Methodology

Cost Estimates

All costs are adjusted to 2024 US dollars using the Bureau of Labor Statistics CPI-U inflation calculator. Original-year (nominal) costs are also provided where available.

Post-9/11 Conflicts (Brown University Methodology)

Cost estimates for post-9/11 conflicts draw primarily from the Brown University Costs of War Project, which provides the most comprehensive accounting of War on Terror expenditures. Their methodology captures costs that Pentagon figures miss:

• Direct war appropriations: OCO funding, supplemental appropriations, and war-specific budget lines
• DOD base budget increases: Permanent spending growth justified by but not limited to the wars
• State Department and USAID: Diplomatic, reconstruction, and stabilization costs
• Veterans Affairs: Current medical and disability costs, plus projected costs through 2050
• Homeland Security: DHS spending attributable to 9/11 and the War on Terror
• Interest on war borrowing: Because wars were debt-financed, interest compounds over decades

This methodology reveals that direct fighting costs are only ~36% of total war costs. The rest — veteran care, interest, DHS, base budget growth — represents the hidden iceberg. This is why the Brown figure ($8T+) for the War on Terror is much higher than Pentagon estimates (~$2T).

Historical Conflicts (Pre-2001)

For historical conflicts, we use Congressional Research Service reports, specifically:

• RL33110: “Costs of Major U.S. Wars” — CRS inflation-adjusted cost estimates
• OMB Historical Tables: Defense spending by year back to 1940
• Academic sources: For pre-1940 conflicts, we use peer-reviewed historical research

Historical cost estimates involve greater uncertainty, especially for conflicts before reliable economic data existed (pre-1860). We note this uncertainty in our data and use ranges where appropriate.

How Inflation Adjustment Works

We use the CPI-U (Consumer Price Index for All Urban Consumers) published by the Bureau of Labor Statistics. For each conflict:

1. Identify the year(s) of spending (or use midpoint year for multi-year conflicts)
2. Apply the CPI-U ratio to convert to 2024 dollars
3. For long conflicts spanning multiple years, we adjust each year's spending separately where data permits

Casualty Data

US military casualty figures use official Department of Defense records via the Defense Manpower Data Center (DMDC) and Congressional Research Service reports (RL32492).

Casualty Categories

• Battle deaths: Killed in direct combat. Most precisely documented. Source: DMDC.
• Total US military deaths: Includes battle deaths, disease, accidents, and other causes during the conflict period. For many historical wars, disease killed more soldiers than combat (e.g., Civil War: ~2/3 of deaths were from disease).
• Wounded: Official DoD figures where available. Significant undercounting is likely for TBI (only diagnosed since ~2000) and PTSD (which was not recognized until 1980).
• Civilian deaths (direct): Killed by violence — bombing, gunfire, crossfire. Sources vary by conflict: Iraq Body Count, Airwars, UN reports, Brown Costs of War.
• Civilian deaths (indirect): Died from war-related disease, displacement, infrastructure destruction, loss of healthcare access. These typically dwarf direct deaths but are estimated with wide uncertainty ranges.
• Enemy combatant deaths: Rarely well-documented. Military “body counts” are notoriously unreliable (Vietnam taught this lesson). We include where credible estimates exist.

Civilian Casualty Methodology

Civilian casualty estimates are inherently uncertain and typically represent conservative lower bounds. We use the following source hierarchy for civilian deaths:

1. Brown University Costs of War Project — Most comprehensive for post-9/11 conflicts
2. Iraq Body Count — Gold standard for documented civilian deaths in Iraq
3. Airwars — Best source for civilian harm from airstrikes globally
4. UN reports — For specific conflicts where UN monitors are present
5. Bureau of Investigative Journalism — For drone strike casualties specifically

When sources provide ranges, we generally use the midpoint or conservative estimate. We explicitly note when figures are estimates vs. documented counts.

Military Spending Data

Annual military spending data comes from multiple sources:

• SIPRI Military Expenditure Database: Global spending data since 1949, using constant-dollar methodology
• OMB Historical Tables: US federal defense spending back to 1940
• DoD Budget Justification Documents: Detailed annual budget requests and final appropriations
• CRS Reports: Analysis of specific spending categories and trends

Note: “Military spending” can be measured different ways. The DoD “base budget” ($886B for FY2024) doesn't include nuclear weapons (DOE), veterans' care (VA), homeland security, intelligence community, or interest on war debt. We distinguish between the DOD budget and the “true national security budget” ($1.4T+) throughout the site.

Overseas Presence Data

Military base counts and troop deployment figures come from:

• DoD Base Structure Report: Official annual inventory of installations
• DMDC Deployment Data: Troop numbers by country
• David Vine's research: (American University, author of Base Nation) — broader definition including “lily pad” bases

Exact base counts vary significantly by definition. The Pentagon's official count (~500) uses a narrow definition. Vine's count (~750-800) includes smaller installations, cooperative security locations, and forward operating sites. We use the broader count with explanation.

Foreign Aid Data

Foreign aid data from USAID Foreign Aid Explorer and CRS reports. We distinguish between military aid (Foreign Military Financing, weapons transfers, IMET) and economic/humanitarian aid.

Computed Fields

For each conflict, we calculate derived metrics to enable comparison:

Source Hierarchy

When sources conflict, we prioritize in this order:

1. Official government records (DoD, CRS, OMB) — Most authoritative for US military data
2. Peer-reviewed academic research (Brown Costs of War, SIPRI) — Most comprehensive analysis
3. Established investigative organizations (Iraq Body Count, Airwars, TBIJ) — Best for civilian harm data
4. Investigative journalism (major outlets with documented sourcing) — Supplementary only

Limitations & Caveats

We acknowledge significant limitations in our data:

• Covert operation costs are often classified and may be significantly understated. The CIA's budget is not publicly disclosed. Covert wars (e.g., CIA operations in Laos, Syria, Libya) are difficult to cost.
• Civilian casualty counts are conservative estimates — true numbers are likely higher. Many deaths go unrecorded, especially in remote areas, conflict zones where journalists cannot operate, and in the aftermath of displacement.
• Long-term costs (veteran care over decades, interest on war debt, environmental remediation) are not fully captured and may add trillions to final tallies. These costs are inherently projections.
• Historical cost conversions involve significant uncertainty, especially for conflicts before reliable economic data existed (pre-1860). CPI-U is imperfect for very long time horizons.
• Congressional authorization classifications may be debated. We classify based on whether a formal declaration or explicit AUMF was passed, but reasonable people may disagree on what constitutes “authorization.”
• “Indirect” deaths from war-related causes (disease, displacement, starvation) have wide uncertainty ranges. The Brown estimate of 3.8M indirect deaths from the War on Terror is a methodological estimate, not a count.
• Ongoing conflicts have incomplete cost and casualty data that will change as they continue.
• Opportunity cost calculations assume fungibility of spending that may not hold in practice. Money saved from defense would not automatically flow to alternatives.
• China's military spending is likely 1.5–2× the officially reported figure. Our global comparison tables use SIPRI estimates, which attempt to capture actual spending.

A Note on Bias

All data collection involves choices — what to include, what to exclude, how to categorize. We approach this work from a perspective that values transparency, constitutional governance, and honest accounting of the costs of war. We strive to use conservative estimates and acknowledge uncertainty.

We are not neutral on whether citizens should have access to this data — we believe they should. We are not neutral on whether the Pentagon should pass an audit — we believe it should. But we are committed to accuracy in the data itself.

If you believe any figure is incorrect, we welcome corrections with sourcing. Accuracy matters more than narrative.