified legislation is enacted in the State Compensation Act listing comprehensive consideration factors, adjudication by courts still relies on administrative agencies explaining the meaning and purpose of their duty norms before judgments can be made. This also highlights the necessity for state compensation adjudication to return to the administrative trial system, enabling courts to exercise higher-intensity investigative powers to demand explanations from agencies.

6. In conclusion, consideration should be given to the distance the State Compensation Act should maintain from traditional civil torts. To achieve the ultimate goal of risk allocation, it should break free from the inherent shackles and limitations of civil torts. Procedurally, state compensation adjudication should return to the public law trial system.