Before the invasion of Iraq, the United States Army’s principal logistics support contract was worth about five million dollars a year.
By 2011 the same contract had paid out more than thirty-seven billion, to a single company, for base services across Afghanistan, Iraq, Kuwait, Djibouti, Jordan, Kenya, Uzbekistan and Georgia. The Commission on Wartime Contracting, reporting the same year, estimated that at least thirty-one billion dollars and possibly as much as sixty billion had been lost to waste and fraud across contingency contracting in the two wars, out of roughly two hundred and six billion spent.
The Logistics Civil Augmentation Program is the part of this subject that is entirely lawful. It is congressionally authorised, competitively bid, publicly audited, litigated in front of the Government Accountability Office, and reported in company filings to the Securities and Exchange Commission.
It also produced, at scale and at public expense, every structural feature the rest of this subject describes: a workforce employed by companies rather than by a state, a subcontracting chain that obscures who is actually performing, payments that end up wherever the local power sits, and a capability that outlived the wars that created it and went looking for customers. The end-user paperwork differs. The airframes, in a number of documented cases, did not.
LOGCAP is not the exception to the ghost-plane economy. It is the same machine with a contract number.
What LOGCAP actually is
The programme was created in 1985 and the first contract was awarded in 1992. The concept is straightforward: rather than maintaining uniformed personnel to run dining facilities, laundries, fuel distribution, housing, waste management and base maintenance in a theatre of operations, the Army contracts for those services and deploys soldiers to do soldiering.
The precedent is older than the programme. Armies have contracted sutlers, teamsters and victuallers for as long as there have been armies, and the twentieth-century model of a self-sufficient military performing its own support was the historical anomaly rather than the norm. What changed in the modern era is scale, legal formality, and the fact that the contractors now operate in the same battlespace as the troops they serve.
The logic is genuinely sound. A dining facility does not require a rifleman. Every service function performed by a contractor is a uniformed position not needed, and in an all-volunteer force with a finite recruiting pool that is a real constraint being relieved.
The programme ran in generations. LOGCAP I went to Brown and Root in 1992, covering the Balkans. LOGCAP II went to DynCorp. LOGCAP III was awarded in December 2001 to Kellogg Brown and Root, sole-sourced, and became the largest of its kind as the wars expanded.
LOGCAP IV was awarded in April 2008 to three companies, DynCorp International, Fluor Intercontinental and KBR, competing for individual task orders, with SERCO running programme management. It was structured as indefinite-delivery, indefinite-quantity, one base year and nine option years, with a ceiling of a hundred and fifty billion dollars.
LOGCAP V arrived in 2019, awarded to Vectrus, Fluor, PAE-Parsons and KBR, with a stated value of eighty-two billion dollars across regional task orders divided by combatant command: Central, European, Indo-Pacific, Africa, Northern and Southern Command, plus Afghanistan.
The task orders have names like Setting the Theater, which is the Army describing, with more candour than it probably intended, what the programme is for. Setting the theater means establishing the physical conditions under which force can be applied, and doing it in advance of any decision to apply force. It is the same function pre-positioned depots served in the Sahara, scaled to a combatant command and put out to tender.
Ceilings, obligations, and the number that matters
A detail worth extracting because it recurs across every large government contract vehicle and is routinely misreported.
LOGCAP IV carried a ceiling of a hundred and fifty billion dollars. The amount actually obligated across all LOGCAP IV task orders was twenty-two billion.
A ceiling is a maximum authorisation, not a commitment. It is the number that appears in headlines and it describes what could be spent rather than what was. The obligation is the real figure and it is frequently an order of magnitude smaller.
That distinction matters for reading this subject generally, because the same confusion runs through every reported contract, every announced constellation and every projected capacity figure in the grey market as well as the white one. The number with the most zeroes is usually an authorisation, and the party quoting it usually knows that.
What does not shrink under the same scrutiny is the actual spend, which was enormous. Two hundred and six billion dollars in contracts and grants across Iraq and Afghanistan by the end of fiscal 2011, of which the Commission’s conservative estimate put thirty-one to sixty billion as waste and fraud, averaging roughly twelve million dollars a day across a decade, with perhaps eighteen billion of it pure fraud.
The people, and the status problem
This is where LOGCAP arrives at the same place Air America did, by an entirely different route.
At various points the contractor population in Iraq and Afghanistan approached or exceeded the uniformed population. By 2011, contractor deaths including local and third-country nationals exceeded military deaths in both countries.
The Commission noted something about that figure that applies directly to everything else in this subject. Contractor deaths are almost certainly higher than the reported total, because the federal statistics are based on filed insurance claims, and many foreign contractors’ employees may be unaware of their insurance rights and therefore never file.
Read that sentence carefully. The official count of people killed doing this work is a count of insurance claims, and the people least likely to file are the third-country nationals who made up the bulk of the workforce and who were recruited through labour brokers in South Asia and East Africa. The count is therefore not a count of deaths. It is a count of successful insurance interactions, which is a different quantity with a systematic bias in a known direction.
That is precisely the Air America problem restated under a compensation statute. The men doing state work under a commercial employment relationship have a status that is worse than a soldier’s and better than nothing, and the gap between those two is where the accounting stops. Air America’s crews fought for decades for retirement credit that has still not been legislated, and the men who flew the post-Soviet freighters had no scheme to be unaware of. LOGCAP’s third-country nationals have an insurance scheme they may not know exists.
Deniability was never the purpose here. The status outcome arrived anyway, because it is a property of the employment structure rather than of anybody’s intent.
The command problem
There is a second consequence that the military has documented itself and that gets less attention than the money.
Commanders depend on contractor support and do not command contractors. A contractor performs according to contract terms administered by a contracting officer, who may be somewhere else entirely and who is not in the chain of command. The Congressional Budget Office has noted the reduced direct authority this produces.
In practice this means a battalion commander whose fuel, food, water and generator maintenance are performed by a company cannot order those services to continue under conditions the contract does not cover. He can request. The contracting officer can modify. The company can decline and invoke force majeure.
That dependency is a form of exposure that no adversary had to create. It was built into the force structure as a cost-saving measure, and it means the sustainment of a deployed army rests on commercial performance rather than on military obedience. During the Iraq drawdown, and again in Afghanistan, that dependency became a planning constraint in its own right: you cannot withdraw faster than the people who load the aircraft are contracted to work.
Where the money went
The Host Nation Trucking contract in Afghanistan is the case that demonstrates what happens when contracted logistics meets a place where the state does not control the roads.
A congressional investigation in 2010 produced a report titled Warlord, Inc., which found a vast protection racket in which warlords, criminals and insurgents extorted contractors for safe passage. Task Force 2010 subsequently traced over three hundred and sixty million dollars in Afghan contracting funds diverted to warlords, power brokers, insurgents and criminal patronage networks, and confirmed that many trucking contractors were making illicit payments that ended up in the hands of the enemy.
The mechanism is worth stating without outrage because outrage obscures it. The United States needed supplies moved along roads it did not control. It contracted the movement to companies. Those companies needed the convoys to arrive. The people who could guarantee arrival were the people who could also prevent it. So the contractors paid them, and the payment was priced into the contract, and the government paid the contract. Roughly three hundred and sixty million dollars is the traced figure. The untraced figure is unknown by definition, because the payments occurred at the bottom of a subcontracting chain in cash, in a country with no functioning financial reporting, and the settlement mechanisms available there do not generate records.
Nobody in that chain had to intend the outcome. It is the same structure as a Chadian airfield collecting fees on transit or a Gabonese registry selling a flag: whoever controls the chokepoint gets paid, and the buyer’s preferences about who that is do not enter into it.
Fuel produced parallel findings. A 2008 congressional examination looked at corruption allegations in Defense Logistics Agency fuel contracts in Iraq, and a subsequent multi-billion-dollar billing dispute with the company feeding American troops in Afghanistan ran for years.
The air piece
The programme’s aviation dimension is the point where the lawful and unlawful halves of this subject become physically indistinguishable.
The United States and its NATO allies discovered during these wars that they did not possess enough outsized strategic airlift. The solution was to charter it, and the aircraft available for charter were Antonov An-124s and Ilyushin Il-76s operated by Ukrainian and Russian companies.
So Western defence ministries spent two decades as major customers of the post-Soviet heavy-lift industry, moving helicopters, vehicles and outsized equipment into and out of theatres on aircraft crewed by the same population of former Soviet military aviators who staffed the grey market.
The same airframes. Frequently the same operators. Sometimes, over a career, the same men. The difference between the two halves of that industry is a procurement document and a compliance department, and the aircraft does not know which one it is flying under. An An-124 moving a helicopter for NATO and an Il-76 moving crates to a desert strip are the same industrial capability serving different customers, and the capability was priced by whichever customer was bidding that week.
That is the most important thing LOGCAP demonstrates and it is rarely stated: the capability is neutral, the market is shared, and the largest single customer of contracted military logistics on earth is the government of the United States.
Why the criticisms are more complicated than they look
The standard account of LOGCAP is a scandal narrative, and it is worth complicating because the scandal version makes the structural point harder to see.
The competition criticism is real and was addressed. LOGCAP III was sole-sourced to one company whose former chief executive was the sitting Vice President when it was awarded, which is a fact that requires no embellishment. The Army responded by restructuring to multiple awards with task-order competition, which is the correct institutional response, and the successor contracts have been competed and protested, with both Fluor and DynCorp litigating the LOGCAP V awards.
The performance criticism is also real. KBR’s record on LOGCAP III included exposing troops to unsafe water and disputes over firefighter pay and benefits, and the company was awarded LOGCAP IV notwithstanding.
And yet the Army kept extending. In 2010 it chose to extend LOGCAP III for base services in Iraq rather than transition to the competitively awarded LOGCAP IV, which tells you that continuity of service during a drawdown outweighed the competition benefits at the moment of decision.
That is the recurring pattern in contracted logistics and it is not corruption. An incumbent performing a function during an operation acquires a position that is hard to displace, because displacement means transition risk in a live theatre. Competitive structure is easy to design and hard to exercise, and the reason is operational rather than venal. The same effect appears in any long-running access relationship where the incumbent supplies something the customer cannot briefly do without.
What the programme was actually buying
Strip away the controversy and the honest assessment is mixed in a specific way.
What LOGCAP genuinely delivered was speed and scale. A contractor can hire globally, move quickly, and surge without a congressional authorisation for end strength. Building the same capability in uniform would have required a much larger army, recruited and trained over years, at a cost the political system would not have authorised.
What it cost was control, oversight capacity, and cost discipline. The Commission’s central finding was tremendous over-reliance, produced by trying to do too much, treating contractors as a free resource, and failing to adapt plans to host-nation conditions.
The word free in that formulation is doing the work. A contractor appears free at the point of use because the requirement is satisfied by signing a task order rather than by finding soldiers. The cost is real, arrives later, and lands in a different budget line from the one the requesting commander is responsible for.
Any system that makes a resource appear costless at the point of decision will overconsume it. That is not a defence contracting insight. It is a general property of institutions, and the wartime version merely runs it at twelve million dollars a day for a decade.
The subcontracting layers
The prime contractor is the name on the contract and frequently not the party doing the work, and the layering underneath is where the programme’s visibility ends.
A prime holding a task order for base operations at a location in Afghanistan subcontracts catering, and the caterer subcontracts food supply, and the supplier subcontracts transport, and the transporter engages local hauliers, and the local hauliers arrange passage with whoever controls the road. By the fourth tier, the contracting officer administering the task order has no visibility at all, and no contractual relationship with the party actually performing.
That is structurally identical to the corporate layering described in the Gulf free zones, with one difference that matters. In the free zone case the layers exist to defeat inquiry. In the LOGCAP case they exist because a global prime cannot directly employ a truck driver in Kandahar, which is an entirely legitimate reason producing an identical result. Opacity does not require intent. It is the default output of any chain long enough, and the chains in this business are long by necessity rather than by design.
The Warlord, Inc. finding is what that looks like at the bottom of the stack. The money leaves Washington as an appropriation, passes through a prime with audited accounts and a compliance department, and arrives four tiers down as a cash payment to a man who controls a section of highway. Every tier is documented. The sum of the documentation describes something nobody at the top authorised.
Subcontracting is also how the workforce composition got where it did. The prime employs Americans. The subcontractors employ third-country nationals recruited through labour brokers, at wages set in labour markets that have no relationship to the contract value, and those are the people who make up the bulk of the headcount and, as the casualty accounting shows, a substantial share of the dead.
The claims that do not hold up
An audit, because this programme is described badly by both its critics and its defenders.
LOGCAP was a hundred and fifty billion dollar contract confuses a ceiling with an obligation. Twenty-two billion was obligated under LOGCAP IV.
Sixty billion dollars was stolen misstates a finding about waste and fraud combined, of which the Commission attributed perhaps eighteen billion to fraud specifically, with the remainder covering projects that produced little or no benefit.
Contractors are mercenaries conflates logistics support with armed security. The overwhelming majority of LOGCAP personnel cooked, cleaned, built, maintained and drove.
LOGCAP was a privatisation project of one political persuasion understates a bipartisan trajectory running from the 1980s through administrations of both parties, with the largest expansions occurring under a Republican administration and the largest restructuring under a Democratic one.
The Army should simply do this itself ignores the end-strength arithmetic, which is the constraint the programme exists to relieve.
Contractors died in smaller numbers than soldiers is contradicted by the Commission’s own finding that contractor deaths exceeded military deaths in both countries by 2011, and by its caution that the true figure is higher.
The scale of LOGCAP waste was unforeseeable is contradicted by a decade of inspector general reporting and by a senator’s observation that the underlying problems identified by the Commission had been known for years.
What LOGCAP is actually telling us
The finding worth carrying is that the ghost-plane economy and the lawful defence logistics industry are not two industries.
They share aircraft types, operators, crews, routes, airfields, freight forwarders, insurance markets, and in several cases the same corporate entities under different contracts. The distinction between them is documentary: whether the movement has a task order and a contracting officer, or an invoice and a company registered somewhere convenient.
That shared foundation explains something the enforcement literature struggles with. Ground a suspect operator and you may be grounding a carrier that holds a defence charter. Restrict a registry and you restrict aircraft that move humanitarian cargo. Designate an operator and you may be designating a company that holds a NATO charter, which is a situation that has arisen more than once and is resolved quietly. The grey market is not a parasite on a healthy body. It is the same body, and the healthy and unhealthy functions run through identical organs.
And the second finding is the one that sets up what follows. Two decades of contingency contracting built an enormous industry: companies, personnel, aircraft, relationships, and above all a population of people who know how to move heavy freight into difficult places at short notice under commercial terms.
Wars end. Contracts do not renew. Capability does not evaporate when the demand that created it disappears, and the capability the United States built for Iraq and Afghanistan was released, fully capitalised and fully trained, into a world that still had plenty of places wanting exactly that service on less scrupulous terms.
A dining facility in Kandahar and a crate of ammunition in Darfur are not the same cargo. They were frequently moved by the same aircraft, flown by the same crews, for companies with the same corporate architecture, and the only reliable difference between the two flights was which document authorised it.