Tuesday, February 4, 2020

The $64 Billion Massachusetts Vehicle Economy

Policymakers and budgetary analysts have long argued that roads are heavily subsidized. The diffusion of spending among federal, state, and local government entities, along with the complexity of indirect costs, make it difficult to understand the fully loaded cost of the vehicle economy. Individual families may track the personal costs of car ownership to their budgets, but they rarely consider the total cost of operating and maintaining the vehicle economy because the vast majority of roads and parking areas are provided free at the point of use. This study is intended to increase transparency regarding road-related spending and to provide a comprehensive estimate of the economic cost of Massachusetts’ vehicle economy.
By B137 - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=48998674
The Massachusetts’ vehicle economy is based on nearly 37,000 miles of public roads, adjacent parking areas, and 4.5 million private passenger cars and light trucks. This system provides conduits for exchange and movement for millions of families, businesses, and services in the State.

[The authors] conclude that the total annual cost of the vehicle economy in Massachusetts is approximately $64.1 billion. Over half this amount, some $35.7 billion, is borne by the public in the form of state budgetary costs, social and economic costs (road injuries and deaths, congestion, and pollution), and the value of land set aside for roads and parking. The remaining $28.4 billion falls on private consumers in the form of financing and operating their vehicles.

Several important conclusions follow from [their] analysis. First, the public costs of the vehicle economy are substantial – amounting to $14,000 per family in Massachusetts, regardless of whether they own a vehicle. Second, this cost is highly subsidized by non-road users – just 1/3 of state budgetary costs (amounting to $5.7 billion) are covered by user fees and gas taxes. Third, the economic impact is regressive: non-car owners typically are lower income and disproportionately from minority communities.

The full cost of the vehicle economy should also throw into perspective the cost of investments in public transportation projects, because the counterfactual of simply relying on roads and vehicles is not free. For example, previous studies have found that that the North-South Rail Link would incur a one-time cost of $3.8 to $12.3 billion.

Publicly borne costs are resources, which include money, time, and land. Massachusetts residents incur these costs regardless of whether they own or operate a vehicle. We categorize this $35.7 billion in annual costs into three categories: public budgetary costs, indirect and direct economic costs, and land value commitments.

[Note that GHG emission costs are estimated at $1.28 billion and pollution costs at $1.2 billion.]
The cost of pollution relied on Parry and Small’s research that established a baseline of $0.019 in pollution damages per mile of vehicle traveled (Parry and Small 2005).15 The number of vehicle miles traveled (VMT) in Massachusetts was identified from Federal Highway Administration data (Office of Highway Policy Information 2018). The baseline cost was multiplied by the annual VMT in Massachusetts to arrive at a yearly cost of vehicle-related pollution.

Greenhouse Gas Emissions
The cost of greenhouse gas emission relied on the widely-cited cost of carbon, $40 per metric ton (Parry and Small 2009; “Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis - Under Executive Order 12866” 2010; “The Social Cost of Carbon” 2017).16 The annual amount of metric tons of carbon dioxide emissions from vehicles in Massachusetts was determined using data from the Federal Highway Administration the Energy Information Administration (“Table MF-21 Highway Statistics 2017 - Policy and Governmental Affairs: Highway Policy Information” 2019; “Greenhouse Gases Equivalencies Calculator - Calculations and References” 2015). The cost of carbon was multiplied by the number of metric tons emitted to determine the cost of greenhouse gas emissions.
Implications for Income Distribution
Policymakers may consider the large subsidy dedicated to the vehicle economy in several dimensions. The public costs of the vehicle economy do not result in equitable access across different social and economic classes in the state. According to the National Equity Atlas, over 90 percent of White households in Massachusetts have access to a vehicle. This compares to 74.3 percent for Black households and 73.7 percent for Latino households (“Car Access - Massachusetts” 2018).

Paris Climate Agreement passes the cost-benefit test

The Paris Climate Agreement aims to keep temperature rise well below 2 °C. This implies mitigation costs as well as avoided climate damages. Here we show that independent of the normative assumptions of inequality aversion and time preferences, the agreement constitutes the economically optimal policy pathway for the century. To this end we consistently incorporate a damage-cost curve reproducing the observed relation between temperature and economic growth into the integrated assessment model DICE [ Dynamic Integrated Climate-Economy model]. We thus provide an inter-temporally optimizing cost-benefit analysis of this century’s climate problem. We account for uncertainties regarding the damage curve, climate sensitivity, socioeconomic future, and mitigation costs. The resulting optimal temperature is robust as can be understood from the generic temperature-dependence of the mitigation costs and the level of damages inferred from the observed temperature-growth relationship. Our results show that the politically motivated Paris Climate Agreement also represents the economically favourable pathway, if carried out properly.
[The authors] find that the 2 °C target represents the cost-benefit optimal temperature for the base calibration (Fig. 2a). This calibration involves the best estimate8 of the temperature–economic growth relation in the past and the original ECS [equilibrium climate sensitivity] value in DICE-2013 of 2.9 °C, which is at the centre of estimates for several decades. Higher ECS values shift the level of target warming for which the mitigation-cost curve diverges to infinity to higher values (Fig. 1), i.e. they incur substantially higher mitigation costs. For ECS of 4 °C, for instance, the 2 °C target becomes too costly. Yet, with an optimal target warming of 2.4 °C the deviation from this target is not large. For smaller ECS values, e.g. of 2 °C, limiting warming further to well below 2 °C is economically optimal. Regardless of the exact ECS, the optimal mitigation efforts promise a significant damage reduction compared to the BAU [business-as-usual] scenario (~14% for ECS of 4 °C, ~10% for ECS of 2.9 °C, and ~8% for ECS of 2 °C). These efforts are, as also claimed by the Paris Agreement, ambitious (Article 3)1 and involve very stringent measures from the outset (Fig. 2c).
Fig. 1: Illustration of universality of the cost-benefit climate analysis.

Cumulative mitigation costs (green curve) and climate damages (black curve) as a function of Earthʼs warming level give the total climate costs (red curve). Mitigation costs diverge for present-day warming and converge to zero for unmitigated warming. The damages are zero for zero warming and increase with temperature. The characteristic steepness of the mitigation curve implies that beyond a certain damage level the economically optimal temperature (which minimizes the total costs) becomes insensitive to a further increase in damages. For example, increasing (black dashed) or decreasing (black dotted) the damage level by half of the initial damage level does not change the economically optimal warming level significantly (grey area).
Fig. 2: Temperature increase, damage costs, and carbon emissions under cost-benefit optimal policy for three different climate sensitivities.
The black curves are associated with the original calibration of the climate sensitivity of 2.9°C; the blue curves with a 2°C climate sensitivity and the red curve with a 4°C climate sensitivity. The inset figures allow comparing the economically optimal temperature development and damage costs with their corresponding values in the BAU scenario.
by Nicole Glanemann, Sven N. Willner & Anders Levermann 
Nature Communications https://www.nature.com/ncomms/
Volume 11, Article Number: 110; (2020); Open Access; Published: 27 January 2020

Monday, January 13, 2020

The Impacts of Harmful Algal Blooms and E. coli on Recreational Behavior in Lake Erie

This paper examines simultaneously the effect of E. coli and harmful algal blooms on recreational behavior using survey data collected from Ohio recreators who visited Lake Erie during the summer of 2016. Using simulation based on latent class models of recreation choice, we find beachgoers and recreational anglers would lose in aggregate $7.7 million and $69.1 million, respectively, each year if water quality conditions were to become so poor that Lake Erie’s western basin were closed. Finally, we recover heterogeneity in recreators’ aversion toward algae and Escherichia coli, with beachgoers more averse to E. coli and anglers more averse to algae.
Eerie Blooms in Lake Erie
by David Wolf 1, Wei Chen 2, Sathya Gopalakrishnan 3, Timothy Haab 4 and H. Allen Klaiber 5
1. Assistant professor, Department of Economics, University of Wisconsin–Eau Claire; wolfdm@uwec.edu
2. Assistant professor, School of Agricultural Economics and Rural Development, Renmin University of China, Beijing, PR China; chen.4863@osu.edu
3. Associate professor, Department of Agricultural, Environmental, and Development Economics, The Ohio State University, Columbus; gopalakrishnan.27@osu.edu
Professor, Department of Agricultural, Environmental, and Development Economics, The Ohio State University, Columbus; haab.1@osu.edu
5, Professor, Department of Agricultural, Environmental, and Development Economics, The Ohio State University, Columbus; klaiber.16@osu.edu
Land Economics http://le.uwpress.org/ via University of Wisconsin Press http://www.uwpress.org/
Volume 95, Number 4; November 1, 2019; pages 455-472

Do Public Benefits of Voluntary Cleanup Programs Justify Their Public Costs? Evidence from New York

This paper contributes to the debate over public benefits and costs of state-funded voluntary cleanup programs, using evidence from property values in New York City. We value site redevelopment separately from cleanup and examine time to capitalization. Using property fixed effects and controlling for time-varying shocks, New York’s Brownfield Cleanup Program added 4% to property values. Off-site gains averaged 5.6% for properties with three units or less and 1.2% for multifamily residences, producing a $579.3 million tax gain that does not exceed the $667.9 million in program spending. Benefits stem from program participation and cleanup, but not from site redevelopment. 
Clean soil stockpile
by Olesya M. Savchenko 1 and John B. Braden 2
1. Assistant professor, Food and Resource Economics Department, University of Florida, Gainesville; savchenko.olesya@gmail.com
2. Professor Emeritus; Department of Agricultural and Consumer Economics, University of Illinois at Urbana-Champaign; jbb@illinois.edu
Land Economics http://le.uwpress.org/ via University of Wisconsin Press http://www.uwpress.org/
Volume 95, Number 3, August 1, 2019; pages 369-390

Investors Are Learning That Clean Tech Pays: The stock performance of green companies shows that climate change is also a business opportunity.


The American companies most reliant on embracing green technology are outperforming every broad measure of the stock market, delivering a greater return last year than all but two (Russia and Greece) of the world’s 94 leading equity indexes.

These are the 92 publicly traded firms with at least 10% of their revenues derived from clean energy, energy efficiency or clean technology, according to data compiled by Bloomberg New Energy Finance. ...

The S&P 500 Index and Russell 3000 gained 31%. ... This exceptional result didn’t come close to the performance of the clean companies, with their combined total return (income plus appreciation) of 40%. Together they were worth $946 billion last year, more than triple their market capitalization at the end of 2010. Whether the investment period is 2, 5 or 10 years, the return is superior by margins of 12%, 37% and 112% for clean companies.

Total Return (Income plus appreciation)
  • ...
    The three biggest companies in this group by market capitalization all derived more than half their revenue from the clean-energy business. They are Nextera Energy Inc., ... operator of commercial nuclear power units and provider of electricity through wind, solar and natural gas; Tesla Inc. ..., manufacturer of battery-powered, electric vehicles; and Universal Display Corp., ... maker of organic light-emitting diode technology, according to data compiled by Bloomberg.

    ... The 19 energy firms in the group produced a 106% total return, 15 times the 7% gain by the overall energy-stock benchmark, the Russell 3000 energy sector. The nine technology companies among the 92 returned 70% when the Russell 3000 technology sector appreciated 46%, and the 16 BNEF-designated utilities earned 34% when the comparable Russell group advanced 26%, according to data compiled by Bloomberg.

    Even U.S.-imposed tariffs on products from China, which manufactures a ... [great deal] of clean-energy equipment, haven’t ... [hurt] the performance of the U.S. companies. Enphase Energy Inc., for example, moved its production to Mexico from China and the Petaluma, California-based maker of renewable energy equipment rallied 452% last year on revenue growth of 96%. Analyst estimates compiled by Bloomberg predict that Enphase sales will increase 26% and 27% annually through 2021.

    By contrast, revenues for the 28 companies in the S&P Energy Index, a mostly fossil-fuel crowd, declined 5% in 2019 and are forecast to grow 4% in each of the next two years.

    One investor who is bullish on clean companies ... manage[s] the most successful mutual fund in the U.S. last year, the Columbia Seligman Communication and Information Fund. The manager, Paul H. Wick, counted three of the cleaner companies among his winners: Advanced Energy Industries, Bloom Energy Corp. and Rambus Inc.

    Wick said that Bloom Energy, for example, reduced the cost of its products “on a continuous basis over the last four or five years” and “as a result will be cheaper than grid power in quite a few jurisdictions, states in the U.S. and overseas.” His fund returned 54% (income plus appreciation) in 2019 and Bloom Energy rallied 144% in the final two months of the year after Wick acquired its shares.

    Among the BNEF technology firms, Universal Display Corp., the ... provider of power-saving lighting products, appreciated 121% as sales climbed 64% last year amid forecasts for 23% and 25% growth in 2020 and 2021. That's a superior outlook compared to the 70 tech companies in the S&P 500 Information Technology Index, which saw sales increase 10% in 2019 with 10% projected for 2020 and 8% for 2021, according to analysts’ estimates.

    Renewable-energy utilities are similarly positioned for growth. Pattern Energy Group rallied 54% last year after the Canada Pension Plan Board agreed to acquire the San Francisco-based renewable power generation firm for $2.6 billion. Revenues increased 10% last year and are expected to rise 11% in 2020 and 3% in 2021, according to analyst estimates compiled by Bloomberg. The 28 companies in the S&P 500 Utility Index reported inferior sales growth of 6% last year and are likely to see only 3% in 2020 and 2% in 2021, according to the estimates.

    City to acquire streetlights, replace them with LEDs ... New lights projected to reduce energy consumption by 70% [in Warwick Rhode Island]

    "If all goes as planned, residents [of Warwick Rhode Island] won’t be waiting for weeks to have a burned-out streetlight replaced; they’ll have better roads, and best of all, the cost of borrowing all the money to get this done will be covered by all the savings."

    Key elements to the plan are city-owned LED streetlights that will save an estimated $750,000 now being paid to National Grid for the maintenance of an estimated 9,000 Warwick streetlights, plus additional energy savings, and a low-cost 10-year loan estimated at 1.3 percent for $10.2 million from the Rhode Island Infrastructure Bank. The loan would fund a ramped up three-year road replacement and paving program, while the savings from streetlights would cover loan debt costs.

    On Monday, the City Council gave second passages to ordinances authorizing the purchase of streetlights, conversion of fixtures to light-emitting diode (LED) technology and the financing of the project through an “appropriation obligation” bond not to exceed $3.2 million and the borrowing of the $10.2 million.....

    In preparation for the streetlight conversion that it is hoped can be completed by this time next year, the city has solicited bids for the acquisition and installation of a system as well as its maintenance. Eight bids have been received for maintenance of the system.... For the acquisition and installation of the system, the city received two bids... Among the issues to be considered in addition to price, equipment and timing are the capabilities of the system and whether they could generate additional savings and be of use to other departments.
    With a basic system, photoelectric cells would control when the lights go on and off. However, there is ... technology to link the lights into a “mesh network” that would give the city the capability to turn on or off lights as well as control their output by neighborhoods. The cloud-based system would come at a higher cost and would require the installation of repeaters and gateways as well as a monthly fee. An alternative is a cell phone-based system where each light is independent of the other and doesn’t require repeaters or gateways. It would have the capability of incorporating cameras. Both systems would provide real-time data on the system identifying outages and allowing ... [regulation of] the output of light.

    The city is looking [at whether] ... the capability of reducing power and perhaps evening blacking some areas of the city during early morning hours could generate sufficient savings to more than offset added upfront costs and operational fees.

    According to Michael D’Amico, financial advisor for the city, the light management option would cost an additional $1 million and the network mesh an added $1.3 million from the basic dusk to dawn photocell system at $3.2 million. At this point he doesn’t see a cost benefit to going with either of the more interactive systems.

    With LEDs it is also easier to regulate the light shed so, for instance, sidewalks and streets can be lit without light spilling into abutting properties.

    As streetlights don’t have individual meters, there’s not a precise knowledge of the energy being consumed. The city now pays National Grid a “tariff” based on light wattage and estimated power usage over the year. In addition it pays a maintenance fee.

    It is estimated once the city replaces all the National Grid lights (it will buy those fixtures for a total cost of about $50,000) with LEDs energy consumption will be reduced by 70 percent.
    Since the city will own the lights, it will assume their maintenance. The [warranty] on the lights that have a projected life expectancy of 10 years would mean the city would have little or no maintenance issues in the first year.

    The cost of maintenance is being viewed as the major cost saving of the program with the potential of offsetting borrowing costs for the mayor’s road program.

    The projected savings of $775,000 would offset about half the debt service costs of both the $3.2 million and $10.2 million bonds, D’Amico said.... 

    When he announced the program, the mayor noted the historically low interest rates available as a common-sense and cost-effective way to address the city’s myriad infrastructure needs. The weighted interest rate for the bonds is expected to be 1.3 percent. Savings from the LED technology will significantly offset debt service costs. 
    The city would spread the repayment of the bonds over ten years.

    Furthermore, by contracting for the maintenance, the burden is not placed on the Department of Public Works and the city can dictate acceptable periods for the work to be completed. A maximum of five days is being considered for the replacement of a nonfunctioning light. Currently it can take several months to replace a streetlight.
    by John Howell
    "City to acquire streetlights, replace them with LEDs; $10.2 million for roads also planned"
    Warwick Online www.warewickonline.com
    Posted Thursday, January 9, 2020

    Sunday, January 12, 2020

    China's Unconventional Nationwide CO2 Emissions Trading System: The Wide-Ranging Impacts of an Implicit Output Subsidy

    China is planning to implement the largest CO2 emissions trading system in the world. To reduce emissions, the system will be a tradable performance standard (TPS), an emissions pricing mechanism that differs significantly from the emissions pricing instruments used in other countries, such as cap and trade (C&T) and a carbon tax. We employ matching analytically and numerically solved models to assess the cost-effectiveness and distributional impacts of China’s forthcoming TPS for achieving CO2 emissions reductions from the power sector.

    We find that the TPS’s implicit subsidy to electricity output has wide-ranging consequences for both cost-effectiveness and distribution. In terms of cost-effectiveness, the subsidy disadvantages the TPS relative to C&T by causing power plants to make less efficient use of output-reduction as a way of reducing emissions (indeed, it induces some generators to increase output) and by limiting the cost-reducing potential of allowance trading. In our central case simulations, TPS’s overall costs are about 47 percent higher than under C&T. At the same time, the TPS has distribution-related attractions. Through the use of multiple benchmarks (maximal emission-output ratios consistent with compliance), it can serve distributional objectives. And because it yields smaller increases in electricity prices than a comparable C&T system, it implies less international emissions leakage.
    by Lawrence H. Goulder, Xianling Long, Jieyi Lu and Richard D. Morgenstern
    National Bureau of Economic Research (NBER) www.NBER.org
    NBER Working Paper No. 26537; Issued in December 2019

    Friday, January 10, 2020

    On the use of Hedonic Regression Models to Measure the Effect of Energy Efficiency on Residential Property Transaction Prices: Evidence for Portugal and Selected Data Issues

    Using a unique dataset containing information of around 256 thousand residential property sales, this paper discloses a clear sales premium for most energy-efficient dwellings, which is more pronounced for apartments (13%) than for houses (5 to 6%). Cross-country comparisons support the finding that energy efficiency price premiums are higher in the Portuguese residential market than in central and northern European markets. Results emphasize the relevance of data issues in hedonic regression models. They illustrate how the use of appraisal prices, explanatory variables with measurement errors, and the omission of variables associated with the quality of the properties, may seriously bias energy efficiency partial effect estimates. These findings provide valuable information not only to policymakers, but also to researchers interested in this area.
    by Rui Evangelista 1, Esmeralda A. Ramalho ad Joao Anrade E. Silva 3
    1. Instituto Nacional de Estatística, rui.evangelista@ine.pt
    2. Universidade de Lisboa, Instituto Superior de Economia e Gestão & REM - CEMAPRE,
    3. Universidade de Lisboa, Instituto Superior de Economia e Gestão & REM - CEMAPRE,
    REM – Research in Economics and Mathematics
    REM Working Paper 064-2019; Lisbon, Portugal; January 2019

    Value of playgrounds relative to green spaces: Matching evidence from property prices in Australia

    We examine the effect on house prices of the presence of a small playground relative to an empty green space using a matching approach combined with hedonic regression analysis. Using a data set of 33,521 property sales in urban Australia, we match properties near small playgrounds to similar properties near two empty, open green spaces which are candidates for playground construction. We control for property characteristics and distance to a wide range of urban amenities and other open spaces. We find that the presence of a playground within 300 metres adds about AU$20,000 (4.6 per cent) to the average property price. The price effect of a playground is larger for houses than apartments and falls with distance from the playground.
    A Melbourne Natural Playground Named Australia’s Best Playground
    • We assess the impact on property prices of the presence of a small playground.
    • We combine hedonic regression with a matching approach.
    • A small playground adds five per cent to the price of properties within 300 meters
    • Playgrounds add more to house prices than non-house property prices.
    • The price impact of playgrounds falls as distance to the playground increases.

    by Robert Breunig 1 Syed Hasan 2 KymWhiteoak 3
    1. Crawford School of Public Policy, Australian National University, Canberra, ACT 0200, Australia
    2. School of Economics and Finance, Massey University, Palmerston North, New Zealand
    3. RM Consulting Group Suite 1, 357 Camberwell Road, Camberwell, VIC 3124, Australia
    Landscape and Urban Planning via Elsevier Science Direct www.ScienceDirect.com
    Volume 190; October, 2019; 103608

    Urban trees, house price, and redevelopment pressure in Tampa, Florida

    We examined the relationship between urban trees and the sales price of single-family homes in Tampa, Florida. We chose Tampa, because the city is facing major redevelopment pressure that may impact the association between trees and house price. In particular, a frequently voiced view in Tampa’s development community is that trees adversely affect the value of houses that are being sold for redevelopment. We estimated hedonic models of sales price controlling for house and neighborhood characteristics and correcting for spatial autocorrelation (n = 1,924). We found that trees within 152m (500 feet) of a house’s lot were significantly associated with higher sales prices. Specifically, a 1-percentage point increase in tree-canopy cover was associated with a total increase in sales price of $9,271 to $9,836 (results were largely insensitive to correction for spatial autocorrelation). Our results demonstrate that, even in a city facing major redevelopment pressure, trees are associated with higher sales prices.

    • Trees on or right next to a house’s lot are not associated with higher sales price.
    • In contrast, houses with more neighborhood trees sell for a price premium.
    • Despite redevelopment pressure, trees are a neighborhood amenity.

    by Geoffrey H. Donovan 1, Shawn Landry 2 and Cody Winter 3
    1. USDA Forest Service, PNW Research Station, 620 SW Main, Suite 502, Portland, OR, 97205, USA
    2. University of South Florida, School of Geosciences, 4202 E Fowler Ave., NES107, Tampa, FL, 33620, USA
    3. Environmental Protection Commission, 3629 Queen Palm Drive, Tampa, FL, 33619, USA
    Urban Forestry & Urban Greening via Elsevier Science Direct www.ScienceDirect.com
    Volume 38; February, 2019; Pages 330-336

    Wednesday, January 8, 2020

    Moral Hazard, Wildfires, and the Economic Incidence of Natural Disasters

    This study measures the degree to which large public expenditures on wildfire protection subsidize development in harm's way. Using administrative firefighting data, we calculate geographically-differentiated implicit subsidies to homeowners throughout the western USA. We first examine how the presence of homes affects firefighting expenditures. These results are used to reconstruct the implied historical cost of protecting each home and to perform an actuarial calculation of expected future protection cost. The expected net present value of this subsidy can exceed 20% of a home's value. It increases with fire risk and decreases surprisingly steeply with development density. A simple model is used to explore effects on expansion of developed areas, density, and private risk-reducing investments. These results demonstrate how policy and institutions influence the costs imposed by a changing climate.
    The Rim Fire burned more than 250,000 acres (1,000 km2) of forest near Yosemite National Park, in 2013https://en.wikipedia.org/wiki/Wildfire
    During 1985–2017 total wildfire property damages in the United States were $51 billion, while direct firefighting costs for federal agencies alone totaled $43 billion.
    A sample of homes including all 8.6 million homes in the western US located near areas of wildland vegetation (44% of all western US residential homes, condos, and apartments)....The range of historical protection costs is large. The average net present value ranges from a few hundred dollars per home for the lowest-cost cells to almost $100,000 per home in the highest-cost cells.
    Using the “suppression only” measure, 50% of WUI homes have expected protection costs under $500, while the highest-risk homes have costs that are much larger. Five percent of homes have expected costs exceeding $3,800. One percent of homes have expected protection costs exceeding $12,700. Using the “suppression plus” measure results in higher costs. The 95th and 99th percentiles of this distribution are about twice as high as for the “suppression only” measure.
    In 2014, California began requiring homeowners in the Cal Fire protection area to pay an equal annual fee of about $150 per year. The fee proved unpopular among homeowners and was suspended in 2017. This study shows that such a fee would need to be much more geographically differentiated in order to correct incentives (as opposed to simply raising revenue). Another lesson from this study is that exempting owners of existing homes could increase the political acceptability of such a policy without reducing its effectiveness, since the protection costs we estimate are not generally high enough to justify abandonment of existing homes. An alternative would be to assign firefighting costs to local governments, which would recover them through property taxes or other measures. This approach would incentivize cities and counties to consider fire protection costs in zoning, land use, and building codes/

    by Patrick Baylis and Judson Boomhower
    National Bureau of Economic Research (NBER) www.NBER.org
    NBER Working Paper No. 26550; Issued in December 2019

    The Private and External Costs of Germany's Nuclear Phase-Out

    Many countries have phased out nuclear electricity production in response to concerns about nuclear waste and the risk of nuclear accidents. This paper examines the impact of the shutdown of roughly half of the nuclear production capacity in Germany after the Fukushima accident in 2011. We use hourly data on power plant operations and a novel machine learning framework to estimate how plants would have operated differently if the phase-out had not occurred. We find that the lost nuclear electricity production due to the phase-out was replaced primarily by coal-fired production and net electricity imports. The social cost of this shift from nuclear to coal is approximately 12 billion dollars per year. Over 70% of this cost comes from the increased mortality risk associated with exposure to the local air pollution emitted when burning fossil fuels. Even the largest estimates of the reduction in the costs associated with nuclear accident risk and waste disposal due to the phase-out are far smaller than 12 billion dollars.

    by Stephen Jarvis, Olivier Deschenes and Akshaya Jha
    National Bureau of Economic Research (NBER) www.NBER.org
    NBER Working Paper No. 26598; Issued in December 2019