Delivering US mail emits:
Ounces of CO2e
Grams of CO2e
1.4 41 per letter
5.1 145 per 6-ounce magazine (165g)
12 350 per 1-pound package (454g). All these are based on:
1.0 27 per piece, for facilities and local delivery, and
0.71 per ounce of mail for air shipment, or
0.71 per gram of mail for air shipment.
Based on a 2001 study (below), which is the most complete.
We divided the CO2e from buildings and local delivery equally across every item mailed, since space, sorting, and handling are roughly equal for each item, regardless of weight. Heavy or large items may need more space, but small items generally pay for faster service, which in its own way may need more space & handling, so equal CO2e per item seemed the best balance, pending more data.
We divided the CO2e from long distance transport by weight, since this CO2e is mostly from air transport, which is sensitive to weight.
The table above adds up these components of CO2e for the postal service. The table below also adds emissions from creating the documents mailed: 1.1 ounces of CO2e to make each ounce of office paper (1.9 for magazine paper) and up to 0.8 ounce CO2e per ounce of paper to print it.
The table below compares those totals for paper mail to total CO2e of electronic alternatives. About 58% of CO2e for electronic alternatives is to make the user's computer and LCD screen. They take 2,140 lb (971 kilos) CO2e to make and are likely to be replaced with new models after two years (30 hours/week active use, 3,100 hours). Thus, making the computer and screen averages 11 ounces (310g) CO2e per hour. The rest is for electricity. Using a solar charger would increase the manufacturing CO2e while eliminating CO2e from electricity.
None of these figures includes making or powering the internet cloud.
Comparing Electronic Alternatives (screen time) with Paper Mail (paper, printing, delivery):
Ounces of CO2e
Grams of CO2e
2.3 64 per greeting card (paper)
2.8 80 per letter (6 pages, paper)
21 583 per 6-ounce magazine (165g, paper)
3 86 per electronic card (10 minutes on screen to select & enjoy)
9 259 per letter (30 minutes on screen)
55 1555 per electronic magazine (3 hours on screen, reading only)
12 346 per electronic magazine (3 hours on iPad, reading only)
Electronic alternatives are best for items which most recipients do not read, like annual reports and bills. All calculations and sources are given in the "Services" tab of our spreadsheet xls.CO2List.org.
All the following are by Michael E. Canes, Logistics Management Institute. There is a 2008 study which seems less complete, showing only fuels used.
This report provides an estimate of USPS greenhouse gas emissions for 2001. The estimate covers a wide variety of USPS activities that generate greenhouse gases, including facility and vehicle operations, contract transport, employee travel, and building landscaping. These cover the great majority of USPS GHG emissions. A few categories, e.g., servicing of chillers, are left out, mainly because the numbers are inconsequential. GHG emissions from Postal Service employee commuting are estimated, but are treated as a separate item and not included in the total.
The USPS has been estimating its greenhouse gas emissions annually since 1998. It also estimated selected previous years’ GHGs, going back to 1985. Other years for which USPS has estimates are 1990, 1991 and 1995.
Interest in emissions of greenhouse gases has been growing for several years. Within the federal government, Executive Order 13123 of June 3, 1999, requires agency facilities to reduce their greenhouse gas emissions by 30% relative to 1990 levels by the year 2010. Also, Executive Order 13149 requires federal agencies to reduce their vehicle petroleum use by 20% relative to 1999 levels by 2005. This latter order too would reduce agency greenhouse emissions. Though USPS is not bound by Executive Orders, it often attempts to follow them in spirit. Thus, these executive orders indirectly have some effect on USPS objectives.
Recently, many Western nations have moved towards ratification of the Kyoto Protocol, an international treaty to control GHG emissions. President Bush has indicated that he will not submit the treaty for Senate ratification, but has committed the U.S. to reducing the GHG intensity of its GDP by 18% by 2012. This would require an annual reduction in the GHG intensity of GDP of about 1.5% per year between 2001 and 2012.
USPS has not as yet put a plan in place to alter the path of its GHG emissions. However, in 2001 it commissioned a report on possible means to reduce GHGs, and its recent agreement with Federal Express in which FedEx will transport mail on behalf of USPS may reduce USPS GHG intensity in the future.
This report proceeds as follows. In section 2 the methodology underlying the GHG estimates is presented. Section 3 describes sources of data, Section 4 presents the results for 2001 and compares these to those from 2000 and earlier years, and Section 5 offers conclusions.
USPS GHGs result primarily from its operation of facilities and its transport of mail. The operation of facilities mainly involves the use of energy in the form of electricity, fuel oil, natural gas and liquid petroleum gases (LPG) , and also includes the disposal of waste products, to landfills, incinerators, or to recycling programs. Transport involves the operation of vehicles and the leasing of space on the vehicles of others. Both are accounted for in the GHG estimates.
Calculation of GHGs from the use of fuel oil, natural gas or LPG at facilities is straightforward. Factors exist to convert the combustion of these fuels into tons of carbon dioxide released, so that all that is necessary is to secure data on the amounts used.
To properly estimate GHGs from electricity use, it is necessary to know what fuels were used to produce the electricity. Because the Postal Service is ubiquitous throughout the United States, it is assumed that the national mix of fuels used to produce electricity also is used to produce the electricity consumed by USPS. Given that assumption, it is straightforward to convert USPS electricity consumption for the year into tons of coal, gallons of fuel oil and billions of cubic feet of natural gas used to product the electricity. These then are converted into GHGs through use of standard conversion factors.
GHGs from waste disposal also are attributed to the operation of facilities. Much of this waste consists of undeliverable mail as well as other paper products. In 1996, when a survey was taken of USPS recycling of its waste products, over 1 million tons of such waste were generated per year.
Waste can be disposed of by incineration, burying at landfills, or by recycling. Though recycling requires some use of energy, it produces fewer GHGs than either incineration or burying at landfills. Unfortunately, the only data available on USPS waste recycling come from its 1996 survey. Neither waste volumes nor the percentage recycled are known since then. To account for GHGs from waste disposal, the proportion of USPS waste recycled in 1996 is assumed to have continued to the present, with the annual volume of waste related to the volume of mail handled. GHG emissions from waste thus are assumed to have increased in proportion to mail volume. As the latter figure has increased by 13% since 1996, so too have GHGs attributed to waste disposal.
This technique might overstate USPS GHGs from waste disposal. At the national level, waste disposal through landfill or incineration has decreased slightly since 1996. If USPS waste volume has increased less than the volume of mail since that time, or if the percentage of USPS waste recycled has increased, then simply assuming an increase in proportion to mail volume overstates USPS GHGs from this source. And, as GHGs from waste disposal make up over 25% of all facility GHGs, such a bias, if present, could have a discernible impact on the overall USPS total.
Finally, an adjustment is made to facility GHGs to account for those facilities that are leased and where USPS does not pay for utilities. In such cases, there is no record of energy use so that it is necessary to estimate it. From survey data, about 4.3% of all USPS floor space is leased with utilities already paid for, and it is assumed that these facilities use the same energy per square foot as all other USPS facilities. Thus, the facility total for GHGs is adjusted by 4.3% to account for these facilities.
Vehicle GHGs result directly from the burning of gasoline, diesel and compressed natural gas (CNG) by the USPS fleet. These are calculated from USPS annual fleet fuel consumption data. Account is taken of biodiesel, which is comprised of 80% diesel and 20% fuel from renewable sources. In particular, 80% of biodiesel consumption is included in the diesel consumption figure. Also, CNG consumption is added to the gasoline consumption figure by converting the CNG on the basis of relative energy content (CNG has about 75% the energy content of gasoline.)
Account also is taken of nitrous oxide and methane from postal vehicles, as both of these gases are GHGs. The method is to multiply the ratio of the Postal Service fleet to the entire U.S. vehicle fleet, times total national nitrous oxide and methane GHGs from mobile sources. These latter are estimated annually by EPA.
In addition to its own vehicles, the Postal Service leases space on commercial air, dedicated air, trucking, rail and ships. An internal USPS study was done of the fuel consumed by commercial air, dedicated air and trucking in FY2001. The study estimated FY2001 fuel use by these sources from contract data for AP5 of 2001, from FY2000 data and from past annual growth. The numbers generated by this study were used to estimate FY2001 GHGs from these sources.
Rail and ship transport GHGs were estimated differently. USPS spending on these forms of transport relative to national spending on freight movement by rail and ship were calculated, and these ratios then applied against total annual fuel used by rail and ships in freight movement. While this methodology yields only very rough estimates of rail and ship GHGs, the numbers are small and therefore any biases do not much affect the total number of USPS GHGs for the year.
The Postal Service also rents the use of vehicles from its rural route carriers. FY2001 rural route fuel use data was estimated by USPS from contracts submitted for renewal at the end of FY2000. This estimate was used to estimate GHGs from this source in the present study.
Employee Business Trips & Landscaping
Two other sources of USPS GHGs are included, employee business trips and landscaping. Data exist for employee airline trips in FY1999, and these were used to estimate fuel used by air carriers, rental cars, and taxis to transport USPS travelers in that year. Since then, GHGs from employee trips have been estimated by assuming the same number of trips per USPS employee as in FY1999, and adjusting each year’s estimate for the total number of Postal Service employees. This method was used for the FY2001 estimate.
A somewhat similar approach is taken towards landscaping. From data produced in a FY1999 survey of the Chesapeake Bay area, an estimate was made of pervious land at retail postal facilities in that area, and the number extrapolated to the national level. Assumptions then were made regarding how much of that land is lawn, and how often the acreage is cut. From that, fuel used for landscaping was estimated, and GHGs calculated. The number for FY2001 is inferred from the FY1999 number based on the respective total number of USPS facilities in the two years.
GHGs from employee commuting are estimated in this report but not included in USPS totals. Such commuting generates a fairly large amount of GHGs, but there is question whether USPS has control over their generation. Also, it can be argued that most employees would commute elsewhere if they were not working for USPS, so that there is no net increase in commuting GHGs from USPS activity.
Employee commuting GHGs were estimated for FY1999 from the number of fulltime and parttime Postal Service employees working in that year, and from assumptions about the average length of their commute, the number of days they work, and the fuel economy of their vehicles. In particular, it was assumed that national average numbers apply to Postal workers, and consequently national average numbers were used for the average length of commute and vehicle fuel efficiency and also for the average number of commuting individuals per vehicle. For FY2001, GHGs from employee commuting have been estimated by taking the total number of USPS employees relative to the number in FY1999.
Excluded Sources of USPS GHGs
A few other possible sources of USPS GHGs have been examined in the past, but excluded from further consideration because the numbers are very small. For example, postal facilities generally have chillers and some leakage of HFCs could be occurring from these. However, the leakage rate is low for facility chillers, and most existing facilities probably still have freon, which is not a greenhouse gas.
Another category is postal parking lots, as the laying and care of asphalt generates a certain amount of GHGs. However, when the amount of GHGs from such asphalt was calculated for 1998, it was too small to matter for the USPS total.
Finally, a certain amount of wastewater is produced at USPS facilities, and energy is required to pump the wastewater out and also to process the waste. However, most of the wastewater at postal facilities is associated with human waste, and since such waste would be produced and disposed of independent of USPS activity it is not included as a source of USPS GHG emissions.
III Sources of Data
Data on fuel consumed at postal facilities was taken from the FY2001 Annual Report on Energy Management and Conservation Programs submitted by the Postal Service to the U.S. Department of Energy (DOE). These data are estimated by USPS from facility expenditures on energy, deflated by current prices. Because this technique is used, the numbers are only approximations to actual facility energy consumption. Better estimates of facility GHGs will be possible when the Postal Service is able to obtain facility energy consumption data more directly.
Most of the vehicle fuel consumption data was obtained from the Postal Service FY2001 Federal Automotive Statistical Tool (FAST) submission to DOE. This submission uses information compiled in the USPS Fuel and Oil Usage by Make/Model National Summary Report, and includes information on biodiesel and CNG consumption.
Data on USPS expenditures on rail and ships were taken from the 2001 Comprehensive Statement on Postal Operations. Other USPS information, e.g, the numbers of facilities and employees, and expenditures on mail, was taken from the USPS Annual Report for 2001.
National data on electricity production and on fossil fuels used to produce electricity were obtained from DOE’s Monthly Energy Review. Data on national nitrous oxide and total US GHGs were taken from EPA’s Inventory of US Greenhouse Gas Emissions and Sinks: 1990-2000 report (draft report read from the EPA website).
Finally, data on rail and ship freight expenditures were taken from US Department of Transportation statistical tables, as were data on rail and ship consumption of fuel by these modes of transport.
The results indicate that facilities and third party transport are the two main sources of USPS GHGs. Postal Service vehicles comprise only about 12 1/2% and “other” sources less than 1%. Thus, to be effective, efforts to deal with USPS GHGs would have to center mainly on facilities and on third party transport.
The recent contract with FedEx to transport a portion of mail on that company’s airplane fleet may result in reduced GHGs from dedicated air. The transport began in August 2001, almost the beginning of the current USPS fiscal year (September 8, 2001). As FedEx is billing USPS for fuel used in the transport of mail, it should be possible to ascertain next year whether fuel reductions in fact occur in FY2002.
In 2000, the latest year for which data is available, U.S. net emissions of GHGs were 6091.5 Tg of CO2 equivalent. Assuming the number did not much change in 2001, USPS was responsible for about .2% of the country’s GHG emissions in that year.
Employee commuting accounted for almost 2 Tg of CO2 equivalent. If this were added to the USPS total, that number would be almost 16 Tg of CO2 eq, and USPS GHGs would make up about .26% of the national total. However, as stated above, it is questionable whether employee commuting GHGs legitimately are attributable to USPS, and hence they are not included in the USPS total.
Overall, USPS greenhouse gases rose by 3.8% between the two years. Also, all of the major sources, specifically facilities, USPS vehicles, and contract transport, increased for the year. For the major categories of contract transport; commercial air, dedicated air and trucking, the USPS contracts office estimated in April 2001 that these had increased 3% between FY2000 and FY2001 and this estimate is used herein.
GHGs per piece of mail and per real dollar spent on mail increased slightly more than GHGs in 2001 because both the volume of mail and the amount spent in real terms declined, by .2% and .3% respectively.
USPS has been tracking its GHGs for several years, and has estimates for selected years going back to 1985. However, estimates before 2000 were based on approximations of fuel used to transport mail by contract airlines and trucking inferred from the volume of mail carried by these sources. With the internal USPS study done on contract transport in FY2001, much better data on fuel used by dedicated air, commercial air and trucking became available for the first time. For that reason, estimates from before 2000 are not comparable to those since.
The increase in total GHGs from 1985 to 1999 was about 50%, and the increase from 2000 to 2001 almost 4%.
GHGs per piece of mail show some increase from 1985 to 1999, and again between 2000 and 2001. However, the increase clearly is much less, only about 4% between 1985 and 1999, and then 4% again between 2000 and 2001. GHGs per constant dollar spent on mail rose by even less between 1985 and 1999, only 1%, though this metric also rose by 4% between 2000 and 2001. One inference is that the volume of mail is an important driver of USPS GHGs. More mail implies more facilities or more intense use of existing facilities and more transport.
However, it appears that volume is not the only driver. Increasing real expenditures on mail probably are related to speed of delivery, higher prices being paid to transmit mail more quickly, overnight or within two days maximum. To the extent such delivery requires more energy per unit, GHGs will rise at a more rapid rate than volume, in particular more in line with real dollars spent.
USPS greenhouse gas emissions rose in 2001, a pattern consistent with data from previous years. In the past, such emissions have been associated with volume and real dollars spent on mail, particularly with the latter. In 2001, however, both the volume and real dollars spent on mail decreased slightly while GHGs increased. In fact, all major categories of USPS GHGs rose, suggesting more energy was used to process and deliver mail in that year.
There appear to be a number of reasons. First, energy used to deliver mail may be related more to the number of customers than to absolute volume. The population of the United States continued to increase in 2001, and USPS’ universal service obligation forces it to deliver to new addresses even if the number of pieces of mail per address may be falling.
Data on the number of rural routes tends to support this reasoning. At the end of FY2000 there were 67,372 rural route carriers, while one year later, at the end of FY2001, there were 69,066, an increase of 2.4%. The number of new rural routes may be a rough proxy measure for the number of new addresses to be served. By implication, more energy was used to deliver mail to more addresses, both through additional carriers and additional transport to pickup spots for these carriers.
USPS facility space also increased in FY2001, by about 6 million square feet or almost 2%. Increased facility space implies more energy used for heating and cooling, and also more for processing equipment, which is energy intensive. Also, estimated energy use at facilities rose on a per square foot basis in 2001, by about 5%. Most likely, this was due to weather conditions, particularly summer heat.
Finally, investment in more energy efficient facilities and vehicles within the past few years has been constrained by overall USPS budget limitations. If these limitations ease somewhat in FY2002, there could be some reduction in energy use from these sources and hence in GHGs.
The United States government is paying more attention to greenhouse gas issues than in the past because of international pressure to conform to the Kyoto protocol, if not to ratify it. It is possible that USPS, along with agencies of the U.S. government, may be asked to do more to control such gases in future than in the past. If that were to occur, USPS has been tracking its GHGs and now has data covering 17 years (1985-2001). This should provide it a head start on any organized effort to deal with GHGs, in particular an understanding of where its GHGs originate, and what the principal factors affecting them might be.
ESTIMATED 2000 GREENHOUSE GAS EMISSIONS OF THE U.S. POSTAL SERVICE AND COMPARISON TO 1999
Estimated USPS greenhouse gas emissions in 2000 were 3.667 million metric tons of carbon (mmtce). Of the 3.667 mmtce, about 38% are attributable to facilities, 12% to vehicles, and 49% to contract transport. The total and the fractional estimates are not directly comparable to previous years’ estimates because more precise data were obtained for 2000 on third party transport and on rural route fuel use. These data were obtained from the USPS transportation contract office, and represent direct amounts of fuel charged to USPS for commercial air, dedicated air, contract trucking and private use of vehicles on rural routes. In previous years, most of these data had to be indirectly estimated, by other means.
The main change from using USPS transportation contract data is a substantial reduction in commercial air fuel use attributed to the carriage of mail. In previous years this metric was estimated by taking the proportion of commercial air revenue ton miles represented by mail carriage and multiplying it by the total fuel use of the commercial airlines. This method charged USPS with its proportionate share of total annual commercial air fuel consumption. From the contract fuel data, it appears that the commercial air lines charge USPS for a much smaller fraction, only about one quarter of its proportionate share of revenue ton miles. Presumably this reflects the commercial airlines’ estimate of the incremental cost that mail carriage represents.
Contract fuel data were available only for a snapshot in time, so that it was not possible to compare the 2000 USPS ghg estimate with previous years using these data. Instead, a second, pro forma estimate for 2000 was made using the previous estimation method for ghgs attributable to USPS from use of commercial air. This allows comparison between 1999 and 2000 even though it overstates total ghgs attributable to USPS for both years. Future estimates will be based on transportation contract data so that it should be possible to compare 2000 to 2001 and subsequent years on a consistent basis.
Facility ghgs increased by 3.9%, vehicle ghgs by 8.7% and contract transport ghgs by 9.2% under this approach, with total ghgs up by 7.5%.
Facility ghgs increased mainly because of an increase in electricity use between the two years. Facility electricity consumption rose from 4.544 to 4.856 million KWh between 1999 and 2000, principally because of an increase in facility square footage, from 291 to 309 million net square feet. As electricity comprises about 87% of facility energy expenditures, changes in electricity consumption largely determine changes in overall USPS facility energy consumption.
Waste disposal ghgs also increased, by a little over 3%. This estimate is made indirectly, as there is no current data on facility recycling and waste disposal. Survey data from 1996 are utilized, with other years estimated from the growth of mail from that year.
Vehicle ghgs rose partly because fuel use data for USPS-leased vehicles were obtained for the first time, but mainly because of an increase in gasoline consumption attributed to private rural route vehicles. In previous years such ghgs were estimated from the number of private rural routes, their average distance in miles, and national light duty vehicle mpg data. In 2000, by contrast, direct data on private rural route fuel use were obtained from USPS. As a result, estimated private rural route fuel consumption increased from 39.4 to 49.5 million gallons, with a commensurate increase in ghgs.
Using the previous method for estimating USPS ghgs from commercial air carriage, total indirect vehicle transport ghgs increased by 9.2%. This was primarily due to an increase in estimated ghgs from air transport, particularly dedicated air. Commercial air ghgs increased from 518 to 540 gallons, while dedicated air increased from 165 to 263 million gallons. Unlike the situation with commercial air, where direct data were obtained from USPS only in 2000, data on dedicated air were obtained from USPS in both years.
Third party truck transport showed a decline of almost 7% in 2000 relative to 1999. Data on truck transport were obtained from USPS in both years. Rail and ship transport were indirectly estimated and showed gains of 6.8% and 7.5% respectively. However, because USPS ships most mail by air and truck, rail and ship ghgs are a small percentage of total third party transport ghgs.
Overall, using the previous estimation method for commercial air, USPS ghgs increased by 7.5% from 1999 to 2000. The principal causes of the increase were the large increase in dedicated air, a somewhat smaller increase in commercial air, an increase in facility electricity usage, and an increase in estimated rural route gasoline usage.
Because part of the 7.5% increase may have been the result of obtaining better data, particularly on third party transport, some caution is in order in interpreting this number. However, with 3.1% growth in mail, more use of dedicated air, and a 6.2% increase in facility floor space from 1999 to 2000, there is little doubt that there was a fairly substantial year to year increase in USPS ghgs.
The better data obtained for 2000 suggest future estimates of USPS ghgs will be more accurate than those for past years. Also, the numbers likely will be somewhat lower, mainly because of the decrease in estimated commercial air fuel use. For example, using the previous estimation method total USPS ghgs would have been 4.784 mmtce in 2000 rather than 3.667 mmtce. The former number would have been about 2.5% of total United States ghgs whereas the latter is about 2.0%.
USPS ghg estimates for 2001 probably will mainly reflect whatever increase may occur in the total volume of mail. However, for 2002, the working agreement between USPS and FedEx should result in a significant net decrease in dedicated air fuel usage, even after attributing increased air fuel use to FedEx to reflect its increased mail carriage. In consequence, there could be an overall reduction in that year.
ESTIMATED 1999 GREENHOUSE GAS EMISSIONS OF THE U.S. POSTAL SERVICE AND COMPARISON TO 1998
Estimated USPS greenhouse gas emissions in 1999 were 4.449 million metric tons of carbon (mmtce). This total represents an increase of 6.3% over 1998, and an increase of 24.7% over a base year of 1990. However, ghgs per piece of mail were up by less; 3.8% from 1998 and 3.0% from 1990. Ghgs per constant $ spent on mail were up 4.9% from 1998 but were down 6.0% from 1990.
In 1999 as in past years, emissions from indirect or third party transport made the single largest contribution to the total. In all, these emissions represented 60% of total USPS ghg emissions. Emissions from USPS facilities represented about 30%, and emissions from vehicles, including rural routes, were a little over 9%. These proportionate contributions have remained stable through the 1990s. In 1990 itself, for example, indirect transport contributed 58%, facilities 31% and vehicles 10%.
In the past, third party transport emissions were estimated indirectly, from data on USPS expenditures on such transport taken as a proportion of national spending. Emissions from dedicated air were estimated by assuming that gallonage per pound shipped was the same for dedicated as for commercial air. For 1999, however, direct data on dedicated air and third party truck transport fuel usage were obtained from USPS. These numbers were somewhat higher than what had been estimated under the earlier procedures. Thus, a part of the increase between 1998 and 1999 may be accounted for by the use of more precise data on third party transport emissions.
USPS facility ghgs for 1999 and 1998 changed very little between the two years, with power and natural gas emissions down slightly and fuel oil and waste disposal emissions up a little. Unfortunately, there is no direct annual data on USPS waste disposal and these emissions are estimated from a 1996 base, adjusted for the volume of mail in any given year. Thus, the increase from 1998 to 1999 simply reflects an increased volume of mail.
Direct vehicle ghgs were up somewhat between 1998 and 1999. This increase took place mainly from increased use of gasoline, both in the USPS-owned and the rural fleets.
By far the largest portion of the increase from 1998 to 1999, however, was for third party air and truck transport. The combination of these two subcategories accounts for fully 91% of the year to year increase (239,000 metric tons of the 262,000 total increment).
Third party air breaks down into two classes; commercial and dedicated air, and a large part of the air increment comes from an increase in the dedicated class. As stated above, direct data on dedicated air fuel usage data was obtained from USPS for the first time, and it appears from these data that previous estimates of dedicated air fuel usage, including the 1998 estimate, were on the low side. In addition, however, USPS reported in its 1998 and 1999 Comprehensive Statements on Postal Operations that spending on air transport increased from $1.34 billion to $1.8 billion between the two years. Such a large increase is at least consistent with a sizeable increase in air transport related ghgs, though it does not directly demonstrate that.
The other category with a large increase was third party trucking, whose estimated emissions rose by 8.9% between 1998 and 1999. A part of this may be due to obtaining direct rather than indirect data. However, the 1999 Comprehensive Statement on Postal Operations reports that USPS expenditures on third party trucking were up by 7.1% between the two years. Such an increase in expenditures on trucking is largely consistent with the estimated year to year ghg increase for this subcategory.
Summarizing, USPS ghgs increased about 6% between 1998 and 1999, largely because of increases in third party air and truck transport. A part of the apparent increase may simply be due to acquisition of better data for 1999, particularly for dedicated air and third party transport, but other indicators, specifically USPS expenditures on air and trucking, tend to support the estimated increments. Other categories showed relatively little change. An ongoing implication of LMI’s annual USPS ghg estimates is that third party transport appears to be a key area to examine. Publicly announced efforts by USPS to rationalize its air and ground transport operations through strategic partnerships with other delivery companies could well achieve important economies of ghg emissions as well as savings in costs.
US POSTAL SERVICE: GREENHOUSE GAS INVENTORY -- 1998
Total USPS GHG emissions in 1998 are estimated at 4.01 million metric tons of carbon equivalent (mm TCE). Of these, 60% resulted from USPS use of 3rd party transport, principally air freight and trucking. Another 32% resulted from facility operation, and the remaining 8% from direct vehicle use. The USPS total is about .2% of US total GHG emissions for 1998. These data are displayed on the next page.
The numbers differ from my preliminary estimate in the following ways. First, I have included facility solid waste disposal, which produces methane in landfills and carbon dioxide through recycling. Second, I have obtained data on facilities rented by USPS where utilities are included in the rent. Emissions from such facilities now are included. Third, I have included methane and nitrous oxide emissions from vehicles, which adds a small amount to direct vehicle emissions. Fourth, more recent data on air transport and on trucking slightly alter the emissions from these 3rd party transport means. And finally, I have put all numbers into metric tons of carbon equivalent rather than CO2 as in the previous memo. This is because both EPA and EIA provide data and express national totals in these terms.
My estimates for per piece mail GHG emissions change accordingly. These now are as follows (CO2 numbers in parentheses for comparison purposes) :
$1 of mail expenditure results in .15 lbs of carbon equivalent (.56 lbs CO2) 1 piece of mail results in .04 lbs of carbon equivalent (.16 lbs CO2) 1 lb of mail results in .37 lbs of carbon equivalent (1.35 lbs CO2)
SUMMARY OF USPS GHG EMISSION SOURCES AND TOTALS FOR 1998 (mm TCE)
Finally, as you asked I have offered opinions on how confident we can be of each of the individual estimates. I did not try to analyze these statistically, but instead base them on how strong the data seem and what sorts of assumptions were necessary to get to an estimate that I feel is defensible.
Facility Energy Use In FY1998, USPS facilities directly consumed 4.5 MWH of electricity, 4.7 million gallons of fuel oil, 5.7 million gallons of LPG and 7.4 BCF of natural gas (USPS, 1998 Annual Energy Management Report). It was assumed that electricity utilized by the USPS was reflective of the national average in 1998, where coal produced 57.1%, natural gas 14.7%, and oil 3.6%. These fuels and the fossil fuels consumed directly by USPS facilities were converted into quads of energy and then into carbon equivalents through conversion factors shown in an attachment hereto. For most space that USPS owns or leases, it is responsible for all utilities. In addition, however, USPS leases space for which the landlord pays utilities (e.g., in shopping centers, small malls and the like). According to data obtained from FMSWIN, as of March 2000 this amounted to 13.5 million square feet, or about 4.3% of the amount where utilities are paid directly by USPS. I assume that this proportion has remained the same over time, and so applies to 1998 data. The amount of carbon equivalent produced by USPS facility energy use therefore was increased by 4.3% to reflect this factor.
The data for these estimates are quite good, so that we can have reasonable confidence that we have accurately represented carbon emissions from facility energy use.
Two other facility-related categories were considered but ultimately ignored. These are emissions related to land holdings and from chillers and refrigerators in buildings.
The holding of large tracts of land could result in both generation and absorption of GHGs. However, for this to be a meaningful source of GHG contribution, the amount of land would have to be quite large, with farming, forestry or animal husbandry typically involved. To ascertain the extent of USPS landholdings, the office of the Facilities Environmental Program Coordinator was contacted. According to that office, the USPS discontinued its land banking program in 1992 and since has acquired land tracts on an as needed basis, when new facilities are planned. Also, most of the tracts acquired are relatively small, with the largest among them no more than around 50-60 acres and many far smaller, with little or no agricultural activity involved. Thus, no greenhouse emissions were attributed to USPS in 1998 from this source.
Chillers and refrigerators in buildings are potential sources of high potential GHGs such as hydroflourocarbons. However, EIA indicates that facility chillers leak at very slow rates, basically only when they are serviced (EIA 1999). And, in older facilities the refrigerants used are not greenhouse gases though they do affect stratospheric ozone. Thus, it is only as these are replaced that their leakage would become a greenhouse factor. Given these facts and that high potential greenhouse gases constitute only 2% of total US GHG emissions to begin with, no attempt was made to estimate USPS GHG emissions from building refrigerants.
Solid Waste Disposal In 1997, a survey was done of USPS solid waste disposal (USPS, 1997). The survey indicated that in 1996 USPS disposed of 525,000 tons of solid waste at landfills while 573,000 tons were recycled. The amount placed in landfills represented .0045% of US total landfill wastes for the year. And, according to a report released by the Paper Task Force (Paper Task Force, 1995) , recycled waste causes about 15% the GHG emissions of landfill waste. In the US in 1996, there were 11.4 million metric tons of methane released from landfills, which equates to 65.1 million tons of carbon equivalent (EPA, 1999). If the USPS was responsible for .0045 percent, then its landfill waste accounted for 293,000 tons, while its recycled waste accounted for another 44 thousand tons, yielding a total of 337 thousand metric tons carbon equivalent for that year.
No data are available for USPS solid waste disposal in other years. However, plausibly such solid waste is related to total mail volume, in part because undeliverable mail constitutes one part of such waste. Total mail volume rose by 5.6% between 1996 and 1998. Assuming a one-to-one relationship between USPS solid waste and total mail handled and also that the Postal Service recycled the same proportion of its solid waste in 1998 as in 1996, emissions related to solid waste also rose by 5.6%, to an estimated 356 thousand metric tons in that year.
The estimate is somewhat tenuous in that we are forced to make assumptions regarding the relationship between solid waste disposal and total mail and also regarding USPS waste recycling behavior. The 1997 Postal Service survey indicated that waste recycling had increased rapidly between 1994 and 1996, reaching 52% in the latter year, and it is possible that the percentage continued to rise after that. It also is possible that USPS solid waste disposal is not linearly related to mail volume. More data would be helpful in this area.
Another issue with these data is that present methane emissions from waste sites largely are produced by past wastes. As we have no data on cumulative USPS (or US) wastes, we are forced instead to use present waste disposal as a proxy. As USPS waste recycling only began a few years ago, it is possible that its cumulative contribution at waste sites is proportionately higher than its present contribution. However, as the waste disposal number represents less than 10% of USPS total GHG emissions for 1998, it does not seem worthwhile to investigate this matter further.
Direct Vehicle Emissions USPS operates a fleet of vehicles, most of which are used for delivery purposes. Data on their gasoline, diesel and natural gas consumption is available from USPS annually. Combined, these accounted for about 278 thousand metric tons of carbon equivalent in 1998.
According to EPA, vehicles also are responsible for the bulk of methane and nitrous oxide emissions in the United States. In 1997, total such emissions were 18.9 mmTCE. In 1998, the USPS vehicle fleet constituted approximately .1% of the nation’s vehicle fleet (200,000 of about 200 million vehicles). Attributing a proportionate amount to USPS and assuming the 1997 national number was duplicated in 1998 (a fair guess, as the 1996 national number was virtually identical to the 1997) gives a total of 19 thousand metric tons of carbon equivalent.
The direct estimate for gasoline, diesel and natural gas rests on a strong base as data is collected directly from USPS areas on vehicle fuel use. The methane and nitrous oxide number is reasonable so long as EPA’s estimate of total vehicle emissions of these gases itself is reasonable.
I also considered emissions of high potential GHGs from vehicle air conditioners. EIA estimates that vehicles leak refrigerants at a rate of 15-30% in total over five years while GM assumes an annual leakage rate of 10% (EIA 1999, p. 52). However, most USPS delivery vehicles do not have air conditioners, and they constitute the great bulk of the USPS vehicle fleet. Subtracting these, about 30,000 USPS administrative vehicles and trucks likely do have them. But as the entire US vehicle fleet produced only about 5 mmTCE from air conditioner leakage in 1998, the contribution of USPS vehicles is lost in the rounding and therefore is here ignored.
Indirect Fuel Consumption
Mail transport involves the combined use of airplanes, ships, trucks and trains, and in addition to its own delivery fleet USPS leases space from 3rd party operators of these various transport modes. Accounting for GHGs associated with such leasing thus is a necessary part of an overall USPS GHG emission estimate.
I chose to estimate these emissions by assuming that USPS payments for 3rd party transport as a proportion of total national payments for such transport accurately reflect USPS proportionate use of these modes. Thus, for example, USPS payments to railroads for freight carriage as a proportion of national payments for such carriage are assumed to reflect USPS proportionate use of railroad space and also of its energy consumption. GHGs associated with such proportionate energy consumption then are attributed to USPS.
Four categories of transport are included in the estimates. These are air, railroad, trucking and shipping. In the case of air transport, both commercial air and a fleet of planes dedicated to USPS are included. Data on USPS use of each of these means of transport were obtained from the Postal Service’s 1998 Comprehensive Statement (USPS, 1998). Data on total national shipments by air and train for 1998 were obtained from individuals at the Air Transport Association and the Association of American Railroads, while data on trucking and shipping were obtained from government publications (Dept. of Commerce, 1999 and Department of Transportation, 1999).
The USPS purchased 2.36 billion ton miles from the nation’s commercial airlines in 1998, which constituted 2.86% of the national total. The nation’s commercial airlines consumed 18.9 billion gallons of fuel in 1998, with USPS thus accounting for 561 million of these gallons. USPS also ran a dedicated fleet which carried 14% the tonnage of scheduled air. I assumed that the fuel consumed per ton of mail moved was the same for dedicated as for commercial air. Adding in the dedicated usage resulted in USPS accounting for a total of 617 million gallons of jet fuel for that year.
USPS paid $1.8 billion for trucking services in 1998, which was .85% of national trucking revenues. According to EIA data, trucking consumed 30.15 billion gallons of diesel fuel in 1998, so that accounting for USPS’ proportionate share results in 256 million gallons for that year.
USPS paid $274 million for rail shipments in 1998. In that year, Class 1 railroads realized $33.2 billion in revenues, 97% of that from freight. Thus, USPS constituted .8 percent of the total. US railroads in 1998 consumed 3.18 billion gallons of diesel fuel in 1998, and USPS’ share comes to 26 million gallons.
Finally, USPS paid $34.5 million for domestic and overseas shipping in 1998. US domestic and international shipping revenues were $26.4 billion in 1997, the last year for which data are available. Assuming 1998 US freight revenues were equal to those in 1997, USPS payments represented .0013% of the US total. Ship fuel consumption in 1998 was 9.4 billion gallons, so that the USPS proportionate share was just over 12 million gallons.
Converting all of these to metric tons of carbon equivalent, USPS purchases of 3rd party transport resulted in a little over 2 million tons in 1998. Of these, airline shipments accounted for about 2/3. Also, 3rd party transport accounted for about 62% of USPS total GHG emissions for the year.
The strength of these estimates is directly related to the methodology employed. I think it is reasonable to attribute emissions from 3rd party transport to USPS proportionate use of such means. However, I have not investigated whether USPS is charged in the same way that others are, and therefore whether its payments accurately reflect the space, weight and distance that its freight makes up. The best data in this category are those from the commercial airlines, which explicitly calculate revenue ton miles from USPS as well as from other shippers and from passenger travel. Luckily this category accounts for better than 50% of all leased 3rd party transport. Rail, trucking and shipping data are accurate so far as they go, but it is less obvious that USPS payments for these services accurately reflect ton miles utilized. Still, assuming USPS lets contracts competitively in these categories, the estimates probably are reasonable.
One other category was considered, namely methane and nitrous oxide emissions from indirect transport. According to EIA data, almost all such emissions in the transport sector come from gasoline usage. Since the above modes generally are powered by jet fuel, diesel or bunker fuel, no emissions of these gases were attributed to indirect transport use by USPS.
National Contribution and per Piece of Mail Estimates
Total US GHG emissions in 1998 were 1803 mmTCE. USPS total emissions were an estimated 4.01 mmTCE, or just a little over .2 percent.
Also in 1998, the USPS delivered 198 billion pieces of mail weighing about 24 billion lbs and resulting in revenues of about $58 billion (USPS 1998 Annual Report). Converting total USPS metric ton emissions to pounds and dividing by the mail totals results in the following per piece of mail factors:
- $1 of mail expenditure results in .15 lbs of carbon equivalent (.56 lbs of CO2) - 1 piece of mail results in .04 lbs of carbon equivalent (.16 lbs CO2) - 1 lb of mail results in .37 lbs of carbon equivalent (1.35 lbs of CO2) Future Data Improvement
In order to provide better estimates in future, collection of the following data might be considered:
- annual jet fuel use in the USPS dedicated fleet (possibly this is already available though I did not come across it) - annual tonnage of USPS solid waste disposal and recycling - annual refrigerant loadings at USPS facilities.
1) The Paper Task Force, “Paper Task Force Recommendations for Purchasing and Using Environmentally Preferable Paper,” Final Report, 1995. (The Paper Task Force was composed of business and environmental group representatives and was led by EDF.)
2) United States Department of Commerce, Bureau of the Census, 1998 Annual Transportation Survey Tables, Table 1, “Motor Freight Transportation and Warehousing Services.”
3) United States Department of Energy, Energy Information Administration, “Emissions of Greenhouse Gases in the United States 1998”, October 1999.
4) ------------------, “Monthly Energy Review,” Tables 7-1, 7-6, A-3, A-4, and A-5.
5) ------------------, “Sales of Kerosine and Fuel Oil,” Table 1, Sales of Distillate Fuel Oil by End Use in the US: 1994-1998.
6) United States Department of Transportation, Bureau of Transportation Statistics, “National Transportation Statistics 99,” Table 2-7, Passenger & Freight Transportation Expenditures.
7) United States Environmental Protection Agency, “Inventory of US Greenhouse Gas Emissions and Sinks: 1990-1997,” Tables A-11 and A-12.
8) United States Postal Service, 1998 Annual Report, “Operating Statistics,” p. 74.
9) ------------------, 1998 Comprehensive Statement on Postal Operations, “Transportation Statistics,” pp. 22-23.
10) ------------------, FY 1998 Annual Energy Management Data Report.
11) ------------------, “Results of the United States Postal Service FY96 Recycling Survey,” September, 1997.
ESTIMATED GREENHOUSE GAS EMISSIONS OF THE UNITED STATES POSTAL SERVICE FOR SELECTED YEARS, 1985 – 1998
Michael E. Canes Logistics Management Institute October 2000
FINDINGS AND CONCLUSIONS Total greenhouse gases (GHGs) of the United States Postal Service (USPS) increased between 1985 and 1998, the last year for which complete data are available. During that period, USPS annual GHGs increased from 2.966 million metric tons of carbon (mmtce) to 4.187 mmtce or by 41 percent. Between 1990 (the international base year) and 1998, USPS GHGs increased by 17 percent.
Over the 1985-1998 and 1990-1998 time periods, the level of activity of the Postal Service also increased. The total number of pieces of mail handled increased by a little over 40 percent between 1985 and 1998, and by 18 percent between 1990 and 1998. Real expenditures on mail (measured as revenues to the USPS in $1996) increased even more, by over 46 percent between 1985 and 1998 and by 29 percent between 1990 and 1998. In consequence, GHGs per piece of mail were roughly stable between 1985, 1990 and 1998 while GHGs per real dollar spent on mail declined between 1985 and 1998, and between 1990 and 1998.
GHG emissions as here analyzed refer to those GHGs which result from the overall operations of the US Postal Service. They include GHGs resulting from facility energy use, solid waste disposal and landscaping; from USPS-owned vehicles; from third party transporters of the U.S. mail; and from Postal Service employee travel.
Among these, the largest contribution comes from third party transport, which accounts for just under 60 percent. Of this 60 percent, air transport accounts for about two-thirds, with trucking the next largest contributor. Facilities account for just over 30 percent, vehicles for around 10 percent, and employee travel for around .5 percent.
There have been changes in the pattern of facility energy use over time. Between 1985 and 1998, the Postal Service has increased its use of electricity and natural gas and decreased its use of fuel oil. Overall energy use at facilities increased roughly in line with the total volume of mail delivered, but less than the increase in facility space.
Direct vehicle emissions rose over the period, but by less than the volume of mail. Also, the vehicle fleet and number of miles driven rose by more than fuel use, indicating an increase in the fuel efficiency of the vehicles used.
GHG emissions associated with third party transport increased by over 45 percent between 1985 and 1998. Over that period there were very substantial increases in the use of trucking and air transport, with a small increase in the use of rail and a slight decrease in the use of ship transport. These data indicate that the way in which the Postal Service delivers mail has changed over time. For example, the relative increase in the use of third party air and truck transport and relative decrease in the use of rail and ship suggest that the USPS has endeavored to transport mail more quickly. The increased use of power at facilities is in part due to the increased use of processing machinery that sorts mail mechanically. This too is intended to deliver mail more quickly. Thus, more energy is consumed but plausibly more value is added for users of the service, increasing the amount they are willing to spend.
Data also indicate that the Postal Service has grown, adding employees, vehicles and facility space. However, energy use per gross square foot has decreased in postal facilities and fuel use has risen more slowly than the number of vehicles, indicating the Postal Service has become more energy efficient. These factors help to explain why, though total GHG emissions have risen, per piece of mail emissions have not and per dollar spent on mail emissions have fallen.
Beginning with the “Earth Summit” in Rio De Janeiro in 1992, international attention has focused on the possibility that gases emitted by the consumption of fossil fuels and by other human activities will affect the long-range climate of the planet. This led to international agreement in Berlin in 1995 to negotiate a post-2000 means to control such gases, agreement in Geneva in 1996 to set binding limits, and a treaty negotiated in Kyoto in 1997 to set target quantities for the industrialized countries to meet by 2008-2012.
The gases included in the Kyoto treaty include carbon dioxide, methane, nitrous oxide, hydroflourocarbons, perflourocarbons, and sulfur hexaflouride. Of these, carbon dioxide comprises well over 80 percent, methane about 10 percent, nitrous oxide about seven percent, and the others combined, two to three percent. The target quantities agreed to in Kyoto are set relative to levels reached in 1990 for carbon dioxide, methane and nitrous oxide, and relative to 1995 for the other gases.
Under United Nations auspices, an Intergovernmental Panel on Climate Change (IPCC) has been established. The Panel is comprised of scientists from around the world commissioned to study scientific aspects of the GHG issue, and to issue findings meant to inform the policy process. The Panel also has devoted effort to establishing protocols for countries to use in measuring their annual emissions of GHGs.
In the United States, actions to deal with GHGs have focused mainly on measurement and on voluntary energy conservation programs. Both EPA and the Energy Information Administration of DOE publish annual estimates of the GHG emissions of the United States. The federal government also has established various energy conservation goals for its own activities and has promulgated certain regulations (e.g., CAFE standards) intended to conserve fuels utilized in the private sector.
To date, there has been relatively little attempt by private firms to measure their GHGs. However, the World Resources Institute (WRI) has initiated a project to define common protocols for corporate GHG reporting, and some corporations already are reporting these (e.g., BP-Amoco and Royal Dutch Shell).
The present study both reports quantities of GHGs resulting from the activities of the United States Postal Service and develops a methodology for their measurement. In most respects, this methodology follows guidelines that WRI is establishing for corporate GHG reporting.
The study also provides a “per piece of mail” GHG estimate. Such an estimate theoretically could be used by other organizations wishing to estimate the GHG emissions that their own activities generate. For this purpose, the per piece of mail estimate is given in two ways, both per piece and per (real) dollar expended on mail. The latter better takes account of the relative size, weight, distance shipped and type of delivery service used. It also may be more useful for GHG estimation purposes if firms maintain records of their spending on mail but not on the number of pieces of mail they ship.
The methodology used to estimate USPS GHGs is explained in full in Appendix A. However, a brief description will help provide a rationale for the reported numbers. Direct energy use in Postal facilities and vehicles was converted to carbon equivalents through the use of standard conversion factors taken from DOE sources. USPS electricity use was broken out by fossil fuel source using national data, and these fuels then converted to carbon equivalents. Solid waste disposal was estimated for one year from survey data and extrapolated to others, with national GHG factors for solid waste disposal then applied to USPS quantities. Landscaping GHGs were estimated from data on pervious acreage at Postal Service facilities and assumptions concerning the fuel consumed in mowing such acreage. GHGs from third party transport were estimated by calculating proportions of freight that mail constituted, and applying these against total fuel consumed. Thus, for example, revenue ton miles of mail flown as a proportion of total revenue ton miles flown by commercial airlines was applied against the airlines’ total annual fuel use to approximate the share of fuel and hence of fuel-related GHGs attributable to the Postal Service. Postal Service employee business travel data were obtained for one year and extrapolated to other years to estimate GHGs attributable to USPS from this source. Also, GHGs from USPS employee commuting were estimated, but were not included in the totals on grounds they likely would have occurred regardless of Postal Service activity. Finally, some categories of GHGs were excluded, mostly because they were too small to matter.
The next sections break out the overall results into four parts: facilities, direct vehicle use, third party mail transport and employee travel. In each case results are presented and analysis given. The methodologies used to make the estimates are presented in the Appendix, and data sources are given in a Bibliography.
USPS GHG EMISSIONS FROM FACILITIES
USPS facilities include main post offices and branches, vehicle maintenance facilities, processing and distribution centers, headquarters buildings, and others. GHGs from these facilities increased by 42 percent between 1985 and 1998 and by almost 20 percent between 1990 and 1998. In carbon terms, this amounts to an increase of 284 thousand metric tons between 1985 and 1998, and of 220 thousand metric tons between 1990 and 1998.
The great bulk of the increase resulted from increased power use. This accounted for 249 thousand incremental tons between 1985 and 1998, and for 149 thousand incremental tons between 1990 and 1998. Emissions from fuel oil decreased over the entire period, while emissions from natural gas increased, as did emissions from liquid propane gas (LPG).
Emissions from solid waste disposal also are attributed to facilities. These increased from 282 to 369 thousand metric tons between 1985 and 1998, but were roughly constant between 1990 and 1998 due to an active recycling program begun by the USPS in the mid-1990’s that more than compensated for an increased volume of waste.
A small amount of emissions result from landscaping, which principally consists of lawnmowing. These decreased slightly over the period as the total number of USPS facilities fell.
Each year’s total is increased by 4.3 percent to account for facilities which USPS leases but for which it does not pay utilities. This increases the annual totals but does not affect year-to-year changes.
The GHG results for facilities reflect their growing use of energy, particularly in the form of electricity. From 1985 to 1998 facility use of electricity grew by 57 percent, and from 1990 to 1998 by 29 percent. GHGs from facility use of power grew at a somewhat slower pace because the proportion of coal used by power generators fell between 1985 and 1999 while the proportion of natural gas rose. Coal use, for example, fell from about 57 percent in 1985 to 53 percent in 1990 and 52 percent in 1998, while gas use rose from 11.8 percent in 1985 to 12.5 percent in 1990 and 14.7 percent in 1998.
Facilities also steadily reduced their direct use of fuel oil after 1985, while generally increasing their use of natural gas. From 1990 to 1998 in particular, gas use more than doubled while fuel oil use about halved. This resulted in a net reduction in GHGs as the burning of gas produces less carbon dioxide per btu than does oil.
The LPG category rose considerably over the period but this is an artifact of the data. In the earlier years LPG was broken out separately from coal and purchased steam but in 1995 and 1998 all of these fuels are lumped together and LPG is taken to represent the category. However, as LPG is a relatively small contributor to USPS GHGs, changes in the composition of this category had little effect on the overall results.
Measured in total btus, USPS facility energy consumption rose by 52 percent between 1985 and 1998, and by 42 percent between 1990 and 1998. This occurred even though the number of facilities did not increase between 1985 and 1998, nor between 1990 and 1998. In fact, the total number of USPS facilities decreased by around 5 percent between 1985 and 1998, from over 40,000 to around 38,000. However, the square footage of these facilities has risen substantially.
In 1985, the total space of US Postal Service facilities was around 189.4 million gross square feet. By 1990 this had risen to 219.3 million and by 1998 to 323.2 million. These numbers indicate facility space growth of 71 percent between 1985 and 1998 and 47 percent between 1990 and 1998. This growth of square footage appears to be the principal reason why facility energy use has grown, and therefore why facility-generated GHGs have risen. On the other hand, USPS has invested considerable effort and monies in facility energy reduction, and GHGs on a per square foot basis have fallen, by 16.6 percent between 1985 and 1998 and by 18.8 percent between 1990 and 1998.
USPS GHG EMISSIONS FROM DIRECT VEHICLE USE
The numbers show a small decrease in gasoline and a large increase in diesel GHGs between 1985 and 1998, and a small increase in gasoline and not quite so large an increase in diesel GHGs from 1990 to 1998. Unfortunately, reported USPS annual gasoline and diesel usage numbers fluctuate considerably from year to year, so that numbers attributed to each individually are less useful than the total of the two. This combined number rose by 12.6 percent between 1985 and 1998, and by 10 percent between 1990 and 1998. Motor vehicles also emit other greenhouse gases, specifically methane and nitrous oxide, and annual Postal Service fleet. Overall, GHG emissions from Postal vehicles rose almost 15 percent between 1985 and 1998 and about 11 percent between 1990 and 1998.
The increase in GHGs from direct vehicle use is the result of an increased number of vehicles traveling an increased number of miles.
From these numbers, the USPS-owned fleet has expanded more or less in line with the total volume of mail, and the number of miles driven has risen proportionately even more. The rural fleet also has expanded in number, though its number of miles driven has not much changed over time, possibly because areas that once were rural have become suburban or exurban, and hence are now served by USPS vehicles and not by rural carriers.
The increase in gasoline and diesel consumption and therefore of GHGs from direct vehicle use is less than the expansion of the USPS fleet or its miles driven. Two possible causes for this are: an increase in the fuel efficiency of the Postal Service and rural fleets, and the substitution of vehicles using alternative fuels for those using gasoline or diesel. Over this time period there have not been large numbers of vehicles using alternative fuels in the USPS fleet and probably few if any in the rural fleet. Thus, increases in fuel efficiency appear to explain the bulk of the differences between increases in fuel consumption, total vehicles and miles driven.
USPS GHG EMISSIONS FROM THIRD PARTY MAIL TRANSPORT
Air transport accounts for the great majority of USPS third party GHG transport emissions. Indeed, air transport is the single largest source of Postal Service GHG emissions overall. These emissions generally increase with time though they dip slightly from 1990 to 1991 and again from 1995 to 1998. The decrease from 1990 to 1991 is consistent with similar decreases for truck and rail, and is probably due to the fact that the total volume of mail decreased slightly between those two years.
Intercity trucking constitutes the next most important category of third party transport GHG emissions, and these have increased rapidly over time. Rail- related emissions have stayed fairly steady through time while ship-related emissions have fallen.
A possible explanation for the pattern of GHGs from third party transport is that there has been increased emphasis on moving mail quickly. Such increased emphasis on speed would motivate increased air transport, with less use of ship transport, which by its nature tends to be slow. It also could explain an increase in the use of trucking, with its flexibility of route, and a relative decrease in the use of rail.
In 1985, the USPS Comprehensive Statement only reported a figure for international ship expenditures. A domestic number was estimated from the rate of growth of domestic ship expenditures between 1990 and 1995, and this was added to the international number.
USPS use of air and truck transport has grown relative to the use of rail or ship. Also, 1998 USPS spending on air or trucking is well over a billion dollars per year whereas that for rail and ship is in the hundreds or tens of millions. These figures help to explain why GHGs from third party transport result mainly from air and truck travel, and why these two categories have grown in importance over time.
USPS GHG EMISSIONS FROM EMPLOYEE TRAVEL
Postal Service employees took about 145,000 air trips in 1999, and paid about $10 million for rental vehicles. As this was the only year for which there was data, the number of air trips and the amount paid annually for rental vehicles for other years was simply related to the number of postal employees in that year relative to 1999. The fact that Postal employee air travel has fallen over time as a proportion of all passenger travel, so that the proportion of total airline fuel-use GHGs allocated to the USPS has similarly fallen.
Results and Discussion
Some have argued that greenhouse gas emissions resulting from USPS employee commuting should be included in the overall totals. They reason that since employees must commute to work in order for the Postal Service to operate, the resulting emissions should be added to other sources in totaling all sources of USPS emissions.
It is, of course, true that most Postal Service employees commute to their jobs. However, such emissions are not the result of postal operations per se. Rather, they result from people choosing to work rather than stay at home. Thus, even if there were no Postal Service, most people employed by USPS would commute elsewhere, to other jobs. In consequence, few if any incremental emissions result from the fact that Postal Service employees commute to their jobs.
Even so, there is reason to include employee commuting GHGs in this report. First, doing so provides information on how large this source is. And second, it provides a basis for assessing what decreases might be achieved through public transit incentive programs or other trip reduction means.
GHG ESTIMATION METHODOLOGY
Annual data on energy usage by postal facilities was obtained from the Postal Service Finance Reporting (PSFR) system and the Facilities Management System database FMSWIN. These data are broken down into various categories: electricity, natural gas, fuel oil, LPG, and, in some years, coal and purchased steam. After 1991 coal and purchased steam no longer are provided separately, and instead they are combined with LPG in an ‘other’ category. As LPG has come to dominate this category, this fuel was taken to represent the entire category in 1995 and 1998. Before that, purchased steam was combined with electricity by converting their btus into megawatt hour (MWH) equivalents while coal was treated as a separate energy source.
Electricity. As the Postal Service is national in scope, it was assumed that electricity purchased comes from the national fuel mix used by utilities and independent power generators.
Given this assumption, electricity usage was converted into consumption of fossil fuels (coal, natural gas and residual fuel oil) , expressed in tons of coal, barrels of oil and thousand cubic feet of gas. These then were converted to millions of tons of carbon equivalent through the use of standard conversion factors taken from DOE sources.
Sample Calculation: In 1998, USPS consumed 4.41 million MWH of electricity. Of this, 51.7 percent or 2.3 million MWH is assumed to come from coal. In the US in 1998, 967 million tons of coal were used to produce 1874 million MWH of electricity. Using that relationship, 1.178 million tons of coal can be attributed to USPS consumption of electricity.
1.178 million tons x 20.5 mmbtus/ton = 24.1 trillion btus, or .0241 quads; then .0241 quads x 25.74 mmtce/quad = .621 mmtce
Other fuels. Direct facility use of natural gas, fuel oil (distillate) , LPG and coal was converted into million metric tons of carbon equivalent (mmtce) through the use of the same conversion factors.
Solid Waste Disposal. Solid waste disposal was included in facility GHG generation. Estimates of total USPS solid waste and of waste recycled for 1995 and 1996 were taken from Greening the Mail, a study conducted under the sponsorship of USPS by the National Task Force on Greening the Mail. In 1996, USPS generated about 1.1 million tons of waste, recycling 52 percent. Unfortunately, 1995 and 1996 were the only years for which data on USPS solid waste disposal and recycling could be obtained. For other years, it was assumed that the ratio of waste generated (in tons) to total pieces of mail in 1996 could be applied. For 1998 it was assumed that 52 percent of generated waste was recycled while for years before 1995 it was assumed that USPS recycled only 10 percent of its waste, below the national average but consistent with language in Greening the Mail indicating that the USPS had only recently begun active recycling programs.
Estimated USPS solid waste disposal at landfills then was divided by national waste disposal and the proportion in any given year applied to total GHG emissions from landfills for that year. Recycled solid waste was assumed to generate 15 percent of the GHG emissions of landfill waste, based on results reported in The Paper Task Force study, a publication of EDF. There is in fact a small difference in recycling emissions depending on how waste is recycled (burning for use as an energy source v. recycling for material reuse) but data were not available as to how USPS wastes were recycled, and the difference is too small to matter for purposes of this study. Finally, note that this way of estimating emissions from solid waste disposal uses annual landfill & recycled solid waste as a proxy for past waste, since it is biodegradable past waste that is producing GHG emissions now.
Facilities at which no utilities are paid. Account was taken of USPS facilities which are rented and for which USPS does not pay utilities directly (e.g., in some shopping centers). According to data obtained from FMSWIN, in early 2000 these accounted for 4.3 percent of the floor space of the USPS facilities for which utilities are paid directly. As this was the only data available regarding the magnitude of this set of facilities, the same percentage was used for all years. Thus, taking account of such facilities raises the total amount of GHGs in any given year but does not change the rate of growth over time.
Landscaping. An estimate of emissions from the use of mowing equipment to landscape postal facilities was included in the analysis. Rohr, et al (1999) conducted a study of the Chesapeake Bay Watershed region in which they estimated the amount of pervious acreage at retail and major postal facilities. Their sample consisted of 2128 facilities and showed that about 26% of facility acreage is pervious and 76% impervious. These estimates were extrapolated in the present study to the national populations of postal facilities and it was assumed that lawn comprised 80% of pervious acreage for retail facilities and 50% for major facilities (which have much larger per unit acreage). From discussions with retail lawn mower salesmen, it was estimated that a power mower cuts 1 acre of lawn per gallon while a riding mower cuts ½ acre per gallon. Power mowers were assumed used at retail facilities and riding mowers at major facilities. Finally, a 30 week growing season was assumed. Because the total number of postal facilities declined slightly between 1985 and 1998, greenhouse gas emissions from this source also declined slightly over that period. In some areas electric lawnmowers and other means to reduce mowing are being introduced, and these measures may further reduce landscaping GHGs in future.
Direct Vehicle Use
USPS-Owned Vehicles. Data on gallonage used by USPS vehicles was obtained from the USPS Fuel and Oil Usage Report, which shows total gallons of gasoline and diesel used in any given year. The data for gasoline and diesel usage taken alone behave strangely, showing increases and decreases that seem implausible, but the two taken together behave reasonably.
USPS Rural Fleet. USPS also contracts with private vehicle owners to deliver mail in rural areas. Annual data was obtained from USPS Delivery Policies and Programs on the number and average distance of rural routes, yielding total miles driven per year by this fleet. It was assumed that the makeup of the rural fleet reflects that of the US vehicle fleet, so that average mile per gallon data for the US fleet reported by DOE could be used to convert rural fleet distance to total gallons of gasoline.
Leased vehicles not counted. Through GSA, USPS leased a few thousand vehicles in 1998 and 1999. However, while survey data for such vehicles is available for those years, there were no data for earlier years. Because of this and because a few thousand vehicles do not much affect the overall results, these vehicles were not included in the totals.
Methane and Nitrous Oxide. Finally, motor vehicles produce not only carbon dioxide but also methane and nitrous oxide, which are greenhouse gases. To take account of these in any given year, the number of vehicles in the USPS vehicle fleet, including the rural fleet, was divided by the total US vehicle fleet, and the resulting proportion multiplied by EPA’s estimated total methane and nitrous oxide emissions from motor vehicles in that year. In round numbers, the USPS fleet, including the private vehicle fleet, constitutes a little over .1% of the entire US vehicle fleet and so that percentage of methane and nitrous oxide emissions from vehicles was attributed to USPS vehicle use.
Sample calculation: In 1998 there were 210,400 USPS owned vehicles plus another 51,200 vehicles used to deliver rural mail. The total US vehicle fleet numbered about 210 million. Methane and nitrous oxide emissions from mobile sources were about 18.9 million mmtce in that year. Accordingly, .125 percent of 18.9 million, or .024 mmtce were attributed to USPS fleet activity in that year.
Third Party Mail Transport
USPS contracts with various private transporters to move mail. These are of four basic types; air, truck, rail and ship. There also is a dedicated air fleet which delivers mail nightly and in additional quantities during the Christmas holiday season. These activities generate GHGs, and therefore it is appropriate to attribute some portion of them to the USPS.
Conceptually, the ideal measure would be the GHGs from incremental flights or ground transport resulting from mail delivery activity. For the USPS dedicated air fleet, clearly all GHGs can be so attributed. However, for commercial air, truck, rail and ship, only a small portion is attributable. The device used for truck, rail and ship to estimate this portion was USPS payments to each mode of transport taken as a proportion of their total revenues from freight movement. For rail and ship, there is a small amount of passenger travel but the bulk of payments to them are for freight movement, and passenger movement was ignored. The proportion of USPS payments for freight movement to total revenues in any given year then was applied against fuel use by the specific mode of travel. For rail and trucks this was taken to be diesel fuel, while for ships it was taken to be residual fuel oil. The resulting gallons then were converted into mmtce.
Sample calculation: In 1998 USPS spent $1.8 billion on third party truck transport, while total US trucking freight revenues were about $212 billion. Also in that year, trucks consumed 30.15 billion gallons of diesel fuel. Taking $1.8 as a proportion of $212 billion and multiplying by 30.15 yields 256 million gallons attributable to USPS for that year. At 42 gallons per barrel, this translates to about 6.1 million barrels of diesel fuel, which then is converted to .704 mmtce through use of conversion factors.
The methodology assumes that transporters such as rail and trucking companies anticipate that a certain proportion of their space will be used for moving mail, and plan their fleets accordingly. This implies that if USPS did not ship mail, these transporters would have smaller fleets that used less fuel. In the long run, therefore, a portion of their fuel use is allocable to USPS, and USPS dollars spent as a proportion of total revenues for a given mode of freight movement is used as the best available proxy measure for this portion.
For air transport, a similar approach was used but somewhat more precise data were available. The Airline Transport Association annually publishes figures for revenue ton miles of the nation’s commercial airlines. The figures break out passengers from freight, and within freight mail is broken out as a separate category. This then allows direct calculation of mail as a proportion of total revenue ton miles flown. As passengers dominate airline movement of tonnage from place to another, they are included in the calculation. The proportion of mail to total revenue ton miles then is applied against scheduled air lines’ consumption of jet fuel for each year, and the resulting gallons converted into mmtce.
Data for the USPS dedicated fleet are sparse before 1995, but become better over time. Total tonnage of mail flown is given in the annual USPS Comprehensive Statement of Operations. To estimate GHGs from this source, it was assumed that fuel used per ton of mail flown is the same for the dedicated fleet as for commercial airlines.
Postal Employee Business Travel
In any given year, Postal Service employees engage in business travel in which they utilize the services of air and ground transport. As GHGs resulting from this travel are attributable to USPS activity, some accounting for their magnitude is appropriate.
Data was available on the number of USPS employee airplane trips taken in 1999 and on expenditures on rental cars in that year. Passenger airplane trips were converted into revenue ton miles using factors made available by the Air Transport Association. These then were taken as a proportion of total revenue ton miles flown by US commercial airlines, and the proportion applied against their total fuel consumption.
For rental car expenditures, it was assumed that rental companies in 1999 were charging $.40/mile on average. This figure then was divided into total USPS expenditures on rental cars to derive total miles driven, and this in turn was converted into gallons using DOE figures on average mileage per gallon in the US fleet.
It also was assumed that any given air trip has an associated ground trip to and from the airport at the embarkation end. To avoid double counting, it was assumed that rental cars are used at the destination end. It was assumed that ground trips to and from the embarkation airport each require the consumption of ½ gallon of gasoline, so that any air trip by a postal employee is assumed to consume one gallon of gasoline at the embarkation end.
Unfortunately, Postal Service data on air trips and on car rental expenditures exist for 1999 only. It was assumed therefore that the number of trips and of car rentals is directly related to the total number of permanent postal employees. Using that assumption, estimates of fuel consumption and resulting carbon emissions were made for the other years covered in this study.
As explained in the text, USPS employee commuting GHGs were estimated but not included in the overall totals. Employees were separated into two categories, full-time and others, with the others assumed to substitute for the full-time one out of six working days per week and during vacations. Rural and city carriers were assumed to work a six-day week, with 80% of Mail Handlers, Clerks and Supervisors also working six days. Employee travel was assumed to occur 50 weeks per year to account for holidays. Rural subs, PM Relief/Leave Replacements and Transitional Employees were treated as others, assumed to substitute for the career employees, and hence these employees were not included to avoid double-counting. Emissions were estimated using U.S. average commute distances (8.5 miles for 1983-1989, 10.6 miles for 1990-1994, and 11.6 miles for 1995-1999, as well as average US fleet economy for each year analyzed, and the US average number of commuting individuals per vehicle (1.14).
Sources of GHGs Excluded from this Study
Several other possible sources of GHGs were examined but dropped from consideration in this study. A small quantity of compressed natural gas is consumed by the postal fleet, but the amount is too small at present to affect the overall results. Postal facilities generally have chillers and some leakage of HFCs could be occurring from these. However, the leakage rate is low for facility chillers, and most existing facilities probably still have freon, which is not a greenhouse gas. Also, the USPS purchases and holds land in anticipation of its expansion needs, but information obtained from the office of the Facilities Environmental Program Coordinator indicates the amounts are small and generally do not involve forestry, animal husbandry or other activities that could generate or absorb significant quantities of greenhouse gases.
Also, as mentioned earlier, energy used in commuting by Postal Service employees was not included in the study on grounds that most if not all such employees would have commuted to jobs in any case. For similar reasons, energy used by Postal Service facilities to dispose of liquid waste was excluded from consideration in the study.
1) The National Task Force on Greening the Mail, Greening the Mail, Final Report, January 1999. (The National Task Force on Greening the Mail was composed of business and government representatives including EPA, USPS and the US Conference of Mayors.
2) The Paper Task Force, Paper Task Force Recommendations for Purchasing and Using Environmentally Preferable Paper, Final Report, 1995. (The Paper Task Force was composed of business and environmental group representatives and was led by EDF.)
3) Rohr, Erica S., Emil J. Dzuray, and Pamela L. Sinha. “Baseline of Postal Service Nonpoint Sources of Nutrients to the Chesapeake Bay,” Logistics Management Institute, PS904L1, February 1999.
4) United States Department of Commerce, Bureau of the Census, 1998 Annual Transportation Survey Tables, Table 1, “Motor Freight Transportation and Warehousing Services.”
5) United States Department of Energy, Energy Information Administration, Emissions of Greenhouse Gases in the United States 1998, October 1999.
6) ------------------, Monthly Energy Review, Table 7-2, “Electricity Net Generation,” and Table 7-6, “Consumption of Fossil Fuels to Generate Electricity”.
7) ------------------, Monthly Energy Review, Table A1, “Approximate Heat Content of Petroleum Products,” Table A3, “Approximate Heat Content of Petroleum Products, Weighted Averages,” Table A4, Approximate Heat Content of Natural Gas,” and Table A5, “Approximate Heat Content of Coal and Coal Coke.”
8) ------------------, Sales of Kerosine and Fuel Oil, Table 1, Sales of Distillate Fuel Oil by End Use in the US: 1994-1998.
9) United States Department of Transportation, Bureau of Transportation Statistics, National Transportation Statistics 99, Table 2-7, “Passenger & Freight Transportation Expenditures”.
10) United States Environmental Protection Agency, Inventory of US Greenhouse Gas Emissions and Sinks: 1990-1997, Table A-11, “Key Assumptions for Estimating Carbon Dioxide Emissions,” and Table A-12, “Annually Variable Carbon Content Coefficients by Year (mmtce/qbtu) ”.
11) United States Postal Service, 1998 Annual Report, “Operating Statistics,” p. 74.
12) ------------------, 1998 Comprehensive Statement on Postal Operations, “Transportation Statistics,” pp. 22-23.
13) ------------------, FY 1998 Annual Energy Management Data Report.
14) -------------------, Fuel and Oil Usage by Make/Model, National Summary, various years.
Like: Facebook, Twitter,
Weibo, Baidu Tieba,
Baidu Like, Baidu Cang, Reddit
Dates are assigned arbitrarily to sort the articles.