When data quality trips power reforms

For things to change on the ground, wrong reporting of numbers needs to be severely penalised

There have been a slew of steps in recent times to improve penetration and quality of electricity supply, and the financial health of utilities, particularly the distribution companies (discoms). However, the achievement on the ground has been far short of the targets envisaged, with timelines getting shifted and targets modified in many cases.

The root cause of such failures? The baseline data used to devise interventions. Aggregate technical and commercial (AT&C) loss is the primary data source to assess the performance of discoms. All schemes — such as, the Restructured Accelerated Power Development and Reforms Programme (R-APDRP), Integrated Power Development Scheme (IPDS), Deen Dayal Upadhyaya Gram Jyoti Yojana (DDUGJY), and, most recently, the Ujwal DISCOM Assurance Yojana (UDAY) — pivot on AT&C loss reduction over three years based on reported numbers (including baseline data used to arrive at these targets).

Disbursement of funds and benefits largely depend on the extent of achievement of targets. However, in case of non-achievement of targets, discoms tend to submit ‘fabricated’ figures instead of ‘actuals’ to ensure they do not lose benefits. This is also because the information reported at the time of scheme inception is often incorrect. The extent of this problem is also visible from the AT&C loss data for fiscal 2015 submitted by discoms to governments, regulators and the Power Finance Corporation under UDAY (see chart). 

# computed from reported distribution loss and collection efficiency as reported to PFC: Data for fiscal 2015

The perception is that discoms report these figures as per their convenience for a particular purpose, whereas the actual losses are much higher. A back-of-the-envelope calculation shows average deviation or misreporting of 1 per cent of AT&C losses would involve manipulating an amount of Rs 40 billion. 

A look at the numbers reported on the UDAY portal or to various authorities in the last three years or so shows the unreliability and inconsistency. For instance, in Uttarakhand, the losses in fiscal 2015 were 18.64 per cent as per UDAY agreement that, instead of declining rose to 29.29 per cent for the December 2016 quarter. This may be because of better or correct reporting in fiscal 2017 through right assessment of unmetered consumption. Such variance is also visible in Punjab and Madhya Pradesh. 

Thus, quality and consistency of reporting is the core issue, whether deliberate or not. The reasons could include, non-availability of actual information with the discom due to low level of metering, especially in rural and agriculture areas and at various energy input levels; no uniformity in procedures for data generation, collation, computing and reporting; inconsistency within discoms itself,where departments follow different processes to capture the same set of data; significant dependence on manual intervention, leaving scope of human error and fudging; camouflaging theft and pilferage as unmetered consumption.

The most important action points, therefore, would be 100 per cent metering at all levels and adoption of a uniform methodology for information processing across states. A shift from manual or staggered information system to a single-point, web-enabled information system would also be desirable. This will afford access to information till the last mile and will also eliminate quality loss due to manual intervention. Information availability can be improved further by enabling a universal platform for obtaining the AT&C loss data. Adoption of common formats, templates, new systems/processes can all enhance the reporting system.

In the interim, a data quality review can be conducted, including sample analysis of the reporting framework adopted by discoms. The review will help in identifying the potential sources of error. Since the schemes typically offer grants/exemptions/subventions, reporting of wrong data should be considered an offence, and accordingly, penalty levied on those responsible.

The author is senior director, CRISIL Infrastructure Advisory