Sachs, Warner (1995) were among the first to claim that «resource curse» is real and that resource abundant economies do indeed grow more slowly than the others. Hundreds of papers were published since then supporting the «resource curse» thesis and offering new explanations of mechanisms and effects that may inhibit growth in resource rich economies. Several recent papers, however (Alexeev, Conrad, 2005; Stijns, 2005; Brunnschweiler, 2006) question the mere existence of the «resource curse» and make it necessary to reconsider the hypotheses about the impact of resource abundance on economic growth. This paper compares various theories of «resource curse» with a special focus on models allowing for the varying — positive or negative — impact of resources on development depending on the quality of institutions and economic policies. Several mechanisms leading to a potentially inefficient use of resources are being examined; it is demonstrated that each of these mechanism is associated with market imperfections and can be «corrected» with appropriate government policies. Empirical evidence seems to suggest that resource abundant countries have on average lower budget deficits and inflation, higher foreign exchange reserves and higher inflows of FDI. Besides, lower domestic fuel prices that are typical for resource rich countries, have a positive effect on long term growth even though they are associated with losses resulting from higher energy intensity. On top of that resource abundance allows to reduce income inequalities. So, on balance, resource wealth turns out to be conducive to growth, especially in countries with strong institutions. However, resource abundance makes democratic political regimes very unstable — they tend to gravitate towards authoritarianism. A game theoretical model is developed to show that democracy in resource abundant countries is inherently unstable and the empirical evidence on the stability of resource democracies is provided.