I am trying to estimate the long run relationship between fiscal policy an economic growth, what is the best way to deal with reverse causality besides finding an IV, in a time series approach (ARDL, ECM, VAR).
The ony way I can imagine is limiting yourself to episodes when fiscal policy is exogenous to economic growth. Check the literature around the work of Alesina and others few years ago about expansionary austerity.
Thanks for the quick answers and references. Additionally i wanted to ask to Jose Perez-Montiel and José-Ignacio Antón , from what i have read in the literature, both of these approaches are used to estimate short-medium run impacts and fiscal multipliers. Do you know if both methods/approaches can be applied to test whether fiscal policy has more of a long run impact on growth. Like some of the endogenous growth models from macroeconomic theory.
For the long run, I don't know a well stablished method besides the usual ones (FMOLS, DOLS, GMM, IV). However, I think your question ties in the so-called "Wagner's law" literature. If you search for empirical papers on this topic, you will probably find something of your interest.