The causal factors of revolution and civil war are straightforward to propose. The interesting part is the quantitative analysis; the validation of the causal model.

The author goes into more detail in Modeling Social Pressures Toward Political Instability (https://escholarship.org/uc/item/6qp8x28p)

On a fun note, I was reading the Wikipedia article on cliodynamics (the discipline whose name the author coined) and saw that the article drew an apt comparison between cliodynamics and Asimov's psychohistory.

I recommend reading the full paper linked by esafak. For those without the time, the brief summary is that political stress is a multiplication of (1) the likelihood of the general populace to mobilize, (2) the likelihood of elites to mobilize, and (3) financial distress at the national level. The primary drivers of each of these are:

(1) real income (its inverse, actually), % of the population that is urban, and % of people in their 20s. i.e. If real income declines, urban population % goes up, or % of the population that is young increases, the mobilization factor goes up.

(2) real income of elites (inverse, again), and elite competition for government offices. i.e., If incomes of elites go down or competition among elites for government offices goes up, the mobilization factor goes up.

(3) debt to GDP ratio, and distrust in the state. i.e., If debt to GDP goes up or people trust the state less, the financial distress factor goes up.

The author provides a worrying chart showing an increasingly steep spike in the overall political stress level of the US, but it stops at 2013 (when the paper was published). I would argue that the financial distress factor has gotten substantially worse in the intervening 12 years, but the the other two factors may have declined due to the resumption of real income increases starting in 2015.