Pakistan’s current economic crisis has once again brought to the fore the issue of electricity subsidies. While on one end of the policy spectrum, economists argue that our fiscal position necessitates the withdrawal of subsidies, on the other end, there are strong advocates for keeping subsidies intact to keep electricity affordable and protect the masses from inflation.

The issue becomes even more dire once we realise that we cannot avoid raising electricity prices indefinitely. Over 65 per cent of Pakistan's electricity is generated using imported fossil fuels, which exposes us to international supply shocks and fuel price fluctuations, and there is no escape from that in the short term.

The costly policies on the generation side — for example capacity payments — make our electricity the most expensive in the region. Add to this mix the inefficiencies on the distribution side — high transmission and distribution losses and poor bill recoveries — further raise the cost of delivering electricity, and it is no wonder that the power sector remains stuck in a high cost-poor service equilibrium.

As economists who have been studying Pakistan’s power sector for quite some time now, we present an alternative to tackle the thorny issue of electricity subsidies, which is based on keeping electricity affordable for those who truly cannot pay for it but in a fiscally prudent manner.

We first argue that the current subsidy regime fails to meet its desired objectives and imposes an unsustainable fiscal burden and, therefore, must be reformed. It is important to communicate this message to the public in an easy and accessible manner, as broad consensus must be built around rationalising the subsidy system to help us move away from populist and simplistic rhetoric. We then propose a system of revamping electricity subsidies based on targeted in-kind transfers in the form of an electricity voucher programme.

What's wrong with the current subsidy regime?

From an economic perspective, the objective of subsidies is to correct under-provision of a socially desirable good or service by the market. If we consider access to electricity a basic necessity and want to ensure a decent level of consumption for every household, we want to have a policy instrument that enables us to achieve this objective, with minimal unintended negative consequences. However, the current subsidy regime fails to deliver on the objective of ensuring access to electricity for multiple reasons.

One of the most important flaws of the current subsidy regime is that it is poorly targeted and benefits richer households disproportionately.

Under the current system, residential customers are billed under an incremental block tariff structure as shown in Table 1. The tariff is increasing in consumption slabs ranging from 0-100, 101-200, 201-300, 301-700 and above 700 units. There is also a lifeline slab with less than 50 units of consumption.

The National Electric Power Regulatory Authority (Nepra) determines a tariff based on the cost of electricity generation and distribution. The Government of Pakistan (GOP) then notifies a consumer tariff which is lower than the Nepra tariff, hence subsidising the electricity price for end consumers. On each tariff slab, the subsidy rate per unit of electricity consumed declines as consumption moves up from one slab to the next.

For example, on the first 100 units, the subsidy rate is Rs6.85 per unit, for the next 100 units the subsidy rate declines to Rs6.35 per unit, and for the next 100 units, the subsidy rate further declines to Rs5.38 per unit. For the additional units consumed above 300, the subsidy is reversed and consumers pay a higher price than the Nepra tariff.

In Figure 1, we simulate the total bill amount based on Nepra determined tariffs (blue line) and GOP determined prices (green line) at all levels of consumption up to 1,000 units per month. The difference between the two is the subsidy amount in rupees that a household gets on their monthly bill (orange dashed line).

The subsidy amount (approximately Rs1,800) is highest for households consuming around 300 units. Figure 2 shows the subsidy amounts as a percentage of the total Nepra determined tariffs at each level of consumption. Although the subsidy percentage declines as consumption increases, due to the incremental pricing structure, even those with high levels of electricity consumption (and income) end up receiving a subsidy. In fact, all consumers consuming up to 933 units are receiving a subsidy.

Who benefits from the subsidy?

Using the nationally representative Pakistan Social And Living Standards Measurement household (PSLM-HIES) survey data, we find the share of households falling in each tariff slab. We then estimate the subsidy burden from each tariff slab using the average per unit subsidy, assuming an average consumption level per household within each slab. This exercise yields some important insights.

The largest group benefitting from the subsidy is households with consumption between 301-700 units, which accounts for 56pc of the consumers or nearly 16.8 million households. Assuming the average household in this group consumes 500 units per month, the total subsidy burden translates into Rs25.8b per month — out of a total monthly subsidy of Rs37.8b for all groups.

Under this structure, the annual subsidy burden would amount to Rs454b. This is almost twice the share of social protection programmes, which were allocated Rs255b in the 2021-22 federal budget. Thus, the current subsidy structure means that even well-off households benefit and in absolute terms, they are the recipients of the major chunk of the power subsidy.

Is the current subsidy system achieving its goal?

If the aim of the subsidy system is to ensure electricity access to the poor and low-income households, it most certainly isn't fulfilling its purpose. These households might receive subsidised electricity, but only when it is available from the grid.

In most parts of the country, electricity is highly unreliable with frequent load shedding undermining the goal of access. In fact, distribution companies practice segmented load shedding — load shedding hours are higher in areas with higher electricity losses and lower recoveries, which are invariably low income neighbourhoods.

Poor service delivery means dissatisfied customers who are not willing to pay bills, and might even resort to informal and illegal connections to meet basic necessities such as running a water pump. Meanwhile, the richer households invest in self-generation such as diesel generators or rooftop solar to insulate themselves from the frequent interruptions.

The electricity sector’s circular debt — inter corporate debt due to outstanding receivables of firms in the energy supply chain — is presently Rs2.5 trillion or 6pc of the GDP. One of the main reasons for circular debt accumulation is not allowing electricity prices to rise in line with rising fuel prices in world markets.

High operational losses and poor recoveries of distribution companies, which are linked to the demand side issues discussed above, also contribute to the circular debt. The electricity sector’s poor financial health means that we cannot afford to pay generation companies for costly electricity.

Plants stay idle even while we have available capacity. As a result, load shedding continues unabated. Moreover, loss-making distribution companies are unable to undertake adequate investment in maintenance and expansion of the grid network. Around 25pc of the population (nearly 50 million people) is still living without an electricity connection. Therefore, we must recognise that the current system is failing to achieve its objectives, even without accounting for the unbearable financial burden of the subsidies.

Thinking of alternatives

One policy option which has been advocated recently is to replace the present subsidy system with targeted cash transfers that poor households can then use for paying electricity bills.

The National Socio-Economic Registry, which covers around 85pc of the country's households, can be used to identify those below a certain eligibility cut off in terms of income or wealth. The biggest challenge of such a system could be that since cash is fungible, people might find it difficult to commit this money to pay electricity bills especially if they are due at the end of the month.

Instead of offering support through cash transfers, we propose a targeted voucher and rebate system. A voucher system offers a simple and non-distortive way to provide in-kind support to households for consuming electricity. Since the support will be applied directly to the electricity bill, it should be easier to build consensus around such a system as opposed to completely eliminating the current subsidy system.

In order to ensure that the support is offered only to deserving households, the government will first need to determine an electricity consumption threshold below which households are to be protected. For example, using the nationally representative PSLM-HIES survey data, we found that electricity consumption in the bottom two quintiles of income distribution was in the range of 250-300 units.

Using the National Socio-Economic Registry and customer’s electricity consumption history from the distribution company, we can define the following two categories of recipients:

• Tier 1: Households with monthly consumption below 200 units receive a voucher of 100 units free electricity and the next 100 units with a rebate of 30pc on the cost recovery tariff set by Nepra. This is equivalent to an effective subsidy of 50pc on every unit, which is higher than the 46pc subsidy received by this group on their monthly bills as shown in Figure 2.
• Tier 2: Households with monthly consumption between 200-300 units receive a voucher of 100 units free electricity and the next 200 units with a rebate of 15pc on cost recovery tariff set by Nepra. This is equivalent to an effective subsidy of 43.3pc on every unit, which is higher than the 38pc-43pc subsidy received by this group on their monthly bills as shown in Figure 2.

Households exceeding 300 units consumption in any month receive no subsidy on any unit.

We propose that the vouchers and rebates should be directly applied to consumer’s bills to avoid unnecessary delays in disbursement of such support. Voucher and rebate units can also be allowed to accumulate over time, for example, if the household’s consumption falls below the tier threshold in one month, the remaining tier units can carry over to the next month. In order to account for seasonal variations in energy demand, tier thresholds can be adjusted upwards.

In addition to providing financial assistance to eligible households, this targeted subsidy equivalent voucher scheme can have multiple positive cascading effects.

First, it will enhance bill payment because vouchers may only be used to pay for electricity bills, improving the utilities' revenue situation. Second, it will relieve fiscal pressure on other budgetary expenditures by reducing the government's circular debt burden. Third, it will encourage households to conserve energy according to their needs as exceeding the threshold consistently means losing the support. Fourth, richer households, who receive no subsidy, will be tempted to install solar as an additional energy source to supplement their energy consumption from the grid. Finally, given the dire state of the energy sector, a low-cost loan arrangement for installing solar systems can be proposed for middle-income households.

In a nutshell, a simplified electricity voucher and rebate system has the potential to achieve a number of goals, including improving electricity access, incentivising electricity conservation, boosting the adoption of clean energy sources and reducing CO2 emissions, while alleviating the financial burden of circular debt.

Header image: Adapted from Andrii Yalanskyi/ Shutterstock.com

The Analytical Angle is a monthly column where top researchers bring rigorous evidence to policy debates in Pakistan. The series is a collaboration between the Centre for Economic Research in Pakistan and Dawn.com. The views expressed are the authors’ alone.