The purpose of the current study is to explain the unstoppable dramatic increases in emissions by proposing a new hypothesis as carbon hysteresis hypothesis. If the emissions have a path dependence or a unit root, a given shock to the emissions will make its equilibrium to a new level in long-run. In this case, the carbon hysteresis hypothesis will be valid. On the other hand, if the emission series is stationary, emissions will resist given shocks in long-run, hence there will be no hysteresis effect. Furthermore, this study reveals the direction of hysteresis. Accordingly, the positive hysteresis indicates continuous increase in emissions while the negative hysteresis indicates continuous decrease in emissions that means carbon reduction.

This study achieves robust results by three different structural break panel unit root strategies. In doing so, the panel unit root tests are performed to test carbon hysteresis on the top-five emitting countries: China, India, Japan, Russian Federation and USA over the time period 1965–2020.The main conclusion is that the carbon hysteresis hypothesis is valid for these countries. Moreover, we estimated six different trend models for each regime of sample period for all countries to determine the direction of the hysteresis. In conclusion, this study proposes a new hypothesis to assist countries on their way to the reduction targets.