The first essay investigates the impacts of Brazil's new biofuel policy, namely RenovaBio. RenovaBio aims to reduce greenhouse gas (GHG) emissions in the transportation sector through expansion of biofuels by requiring tradable decarbonization credits (CBIOs) associated with the GHG reductions achieved by biofuels. We model the CBIO and fuel markets using a partial equilibrium model to simulate the impacts of RenovaBio. As a result of the implementation of RenovaBio, biofuel consumption increases, and petroleum product use decreases to meet the CBIO compliance targets. Perhaps because of reduced targets during the pandemic, CBIO prices remain low, but our estimates highlight the potential for higher prices and large impacts. Results also show that RenovaBio can cause 7.2 percent of GHG abatement with an annual average compliance cost of R$ 36.3 billion. Total compliance costs are reduced if CBIO compliance targets are relaxed or there is good progress reducing biofuel carbon intensities. The second essay investigates the pass-through of the biodiesel-diesel price spread and BTC to the RIN price and estimate whether the market expectations about the reinstatement of the BTC have affected the RIN prices. The USA biodiesel tax credit (BTC) is a subsidy that could affect the cost of biofuel mandates in that country, but with uncertainty because the BTC is not continuously in place and is otherwise retroactively applicable. The biofuel mandate compliance certificate (RIN) price presumably depends on the price of fuels, such as biodiesel and diesel, as well as the BTC. Our estimated results shows that market participants' responses to the uncertainty of BTC might be different across time. These results could raise questions about whether an inconsistently applied BTC in the context of a blend mandate encourages biofuel expansion. The last essay develops aggregate supply response of Korean rice farms from the results of farm-level estimation. We first estimate supply components using detailed farm-level data. The farming area, land productivity, and farm exit are examined to see how those components are affected by the rice price and other determinants. We also estimate the debt-to-asset ratio that could influence farming area and farm exit. Then, we simulate the short- and long-run supply curves implied by aggregating farm-level responses to price changes. Given the price changes in our simulations, the price elasticities of Korean rice production are between 1.00-1.79 in the short run and 1.57-2.77 in the long run. The decomposed aggregate supply response into its three components shows that supply response might be under-estimated when we focus on only a component of total supply.