Regulatory Constraints and Market Equilibrium: Evidence from Hong Kong's IPO Allocation Reform
Year: 2026
Type: preprint
Abstract: Allocation regulation does more than reserve shares for particular investor groups. By changing the contracts available to issuers and underwriters, it changes the equilibrium division of committed capital, informed order collection, and retail participation. Hong Kong's August 2025 IPO reform makes this distinction visible. It required at least 40% of initially offered shares to enter non-cornerstone bookbuilding and introduced a choice between demand-contingent public clawbacks and a fixed, no-clawback alternative. We reconstruct allocation decisions for 161 operating-company IPOs in the first post-reform year, with exact final three-part allocations for 146. After prospectus-level and exchange-source reconciliation of 20 Chapter 18C specialist-technology issuers, 139 of 141 ordinary offerings selected the no-clawback Mechanism B; all 139 began with a 10% public tranche. Among 112 B offerings with cornerstone participation, 32 clustered within one percentage point of the initial 40–50–10 vertex, including 14 within 0.1 percentage point. The propensity to select the main near-vertex category rises from 10.7% to 60.7% across the smallest and largest proceeds quartiles, and the positive size gradient survives tighter vertex definitions. Of the 29 initially near-vertex offerings with exact final allocations, 25 remain near the vertex. A two-stage contracting model connects selection of the regulatory mechanism with conditional selection of the allocation vertex: statutory minima shape the feasible contract, while the 50% residual cornerstone tranche is not itself legislated. The secondary-market evidence reveals a separate price-adjustment margin: median first-day gain from the offer is 22.49%, while in the 160-issuer sixty-session panel, adjusted for documented share subdivisions, median aftermarket log performance relative to the Hang Seng Index is −7.57 percentage points, with 94 issuers underperforming. Its simple two-sided sign-test p-value of 0.0325 is sensitive to horizon multiplicity, independent-price matching, and overlapping calendar windows. Conditional regressions do not precisely identify an allocation–aftermarket association, particularly given limited mechanism-choice comparability. Regulation can standardize primary-market contracts while leaving subsequent valuation to a distinct, heterogeneous price-discovery process.
Authors: Yang, Fuli, Liu, Yu
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7225878182
