State public pension systems manage over 6 trillion in assets while facing approximately 1. 3 trillion in unfunded liabilities, raising fundamental questions about governance effectiveness and fiduciary accountability. This Article examines these challenges through the analytical framework of governance, applying insights from public choice, principal-agent models, and corporate governance to argue that details in implementation significantly impact outcomes. We offer Ohio’s State Teachers Retirement System (STRS) as a case study. Recent governance reforms in Ohio—including legislative restructuring of the STRS board and enactment of fiduciary standards legislation modeled on the American Legislative Exchange Council’s framework—suggest the potential for earlier governance problems. Reported investment returns exceeded audited returns in 19 of 20 fiscal years examined (2003-2022), with an average annual overstatement of 0. 33 percentage points and a single exception in fiscal year 2020, resulting in approximately 9. 3 billion in cumulative overstatement. By contrast, the Ohio Public Employees Retirement System (OPERS) exhibited minimal, bidirectional discrepancies (0. 08 percentage points), consistent with measurement variation rather than systematic bias. This divergence reflects predictable institutional failures: performance bonuses tied to unaudited metrics create concentrated benefits for administrators while dispersing costs across beneficiaries and taxpayers. This Article proposes tying performance-based compensation exclusively to CPA-audited returns derived from net fiduciary position, arguing that this incremental reform would reduce information asymmetry and restore transparent accountability without requiring wholesale institutional restructuring. This analysis contributes to ongoing scholarship on pension governance by demonstrating how institutional design predictably shapes reporting behavior and by offering grounded reform that addresses incentive misalignment while preserving pension system architecture.
Mendenhall et al. (Fri,) studied this question.