Abstract

This paper investigates the optimal path of subsidies when a new business, such as electric vehicles (EV), with low externality and market competitiveness, competes against conventional businesses, such as internal combustion engines (ICE), with high externality and market competitiveness. I build a dynamic model that addresses market competition, the channel of sales leading to technology improvements, and the externality of carbon emissions. This paper compares dynamic purchase and investment subsidies with currently fixed subsidies in the US and no policy. To run the model, I utilize collocation methods and use coefficients from the estimation of aggregated real market data: 531 vehicle models sold in the US from 2008 to 2019, as well as SEC filing reports. I find that the optimal path should begin at high values and then approach zero as the capital of EV firms increases. Both purchase and investment subsidies increase the total social net benefit compared to the constant subsidy, and also more, compared to no policy. It suggests that constant subsidies based on the Pigouvian concept will not fully correct the externalities from carbon emissions in the automobile market, which has a strong potential for improvement via investments.

JEL codes: Q58, Q55, H23, R48, L98 

Abstract

Economic activity consumes large amounts of energy, making modern economies especially vulnerable to energy price shocks. Following such a shock, firms adjust by reducing output or energy used per unit of output. Here we use annual panel data for the 50 US States to examine how local industries have adapted to energy price changes since 2000. We estimate a recursive dynamic system that models how energy prices affect output both directly and indirectly through induced capital adjustment and energy intensity. We find that energy price increases reduce local GDP in the short run but also lead to long-term increases in energy efficiency. Output contracts in the short run, but dynamic efficiency gains mitigate these losses: on average, improvements in energy intensity offset between 2% to 5% of the short run GDP decline. We interpret these dynamics as evidence of induced energy-saving technological adjustment, whereby endogenous efficiency adaptation leads to a decoupling of energy consumption from economic growth as energy price rise at the State level.

JEL codes: Q43, R11, Q40, O44 

Abstract

State governments use various policies to encourage energy production from non-traditional sources to promote clean energy transitions, including renewable portfolio standards (RPS). Despite their widespread adoption, RPS impacts on farmland remain understudied. Adoption by farmers is especially important as they often have the land needed to accommodate new forms of electricity production. This paper examines how state RPS targets influence energy adoption on farms and economic outcomes, using panel data from 50 US states for 2012, 2017, and 2022, and a system of equations with inverse probability weighting to address endogeneity and selection biases. The results suggest that on-farm renewable energy (RE) adoption is significantly higher, and net cash farm income is also higher, in states with higher RPS. This additional income gain is mostly driven by gross income growth rather than cost reduction. In addition, state RPS are expanding, stimulated by higher electricity prices, regardless of political orientation. These findings contribute to the understanding of how state RPS affects on-farm RE adoption, farm survival, and regional economic resilience. Beyond understanding the determinants and effects of RPS in general, our paper contributes to the literature on farm adoption of different types of RE, effects of RE on farm incomes, and the role of farm size.

JEL codes: Q15, Q42, Q48, R11 

Abstract

Despite federal mitigation efforts since the 1950s, rural poverty rates have remained 3–5 percentage points higher than those in metropolitan areas in the United States. This study uses panel data to examine whether USDA Rural Development (RD) programs reduced poverty in nonmetro counties from 2004 to 2022. We apply a Bartik instrument based on national industry shocks and historical RD investment composition to identify program effects. Greater per capita RD investments cause reductions in county poverty rates over time. This effect spans RD investment types, with housing programs showing the strongest impact. Results underscore the importance of targeted rural development for effective federal strategies. 

JEL codes: I30, R11, R58

Abstract

This paper quantifies peer effects of electric vehicles (EV) by area and subgroups in Michigan, USA. The peer effect has been widely used to explain consumer behavior, and it plays an important role in product adoption under imperfect information. Consumers may initially hesitate to purchase an EV; however, if they observe more EVs in their neighborhood, on the road, or in parking lots, their reluctance to buy one may be alleviated. I build a system of equations to identify peer effects from lagged cumulative EV adoption by area and subgroups of type and brand values, with other effects from charging stations and socio-economic controls. This paper provides evidence that peer effects from the cumulative adoption of a specific model play a substantial role in the new adoption of the same model. While peer effects also increase overall adoption within the same ZCTA, competing types or brand values negatively affect the adoption of new EVs. I discuss the implications of these findings and how they affect the real impact of subsidization with peer effects. This paper also finds that the impact of subsidization would be significantly larger when peer effects are considered.

JEL codes:  D12, Q40, Q50, R11 

Abstract

How do firms adjust investment in response to energy price shocks, and what are the implications for energy and carbon intensity? This paper studies energy prices as a dynamic cost shock that reshapes firms’ capital allocation between production and low-carbon technologies. We develop a framework in which forward-looking firms choose investment in energy-efficient capital. This paper studies U.S. firm-level investment facing energy price shocks and provides new evidence on the investment channel linking external shocks to firm performance. While prior work emphasizes the role of energy price on firm investment at the macro level (national or sector-level), we develop and estimate a structural framework from energy price to energy and carbon intensity using firm- and establishment-level micro data. First, the article explores the theoretical background of firms’ investment decisions under external energy price shocks using recursive dynamic models. Second, the empirical analysis combines establishment-level data on manufacturing plants from the 2014, 2018, and 2022 Manufacturing Energy Consumption Survey with firm-level information from the 2023 Annual Business Survey, which uniquely measures energy use, emissions, and relevant investment. Identification exploits plausibly exogenous state-level variation in energy price shocks across time. We complement the reduced-form analysis with a recursive dynamic model of firm investment. Results show that higher energy prices induce an increase in investment in energy-efficient and low-carbon technologies, leading to significant reductions in firm-level energy and carbon intensity. The effects are stronger for energy-intensive firms, and persistent high energy prices, manifest in greater impacts on carbon intensity. Counterfactual simulations indicate that investment subsidies tied to energy price shocks amplify capital reallocation toward cleaner technologies and improve efficiency relative to untargeted policies. These findings highlight the importance of proper policy implementation in firm-level investment responses and provide micro-founded evidence on the role of energy price on firm investment and state-level economic growth.

JEL codes:  D22, L60, Q41, Q52 

Abstract

How do subnational emissions trading systems (ETS) affect firms’ energy and carbon intensity, and through which mechanisms? In the U.S., the only activated emissions trading policy is the state-level emission trading scheme (ETS) administered mainly in coastal Western and Northeastern states. This paper studies U.S. state-level ETS programs and provides new evidence on the investment channel linking carbon pricing to firm performance. While prior work emphasizes national/international level analysis or reduced-form models, we develop and estimate a structural framework in which firms choose between abatement investment and permit purchases under state-level carbon pricing. First, the article explores the theoretical background of firms’ investment decisions under state-level emission trading schemes using game theory and dynamic optimization. Second, the empirical analysis combines establishment-level data of manufacturing plants from the 2014, 2018, and 2022 Manufacturing Energy Consumption Survey with firm-level information from the 2023 Annual Business Survey, which uniquely measures energy use, emissions, and relevant investment. We exploit adoption of state ETS policies using a staggered difference-in-differences design and complement it with a structural dynamic model of firm investment under carbon pricing. Results show that ETS exposure significantly reduces firm-level energy and carbon intensity. The primary mechanism operates through increased investment in energy-efficient and low-carbon technologies rather than through permit purchases alone. Higher permit prices amplify this effect by raising the marginal return to abatement capital. Counterfactual simulations indicate that broader adoption of ETS would generate substantial reductions in energy and carbon intensity at relatively low cost, consistent with second-best efficiency gains under incomplete policy coverage. These findings highlight the importance of carbon pricing policies in firm-level investment responses and provide micro-founded evidence on the role of subnational ETS in driving decarbonization.

JEL codes:  L51, Q54, Q55, Q58

Abstract

Transnational Municipal Networks (TMNs) on climate and energy issues have grown significantly in the 21st century. There is evidence that local governments joining these TMNs achieve targeted outcomes such as a reduction in energy consumption or carbon emissions. Nevertheless, the actual mechanisms by which TMNs affect energy consumption or carbon emissions have not been studied well. This paper studies how local governments’ participation in global climate networks shapes firm-level energy and carbon intensity by altering investment incentives. While past literature utilized reduced form models linking TMN membership to energy consumption of aggregate local energy consumption, we identify how TMN membership affects firm-level investment and energy consumption. Our research question is: How do TMNs on climate and energy issues affect firms’ energy and carbon intensity, and through which mechanisms? To address this question, we develop a system of equations and connect it to micro-level data for empirical validation. The empirical analysis links establishment-level data on manufacturing plants from the 2014, 2018, and 2022 Manufacturing Energy Consumption Survey with firm-level information from the 2023 Annual Business Survey, which uniquely measures energy use, emissions, and relevant investment. We also use a city networks membership dataset provided by Drs. Michele Acuto and Benjamin Leffel. Identification exploits staggered adoption of TMN membership across cities and counties in a difference-in-differences framework. To isolate mechanisms, we also estimate a system of equations that jointly model investment and energy and carbon intensity outcomes. Results show that TMN membership significantly reduces establishment-level energy and carbon intensity. The effects operate primarily through increased investment in energy-efficient and low-carbon technologies. The impact is amplified when local and state governments participate simultaneously, consistent with complementarities in policy signals and expectations. These findings suggest that TMNs influence firm behavior by coordinating policy environments and reducing uncertainty about the direction of climate policy. This paper provides micro-founded evidence on how non-binding, network-based governance affects firm decision-making. More broadly, it highlights the role of policy networks in driving the energy transition, offering a new perspective on the effectiveness of decentralized climate governance. This presents that local governments voluntarily achieve policy outcome via their network even without the support from federal governments. Federal governments may increase the policy effectiveness to support local governments to join the TMNs or stimulate the network participation.

JEL codes:  Q54, Q48, R58, D22, H73

Abstract

The environmental justice literature has examined the association between economic disadvantage and industrial emissions, but rarely examined is the association between rurality, open space, and emissions. In a fossil fuel economy, open space may be valued to disperse industrial disamenities and reduce public opposition to emission-intensive production. However, open space may also be valued economically to produce less energy dense renewables that are space intensive. This research uses the 2014 and 2018 versions of the Manufacturing Energy Consumption Survey to investigate local characteristics that are strongly associated with poor carbon performance. Quantile regression is used to identify whether industrial activities that produce the most carbon pollution may be more amenable to transitioning to renewable energy.

Abstract

The Office of Science and Technology Policy requested questions on both R&D and use of renewable energy and battery storage technologies in the 2022 Annual Business Survey. The definition of R&D for process innovation focuses on creating, improving, or adapting existing production or delivery methods, including significant changes in techniques or equipment that would include evaluating or implementing onsite renewable energy. Use of onsite renewable energy, with or without battery storage, may provide evidence of an energy transition from fossil to renewable sources if adopted at significant scale. The critical regional science question investigated in this paper is where these emerging phenomena are located. Space-intensive renewable energy production is likely to be cost prohibitive in more densely populated areas where the opportunity cost of open space is higher. The one limitation of the current analysis is the inability to connect local characteristics to renewable energy initiatives of multi-plant firms. However, the phenomenon in large single-unit firms may be informative while we wait for data on establishments in multi-unit firms available in the 2022 Manufacturing Energy Consumption Survey. 

JEL codes: Q42, R14, C11 

Abstract

The longer society waits to reduce carbon emissions on an eventual path to net zero, the more critical environmental innovation becomes to averting a catastrophic rise in global temperatures above 2°C. Innovation surveys such as the EU Community Innovation Survey (CIS) and the US Annual Business Survey (ABS) now include questions on environmental innovation related to reducing a firm’s carbon footprint. However, as with all self-reported innovation questions that comport with the Oslo Manual (OECD/Eurostat 2018), the requirements for reporting are only that the innovation is available to users and that it comprises a substantive change from what was done before. Neither the success nor the type of environmental innovations are investigated in the innovation surveys making the findings unpersuasive due to potential social desirability bias (Tourangeau, et al. 2000) and of limited usefulness for informing environmental innovation policy related to the energy transition.

Abstract

In recent decades, investments in controlled environment agriculture (CEA) have increased markedly, resulting in the number of CEA operations doubling, even as the share of CEA in total agriculture remains low. The productivity of CEA is exceptionally high compared to conventional outdoor agriculture, and it provides stable production conditions regardless of weather conditions. Increased adoption of the technology is expected over time as more farmers move up the learning curve because CEA is still at the beginning of the technology adoption phase in the US. In this paper, we investigate how CEA affects farm survival and preservation of farmland using panel data for 50 US states from 1998 to 2019. Data comes from the USDA Census of Agriculture, Censuses of Horticultural Specialties, Horticultural Specialties Reports, Vegetable Annual Summary Reports, and the US EIA State Energy Data System. We use a system of equations to investigate (1) how climate change has affected farmers’ decisions on CEA adoption, and (2) how CEA adoption contributes to farmers’ income and energy consumption. We find that more severe climate conditions lead farmers to install CEA, and CEA adoption has increased farmers’ income substantially more than associated cost increases. CEA adoption also decreases energy consumption. These results demonstrate that CEA simultaneously contributes to farmers’ income and energy security, suggesting important implications for agricultural sustainability and climate adaptation strategies.

JEL codes: Q15, Q16, Q19, Q49 

Abstract

Trade shocks at the national level are external to and largely unavoidable for local economies, and in the past shocks associated with Chinese imports have profoundly affected local businesses. These impacts are observable in local economies in poverty and unemployment rates, economic growth, and population (out-)migration. A county having more industries influenced by Chinese imports has higher poverty. Intense competition with Chinese imports may force firms to exit the market, or reduce employment, and consequently, workers may move to other regions. Here we analyze the short-term and long-term impacts of trade-induced employment and income shocks on about 3,100 US counties between 2002 and 2022. Bartik-type instruments are constructed to indicate trade shock on local economies by applying a Chinese import shock at the three-digit NAICS code to the share of employment and payrolls at the county-level three-digit NAICS code. We find that a trade shock significantly increases the poverty rate, and the effect is worse in denser areas. More people aged over 65 absorb the impact of trade shock on the poverty rate, while more people under 5 increase the poverty rate with the shock. Wage employment-dependent local economies are vulnerable to the trade shock, while more self-employment can alleviate the negative from the trade shock. Overall, import shocks worsen local poverty, and local areas with more children and wage employees are also more vulnerable to such external forces.

JEL codes: F14, R11 

Abstract

Microdata from surveys that have been used successfully to study local characteristics associated with innovative behavior are largely silent on local innovation rates for all but the largest substate areas. This is because estimating a proportion from a binomial distribution requires a sampling fraction that is cost prohibitive outside of large conurbations. However, there is growing policy interest in the causes and consequences of innovation at the local level across OECD member countries as new initiatives are introduced to ensure that all regions in a country fully participate in the innovation economy. This study summarizes development of small area innovation rate estimation in which detailed innovation data in the Annual Business Survey (ABS) “borrow strength” from the much larger Economic Census (EC) to more accurately estimate innovation incidence at the local level. It does this by combining observed innovation in ABS with modeled innovation in EC using auxiliary variables that are common to both datasets. Bayesian spatial dependence in concert with random effects that allows partial pooling greatly increases the efficiency of small area innovation rate estimates even for areas with a relatively small number of direct observations. Possible use cases of the small area innovation rate estimation motivated by the Regional Innovation Engines program at the U.S. National Science Foundation that funded this research are demonstrated.