Can Commercial LLMs Be Parliamentary Political Companions? Comparing LLM Reasoning Against Romanian Legislative Expuneri de Motive

TL;DR

This study assesses the reasoning capabilities of various commercial LLMs against Romanian legislative explanations, revealing strengths in frontier models and limitations due to contextual ignorance and confabulation.

Contribution

It introduces a novel evaluation framework comparing LLM reasoning to official legislative reasoning and conceptualizes the legislator-LLM relationship through principal-agent theory.

Findings

Frontier models outperform open-weight models in semantic similarity scores.

All models show task-dependent confabulation, especially on politically idiosyncratic proposals.

Failures in reasoning compound across bounded rationality levels, influenced by training data coverage.

Abstract

This paper evaluates whether commercial large language models (LLMs) can function as reliable political advisory tools by comparing their outputs against official legislative reasoning. Using a dataset of 15 Romanian Senate law proposals paired with their official explanatory memoranda (expuneri de motive), we test six LLMs spanning three provider families and multiple capability tiers: GPT-5-mini, GPT-5-chat (OpenAI), Claude Haiku 4.5 (Anthropic), and Llama 4 Maverick, Llama 3.3 70B, and Llama 3.1 8B (Meta). Each model generates predicted rationales evaluated through a dual framework combining LLM-as-Judge semantic scoring and programmatic text similarity metrics. We frame the LLM-politician relationship through principal-agent theory and bounded rationality, conceptualizing the legislator as a principal delegating advisory tasks to a boundedly rational agent under structural information asymmetry. Results reveal a sharp two-tier structure: frontier models (Claude Haiku 4.5, GPT-5-chat, GPT-5-mini) achieve statistically indistinguishable semantic closeness scores above 4.6 out of 5.0, while open-weight models cluster a full tier below (Cohen's d larger than 1.4). However, all models exhibit task-dependent confabulation, performing well on standardized legislative templates (e.g., EU directive transpositions) but generating plausible yet unfounded reasoning for politically idiosyncratic proposals. We introduce the concept of cascading bounded rationality to describe how failures compound across bounded principals, agents, and evaluators, and argue that the operative risk for legislators is not stable ideological bias but contextual ignorance shaped by training data coverage.