Traversing the Short-Sale-Ban and the No-Ban Regime : A New Tale of the Old Overvaluation Story

Recent studies add to the evidence against Miller's (1977) overvaluation theory, further dwarfing the already sparse evidence for the theory. In this paper, we identify several issues untouched so far. Addressing these issues simultaneously, we conceive a novel empirical identification strategy which (1) looks at short-sale bans rather than short-sale constraints; (2) treats a short-sale ban (or no ban) as a regime rather than as an event; (3) probes the same stock that traverses the ban and no-ban regimes rather than different stocks concurrently under different regimes or the same regime; and (4) allows for possible endogeneity in regulators' decision to change the short-selling regime for a stock. Employing data from the Hong Kong market, we find robust and strong evidence for Miller's theory. Stocks earn significantly higher abnormal returns in the ban regime than in the no-ban regime. The ban regime and wider dispersion of investor opinions reinforce each other in further increasing current, while further reducing subsequent, returns to individual stocks. The tick rule acts as a short-sale constraint to impair the negative price effects of the no-ban regime