How to assess the growth effect of structural reforms?

This paper surveys tools to analyse the macroeconomic impact of structural reforms, with a focus on two general approaches: (i) econometric estimates, linking indicators of policy change to macroeconomic outcomes, and (ii) structural macroeconomic models simulating the effect of reforms on economic activity and other variables of interest. Both approaches have advantages and drawbacks. The data-driven approach (i) is flexible with respect to the policy variables that can be linked to macroeconomic outcomes, but parameter estimates are sometimes difficult to interpret from the perspective of economic theory. Structural model simulations (ii), by contrast, provide clarity on the transmission channels, but models are simplifications, and reforms must therefore often be approximated by intermediate outcomes, derived from "satellite tools", which tends to be labourious, especially in a multi-country setting, and imperfect. Hence, model-based analysis often recurs to stylised scenarios, such as narrowing the gap towards best performers, illustrating potential gains from reforms instead of analysing actual policy trajectories. The drawbacks of the two approaches suggest that ex-ante analysis of structural reform effects at the macro level will have to remain largely qualitative at present. Without commitment to a detailed reform agenda (measures and coverage) and appropriate metrics providing a reliable quantification of reforms, an ex-ante assessment of structural reforms can provide plausible orders of magnitudes for the effects at best.