This paper develops a limit-cycle-based theory of debt fluctuations through a simple endogenous growth model. Public debt and deficit are introduced by relaxing the balanced-budget rule hypothesis, and assuming a simple fiscal rule. Our main result is that fiscal rules can be destabilizing, leading to (i) multiple equilibria-four balanced-growth paths can emerge-, (ii) endogenous public debt cycles, which appear both in the short and the long run, and (iii) hysteresis phenomena arising from extreme sensitivity of changes in parameters. We also reveal that a balanced-budget rule does not preclude large aggregate fluctuations. Finally, our calibration exercise highlights that our model produces asymmetric cycles consistent with observed stylized facts.