Climate and the Economy

“Does the Environment Still Matter? Daily Temperature and Income in the United States,” by Tatyana Deryugina and Solomon M. Hsiang. December 2014. NBER #20750.

“Will Adaptation to Climate Change be Slow and Costly?” by Alan Barreca, Karen Clay, Olivier Deschenes, Michael Greenstone, and Joseph S. Shapiro. January 2015. SSRN #2552786.

For an economist, the relevant question about climate change resulting from increased carbon concentrations in the atmosphere is whether there are negative economic consequences on net. In 1992, Nobel economics laureate Thomas Schelling argued in the American Economic Review that, for developed countries, “the impact [of climate change] on economic output will be negligible and unlikely to be noticed.”

Since Schelling’s observation, some economists have tried to argue that developed countries would suffer net costs that were more than negligible. To buttress their arguments, they’ve developed Integrated Assessment Models (IAMs) that combine climate modeling with the modeling of future economic outcomes as a result of climate change. These IAMs can produce estimates of the damages from specific levels of carbon emissions, and thus can be used to calculate appropriate Pigovian taxes. Rather than negligible economic effects for developed nations, these models typically estimate damages in the order of about $20 per ton of carbon air emissions.

But IAMs are subject to intense criticism. MIT economist Robert Pindyck, writing in these pages (“Pricing Carbon When We Don’t Know the Right Price,” Summer 2013), states:

The typical IAM has a loss function that relates temperature increases to reductions in GDP. But there is no economic theory behind the loss function; it is simply made up. Nor is there data on which to base the parameters of the function; instead the parameters are simply chosen to yield moderate losses that seem “reasonable” (e.g., 1 or 2 percent of GDP) from moderate temperature increases (e.g., 2° or 3°C).

More recent papers attempt to improve upon the lack of knowledge in conventional IAMs. The paper by Tatyana Deryugina and Solomon Hsiang considers whether small variations in weather outcomes at the U.S. county level actually affect economic outcomes. It examines daily temperature variation and per‐​capita income from 1969 through 2011. The possible range of average temperatures is divided into “bins” with a 3 degree Celsius range. The regressions include year and county fixed effects and the distribution of rainfall; thus the effect of temperature on income is the result of an increase in the number of days a county is in a higher‐​temperature “bin” in some years relative to others.

People who live in areas with the largest number of hot days have a much lower mortality rate from hot days than people who live in areas with fewer hot days.

The results show that personal income per capita “increases slightly as temperatures rise from cool to moderate,” and then decreases “linearly at temperatures above 59 degrees F.” Each 1 degree centigrade increase in daily average temperature above 59 degrees Fahrenheit reduces average daily income by 1.7 percent over the period 1969–2011.

The authors use the coefficients estimated from the 1969–2011 temperature data to conduct 20‐​year simulations using 44 different temperature projections for the period 2080–2099 under “business‐​as‐​usual” (that is, no policy change) assumptions. They determine that increased temperatures at the end of the 21st century would reduce average incomes by 2.31 percent at the end of 2099. The authors don’t say whether they consider that loss to be negligible, nor do they stipulate what carbon tax would be appropriate to avert that loss.

The second paper, by Alan Barreca et al., examines the effect of the adoption of air conditioning on mortality to provide insight into the effects of possible future adaptations to higher temperatures. It analyzes monthly mortality rates and daily temperature from 1900 to 2004 and finds that the effect of extreme heat on mortality is smaller in generally hotter areas.

Extreme heat is defined as the number of days in a month (each year) that a state experiences an average temperature above 90 degrees F. The regressions include state‐​month fixed effects so the effect of extreme heat stems from the variation in the number of above‐​90 days around its long‐​term average. In the 10 percent of states with the highest above‐​90 frequency, the mortality effect of one day above 90 is only a 0.68 percent increase, while in the coldest 10 percent of states the effect of a 90 degree day is a 31 percent increase in mortality.

This evidence is consistent with long‐​term adaptation. People who live in areas with the largest number of hot days have a much lower mortality rate from hot days.

The authors then estimate the effect on the data pre‐ and post‐​1960, when the diffusion of residential air conditioning started. The effect of above–90 degree days on mortality declined 90 percent in both hot and cold states from the earlier era to the latter. Thus the effect of heat on mortality in cold states is still 10 times the effect in hot states after 1960. The effect of an additional 90‐​degree day in cold states from 1980 to 2004 is the same as it was in the hottest 10 percent of states from 1900 to 1939. Adaptation does happen, but it is slow and costly.

Command and Control Versus Externality Taxes

“Toward a Pigovian State,” by Jonathan S. Masur and Eric A. Posner. February 2015. SSRN #2559393.

Why don’t government agencies use Pigovian taxes to remedy externality problems? The conventional wisdom is that they are not permitted to do so by the statutes that govern their actions. In this paper, Jonathan Masur and Eric Posner argue that, to the contrary, current laws do allow agencies to do this.

An agency like the U.S. Environmental Protection Agency could levy a tax in the following way: Set an emissions limitation at zero. Set the fine for emitting a unit of pollution equal to the appropriate Pigovian tax. Initiate an administrative proceeding against all polluters to collect the fines. Voilà—the fines would work exactly like a Pigovian tax.

The remainder of the paper explores regulatory law in a variety of policy areas to convince the reader that command‐​and‐​control regulation could be replaced by externality prices, and such actions would be upheld by the courts.

Shale Gas Development and Housing Values

“Shale Gas Development and Housing Values over a Decade: Evidence from the Barnett Shale,” by Jeremy G. Weber, J. Wesley Burnett, and Irene M. Xiarchos. July 2014. SSRN #2467622.

Most discussions of the costs and benefits of shale gas development describe diffuse benefits for consumers and concentrated costs for the producing areas. Those costs result in externalities from “industrial” development in areas that have not typically experienced them. That tradeoff influences policymakers’ decisions on whether to allow shale oil extraction. For instance, an evaluation indicating high localized costs led New York Gov. Andrew Cuomo (D) to ban horizontal fracturing shale drilling in the state in December 2014.

But do producing areas actually experience net costs? In Texas the value of oil and gas rights is part of the local property tax base. Thus localities receive tax revenues from oil and gas development and can finance local public amenities without increasing property taxes.

How large are those benefits? House values include all public and private costs and benefits that flow from occupancy of a particular home. The Barnett shale area splits the Dallas–Fort Worth area in half; all of the wells are in the western part of the metro area. The authors of this paper seize on that geological accident to determine whether shale development produces net benefits and thus increases housing values.

Over the entire 1997–2013 period, shale ZIP codes in the Dallas–Fort Worth area appreciated 5 to 6 percentage points more than non‐​shale ZIP codes. By 2012 the local tax base of ZIP codes with shale had expanded by $82,000 per public school student. However, within shale ZIP codes, greater well density was associated with less appreciation, indicating some local disamenities. Nonetheless, these results suggest that improved local finances have more than offset whatever disamenities result from shale development for the typical homeowner.

E‑Cigarette Taxation

“Should E‑Cigarettes Be Taxed?” by Alex Brill, Sally Satel, and Alan D. Viard. October 2014. SSRN #2515026.

In these pages, Jonathan Adler et al. recently argued that, for political reasons, heavy taxation of electronic cigarette is forthcoming (“Bootleggers, Baptists, and E‑Cigs,” Spring 2015). This paper, by Alex Brill et al., considers these taxes from a public policy perspective and finds the economic case for such taxation to be lacking.

The federal cigarette tax (currently about $1 a pack) is now five times higher in real terms than it was 32 years ago. Federal revenue from the tax in 2013 was $16 billion, or 0.6 percent of all federal revenue. State taxes range from 17 cents per pack in Missouri to $4.35 in New York. Combined taxes in Chicago are now $7.17 per pack, the highest in the country.

The original rationale for tobacco taxation was that it was a luxury good (rather than for health reasons). But today, tobacco use is inversely related to income, which contradicts the notion of cigarettes as a “luxury.” In fact, cigarette taxes are regressive; 3.2 percent of household income is spent on tobacco in the lowest quarter of the income distribution while only 0.4 percent is spent in the highest quarter.

Both the externality and paternalism rationales now used to justify tobacco taxation depend on adverse health effects, but e‑cigarettes don’t have such effects. There are no carcinogenic combustion gases or particles inhaled either by the consumer or second‐​hand by other people. The only recognized health concern for e‑cigarettes is that the consumer inhales nicotine, but, in the words of Mitch Zeller, the head of the U.S. Food and Drug Administration’s tobacco division, “I’m not saying nicotine is benign, but compared to the risk from regular tobacco it pales.”

According to the authors,

It would make sense to tax e‑cigarettes if they pose serious health risks or lead to an increase in smoking. As discussed, however, the evidence does not point to serious health risks or to e‑cigarettes serving as a gateway to smoking. Therefore e‑cigarettes should not be taxed at this time.

Effects of Extended Unemployment Insurance

“The Effect of Extended Unemployment Insurance Benefits: Evidence from the 2012–2013 Phase‐​Out,” by Henry S. Farber, Jesse Rothstein, and Robert G. Valletta. January 2015. SSRN #2558363.

The spike in unemployment during the last recession rekindled the debate over disincentives from unemployment insurance. On one side, people argue that the duration of unemployment benefits should be extended in recessions because of the moribund jobs market. On the other side, people argue that extending the benefits would reduce the incentive for the unemployed to find jobs, which would perpetuate their unemployment.

This paper, by Henry Farber, Jesse Rothstein, and Robert Valletta, attempts to determine how severe was the employment disincentive effect of unemployment insurance in the wake of the last recession. The authors find that the availability of extended benefits did reduce the monthly exit rate from the program (which is normally around 20 percent) to about 17 percent. But the decline did not adversely affect exit due to employment; rather, it affected exit due to the beneficiary dropping out of the labor force. That is, when extended benefits are ended, people do not increasingly find work; they increasingly drop out of the labor force.

According to the authors,

UI extensions have not had large moral hazard effects on recipients’ job‐​finding rates, either during the worst period of the Great Recession or during the subsequent recovery…. UI extensions around the Great Recession had very limited impacts on labor market efficiency.