Abstract
Researchers and environmental policy advocates have raised questions regarding the distributional impacts of emissions trading programs, a.k.a. “cap-and-trade”. While previous research has been careful to identify the causal effect of emissions trading on emissions reductions (Fowlie et al. in Am Econ Rev 102(2):965–993, 2012, hereafter FHM), we argue that existing estimates of differential impacts on demographic groups have relied on unrealistic assumptions regarding pollution dispersion. In this paper, we estimate the emissions reduction due to the RECLAIM cap-and-trade program in Southern California following the identification strategy of FHM, but we relax the assumption of uniform dispersion surrounding point sources. We model the transport of effluents using a state-of-the-science dispersion model to determine the areas impacted by emissions from each source. Importantly, conditional on race and ethnicity, we find that higher income areas receive larger reductions in pollution under cap-and-trade. Furthermore, conditional on income (or poverty rates), we find that Blacks benefit while Hispanics lose relative to whites under RECLAIM.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
An exception is a working paper by Sullivan (2016), who also uses a dispersion model in the LA Basin. He also finds that poorer households benefit less from air quality improvements.
South Coast Air Quality Management District.
Emission levels in Period 1 is an average of 1990 and 1993 had data from both years are available. Otherwise, emissions levels from either 1990 or 1993 are used. The same is done for year 2004 and 2005 in Period 2. In addition, we independently geocoded each facility, and when replicating the results in FHM, our coefficient estimates are slightly different. Furthermore, there are ten sources that were dropped in FHM’s analysis because they are offshore, do not intersect census blocks, or are uninhabited, but when we model the dispersion of pollution from these sites there are affected communities. These ten firms are included in our analysis.
When we include the power plants into our regression, we do not find significant correlation between any socioeconomic group and \(\mathrm{NO}_x\) reduction. The results can be found in the Appendix Table 6. Appendix Table 7 show results excluding electric generating units, under the assumption that \(\mathrm{NO}_x\) disperses uniformly within a 1-mile radius around each facility.
Although there are fourteen power producers, Table 1 summarizes only thirteen power producers as one firm, Riverside Canal Power Company, did not report emissions after 2001.
To be specific, HYSPLIT gives the average air pollution concentration level at every 0.01 latitude and longitude grid cell, which is roughly equivalent to a 1 by 1 \(\hbox {km}^{2}\). We assume that \(\mathrm{NO}_x\) is being emitted at 65 m above ground level. This is the minimum smokestack height required by the EPA. For each HYSPLIT run, the output concentration level is the average concentration level from 0 to 100 m above ground level, 12 h after a puff of air is emitted from a given source.
While dispersion patterns vary over the year, in estimates not shown here we find no significant difference in the estimates by season. Results estimated separately by season are available from the authors.
See “Appendix” section for a detailed description of how HYSPLIT simulation results are combined with census block data to calculate affected neighborhoods.
Average relative change in emissions, weighted by each demographic group, are calculated from \(\sum _f (R_f D_{f})/\sum _f D_f\) where \(R_f\) is relative change for each facility f, and \(D_f\) are affected people in each demographic group from facility f.
For each census block b with \(D_{fb}\) people affected by each firm f, the block-level, population-weighted relative change is \(\sum _f (R_f D_{fb})/D_{fb}\).
Within each census block, the fraction of number of people affected by a single firm relative to the number of people affected by all firms is the same regardless of demographic group. For example, in census block b, the fraction of blacks affected by firm 1 relative to blacks affected by all firms is equal to the fraction of Hispanics affected by firm 1 relative to all firms. For example, \(\dfrac{{ Black}_{1b}}{\sum _f {{ Black}_fb}} = \dfrac{{ Hispanic}_{1b}}{\sum _f {{ Hispanic}_fb}}\). Let \(D_{fb} = F_{fb}\sum _f {D_fb}\) where \(F_{fb}\) is a constant fraction representing firm f’s effect on census block b. Therefore, census block level population weighted relative change can be expressed as \(\dfrac{\sum _f (R_f F_{fb}\sum _f {D_fb})}{\sum _f D_{fb}} = \sum _f R_f F_{fb}\) which is independent of the number of people living within each census block. Differences in rankings between top beneficiaries or losers from cap-and-trade are only due to the fact that same census blocks do not have certain demographic groups living in them, \(D_{fb}=0\), and are omitted from the analysis.
Population density weighted relative change for each census block b is \(\sum _f R_f D_{fb} / A_b\) where \(R_f\) is the relative change of firm f, \(D_{fb}\) is the number of people in block b affected by facility f, and \(A_b\) is the area of block b.
A list of census blocks that benefit or lose from cap-and-trade relative to command-and-control is available from the authors.
Our results correspond to specifications (3), (5), (6), and (7) in FHM’s Table 7.
Additional results are shown in the Appendix to contrast our results with FHM’s main specification, including estimates that include electric generating units in the sample, and including their definition of Minority (Hispanic and Blacks).
FHM cluster at a higher level, but the result is only ten cluster groups, which may be too few, particularly in unbalanced cases (Cameron and Miller 2015).
14.60 is a linear combination of 16.64 and −2.04 which are the coefficients on Treatment \(\times \) %Poverty and %Poverty, respectively.
3.99 is a linear combination of −4.46 and 0.47 which are the coefficients on Treatment \(\times \) %Black and %Black, respectively.
2.49 is a linear combination of 2.97 and −0.48 which are the coefficients on Treatment \(\times \) %Hispanic and %Hispanic, respectively.
In results not shown here, we estimated similar specifications but including interactions of income with indicators for different racial/ethnic compositions. These results are generally insignificant.
We independently geocoded the sources regulated by RECLAIM, and when replicating the results in FHM our coefficient estimates are slightly different from FHM. Note that we have 821 observations in Table 7 and column (1) Table 8 whereas FHM have 822 observations in their regressions excluding electric facilities. We independently geocoded each facility and finds no population living within a 1-mile radius of Raisch Products located on 1444 Borregas Avenue, Sunnyvale CA 94089. This observation is omitted from our analysis.
References
Abadie A, Imbens GW (2006) Large sample properties of matching estimators for average treatment effects. Econometrica 74(1):235–67
Baker J (2010) A cluster analysis of long range air transport pathways and associated pollutant concentrations within the UK. Atmos Environ 44(4):563–571
Banzhaf HS, Walsh RP (2008) Do people vote with their feet? An empirical test of Tiebout. Am Econ Rev 98(3):843–63
Banzhaf HS, Walsh RP (2013) Segregation and Tiebout sorting: the link between place-based investments and neighborhood tipping. J Urban Econ 74:83–98
Burtraw D, Szambelan SJF (2009) US emissions trading markets for \(\text{SO}_{2}\) and \(\text{ NO }_{x}\). In: Resources for the future discussion paper (09-40)
Cameron AC, Miller DL (2015) A practitioner’s guide to Cluster–Robust inference. J Hum Resour 50(2):317–372
Card D, Mas A, Rothstein J (2008) Tipping and the dynamics of segregation. Q J Econ 2:177–218
Conley Timothy G (1999) GMM Estimation with Cross Sectional Dependence. Journal of Econometrics 92:1–45
de Nevers N (2010) Air pollution control engineering. Waveland Press, Long Grove, Illinois
Draxler RR, Hess GD (1997) Description of the HYSPLIT4 modeling system. In: NOAA technical memorandum ERL ARL-224. NOAA Air Resources Laboratory, Silver Spring, MD
Draxler RR, Hess GD (1998) An overview of the HYSPLIT4 modeling system of trajectories, dispersion, and deposition. Aust Meteor Mag 47:295–308
Draxler RR (1999) HYSPLIT4 user’s guide. In: NOAA technical memorandum ERL ARL-230. NOAA Air Resources Laboratory, Silver Spring, MD
Fowlie M, Holland SP, Mansur ET (2012) What do emissions markets deliver and to whom? Am Econ Rev 102(2):965–993
Fullerton D (2011) Six distributional effects of environmental policy. Risk Anal 31(6):923–929
Fullerton D, Heutel G (2007) The general equilibrium incidence of environmental taxes. J Public Econ 91(3):571–591
Gebhart KA, Schichtel BA, Malm WC, Barna MG, Rodriguez MA, Collett JL Jr (2011) Back-trajectory-based source apportionment of airborne sulfur and nitrogen concentrations at Rocky Mountain National Park, Colorado, USA. Atmos Environ 45(3):621–633
Grainger C (2012) The distributional effects of pollution regulations: do renters fully pay for cleaner air? J Public Econ 96(9–10):840–852
Grainger C, Kolstad C (2010) Who pays a price on carbon? Environ Res Econ 46(3):359–376
Junker C, Wang J-L, Lee C-T (2009) Evaluation of the effect of long-range transport of air pollutants on coastal atmospheric monitoring sites in and around Taiwan. Atmos Environ 43(21):3374–3384
Jacobson MZ (2012) Air pollution and global warming: history, science, and solutions, 2nd edn. Cambridge University Press, Cambridge
Parry IW, Sigman H, Walls M, Williams RC III (2006) The incidence of pollution control policies. Int Yearb Environ Resour Econ 2006(2007):1–42
Ringquist EJ (2011) Trading equity for efficiency in environmental protection? environmental justice effects from the \(\text{ SO }_2\) allowance trading program. Soc Sci Q 92:297–323
Sethi R, Somanathan R (2004) Inequality and segregation. J Polit Econ 112:1296–1321
Sullivan, DM (2016) The true cost of air pollution: evidence from house prices and migration. In: Working paper. Available http://heep.hks.harvard.edu/files/heep/files/dp69_sullivan.pdf. Accessed 29 March 2017
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 Calculating Affected Neighborhoods
A puff of air that is being emitted from a facility affects G 0.01 latitude and 0.01 longitude grid points. Each grid point g receives an aggregate impact of \(c_g\). Emissions from each facility affect B census blocks in California.
Figure 9 illustrates a sample puff of air that disperses to 9 grids (shown in the squares) and affects 3 census blocks (shown in the hexagons). In order to calculate demographic groups that are affected by this puff of air, we have to calculate affected groups within each grid - census block intersecting area; such as the blue area where grid g3 and block b2 intersects.
Sample HYSPLIT grids and census blocks. Notes the squares show sample 0.01 latitude and 0.01 longitude grid points affected by a sample dispersion. The hexagons show sample census blocks
Let \({ area}(g \cap b)\) represent the size of an area in census block b that is affected by grid point g. The effect of emissions in grid g that affects census block b is \(c_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)}\).
The net effect of grid g on all census blocks is \(\sum _{b=1} ^B \left\{ c_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)} \right\} \).
When a facility affects G grids, the total effect of that facility on all census blocks is \(\sum _{g=1} ^G \sum _{b=1} ^B \left\{ c_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)} \right\} \).
We scale this effect such that \(\sum _{g=1} ^G \sum _{b=1} ^B \left\{ \hat{c}_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)} \right\} =1\). Hence, \(\hat{c}_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)}\) would represent the fraction of the effect of a facility’s emissions on a concentration grid-census block intersecting area.
Assume that demographic of interest has \(D_b\) people in census block b. Under the assumption that people distributes uniformly within census block b, \(D_b \times \dfrac{{ area}(g \cap b)}{{ area}(b)}\) people are affected by grid g in census block b.
The number of people affected by a facility’s emission on the concentration grid g - census block b intersecting area is the product of number of people living in that area and the fraction of effect that it receives from that facility, \(D_b \times \dfrac{{ area}(g \cap b)}{{ area}(b)} \times \hat{c}_g \times \dfrac{{ area}(g \cap b)}{{ area}(g)}\).
The total effect of a facility on demographic group D is the sum of such effects across all concentration grid - census block intersecting areas.
1.2 Additional Results
Table 6 shows our main specifications but including electric firms in the sample. Table 7 shows results using FHM’s 1-mile data specification but excluding electric firms. Table 6 shows results using HYSPLIT-simulated data, but Blacks and Hispanics are included as a single minority group (the first column shows results using FHM’s 1-mile radius for comparison).Footnote 23
Rights and permissions
About this article
Cite this article
Grainger, C., Ruangmas, T. Who Wins from Emissions Trading? Evidence from California. Environ Resource Econ 71, 703–727 (2018). https://doi.org/10.1007/s10640-017-0180-1
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10640-017-0180-1