In Minsky's Financial Instability Hypothesis (FIH), financial fragility of non-financial firms tends to increase endogenously over the cycle along with the macroeconomic leverage ratio. This analysis has been criticized for two main complementary reasons: firstly, it does not duly consider the aggregate pro-cyclicallity of profits; secondly, due to an overly aggregate analysis, some inferences about the relation between aggregate leverage and systemic fragility are potentially misleading. In this paper, we take these criticisms into account by building an agent-based stock-flow consistent model which integrates the real and financial sides of the economy in a fundamentally dynamic environment. We calibrate and simulate our model and show that the dynamics generated are in line with empirical evidence both at the micro and the macro levels. We create a financial fragility index and examine how systemic financial fragility relates to the aggregate leverage along the cycle. We show that our model yields both Min-skian regimes, in which the aggregate leverage increases along with investment, and Steindlian regimes, where investment brings leverage down. Our key findings are that the sensitivity of financial fragility to aggregate leverage is not as big as assumed in the literature; and that the distribution of profits amongst firms does matter for the stability of the system, both statically (immediately for financial fragility) and dynamically (because of the dynamics of leverage).