Notes and Sources

Cost of National Security Counters

Last updated May 2022

The Cost of National Security counters examine the costs of the United States’ post-9/11 wars. These wars – in Afghanistan, Iraq, and expanding to Syria and beyond – were the longest wars in United States history. While the troops have come home, their costs have not ended. The counters show their cost in dollars, according to the Costs of War project at Brown University’s Watson Institute for International and Public Affairs.

All counters present non-inflation-adjusted dollars and budgeted amounts through the end of fiscal year 2022.

More About the Cost of War Counters

The war cost counters include an estimate of appropriated spending on the war on terror since September 11, 2001 through the end of the 2022 fiscal year. The last troops came home from Afghanistan in September 2021, but costs of the withdrawal, interest on debt, and care for veterans continue. Some of these costs will continue for decades.

Aside from the total costs of these wars, we present four components of this spending:

• Military Costs of War: These include war funding appropriated as Overseas Contingency Operations for both the Departments of Defense and State, as well as a portion of the base Department of Defense budget estimated to represent increased military spending due to ongoing wars. 
• Homeland Security Costs of War: Through FY 2017, these represent the spending under specific Homeland Security missions associated with counterterrorism, related to preventing and disrupting terror attacks, responding and recovering from terrorist incidents, and protecting Americans and infrastructure. 
• Veterans’ Care: Estimated and/or reported costs for medical care and other services and supports for veterans of the 9/11 wars, including the accrual of future costs through 2050.
• Interest on war debt: Estimate of interest on debt taken out for war spending.

For more information about any of these items, please visit Costs of War.

Local and State Estimates

To estimate how much each locality pays for the costs of national security, NPP assigns each state, county, city, and congressional district in the U.S. a percentage that represents the area's share of the total U.S. income tax bill.

States

To assign the state percentages, we use state-level data from the IRS’s Individual Master File System. Specifically, Historic Table 2: Individual Income and Tax Data, by State and Size of Adjusted Gross Income, Tax Year 2012: http://www.irs.gov/uac/SOI-Tax-Stats-Historic-Table-2 ^[i].

To get the percentage for each state, we divide its Income tax amount by the US Income tax amount ^[ii].

Zip Code Tabulation Area (ZCTA)

Zip code tabulation areas (ZCTAs) are the Census Bureau's version of postal service zip codes. The basis for all sub-state multipliers is ZCTA rather than the USPS zip code, because ZCTAs can be allocated to other geographic areas of interest, such as congressional districts and counties.

To determine the percentage of total federal individual income tax contributed by each ZCTA, we start with Individual Income Tax ZIP Code Data compiled by the IRS: http://www.irs.gov/uac/SOI-Tax-Stats-Individual-Income-Tax-Statistics-ZIP-Code-Data-(SOI) ^[iii]. We use tax year 2012 numbers.

To get the percentage of total taxes paid for each ZCTA, we do the following:

1. Convert the U.S. Postal Service zip codes on the IRS report to their corresponding ZCTAs using this crosswalk file.
2. Sum column A06500 (Income Tax Paid After Credits) of the IRS report by ZCTA ^[ii].
3. For each ZCTA, divide the amount paid (from step 2) by the total U.S. income taxes ^[iv].

Counties, Places, and Congressional Districts

The Individual Income Tax ZIP Code Data is also the basis for the county, place, and congressional district taxes paid. To get those numbers, we follow steps 1 and 2 above. Then we match the ZCTAs with county, place, and 113^th congressional district boundaries using the Missouri Census Data Center's MABLE Geographic Correspondence Engine with Census 2010 Geography .

MABLE maps each ZCTA to a county or counties. Separate downloads do the same thing for places and congressional districts. When a ZCTA covers more than one county, place, or congressional district, MABLE provides an allocation to describe how much of the ZCTA belongs to each. This allocation is Census 2010 population-based.

For example, consider ZCTA 19317 in Pennsylvania. According to MABLE, 45% of the 19317 population live in Chester County, and 55% live in Delaware. Therefore, when determining the income taxes paid for each county, 45% of the 19317 amount will be assigned to Chester, and 55% will be assigned to Delaware.

Once we have the allocations from MABLE, we do the following calculations three times—once each for counties, places, and congressional districts.

1. For each ZCTA, multiply the taxes paid (from the ZCTA-summarized IRS data) by the allocation percentage for its corresponding counties/places/congressional districts.
2. Summarize the results by county/places/congressional district to get the total income taxes paid for that area.
3. Divide each area's total taxes by the total U.S. income taxes. This number is the sum of column A06500 in the original IRS taxes by zip code file.

FOOTNOTES

ⁱ This site uses income tax numbers to approximate each locality's contribution to the designated programs, which are  funded by federal funds.

ⁱⁱ Income tax amount reflects income taxes owed after the deduction of non-refundable credits. The amount has not been reflected to adjust for the earned income credit or for refundable credits such as the health coverage credit, and additional child tax credit. A complete list of these refundable credits is on the payments section of the 1040.

ⁱⁱⁱ To protect individual identities, some returns and zip codes are excluded from IRS zip code-level reports. See the IRS documentation for details (MS Word doc).

^iv For the few ZCTAs that contain two states, we assign the ZCTA to the state that contains most of its population. This assignment doesn’t change the trade-offs calculations—if affects only how the user would search for that ZCTA/”city”.