Behind a rusted blue gate, inside a dilapidated building, next to a disorganised heap of files sits an equally dishevelled middle-aged man. He has bloodshot eyes and a thick film of sweat on his forehead. Peshawar can experience uncomfortable autumn humidity and further raising the temperature in his small office are four irate people, assailing him at the same time.

The man’s name is Muhammad Jamal and he is one of the many executive engineers (XEN) at the Peshawar Electric Supply Company (Pesco). He oversees linemen, metre readers, grid station workers and subdivisional officers (SDOs) in the area under his charge. That means when you are extremely unhappy with the people involved in the provision of electricity – which in Pakistan happens to be all the time – he is the person you visit to complain. What he doesn’t handle is policy. That comes from the top.

One of the complainants in Jamal’s office is a woman called Amina Bibi. She thinks she has been overcharged on her previous month’s bill. A widow and the mother of three children, she works as a seamstress and wants to know why her bill has made a sudden leap in numbers. In a subdued tone, the XEN tells her the electricity price has gone up. There is nothing he can change. If it was a fault in the metre reading, he would be happy to oblige but, as she herself admits, the metre reading is correct.

Most of the electricity being used in shops and small businesses is through direct connections, called the kunda system. These direct connections steal electricity by attaching metal hooks to transmission lines, bypassing the need for a metre.

Another complainant is a plump man with a thick, oily moustache representing an entire neighbourhood in Nishterabad area. He says the electricity transformer blew up in his neighbourhood a week ago and no replacement is in sight. Jamal gives him a sullen look. There is a dispute over how the transformer broke down. The area’s SDO has written to the XEN that hooks were attached to the transformer to illegally provide electricity for a wedding.

“Now you tell me what to do,” Jamal addresses the man. “To save a few 1000 rupees on renting a generator, you end up overloading our transformer and costing us tens of thousands of rupees in repairs.”

The last two complainants, a pair of young men, remonstrate from the waiting benches. They are loud and agitated. At the first mention of the word press, they start addressing me instead. The language changes fluidly from Pashto to Urdu, sensing that I am not a native speaker.

There is a flurry of words and it is hard to figure out who has been wronged and how. The more vocal of the two men shows me a bill with a cascading list of arrears. He says his area’s SDO is trying to rob him. He makes 15,000 rupees a month, so how is he supposed to pay this 50,000-rupee bill? He has been here all day and the XEN did not listen to him even once. It is important that the press at least raise its voice for the common people, he pleads.

Then he switches back to Pashto and starts admonishing the XEN who responds in a terse one-liner: “You haven’t been paying your bills”. When the complainants leave, Jamal rises from his desk and unfolds a long piece of paper in front of me. “Look”, he says.

In the last 13 months, the young complainant paid only a single bill. His electricity has not been cut because he gathers a mob at his house every time the SDO sends linemen over. The XEN says he is willing to make concessions, to waive some fines, but the complainant only wants to pay the last month’s bill. He will go to court and argue the metre reader has been fudging the numbers and he will likely get a stay order — the Pesco cannot recover about 15 billion rupees worth of bills annually because of stay orders.

This is the daily life of an XEN.

In an already existing culture of politicised appointments, the dissolution of Wapda, says Cheema, felt like a funeral procession of merit and qualification. When he last dealt with the board of directors at Lesco, only two out of the nine directors were engineers — the rest were retired bureaucrats and businessmen.

In many similar offices across Peshawar, the wall behind a XEN’s desk is covered with whiteboards detailing recovery figures from the area under his charge. It is a key policy instrument of PESCO’s new management to overturn a reputation for poor recoveries and high line losses — the main reasons why the company incurs annual losses of more than 15 billion rupees.

In the new energy policy formulated by the ruling Pakistan Muslim League–Nawaz (PMLN) in 2013, one of the main highlights is the problem in recoveries. The government has come up with the idea of performance-based contracts for the heads of the distribution companies (Discos). These are short-term appointments which can only be extended if recovery targets are met.

While some areas in Khyber Pakhtunkhwa are genuinely tricky to recover electricity bills because of security concerns, recoveries remain staggeringly low even in relatively safe urban centres such as Peshawar. In October, 2014, when I was entering the city from the M5 Motorway, the first thing to greet me beyond the toll plaza was an enormous billboard discouraging electricity theft. This social awareness campaign extends to television and print media as well.

A recent report published by the Institute of Public Policy, at the Beaconhouse National University, Lahore, puts the economic cost of load shedding – and the fall in industrial/agricultural production and commercial activities that the lack of electricity causes – at an estimated 1.4 trillion rupees a year.

The last time Pesco workers went to Hazar Khwani, Kalashnikovs came out. Fazal Elahi, a local member of the provincial assembly, had to step in to broker a deal. The residents were offered a flat rate — 700 rupees a month per household for at least the next six months.

Once that period expires, further negotiations will ensue.

Conversely, the industrial feeders of Shabqadar and suburban areas such as Hayatabad have a good record for recoveries — generally around 80 per cent. Recoveries in commercial areas, however, are surprisingly poor. Most of the electricity being used in shops and small businesses is through direct connections, called the kunda system. These direct connections steal electricity by attaching metal hooks to transmission lines, bypassing the need for a metre.

People can either do it themselves or they can give a lineman about 4,000 rupees in bribes to do it for them — plus 2,000-3,000 rupees a month to ignore the transgression.

This overloads the system. A transformer can only cater to a specific number of electricity connections. Unaccounted connections put an additional burden on it, higher than its capacity to handle, leading it to trip — sometimes sustaining permanent damage. This is what causes long bouts of unannounced load shedding — the irony being that the same people who overload the transformer, then complain, often violently, about the lack of electricity.

Three years ago, police arrested some 30 protesters for attacking a Pesco office and attempting to destroy official records during load shedding protests in Tank. None of them were officially connected to the grid.

These problems are not specific to PESCO. Four years ago, an XEN at the Lahore Electric Supply Company (Lesco) was shot dead in Lahore’s Township area over a billing dispute. Earlier this year, two workers of the Multan Electric Power Company (Mepco) were killed by a mob over a disconnection issue. The offices of the Faisalabad Electric Supply Company (Fesco) have been ransacked in protests over load shedding so many times that the officials there wonder if they should move to Sargodha.

In Karachi, K-Electric found the towns of Orangi and Gadap so unmanageable that it outsourced billing in these areas to local power brokers: small businessmen, traders, transporters, people who have influence in the community and enough muscle power to make collections.

This is the story of electricity in Pakistan — a daily struggle between suppliers and consumers. Heated words, abuses, lamentations, even lyrical verses dedicated to candlelight and mosquito bites — there is a growing culture centred on the lack of electricity with its own rituals and idioms.

A recent report published by the Institute of Public Policy, at the Beaconhouse National University, Lahore, puts the economic cost of load shedding – and the fall in industrial/agricultural production and commercial activities that the lack of electricity causes – at an estimated 1.4 trillion rupees a year. This is a conservative estimate because it does not take into account the missed opportunities of investment that result from an unreliable power supply. It, however, still comes to about seven per cent of the Gross Domestic Product (GDP). The report further estimates that load shedding cuts one to two per cent from the annual GDP growth rate — in plain words we are talking about hundreds of billions of rupees worth of production, services and jobs which are essentially lost every year.

Institutionally, things are complicated. The national grid no longer belongs to a single entity, often making it hard to decide who exactly to abuse. It cuts across a morass of bureaucratic organisations which collude and collaborate but also undercut and undermine each other.

There was a time when things were simpler, when all of this was done under the umbrella of a single entity — the Water and Power Development Authority (Wapda).

Tahir Basharat Cheema is a career Wapda man. He talks quickly and brusquely, with little patience for formalities. He started his career in what he calls a different world, almost 40 years ago, as a grade 17 SDO right after completing his graduation in electrical engineering. With slow but steady vertical unbundling (the intimidating technical term for dividing generation, transmission and supply among many entities) of Wapda post-1998, he moved to Pakistan Electric Power Company (Pepco) and eventually retired as possibly its last managing director.

Formed to oversee Discos, Pepco is in the process of dissolution. Each Disco – now with its own chief executive officer and board of directors, as well as directorates for human resources, finance and assorted other things – wishes to be independent. That will formally be the end of Wapda’s power wing. The Pepco office that lingers on in the imperiously statured halls of Wapda House in Lahore will presumably be shut down for good.

For many from the old guard, this unbundling at the behest of the World Bank – which brought it in as a conditionality to Nawaz Sharif’s last government in 1997-1999 for a 750-million-dollar aid package meant for improving the energy infrastructure – is somewhat akin to a theatrical tragedy, and the root cause of a lot of the power ills.

The Power Wing, at its peak, directly employed 125,000 people, about 80 per cent of Wapda’s total staff. This huge number of people had to be distributed, initially, over 12 organisations including three generation companies (Gencos), eight Discos, and a transmission company, the National Transmission and Dispatch Company (NTDC). The number of such organisations has since increased.

Setting up independent administration and bringing in directorates from outside at each of these small organisations was never going to be easy. In an already existing culture of politicised appointments, the dissolution of Wapda, says Cheema, felt like a funeral procession of merit and qualification. When he last dealt with the board of directors at Lesco, only two out of the nine directors were engineers — the rest were retired bureaucrats and businessmen. It is a similar story across the board.

Also problematic was the division of Wapda’s research and training centres, such as the Wapda Engineering Academy in Faisalabad. When it was a single organisation, Wapda-nominated employees from all over the country to undergo training at these centres. Now the Discos haggle over staff allocations and funding for courses. It is no surprise, then, that the total number of courses offered at the Faisalabad academy has dwindled from 1,500 in 2004 to 700 in 2013.

Fewer courses mean fewer up-to-date engineers. In the good old days, training courses prepared a whole crop of engineers who could handle the most complex technologies. Take, for instance, the bulb turbine technology which made power plants such as the one at Chashma Barrage possible. It was challenging to produce electricity through this highly sophisticated technology but, some 25 years ago, it became possible because Wapda had engineers specifically trained for the purpose.

When he joined Wapda in 1975, the organisation was already politicised...This was Ayub Khan’s era, when a new class of ambitious civil servants was on the ascendancy.

Monitoring was also easy in an integrated organisation — the Wapda monitoring teams were not beholden to local electricity boards, predecessors to the modern day Discos. It now happens internally at the Discos which do not want to share their reports with Pepco’s central monitoring desk because poor performance by lower grade workers reflects badly on the management. There is a general feeling among Pepco officials that corruption and electricity theft have become worse, not better, since Wapda’s dissolution.

For Cheema, corruption is only one, insignificant, problem among many from which the power sector suffers. “There is no more corruption in the power sector than there is in any other industry. It is not the reason for the near collapse of the national grid. Incompetence is.” Coming from a man who was lavished with a Pride of Performance Award in 1991 for his role in catching electricity theft, this sounds like an indictment. For him, the fallout of corruption is manageable, the damage wreaked by mismanagement and poor policy is not.

When he joined Wapda in 1975, the organisation was already politicised. Many lucrative appointments and postings opened up after the Indus Basin Project began, resulting in megastructures worth billions of rupees, such as the Mangla and Tarbela dams. This was Ayub Khan’s era, when a new class of ambitious civil servants was on the ascendancy.

“The 1960s was the age of the generalists. You could put a bureaucrat with a humanities degree in charge of some engineers and get a lot accomplished. But the world has moved on,” says Cheema. He calls the 1980s the age of the specialists – engineers around the world were entering management positions – but Pakistan’s bureaucracy was not willing to concede the space it had carved for itself.

Things were made worse by the ministry of water and power. Politicians were beginning to realise that the electrification of rural areas had a significant impact on votes. Pakistan’s rural electricity coverage far exceeds its neighbours. According to a 2013 World Bank report, titled Global Tracking Framework, 60 per cent of Pakistan was connected to the national grid in 1990, a figure that rose to 91 per cent by 2010. In India, the same report said, electricity coverage was 51 per cent in 1990 and 75 per cent in 2010.

These numbers aren’t meant to be flattering. Pakistan has been overstretching its grid infrastructure. Rural electrification has not happened as a result of holistic improvement in the system, putting additional burden on the existing grid infrastructure which sees little proportional improvements.

In order to ensure the system of political patronage remains intact, Cheema says, the ministry is always trying to micromanage a technical industry it does not fully understand. Its interference does not stop just at managerial appointments but also goes deep down to the transfer, postings of even metre readers and linemen.

An anecdote may illustrate how. A director at the Wapda House, the ministry’s headquarters in Lahore, once wanted to get in touch with a member of the National Assembly (MNA) from Bahawalpur. He called some bureaucrats in the provincial government who, as these things go, called some other bureaucrats until a lot of telephones in a lot of offices chimed in to ask the same question — yet nobody seemed to have the parliamentarian's telephone number. Since the matter was urgent, the director became a little creative and called the XEN in the legislator’s home town and asked him to immediately suspend the senior most metre reader there. Within an hour, the MNA was on the office line asking why the poor metre reader was being sacked.

“Unbundling was not the answer in these circumstances. It has just made political patronage easier and produced more seats to be filled by cronies,” says Cheema. “The world is moving back to vertical integration [which brings generation, transmission and supply under a single chain of command]. Power utilities in developed countries, which were decentralised many years ago, are now lobbying for mergers and takeovers,” he adds.

This is due to the fact that electricity cannot be stored. Generation and supply have to be synchronised — if one is unreliable, the other will inevitably suffer. “Companies don’t want to buy and sell electricity like an intermediary. They want to make their own generation decisions,” says Cheema.

This is in stark contrast to what international donors were advising in the 1990s. When the World Bank was championing vertical unbundling, this was accompanied by Independent Power Producers (IPPs) — private companies which stepped in to generate electricity and sell it to the national grid.

It appealed to the government at the time. Electricity shortfall had reached a historic milestone of 2,000 megawatts by 1995 due to increasing demand but there was little appetite for more public spending on power projects. The IPPs seemed like a perfect fit to fill this investment void. Soon this led to another type of private power producer.

Cheema was still serving as a Pepco director when the Rental Power Plant (RPP) fiasco unfolded. These plants were meant to be short-term IPPs licensed for a period of five years to address the acute electricity shortfall the country was facing in the late 2000s. These plants could be commissioned within six months.

The government started issuing licences for RPPs in 2008. By the next year, 19 RPPs with a collective capacity of 2,800 megawatts had been commissioned. By 2012, however, they were producing only 120 megawatts of electricity. The Supreme Court of Pakistan officially decommissioned a lot of them after a suo motu hearing, calling the entire RPP scheme a mega scam.

When the apex court ordered registration of corruption cases against the then minister for water and power, Raja Pervez Ashraf, for alleged embezzlement of 22 billion rupees, Cheema and many of his colleagues were likewise embroiled in the fallout. The Supreme Court’s ire came down on everyone in the power sector for letting the RPPs go ahead. In his own words, it is the politicians who make the decisions but the engineers who have to answer for them.

The ministry is held in reverence because it appoints the chiefs of the Discos. It also appoints people at Pepco who, in turn, handle all senior-level postings at the distribution companies. The Discos have jurisdiction over employees only in grade 18 and below.

People in the power sector say Federal Minister for Water and Power, Khawaja Asif, is largely an absentee minister. Decisions are made in the Prime Minister’s Office, up until recently on the advice of his energy adviser Dr Musadik Malik — referred to as ‘pharmacist’ by the Wapda engineers for his lack of credentials relevant to the power sector.

Once the decisions are made, the contentious state minister for water and power, Abid Sher Ali, acts as the enforcer on the ground. This is, roughly, the policy mechanism that led to the disaster at Nandipur where the government’s much advertised new thermal power plant ran for only five days and produced electricity at the rate of 42 rupees per unit.

When I wanted to visit the plant a couple of months ago, I had trouble finding the place. The directions were simple enough; it was meant to be dab in the middle of the main road that connects Gujranwala with Sialkot but all I could see was rubble. About 20 minutes of confusion later, I was told by a security guard sitting next to a mound of dirt that the rubble was the site.

There, however, is a plant inside. Encircled in dirt, construction cranes and empty office spaces, there are boilers attached to an intricate network of pipes and wires, to supply fuel to the machines and to take away the generated electricity. These contraptions have still not reached a Commercial Operations Date — by when an engineer is satisfied that all tests have been run and the machinery is ready to be used. There is also no means for treating furnace oil on the site, a consequence of the dispute between the government and the project’s original Chinese investors, Dongfang Electric Corporation, which became uncertain of the payment plans midway through the project and left. The investor’s departure had left much of the plant’s machinery marooned at Karachi Port for a good three years. The project was initially to cost 23 billion rupees but has hit more than double that amount. The machines are now run on diesel which produces electricity at a record cost.

When Prime Minister Nawaz Sharif decided to go ahead with the inauguration of the plant in May this year despite all the problems the project was facing, the decision was purely political. This, officials say, would never have happened with an engineer in charge.

At policy level, the International Monetary Fund (IMF) is also pressing the government to reduce the difference between what the Discos want to charge from the consumers and what the government feels the consumers are able to pay — industry insiders call it a tariff differential subsidy. The government pays huge sums of money to the Discos, 150-200 billion rupees a year, to keep electricity affordable.

There are problems with the subsidy. While lower income groups get a higher percentage of subsidies, in real terms, the amount of subsidy spent on middle-income consumers is much bigger. For example, when people consuming 100 units a month get two rupees off on every unit of electricity they consume, they are getting a 50 per cent subsidy; when people consuming 300 units get 4.5 rupees off on every unit of electricity they consume, they are getting a 40 per cent subsidy. The actual government spending on the latter completely dwarfs the one on the former.

Moreover, the share of electricity from the national grid is determined by the average demand at each Disco, not on how it performs, for instance, on recovering bills. This means that the federal government is inadvertently subsidising the worst performing Discos the most.

In 1980, one Bashiruddin Mahmood of the Pakistan Atomic Energy Commission recommended using jinns. He argued these fiery beings were likely constituted of some form of gas that could act as a free and infinite source of fuel to power turbines.

At the moment, the Discos include their poor recoveries in tariff determination by the National Electric Power Regulatory Authority (Nepra). This means that Nepra, while setting prices, compensates a Disco for its inability to recover money from its consumers, passing the burden onto the paying consumers and the government.

The IMF wants Discos to be financially responsible and improve their performance. As with the World Bank before, energy-related loans are directly tied to these demands.

In order to improve performance, the Discos say they need a bigger workforce and newer technology. Most Discos quote a workforce shortage of 5,000 -7,000 people. They also cite the poor condition of 132 kilovolt and 66 kilovolt transmission lines which run within towns and cities. Some of these lines were laid out in 1965. Technical losses – the electricity lost due to poor insulation, worn-out wires and conductors – are linked to the age of the transmission lines which in extreme cases are no longer safe to work with.

Changing these lines requires money which the Discos do not have, the government cannot find and the consumers certainly cannot pay. Adding supply losses to the tariff determination equation, which currently only reflects generation and transmission costs, will spike electricity rates a great deal. It will be the sort of unpopular political move that could end a government quicker than any rally or sit-in.

As a company secretary at Lesco explains to me, if a Disco has three million consumers, half of those will fall under the category which consumes less than 50 units a month. They pay a nominal electricity price, well below the average cost of production. But they will still want a 24-hour supply to their homes.

That, then, becomes a question about the nature of the organisation. If a Disco is a government-owned company then the government should subsidise these consumers. If it is a private company then it will not be bothered about consumers — not at the prevailing prices.

One obvious solution would be to reduce the average cost of production but that will require correcting Pakistan’s energy mix (the breakdown of electricity according to the sources it is coming from). Forty years ago, Pakistan’s energy mix was roughly 70 percent hydroelectricity and 30 per cent thermal power. It stands reversed now — at 33 per cent hydro-electricity and 61 per cent thermal power. Out of the latter, 21 per cent electricity is produced with gas and 40 per cent with furnace oil — which is one of the costliest possible inputs. The remaining six per cent is being produced from nuclear sources and coal.

Government reports on the energy mix state that more than 40 per cent of the world’s energy is produced from coal whereas in Pakistan coal-produced electricity is at a masochistically impressive one per cent. Finding cheaper sources of electricity is a long-standing problem that has seen some innovative suggestions at times. In 1980, one Bashiruddin Mahmood of the Pakistan Atomic Energy Commission recommended using jinns. He argued these fiery beings were likely constituted of some form of gas that could act as a free and infinite source of fuel to power turbines.

Back in the world of sanity, however, policy recommendations still focus on hydroelectricity projects.

As you drive along Tarbela Road hoping to catch sight of the world’s largest earth-filled dam on the horizon, you come across a small town called Ghazi first. Situated just inside the border of Haripur district in Khyber Pakhtunkwa’s Hazara division, Ghazi’s modest bazaar offers fish, boat rides, maps and a rooftop diner advertising a heart-winning view.

All tokens to Tarbela’s now somewhat curtailed tourist appeal. Security concerns have turned the dam into a garrison. Last year, militants fired rockets on one of the grid stations, at Sheikh Muhammadi, Peshawar, which is directly connected to the dam.

The sight of the dam is still majestic, a raging river calmed and made as still as a puddle. The water is piercing blue, an effect of the clear sky above, I am told.

Tarbela’s power station is nestled under the right bank of the dam. Its 14 generators produce a total of 3,478 megawatts of electricity, making it the single largest producer of electricity in the country, accounting for 15 per cent of the total generation capacity that Pakistan has. A fourth extension project is currently underway, to add production capacity for another 1,400 megawatts.

This is also the cheapest electricity produced, at an approximate rate of 1.5 rupees per unit, compared to eight rupees to 10 rupees per unit for gas-powered turbines and 20-24 rupees per unit for electricity produced with furnace oil. It, therefore, comes as a surprise when the authorities say water is their primary concern and energy is just a by-product.

Wapda has proposals for setting up projects which can produce 60,000 megawatts of hydroelectricity — from sites on the Indus, Swat, Kunhar and Chitral rivers. But they are going to take billions of dollars in investment and long construction times.

“The Indus River System Authority (IRSA) gives us the indent of water level that must be maintained at Tarbela. We follow it religiously,” explains Engineer Riaz Hussain, a senior Wapda official at the dam. The indent provided by IRSA, which is responsible for overseeing water reservoirs, link canals and headworks, is made keeping in mind the amount of water that is to be stored at a reservoir based on irrigation needs and provincial water sharing agreements.

At the time of my visit in late September, Tarbela was producing 2,525 megawatts. Some of the generators were offline due to a higher than normal indent — more water was being stored and less let out. The monsoon and flooding season is a delicate time for the reservoir. There is a dual incentive to store water, both for the winter crops and to avoid flooding downstream.

Not all hydro-electricity plants are attached to water reservoirs. Hussain tells me of the terrific success of the Ghazi-Barotha project, just a few kilometres south of Tarbela, producing 1,450 megawatts of electricity and built at a quarter of Tarbela’s cost (adjusted for inflation). It is a run- of-the-river project in which water rushes at a great speed through a diversion tunnel to run the power turbine set up at its end, before coming back into the river. Ghazi-Barotha involved minimal resettlement of people and has created no irrigation or provincial water sharing disputes — all huge problems in the construction of other hydropower projects. But run-of-the-river sites are harder to find; they often need an upstream reservoir to begin with.

Wapda has proposals for setting up projects which can produce 60,000 megawatts of hydroelectricity — from sites on the Indus, Swat, Kunhar and Chitral rivers. But they are going to take billions of dollars in investment and long construction times. They are not as politically expedient as thermal power projects, which can be functional in three years.

Take, for instance, the Neelum-Jhelum project. It was commissioned in 2008 and is still not complete. Located in Muzaffarabad district in Azad Jammu and Kashmir, it is being delayed by a complex set of reasons, one of them being Pakistan’s challenge in international courts to the Kishanganga Hydropower Project in Indian-administered Kashmir. The Indian project involves diversion of water from the Neelum river to another tributary of the Jhelum river, causing a 27 per cent water deficit for the Neelum-Jhelum project.

Similarly, Diamer-Bhasha Dam, where construction was to begin in 2011, is now being advertised as part of Wapda’s 2025 Vision Plan, all but confirming that the project will take at least another decade to complete. Delays have already increased its estimated cost from 11.2 billion dollars to 14 billion dollars.

After drawn-out resettlement deals for the people that this project will potentially displace, it is now mired in an equally drawn-out argument over royalties. Both Khyber Pakhtunkhwa and Gilgit Baltistan are bickering vociferously over the ownership of the land where the powerhouse for the project is to be located since it is that location which determines who gets royalties from power generation.

The World Bank, a creditor for the project together with the Asian Development Bank, says Diamer-Bhasha Dam will also require a No-Objection Certificate from India. That is because most of it is to be situated in Gilgit-Baltistan which, according to the United Nations resolutions, is part of the state of Jammu and Kashmir that remains disputed between India and Pakistan.

Big water projects can be intrusive and give back little to the community they intrude on. The people of Ghazi, despite being wedged between two megaprojects, experience severe power outages. The 5,000 megawatts produced around them go directly to the national grid and only a tiny part of them comes back.

This is also a problem with the coal-based Pakistan Power Park in Gadani, a coastal settlement in Balochistan, for which money is currently being arranged through various investors. While the location of the 6,600 megawatts project – consisting of 10 powerhouses of 660 megawatts each – is suitable for reducing the cost of transporting the imported coal, the power it will generate is going to go largely to Punjab. As the Sindh government pointed out earlier this year in an official letter, the pollution caused by the project is going to badly hit the coastal settlements and Karachi, which is only a few kilometres away.

The muddle of regional disputes and long construction times required for hydroelectricity projects is what mainly led to the ready acceptance of the IPP model. This is also a reason why the Private Power Policy initiated in 1994 was such a vague paper. Weary of more delays, it did not specify the type of technology, fuel or location for the IPPs which, therefore, vary wildly in terms of generation costs and output capacity. The government also promised to purchase electricity generated for the next 30 years with sovereign guarantees on payments — this means the government will pay to the producers in case NTDC is unable to.

Some Wapda officials say the other purpose of the vague policy was to facilitate commissions and kickbacks in issuing commercial licences for the IPPs. Licenses were issued to anyone who applied for them — even textile mill owners became power producers.

Incidentally, Wapda was not made a stakeholder in the 1994 power policy. Among other things, this meant that Wapda’s thermal power generation costs began exceeding those of the IPPs due to a dearth of investment.

The Jamshoro Power Company Limited (JPCL) can be found on the Indus Highway in Sindh, a short distance from the city of Jamshoro. The driveway that leads to the administrative buildings is expansive enough to be mistaken for a runway. The buildings themselves are quite large, with four towering boiler chimneys providing a suitable backdrop.

It is only when you get closer that the signs of age and disrepair belie the spectacle from afar. Once inside, the obligatory portraits of Jinnah remove any lingering doubts that this is a government company. It is not easy getting in. Like everybody else in the power sector, the company is weary of bad press. (In my electricity travels, I am told by one public relations officer that he is not allowed to talk to the press, another asks me not to tell anyone he gave me contacts for senior engineers; in fact, he implores me to forget we ever met. At the JPCL, they don’t even have a public relations officer. The head of the human resources and administrative department, Jamil Ahmed Nizamani, is left to deal with the media.) Media scrutiny is even more of an anathema to thermal plants where generation is sporadic, prone to stoppages when payments pile up and oil companies refuse to supply fuel.

In my electricity travels, I am told by one public relations officer that he is not allowed to talk to the press, another asks me not to tell anyone he gave me contacts for senior engineers; in fact, he implores me to forget we ever met.

Nizamani is a veteran of the thermal industry. Before JPCL, he served at another generation company in Guddu. Then, following some of his colleagues, he went abroad for a while, to Saudi Arabia and Japan, both countries where thermal power is huge.

He comes from a village called Matli. The villagers see the engineering university in nearby Jamshoro as a gateway to a better life. Many of Nizamani’s friends became Wapda engineers — almost all of them are now working abroad. While most may have traded in their government pay scale for more zeroes on their monthly cheques, some have been exported on government recommendations as well.

The constant outflow of experienced personnel obviously leaves vacuums, perhaps not at the lower level where the influx of newer engineers is still adequate, but at the higher rungs of the power sector ladder.

The JPCL has four power generating machines — a 250-megawatt capacity Japanese machine and three Chinese ones meant to produce 200 megawatts each, all imported 23 years ago. The plant has not achieved its full operational capacity in the entire last decade because its machinery needs regular repairs. But funding is often not forthcoming. For five years during the Musharraf regime, maintenance funds were not released even once.

Having to run the turbines on faulty parts caused long-term damage and deterioration in generation capacity. By 2009, only one of the four turbines could be operated, producing 200-250 megawatts of electricity. Unrepaired turbines also use more oil for every unit of electricity they produce than the ones in good condition. In 2012, Nepra was alarmed by an “excessive use of furnace oil” at various Gencos. Nepra investigators wanted to call it embezzlement but the Gencos called it helplessness. Nothing was being stolen, everything was just being wasted.

Things would have been even worse if USAID hadn’t stepped in with the promise of major investment five years ago. A technical team led by Rustam Ali Ghauri, the engineer in charge of procurements at JPCL, gave USAID a financial assessment for the complete rehabilitation of the plant. The donor deemed that assessment too expensive — the best they could offer was “minimum replacements required to reclaim installed capacity”.

As the power sector’s fuel bill has reached a historical high of 238 billion rupees this year, PSO has already defaulted on some international payments.

This involved importing 23 new parts, still costing a steep 19.3 million dollars. The last part arrived this October. Once all of these parts are fixed, the plant will finally function close to its installed capacity for the first time in 12 years — that is, if it keeps getting its full supply of fuel. In some months, the JPCL ends up getting only a 1,000 tonnes of the 3,000 tonnes of furnace oil it requires.

That supply is linked to 500 billion rupees worth of circular debt, money that the Discos owe to the Central Power Purchasing Authority (CPPA), a subsidiary of the NTDC, which in turn owes that money to the power generators. When nothing else works, the water and power ministry has to pay some money to the power generators so that generation can continue and supply does not stop altogether.

In a bout of political triumphalism during its first year in power, the ruling Sharif government eradicated the circular debt but the paper chain around the power sector’s neck is now back to a figure of 577 billion rupees. Fuel payments – owed by power producers to Pakistan State Oil (PSO) – make up a big chunk of it. As the power sector’s fuel bill has reached a historical high of 238 billion rupees this year, PSO has already defaulted on some international payments.

In stark opposition to the Gencos, stands the Hub Power Plant in Lasbela, Balochistan. Run by a private company, Hub Power Company (Hubco), the plant employs a staff of 300 people — about one fourth of staff strength at the JPCL. Maintenance is outsourced to a multinational, GDF Suez — an expensive arrangement but one that keeps turbines humming healthily.

Like the Gencos, however, the cost of fuel used at Hub Power Plant is also high and the company has to deal with the same government entities which make delayed payments. Khalid Mansoor, Hubco’s chief executive officer, says these payments are the primary problem in the power sector.

The numbers, indeed, don’t add up.

Twice in three years, the IPPs have invoked sovereign guarantee. In 2012, when eight IPPs first invoked sovereign guarantees, the dispute had to be settled in the Supreme Court which ordered the government to immediately chalk out a payment schedule for 60 billion rupees. Two years later, in June 2014, sovereign guarantees were invoked again. Now 15 IPPs demanded the payment of 66 billion rupees. Finance Minister Ishaq Dar has assuaged the IPPs with promises of a timely repayment plan but promises about promises can’t engender patience and chances of future litigation remain high.

For the future, some monitoring solutions are coming from USAID. It is giving metre readers new handheld units that scan metres and automatically record readings.

Hubco, in the meanwhile, is weighing its options. Last year, it inaugurated Pakistan’s first private hydropower project, in Laraib, a modest run-of-the- river plant downstream from Mangla, producing 84 megawatts of electricity. For a company trying to move away from expensive fuels, it is an important step.

The company is also looking at the possibility of shifting its production to coal. Mansoor, who was also the first chief executive officer of the Sindh Coal Mining Company a few years ago, says Hubco is trying to make capacity building investments in coal mining. There, however, are obstacles — sometimes originating from the least expected sources. Western lenders, for instance, discourage the use of coal as part of their drive to reduce global carbon emissions.

At other times, the obstacles are predictably straightforward. Investors are extremely weary of being caught in Pakistan’s wash-rinse-and-repeat cycle of circular debt. They look at the system into which any potential electricity is going, and they don’t like what they see.

Multan is an annual host to extremely hot, and extremely long, summers.

Kamran Zahoor, an XEN at Mepco’s head office, is a soft-spoken but energetic man. He hops around from office to office delivering people’s complaints to the relevant officials. He says summer always brings charred transformers, precautionary load shedding and phone calls from people demanding he fire everyone at the facility, including himself.

During my visit to Mepco, an ebullient woman is sitting in Zahoor’s office. The metre installed in her home is not working properly and she has proof of erratic readings. Zahoor agrees with her and is just signing off on papers before the metre can be replaced. He thinks the heat must have damaged it.

She disagrees. In her opinion, the metre reader has been at it with his tools to coerce her into paying him bribes. She looks around the room conspiratorially and declares that these lower grade workers are all vultures. They feed off the vulnerable.

Zahoor smiles and shakes his head, telling her that the staff working under him are very straightforward. The woman laughs and says, yes, they are very straightforward: “Bilkul jalebi ki tarha seedhe.”

This is not an unpopular or uncommon opinion in the country. If you broach the subject of electricity supply in a casual conversation, you can be assured of being regaled with tales of a metre reader arriving on the job in a Pajero. That what a lineman really fears isn’t death, but a promotion. Desk jobs don’t offer the same opportunities for selling illegal connections. Then there is that well circulated tale of an MNA who once worked as a Wapda field worker.

The truth, however, may be a bit less sensational.

“What most people call corruption is just inhuman workload. Overbilling does not happen because the metre reader is looking for bribes; it happens because the metre reader never takes a reading.”

So opens my conversation with Gauhar Taj, the central chairman for the All Pakistan Wapda Hydro Workers Union. He says electricity theft and poor working conditions are two sides of the same coin. “High-loss feeders are always those that do not have a complaints office nearby, where the workers cite a lack of training and equipment, where there is a neglect of management.”

He cites the example of his own village. “When I joined Wapda in 1977, there was one complaint office and two transformers in my village. These were looked after by six people. Now there are 300 transformers in the area, and they are being looked after by three people.”

In her opinion, the metre reader has been at it with his tools to coerce her into paying him bribes. She looks around the room conspiratorially and declares that these lower grade workers are all vultures.

The union does its own research on staff shortages. If a worker does eight hours of honest work a day, he can do about 800-1,000 readings per month, depending on the area, reads the research. Likewise, it adds, a worker can reasonably be expected to distribute 1,500 to 2,000 bills per month. No more. The actual burden of bill distribution per worker is 10,000 per month and every metre reader is responsible for 7,000 readings a month. All of this leads to one thing: making up numbers sitting at a desk.

It leads to another thing too. If the workers do physically attempt to meet their job requirements, there is a high incidence of accidents, mutilations and deaths.

The union is headquartered at Bakhtiar Labour Hall, named after its founder, on Nisbat Road, Lahore. Bashir Bakhtiar was the leader of the Electricity Boards Union formed in 1935, headquartered at Simla in Himachal Pradesh. After Partition, he set up a new headquarter in Lahore and over the years it has become the single biggest workers organisation in the country.

The collective strength of the union is a source of pride at Bakhtiar Labour Hall. Pamphlets detailing the history all mention an incident in 1969 when, under the leadership of Bakhtiar, Wapda workers cut off electricity to the president’s house and the army headquarters, precipitating the formal end of Ayub Khan’s dictatorship.

The other printed material at the hall is more concerned with contemporary rulers, specifically Nawaz Sharif. A 2013 flyer cautions workers to remain vigilant and loyal to the union management as the federal government is back in the hands of the, ‘capitalist, right-wing, anti-union PMLN.’ The language is harsh — the PMLN leaders are likened to devils and rapists, but the fears aren’t entirely unfounded. The PMLN’s power policy is, indeed, privatisation-oriented and is planned entirely around foreign loans/aid.

That is why, even though Wapda workers are now spread over more than a dozen companies and organisations, they remain united in electing a single Collective Bargaining Agency (CBA) that has the right to negotiate with the management and the government on the workers’ behalf. It finds strength in numbers. If one worker goes on strike, the company can keep functioning smoothly; if 20 workers go on strike, the employer has a problem.

At the union’s Lahore head office, I am told that 67 linemen were killed in 2014 alone. Almost a similar number of workers have been disabled. Assorted injuries leading to temporary hospitalisation are comfortably beyond a hundred. The union has been running campaigns to raise awareness for workers’ safety, imploring linemen to disobey an order if proper safety protocols for climbing a pole have not been put in place. But when faults break out, complaints start stampeding in and there are only a handful of people to attend to them. Management often foregoes safety measures to save time and its own reputation.

In March this year, Pesco issued tenders for the sale of two Peshawar sub-divisions, Rehman Baba and Gulbela, suffering 90 per cent electricity losses, to private investors, somewhat similar to the franchising of difficult billing areas in Karachi. The union opposed the sale, as it would have meant an immediate curtailing and replacement of workers. The workers, instead, lobbied to be allowed to run the subdivisions.

Pesco ceded. By the end of May, in just three months, 90 per cent losses were converted into 90 per cent recoveries. Mustaja Mazdooryar, the provincial secretary of the workers’ union in Khyber Pakhtunkhwa, was in charge of the cleanup. “We replaced the old management and promoted people from within the workers. We made it a policy to treat consumers with respect. Anyone coming in with a complaint would be given tea and undivided attention,” he says.

The union called on the district police to help with non-paying customers. It also held internal inquiries and fired workers over corruption. Mazdooryar says in the first month a lot of ministers called asking what the union was trying to do. He would politely reply, “Just our jobs”.

Now, he says, Pesco’s administration is offering us more subdivisions. “We haven’t taken a decision yet. We are a union, not a company.”

For the future, some monitoring solutions are coming from USAID. It is giving metre readers new handheld units that scan metres and automatically record readings. The American donor is also giving Discos new metres to install — ones with SIM cards that send live readings from the metre to a central server. Any attempt to alter the readings will cause an automatic server enforced shutdown. This will end reliance on metre readers, or the SDOs, as primary data entry points.

USAID is on an extensive energy sector investment drive at the moment. You can hardly visit a Disco without running into its emblem somewhere. The heaviest investment is happening in Multan; officials there even drink coffee from USAID mugs.

The aid agency has set up a brand new power cell in Mepco’s head office. It is called the Power Dispatch Center and is equipped with five LCD screens mounted on a wall in front of a row of desks, showing live information coming in from grid stations, feeders and the new SIM-based metres.

The screens show total wattage being supplied and the power factor associated with each grid station, giving a rough indication of technical losses being experienced in transmission. They also show the total number of feeders, over 1,000 in Mepco’s jurisdiction, and their breakup, between domestic, rural, and commercial. It is the rural tube-well column that accounts for about 566 of the total feeders.

More importantly, the system makes it easy for the desk monitoring team to catch a fault well before anyone has called in to complain about a power outage. The USAID calls it the Load Data Improvement Project and it has been offered to every distribution company.

Some, like LESCO, have not taken up the offer. It does not feel monitoring is its biggest problem and is looking for a direct form of investment — in the company itself.

Which brings us to K-Electric, formerly known as the Karachi Electricity Supply Company (KESC). When the Private Power Policy was formulated in 1994 and the first steps taken towards both private power production and decentralisation of electricity distribution, KESC was separated from Wapda and, at the time, put under military charge. It was an interim administrative measure that, like all interim military measures, lasted a little too long. After 11 long years, in 2005, the government finally attracted a consortium of Saudi and Kuwaiti investors to buy 71 per cent of the company’s shares.

The initial investment was not a success story but it was a landmark moment for the company itself. Its fate was about to diverge drastically from the rest of the country’s, where direct foreign investment remains a distant dream. When the Saudi consortium sold its shares to the Abraaj Group in 2009, privatisation started paying dividends. Usama Qureshi, a deputy director at K-Electric, explains, “The 2005 takeover ended up a half measure. The investors had promised a 361 million dollar investment; 250 million dollars went to the government in actual buying of the shares. At the time, the remaining money was not sufficient to overturn KESC’s fortunes.”

The problem was compounded by other factors. “Consumer banking was at its peak then and electric appliances were becoming affordable; air-conditioning, for instance, stopped being a luxury and became a necessity. The demands on Karachi’s grid network were immense,” says Qureshi.

Abraaj came in with an additional 361 million dollar offer.

Since the subsequent takeover, K-Electric has added 1,000 megawatts to its grid, primarily produced from gas, while turning a 120-billion-rupees loss in 2005 into a 120-billion rupees profit by 2012. The company’s share price has shot up from 1.5 rupees per share to eight rupees. A surprising 58 per cent of Karachi’s feeders are now exempt from load shedding, the highest figure for any city in the country. In 2009, when the new administration took over, that figure was only 23 per cent.

K-Electric can now claim to be the only vertically-integrated power company in Pakistan. It generates about 2,000 megawatts of electricity on its own, although it still takes a further 650 megawatts from the national grid. But privatisation hasn’t come without its hitches.

Since the subsequent takeover, K-Electric has added 1,000 megawatts to its grid, primarily produced from gas, while turning a 120-billion-rupees loss in 2005 into a 120-billion rupees profit by 2012.

In 2011, the company had 18,000 employees on its payroll. Many of these were considered non-core workers, employed in civil works and assorted vocations such as gardening etc. The investors wanted to cut costs where they could so they offered a severance package to around 4,500 employees. Some of them took the money, others were unhappy and called in the union. An already sputtering grid network was paralysed by the ensuing strikes. Negotiations went on for weeks. Mostly fruitlessly. The company wanted to establish the writ of the new management, the union wanted to show its strength. Things inevitably got violent.

“During one protest outside our head office, the protesters started breaking windows and setting fire to our cars in the parking lot,” recalls Qureshi. One such car was put up on the roof of the entrance to K-Electric, as a reminder of what the company had gone through during its baptism for privatisation. The car has only recently been taken down.

The clash shows the less pleasant face of the union. (In Punjab, the union has often been accused of siding with corrupt workers in the event of inquiries. Solidarity is seen as an indispensable tool of negotiation with the management — solidarity, sometimes, at the expense of integrity.)

When the dispute did not end, the Sindh Government got involved and, as Qureshi says, the management was forced to restore the workers, “on gun point”. It was clear that privatisation had not brought the autonomy of decision-making that the management had expected, failing to provide them with an escape from the politicisation that the power sector faces in the whole of the country.

The federal government, at least, has a case for that. Despite being serviced by a private company, Karachi’s consumers get subsidised electricity as well. There would be an uncontrollable outcry if Karachi’s denizens have to buy electricity at non-subsidised rates while everyone else in the country keeps enjoying subsidies.

According to a 2013 report by the Pakistan Institute of Development Economics (Pide), tariff differential subsidy given to K-Electric is the biggest amount given to any distribution company in the country, approaching 50 billion rupees a year. This is because, price-wise, K-Electric acts like a private utility provider, adding the cost of supply and its human resources to tariff determination — unlike the Discos. The higher price necessitates a bigger subsidy. So some of the profit K-Electric is showing on its books is being made off the federal government.

The five year agreement for the 650 megawatts that K-Electric gets from the national grid is coming to an end in 2015. With the belligerent state minister Ali already having locked horns with the company’s management over accusations of taking more electricity than its share, and the PMLN’s stronghold in Punjab suffering the largest electricity shortfall, a similar arrangement for the future looks unlikely at the moment. Karachi’s five years of living up to its city-of-lights moniker might be in peril soon.

Khurshid Ahmed is the current patron of the All Pakistan Wapda Hydro Workers Union. He is a white-haired, white-bearded man and not in the best of health these days. The 56 years he has been involved with the power sector and the union have taken their toll.

“Electricity is the hardest public work,” he says. On the day I visit him he is arranging compensation for three injured linemen in Lahore during the monsoon storms in August. He is also on the phone coordinating a strike in Sheikhupura to protest an attack on the Wapda workers.

As the leader of the biggest workers union in Pakistan, Ahmed has direct access to the Ministry of Water and Power. He says he has even talked to Prime Minister Sharif about the questionable wisdom of investing in new projects without improving the existing infrastructure and addressing the workers’ issues. “Say you add 10,000 megawatts to the existing system; that means more payments, more government guarantees; and when that output goes through a network with 25-30 per cent theft and losses, it means more circular debt,” he says.

Opinions on how to fix the energy crisis remain divided. Some, like Ahmed, believe adding more power to a grossly inefficient system will create new problems instead of solving old ones. Others argue in favour of an alternate micro-grid system consisting of small, community-based renewable energy projects. Give incentives to businessmen to invest in renewable energy and hook their plants up to smaller, more localised grids, thereby foregoing costly transmission losses, is how this argument goes.

Others still argue for a complete reversal to hydropower, or see Thar’s coal as a saviour. The National Power Policy released by the government last year has a bit of everything mentioned, in addition to a solemn promise to privatise the inefficient Gencos and Discos.

Pakistan’s maximum power generation capacity stands at 23,500 megawatts while maximum demand stands at 23,000 megawatts. Operational capacity, however, normally hovers around 16,500 megawatts, causing a shortfall of 5,500 megawatts. We can just about break even by rehabilitating the existing operational capacity to its maximum potential, but there are ill omens for the future.

Growth rate in power generation hovered around 7.5 per cent between 1970-2000 but under the Musharraf regime it sharply dropped to 2.7 per cent. While additions to the total installed capacity slowed down, demand did not; it has increased by six per cent since 2000. On current projections, it will hit a 30,000 megawatt figure by 2020. That is just six years away.

This just highlights the dual nature of Pakistan’s power woes — both extra capacity and a complete overhaul of the existing capacity are desperately required.

As winter falls on Lahore, the city is settling into a steady routine of eight hours of daily blackouts in the economically privileged suburban areas. In the parts of the city where bills are not paid as frequently as in the posh localities, the load shedding schedule can be as long as 12 hours a day. The drop in electricity usage during December and January means that annual maintenance of the grid network will be starting soon. This entails six hours of additional load shedding on any given day of the week.

And this is Lahore, the heart of Pakistan, that we’re talking about. Out in the great expanse of the country, there are places where people have to wait for five hours just to see an hour’s worth of electricity. In the next six years, if the government cannot find a way to reduce the number of hours of daily load shedding, it might just have to find a way to increase the number of hours in a day.

Originally published in Herald's December 2014 issue.