What is the effect of natural resources on economic development? Existing work focuses on what happens when countries experience a positive shock in the form of natural-resource windfall income. By comparison, we know little about how negative revenue shocks affect economic outcomes. This dissertation proposes that the effect of negative shocks on development depends on how political elites perceive the shock. If the shock is long-lasting, elites have an incentive to compensate for lost revenues and improve the institutional environment for economic activities. In contrast, if the decline in revenues is temporary, elites make up for the shock in ways that may worsen the business environment. I support this theory with evidence from two negative shocks that affected Russia's oil- and gas-producing regions. The first shock resulted from a tax reform in 2002 that centralized regional resource revenues. I compare the effect of this institutionalized shock with that of a more transient negative shock in 2014, the result of a commodity price decline. Employing a difference-in-differences design, I show that the first shock led to improvements in oil- and gas-producing regions' business environment and increased private investment in fixed capital and small business entry. In contrast, the post-2014 shock did not stimulate such changes. I explore the mechanisms behind these findings using a most-different case study of two Russian regions: Tyumen' and Tatarstan. These regions are dissimilar in their political, business, and social structures, but both suffered fiscal declines in 2002 and 2014. The regional governments saw the former as a permanent policy shift and the latter as a temporary swing in the market. Consequently, after 2002, both regions offered tax benefits to firms, strengthened their infrastructure for business support, reduced administrative burdens, and transferred federal grants to businesses. They did not make similar improvements after the commodity shock of 2014. In addition to examining Russia’s experience with negative shocks, I estimate the average short-term effect of negative shocks to various types of resource revenues on economic freedom in a cross-national setting. I apply a first-differences model and find that negative shocks are generally associated with improvements in institutions and economic freedom.