Although broadband is essential for modern economic activity, it remains unaffordable for many U.S. households, prompting historic public investments. This paper evaluates the effects of two flagship federal initiatives——the Affordable Connectivity Program (ACP) and the Broadband Equity, Access, and Deployment (BEAD) program——on welfare and market structure. Using detailed data on markets, providers, and broadband products, I uniquely combine a continuous-treatment difference-in-differences design with a structural model of demand, entry, and product choice to quantify impacts and counterfactuals. The ACP, a monthly household subsidy, increases social surplus and generates benefit-cost ratios of 1.28–1.91 per dollar spent, while leaving the market structure largely unchanged. In contrast, the BEAD program subsidizes providers’ fixed costs, inducing firm entry and expanding product variety. As firms extend services to high-cost areas, marginal costs and prices increase, reflecting deployment costs. Welfare gains per dollar spent range from $6.7 to $10.7 and peak at intermediate subsidy intensities. These findings reveal a key policy trade-off: demand-side subsidies yield immediate, cost-effective affordability gains, while supply-side subsidies reshape market structure for long-run efficiency, although at higher fiscal costs.
This paper introduces a new method that integrates machine learning techniques with traditional econometric tools to correct bias in two-step estimation for dynamic discrete choice models under incomplete information. Unlike most existing approaches that primarily address bias from conditional choice probabilities while taking the transition matrix as given, this method simultaneously addresses both sources of bias. I propose a two-step estimator that jointly includes the conditional choice probabilities and transition matrix as first-step estimates, providing a robust, non-iterative solution. The estimator achieves square root N consistency and asymptotic efficiency. In addition, I establish a simple connection between Neyman orthogonality and the Zero-Jacobian property, a key result that guarantees consistent and efficient estimators in single-agent models. Monte Carlo simulations using a common benchmark data-generating process demonstrate the good performance of the proposed method.
Mobile money is a core component of Senegal’s financial system, and provides essential transaction services to a largely unbanked population. As these services are delivered through retail agents, competition among operators at the store level directly influences prices and access. The recent entry of Wave, followed by sharp fee reductions from incumbents, highlights the sector’s sensitivity to competitive pressure. This paper analyzes how market structure shapes competitive behavior in Senegal’s mobile money industry. Using a novel dataset of 1,553 retail stores, I estimate a static discrete-choice entry game under complete information and obtain bounds on profitability and competitive effects through moment inequalities. I then use the estimated structure to assess how Wave’s entry affected competitive conditions and to simulate alternative market environments. Preliminary results show that larger stores are more likely to host multiple operators and that competitive interactions among firms are significant. The findings offer the first supply-side evidence of competition in Senegal’s mobile money sector.
We leverage a convex programming strategy and propose a new estimation strategy to estimate a discrete choice models which is robust to standard specifications.