Structural time series models and synthetic controls—assessing the impact of the euro adoption

So far, empirical research on an ex-post benchmark of the euro adoption has relied on the synthetic control method by Abadie & Gardeazabal (Am Econ Rev 93:112-132, 2003) and Abadie et al. (J Am Stat Assoc 105:493-505, 2010, Am J Polit Sci, 59:495-510, 2015). However, the evidence obtained with this method is not overly consistent, leading to the conclusion that the method is not too robust to different settings of the adjustment screws. Using a new method developed by Harvey & Thiele (J Appl Econ 36:71–85, 2021) based on structural time series models, I find that France and Italy are clear losers from the euro while there are no real winners until 2019. Spain, Netherlands, Greece gain in the period before the financial crisis, but afterwards they lose. In fact, only the German economy is robust to the two crises but it loses until 2008. Relating the theories of optimum currency areas to the estimated gaps of the euro adopters from their synthetic controls, I find that openness, real convergence, and net migration are the main drivers of gains from a euro adoption, while the main drivers of losses are low levels of competitiveness, fiscal instability and labour market rigidity. Examining the crisis channels of the financial and euro crisis shows that fiscal instability, labour market rigidity and also business cycle synchronization cause large losses during the crises. Net migration helps to dampen these shocks. While openness is beneficial during the pre-crises period, it cannot help dampen the shock of the crises.