A bstract In the near future, the neutrinoless double-beta (0 νββ ) decay experiments will hopefully reach the sensitivity of a few meV to the effective neutrino mass | m ββ |. In this paper, we tentatively examine the sensitivity of future 0 νββ -decay experiments to neutrino masses and Majorana CP phases by following the Bayesian statistical approach. Provided experimental setups corresponding to the experimental sensitivity of | m ββ | ≃ 1 meV, the null observation of 0 νββ decays in the case of normal neutrino mass ordering leads to a very competitive bound on the lightest neutrino mass m 1 . Namely, the 95% credible interval in the Bayesian approach turns out to be 1 . 6 meV ≲ m 1 ≲ 7 . 3 meV or 0 . 3 meV ≲ m 1 ≲ 5 . 6 meV when the uniform prior on m 1 / eV or on log 10 ( m 1 / eV) is adopted. Moreover, one of two Majorana CP phases is strictly constrained, i.e., 140° ≲ ρ ≲ 220° for both scenarios of prior distributions of m 1 . In contrast, if a relatively worse experimental sensitivity of | m ββ | ≃ 10 meV is assumed, the constraint on the lightest neutrino mass becomes accordingly 0 . 6 meV ≲ m 1 ≲ 26 meV or 0 ≲ m 1 ≲ 6 . 1 meV, while two Majorana CP phases will be essentially unconstrained. In the same statistical framework, the prospects for the determination of neutrino mass ordering and the discrimination between Majorana and Dirac nature of massive neutrinos in the 0 νββ -decay experiments are also discussed. Given the experimental sensitivity of | m ββ | ≃ 10 meV (or 1 meV), the strength of evidence to exclude the Majorana nature under the null observation of 0 νββ decays is found to be inconclusive (or strong), no matter which of two priors on m 1 is taken.