Digital platforms shape not only what news people see but also what news outlets choose to produce. This paper quantifies the equilibrium effect of algorithmic curation on the supply of information. Linking print and online headlines from major U.S. newspapers, I document that online headlines are systematically more ideologically slanted and carry more emotional tone than their print counterparts. Exploiting Facebook’s announcement of a major algorithmic change as a quasi-experiment, I show that online headline slant immediately converged toward print slant, consistent with platform algorithms shaping newsroom output. Using Facebook posts and engagement data, I build a structural model of users, algorithms, and media outlets. I find that readers value both like-minded and unexpectedly slanted posts, but platforms overweight these forces, accounting for about 20% of the slant in user exposure.  Media outlets add extra online slant solely to maximize the viewership allocated by the platform. Counterfactual simulations show that removing the algorithm’s amplification of ideological signals would eliminate roughly 90% of excess polarization and substantially flatten exposure across users with different ideologies. These results provide the first quantitative evidence that platform algorithms play a central causal role in shaping the supply of news slant, and highlight a trade-off between engagement and informational diversity that is central to the design and regulation of digital platforms.

We examine the strategic interaction between an expert (principal) maximizing engagement and an agent seeking swift information. Our analysis reveals: When priors align, relative patience determines optimal disclosure—impatient agents induce gradual revelation, while impatient principals cause delayed, abrupt revelation. When priors disagree, catering to the bias often emerges, with the principal initially providing signals aligned with the agent’s bias. With private agent beliefs, we observe two phases: one engaging both agents, followed by catering to one type. Comparing personalized and non-personalized strategies, we find faster information revelation in the non-personalized case, but higher quality information in the personalized case.

Central limit order books, standard in traditional exchanges, are impractical on blockchains. Decentralized exchanges like Uniswap and Curve instead use Automated Market Makers (AMMs) that set prices via functions of token quantities. To study price discovery and market impact in this setting, we characterize the optimal liquidity provision of AMM participants. Theoretically and empirically, we show that price impact depends on both trade size and the dynamics of liquidity provision. Using data from 31 large Uniswap v2 pools, we find that liquidity providers adjust positions in ways that offset prior trades, especially when those trades are likely uninformed.