Assessing regulatory capability in Tamil Nadu electricity regulation: Evidence from appeals

by Bhavin Patel and Renuka Sane.

The Indian journey to decarbonisation faces the roadblocks of electricity policy. One of the critical impediments faced is that of state capability in electricity regulation. There is a new body of knowledge in India, in regulatory theory, that can usefully be brought to bear on the problem of improving electricity regulation. One element of this field is the question of assessing the state of regulatory capability. At any point in time, how would we judge the extent to which electricity regulation in a certain state is working well?

Electricity is regulated by the Central and State Electricity Regulatory Commissions (ERCs), which perform legislative, executive, and quasi-judicial functions. The concentration of power and the lack of democratic accountability in the new administrative state raise concerns about their functioning in modern economies. In India, the ERCs have quasi-judicial powers that are even broader than those of other regulators such as the SEBI. With one electricity regulator — the TNERC — in focus, we took up the research strategy of assessing regulatory capability through the analysis of regulatory orders that go to appeal. A key metric of the quality of orders the regulator passes is how well its orders fare at an appellate forum. If the regulator succeeds in defending its decisions, it signifies that the orders are well-reasoned and have followed the procedures required by the applicable law (whatever they be).

In a recent paper, Performance at the Appellate Tribunal as an indicator of regulatory capacity: The case of TNERC at APTEL, we study the overturn rate of the orders of the Tamil Nadu Electricity Regulatory Commission (TNERC) at the Appellate Tribunal for Electricity (APTEL) between 2013 and 2023.

We find that TNERC failed to hold its ground in 52% of appeals. Of these, about half were remanded back to the TNERC, suggesting that it did not do an adequate job of bringing the evidence necessary to decide a matter, despite being conferred the powers of a Civil Court for such purposes under Section 94 of the Electricity Act, 2003. Further, the TNERC lost 86% of matters that involved issues related to how it uses its regulatory powers.

These results raise concerns about the exercise of quasi-judicial authority by the TNERC. There is considerable knowledge in the field of regulation in India, on how such deficiencies can be addressed. If policy makers were to take up such strategies, it would help improve private sector confidence for electricity investment in Tamil Nadu.

Our work is limited to orders that were appealed. There is a selection bias, in that regulated persons make rational choices when deciding to appeal. In the extreme, it is possible to argue that all sound orders are not appealed, and only defective orders are appealed, so the defect rate is just the appeal rate (which is about 8% with appeals to APTEL that we measure, and some appeals that go to the High Court under writ jurisdiction which we do not measure). We suspect the defect rate is higher. Considerations in appealing include issues such as errors in the regulatory order which create a high chance of winning, the expense in appealing, the rupee value at stake multiplied by the probability of winning, and the non-rule-of-law pressures from TNEB in favour of not appealing. The appealed orders, that are studied by us, offer valuable insights into the limitations of state capability in regulation in TN electricity. Such measurement should motivate regulatory reform, and it should be used in measuring the extent of progress in regulatory reform.

Our approach can be usefully applied in analysing and improving electricity regulation in other states. To assist this process, we have also released a detailed Manual For Reviewing Regulatory Orders: Orders of the Tamil Nadu Electricity Regulatory Commission at the Appellate Tribunal for Electricity. This will enable replication of our results for Tamil Nadu, and porting these methods to other states.

The authors are researchers at TrustBridge.