Our 2 Cents

Corruption in the Indian Electricity Sector

Wednesday, December 4, 2019
ByMeera Mahadevan

Meera Mahadevan is a postdoctoral researcher at emLab.

India faces huge electricity shortages despite having enough generation capacity to supply to the entire country. In 2013, 20% of all electricity generated in India was lost due to transmission and distribution losses. There are a number of reasons for this including inefficiencies in power purchase contracts, poor infrastructure, electricity theft, and lack of universal metering. However, a major problem affecting the sector continues to be corruption, and more specifically politically-motivated corruption. The electricity sector is vulnerable to political influence at virtually every level, from higher level policy to the transmission (e.g. discretion over which areas are allowed to overconsume from the national grid), distribution, and retail levels (e.g. politicians rewarding voters with cheaper electricity, soliciting farmers’ votes with promises of free or unmetered electricity). 

The financial losses experienced by the Indian Electricity Sector are staggering. According to the World Bank, theft from public utilities in developing countries amounts to more than $33 billion a year, with India contributing a significant portion of this. Corruption in capital expenditure alone is about $8 billion a year.  These numbers likely only capture the measurable losses from corruption in the electricity sector. This number may be exponentially higher when taking into account the welfare losses associated with politically motivated redistribution of electricity services to the rich or selectively allowing theft to occur in neighborhoods supportive of the election winners. 

These phenomena are not limited to India, but are pervasive across other developing countries. A significant portion of such corruption goes undetected, perpetuating a vicious cycle of poor electricity quality and an opportunity for politicians to exploit in exchange for votes. The lack of detection is facilitated by the fact that it is difficult to establish causality between political action and electricity outcomes. The micro-level data that may help with this task (electricity consumption records or details of power purchase agreements) is hard or impossible to obtain, and even if the data is available, it is difficult to judge how much it can be trusted and whether it has been manipulated in some way. 

The good news is there are newer and increasingly more popular sources of data we can harness to verify electricity consumption data. In my work, I contrast administrative electricity consumption records with satellite nighttime lights data and show that regions aligned with the incumbent party are rewarded with electricity subsidies. The electricity consumption records reflect an overall lower consumption level in regions that narrowly voted for the winners in the last election. However, the satellite data paints a different picture: actual electricity consumption in these same regions is higher than others. 

What explains this discrepancy between the two data sources? Digging deeper, I find that the administrative data demonstrate patterns consistent with data manipulation (e.g. a disproportionately high number of accounts in areas supporting the election winner appear to consume electricity units in multiples of ten). These patterns imply that these regions consumed more electricity than others, but were billed for 40% less, indirectly subsidizing their consumption. Furthermore, the groups who benefit from this subsidy are usually urban dwellers or large businesses. This kind of corruption is hard to detect by on-the-ground auditors for instance, who examine administrative data alone.

I find that in a single state alone, this kind of politically motivated subsidy significantly hurts utility revenues and creates a welfare loss large enough to power 3.7 million additional rural homes. Further contextualizing this is the fact that 200 million individuals in India still lacked an electricity connection in 2018. Detecting corruption and estimating its impact is an important step in fixing some of these systemic issues. Part of the problem is that politicians have incentives to keep the electricity sector broken. Previous research has found that incumbent politicians are more likely to win in areas with a large amount of power losses, and therefore where there is greater demand for politicians to “fix” the problems in the electricity sector. When their fiscal capacities are limited, public utilities can provide a valuable political tool. A true separation between electricity and politics does not appear likely under the current configuration since almost all state utilities in India are state-owned and therefore beholden to politicians and bureaucrats. 

This is an avenue with a lot of future research and policy potential, with opportunities to rethink the role of private electricity provision, design an incentive-compatible utility provisioning framework, and improve monitoring of the sector.

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.