# The Military Working Dog Program: The Mission Worked, the Kennels Did Not A DoD Inspector General review found four military working dogs died in aging Air Force kennels from FY2021 to FY2023, inside a program costing up to $85,000 per dog. Author: J.A. Watte Published: July 18, 2026 Source: https://jwatte.com/blog/military-working-dog-program/ --- Walk the kennel rows at Joint Base San Antonio Lackland in Texas and you are looking at the front end of nearly every bomb dog and patrol dog the United States military owns. This is where the animals are bred, tested, matched to handlers, and pushed through detection courses, and it is where they come back to when a war ends. It is also, according to the Department of Defense's own Inspector General, a place where dogs have been kept in kennels built more than forty years ago, some with mold and standing waste and no real protection from Texas heat. The program is the DoD Military Working Dog (MWD) Program, run for the whole department by the Air Force, and its watchdog file now carries a hard number: four working dogs dead from inadequate housing conditions between fiscal year 2021 and fiscal year 2023. Set against that is a mission almost nobody disputes, dogs that find explosives no sensor reliably matches and that handlers credit with keeping people alive. Both of those things are true at once, and that is the whole story. ## What the program is and why it exists The Military Working Dog Program is the single Defense-wide pipeline that breeds, buys, trains, assigns, and eventually retires every military working dog used by all the services. Instead of the Army, Navy, Air Force, and Marine Corps each running their own separate breeding and training operations, the department consolidates the whole thing so that a bomb-detection dog in the Army and a patrol dog in the Air Force come off a common line, meet a common standard, and flow back through a common adoption channel when their working days are done. The animals do real work. A trained detection dog can locate explosives, weapons, and narcotics that fixed sensors miss, and it does it fast, in places and conditions where machines struggle. Patrol dogs add a deterrent and a search capability that a person alone cannot provide. That is why the capability has survived decades of budget pressure while flashier programs got cut: it is cheap by defense standards, it is hard to replicate with hardware, and the people who deploy alongside these dogs will tell you the payoff is measured in lives. Because the whole enterprise runs under one roof, it also lives or dies on the quality of that roof. Centralization is efficient when the central manager is doing its job. It concentrates the failure when the central manager is not. The 2026 Inspector General findings are, at bottom, a report about what happens when a consolidated program lets its physical plant and its staffing fall behind the mission it is still being asked to perform. ## The mechanism: one Executive Agent, one line at Lackland Under Department of Defense rules, the Secretary of the Air Force is the DoD Executive Agent for the Military Working Dog Program. That designation is recorded in the DoD Executive Agent registry, which lists the Air Force as the responsible component, dates the Executive Agent function to September 7, 1983, and names the governing issuance as DoD Directive 5200.31E, DoD Military Working Dog (MWD) Program, reissued August 10, 2011. In plain terms, one service is legally on the hook for resourcing, training, utilization, and final disposition of the dogs for everybody. Inside the Air Force, the responsibility runs down to the Director of Security Forces at Headquarters Air Force, the office known as AF/A4S. The physical center of gravity is the 341st Training Squadron at Lackland. Nearly all dogs and handlers move through there. That single point of control is exactly what makes the program efficient on paper and exactly what makes a shortfall at Lackland show up everywhere else. One worth noting bookkeeping detail: the Executive Agent registry page that carries this designation now displays Department of War branding, reflecting the 2025 rename of the department, and was last updated in March 2026. The directive number, the 2011 issuance date, the Air Force as the responsible component, and the 1983 origin of the Executive Agent role are all unchanged. In this article I use Department of Defense and DoD for historical accuracy, because that is the name attached to the directive and to the fiscal years in question. ## The money Start with scale. As of February 15, 2022, the Government Accountability Office counted about 5,100 working dogs across federally managed programs, and the Defense Department held nearly 1,800 of them (1,769 in the GAO's tally). That made DoD the second-largest holder in the federal government, behind the Department of Homeland Security at 2,942. A precision note worth keeping straight: the single largest individual program is the Transportation Security Administration at 1,097 dogs, so DoD is the largest holder across its services rather than the operator of the biggest single kennel. The roughly 5,100 count also covers only federally managed dogs, so the true universe, once contractor-managed and State Department partner dogs are added, is somewhat larger. Now the per-dog cost, which is where reporting often gets loose. The GAO states directly, on page 6 of its report and attributed to agency officials, that procuring and training a dog can cost approximately $65,000 to $85,000 per dog. That is an estimated acquisition-and-training figure, not a contract ceiling and not a single obligated number. A fully trained specialized detection dog can be valued well past $150,000 once you account for the full course and the capability. And a commonly quoted $6,500 press figure is misleading if you read it as the price of a finished dog: it covers only one dog's basic course of roughly ninety days, not a specialized detection animal. These are program operating and life-cycle costs. Nobody writes a single check labeled Military Working Dog Program, so the honest way to hold the money is as a per-animal estimate multiplied across a population of nearly 1,800, plus the kennels, veterinary care, staffing, and retirement that the 2026 findings say were being shortchanged. ## The watchdog critique Two oversight files carry the weight here, and they point in the same direction. The first is the GAO's 2022 report, GAO-23-104489, Working Dogs: Federal Agencies Need to Better Address Health and Welfare. Looking across the government, the GAO found that about half of federally managed dog programs and about half of contractor programs did not even address abuse and neglect, rest limits, or retirement and euthanasia decisions in their policies or contracts. It issued 19 recommendations. That is a governance gap, the kind of finding that says the rules on paper were thin before you ever get to conditions on the ground. The second file is worse, because it is specifically about the program the Air Force runs. In DODIG-2026-057, released in February 2026, the Department of Defense Inspector General found neglect inside the MWD program itself. At 10 of 12 sites the evaluators visited, the kennel facilities were aging and unsatisfactory, many built more than forty years ago, with mold, standing waste, and inadequate protection from heat. Four dogs died from inadequate kennel and housing conditions between fiscal year 2021 and fiscal year 2023. Reporting on the same review found that at Lackland roughly 22 percent of the 520 dogs there were infected with an intestinal disease. Staffing was thin enough that dogs were getting something like ten-minute walks four times a week or less, against a standard that called for roughly five hours a day of activity and enrichment. I want to be careful with two details here. The primary redacted report is hosted on a Defense Department server that blocks automated access, so the four-death figure and the site count are drawn from two independent outlets each citing DODIG-2026-057, and they are consistent. I am not attaching a specific medical cause to the deaths, because the public reporting attributes them to the housing conditions rather than to a named illness, and I could not confirm a specific cause from a primary source. There is a longer pattern behind these two reports. In 2018, a separate Inspector General report, DODIG-2018-081, found that the Army mishandled the disposition of its Tactical Explosive Detection Dogs, so that former handlers who wanted to adopt the specific dogs they had deployed with could not get them back. That is a failure under Robby's Law, the statute (10 U.S.C. 2583) written precisely to make sure retired military dogs get placed into homes rather than lost in the system. When the same enterprise fumbles both the front end (kennels and care) and the back end (adoption and disposition), the problem is not one bad kennel. It is a management posture that treated the animals as an afterthought to the mission they were performing. ## The public-good defense Now the other side, stated as plainly as the criticism. The capability is real and hard to replace. There is no off-the-shelf sensor that matches a trained detection dog across the range of explosives and narcotics and field conditions the military faces, and the people who have deployed with these dogs are not sentimental when they say the dogs saved lives. That value does not evaporate because the kennels were bad. The consolidated structure, for all that it concentrates risk, also does real good. Putting breeding, training, and standards under one Executive Agent at Lackland gives every service a common pipeline instead of four uneven ones, and it gives retired dogs a single adoption channel. Robby's Law and the mandatory annual disposition report to Congress exist so that retired dogs get placed into homes, and the fact that the 2018 report could name the Army's failure by number is evidence the reporting requirement works. Oversight caught the kennel problem too. The Inspector General went out and visited sites, and the Air Force did not dispute the 2026 findings. It requested fiscal year 2025 funding to add caretakers and to meet the enrichment requirements the report said were being missed. That is the accountability loop closing rather than jamming: an independent watchdog documents a failure, the responsible service accepts it and asks for money to fix it, and the record is public. A program that hides its problems does not do that. ## Reading the ledger Hold both columns open and do not average them into a grade. On one side, a genuinely valuable, comparatively cheap capability, roughly 1,800 dogs that no machine reliably replaces, run through a common pipeline with a legal channel to send retirees home. On the other side, an Executive Agent that let the physical plant rot for forty years, staffing so thin the animals got a fraction of the activity their own standard required, an intestinal-disease outbreak at the flagship base, and four dogs dead from housing conditions inside the very years the department was still deploying their peers. The mission worked. The stewardship of the animals doing that mission did not, and it took an Inspector General walking the kennel rows to force it into daylight. The right reading is not that the Military Working Dog Program is a boondoggle. It is that a good and necessary program was run cheaply in the wrong place, on the backs of the dogs, until the watchdog made the bill visible and the Air Force finally went looking for the money to pay it. ## Related reading - [Why the improper-payments and high-risk list matters](/blog/gao-high-risk-list-improper-payments/): the same GAO oversight machinery that flags federal waste is what documented the working-dog welfare gaps. - [Why the Pentagon has never passed an audit](/blog/pentagon-never-passed-audit/): a department that cannot fully account for its money is also the one that let forty-year-old kennels slide. - [The index of programs, lifelines, and boondoggles](/blog/public-money-programs-index/): the full catalog this entry belongs to. - [The working ledgers](/blog/the-working-ledgers/): the running record where spent, projected, and contested figures are kept side by side. ## Fact-check notes and sources - The Secretary of the Air Force is the DoD Executive Agent for the Military Working Dog Program, a role dated to September 7, 1983 and governed by DoD Directive 5200.31E (reissued August 10, 2011), with responsibility delegated inside the Air Force to the Director of Security Forces (AF/A4S). The registry page now carries Department of War branding after the 2025 rename, but the directive number, 2011 date, Air Force component, and 1983 origin are unchanged. [DoD Executive Agent Registry, entry 71](https://dod-executiveagent.osd.mil/Agents/ViewAgent.aspx?agentId=71); [DoD Directive 5200.31E](https://www.esd.whs.mil/Portals/54/Documents/DD/issuances/dodd/520031e.pdf). - The load-bearing figure: four DoD working dogs died from inadequate kennel and housing conditions between FY2021 and FY2023, with aging and unsatisfactory facilities at 10 of 12 sites visited and enrichment cut to roughly ten minutes four times a week against a standard of about five hours a day, per DODIG-2026-057. The primary redacted report is hosted on a Defense server that blocks automated access, so these figures are drawn from independent press coverage citing the report; no specific medical cause is attributed to the deaths in that coverage, so none is stated here. [Military Times reporting on DODIG-2026-057](https://www.militarytimes.com/news/your-military/2026/02/24/military-dogs-are-facing-substandard-conditions-leading-to-deaths-report-finds/); [DoD OIG report page, DODIG-2026-057](https://www.dodig.mil/reports.html/Article/4409688/evaluation-of-the-dod-military-working-dog-programs-management-of-canine-welfar/). - Reporting on the same review found roughly 22 percent of the 520 dogs at Lackland were infected with an intestinal disease. This is a reported figure tied to the flagship base and should be read as such. [Military Times reporting on DODIG-2026-057](https://www.militarytimes.com/news/your-military/2026/02/24/military-dogs-are-facing-substandard-conditions-leading-to-deaths-report-finds/). - As of February 15, 2022, the GAO counted about 5,100 federally managed working dogs, with DoD holding nearly 1,800 (1,769), second only to DHS at 2,942; the single largest program is TSA at 1,097. The GAO also found about half of federal and half of contractor programs did not address abuse, neglect, rest, or retirement, and it made 19 recommendations. Procuring and training a dog is estimated at approximately $65,000 to $85,000 (GAO page 6, attributed to agency officials), an acquisition-and-training estimate, not a contract ceiling; specialized detection dogs are valued well past $150,000, while the commonly cited $6,500 figure covers only a roughly ninety-day basic course. [GAO-23-104489, Working Dogs](https://www.gao.gov/products/gao-23-104489). - A 2018 Inspector General report found the Army mishandled disposition of its Tactical Explosive Detection Dogs, blocking former handlers from adopting the dogs they deployed with, a failure under Robby's Law (10 U.S.C. 2583). [DODIG-2018-081](https://media.defense.gov/2018/Mar/01/2001884674/-1/-1/1/DODIG-2018-081_FINAL_REDACTED.PDF). *This post is informational and journalistic, drawn from public records, and is not legal, financial, or policy advice, with dollar and population figures attributed to their stated fiscal year or reporting date.* --- Canonical HTML: https://jwatte.com/blog/military-working-dog-program/ RSS: https://jwatte.com/feed.xml JSON Feed: https://jwatte.com/feed.json Hero image: https://jwatte.com/images/military-working-dog-program.webp