Crisis? What crisis? Measuring economic crisis in political science

Abstract

An influential body of scholarship in political science has investigated the impact of economic crisis on various political outcomes. The vast majority of these studies rely on annual growth rates (AGR) to specify economic crisis. I argue that this canonical approach comes with several logical shortcomings. It leads to misguided impressions of crisis severity; it makes no distinction between rapid expansion years and rapid recovery years; and it disregards the financial dimension of economic crises. I present and discuss three alternative approaches of measuring economic crisis, imported from economics: economic shocks, economic slumps, and measures of financial crises. Examples from the regime instability literature demonstrate that these alternative crisis measurements provide results that are theoretically more nuanced and empirically more robust. On this basis, the article encourages researchers to pay more attention to the way they measure economic crisis in general and to supplement the AGR approach with alternative crisis measures in particular.

Economic crisis and regime transitions from within

“…Our benchmark measure is therefore continuous GDP per capita growth in a year. 4 Yet, as discussed by Krishnarajan (2019a), this approach also has its shortcomings. We thus test multiple alternative measures, including dummy operationalizations requiring different thresholds for depth and longevity for an economic contraction to be registered as a crisis.…”

Commodity terms of trade shocks and political transitions

“…Additionally, the change in the moving-average index helps us to identify protracted income declines. For example, 5% growth in the current CTOT index following 2 years of 10% decline would result in  CTOT Δ −5% jt and indicate that the terms of trade overall deteriorated during the last 3 years (Krishnarajan, 2019b In Figure 2, we depict the average polity2 democracy scores for the six regions of countries in our dataset since 1962. 8 Remarkably, all of the six regions are characterized by S-shaped evolutions.…”

“…Additionally, the change in the moving‐average index helps us to identify protracted income declines. For example, 5% growth in the current CTOT index following 2 years of 10% decline would result in and indicate that the terms of trade overall deteriorated during the last 3 years (Krishnarajan, 2019b). 7…”