A better understanding of why some households get ahead while others fall behind is crucial for achieving further worldwide poverty reduction in future. Who moves in and out of poverty over time has been the focus of a growing literature on household level poverty dynamics. A central theme in this literature has been the role of shocks and their impact on households’ risk management strategies. Risk and uncertainty diminish the current level of households’ economic well-being as well as their prospects for the future. This is particularly true when livelihood-generating activities are highly stochastic, as they are for many rural households in developing countries, and when mechanisms to insure against risk are either non-existent or insufficient. Moreover, the ability to manage risk is becoming more important as economic factors that are key to the poor, e.g., food prices, have become more volatile. Households face two distinct sources of risk: First, covariate risk such as drought, rainfall, pests or civil war affect the entire population. These types of risks have been the focus of much of the existing literature on welfare dynamics. Second, idiosyncratic or household-specific risk such as medical expenses or crop and livestock losses. Evidence suggests that such idiosyncratic risk often dominates covariate risk and that the latter requires different strategies to mitigate. This paper examines the importance of idiosyncratic risk for households in rural Ghana and the effectiveness of various ex ante and ex post risk management strategies in mitigating this risk, including support from social networks, self-insurance in the form of savings, accessing credit and diversifying livelihood activities. This paper uses an unusually rich set of panel data from rural Ghana covering the period between 1997 and 2009 and containing information on a large variety of formal and informal insurance mechanism, including on access to social networks, self-insurance, income diversification and asset sales. It quantifies the impact of different types of idiosyncratic shocks on households’ ability to get ahead, examines the individual and joint effect of households’ access to different types of insurance mechanisms and assesses the relative effectiveness of different risk management strategies on households’ ability to bounce back from negative shocks and to escape poverty. This paper makes three main contributions. First, it examines the impact of idiosyncratic risk on household level welfare dynamics, focusing on how idiosyncratic shocks affect welfare dynamics paths and whether those effects are mitigated differentially by different risk management strategies. Facing uninsured risk ex ante and coping with consequences of risk ex post households employ a range of strategies to minimize their exposure to risk and to smooth their consumption over time. These include mechanism such as self-insurance through savings (and precautionary savings), formal and informal insurance and credit, social networks, and labor market access and income diversification. Existing studies typically only have data for one of these insurance mechanisms. The second contribution of this paper is examine each of these risk management strategies individually as well as jointly. The unique panel data makes it possible to both assess the relative importance of these mechanisms as well as identify complementarities between them. This extension beyond a single risk management mechanism allows a more appropriate modeling of the behavior of households that face risk as when one or more risk management mechanisms are unavailable households resort to other mechanisms. A corollary question addressed is whether there are critical thresholds that establish some minimum effective scale of such risk management strategies. Existing studies impose strong assumptions on the low-order polynomial relationship between risk management strategies and welfare dynamics. This paper’s approach allows for such potentially highly non-linear threshold effects by using innovative semi-parametric panel data estimators. The paper also explores the interaction, substitution and complementarities between different risk management strategies. Third, this paper contributes to the emerging literature on household level welfare dynamics in Subsaharan Africa by providing a case study from rural Ghana. The results suggest varying degrees of effectiveness for the different mechanisms households use to reduce and mitigate risk. Having own savings and being able to draw on larger social networks offers a statistically significantly better chance of overcoming the consequences of negative shocks. Credit access also seems to help though the evidence is less strong. A greater diversity of income sources is associated with greater gains in expenditure but income diversification does not seem to help in overcoming shocks. This could be a reflection of many households diversifying into low-return activities that are positively correlated. Overall the results suggest some effectiveness of these formal and informal insurance mechanisms. However, even in their combination they are not sufficient in helping households overcome negative shocks and to ensure sustained improvement in well-being over time. Thus, while idiosyncratic risks can be partly ensured at the local level, there is a need for policy to supplement these mechanisms through social policy.