We test for arbitrage opportunities and market efficiency in the Hong Kong money, stock, and real estate markets. We find that the money market, represented by the Exchange Fund Bills, stochastically dominates both the stock and real estate markets, represented by the Hang Seng Index and the Property Price Index, respectively, and the real estate market stochastically dominates the stock market. Hence, it appears that sometimes investors could achieve higher levels of expected utility but not higher (expected) wealth by investing in an individual asset rather than a portfolio on the efficient frontier, inferring that there is no arbitrage opportunity in the markets but the markets are not efficient. We further compare the performance of each individual market to the performance of portfolios on the historically derived efficient frontier. We find that the money market second-order stochastically dominates nearly all the efficient frontier portfolios, while, on the other hand, none of the efficient portfolios stochastically dominates the money market. Interestingly, the money market also second-order stochastically dominates equal weighting; that is, the naïve 1/N portfolio. To test the robustness of the effect of COVID-19 in our study, we divide the period in our study into two subperiods: the pre-COVID-19 (sub)period, and the COVID-19 (sub)period. We find that the conclusions drawn from the pre-COVID-19 period are the same as those drawn from the entire period and the conclusions drawn from the COVID-19 period are the same as those drawn from the entire period except that the money market does not second-order stochastically dominate nearly all the efficient frontier portfolios, but only second-order stochastically dominates some of the efficient frontier portfolios. Our findings, therefore, question diversification benefits in the Hong Kong capital market during our sample period between 1993 and 2021, including both the pre-COVID-19 period, and the COVID-19 period. The findings in our study are useful for investors for their investment and useful for policymakers for their policymaking on the assets studied in our paper. The last finding in our paper shows a new puzzle in financial theory that the traditional financial theory is not always held
