Drawing on new empirical analysis of 30 years of structural reforms across the OECD, this paper sheds light on the impact of reforms over time, identifies the horizon over which their full effects materialise, and investigates whether such effects vary with prevailing economic conditions and institutions. Impulse responses of aggregate outcomes (GDP growth, employment rate) to various labour, product market and tax reforms are estimated at different horizons. This analysis indicates that the benefits from reforms typically take time to fully materialise. When significant effects are found in the short run, reforms seldom involve significant aggregate economic losses; on the contrary they often deliver some benefits. The absence of major depressing effects does not lend support to the view that reforms should be in general accompanied by substantial macroeconomic policy easing in order to deliver some short-term gains. Nevertheless, there is also tentative evidence that some labour market reforms (e.g. of unemployment benefit systems and job protection) pay off more quickly in good times than in bad times, and can even entail short-term losses in severely depressed economies.