Who Regulates the Regulators

Pritiraj BrahmaFounding Partner, Pritiraj & Partners · 23 April 2026

India built independent regulators for its most important sectors and then left their appointments, budgets and accountability inside the ministries they were meant to check.

ProvenancePublished 23 April 2026; rebuilt 11 August 2026 to replace argument from statute with attributed figures, and revised 31 August 2026 to add two figures and a provenance line. Tribunal chairperson vacancies are from a reply in Parliament, 2025; the Income Tax Appellate Tribunal member posts from the Parliamentary Standing Committee on Personnel, Public Grievances, Law and Justice, March 2026; the Supreme Court notice from May 2026. Pendency figures are government figures as reported in August 2026 and are a position, not a series. The findings on appointment, funding and capacity describe institutional design across sectors and are not attributed to a single regulator’s accounts; they should be read as characterisation, not measurement.

An independent regulator whose chairman is appointed by the ministry, whose budget is sanctioned by the ministry, and whose staff are deputed from the ministry is not independent. It is a department with better stationery.

Between the early 1990s and today India built a regulatory state: electricity regulatory commissions at Centre and state, a securities regulator, a telecom regulator, an insurance regulator, a pension regulator, sectoral bodies for petroleum, ports, real estate, food safety and competition. The logic was sound — as the state moved from producing to overseeing, expertise and independence had to be located somewhere other than the ministry that had previously run the industry.

The construction was left unfinished in three places, and the same three recur across almost every regulator.

Five findings anchor this analysis:

1. Appointment remains executive. Selection committees vary, but in most cases the effective appointing authority is the Union or state government, often the administrative ministry for the sector. A regulator whose reappointment or post-retirement prospects depend on the entity it regulates faces an obvious structural conflict, whatever the individual's integrity.
2. Funding is rarely autonomous. Where a regulator depends on annual budgetary sanction rather than a dedicated fund or levy, its most binding constraint is the goodwill of the department it oversees. Regulators with independent revenue behave differently from those without, and the difference is visible in their willingness to issue adverse orders.
3. Capacity is thin and largely borrowed. Many commissions run substantially below sanctioned strength and staff themselves through deputation from government. An officer on deputation returns to the parent cadre; the incentive to write an order the parent department dislikes is correspondingly weak. Sectoral expertise — engineers, economists, actuaries, data scientists — is scarce at regulator pay scales.
4. Accountability runs to the executive rather than to the legislature. Regulators typically report through a ministry. Parliamentary committees examine them episodically. There is no routine mechanism by which a regulator must explain its decisions, its pendency and its enforcement record in public on a fixed calendar.
5. The appeal route is itself unstaffed. A regulator's orders are meant to be tested before an appellate tribunal rather than reviewed by the ministry that appointed its members, and that layer is chronically vacant. A 2025 reply in Parliament recorded seven tribunals functioning without a chairperson — the Central Government Industrial Tribunal, CESTAT, the Appellate Tribunal for Forfeited Property, the Debt Recovery Tribunal, the Debt Recovery Appellate Tribunal, the Railway Claims Tribunal and the Armed Forces Tribunal. In March 2026 the Parliamentary Standing Committee on Personnel, Public Grievances, Law and Justice recorded 27 member posts under recruitment at the Income Tax Appellate Tribunal alone, 15 Accountant and 12 Judicial; in May 2026 the Supreme Court issued notice to the Union on a petition over the same vacancies. Pendency has followed: 2,33,901 cases before the Debt Recovery Tribunals, 71,454 before CESTAT, 69,102 before the Central Administrative Tribunal and 42,502 before the ITAT, on government figures reported in August 2026. The forum built to relieve the courts now carries a docket of the same character.

Where this shows up

The pattern is clearest in electricity, which is why it recurs in this Review. State electricity regulatory commissions are statutorily required to determine tariffs. Where tariff orders are delayed, suppressed or issued without the cost basis being followed, the distribution utility accumulates a regulatory asset — a receivable it is permitted to book but not to collect. The Sixteenth Finance Commission's conditions specifically address timely tariff orders and the non-creation of new regulatory assets, which is a tacit acknowledgment that the regulator has not consistently been able to do the job the statute assigns it.

The mechanism is not corruption. It is that a commission whose members are appointed by a state government, funded by it, and staffed from it, is being asked to require that government to raise prices on its own voters. The design places an institution in a position no institution can hold.

Why electricity is the hard case, and what the easy case shows

The three defects above are not equally severe everywhere, and the variation is diagnostic. Compare the securities regulator with a state electricity regulatory commission — bodies of broadly similar statutory form, with markedly different capacity to act.

The securities regulator is funded substantially from fees and charges levied on the market it oversees rather than from an annual departmental sanction, and its orders are challenged before an appellate tribunal rather than reviewed by the ministry that appointed its members. A state electricity regulatory commission typically depends on the state budget, is staffed in significant part by officers on deputation from the state, and regulates a distribution utility the state itself owns. The statutory language granting independence is comparable in both cases. The behaviour is not, and the difference tracks the funding and the ownership rather than the drafting.

That points to a defect deeper than any of the three named above, and one no appointment reform reaches. Where the state owns the regulated entity, the conflict is structural rather than procedural. A commission required to order a state-owned discom to raise tariffs is being asked to instruct its own appointing authority, funder and employer to raise prices on that authority's voters. No selection committee composition survives that arrangement, and no cooling-off period touches it. Separating ownership from regulation is the reform; independence of the regulator is a partial substitute for it.

There is a further piece of evidence, and it is close to an admission. The Sixteenth Finance Commission conditions money on timely tariff orders and on the non-creation of new regulatory assets. Consider what that means institutionally. A statutory regulator already holds a legal duty to determine tariffs on a cost basis; the Electricity Act assigns it. When a Finance Commission finds it necessary to attach a fiscal condition requiring that this duty be performed, the Union is using a grant instrument to obtain compliance the regulatory statute was written to secure directly. That is not a criticism of the Commission, whose design is sound and whose leverage is real. It is evidence that the regulatory settlement has failed at its intended point, and that the failure is being routed around rather than repaired.

The route around works, and it is worth noting why: a conditional grant creates a consequence, which is the element the regulatory design omitted. A regulator with a statutory duty, no funding autonomy and no accountability calendar is a body with an obligation and no mechanism. Attaching money to the obligation supplies the mechanism from outside. The durable answer is to build it inside — which is what a dedicated levy, a legislative accountability hearing and a permanent technical cadre together amount to.

One statute, thirty-five outcomes

The clearest published evidence that the defect is architectural rather than personal comes from the youngest of these regulators, where one central statute was implemented by every state inside a decade. At the fifth meeting of the Central Advisory Council under the Real Estate (Regulation and Development) Act in September 2025, the Ministry of Housing and Urban Affairs recorded that 35 states and Union Territories had constituted Real Estate Regulatory Authorities, 29 had established appellate tribunals and 27 had appointed adjudicating officers. Around 1.51 lakh projects and 1.06 lakh agents stood registered, and roughly 1.47 lakh complaints had been disposed of.

Two things follow. First, in six states and Union Territories a homebuyer has a regulator but no tribunal to appeal to, and in eight there is no adjudicating officer to award compensation. The remedy the Act promises does not exist in part of the country — an omission of appointment, not of law, and precisely the defect this Review keeps finding: a task assigned to nobody in particular. Second, where the machinery does exist, performance varies by a wide margin. On the Ministry's own RERA tracker, Uttar Pradesh had disposed of 86.71 per cent of 60,021 registered complaints, Maharashtra 82.03 per cent of 34,485, Gurugram 93.62 per cent of 17,893 and Karnataka 81.54 per cent. One statute, one template, one decade — and a spread that tracks staffing and state practice rather than drafting.

The counter-case, honestly stated

There is a real argument on the other side, and it is democratic rather than administrative. Regulators are unelected. Tariffs, spectrum prices and insurance rules have large distributional consequences, and it is not obvious that such decisions should sit wholly beyond the reach of an elected government. Excessive insulation produces its own pathology: a technocratic body accountable to nobody, captured over time by the industry it regulates — which is the standard failure mode of regulators in mature economies.

That objection is correct and it argues for a specific answer rather than against independence. The right settlement is not insulation from politics; it is insulation from the executive combined with accountability to the legislature. Government sets policy through published directions; the regulator applies it through reasoned orders; Parliament or the state assembly examines the record annually in public.

What we would do

1. Put appointments through a fixed, published process with a non-executive majority. A selection committee where the administrative ministry does not hold the deciding voice, with candidates and reasons published.
2. Bar post-retirement employment in the regulated sector for a defined period. A cooling-off period of two to three years, applied uniformly. This is the single cheapest anti-capture measure available and it is applied inconsistently at present.
3. Fund regulators from a dedicated levy, not annual sanction. A small levy on the regulated activity, with the regulator's budget approved by the legislature rather than the ministry. Financial independence is the precondition for every other kind.
4. Build permanent technical cadres. Direct recruitment at market-competitive terms for economists, engineers and analysts, with career paths inside the regulator. Deputation should be the exception, not the staffing model.
5. Require an annual public accounting. Every regulator should table an annual report covering orders issued and their timeliness, pendency, enforcement actions, and compliance with its own statutory deadlines — examined in a public hearing by the relevant legislative committee.

India built the institutions. It did not finish building the conditions under which they can act. That is the same finding this Review reaches in policing, in the districts and in the courts: the design is sound, the discretion that would make it work was never handed over, and the failure is then described as a failure of the institution rather than of the settlement around it.