Phormium tenax (harakeke, New Zealand flax: Asphodelaceae) has long been considered indigenous to New Zealand (including the Chatham Islands) and Norfolk Island. However, the indigeneity of P. tenax on Norfolk Island, in particular, has been challenged by an alternative hypothesis that it was introduced by East Polynesians prior to European colonisation. We tested this alternative hypothesis using a dated phylogenetic tree. We also tested whether dated phylogenetic trees of Phormium could be reconciled with vicariance explanations of its distribution. We examined near-complete plastid genome sequences of P. tenax, P. cookianumand related plants. We then undertook Bayesian and likelihood estimation of the age of the divergence between Norfolk Island and New Zealand accessions using fossil calibration, and separately using biogeographic calibration assuming vicariance explanations of trans-oceanic distributions. DNA sequences of Norfolk Island plants were invariant and nested well within the wider diversity of P. tenax. Estimates of divergence times using fossil calibration did not exclude a common ancestor as recent as the second millennium CE. DNA sequences of Chatham Islands P. tenax were also nested within the diversity of P. tenax from New Zealand, but age estimates were older for their divergence from New Zealand plants (around 20,000–400,000 years). Biogeographic calibrations resulted in extremely ancient ages (tens of billions of years) of deeper nodes within the tree or several orders of magnitude variation in rates among lineages. Our results are consistent with translocation of harakeke by East Polynesian people, but our analyses cannot exclude a Late Quaternary natural dispersal event, which might result in similar genetic patterns. Biogeographic calibrations based on the break-up of Gondwana imply major departures from contemporary ideas of Earth’s history, or orders of magnitude rate variation among lineages.