Re-engineering Afghanistan

The sun is poking through mist on a humid Wednesday morning in early April as we drive west through the city of Kandahar, in southern Afghanistan. The place doesn't look too bad, considering that just a couple of days ago rioters demolished stores and homes here after word got out that a minister in Florida had burned a Koran. In Kandahar alone 16 people were killed and 128 wounded.

Our destination is the only electrical substation within city limits, a creaky relic from the 1970s outfitted with Bulgarian switchgear and other electrical esoterica, including a 25-megavolt- ampere transformer that's so old it has to be hosed down on hot days to keep it from overheating. I'm traveling with Chief Warrant Officer 5 Thomas Black, the deputy commander of the U.S. Army Corps of Engineers task force headquartered at Kandahar Airfield, the huge coalition air base to the southeast. He wants to make some voltage measurements at the plant.

I'm tagging along because I am trying to determine whether the hundreds of millions of dollars committed to electrical equipment and construction and fuel have made any difference at all in the vast, chaotic, decadelong effort to weaken the Afghan insurgency, stabilize this notoriously failed state, and push it in the general direction of the 21st century.

Our group is traveling in three military vehicles called MRAPs (mine resistant ambush protected). The one we're in weighs about 16 tons and costs about US $500 000. We're rolling down Kandahar city's main road, a broad thoroughfare with a parklike median that's lush with flowers and trees. I stare out through one of the MRAP's tiny trapezoidal rear windows, as thick as a Dean Koontz novel.

For now, at least, the streets are calm. Small groups of men and boys sit in the median, tending the plants or talking in the late-morning languor. A smoky, musty smell hangs in the air. The street is lined with small, dusty, crumbling, cubelike, open-front sidewalk stores selling produce, meat, hardware, tires, bricks, poles and thatch, and other goods. During the drive I see goats, sheep, camels, donkeys, horses, and dogs. Outside the stores are men wearing loose Afghan qmis shirts and shalwar pants. Some men glare at us; most ignore us. Scattered groups of young boys throw rocks at us. The very few women out are in full burqas, gun-metal gray, a gaping mesh oval over the eyes and nose.

Kandahar is Afghanistan's second city, behind Kabul in the north. But where Kabul has nearly round-the-clock power now, Kandahar's 850 000 people have it only fitfully, about 40 megawatts at most. For comparison, NATO's airbase nearby, with about 30 000 people, has about 100 MW.

Kandahar is primitive by any standard, a backwater of a backwater. Afghanistan ranks in the bottom 10 percent of the world in electricity use per capita, according to the U.S. Central Intelligence Agency's World Factbook. And that deficiency is much more acute in the south than in the north. Most Kandaharis make do with 4 to 6 hours a day at most. The transmission and distribution systems throughout the south are frayed and dilapidated, spliced and respliced by hand so many times that the lines can now handle only about two-thirds of the current they could 25 years ago. Many feeders in the region experience multiple outages a day, typically because two lines have slapped together or because the current has arced or "flashed over" old and cracked insulators.

Here in southern Afghanistan, the local linemen scramble up utility poles in baggy garb and sandals, without safety belts, rubber gloves, or any other protective gear. Only a small fraction of the workers can read, their boss will later tell me.

As they did in Iraq, the coalition forces in Afghanistan have spent or committed tens of billions of dollars to building and refurbishing infrastructure, including roads, schools, hospitals, and water systems. It is an enormous effort involving dozens of government agencies and hundreds of contractors. Among the government organizations, the U.S. Agency for International Development (USAID) is dominant, but the U.S. Army Corps of Engineers and other military organizations have big roles as well. The Asian Development Bank and the governments of India and Germany have also funded a few large projects.

Having spent an estimated $55 billion on Afghan reconstruction over the past decade, the U.S. government is the largest contributor by far. And, as in Iraq, much of that money has been wasted or badly spent, particularly in the electrical sector. The goals of reconstruction are pretty basic: to stimulate economic activity and create jobs; to make life more comfortable and secure; to give people a more attractive alternative to the typically medieval societies imposed by insurgents. And last, but certainly not least, to win the allegiance of citizens and build their confidence in fledgling government institutions and officials who are all too often bungling or corrupt. Or both.

"If you electrify a village, a carpenter can have power tools and work after the sun goes down," notes Thomas Barfield, a professor of anthropology at Boston University who specializes in Afghanistan and reconstruction. "Electricity transforms the way people live. Once they get a taste of electricity they can't help but wonder, 'Which political system is going to deliver these things to me?' The Taliban is going to deliver a 6th- century state, while NATO is offering a 21st-century state."

And yet in the electrical sector, especially, the attempts to modernize Afghanistan have been dismal, despite the fact that the agencies and many of the contractors involved have had years of reconstruction experience in Iraq.

Of the many reasons for the disappointing results, a few stand out. The standard contracting methodology used by U.S. government development agencies in Afghanistan does nothing to discourage overspending and inefficiency. In addition, Afghanistan has a long history of relying on foreign powers, often invading ones, for its major infrastructure. This reliance has fostered a culture of passivity and a crippling deficiency of homegrown engineering expertise.

In the electrical sector, however, these factors pale alongside a couple of others. The Afghan national electric utility is unable to collect enough revenue to sustain its own operations and it has trouble simply keeping records consistently. And one of the most serious problems, according to analysts, officials, and engineers interviewed in Afghanistan and the United States, is the incompetence of USAID, an independent U.S. government agency that receives guidance from the Secretary of State. It is the dominant development organization in Afghanistan, and it has, according to these observers and even its own internal documents, made major missteps in every significant electrical construction project it has undertaken in the country.

"USAID has grossly underperformed in electrical reconstruction," says Charles Tiefer, an attorney and contracting expert who serves on the Commission on Wartime Contracting, a panel appointed by the U.S. Congress to analyze why so much money was misspent in Afghanistan and Iraq.

"What you've got is an agency that has become a contract- management agency, and they're not even very good at that," adds Dov S. Zakheim, another Wartime Contracting commissioner. "They threw money at contracting with less-than-great results."

The Wartime Contracting commission released its final, 248-page report (PDF) at the end of August. It concluded that between $31 billion and $60 billion (PDF) of the approximately $206 billion spent on contracting in Iraq and Afghanistan was wasted, because of badly conceived projects, poor planning and oversight, and criminal behavior, such as fraud and corruption. (The $206 billion figure includes contracts for logistics and security, in addition to reconstruction.) The report also says that "money lost as a result of the inability to sustain projects could easily exceed the contract waste and fraud already incurred."

Regardless, USAID is plowing ahead. The agency recently embarked on a long-planned $1.2 billion project to give Afghanistan a modern national electrical grid. With U.S. and other NATO forces starting their drawdown from Afghanistan, the grid would be a superb legacy. But not much in the agency's record in Afghanistan suggests that the project will be completed competently and efficiently.

USAID's struggles prompt questions about what, exactly, diplomatic and military officials can take away from the Afghan reconstruction experience. In Afghanistan, as in Iraq, the coalition tested new and sometimes radical strategies in counter insurgency, and through it all reconstruction was a cornerstone. A half dozen or more development agencies used scores of contractors and subcontractors to perform electrical and other work in Afghanistan and in Iraq, too. As the U.S. government itself has disclosed in thousands of pages of reports and analyses, far too little was achieved for the more than $100 billion spent reconstructing the two countries—even after taking into account the inevitable failures and frustrations of war-zone construction.

So, did reconstruction make a difference in the counterinsurgency? Was it worth the money and lives it cost?

We arrive at the Kandahar city substation, which is called Breshna Kot. There's a collection of diesel generators here, but none of them are running. I'm told that the utility has not been able to negotiate a contract for diesel fuel at acceptable terms. Apparently, the utility simply can't afford to pay what the suppliers charge for fuel. That's not much of a surprise; the Afghan national utility, Da Afghanistan Breshna Sherkat (DABS), is stunningly unprofitable. It charges about 6 to 10 Afghanis (about 12 to 20 U.S. cents) per kilowatt-hour, a third of what it would need simply to break even, NATO engineers tell me. And that's if most customers were actually paying for their electricity. Which they're not. At the moment, the World Bank and the U.S. government are covering the diesel fuel costs, which run well into the tens of millions of dollars a year.

Breshna Kot is what our MRAP crew is referring to as "semisecure," so we have arrived in four MRAPs instead of three. The MRAPs are left running, and we are told to leave our body armor on because the soldiers running security for our mission are skittish about the situation around us in the city. Chief Black and I are standing in the switchyard, our backs to the substation office. Besides the Bulgarian 25-MVA transformer in front of us, there are some vacuum disconnect switches, also Bulgarian. This gear connects several city feeders to an old and threadbare 110-kilovolt line that brings power from the hydro plants at the massive Kajaki Dam, 180 kilometers to the northwest, deep in Taliban country.

But right now there is no hum at all, which means there is no power coming from Kajaki.

Suddenly there is a loud thunk, and then another. Black smiles. The first thunk was the sound of the transformer being energized. The second was the substation's electrical bus being energized. Power is flowing again.

Now there's a faint hiss in the morning air. It comes from mild electrical arcs. It's the sound of decrepit insulators and incipient failure.

A rotund man with a flowing gray-and-white beard and a dazzling smile joins us. He is engineer Fazal Ahmad, the power director for Kandahar province for DABS. Ahmad explains that the power from Kajaki had been interrupted when the 110-kV line was damaged during fighting in Lashkar Gah, the capital of Helmand province.

We go into the substation building to look at the electrical bus, which interconnects the big transformer and also the now-dormant diesel generators to the lines that feed the city's seven electrical districts. "This is—what do you call it?—magic," Ahmad tells me, his eyes twinkling. Normally, a substation bus is a continuous bar of copper, but this one is broken up into five removable pieces, which allows Ahmad's technicians to channel power from either the Kajaki line or from the diesels to several of the city's feeders. This odd mix-and-match arrangement is necessary in order to keep the two power sources separate. They aren't in synchrony, so connecting them could blow the disconnect switches. Or worse.

Illustration: Bryan Christie Design

NORTH-SOUTH CONNECTION: USAID’s long-term plan for the Afghan electrical grid includes a line [in orange] that will connect networks in the north and south for the first time. Click to enlarge.

The exposed copper bars, at 20 000 volts, give Black the heebie-jeebies. He emphatically warns Ahmad and me to be careful as we go behind the big metal cabinets to look at the baroque and dusty apparatus that makes the magic. I am struck by the resourcefulness of it, even if it's all kind of harebrained and rickety. "One substation is divided into five parts," Ahmad explains. "That's why I call it magic."

He wants to tell me one other thing, too. Out in the yard, beaming, he points to the big 25-MVA transformer. The transmission system linking the substation to Kajaki, including that transformer and the line itself, was engineered and installed by Afghans 15 years ago during the Taliban period, he wants me to know. His pride is touching.

As well it may be. Hardly any heavy-electrical installations in the country have been engineered by Afghans. During a briefing at the Army Corps of Engineers compound at Kandahar Airfield, I'd asked Black to tell me who is doing the engineering on the scores of electrical projects scattered around the country. "We're doing it," he answered, and by "we" he meant NATO—including the Corps of Engineers and various Western electrical contracting firms. "Before us, the Russians were doing it," he added. And, I later learned, before the Russians, the Americans were doing it. As near as I can tell, Ahmad is the only degreed electrical engineer working for DABS in all of southern Afghanistan.

A little after 11:00 a.m. a member of our security detail approaches Black. It seems an improvised explosive device has been found in the road about 2 kilometers away. The security guys urge us to wrap up our business and get out of there. But no visit in Afghanistan is complete without tea, so we repair to Ahmad's little office. There are bright turquoise walls, folding chairs, a scarred Formica coffee table, a desk heaped with blueprints and reports, and a neatly made bed in the corner for nights when it's not safe to travel. As the security team stalks around outside, we sip tea and eat sugary dates and cookies.

Electrically, Afghanistan is not one country but many. Individual feeders within a district, particularly in the south, are typically not synchronized with each other, let alone larger grids. Even if the country did have large-scale grids, power- sharing agreements would probably be difficult because of the lack of trust that often strains relationships among different ethnic groups, tribes, and provinces.

The reason why Kabul and the surrounding areas have round-the-clock electricity is that the northern provinces benefit from transmission-line connections to Afghanistan's northern neighbors: Turkmenistan and Uzbekistan (and soon Tajikistan). Those central Asian countries have a surfeit of electricity from massive hydro plants built by the Soviets decades ago. But there are no connections between Afghanistan's northern power grids and those in the south, so any surpluses of power in the north can't be shared.

What the south has, first and foremost, is the hydroelectric plant at the Kajaki Dam, which is sizable but decrepit. There is also a collection of smaller hydro plants and a smattering of diesel-fueled plants—including two brand-new ones just installed by the Corps of Engineers. These two plants, which flank Kandahar city, are each 10 MW. They were built as part of a $94 million contract that also covers one year of fuel and operation and maintenance expenses. The tiny, very fortunate minority of Kandahar residents who happen to live along one of the distribution feeders from either of these plants have electric power pretty much all the time. Anywhere else in the city, you can count on perhaps 6 hours of electricity a day at most.

It is impossible to understand the unfulfilled promise of electrification in southern Afghanistan without an appreciation of the tortuous and improbable history of the Kajaki hydroelectric plant. Built 36 years ago with funding from USAID, the plant was intended mostly to power Kandahar. But it hasn't really worked out that way. Kajaki is in Helmand province, adjacent to Kandahar province, and it generates as much as 33 MW, of which 7 to 10 MW typically end up in Kandahar city, according to Chief Black.

The rest is siphoned off in Helmand, in some places by taps placed by the Taliban. A report last year in The Wall Street Journal disclosed that in Taliban-controlled districts in Helmand, residents paid the Taliban directly for their electricity, and that much of the electricity was being used to pump water to irrigate opium poppies for the drug trade, which further benefited the Taliban.

The Kajaki Dam is so strategically important to Helmand and Kandahar that it was among the early targets bombed by U.S. warplanes as the war against the Taliban began in October 2001. The hydroelectric facility has two turbines, known as units 1 and 3, and space between them for a third unit, which would be known as unit 2. USAID kicked off the refurbishment of Kajaki in December 2003 by signing a contract with the Louis Berger Group, a U.S.-based contractor. Louis Berger then subcontracted with Voith Hydro to rehabilitate unit 1 and with a Chinese firm called China Machine-Building International Corp. to fix unit 3. CMIC was also engaged to build the turbine generator, transformers, and related gear for unit 2, which was to boost the output of Kajaki from 33 MW to 51.5 MW. CMIC had come to Kajaki during the Taliban years, when the Taliban had no idea how to operate and maintain it. [See answers submitted by the Louis Berger Group in response to written questions from IEEE Spectrum.]

In 2006 another U.S. contractor, Black & Veatch Corp., entered the picture. It joined forces with Louis Berger, and USAID awarded this Joint Venture, as it was known, a $1.4 billion contract for work in Afghanistan, including the ongoing restoration of Kajaki. Meanwhile, the Taliban was strengthening its grip on the area around Kajaki, with frequent rocket attacks on the site all through the summer of 2006.

In 2008 CMIC finished building the equipment for unit 2, and that gear, weighing 220 metric tons, was delivered to Kajaki in a legendary operation code-named Eagle's Summit. The area around the dam was under Taliban control, and the equipment was too heavy to fly in by helicopter. So to get the gear from Kandahar Airfield to the dam, 60 British officers spent four months devising a daring land operation that was later declared the largest logistical mission since World War II.

The mission, which involved some 4500 soldiers, began in late August with heavy bombing of Taliban positions and raids by U.S. and British special forces on Taliban-held villages along the Helmand River. On 27 August, a 100-vehicle convoy including the trucks carrying the gear set out from Kandahar and drove into the desert, usually going just a few kilometers per hour. Well ahead of the convoy was another, "dummy" convoy of about 40 Danish vehicles that took the main road near the dam, drawing fire and diverting attention from the real convoy. Bomb-disposal teams cleared scores of improvised explosive devices along the route, including 11 in a single short stretch of road not far from the dam. The convoy transporting the massive electrical gear, which was hidden in multiple shipping containers covered with posters emblazoned with quotes from the Koran, took an alternate route through the mountains mapped earlier by a British reconnaissance team. At several points the convoys were rocketed or mortared, but on 2 September, the trucks rolled into the yard at the dam with the equipment undamaged. The casualty count for the 12-day mission was one coalition soldier dead and eight wounded and approximately 200 Taliban killed.

Sadly, the heavy-electrical gear still sits unused at Kajaki. One month after Eagle's Summit, the Chinese engineers from CMIC abruptly left Kajaki. According to a USAID official in Kabul, the engineers realized that the Taliban could not be cleared anytime soon from the area around the road leading to the plant, which meant there was no chance that NATO could deliver the hundreds of tons of concrete needed to install the third hydroturbine.

It would get worse for USAID. That autumn, while the Kajaki third-turbine project was falling apart, another huge project started floundering. It stemmed from a 2006 decision by the George W. Bush administration to build a power plant near Kabul as part of an effort to get President Hamid Karzai reelected. The idea that the plant would help Karzai was based on "a misreading of Afghan politics," says Boston University's Barfield. Nevertheless, the notion that the plant was vital to Karzai's success was so entrenched that Afghans started mocking it as "Karzai's winter coat."

In a war zone, the rules of logic sometimes seem suspended. Still, it is hard to understand why anyone thought it was a good idea to build, in a wretchedly poor country, a plant that would consume vast quantities of extraordinarily costly diesel fuel.

The diesel plant, which was built at a village northeast of Kabul called Tarakhil, quickly became emblematic of so much of what went wrong with reconstruction in the Iraq and Afghanistan wars. Much of the story has been documented by Pratap Chatterjee of the CorpWatch news service and Marisa Taylor of McClatchy Newspapers, among others.

As at Kajaki—and, indeed, for virtually every single one of USAID's electrical projects—the primary contractor was the Berger–Black & Veatch Joint Venture. In this alliance, Black & Veatch took the lead for the electrical projects. But in keeping with the U.S. government's modus operandi in Afghanistan (and before Afghanistan, in Iraq), the prime contractor for the project wouldn't actually build the plant. For that, it would hire a subcontractor. And whereas the Joint Venture enjoyed a "cost-plus" contract with USAID that guaranteed that all of its expenses would be covered and a profit paid on top of them, the subcontractor would be on a "firm-fixed price" contract that made no such guarantees (more on that later).

In May 2007, USAID had formally agreed with the government of Afghanistan to build the plant. And yet it wasn't until 14 June 2008—13 months after that agreement—that Black & Veatch signed a $62 million, firm-fixed price contract with Symbion Power to build the plant. Symbion, a U.S.-based electrical contractor, had compiled an impressive record in Iraq. The delay notwithstanding, USAID and Black & Veatch wanted the first 70 MW in place and operating within six months—by the end of December 2008—and the rest by April 2009. It was an aggressive if not unrealistic schedule.

Why did 13 months go by before Symbion got the subcontract? A 10 November 2009 audit (PDF) by USAID's inspector general says that the plant was delayed because USAID had failed to secure the title to the land where it intended to build the plant. The title problem plagued the project intermittently from May 2007 until April 2008. The audit also faults USAID for writing a project plan so vague that it did not require the contractor to provide "specific deliverables with concrete delivery dates."

"Furthermore," the audit goes on, "the contract contained no consequences for failing to provide power by the specified milestone dates." It's a lapse that the USAID inspector general said was "compounded by the inexperience of the original mission personnel tasked with preparing the statement of work."

In response to a written question, USAID's press office blamed the delay in part on a local tribe. "In early spring of 2008…the local tribe [Hotak] disrupted the construction of the perimeter wall and it took GIRoA [Government of the Islamic Republic of Afghanistan] until April of that year to settle the issue." [You can read the full text of Spectrum's questions, and USAID's answers, online.]

As construction proceeded in the fall of 2008, U.S. State Department officials in Kabul suddenly took an interest in the plant. The urgency increased when rumors began flying around the site that George W. Bush, who would make his final visit as president to Afghanistan that December, wanted to flip the switch at the inauguration of the Tarakhil plant.

But at Tarakhil, Symbion was grappling with several problems. The major ones were technical. First, despite USAID's policy of favoring U.S. manufacturers, the units had been manufactured in Germany. So all the piping, connectors, and other hardware that Symbion needed to install them were in metric units. That meant Symbion couldn't use imperial-unit hardware, which was cheaply and easily available in Pakistan or Dubai. Second—and worse—the construction drawings and plans of the power plant, supplied by Black & Veatch, were "riddled with errors," according to Symbion's CEO, Paul Hinks. Hinks also said that Black & Veatch has denied this allegation.

Photo: Petty Officer 2nd Class Jonathan Chandler

[Spectrum submitted 34 questions in writing to Black & Veatch's corporate communications department. In response, Black & Veatch sent a message declining to answer the questions, citing the stipulations of an ongoing arbitration case between the company and Symbion. You can read the full text of Black & Veatch's response. In the arbitration case, Symbion is seeking about $2 million from Black & Veatch for work on the Tarakhil plant. At press time, the case was being arbitrated by the International Court of Arbitration of the International Chamber of Commerce.]

On 29 October 2008, with Tarakhil far from finished, it fell to Michael McGovern, the head of the Berger–Black & Veatch Joint Venture, to inform William Wood, the U.S. ambassador to Afghanistan, and Michael J. Yates, USAID's mission director in Afghanistan, that the Tarakhil project wouldn't achieve the 70-MW milestone by the end of the year. In mid-January came a flurry of telephone conversations and letters between USAID's Yates and Len C. Rodman, chairman of Black & Veatch Corp., the parent company of Black & Veatch Special Projects Corp., the subsidiary working for USAID. Copies of the letters and memoranda summarizing the telephone conversations were obtained under the Freedom of Information Act (FOIA). In one of them, a letter to Rodman dated 24 January 2009, Yates expresses USAID's "extreme dissatisfaction over (1) the delay in the construction of the 100 MW Kabul Power Plant project and (2) the inability of the Louis Berger Group–Black & Veatch Joint Venture (JV) to provide USAID with critical information in a timely manner."

The FOIA documents do not mention any private discussions that executives of Black & Veatch and Louis Berger might have had that January. But subsequent events suggest that Black & Veatch and Berger decided to assign the blame to Symbion. In a telephone conversation on 16 January 2009, summarized in the documents, Rodman addressed the delays at Tarakhil, telling Yates that "this situation is a massive failure of the subcontractor." It was an odd assertion, because on 15 October, one of Rodman's direct reports, Robert N. Bell, the senior vice president in charge of Black & Veatch's operations in Afghanistan, had sent an e-mail to Paul Hinks that said: "We all recognize what a great effort Symbion is making within their own management and are simply being hampered by the Afghan locals simply being unreliable and unwilling to work. All of us are 100 percent behind you and supportive of you and please do not ever think differently."

In January, Black & Veatch fired (and then rehired and demoted) the Tarakhil project manager, Jack Currie. Currie's replacement, newly arrived at Tarakhil, promptly concluded that Symbion had already been paid too much money and refused to pay Symbion for two invoices totaling $1.85 million.

This was strange, too, in view of the fact that the Berger–Black & Veatch Joint Venture had already approved the payments to Symbion, had already passed the costs on to USAID, and indeed had already received compensation for them from USAID, according to subsequent congressional testimony and the USAID inspector general's report of 10 November 2009. The same report faulted Black & Veatch for not returning the money, which then amounted to over $2 million with interest, to the U.S. government.

On 19 May, after working at Tarakhil for nearly five months without pay, Hinks says, Symbion sent a letter to the Joint Venture terminating the contract on the grounds of material breach. In June 2009, the dispute over the uncollected invoices went to arbitration before the International Court of Arbitration. As this article went to press, a decision was expected at the end of September.

The Tarakhil plant was finally completed, by Black & Veatch, on 31 May 2010. The 10 November 2009 audit by USAID's inspector general found that cost overruns on Tarakhil amounted to $39 million, mainly because of construction and other delays. It's unclear what the total expenditures for the plant were, but almost everyone agrees they added up to more than $300 million.

And since its completion a year and a half ago, the Tarakhil plant has hardly been used. The main reason is that its diesel fuel must be imported by truck at great cost. This past April the electricity generated at the Tarakhil plant cost around 42 cents per kilowatt-hour, according to engineers and development specialists interviewed in Afghanistan. For comparison, the electricity from Turkmenistan and Uzbekistan costs around 6 cents per kilowatt-hour, a USAID official said at a briefing in Washington, D.C., on 27 June.

In an interview with the Associated Press shortly after the Tarakhil project was completed, Mary Louise Vitelli, an energy and development specialist, referred to it as "sinful." Vitelli had worked with the World Bank as an advisor to the Afghan government when the plant was being built.

Who is to blame for the idea of building this ill-conceived and absurdly oversize diesel plant in the first place? Though the Bush administration first proposed the plant, Black & Veatch seized on the idea and kept it alive. "Black & Veatch were the ones that identified the need for the Tarakhil diesel plant," a provincial reconstruction officer wrote in a letter to Spectrum. "Their power flow studies clearly indicated that the imported power from the 'stans to the north would be insufficient and not available in a timely fashion. Turns out the imported power arrived prior to diesel plant operations such that the diesel plant was obsolete before it even started."

In other words, Black & Veatch, a contractor that stood to make (as it turned out) roughly $25 million by building a huge diesel power plant, was the very same organization that also "proved" that the plant was necessary. USAID did not perform or commission any analysis to try to determine independently whether the mammoth plant was actually needed.

In its response to a question from Spectrum, USAID's press office wrote that "the plant has since proven its worth." As evidence, the response goes on to cite a May 2010 "rock slide [that] interrupted the flow of imported power from Central Asia to Kabul….[T]he Tarakhil plant was run to seamlessly fill the gap in supplies in a way that few if any consumers in Kabul even knew that the NEPS [North East Power System] line had been damaged."

USAID's figures (PDF) do indeed show that on 17 May 2010, the Tarakhil plant generated 437.0 megawatt-hours—far above its typical daily output of 0 MWh. That figure means that if the plant was running at full capacity, it ran for about 4 hours that day. The avoidance of inconvenience to the residents of Kabul is certainly laudable. But does it even begin to justify an expenditure of $300 million? You decide.

You can't read the story of Tarakhil and conclude that USAID, its State Department overseers, and other U.S. government agencies learned anything significant from the many mistakes made in the Iraq reconstruction effort, even though those mistakes were documented in thousands of pages of U.S. government reports.

First, Tarakhil was a politically motivated project from the start. In the rush to get the plant done and help Karzai win reelection, common sense seems to have been cast aside. In his report on the Tarakhil project, the USAID inspector general wrote, "Mission officials commented that they had been under extreme political pressure to deliver a specified amount of additional power before the Afghan elections in the winter of 2008–2009."

Similarly, in Iraq in 2003, the State Department and the U.S. military embarked on a massive, accelerated program that resulted in the installation of dozens of generators all over Iraq—plants that, like Tarakhil in Afghanistan, have been for the most part seldom used. The program that built those plants had been driven primarily by political circumstances rather than engineering and logistical realities. The State and Defense departments in 2003 were struggling to boost Iraqi generating capacity as quickly as possible, to help build confidence in Iraq's fledgling and struggling governmental institutions. That urgency led to the decision to put in dozens of combustion-turbine generators, which could be installed relatively quickly but worked best with a fuel—natural gas—largely unavailable in Iraq. They also required maintenance expertise that no Iraqis had.

By late 2005, it was clear that the combustion-turbine idea in Iraq was a bad one and that a much smarter plan would have been to install more of what the Iraqi engineers and technicians were comfortable with: steam-thermal generators [see "Re-engineering Iraq," IEEE Spectrum, February 2006]. And yet in late 2006—a full year after the combustion-turbine plan was clearly starting to come unraveled in Iraq—USAID and the State Department were making exactly the same mistake in Afghanistan. They were embarking on a project to install a plant whose diesel fuel would have to be imported, at great cost. In one of the world's poorest countries, they intended to build a massive facility whose operating costs wouldn't make sense even in Chicago or Frankfurt.

Had any development officials at USAID and State bothered to heed the lessons of Iraq, a better plan would have been obvious. It would have been based on the fuels and resources available in Afghanistan and the technologies that the Afghans had some experience with. The resulting plan would have relied on hydropower supplemented with some natural-gas-fired combustion turbines.

Afghanistan is a mountainous country with enormous hydroelectric potential—an estimated 23 000 MW, according to the country's Ministry of Energy and Water. Indeed, a parallel but smaller reconstruction effort in Afghanistan, run by military officials and financed with funds from the U.S. Commanders' Emergency Response Program, has installed or repaired dozens of "micro" and "mini" hydroelectric plants. The results with these small facilities have been excellent, according to David Westphalen, the Australian colonel who is the chief engineer of the NATO force in Afghanistan and who spoke with Spectrum at the NATO Afghanistan headquarters near Kabul International Airport.

"It's about building for what the locals can deal with and use," said Westphalen. "We go for things that are easy to maintain."

The diesel plants, on the other hand, with their ultrahigh operating expenses, will certainly be abandoned the moment NATO forces leave. So why did coalition forces spend hundreds of millions of dollars on them? Because they wanted electricity fast, to stabilize population centers as part of their counterinsurgency strategy, and hydro plants can't be built fast.

Even so, there were quick-turnaround options other than diesel. A patchwork of gas fields near a village called Sheberghan, in northwest Afghanistan, contains an estimated 120 billion cubic meters of natural gas (PDF)—enough to fuel generation to the tune of hundreds of megawatts for decades. In February 2008, the Berger–Black & Veatch Joint Venture began working on a USAID-funded project to "design and support the development" of the Sheberghan gas fields to provide power to Kabul. A year into the project, however, USAID rated its satisfaction with the Joint Venture's work: It gave the organization a "zero" rating, on a scale of zero to five. The rebuke was disclosed this past February (PDF) at a hearing in Washington of the Congressional Commission on Wartime Contracting.

A few months later, on 1 June 2009, USAID killed the project, after issuing many memos complaining bitterly about the Joint Venture's lack of progress. The termination came a month after the project was supposed to have been completed. Very little had been accomplished. But $7.1 million of the project's total estimated cost of $11.9 million had been spent.

In February 2008, before construction at Tarakhil got under way and before Symbion fell out with Black & Veatch, Symbion CEO Hinks threw a party at the Heetal Plaza Hotel in Kabul. At the party, Hinks was approached by Jack Whippen, the head of Black & Veatch's organization in Afghanistan. Hinks says that Whippen told him that he, Whippen, would soon ask Hinks to submit an "expression of interest" to bid on a huge project to refurbish and upgrade the main electrical transmission system in southern Afghanistan.

At the time, Black & Veatch was under pressure from USAID to get more bidders for projects. So Whippen's disclosure didn't surprise Hinks. But what Whippen said next sure did: "We both know you can't do this job" is how Hinks remembers it. To Hinks, the message was clear: Whippen did not want him to actually submit the expression of interest.

But the job was precisely the kind at which Symbion excelled. The project was part of an initiative referred to loosely as the Southeastern Power System, or SEPS. It involved repairing and replacing as necessary the 110-kV transmission line running from the Kajaki hydroelectric plant to Kandahar and other sites. Before he came to Afghanistan, Hinks had managed work in Iraq that fulfilled the requirements of 15 contracts that together resulted in the installation of more than 1000 km of transmission lines and 12 substations. The work included a line through Al Anbar province at the height of the often insanely violent insurgency there.

If there had been an attempt to dissuade a qualified and responsible bidder from bidding on a government contract, it would have raised all sorts of red flags. [Messages sent to Whippen's e-mail address were not returned. Black & Veatch declined to make Whippen and other executives available to Spectrum for interviews.]

When the time came, Hinks wasn't deterred. "I thought, 'To hell with it. We're gonna bid,' " he recalls. He responded with an expression of interest, the first step in a bid. In the end, none of it mattered, because in the summer of 2008 Black & Veatch scrapped the bids without explanation.

Not long after, Whippen suggested Hinks talk to Haidar Barak, the CEO of the Afghan Electrical Power Corp. (AEPC), an Afghan-Canadian joint venture. "He [Whippen] drove us in AEPC's direction," Hinks says. Then, on 24 August 2008, Black & Veatch again put SEPS out for bids. Hinks and Barak agreed to form a joint venture to bid on the SEPS project. In the hectic weeks leading up to the final bid, as Hinks was fretting about being underbid by a potential competitor from Turkey, Barak suggested that they add at least $5 million to their $105 million bid. Stunned, Hinks asked why. In an e-mail response, Barak concluded: "I hope we are not leaving money on the table." Hinks, nonplussed, refused to increase the bid.

Weeks later, the odd suggestion started to make sense to Hinks when he learned that the Symbion-AEPC bid was the only one submitted for the project. Had Barak been told that there were no other bidders for the SEPS contract? USAID's own inspector general issues a handbook for contractors that warns explicitly about such a scenario.

Reached at his office in Kabul, Barak denied he had any such inside information. He had "no idea what he [Hinks] is referring to," he said. Then he added that "we had many, many phone conversations about price going up and price going down….It's a normal course of business in bid preparation to go up and down in your suggested price."

Later, Hinks and Barak met with Whippen and three USAID officials in Kabul about the SEPS project. "Out of the blue," Hinks says, Whippen suggested that they build a 220-kV line rather than rebuild the existing 110-kV line between Kajaki and Kandahar, which was what the contract specified. Hinks immediately realized it would add greatly to the cost and duration of the project and advised against it. "It made no sense," he says of the 220-kV idea.

Despite all the maneuvering, Black & Veatch did not issue a subcontract for SEPS. The "whole thing fizzled out" by mid-2009, Hinks says.

After hearing nothing about SEPS for months, Hinks wrote to Karl Eikenberry, who had been recently appointed U.S. ambassador to Afghanistan, urging him to reconsider and restart the project. Eikenberry passed the message to William M. Frej, USAID's mission director in Afghanistan, who in an e-mail to Hinks dated 29 January 2010 said that the SEPS project was "a high priority" and that as soon as the security situation around Kajaki improved, "the United States Agency for International Development will contract services for this project through its normal full and open procurement procedures." Frej repeated the pledge in another e-mail to Hinks on 2 February.

It turned out to be a hollow promise. Subsequent e-mails from Hinks prompted either no response or one from a low-level bureaucrat at USAID. Finally, in December 2010, USAID disclosed its intention to "sole source," to Black & Veatch, the entire $266 million contract for the SEPS initiatives (which USAID was now calling the Kandahar Helmand Power Project). Frej's assurances notwithstanding, there was no competitive bidding at all. In choosing Black & Veatch, USAID bypassed not just Symbion but also several other qualified contractors to select a company that the agency had excoriated in dozens of memos for its performance on the Tarakhil plant, the Sheberghan gas-field study, and the Kajaki hydroelectric facility.

(When USAID disclosed the contract for the Kandahar Helmand Power Project, Louis Berger was not mentioned. On 5 November 2010, after an investigation by the U.S. Department of Justice into fraudulent overbilling for projects outside of Afghanistan, Berger agreed to pay $69 million to settle a combination of criminal penalties and the findings of a civil investigation.)

In its responses to Spectrum's questions, the USAID press office said that Black & Veatch was chosen mainly for expediency. It wrote: "[I]n this case, the usual bid process would have taken six to nine months"—more time than the agency could spare. "Black & Veatch was already mobilized within the theater of operations. Mobilizing a contractor not in the region under these circumstances would have resulted in significant delays."

The rationale doesn't make sense. The SEPS project was formally described in February 2008. Over the next two and a half years, USAID's contractor, Black & Veatch, issued and then abandoned or failed to act on requests for proposals for the project, typically without notice or explanation. And mobilizing a different contractor to take over SEPS could have been done in a few weeks, according to military engineers who have supervised large projects in Afghanistan.

In relying on a single contractor for all its significant electrical work, USAID has managed to introduce a new wrinkle to war-zone reconstruction—one that was absent from Iraq, despite all of its woes. But this problem has just added to the larger morass, because essentially all the major problems in Iraq seem to have been transferred to Afghanistan.

The most fundamental of these is the use of cost-plus contracts (PDF). For any project it wants done, USAID hires a contractor, which is awarded a cost-plus contract. The contractor in turn hires subcontractors, which may also hire subcontractors. All of those "subs" operate under firm-fixed price contracts. "Some contracts have so many subs that by the time the funding reaches the sub-sub-sub contractor, there isn't enough $$ left to actually pay for what work is required," the provincial reconstruction officer wrote to Spectrum.

But never mind that for now. The basic problem is at the top: "In effect, USAID depends on contractors to help it with contractors," says Tiefer, the attorney and Wartime Contracting commissioner.

"USAID lets contracts which it essentially doesn't have the capacity to supervise or even to understand," adds Boston University's Barfield.

Under a cost-plus contract, the prime contractor keeps tabs on all of its expenses for everything associated with the project and bills the government agency for those expenses. In addition, the contractor is given a margin—typically 8 percent—as profit. The contract may also specify bonuses for goals; for example, if all the deadlines are met, the profit might be greater than 8 percent.

The problem with this setup is obvious: The more a contractor spends, the more money it makes as "profit." Cost-plus contracting was used in Iraq and was harshly criticized in at least half a dozen government reports.

A firm-fixed price contract, on the other hand, is what it sounds like: The subcontractor agrees to accomplish a task by a certain date and for a certain fixed sum. If the company encounters major and unforeseen problems, renegotiation is possible, but there are no guarantees that the company will get any financial concessions.

Prime contractors have argued that the vagaries and exigencies of war-zone reconstruction make cost-plus contracts a necessity. But that argument fails to account for the scores of subcontractors that have managed, in the face of the very same vagaries and exigencies, to stay in business under fixed-price contracts.

On a humid afternoon in April, I visit the Shorandam Industrial Park, east of Kandahar city. More than five years ago, a contractor hired by USAID tried to install ten 660-kW diesel engine-generators here. But the agency got into a dispute with the contractor and later claimed that some of the engine-generator sets were damaged by the blast of an improvised explosive device. The units went into storage, but now USAID has hired another firm—Black & Veatch, as it happens—to install them.

A U.S. military engineer shows me the 10 concrete pads, one for each generator, that USAID's original contractor had installed. The pads are laid out in a line with about 2 meters of separation between pads. Given the local climate, the engineer tells me with a shake of his head, the spacing between the pads should be at least twice that, to ensure enough airflow for adequate cooling.

Four months later, in early August, I get word from Kandahar that Black & Veatch has at last installed the diesel generators—on the very same closely spaced concrete pads that had been laid out by USAID's previous contractor.

There's another, brand-new diesel-engine facility here; it is one of the 10-MW plants put in recently by the Army Corps of Engineers. The Corps's contractor, IAP Worldwide Services, needed just four months to install it. I'm impressed but also puzzled. Why have two different U.S. government agencies hatched plans for two different diesel generating facilities in the same location but with different transformers, switches, contractors, and manufacturers? Nobody seems to know.

The day before, a NATO development official handed me a confidential document spelling out USAID's intention to put DABS, the Afghan utility, in charge of not only the huge SEPS effort but also NEPS, the parallel and similarly mammoth enterprise in the north. The utility would, at least nominally, control $906 million of the $1.2 billion budget for the projects. According to the document, USAID would act as a sort of supervisor and intermediary between DABS and the Afghan Ministry of Finance.

I'm stunned. Nothing I've seen so far suggests that DABS is anywhere near capable of managing the work of dozens of contractors and subcontractors on a series of huge and complicated infrastructure projects. Over the next several days, I speak with half a dozen development experts not affiliated with USAID, and all of them think it is a terrible idea. "It's almost like we're setting them [DABS] up for failure," one of them says.

I ask the NATO official why USAID wants to give DABS more control, and the answer is, in a word, politics. The agency is under pressure from its State Department overseers to show evidence of progress before the troop pullouts are well under way. But at the same time, the administration of Afghan president Hamid Karzai has been insisting that it be put in charge of more of the budget and agenda for reconstruction. (Never mind that of the 20 to 25 percent of development funds given to the Karzai administration to manage last year, it was able to spend only about a third, according to the NATO official.) Giving DABS and the Afghan Finance Ministry control of the huge NEPS/SEPS project would seem to accommodate both parties.

But Ahmad, who is DABS's only degreed electrical engineer in Kandahar province, is skeptical of the plan. As a senior executive of DABS and power director for this part of Afghanistan, Ahmad would be in the thick of any supervisory role DABS would have of the NEPS/SEPS effort. So in a meeting here at Shorandam, I ask him about NEPS and SEPS. He shakes his head dismissively and says it would take more than 10 years to complete these huge initiatives—too long to help Afghanistan in its current precarious condition.

He'd much rather see the money and effort go toward building more small hydroelectric plants. "What we need are microstations, to have our own power," he says, before reeling off a list of rivers and locations in Afghanistan that he feels are ripe for hydroelectric development.

As a reporter, I have over the years dug up the occasional isolated and carefully concealed incident suggesting incompetence or wrongdoing. But I've never reported on anything quite like USAID in Afghanistan, where the examples of ineptitude, poor decisions, and apparent impropriety sometimes seemed to come swarming at me like targets in a video game.

And the problems and failures aren't limited to USAID and to electricity. "The problems are significant; widespread; and, in my opinion, insurmountable given any agency that currently exists," wrote the provincial reconstruction officer in his letter to Spectrum. "I went, I saw," he added. "Every day I was there, the problem became more complex—like the insect in the spider web—turn and twist, looking for a way out, only to make matters worse."

All the while, year after year, the Bush and then Obama administrations let it go on. "I'm a little surprised," says Boston University's Barfield. "Why hasn't anybody in the White House sorted this out?" By now, he says, somebody should have given USAID an ultimatum: "If you can't do it, we'll find somebody who can." But he doesn't expect it to happen. "I don't see that kind of political will," he says.

Does it matter? Yes, it does. Put aside for the moment the $53 billion (including security costs) that the United States made available to reconstruct Iraq and the $55 billion it has spent on Afghanistan. Consider instead the hundreds of people who died trying to bring modern infrastructure to the people of those two countries, according to the website Icasualties.org. It compounds the tragedy that some of them died in the service of something as poorly run as USAID's projects in those countries.

Something else perished as well: A rare and potentially momentous opportunity was squandered. Counterinsurgency (PDF) is complicated and messy and hard, as I was told over and over again in Afghanistan. As theories about counterinsurgency were endlessly debated and tried out in Iraq and Afghanistan, some unconventional ideas were proposed and put to the test. None was more radical than the proposition that helping ordinary people become more comfortable and productive could be as valuable, in military terms, as killing bad guys.

Was that proposition right? We will never know for sure.