Is PbZrO$_3$ really antiferroelectric?

Lead zirconate (PbZrO$_3$) is considered the prototypical antiferroelectric material with an antipolar ground state. Yet, several experimental and theoretical works hint at a partially polar behaviour in this compound, indicating that the polarization may not be completely compensated. In this work we propose a simple ferrielectric structure for this lead zirconate. First-principles calculations reveal this state to be more stable than the commonly accepted antiferroelectric phase at low temperatures, possibly up to room temperature, suggesting that PbZrO$_3$ may not be antiferroelectric at ambient conditions. We discuss the implications of our discovery, how it can be reconciled with experimental observations and how the ferrielectric phase could be obtained in practice.