In Marc Lavoie's Post-Keynesian Economics: New Foundations, he has an interesting discussion in Section 6.10.4, which is labelled "Minsky or Steindl Debt Dynamics?" The Minsky dynamics are the well-known Financial Instability Hypothesis (link to primer), while the Steindl dynamics refers to the discussion in Maturity and Stagnation in American Capitalism by Josef Steindl. Lavoie's discussion raises some issues with the limitations of aggregated analysis in this context. This is a brief comment on this topic.
Lavoie's discussion is based on some calculus that demonstrates that we can have two stable regimes for growth: one in which the debt-to-capital ratio is rising, and one where it is falling. He identifies the rising debt ratio regime as being associated with Minsky, and the falling debt regime with Steindl.
As I argued in my discussion of the Financial Instability Hypothesis, its most natural interpretation is in terms of agent-based models. (Minsky used the term "unit" in his texts extensively, so "unit-based models may be more fitting.) In a growing economy, it is unclear that aggregate debt ratios will match the intuition that higher risk lending always corresponds to higher debt ratios.
This is because in an environment of fast growth, debt principal can be "inflated away" by greater nominal income growth (which may either be inflation or real growth). Even if interest rates are raised to compensate for higher inflation, the principal amount is still being whittled away by inflation, which shows up in the ratio.
In order to control for these effects, we need to look at the trends in debt service ratios to get a better handle on the riskiness of lending.
Furthermore, even if inflation is not reducing debt ratios, the aggregates may be hiding trends for risk. Although the usual framing of the Financial Instability Hypothesis is that there is an across-the-board rise in risk-taking, it seems possible that dominant mature industries could reduce leverage as they no longer face large fixed investments to finance. This could reduce debt ratios in aggregate as newer industries pursue more aggressive strategies. Since the credit markets are sensitive to tail risk, it does not matter much if low risk borrowers become even lower risk if troubles are brewing elsewhere.
In summary, we need to be cautious about putting too much emphasis on aggregate debt ratios as a shorthand for riskiness of borrowing.
(c) Brian Romanchuk 2019