This study investigates the effects of the World Bank's exogenously-determined income threshold for eligibility for concessionary International Development Association (IDA) loans on the allocations of bilateral donors. The donors might interpret the World Bank's policies and allocations across recipients as informative signals of where their own aid might be used most effectively. Alternatively, other donors might compensate for reduced IDA allocations by increasing their own aid. This paper shows that the signaling effect dominates any crowding out effects. The analysis uses panel data with country fixed effects and finds that aid from the bilateral donor countries is significantly reduced after countries cross the IDA income cutoff, controlling for other determinants of aid. Allocations by other donors are not sensitive to actual IDA disbursements, only to the IDA income threshold. Because crossing the income cutoff for eligibility significantly reduces aid levels from other donors as well as from the World Bank, government officials in recipient countries may have an incentive to manipulate their national accounts data to understate per capita income when it is near the IDA threshold. However, tests for "bunching" of observations just below the income threshold find no evidence to support data manipulation concerns. These findings suggest that graduation from IDA should be an even more gradual process than it already is, to dampen the sharp drops in aid experienced by countries after crossing an arbitrary income threshold.