Across the developing world, the deforestation of primary, native forests supports economic activity at the cost of environmental degradation with global, long-term consequences. Events like record-setting wildfires in the Amazon highlight the potential for deforestation to drive ecosystems past potentially irreversible tipping points, such as a shift from tropical forest into grassy savanna. Investments in reforestation with secondary forests are one potential tool to avert or delay tipping points, but the success of this strategy may depend on the degree to which secondary and primary forests are substitutes in the production of ecosystem services. We develop a dynamic optimization model and numerical simulation to explore how deforestation, reforestation, and the degree of substitutability between forest types affects the likelihood that a forest system will cross a tipping point. In the context of the Brazilian Amazon, we find that efforts to ensure secondary forests better mimic primary forests only yield a small improvement in terms of avoiding or delaying ecosystem collapse. Rather, the most significant effects on tipping points arise from an increase in the relative costs of clearing primary forests and/or a decrease in the tenure costs associated with protecting secondary forest investments. Our results highlight in particular the importance of tenure costs as a policy target to reduce the risk of ecosystem collapse. (with : Gregory Amacher, Virginia Tech - Philippe Delacote, INRAE - Haoyu Wang, Virginia Tech)