Varieties of Forecasts: Predicting Adverse Regime Transitions

Abstract

This article introduces the V-Forecast project, the forecasting intuitive of the Varieties of Democracy (V-Dem) institute. In this the initial year of the V-Forecast project, we provide two-year ahead forecasts of the risk of adverse regime transitions (ARTs) for 169 countries. ARTs are substantial movements of a country's regime towards more authoritarian governance, whether authoritarian reversals in a democracy, or further autocratization in an already nondemocratic country. Examples include Hungary and Poland over the past few years, which are prominent cases in a more widespread and worrying global trend over that effects a signicant fraction of the world's population. Yet so far, there has been no public forecasting system for anticipating new ARTs and identifying countries most at risk. We describe an effort that forecasts ARTs - operationalized using the Regimes of the World (RoW) categorization - with an ensemble model that leverages V-Dem and several additional external data sources. Despite being rare events with a roughly four percent baseline chance over any two-year period, in test forecasts the model is able to achieve good accuracy.