Presentation Title: State subsidies and the spatial allocation of production: Evidence from the U.S. manufacturing industry State governments use production subsidies to attract companies and facilitate economic development. Such programs benefit state residents by increasing local labor demand, but may also encourage firms to pursue suboptimal production strategies. To assess the net welfare impact of these competing effects, I develop a general equilibrium framework with multiregional production, rich firm heterogeneity, and production subsidies that vary across states and between firms. I show that model predictions are consistent with evidence from the US manufacturing sector, and calibrate the model using a representative sample of firm-level balance sheet and subsidy data. I find that eliminating subsidies would increase total US welfare by 1.1%. While net positive for the nation as a whole, eliminating subsidies hurts peripheral states in the Deep South and Northwest, which offer the highest subsidies as a share of output.

Presentation Title: Demographic distributional effect of high-speed railway: Evidence from Taiwan High-Speed Railway (HSR) has become an important mode of transportation in many countries and regions since the end of the Second World War, but its economic and demographic distributional effects have not yet been fully understood. Employing Taiwan’s data as the empirical example, this paper systematically evaluates the demographic effects of the HSR developments. To this end, this paper first revises the traditional gravity model of population migration by using econometric techniques such as interaction-augmented regression discontinuity design (RDD). With the modified model, this paper discovers that Taiwan HSR significantly increased the level of demographic interactions between different counties. Nonetheless, applying mathematical methods used in the trade balance econometric equations, it is found that the increases in migration from suburban and rural areas to cities are larger than those in the opposite direction, indicating that Taiwan is changing from experiencing demographic dispersion to urban agglomeration due to the availability of the HSR system. To substantiate these results, this paper constructs the counterfactuals of the log level population sizes of two peripheral HSR counties with in the absence of the HSR project, and confirms the severity of demographic decline faced by the non-urbanized areas.

Presentation Title: An empirical asset pricing model: Accounting for the sector-heterogeneity of risk Stock returns are generally difficult to explain, as they’re comprised of many discrete channels of risk. Empirical Asset Pricing Models (EAPM), such as the Fama-French Five-Factor model (FF5), are used to partition these channels across a series of systematic risk-factors - such as company size (total market equity), value (book-to-market ratio), investment, and operating profitability. Prior EAPMs only account for how such factors contribute to risk at the market-level, ignoring any potential variation across sectors. In this paper, we propose a Sector-Heterogenous Model (SHM) which directly accounts for this variation, by generalizing the Fama French methodology to sector-subsets of stocks. We find that risk is meaningfully heterogenous across sectors for each of the factors in the FF5, with different subgroups of factors being statistically significant within each sector. In a direct comparison of explanatory power, we find that the SHM outperforms the FF5 - improving Adjusted-R2 by an average of 5% for stocks across all sectors. We then demonstrate applications of sector-heterogeneity to stockpicking, developing a High-Beta portfolio strategy using the SHM-Beta which outperforms the S&P 500 in backtesting. We conclude that there is meaningful sector-heterogeneity of risk in the market, and that this information is materially useful to investors.

Presentation Title: Crop production and climate change: The importance of temperature variability The impact of climate change on agriculture has risen to the forefront of academic inquiry across the natural and economic sciences. In this paper, we call attention to the manner in which researchers specify the relationship between warming temperatures and crop production. While many have investigated the effect of rising mean temperatures on crop production, fewer have considered measurements of temperature variability: daily minimum temperature, daily maximum temperature, and diurnal temperature range. In this paper, we use state-level panel data from the U.S. Department of Agriculture to compare econometric models of crop production specified only with mean temperature against those that also feature temperature variability. We find that the inclusion of temperature variability produces more optimistic estimates of the causal effect of climate change on American crop production from 1960 to 2004. We also find that temperature variability improves the fit and accuracy of standard longitudinal forecasting models. These results suggest that temperature variability is an underappreciated but useful tool for explaining the relationship between climate change and crop production.