Within post-Keynesian macroeconomic theory, the contribution by Marglin and Bhaduri (Bhaduri and Marglin 1990; Marglin and Bhaduri 1990) on the relationship between income distribution and growth has progressively asserted itself as a benchmark model, a reference point that has originated and still gives rise to plenty of theoretical and empirical works. Given this popularity, in the related literature it is often claimed that the only open question left is an empirical one: to assess econometrically whether a particular economy is wage- or profit-led. In this essay, I will argue that some theoretical issues, related to this model and to the literature inspired by it, can nonetheless be raised. In particular, the treatment of investment appears to be the least convincing aspect of the approach à la Marglin–Bhaduri. More specifically, it seems possible to raise some doubts about an independent long-run influence of the profit rate or of the profit share on investment, an influence that is not in general justified or explained in detail by this literature and that to some extent is simply taken for granted. It will be shown that, if the Marglin–Bhaduri model is integrated with an explicit consideration of the autonomous components of demand, income distribution does not exert any permanent influence on the rate of growth of the economy and on the rate of accumulation. Matching this result with the usual assumption, made in post-Keynesian models of growth and distribution, that capacity utilization is the adjusting variable in equilibrating investment and savings leads to paradoxical results that question the plausibility of an accumulation function like the one used in the Marglin–Bhaduri model.