Research

I study how Initial Public Offering (IPO) decisions propagate through firm networks formed by common investors. Using a dynamic network constructed from Crunchbase data for 2000--2023, I show that firms are more likely to go public after directly connected peers complete IPOs. These spillovers are stronger when firms share a lead investor, but weaker when their peers have already been exposed to IPO activity through more distant network connections. Peer IPOs are also associated with higher valuations and share prices, and with firms subsequently attracting new investors and receiving additional funding. I develop a dynamic learning model in which peer IPOs provide connected private firms with information about how public markets value their common investors. This learning raises the value firms expect to obtain from going public and can account for the main empirical patterns, including stronger spillovers through important investors and weaker effects when similar information has already arrived through the network.

I study how informal risk-sharing networks shape households' decisions to undertake risky internal migration. I develop a network model in which households face uncertain rural income, choose whether to migrate at a cost, and share resources with connected households. These links provide insurance, but they can also reduce the private return to migration when migration income is shared. Under CARA preferences, neighbors' migration choices do not affect a household's own decision. Under DARA preferences, however, resources received through the network change households' willingness to bear migration risk. When remittances are shared, a neighbor's migration lowers the migration threshold, generating strategic complementarity. Forming an additional link has an ambiguous effect, reflecting the trade-off between greater insurance and stronger sharing obligations.

This paper studies how Social Media intermediation affects professional information production and voters’ attention. Traditional Media chooses the precision of professional news, while Social Media chooses how faithfully to transmit this information to its users. Voters then decide how much additional costly information to acquire. Social Media also provides personalized information, so its users may be more or less informed and attentive depending on whether personalization offsets the loss from intermediation. Social Media penetration changes Traditional Media’s incentives to invest in reporting quality, while transmission quality shapes voters’ residual uncertainty and attention. Revenue sharing creates a further trade-off: a larger share accruing to Traditional Media strengthens its incentive to produce information but weakens Social Media’s incentive to transmit it. As a result, professional reporting quality can be highest at an intermediate revenue share.