How annual incomes in the shadow of oil refineries compare to state and regional prosperity
Typically, we analyze the potential economic impacts of oil refineries by simply quantifying potential and/or actual capacity on an annual or daily basis. Using this method, we find that the 126 refineries operating in the U.S. produce an average of 100,000-133,645 barrels per day (BPD) of oil – or 258 billion gallons per year.
In all of North America, there are 158 refineries. When you include the 21 and 27 billion gallons per year produced by our neighbors to the south and north, respectively, North American refineries account for 23-24% of the global refining capacity. That is, of course, if you believe the $113 dollar International Energy Agency’s 2016 “Medium-Term Oil Market Report” 4.03 billion gallon annual estimates (Table 1 and Figure 1).
Table 1. Oil Refinery Capacity in the United States and Canada (Barrels Per Day (BPD))
|Average Capacity||133,645 BPD||104,471 BPD||228,417 BPD||139,619 BPD|
|Low||Foreland & Silver Eagle Refining in NV & WY, 2-3K BPD||Prince George & Moose Jaw Refining in BC and SK, 12-15K BPD||Pemex’s Ciudad Madero Refinery, 152K BPD||—|
|High||Exxon Mobil in TX & LA, 502-560K BPD||Valero and Irving Oil Refining in QC & NS, 265-300K BPD||Pemex’s Tula Refinery, 340K BPD||—|
|Median||100,000 BPD||85,000 BPD||226,500||109,000|
|Total Capacity||16.8 MBPD||1.8 MBPD||1.4 MBPD||22.1 MBPD|
Census Tract Income Disparities
However, we would propose that an alternative measure of a given oil refinery’s impact would be neighborhood prosperity in the census tract(s) where the refinery is located. We believe this figure serves as a proxy for economic justice. As such, we recently used the above refinery location and capacity data in combination with US Census Bureau Cartographic Boundaries (i.e., Census Tracts) and the Census’ American FactFinder clearinghouse to estimate neighborhood prosperity near refineries.
Our analysis involved merging oil refineries to their respective census tracts in ArcMAP 10.2, along with all census tracts that touch the actual census tract where the refineries are located, and calling that collection the oil refinery’s sphere of influence, for lack of a better term. We then assigned Mean Income in the Past 12 Months (In 2014 Inflation-Adjusted Dollars) values for each census tract to the aforementioned refinery tracts – as well as surrounding regional, city, and state tracts – to allow for a comparison of income disparities. We chose to analyze mean income instead of other variables such as educational attainment, unemployment, or poverty percentages because it largely encapsulates these economic indicators.
As the authors of the UN’s International Forum of Social Development paper Social Justice in an Open World wrote:
In today’s world, the enormous gap in the distribution of wealth, income and public benefits is growing ever wider, reflecting a general trend that is morally unfair, politically unwise and economically unsound… excessive income inequality restricts social mobility and leads to social segmentation and eventually social breakdown…In the modern context, those concerned with social justice see the general increase in income inequality as unjust, deplorable and alarming. It is argued that poverty reduction and overall improvements in the standard of living are attainable goals that would bring the world closer to social justice.
Environmental regulatory agencies like to separate air pollution sources into point and non-point sources. Point sources are “single, identifiable” sources, whereas non-point are more ‘diffuse’ resulting in impacts spread out over a larger geographical area. We would equate oil refineries to point sources of socioeconomic and/or environmental injustice. The non-point analysis would be far more difficult to model given the difficulties associated with converting perceived quality of life disturbance(s) associated with infrastructure like compressor stations from the anecdotal to the empirical.
Primarily, residents living in the shadow of 80% of our refineries earn nearly $16,000 less than those in the surrounding region – or, in the case of urban refineries, the surrounding Metropolitan Statistical Areas (MSAs). Only residents living in census tracts within the shadow of 25 of our 126 oil refineries earn around $10,000 more annually than those in the region.
On average, residents of census tracts that contain oil refineries earn 13-16% less than those in the greater region and/or MSAs (Figure 2). Similarly, in comparing oil refinery census tract incomes to state averages we see a slightly larger 17-21% disparity (Figure 3).
Oil refinery income disparities seem to occur not just in one region, but across the U.S. (Figure 4).
The biggest regional/MSA disparities occur in northeastern Denver neighborhoods around the Suncor Refinery complex (103,000 BPD), where the refinery’s census tracts earn roughly $42,000 less than Greater Denver residents1. California, too, has some issues near its Los Angeles’ Valero and Tesoro Refineries and Chevron’s Bay Area Refinery, with a combined daily capacity of nearly 600 BPD. There, two California census associations in the shadow of those refineries earn roughly $38,000 less than Contra Costa and Los Angeles Counties, respectively. In the Lone Star state Marathon’s Texas City, Galveston County refinery resides among census tracts where annual incomes nearly $33,000 less than the Galveston-Houston metroplex. Linden, NJ and St. Paul, MN, residents near Conoco Phillips and Flint Hills Resources refineries aren’t fairing much better, with annual incomes that are roughly $35,000 and nearly $33,000 less than the surrounding regions, respectively.
Click on the images below to explore each of the top disparate areas near oil refineries in the U.S. in more detail. Lighter shades indicate census tracks with a lower mean annual income ($).
Clearly, certain communities throughout the United States have been essentially sacrificed in the name of Energy Independence and overly-course measures of economic productivity such as Gross Domestic Product (GDP). The presence and/or construction of mid- and downstream oil and gas infrastructure appears to accelerate an already insidious positive feedback loop in low-income neighborhoods throughout the United States. Only a few places like Southeast Chicago and Detroit, however, have even begun to discuss where these disadvantaged communities should live, let alone how to remediate the environmental costs.
Internally Displaced People
There exists a robust history of journalists and academics focusing on Internally Displaced People (IDP) throughout war-torn regions of Africa, the Middle East, and Southeast Asia – to name a few – and most of these 38 million people have “become displaced within their own country as a result of violence.” However, there is a growing body of literature and media coverage associated with current and potential IDP resulting from rising sea levels, drought, chronic wildfire, etc.
The issues associated with oil and gas infrastructure expansion and IDPs are only going to grow in the coming years as the Shale Revolution results in a greater need for pipelines, compressor stations, cracker facilities, etc. We would propose there is the potential for IDP resulting from the rapid, ubiquitous, and intense expansion of the Hydrocarbon Industrial Complex here in the United States.
N. American Hydrocarbon Industrial Complex Map
Footnotes and Additional Reading
- The Suncor refinery was implicated in a significant leak of tar sands crude associated benzene into the South Platte River as recently as 2013. According to Suncor’s website this refinery “supplies about 35% of Colorado’s gasoline and diesel fuel demand and is a major supplier of jet fuel to the Denver International Airport. The refinery is also the largest supplier of paving-grade asphalt in Colorado.”
- New York Times story on the growing footprint of BP’s Whiting Refinery: Surrounded by Industry, a Historic Community Fights for Its Future
By Ted Auch, PhD – Great Lakes Program Coordinator, FracTracker Alliance