class: center, middle, inverse, title-slide # Econ 330: Urban Economics ## Lecture 16 ### John Morehouse ### May 26th, 2021 --- class: inverse, center, middle # Lecture 18: Urban Sorting & The Environment --- name: schedule # Schedule ## Today -- 1) .hi.purple[Household Sorting, Land-use Restrictions, and Carbon Emissions] - Walkthrough of [Colas & Morehouse (2021)](https://rawcdn.githack.com/johnmorehouse/Papers/6fe51365bf2cd0debb4b6c962c23683222349892/colas_morehouse_jan21.pdf) -- -- ## Upcoming - .hi.slate[Book Report Due May 30th] - .hi.slate[HW4 due June 6th] -- --- # Intro Households emit carbon and other pollutants for many reasons --- count: false # Intro Households emit carbon and other pollutants for many reasons. 1) .hi[Use of appliances]. Main Culprits: - Air conditioning (electricity) - Heating (natural gas) --- count: false # Intro Households emit carbon and other pollutants for many reasons. 1) .hi[Use of appliances]. Main Culprits: - Air conditioning (electricity) - Heating (natural gas) 2) .hi.purple[Driving] --- count: false # Intro Households emit carbon and other pollutants for many reasons. 1) .hi[Use of appliances]. Main Culprits: - Air conditioning (electricity) - Heating (natural gas) 2) .hi.purple[Driving] We will focus on 1). The amount of electricity and natural gas a household consumes varies considerably across cities --- # Energy Use and Climate <img src="images/cbsa_weather.png" width="90%" height="90%" style="display: block; margin: auto;" /> --- # Emissions The utilization of energy (natural gas, electricity) leads to carbon and other pollutant emissions --- count: false # Emissions The utilization of energy (natural gas, electricity) leads to carbon and other pollutant emissions - Natural gas directly emits carbon (so the emissions factor is independent of location) - Electricity needs to be produced at a power plant --- count: false # Emissions The utilization of energy (natural gas, electricity) leads to carbon and other pollutant emissions - Natural gas directly emits carbon (so the emissions factor is independent of location) - Electricity needs to be produced at a power plant. - Power plants use different fuels - Different fuels `\(\implies\)` different emissions factors for electricity use across locations --- count: false # Emissions The utilization of energy (natural gas, electricity) leads to carbon and other pollutant emissions - Natural gas directly emits carbon (so the emissions factor is independent of location) - Electricity needs to be produced at a power plant. - Power plants use different fuels - Different fuels `\(\implies\)` different emissions factors for electricity use across locations Example of clean energy? Dirty Energy? --- # Plant Types <img src="images/plant_types.png" width="90%" height="90%" style="display: block; margin: auto;" /> --- # NERC Regions <img src="images/nerc_map.png" width="90%" height="90%" style="display: block; margin: auto;" /> --- # So Far Household carbon emissions vary by location due to --- count: false # So Far Household carbon emissions vary by location due to 1. Temperature Differences --- count: false # So Far Household carbon emissions vary by location due to 1. Temperature Differences 2. Variation in the carbon intensity of local power plants --- count: false # So Far Household carbon emissions vary by location due to 1. Temperature Differences 2. Variation in the carbon intensity of local power plants .qa[Question]: A representative household in San Diego is responsible for 16.71 thousands pounds of carbon emissions per year from electricity and natural gas usage. How much is the same family in Memphis responsible for? --- # Data <img src="images/use_tab.png" width="80%" height="80%" style="display: block; margin: auto;" /> --- # Land-Use Restrictions Cities also vary in the amount of land use restrictions they have. In some cities, land use restrictions are tight, meaning lots of costs and restrictions from building more housing. - Measured by the [Wharton Land Use Regulation Index](http://realestate.wharton.upenn.edu/working-papers/a-new-measure-of-the-local-regulatory-environment-for-housing-markets-the-wharton-residential-land-use-regulatory-index/) - What we have talked about before! --- # Policy and Emissions <img src="images/wharton_ca.png" width="90%" height="90%" style="display: block; margin: auto;" /> --- # Recap __So Far__ - Cities vary drastically in terms of average household carbon emissions --- count: false # Recap __So Far__ - Cities vary drastically in terms of average household carbon emissions - This variation stems from: - Differing carbon intensities of local power plants - Spatial variation in the _marginal benefit_ of energy consumption --- count: false # Recap __So Far__ - Cities vary drastically in terms of average household carbon emissions - This variation stems from: - Differing carbon intensities of local power plants - Spatial variation in the _marginal benefit_ of energy consumption - Cities with lower per capita carbon emissions have stricter land-use regs (on average) --- count: false # Recap __So Far__ - Cities vary drastically in terms of average household carbon emissions - This variation stems from: - Differing carbon intensities of local power plants - Spatial variation in the _marginal benefit_ of energy consumption - Cities with lower per capita carbon emissions have stricter land-use regs (on average) __Up Next__ - Develop a quantitative model to estimate the impact of land-use restrictions on national carbon emissions. Specifically, relax land-use regulations in CA to a "reasonable" level --- # Model: Overview 1) __Locations__ vary by --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants - Marginal benefit of energy use for different energy types --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants - Marginal benefit of energy use for different energy types 2) __Households__ - Demand a composite consumption good, housing, and energy - Decide where to live (and thus work) --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants - Marginal benefit of energy use for different energy types 2) __Households__ - Demand a composite consumption good, housing, and energy - Decide where to live (and thus work) - Vary by demographic, birth locations, and an unobservable component (to the researchers) to location preferences --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants - Marginal benefit of energy use for different energy types 2) __Households__ - Demand a composite consumption good, housing, and energy - Decide where to live (and thus work) - Vary by demographic, birth locations, and an unobservable component (to the researchers) to location preferences 3) __Emissions__ - Use of energy leads to carbon emissions - Emissions factors for electricity use vary by location --- count: false # Model: Overview 1) __Locations__ vary by - Wages, rents, energy prices, and amenities - Carbon Intensity of regional power plants - Marginal benefit of energy use for different energy types 2) __Households__ - Demand a composite consumption good, housing, and energy - Decide where to live (and thus work) - Vary by demographic, birth locations, and an unobservable component (to the researchers) to location preferences 3) __Emissions__ - Use of energy leads to carbon emissions - Emissions factors for electricity use vary by location --- # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` --- count: false # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` - Rents in each city are given by `\(R_j = (1+k_j)*L_j\)` where: --- count: false # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` - Rents in each city are given by `\(R_j = (1+k_j)*L_j\)` where: - `\(k_j\)` is the level of land-use restrictions in location `\(j\)` - `\(L_j\)` is the population of location `\(j\)` --- count: false # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` - Rents in each city are given by `\(R_j = (1+k_j)*L_j\)` where: - `\(k_j\)` is the level of land-use restrictions in location `\(j\)` - `\(L_j\)` is the population of location `\(j\)` --- count: false # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` - Rents in each city are given by `\(R_j = (1+k_j)*L_j\)` where: - `\(k_j\)` is the level of land-use restrictions in location `\(j\)` - `\(L_j\)` is the population of location `\(j\)` --- count: false # Simplified Model Suppose we have two locations. Utility in location `\(j\)` is given by: `\begin{align*} U(w_j, R_j, P^{elec}_j) = w_j - \frac{1}{2}* R_j - \frac{1}{10}P^{elec}_j \end{align*}` - Rents in each city are given by `\(R_j = (1+k_j)*L_j\)` where: - `\(k_j\)` is the level of land-use restrictions in location `\(j\)` - `\(L_j\)` is the population of location `\(j\)` - Carbon emissions in location 1: `\(10\)` per 1 unit of electricity - Carbon emissions in location 2: `\(5\)` per 1 unit of electricity - Denote carbon emissions per unit in city j as `\(\delta_j\)` --- # Example Suppose `\(k_1 = 2\)`, `\(k_2 = 4\)`. Also suppose `\(P^{elec}_1 = 10\)` and `\(P^{elec}_2 = 10\)`. - Assume `\(w_1 = 200, w_2 = 205\)` and total population is fixed at 1000. .qa[Q1]: Compute equilibrium population levels in each city --- # Example --- # Example Furthermore, let `\(e^*_j\)` be the households optimal consumption of electicity in location `\(j\)`. This is given by: `\begin{align*} e^*_j = \frac{\frac{1}{10}\times w_j}{P^{elec}_j} \end{align*}` .qa[Q2]: Compute emissions in each city and total emissions. Hint: Emissions in each city is given by: `\begin{align*} \text{Emis}_j = \delta_j\times e^*_j \times L^*_j \end{align*}` --- # Example Recompute total emissions when `\(k_2 = 2\)`. --- # Example --- # Example --- # Local Pollutants In addition to carbon emissions, I have mentioned "other pollutants" from HHs --- count: false # Local Pollutants In addition to carbon emissions, I have mentioned "other pollutants" from HHs - We focus on emissions of `\(PM_{2.5}\)` - Widely studied and regulated measure of air quality - At the helm of US air quality regulation since late 90s - Associated with detrimental health and economic outcomes such as: higher infant mortality, increased cognitive decline in seniors, reduced property values --- # Local Pollutants - `\(PM_{2.5}\)` is considered a _local_ pollutant because it affects those directly exposed - Emissions of PM2.5 from a given location can impact air quality locally, regionally, and nationally - [Ed and I](https://rawcdn.githack.com/johnmorehouse/Papers/4bc0512d24117cc3421facaf5c0b52c5addf7578/morehouse_rubin_feb24.pdf) estimate that 90% of particulate matter emissions from coal-power plants leave their state of origin within 48 hours --- count: false # Local Pollutants - `\(PM_{2.5}\)` is considered a _local_ pollutant because it affects those directly exposed - Emissions of PM2.5 from a given location can impact air quality locally, regionally, and nationally - [Ed and I](https://rawcdn.githack.com/johnmorehouse/Papers/4bc0512d24117cc3421facaf5c0b52c5addf7578/morehouse_rubin_feb24.pdf) estimate that 90% of particulate matter emissions from coal-power plants leave their state of origin within 48 hours - Thus: Electricity demand in CA leads to an increase in PM from power plants in CA - The additional emissions also travel to other states --- # So What? So: we want to model how HH energy consumptions impacts `\(PM_{2.5}\)` - But we are economists and do not know how to model chemical transport! - Luckily, we don't have to. We employ the [Intervention Model for Air Pollution](https://www.inmap.run/) (InMAP) --- count: false # So What? So: we want to model how HH energy consumptions impacts `\(PM_{2.5}\)` - But we are economists and do not know how to model chemical transport! - Luckily, we don't have to. We employ the [Intervention Model for Air Pollution](https://www.inmap.run/) (InMAP) __InMAP__ is an "off-the-shelf" chemical transport model that maps emissions to concentration across the US - Very smart -- takes local atmospheric conditions, stack height of emissions, and chemical decay into account - Very easy to use, as well. Python code already written for you to use. Just feed it your data --- # Model Fit - Mark and I estimate the model on publicly available data - This part is the hard part, so I won't go into detail --- count: false # Model Fit - Mark and I estimate the model on publicly available data - This part is the hard part, so I won't go into detail - Can check how well the model fits the data by solving the equilibrium in the model and compare the model's outputs to the data. --- # Model Fit <img src="images/mfit2.png" width="80%" height="80%" style="display: block; margin: auto;" /> --- # Counterfactual - We estimate the model on publicly available data. Then: __Main Counterfactual__: Relax land-use regulation in CA to match the national median --- count: false # Counterfactual - We estimate the model on publicly available data. Then: __Main Counterfactual__: Relax land-use regulation in CA to match the national median __Mechanism__: --- count: false # Counterfactual - We estimate the model on publicly available data. Then: __Main Counterfactual__: Relax land-use regulation in CA to match the national median __Mechanism__: - Mechanically, relaxing land-use regs decrease rents in CA - Households move to CA due to cheaper rents --- count: false # Counterfactual - We estimate the model on publicly available data. Then: __Main Counterfactual__: Relax land-use regulation in CA to match the national median __Mechanism__: - Mechanically, relaxing land-use regs decrease rents in CA - Households move to CA due to cheaper rents - Economy reaches a new equilibrium in rents, wages, and location choices --- count: false # Counterfactual - We estimate the model on publicly available data. Then: __Main Counterfactual__: Relax land-use regulation in CA to match the national median __Mechanism__: - Mechanically, relaxing land-use regs decrease rents in CA - Households move to CA due to cheaper rents - Economy reaches a new equilibrium in rents, wages, and location choices __Carbon Emissions__: - .hi[CA]: low marginal benefit of energy use (temperate climate) - Carbon efficient power plants - Thus, should expect equilibrium emissions to fall as a result --- # Graph .pull-left[ <img src="images/housing_market_pre.png" width="90%" height="90%" style="display: block; margin: auto;" /> ] --- count: false # Graph .pull-left[ <img src="images/housing_market_pre.png" width="90%" height="90%" style="display: block; margin: auto;" /> ] .pull-right[ <img src="images/housing_market_post.png" width="90%" height="90%" style="display: block; margin: auto;" /> ] --- # SB501 .pull-left[ <img src="images/SB501.png" width="90%" height="90%" style="display: block; margin: auto;" /> ] .pull-right[ <img src="images/SB502.png" width="90%" height="90%" style="display: block; margin: auto;" /> ] --- # Results <img src="images/counter_new.png" width="75%" height="75%" style="display: block; margin: auto;" /> --- # Results <img src="images/pm_changes.png" width="90%" height="90%" style="display: block; margin: auto;" /> --- # Conclusion ## Takeaways - Land-use policies that change where workers live have important implications for national carbon emissions --- count: false # Conclusion ## Takeaways - Land-use policies that change where workers live have important implications for national carbon emissions - Estimated the impact of relaxed land-use regulations in CA - Predicted national household carbon emissions fall by .6% - Doesn't seem like much but this is equivalent to a savings of $310 million dollars annually (using Social Cost of Carbon) What do we not do? Do you have any thoughts/comments/questions? --- exclude: true