To begin with, we would like to thank Peter Skott for extensive discussions about the different approaches to the topic of cyclical growth. We have benefitted greatly from his detailed comments. As Skott outlines, the Goodwin–Kaldor and Goodwin–Kalecki frameworks share many features, and it is our hope to continue this constructive exchange.

Our argument was laid out partly in response to Skott and Zipperer (2012). Therein, a model is presented as a ‘Kaleckian’baseline. That model does not seem to be the relevant starting point for discussion of a Kaleckian cyclical growth model. Rather, in our opinion, Barbosa-Filho and Taylor (2006) provide that starting point–not despite, but because of its Goodwinian features. We have extended the latter to include an unstable but bounded accelerator, to show that it is straightforward to generate cyclical growth in that framework.