Infrastructure laws in India do not form a single code, and the reader who goes looking for one will misunderstand the field. There is no Infrastructure Act. What exists is a shelf of sector statutes, one cluster for each network industry, and the statutes differ from each other in the one respect that decides almost everything practical about a sector: who governs it. Some sectors are run by an independent regulator created by statute, a body that licenses operators, fixes or approves the charges they levy, and answers appeals through a specialised tribunal. Other sectors are run by a government department that is at once the owner of the network, the operator of the service, the maker of the rules and the judge of complaints against itself. Electricity belongs to the first group. The railways and the national highways belong to the second. Ports, airports, oil and gas, road transport, mining and atomic energy are distributed between the two models, or split inside themselves, and the distribution is not random.

The organising distinction of the entire field is whether the State still operates the service. Where the State withdrew from operations and admitted private and competing operators, Parliament built a regulator to stand between the operator and the consumer, because a market with a former monopolist and new entrants cannot be refereed by a ministry that owns one of the players. Where the State remained the operator, no referee was built, and the department continues to set its own charges, investigate its own failures and decide its own liability. That single fact explains why a power consumer in a State capital can appeal a tariff order to a tribunal in Delhi while a railway passenger has no tariff forum at all, why an airport’s aeronautical charges are determined by a statutory authority while a port’s were for decades fixed by a tariff body that the port statute has since discarded, and why the competition between governance models, rather than any list of Acts, is the thing a reader must grasp first. This article maps the whole field on that distinction: the constitutional entries that scatter infrastructure across all three Lists of the Seventh Schedule, the statute set sector by sector, the land question that every project shares, the contract layer that substitutes for a missing partnership statute, and the dispute forums that decide what happens when projects fail.

Infrastructure laws in India begin in the Seventh Schedule

The reason Indian infrastructure is governed so unevenly is written into the Constitution before any statute is reached. Most fields of legislation sit mainly in one List. Infrastructure refuses to. Its subjects are scattered across the Union List, the State List and the Concurrent List in a pattern that follows the geography of each network rather than any theory of regulation, and the full map of that division is set out in the article on the three lists of the Seventh Schedule and how subjects are distributed between them. Electricity is a Concurrent subject. Roads are split between national highways in the Union List and every other road in the State List. Mines are split by a declaration device: the Union entry takes control only to the extent Parliament declares expedient, and the State entry operates in the space the declaration leaves. Railways and atomic energy are exclusively Union subjects from end to end. A reader who holds that map can predict, for almost any sector, which legislature wrote its statute, which government administers it, and where the friction between Centre and State will appear.

The Union List entries carry the network industries that cross State boundaries by their nature. Entry 22 covers railways. Entry 23 covers highways declared by or under law made by Parliament to be national highways, which is why the National Highways Act, 1956 proceeds by declaration: a road becomes a national highway when the Central Government declares it one, and the declaration is what moves the road from the State List to the Union List. Entry 24 covers shipping and navigation on inland waterways declared by Parliament to be national waterways, as regards mechanically propelled vessels, and Entry 25 covers maritime shipping and navigation, including shipping on tidal waters. Entry 27 covers ports declared to be major ports, together with the constitution and powers of their port authorities. Entry 29 covers airways, aircraft and air navigation, and the provision and regulation of aerodromes. Entry 30 covers the carriage of passengers and goods by railway, sea or air, or by national waterways in mechanically propelled vessels. Entry 6 covers atomic energy and the mineral resources necessary for its production. Entry 52 covers industries whose control by the Union is declared by Parliament to be expedient in the public interest, the entry on which the industrial licensing statutes rested. Entry 53 covers the regulation and development of oilfields and mineral oil resources, together with petroleum and petroleum products. Entry 54 covers the regulation of mines and mineral development to the extent to which such regulation under Union control is declared by Parliament to be expedient in the public interest, and the Mines and Minerals (Development and Regulation) Act, 1957 contains exactly that declaration, which is why a subject that also appears in the State List is in practice governed by a central statute.

The State List entries are the residue and the foundation at once. Entry 13 covers communications, meaning roads, bridges, ferries and other means of communication not specified in the Union List, together with municipal tramways and ropeways. Every State road, district road and village road stands on this entry, which is why the physical road network of India is overwhelmingly a State subject punctured by declared national highways. Entry 23 covers the regulation of mines and mineral development, subject to the provisions of the Union List with respect to regulation and development under Union control, so the State power over mining is real but defeasible, and it survives most completely for minor minerals, which the central statute leaves largely to State rules. Entry 24 covers industries, subject to Entries 7 and 52 of the Union List. Entry 17, on water supplies, irrigation and canals, drainage and embankments, water storage and water power, subject to the Union entry on inter-State rivers, governs the water infrastructure that this article touches only at its edges, because water power is State while the electricity it generates is Concurrent, a division that produces its own celebrated frictions.

The Concurrent List holds the subjects on which both legislatures may act and on which the sectors’ daily law is often made. Entry 38 is electricity, a single word that carries the entire power sector. Entry 35 covers mechanically propelled vehicles, including the principles on which taxes on such vehicles are to be levied, and the Motor Vehicles Act, 1988 rests on it, administered in large part by the States even though Parliament wrote it. Entry 20 covers economic and social planning, the peg for plan-era industrial and infrastructure programming. Entry 31 covers ports other than major ports, which is why the minor ports of the coastline are a Concurrent subject administered almost entirely by State maritime boards. Entry 32 covers shipping and navigation on inland waterways as regards mechanically propelled vessels, and the carriage of passengers and goods on such waterways, subject to the Union entry for declared national waterways, so the same river can be Union water for a declared stretch and Concurrent water elsewhere. Entry 42 covers the acquisition and requisitioning of property, moved into the Concurrent List by the Seventh Amendment, and it is the entry under which both Parliament and State legislatures enact acquisition law, a point that matters when States frame their own acquisition statutes or amend the central one for their territory.

Can a State make its own law on electricity or mines?

Yes, within limits that differ by subject. Electricity sits in the Concurrent List, so a State legislature may legislate on it, and State amendments and rules fill large parts of the working law. Mining is different: the State entry is subject to the Union declaration in the 1957 Act, so State power contracts where the central statute occupies the field.

The practical consequence of the scatter is that no ministry and no single legislature sees the whole field. The Ministry of Power administers a Concurrent subject through central and State commissions. The Ministry of Railways administers an exclusive Union subject through a department that is itself the operator. The Ministry of Road Transport and Highways administers declared highways under a Union entry while every connecting road belongs to a State public works department. Ports divide at the waterline between major and minor. The constitutional map does not merely permit this variety; it guarantees it, and every statute described below is one legislature’s answer to the slice of the map it was given.

How the field was built: five phases of one argument

The statute book reads as a single argument conducted over five phases, and placing a sector statute in its phase explains most of its design. The first phase is the founding departmental settlement of the early Republic. The National Highways Act of 1956, the Mines and Minerals Act of 1957, the Atomic Energy Act of 1962 and the Major Port Trusts Act of 1963 share a common assumption so complete it is never stated: the State will build the network, own it, operate it and answer for it, and the statute’s business is to equip a department with the powers of acquisition, construction and management that the task requires. Regulatory language is absent because there is nobody to regulate. The user’s protection in these statutes is not a forum but a liability clause and the political process, and the charges the user pays are departmental decisions taken inside the general budgetary discipline of the State.

The second phase is the plan-era deepening of the same settlement, in which the department acquires corporate and board forms without surrendering control. The State Electricity Boards under the Electricity (Supply) Act, 1948, the port trusts under the Act of 1963 and the railway administration under its own departmental hierarchy are all devices for giving a commercial operation a distinct accounting personality while keeping appointment, tariff and investment inside government. The legal signature of the phase is the government company and the statutory board whose members the government appoints and whose surpluses and deficits the government absorbs. It is the phase that built most of the physical network the country still uses, and it is also the phase whose finances failed first, because a board that sets its own tariffs under political instruction will set them below cost and finance the difference by neglecting the asset.

The third phase is the turn of 1991 and the decade that followed it, in which the operating monopoly was broken in sector after sector and the law had to answer a question it had never faced: on what terms may a private operator enter a network the State built, and who protects the entrant against the incumbent and the consumer against both. The legislative answers of this phase are the regulator statutes in embryo and in full: the Electricity Regulatory Commissions Act, 1998, the amendment of 1997 that created the Tariff Authority for Major Ports, the opening of highway construction to concessionaires under the Authority’s programme, and the first model concession agreements, which belong to this phase even though they are documents rather than statutes. The phase’s governing idea is that credibility is an institution or a document, deliberately manufactured, and that private capital prices its absence.

The fourth phase is the consolidation of the regulator generation and the great build-out of the contract layer. The Electricity Act, 2003 completes the power sector’s reconstruction. The Petroleum and Natural Gas Regulatory Board Act follows in 2006 and the Airports Economic Regulatory Authority Act in 2008. The highways programme reaches its full concession scale, the ports admit private terminals as the standard method of adding capacity, and the model documents multiply across sectors. The phase’s failures are as instructive as its statutes: the build-out was financed on projections of traffic, demand and fuel supply that the following decade disappointed, and the documents and institutions of the phase were stress-tested by events their drafters had treated as remote.

The fifth phase is the law of stress and correction, and it is the phase whose instruments are least sectoral. The Insolvency and Bankruptcy Code, 2016 supplies a general process for the failed project company wherever it sits. The Major Port Authorities Act, 2021 rearranges the ports settlement toward port autonomy and away from the tariff authority its predecessor phase created. The arbitration jurisprudence culminates, for this field, in the Constitution Bench decision of 2024 on who may appoint the tribunal that hears the State’s own contract disputes. And the acquisition settlement of 2013 and 2015, which belongs chronologically across the fourth and fifth phases, re-prices the land beneath every sector at once. A reader who can date a provision to its phase can usually infer its logic: departmental ownership from the first two, regulated entry from the third, contractual scale from the fourth, and failure management from the fifth.

The anatomy of an Indian infrastructure regulator

The regulator statutes differ in their sectors but share an institutional anatomy, and the shared parts are where their real powers and real limits sit. Composition first. Each body is a small commission of a chairperson and members appointed by the Central or the State Government, as the sector requires, on the recommendation of a selection committee whose own composition the statute prescribes, drawn from the judiciary, the civil services, the industry and the academy in proportions that vary by statute. Tenure is fixed and short, typically a term of years with an age ceiling, and reappointment is restricted or barred in several of the statutes, a design intended to free the member from pleasing the appointing government in hope of renewal. Removal is confined to stated grounds such as insolvency, conviction, incapacity and abuse of position, in some statutes after an inquiry by a judge of the Supreme Court, so a member cannot lawfully be dismissed for an unwelcome tariff order. These are genuine independence devices, and they operate; the qualification the field records is that they protect the member after appointment and do nothing about the appointment itself, which remains the government’s, and a government that cannot remove a regulator can wait one out, leave a vacancy open, or appoint the compliant. The vacancy is the Indian regulator’s characteristic wound: commissions across the sectors have functioned for long periods below strength, and appellate tribunals likewise, because the appointing power and the regulated power are the same hand.

Function next. The infrastructure regulator’s powers assemble from a standard kit: the licensing or authorisation power, which controls entry; the tariff power, which controls price; the standard-setting power, which controls quality and technical terms; the information power, which compels the regulated entities to file, publish and disclose; the adjudicatory power, which decides disputes of the classes the statute assigns; and the penalty power, which fines and, in electricity, supports a dedicated officer for investigation and enforcement within the sector’s own scheme. No Indian infrastructure regulator holds all six in full strength. The petroleum board authorises and sets standards but holds no retail price. The airports authority determines aeronautical tariffs but leaves the airport’s commercial revenues and the airline’s fares outside its order. The electricity commissions come closest to the full kit, and even they adjudicate only between the classes of person the Act names. Reading a regulator statute for its kit, and noting which instrument is missing, is the fastest reliable way to predict what the body will fail to control.

Accountability runs on three rails, none of them a user’s remedy. The regulator’s accounts are audited by the Comptroller and Auditor General and its annual report is laid before the legislature, which subjects it to the scrutiny of parliamentary or assembly committees. Its decisions are appealable to the sector tribunal and then, on law, to the Supreme Court, which is the accountability that bites in individual cases. And the government retains a power, present across the regulator statutes, to issue directions on questions of policy: under the Electricity Act, 2003 the appropriate government may direct the commission in the public interest on a question of policy, the commission is guided by such directions, and the government’s decision on whether a question is one of policy is made final by the statute, while the direction cannot decide a case. The boundary between a policy direction and an instruction about outcomes is the constitutional fault line of the regulator model, litigated rarely and watched constantly, because a direction power used to set a tariff would collapse the model back into departmental administration while leaving its forms standing.

Entry and access: the second question every sector statute answers

Beneath the tariff question sits a prior one that decides whether a sector has competitors for a regulator to regulate at all: on what terms may a new operator enter, and on what terms may it use the network that already exists. The sector statutes answer it so differently that the answers amount to a second map of the field, laid over the first. Electricity gives the most liberal answer in the set. Under the Electricity Act, 2003, generation is delicensed: no licence is required to establish, operate or maintain a generating station, with a technical consultation requirement for hydroelectric projects, the single surviving echo of the licensing era, while transmission, distribution and trading in electricity remain licensed activities. The access half of the answer matches: the Act imposes open access, the duty of a transmission or distribution licensee to allow a generator, a licensee or an eligible consumer to use its network against the regulated charges, phased in by the commissions’ regulations, so that the incumbent’s wires became a common facility in law and the market the regulator polices is a market in the use of a shared network. The design is the regulator model’s foundation document: entry free at the point of production, access guaranteed at the point of transport, price regulated at both.

Oil and gas answers through authorisation and declaration. The petroleum board authorises the laying and operation of pipelines and city gas networks, and the statute empowers it to declare a pipeline a common carrier or contract carrier, which obliges its owner to accept third party capacity on regulated terms, so access follows designation rather than arising generally. City gas authorisations carried, under the Board’s regulations, a period of marketing exclusivity for the authorised entity in its geographical area, an incubation device with no equivalent in electricity, defended as the price of persuading distributors to build networks whose demand was unproven. The Board’s limits were marked by litigation early: in the Indraprastha Gas case decided by the Delhi High Court in 2012, the court set aside the Board’s attempt to fix the network tariff and compression charge of the Delhi gas distributor in the manner it had chosen, holding that the power asserted had not been conferred in that form, a judgment the field reads for the general proposition that an Indian regulator holds the powers its statute enumerates and not the powers its sector seems to need. Airports answer the entry question with the concession rather than the licence: entry to a major airport is by a single concession over public land to one operator for a long term, with the aeronautical tariff then regulated by the authority, so competition, where it exists, is competition for the market at the bid rather than competition within it. Mining answers by auction since the amendment of 2015 made auction the mandatory route for the grant of concessions for the principal minerals, ending the discretionary allotment whose exercise in coal had been condemned and undone by the Supreme Court in the coal allocation judgments of 2014, which cancelled the allotments of more than two hundred blocks for arbitrariness in the screening process. Highways answer by bid for a defined stretch on the model documents. The railways, alone in the set, give no general answer at all, because no general entry exists to be answered for: entry is a set of specific permissions the administration grants at its own edges, on its own terms.

Arranged on a single scale, the entry answers run from the open to the closed, and the scale is a serviceable summary of the whole field. At the open end stands electricity generation, delicensed altogether, where the entrant needs land, fuel, a grid connection under open access and a buyer, and the State’s permission is the exception rather than the rule. Next stand the authorised networks of oil and gas and the auctioned concessions of mining and highways, where entry is competitive and periodic but the terms, the duration and the exclusivity are set in advance by the authority that grants them. Next stand the single concessions of airports and port terminals, where entry is a rare event decided by one bid for one asset for a generation. And at the closed end stand the railways’ operating core and atomic energy, where entry is not a legal category at all. A reform that moves a sector along this scale changes its law more deeply than any amendment of its charges, because it changes who may exist in the sector; and the direction of travel in the statute book since 1991 has been, sector by sector and never completely, from the closed end toward the open.

Federal friction in operation

The constitutional scatter described at the start of this article is not a classroom point; it produces a working friction in at least four places, and a guide to the field should name them. The first is electricity’s Concurrent character working against the 2003 Act’s own timetable. Because distribution is administered by the States and the licensees are mostly State owned, the unbundling of the State Electricity Boards, the constitution of State commissions and the introduction of open access all depended on State action under a central statute, and they proceeded at State speeds that diverged widely, with the Act’s restructuring deadlines extended and its mechanisms adapted State by State. The central statute supplied the destination; the federal structure set the pace, and any account of the power sector that reads the Act as self-executing misreads the field.

The second friction is mining, where the friction runs the other way. The State entry on mines and minerals is subject to the Union declaration, and the 1957 Act occupies the field for the major minerals, so the State’s role, though operationally dominant in the grant and administration of leases, is exercised inside a central code that prescribes even the royalty schedule. When the States’ own discretionary allotments of coal blocks were cancelled by the Supreme Court in 2014 and Parliament answered with mandatory auctions in 2015, the sequence showed the declaration’s real content: the Centre can convert a State-administered field into a centrally scripted one whenever it legislates to the extent of its entry, and the States’ remedy is political rather than constitutional.

The third friction is land. Acquisition is Concurrent, and the 2013 Act is a central statute on a Concurrent entry, so States may legislate for their own territories with the assent of the President under Article 254(2), which several States obtained for amendments diluting or adapting the central Act’s consent and assessment provisions, while the Union’s own sector statutes continue to run their separate procedures in every State regardless. The result is a genuine patchwork at the exact point where the citizen meets the field: the same highway programme can acquire under the central sector code in one State and interact with a State-amended compensation environment in the next.

The fourth friction is the oldest and the quietest. Water is a State subject and electricity a Concurrent one, so the hydroelectric project sits across the seam: the State controls the river and the water, the project’s economics are regulated under the central electricity framework, and the 2003 Act itself preserves the seam by requiring the techno-economic consultation for hydro projects noted in the entry discussion above. Inter-State rivers add the Union’s Entry 56 power to regulate and develop inter-State rivers and river valleys to the extent Parliament declares, a declaration made for particular valleys by particular statutes rather than generally. No reform proposal in the field is older than the suggestion that this seam be restitched, and none has been taken up, which the map records without surprise: the scatter of the Seventh Schedule is not an accident awaiting repair but a settlement each level of government defends where it holds the stronger entry.

Two governance models divide the statute book

Lay the sector statutes side by side and two templates emerge. The first template is the regulator statute. Parliament creates a commission or authority as a body corporate, separates it from the ministry, gives it the licensing power, the tariff power and the power to penalise, builds an appellate tribunal above it, and confines the government’s role to policy directions on questions of policy and to appointments and budgets, which are the levers that remain. The Electricity Act, 2003 is the complete instance: the Central Electricity Regulatory Commission and the State Electricity Regulatory Commissions license transmission and trading, determine tariffs for generation and transmission within their jurisdictions, adjudicate disputes between licensees and generating companies, and answer to the Appellate Tribunal for Electricity. The Petroleum and Natural Gas Regulatory Board Act, 2006 and the Airports Economic Regulatory Authority of India Act, 2008 follow the same template in narrower fields. The second template is the departmental statute. Parliament creates no regulator at all. The Railways Act, 1989 vests the administration of the railways in the Central Government acting through the railway administration, which operates the trains, frames the rules, classifies the goods, fixes the fares and freights through government notification, and carries its own liability regime inside the same Act. The National Highways Act, 1956 with the National Highways Authority of India Act, 1988 is a softer version of the same template: the Authority is a statutory body, but it is a project-executing agency of the government, not a regulator of others, and no independent body stands between the highway user and the department that builds and tolls the road.

Between the templates lie the hybrids, and the hybrids are where the field’s real texture is. Major ports were regulated for tariff purposes by a Tariff Authority for Major Ports created by an amendment of 1997 to the Major Port Trusts Act, 1963, while the ports themselves were run by port trusts that were departmental in character; the Major Port Authorities Act, 2021 then removed the tariff authority from the major ports and handed rate-fixing to each port’s own Board, moving the sector from the regulator template toward the departmental one at the very moment its traffic was being opened wider to private terminals. Road transport is administered by the States under a central Concurrent statute, so the model is departmental but federal: the registering authority, the transport commissioner and the State transport undertaking all belong to State governments, and no central body regulates the sector’s economics at all. Mining combines a central statute with State administration: Parliament wrote the Mines and Minerals (Development and Regulation) Act, 1957, but leases are granted by State governments, auctions are conducted by State governments, and the royalty and district mineral foundation machinery runs through State hands, with a central tribunal route for revision. Atomic energy is departmental in its purest surviving form: the Atomic Energy Act, 1962 (Act 33 of 1962) concentrates control in the Central Government and the Department of Atomic Energy, and the Atomic Energy Regulatory Board, which performs the safety function, was created by an executive order of 1983 rather than by statute, so the sector’s safety regulator holds no statutory charter of its own.

The sequence in which the models arrived is itself evidence. The departmental statutes are the older ones: the National Highways Act of 1956, the Mines and Minerals Act of 1957, the Atomic Energy Act of 1962, the Major Port Trusts Act of 1963, the Railways Act of 1989 replacing an Act of 1890. The regulator statutes cluster in a single reform generation: the Electricity Regulatory Commissions Act of 1998 and its successor the Electricity Act of 2003, the Petroleum and Natural Gas Regulatory Board Act of 2006, the Airports Economic Regulatory Authority Act of 2008. Between the two generations lies the liberalisation of the 1990s, and the correlation is the article’s central claim in embryo: the regulator statutes appear exactly where private capital was invited into a network the State had monopolised, because a private entrant will not invest against a referee who is also the incumbent. Where no invitation was issued, or where the invitation stopped at construction contracts rather than operations, the departmental template survived untouched.

The regulator sectors: electricity as the complete instance

Electricity is the sector in which the regulator template was built first and built most fully, and Articles 252 and 253 of this series carry its detail: the pillar treatment in the Electricity Act, 2003 explained owns the statute’s provisions, and the pre-2003 lineage of the Indian Electricity Act, 1910 and the Electricity (Supply) Act, 1948 is treated in the article that follows it. For the field map, three features matter. The first is the reason the template was needed. Under the Electricity (Supply) Act, 1948 (Act 54 of 1948), electricity was supplied by State Electricity Boards that were vertically integrated departmental monopolies: they generated, transmitted and distributed power, advised the government that owned them, and in substance regulated themselves, with tariffs set as an act of government and cross-subsidy carried as an unexamined habit. The Electricity Regulatory Commissions Act, 1998 (Act 14 of 1998) broke that arrangement by creating the Central Commission and enabling State Commissions, and the Electricity Act, 2003 (Act 46 of 2003), in force from 10 June 2003, repealed the earlier statutes, unbundled the boards, opened access to the transmission network, and made tariff determination a reasoned, appealable act of an independent commission rather than a budgetary decision of a department.

The second feature is the completeness of the separation. Under the 2003 Act, the appropriate commission determines the tariff for generation, transmission and distribution within its sphere, grants licences, adjudicates disputes between licensees and generating companies, and enforces its orders, while policy remains with the governments: the Central Government frames the National Electricity Policy and the tariff policy, and State governments may direct subsidies, but a subsidy must be paid for through the budget rather than imposed on the tariff by silence, and a government direction binds the commission only on a question of policy and not on the outcome of a case. The third feature is the appellate chain. The Appellate Tribunal for Electricity, constituted under the 2003 Act, hears appeals from the commissions and, by cross-reference in other statutes, from the petroleum board as well, and its orders are appealable to the Supreme Court on a substantial question of law. A sector in which a consumer association can challenge a tariff order before a tribunal, and the tribunal can set it aside for want of reasoning, is a different legal world from a sector in which the charge is whatever a notification says it is, and the whole distance between the two governance models can be measured on that one axis.

The electricity settlement also shows the template’s internal limit, which recurs across the field. The commissions regulate the economics; they do not run the sector, and they do not control the two decisions that most determine its health, which are the appointment of their own members by the governments they regulate and the refusal of distribution licensees, most of them still government-owned companies, to operate at arm’s length from their owners. The regulator model in India is therefore real but bounded: independent in adjudication and tariff, dependent in constitution and finance. That bounded character is not a drafting accident shared by coincidence across statutes. It is the standard Indian compromise between the investor’s demand for a neutral referee and the State’s refusal to surrender the sector, and it reappears, with the proportions varied, in oil and gas and in airports.

Two working features of the electricity settlement complete the headline picture, because both recur as ambitions in the other sectors. The first is the consumer staircase: the distribution licensee must maintain an internal grievance forum, the Act’s framework adds a Consumer Grievance Redressal Forum for each licensee and an Electricity Ombudsman appointed by the State commission above it, so that a billing or supply grievance climbs two specialised rungs before it becomes litigation, and most grievances end on those rungs. No other sector in this field has built an equivalent staircase inside its statute, and the absence in the departmental sectors, where the user’s staircase has one step and it is the High Court, is among the strongest practical arguments the regulator model can make for itself. The second feature is harmonisation by conference: the central and State commissions meet in a Forum of Regulators, an arrangement the electricity framework supports, in which model regulations and common approaches are negotiated among commissions that no superior body commands. The Forum has no power to bind a State commission, and its products persuade rather than govern, which makes it the federal system’s characteristic contribution to regulatory design: coordination as a standing institution, in place of the hierarchy the Constitution’s scatter forbids.

A third feature deserves record because it is the point at which politics re-enters the regulated sector by the front door the statute leaves open. State governments remain free to promise electricity below the tariff the commission determines, and the 2003 Act regulates the promise rather than forbidding it: under Section 65, where a State government requires the grant of a subsidy to a class of consumers, the affected licensee is compensated by the government paying the subsidy amount in advance, in a manner the State commission directs, and the commission’s tariff order proceeds on the unsubsidised cost. The design is the regulator model’s treaty with electoral politics: the commission keeps the price honest and public, the government keeps the power to buy the price down, and the cost of the promise appears in a budget where it can be counted instead of accumulating invisibly in a licensee’s balance sheet. Where the treaty holds, the sector’s accounts show the subsidy as expenditure. Where State governments have promised without paying in full and on time, the model’s oldest disease, the gap between tariff and cost financed by neglect, reappears inside the regulated form, which is the standing proof that a regulator disciplines a government only to the extent the government accepts the discipline in its own payments.

The regulator sectors: oil and gas, and airports

The Petroleum and Natural Gas Regulatory Board Act, 2006 (Act 19 of 2006) created the Board that regulates the downstream half of the petroleum sector: refining, processing, storage, transportation, distribution, marketing and sale of petroleum, petroleum products and natural gas, with the production of crude oil and natural gas expressly outside its remit. The Board was established in June 2007 and the Act was brought into force on 1 October 2007. Its working instruments are authorisation and access: it authorises entities to lay, build, operate or expand pipelines as common carriers or contract carriers and to build city or local natural gas distribution networks, it declares pipelines to be common or contract carriers, it secures third-party access to them, it registers entities, lays down technical and safety standards, and it protects consumer interests and fosters competitive markets in the activities it covers. Its tariff work has a shape worth noting for the map, because it is regulation of charges between businesses rather than charges to households: the Board determines transportation tariffs for common carrier pipelines and, by its tariff regulations, has moved the interconnected national gas grid toward a unified tariff, in which the revenue requirements of the interconnected pipelines are pooled and the shipper pays by zone rather than negotiating or litigating each pipeline’s own tariff line by line. The device answers the specific pathology of a network built pipeline by pipeline under separate authorisations, in which gas grew more expensive with every border it crossed. It is also the petroleum board’s clearest demonstration of the regulator’s distinctive competence: no ministry negotiating bilaterally, and no court, could have pooled the tariffs of competing pipeline owners into one schedule, because the pooling is precisely a compulsory redistribution among licensees that only a statutory authority with tariff power can impose and defend. What it does not do is equally instructive. It does not set the retail price of petrol or diesel, which moved through administered pricing into a market-linked regime by executive decision rather than by any regulator’s order, and it does not govern the upstream, where exploration and production remain with the Directorate General of Hydrocarbons as a technical arm of the ministry and with contract administration under the production sharing contracts. The upstream rests on an older statutory pair that predates the regulator idea entirely: the Oilfields (Regulation and Development) Act, 1948 (Act 53 of 1948), which stands on Entry 53 of the Union List and provides for mining leases for mineral oils, and the Petroleum and Natural Gas Rules, 1959, under which petroleum exploration licences and mining leases are granted and royalties and dead rent are charged. Entry into the upstream is therefore by a licensing and contract system administered by the ministry: the explorer holds a licence or lease from the State, a production sharing contract with the Centre where the regime requires one, and sells into a pricing environment that executive decisions shape, with no commission between the producer and the government that is also its licensor, contract counterparty and principal customer through the public sector refiners. Read against the downstream board, the sector contains the field’s whole history in cross-section: a plan-era licensing statute for the resource, an administered price era that came and went without a statute of its own, and a regulator generation statute for the pipes, each layer addressed to the problem of its decade and none of them displaced by its successor. Oil and gas is therefore a half-regulated sector: a genuine statutory regulator for the pipes and the networks, departmental administration for the wells and the pump price. The namable asymmetry of the field appears inside this one sector, because the State exited transportation infrastructure and kept production, and the regulator follows the exit line exactly.

The Airports Economic Regulatory Authority of India Act, 2008 (Act 27 of 2008) regulates the economics of the major airports: it determines the tariff for aeronautical services at airports above the statutory passenger threshold, monitors the service quality standards, and decides the development fee and passenger service fee components that the older Airports Authority regime had handled as administration. The threshold is the sector’s telling detail. As enacted, the Authority’s remit ran to airports handling at least 1.5 million passengers a year or designated by notification; the amendment of 2019 raised the floor to 3.5 million passengers a year, which removed a tier of growing airports from independent tariff regulation and returned their charges to negotiation under the Airports Authority of India Act, 1994 framework. A legislature willing to narrow its own regulator’s jurisdiction by threshold, in a sector built on public private partnerships at Delhi, Mumbai, Bengaluru, Hyderabad and elsewhere, demonstrates the bounded character of the Indian regulator model from the other side: the regulator exists because concessionaires and airlines each demanded a neutral tariff forum, and its perimeter is set by how much neutrality the State is prepared to accept at any given time. Appeals from the Authority, which originally lay to a dedicated appellate tribunal, were routed by the tribunal mergers carried in the Finance Act, 2017 to the Telecom Disputes Settlement and Appellate Tribunal, a reminder that the appellate architecture of the regulator sectors is itself periodically rearranged by finance legislation rather than by the sector statutes.

The airport settlement’s structure explains why a tariff regulator was unavoidable there. The Airports Authority of India, constituted under the Airports Authority of India Act, 1994 (Act 55 of 1994), was the departmental operator of the country’s airports; the restructuring of the Delhi and Mumbai airports in 2006 transferred their operation, management and development to joint venture concessionaires for long terms against a revenue share, and greenfield concessions at Bengaluru and Hyderabad had preceded them, so the sector acquired, within a few years, private operators holding exclusive positions at the country’s principal gateways, an Authority that was landlord, licensor and former operator at once, and airlines and passengers paying charges that no independent body examined. The Act of 2008 is the direct answer to that configuration, and its confinement to aeronautical charges, leaving the operator’s commercial revenues from retail, property and other non-aeronautical activity outside the tariff determination except as the single till or hybrid till methodology of the tariff order takes them into account, is the negotiated boundary of the answer: the regulator prices the monopoly service, the market prices the rest, and the definition of the boundary is the sector’s standing litigation. Set beside electricity, where the commission prices the whole of the licensee’s regulated business, the airport settlement is the regulator model at its most deliberately partial, adopted for one class of assets in one generation of transactions and extended by threshold rather than by principle.

Aviation also supplies the field’s one clean separation between economic and safety regulation, and it is worth recording because the other sectors mostly fuse the two or neglect the second. The airports authority regulates the economics of the major airports; the safety of flight, the certification of aircraft and aerodromes and the investigation of accidents belong to the aviation safety establishment under the Aircraft Act, 1934, whose principal bodies, the Directorate General of Civil Aviation, the Bureau of Civil Aviation Security and the Aircraft Accident Investigation Bureau, were given statutory form by the Aircraft (Amendment) Act, 2020. The electricity sector separates safety less institutionally: the Central Electricity Authority, a statutory body continued under the 2003 Act, carries the technical and safety standard functions alongside its planning role, inside the same sector framework as the commissions but distinct from them. Railways, as the accident inquiry discussion has shown, separates accident investigation by the administrative device of housing the Commissioner under another ministry. Three sectors, three different answers to the same design question, and only aviation’s is a legislature’s considered separation of the referee of prices from the referee of safety.

The departmental sectors: the railways

The Railways Act, 1989 (Act 24 of 1989), which commenced on 1 July 1990 and replaced the Indian Railways Act, 1890, is the governing statute of the largest infrastructure network in the field, and it creates no regulator. The railway administration, meaning the Central Government acting through the Railway Board and the zonal railways, is the operator of the trains, the owner of the track, stations and rolling stock, the classifier of goods for freight, the fixer of fares and freight rates, the licensor of what little private participation the network admits, and the bearer of a statutory liability regime for accidents, death, injury and loss of goods that the same Act defines. Claims for compensation for accidents and for loss, destruction or deterioration of goods go to the Railway Claims Tribunal under the Railway Claims Tribunal Act, 1987, which is a forum for compensation, not an economic regulator. Safety investigation sits in a deliberately awkward place: the Commissioner of Railway Safety, who inquires into serious railway accidents, functions under the Railways Act but is placed administratively under the Ministry of Civil Aviation precisely so that the inquiry into a railway accident is not conducted inside the ministry whose organisation caused it, an admission inside the departmental model that some functions cannot credibly be self-performed.

Who investigates a serious railway accident in India?

The Commissioner of Railway Safety inquires into serious railway accidents, holding a statutory inquiry and reporting to the Central Government. The office functions under the Railways Act, 1989, but is administered through the Ministry of Civil Aviation rather than the Ministry of Railways, so that the investigation of the operator is not conducted by the operator’s own ministry.

The tariff position follows the model exactly. Passenger fares and freight rates are set by the railway administration and notified by the government; no commission tests them against cost, no tribunal hears a freight user’s appeal against a classification, and the cross-subsidy by which freight earnings carry passenger losses is a policy outcome inside the department’s own accounts rather than a reasoned order open to challenge. The case for this arrangement is made seriously and deserves its strongest form. The railway network is a single integrated system with safety interdependencies at every junction; its passenger fares carry a social obligation to move people at prices far below cost, which a commercial regulator applying cost-reflective tariff principles would either forbid or expose as a subsidy someone must fund; and a regulator superimposed on a network monopoly that remains publicly owned would regulate a single entity that is also its owner, which is a weaker discipline than the political accountability the department already answers to. The committee chaired by Bibek Debroy on mobilising resources for major railway projects and restructuring the Ministry of Railways, which reported in 2015, recommended an independent regulator, returning to a proposal first made by the expert group chaired by Rakesh Mohan in 2001 and repeated by the National Transport Development Policy Committee, and the Union Cabinet in April 2017 approved the creation of a Rail Development Authority to advise on pricing of services, protect consumer interests, promote competition and set performance standards. The Authority was to be created by executive order rather than by statute, its functions were recommendatory rather than binding, and it was not in the event set up as a working body. The sequence is the departmental model’s best self-portrait: three decades of committee consensus that an integrated operator should not price its own services unexamined, answered by an approval that changed neither the statute nor the tariff.

Why does electricity have an independent regulator but the railways do not?

Because the State left the operations in one and stayed in the other. Electricity was unbundled and opened to private generation and traders, and a market with rival operators required a neutral body to license them and fix their charges. The railways remained a departmental operator, so no forum was created, and fares, freight and liability stayed inside the department.

The counter-reading should also be carried fairly. Defenders of the departmental model point out that the electricity commissions regulate a sector in which the State also remains the dominant owner through its generation and distribution companies, so the contrast is not between a market and a monopoly but between two monopolies, one of which has an appeals process attached; that a rail regulator without a statutory charter, binding powers and control over access to the track would be a commentator rather than a regulator; and that the genuine structural question, which is whether train operations should be separated from track ownership so that rival operators can be regulated at all, is one no committee recommendation has yet persuaded a government to take. The field map should record the asymmetry without pretending it is an oversight. It is a choice, renewed each time a tariff is notified instead of determined, and its persistence in the sector with the largest number of daily users is the strongest single exhibit for the article’s claim that regulation in Indian infrastructure tracks the State’s withdrawal from operations and not the importance of the service or the number of people who depend on it.

Two further railway facts belong on the map because each tests the model. The first is the merger of the railway budget with the general budget from the budget of 2017-18, ending a presentation that had run since the 1920s on the recommendation of the Debroy committee. The merger was an accounting and accountability reform, not a regulatory one, but it belongs in this article because it removed the last formal device by which the department’s finances were presented as a self-contained commercial account, and with it the occasion on which fares were traditionally announced and defended in Parliament. The departmental model’s political discipline, never strong on price, lost its one scheduled annual exercise. The second is the admission of private operations at the network’s edges. Private container train operators were licensed from 2006 under an executive policy, running their own rakes on the public track against haulage charges the administration sets; freight terminals and private sidings multiply the pattern; and the Dedicated Freight Corridor Corporation of India, a government company, was incorporated to build and operate the eastern and western freight corridors as a distinct infrastructure manager on the same network. Each arrangement is instructive in the same direction: private capital has been admitted to use the department’s track, haul on it and build beside it, under charges the department fixes and terms the department drafts, which is participation without regulation, the exact inverse of the electricity settlement in which the regulator came first and the entrants followed.

The departmental sectors: national highways

The National Highways Act, 1956 (Act 48 of 1956) declares highways to be national highways, vests them in the Union, and provides for their development and maintenance by the Central Government, which may entrust the work to an authority or agency. The National Highways Authority of India Act, 1988 (Act 68 of 1988) created the Authority that became the sector’s executing agency, and the Authority builds, maintains and operates the national highway network, in large part by awarding stretches to concessionaires who finance, build and operate them for a concession period in return for toll or annuity payments. The Authority is a statutory body with real institutional weight, but in the taxonomy of this article it belongs on the departmental side of the line, because it regulates nobody: it is the government’s own developer and asset manager, it procures and supervises the concessions, it collects or supervises the collection of toll, and the fee for using a national highway is levied by the Central Government under Section 7 of the 1956 Act through the fee rules made under it, principally the National Highways Fee (Determination of Rates and Collection) Rules, 2008, and not determined by any body independent of the government that owns the road. The detailed treatment of this architecture, its declaration device and its compensation code is carried in the National Highways Act explained, and this guide holds it at the level the map requires.

The highways sector is also where the contract layer described later in this article was developed furthest, because a departmental sector that wants private capital without a regulator must manufacture credibility out of documents. The model concession agreements for build, operate and transfer toll projects, for annuity projects and for the hybrid annuity model that followed, the lenders’ substitution rights when a concessionaire fails, the escrow and termination-payment machinery, and the authority’s own conciliation committees of independent experts for disputes with contractors, are all substitutes, clause by clause, for the standing institutional assurance that a regulator provides in electricity. They worked well enough to build the largest highway programme in the country’s history, and their weakness is the mirror image of their strength: every protection a concessionaire enjoys exists because its contract says so, on terms the government side drafted, and a change of policy or a stressed balance sheet tests the documents rather than an institution. The comparison disciplines both models. Electricity bought investor confidence with an institution and argues about its appointments forever; highways bought it with drafting and argue about the drafting in every arbitration.

The toll itself shows the departmental method in its purest administrative form. The fee rules of 2008 prescribe the base fee by class of vehicle and by the character of the section, distinguishing ordinary stretches from structures such as bridges, tunnels and bypasses whose cost the fee is meant to recover, and they revise the base rates annually by reference to movements in the wholesale price index, so the charge at a plaza changes by formula and notification rather than by any body’s adjudication. Exemptions and discounts, for local traffic, for return journeys within a day and for monthly passes, are likewise prescribed by rule. A user who believes the fee at a particular plaza exceeds what the rules allow has no tariff appeal and no commission to petition; the remedy is a representation to the Authority and, failing that, the writ jurisdiction, on the administrative law grounds of illegality and irrationality. The contrast with a distribution licensee’s consumer, who can carry a metering or tariff grievance through the licensee’s internal forum, the consumer grievance redressal forum the electricity framework maintains and the ombudsman above it, before ever reaching the commission, is the contrast between a sector that built a staircase of forums and a sector that built a plaza.

The Authority’s finances complete the departmental picture. The highway programme is funded from the budget, from borrowings the Authority raises on the strength of its toll and annuity receivables, from the share of the cess on petrol and diesel that is credited to the Central Road and Infrastructure Fund and allotted to national highways, and from instruments that recycle finished assets: under the toll, operate and transfer model, a completed and tolled stretch is concessioned for a long term against an upfront payment, and the proceeds fund new construction, so the department sells tolls not yet collected to pay for roads not yet built. Each instrument is an executive and financial device created by policy and documentation. None required, or received, a financing statute of its own beyond the fund legislation, and the contrast with the electricity sector, where the tariff order itself is the financing instrument and must therefore be reasoned and appealable, states the two models’ deepest difference: in the regulated sector the price is the policy, and in the departmental sector the budget is.

The highways map has a second tier that the national programme’s prominence obscures. State highways and major district roads, which carry a large share of the country’s road traffic, belong to the State List and are governed by State highway legislation and State public works departments, with their own acquisition, ribbon development and encroachment provisions varying State by State; a minority of States have constituted road development corporations or boards on the departmental model, and none has created a road regulator. A project road may therefore pass in a single journey from a national highway under the 1956 Act, tolled under central fee rules and built by a concessionaire under a model agreement, to a State highway under a State Act, untolled or tolled by State decision and built by a department, with the traveller’s legal position changing at a boundary no signboard marks. The contrast is the field in miniature: two governance arrangements for the same physical activity, divided by a declaration in a gazette.

Ports, waterways, road transport, mining and atomic energy

Ports divide at the constitutional line between Entry 27 of the Union List and Entry 31 of the Concurrent List, and the two halves live under different law. The major ports were governed by the Major Port Trusts Act, 1963 (Act 38 of 1963), under which each port was run by a Board of Trustees and its tariffs were fixed by the Tariff Authority for Major Ports under the 1997 amendment, by reference to cost-based guidelines that port users and terminal operators alike criticised as cramping commercial pricing. The Major Port Authorities Act, 2021 (Act 1 of 2021), which received assent on 17 February 2021, replaced the 1963 Act for the eleven major ports it names, the twelfth major port, Kamarajar, standing outside it as a company under the Companies Act, and recast each port trust as a Board of a Major Port Authority with greater financial and operational autonomy, gave the Board the power to fix the scale of rates for the port’s services and assets, and so removed tariff-setting from the independent authority to the port itself, subject to the Act’s framework and to reference tariffs for public private partnership concessions granted under the earlier regime. Disputes of the kind specified in the Act go to an Adjudicatory Board constituted under it, with an appeal to the High Court. Private operation inside the major ports long predates the 2021 Act: berths and terminals were concessioned to private operators from the 1990s onward under the landlord pattern, in which the port retains the harbour, the channels and the land and the operator builds and runs the terminal, and those concessionaires priced their services within the tariff authority’s scales for the port, a regime under which the same container could attract a regulated ceiling set port-wise while the operator’s commercial problem was terminal-wise. The later tariff guidelines for such operators moved them toward market-linked rates against a reference tariff, and the 2021 Act’s transfer of rate-fixing to the port Board completes the movement: the landlord port has become, in law, the price setter for its own estate, supervised by its statute and its ministry rather than by a specialist tariff body. Beneath both halves of the ports map lies one older common layer: the Indian Ports Act, 1908 (Act 15 of 1908), which remains the general statute for the conservancy of ports, the powers of port officers, the safety of shipping alongside and the levy of port dues at ports to which it is applied, major and minor alike, a reminder that the field’s strata include a colonial conservancy code that the trust, authority and board statutes of the following century were built over rather than substituted for. The minor ports, which carry a large and growing share of the country’s cargo, are not touched by the central port statutes at all: they are administered by the maritime States under their own arrangements, through State maritime boards or port departments, under the Concurrent entry, with tariffs and concessions set by State governments. One coastline therefore contains both templates and a federal residue: a central port whose board fixes its own rates, a tariff authority abolished for those ports, and State ports that never had one.

Waterways follow the declaration pattern. The National Waterways Act, 2016 (Act 17 of 2016) declared 111 inland waterways to be national waterways, building on the five declared individually under earlier Acts, and development and regulation of the declared waterways rest with the Inland Waterways Authority of India, created by the Inland Waterways Authority of India Act, 1985 (Act 82 of 1985), a developmental authority of the departmental type. The Inland Vessels Act, 2021 (Act 24 of 2021) replaced the Inland Vessels Act, 1917 (Act 1 of 1917) and unified the regime for mechanically propelled vessels on inland waters, certification, survey and safety, administered substantially through the States. The development side of the waterways settlement is project-based: the Authority executes capacity augmentation on the declared waterways, of which the largest instance is the project for National Waterway 1 on the Ganga between Haldia and Varanasi, undertaken with World Bank assistance to build terminals, navigational aids and a guaranteed draft on a working river. The 2021 vessels legislation carried a federal argument of its own, since a uniform central regime for vessels on waters that are Concurrent and largely State-administered was opposed in several States as an encroachment, and defended by the Centre as the condition of treating a river system that crosses States as one waterway; the Act as passed keeps the administration substantially with the States while prescribing the uniform frame, the standard Indian resolution of a Concurrent subject in a network industry. Road transport is the field’s most State-heavy sector: the Motor Vehicles Act, 1988 (Act 59 of 1988), substantially amended by the Motor Vehicles (Amendment) Act, 2019 (Act 32 of 2019), is a central Act on a Concurrent entry, but registration, licensing, permits and enforcement belong to State transport departments, stage carriage economics are set by State governments, and no regulator, central or State, determines the sector’s charges. The sector’s operating institutions are themselves statutory and departmental in the oldest style: the State transport undertakings that run the bus networks are road transport corporations constituted under the Road Transport Corporations Act, 1950 (Act 64 of 1950), a central statute that creates the corporate shell and leaves the corporation inside the State government’s administrative family, its fares approved by the government that owns it and its losses carried as a social obligation of the State. Private stage carriage operates beside the corporations on permits granted by the State and regional transport authorities under the Motor Vehicles Act, with route and fare decisions taken by those authorities, so the sector’s competitive fringe is licensed by the same administration that owns its largest operator, the precise conflict the regulator template exists to dissolve, persisting undisturbed in the sector with the largest daily ridership in the field after the railways. The road safety apparatus, strengthened by the amendment of 2019 with a National Road Safety Board in advisory form, follows the pattern of the railway safety arrangement: expertise assembled beside the operating administration rather than authority constituted above it. Mining stands under the Mines and Minerals (Development and Regulation) Act, 1957 (Act 67 of 1957), treated in depth in the MMDR Act and mining law in India: a central statute containing the Entry 54 declaration, administered by State governments that grant and auction mineral concessions, collect royalty at rates the central Act’s Second Schedule prescribes, run the District Mineral Foundations created by the 2015 amendment, and answer for revision to the central tribunal machinery. The Foundation deserves its sentence on this map because it is the field’s one systematic device for returning a share of a sector’s rent to the people the sector displaces: a trust in each mining district, funded by contributions from leaseholders fixed by rule as a fraction of royalty, and applied to the benefit of mining-affected persons and areas, drinking water, health, education and infrastructure among them. The amendment of 2023 added an exploration licence for specified deep-seated and critical minerals, an instrument aimed at the field’s exploration deficit, in which the licensee explores and is rewarded from the eventual auction of the block rather than by a right to mine it, a deliberate separation of the explorer’s reward from the concession that the older statute had always fused. Atomic energy closes the set as the departmental model in undiluted form: Entry 6 of the Union List, the Atomic Energy Act, 1962, exclusive central control over production, development, use and disposal, a department that is itself the operator through its public enterprises, and a safety board created by executive order, with civil liability for nuclear damage separated into the Civil Liability for Nuclear Damage Act, 2010 (Act 38 of 2010). The safety arrangement states the departmental model’s limit in its sharpest form. The Atomic Energy Regulatory Board was constituted by a presidential order of 1983 under the Atomic Energy Act, and it regulates the safety of the installations of the department to which it is administratively attached; a Bill to replace it with a statutory Nuclear Safety Regulatory Authority was introduced in Parliament in 2011 and lapsed without enactment, so the one infrastructure sector in which the cost of regulatory failure is least forgivable is the one in which the regulator holds no statutory charter, a position defended on the ground that the Board’s technical authority within the atomic energy establishment is real, and criticised, including by the Comptroller and Auditor General in a performance audit of the Board, on the ground that legal independence cannot be borrowed from the body being regulated. The register note for this cluster asks for positions to be carried in their proponents’ terms and none adopted, and the article complies; the structural fact itself, an executive-order regulator inside an operating department, is not a position but a datum of the map.

Regulator or department: the field on one page

The table states the article’s argument in its tariff column. Read down that column and the pattern assembles itself: independent tariff determination appears in electricity, in the downstream of oil and gas, and at the major airports, the three places where the State ceased to be the principal operator; the charge is set by the operator or the owning government in railways, highways, ports after the 2021 Act, road transport and mining royalty, the places where it did not.

Sector	Principal statute	Who sets tariffs or charges	Appellate route	Acquisition route	Principal dispute forum
Electricity	Electricity Act, 2003	Central and State Regulatory Commissions	Appellate Tribunal for Electricity, then the Supreme Court	Sector provisions with the general acquisition framework	Commissions for licensee disputes; arbitration for contract disputes
Oil and gas (downstream)	Petroleum and Natural Gas Regulatory Board Act, 2006	The Board for regulated activities; retail fuel prices outside the Board	Appellate Tribunal for Electricity, then the Supreme Court	Pipeline right-of-user statutes and the general framework	The Board for entity disputes; arbitration for contract disputes
Airports	Airports Economic Regulatory Authority of India Act, 2008	The Authority at airports above the passenger threshold	Telecom Disputes Settlement and Appellate Tribunal	Airports Authority framework with the general acquisition law	The Authority; arbitration under concession documents
Railways	Railways Act, 1989	The railway administration by government notification	No tariff appeal; Railway Claims Tribunal for compensation	Chapter inserted for special railway projects in the 1989 Act	Railway Claims Tribunal; arbitration under contract conditions
National highways	National Highways Act, 1956 and the Authority Act, 1988	The Central Government by fee rules under the 1956 Act	No tariff appeal; writ jurisdiction of the High Courts	Sections 3A to 3J of the 1956 Act	Arbitration under the concession; statutory authority for acquisition disputes
Ports	Major Port Authorities Act, 2021	The Board of each Major Port Authority	Adjudicatory Board under the Act, then the High Court	Port and general acquisition framework	Adjudicatory Board; arbitration under concession documents
Road transport	Motor Vehicles Act, 1988	State governments and transport authorities	Statutory transport appellate tribunals of the States	General acquisition law	State transport authorities and tribunals
Mining	Mines and Minerals (Development and Regulation) Act, 1957	Royalty by the central Act’s schedule; prices by the market	Revision to the Central Government’s tribunal machinery	General acquisition law and the coal and mines acquisition statutes	Revisional authority; arbitration under concession and contract terms

Two cells of the table deserve a sentence each before the argument moves on. The mining row’s tariff cell is the honest exception that tests the article’s claim: royalty is fixed by a central schedule, not by a regulator and not by the operating State, because royalty is not a price for a service but a payment for a publicly owned resource, and the field’s logic treats it as rent rather than tariff. The airports row’s appellate cell is the field’s untidiest: an economic regulator for airports whose appeals travel to a tribunal built for telecom and broadcasting disputes, placed there by a finance Act’s tribunal merger rather than by any judgment about sectoral fit.

Land: the shared foundation, taken sector by sector

Every sector statute in this field eventually meets the same physical question, and it is the question on which projects most often stall: under what law is the land taken, by what procedure, and at what compensation. The general acquisition statute, the Right to Fair Compensation and Transparency in Land Acquisition, Rehabilitation and Resettlement Act, 2013 (Act 30 of 2013), whose scheme is treated in full in the LARR Act, 2013 explained, is not the statute under which most infrastructure land is in fact acquired. Infrastructure land is acquired principally under the sector statutes themselves. Land for a national highway is acquired under Sections 3A to 3J of the National Highways Act, 1956, a self-contained code inserted by amendment in 1997, under which the Central Government notifies its intention, hears objections, declares the land vested in itself free of encumbrances on publication of the declaration, and has the amount determined by the competent authority, with a reference to an arbitrator appointed under the Act and thereafter to the courts on the statutory terms, and with no social impact assessment and no consent requirement of the kind the 2013 Act imposes on its own acquisitions. Land for a special railway project is acquired under the chapter inserted into the Railways Act, 1989 by the amendment of 2008, Sections 20A to 20P, on a similar pattern of notification, objection, declaration, vesting and determination by a competent authority. Urban rail has its own instrument in the Metro Railways (Construction of Works) Act, 1978, another of the Fourth Schedule enactments, under which land for a metro railway is acquired through a comparable sector procedure, which is why a metro project and the general acquisition Act meet on the same two-instrument terms as a highway. The pattern generalises across the set: Parliament’s habit, whenever it builds a network, is to write the network’s land code into the network’s own statute or a companion statute, because a linear project administered through the general acquisition machinery of each district it crosses was, in the plan era’s experience, a project that did not finish. Land for pipelines is taken as a right of user under the Petroleum and Minerals Pipelines (Acquisition of Right of User in Land) Act, 1962, which acquires the right to lay the pipeline while ownership remains with the landowner. The Electricity Act, 2003 carries its own approach through the powers it confers by reference to the telegraph framework for placing lines, while generation and other project land moves under the general law or State arrangements. The practical rule for any reader tracing a project is therefore to begin in the sector statute, not in the general Act: the procedure that will actually be followed, the authority that will conduct it and the forum that will hear the compensation dispute are all found there.

The 2013 Act reaches into every one of these procedures, and the mechanism of its reach is the most consequential single detail in the land law of infrastructure. The sector acquisition statutes were not repealed, and the 2013 Act as enacted largely exempted them: Section 105 left the Act’s provisions inapplicable, in the first instance, to the enactments listed in its Fourth Schedule, thirteen enactments that include the National Highways Act, 1956, the Railways Act, 1989, the Electricity Act, 2003 and the Atomic Energy Act, 1962, but required the Central Government, within one year of the Act’s commencement, to issue a notification applying the Act’s provisions on the determination of compensation and on rehabilitation and resettlement to those scheduled enactments. The notification route failed amid the wider contest over the 2013 Act: the amendment effort of the winter of 2014 and 2015, carried forward by ordinance more than once while Parliament declined to pass the Bill, would have rewritten the scheduled sectors’ position on different terms, and it was allowed to lapse in 2015, leaving the order route as the only available instrument, and the connection was instead completed by the Right to Fair Compensation and Transparency in Land Acquisition, Rehabilitation and Resettlement (Removal of Difficulties) Order, 2015, issued as S.O. 2368(E) on 28 August 2015 under the Act’s removal of difficulties power and expressed to operate from 1 January 2015, which applied the compensation provisions of the First Schedule and the rehabilitation and resettlement provisions of the Second and Third Schedules to acquisitions under the Fourth Schedule enactments. Article 257 of this series owns the detail of that settlement. Its consequence for the field map is exact and frequently misstated in both directions: an acquisition for a national highway follows the highways procedure, its stages, its competent authority and its arbitrator, and not the 2013 Act’s procedure of social impact assessment and consent; but the money is computed on the 2013 Act’s terms, its market value determination, its multipliers and solatium under the First Schedule, and the affected family’s rehabilitation and resettlement entitlements come from the Second and Third Schedules. The compensation half of the settlement changed the economics of every linear project. Under the First Schedule, the market value of the land is determined on the statutory basis, multiplied in rural areas by a factor that rises with the distance of the land from urban centres within the prescribed band, and solatium is payable at one hundred per cent of the compensation amount, a doubling device the 1894 Act’s successor world had known at thirty per cent; the consequence is that a highway’s land bill under the highways procedure, computed on 2013 terms, bears no resemblance to the sums the same code produced before the order, and project costs, bid design and the authority’s land banking strategy have all been repriced around the settlement. The procedure half, meanwhile, retained the sector code’s speed: no social impact assessment, no consent clause, vesting on declaration. Critics of the settlement attack exactly this asymmetry, a generous price reached by a summary process; defenders answer that a linear project cannot be held to a district-by-district consent architecture and that the money is where the landowner’s real interest lies. Both descriptions are accurate about the half they describe, which is why the settlement has endured. Pages that apply the general statute directly to a highway or railway acquisition get the procedure wrong; pages that treat the sector statute as untouched by 2013 get the compensation wrong. The two halves of the acquisition are governed by two different instruments, and identifying which half a dispute is about is the first analytical step in any infrastructure land case.

One feature of the highways code deserves its own scrutiny, because it connects the land layer to the arbitration question that dominates the contract layer. Where a landowner disputes the amount determined by the competent authority for a national highway acquisition, the National Highways Act refers the determination to an arbitrator appointed by the Central Government, the very government acquiring the land and paying the amount. The design was defended as expert and fast, and for years it operated as the standard illustration, in acquisition practice, of the State appointing the judge of its own liability. The constitutional discomfort with unilateral appointment that the Supreme Court worked out in the contract cases, culminating in the Constitution Bench decision of 2024 discussed later in this article, speaks directly to the instincts behind that design, although the acquisition arbitrator’s position rests on the statute and on the landowner’s surviving remedies to the courts on the statutory terms rather than on the contract doctrine the 2024 case decided. Electricity supplies the complementary case from the other direction: for placing transmission and distribution lines, Section 164 of the Electricity Act, 2003 empowers the appropriate government to confer on a licensee the powers of the telegraph authority under the Indian Telegraph Act, 1885, so that lines are laid under a nineteenth century power to place and maintain telegraph lines, with compensation for damage rather than acquisition of the soil, an arrangement that treats most grid land questions as questions of access and damage instead of ownership, and reserves full acquisition for the substations, generating stations and project sites where ownership is genuinely needed.

Two further features of the land layer cut across the sectors. The first is the federal one. Acquisition and requisitioning of property is Entry 42 of the Concurrent List, so State legislatures may also enact acquisition law, and several States have framed their own statutes or obtained amendments of the 2013 Act for their territories through the route of presidential assent, which is why the compensation position for the same sector can differ across a State boundary. The second is that the sector statutes’ vesting provisions are drafted for speed: land vests in the government free from encumbrances on the declaration, possession follows on the statutory timetable, and the landowner’s remedies run to the amount rather than to the taking, a design that suits linear projects, in which a highway or a railway line cannot be re-routed around a single contested parcel, and that correspondingly concentrates all litigation energy on valuation. The account given here states the general position across sectors; a live acquisition turns on the notifications and awards in its own record, and the limitation periods and reference procedures in the sector statute are unforgiving of delay, so the affected landowner’s position needs advice taken on those documents rather than on any general map, including this one.

The contract layer: concession without a partnership statute

Above the land and under the tariffs sits the layer in which most infrastructure is actually built and financed, and its defining fact is a negative one: India has no statute governing public private partnerships. No Act defines a concession, prescribes how one is awarded, allocates its risks, regulates its renegotiation or governs its termination. A concession agreement is, in law, a contract, and it rests on three supports: the general law of contract, the sector statute under which the awarding authority holds and can part with the rights being conceded, and the administrative documents by which the government standardises its own bargaining. Those standardising documents are the model concession agreements, prepared and revised administratively for each sector and each project model, for toll and annuity and hybrid annuity highways, for port terminals, for airports and for other sectors, and they carry the real constitutional law of the partnership economy: the allocation of traffic risk and construction risk, the change in law clause, the escrow of revenues, the lenders’ right to substitute a failing concessionaire, the termination payments calibrated to the cause of termination, and the dispute clause. The appraisal presentations through which project sanctions are examined can be opened and read without proprietary software through the ReportMedic presentation viewer for infrastructure appraisal decks, a small practical point that matters because the documentary record of a project, from appraisal to concession, is the closest thing the field has to a public code.

What is a model concession agreement in law?

A model concession agreement is not law. It is a contract template published by the government for a sector and a project model, which the awarding authority adapts for a bid. Its clauses bind because the concession incorporates them as contract terms, not because any statute enacts them, and departure from it is made by drafting, not by amendment.

The contract families through which the layer operates repay a compact description, because the same four shapes recur across sectors and the differences among them are differences in who carries the demand risk. In an engineering, procurement and construction contract, the authority pays for the asset outright and carries every risk after completion; the private party is a builder, not a partner, and the arrangement belongs to procurement law rather than to partnership at all. In a build, operate and transfer toll concession, the concessionaire finances and builds the asset, collects the user charges for the concession period and carries the traffic risk, which is the model’s strength as an incentive and its weakness as a financing proposition when traffic disappoints. In the annuity variant, the authority pays the concessionaire fixed half-yearly amounts and keeps the toll itself, so the demand risk returns to the State and the concessionaire’s discipline narrows to construction quality and maintenance. The hybrid annuity model, adopted for national highways from 2016, splits the construction cost between an authority contribution during construction and annuities thereafter, a design written directly out of the failure of the pure toll concessions that preceded it. Around these families runs the supporting apparatus of viability gap funding, under which a capital grant from the government bridges the distance between a project’s economics and its social value, awarded through an inter-institutional appraisal process under a scheme first framed in 2005 and revised since. None of these instruments is a creature of statute. Each is a policy and a drafting practice, revised by committee and circular, and the rights of the parties in any given project are exactly and only what the executed documents say.

The absence of a partnership statute has been examined rather than merely noted. The committee on revisiting and revitalising the public private partnership model of infrastructure development, chaired by Vijay Kelkar, which reported in 2015, recommended a formal framework for post-award renegotiation, an institutional mechanism for the partnership programme, and consideration of dedicated legislation, and the renegotiation question it identified remains the layer’s open wound: a concession priced on traffic or demand projections that fail, or on a policy environment that changes, has no statutory route to revision, so revision happens by negotiation under the shadow of termination, by relief clauses drafted in advance, or not at all, and each route distributes the loss differently between the conceding authority, the concessionaire, its lenders and the user. The honest assessment cuts in both directions, as the brief for this cluster requires. The flexibility of the contractual model is real: one legal technique has served toll roads, annuity roads, ports, airports, transmission lines and urban infrastructure without a fresh statute for each, and models are revised as experience accumulates, which a statute would make slower and more political. The costs are also real: drafting quality varies, the standard terms are written by one side, renegotiation has no published discipline, and the citizen whose toll or tariff pays for the project is not a party to the document that governs it. A field whose central transactions rest on contract is a field whose law is found in documents most readers will never see, which is precisely why the map of institutions and forums, which are public, carries the analytical weight in this article.

Disputes: arbitration, regulators and the boundary between them

An infrastructure dispute can begin in one of three kinds of forum, and routing it correctly is the practical skill the field demands. Contract disputes under a concession, a construction contract or a supply arrangement travel the route the contract prescribes, which in this field means arbitration under the Arbitration and Conciliation Act, 1996 (Act 26 of 1996), with the courts in support and in supervision. Regulatory disputes, by contrast, belong to the statutory forum the sector statute creates: disputes between licensees and generating companies in electricity go to the appropriate commission and not to an arbitral tribunal of the parties’ choosing, tariff and authorisation disputes go to the petroleum board and the airports authority in their spheres, and acquisition disputes go to the competent authority and the statutory arbitrator or reference court the sector statute names. The boundary is jurisdictional, not consensual: an arbitration clause cannot pull a matter out of a statutory forum the legislature has made exclusive, and a regulator cannot decide a pure contract claim merely because the contract concerns its sector. The recurring litigation of the field is boundary litigation, fought over whether a dispute is in substance regulatory or contractual, and its outcomes decide whether a claim is heard by a commission applying a tariff order or by a tribunal applying a concession.

Within the contractual territory, a single question has reshaped the documents: who chooses the arbitrator. Standard conditions in government and public sector contracts long provided for appointment by the authority itself, or selection from a panel the authority curated, sometimes from among its own serving or retired officers. The Supreme Court dismantled the strongest form of that practice in stages. In TRF Limited v. Energo Engineering Projects Limited, decided in 2017, the Court held that a person who is himself ineligible to act as arbitrator, because of an interest in the outcome, cannot nominate another in his place. In Perkins Eastman Architects DPC v. HSCC (India) Limited, decided in 2019, it held that a party with an interest in the dispute cannot unilaterally appoint a sole arbitrator, because the appointment power is itself an advantage in the contest. The position was then restated for the whole field by a Constitution Bench of five judges in Central Organisation for Railway Electrification v. M/s ECI SPIC SMO MCML (JV), decided on 8 November 2024, which arose, fittingly for this article, out of a railway contract. The majority opinion of Chandrachud CJ, for himself and Pardiwala and Misra JJ, held that a clause in a public private contract that allows one party unilaterally to appoint a sole arbitrator, or compels the other party to choose its arbitrator from a panel that party has curated, offends the principle of equal treatment of parties and, in contracts with the State and its instrumentalities, Article 14 of the Constitution, and it overruled the earlier railway panel decision that had upheld such clauses; Hrishikesh Roy and Narasimha JJ wrote separate opinions, concurring on central points while differing on parts of the route, with Narasimha J resting the conclusion principally on the Arbitration Act itself rather than on Article 14. The Court directed, in paragraph 170.7 of the majority opinion, that the law laid down would apply prospectively to arbitrator appointments made after the date of the judgment, a direction the Court expressed in relation to three-member tribunals. Panels are not forbidden: a public authority may maintain one, provided it is broad based and the other party is not compelled to choose from it. For the field, the consequence is documentary and immediate: thousands of concession and works contracts carry appointment clauses the judgment governs, and the arbitration clause of an Indian infrastructure contract can no longer be read as a routine annexure, because it is the clause that decides whether the forum will be believed.

The regulatory side of the boundary has its own settled discipline, and it is one of deference with a hard edge. Courts reviewing a commission’s tariff order do not rework the tariff: the determination of price by an expert body applying a statutory methodology is treated as a function into which the court’s own view does not enter, and interference is confined to illegality, procedural failure and perversity. In PTC India Limited v. Central Electricity Regulatory Commission, decided in 2010, the Supreme Court worked the connected boundary between the Central Commission’s power to make regulations and its power to decide cases, holding that the regulatory power must be exercised through the instruments the Act prescribes and that a tariff order cannot be displaced by guidelines that were never made into regulations. The pair of propositions, that a commission acting within its statutory instruments is near-final on price and that a commission stepping outside them is not protected at all, is the judicial settlement under which the whole regulator model operates, and it is the reason the drafting of regulations, an activity no newspaper reports, is among the most consequential law-making in the field.

Where does an appeal against an electricity tariff order go?

An appeal against an order of a State or the Central Electricity Regulatory Commission lies to the Appellate Tribunal for Electricity, the tribunal the Electricity Act, 2003 constitutes for the sector, which also hears appeals from the petroleum and natural gas board. A further appeal lies to the Supreme Court only on a substantial question of law.

The third forum family is the constitutional one, and it is busier in the departmental sectors than anywhere else in Indian public law. Where no regulator and no tariff appeal exist, the aggrieved user’s remedy against a fare, a toll or a classification is a writ petition to the High Court under Article 226, tested on the standards of administrative law, and the departmental sectors generate their litigation accordingly, in challenges to toll notifications, to levy and exemption decisions, and to the fairness of departmental determinations. The writ court does not fix the charge; it examines the legality of the fixing. That remedial thinness is part of the price of the departmental model, and it is rarely counted in the model’s defence.

Two softening devices deserve record before the forums are left, because both show the departmental sectors borrowing the regulator model’s clothes. The highway authority maintains conciliation committees of independent experts, to which disputes with concessionaires and contractors can be referred for a negotiated settlement before arbitration is invoked, an acknowledgment inside the largest departmental programme that a standing body of neutrals, even a recommendatory one, resolves disputes that bilateral negotiation cannot. And across the regulator statutes, the civil courts are barred from entertaining matters the sector authority is empowered to decide, a standard clause that protects the specialist forum’s exclusivity and routes the disappointed party to the tribunal instead. Neither device changes the taxonomy. Conciliation advises and cannot bind; a jurisdiction bar protects a forum that exists. The departmental sectors have the first and mostly lack the second, having no forum to protect.

Stress, renegotiation and the insolvency interface

The recurring failure mode of the contract layer is the stressed project: a concession whose revenues fall short of the projections on which it was financed, a highway section whose traffic never arrives, a power project whose fuel or purchaser fails. The statutes of the field were not written for this failure, so its law is an interface between documents and a code. The Insolvency and Bankruptcy Code, 2016 (Act 31 of 2016) supplies the code: on a default, a financial creditor, an operational creditor or the corporate debtor itself may set the insolvency resolution process in motion before the National Company Law Tribunal, a moratorium descends on suits and enforcement against the concessionaire company, and the concession, which is usually the company’s only asset, falls to be dealt with inside a resolution process whose object is the company’s revival or its sale as a going concern. The interface generates the field’s characteristic frictions. A concession is a contract with a government authority, terminable on defaults that the insolvency process suspends; the authority is at once a counterparty, sometimes a creditor, and the guardian of a public service that cannot stop while lawyers argue; and the lenders, whose step-in and substitution rights were drafted into the concession for exactly this contingency, find those rights operating inside a collective process the concession documents did not contemplate. The Supreme Court has held that the Code prevails over inconsistent statutes and instruments by force of its own overriding clause, and the working position that has emerged is that essential public services are protected by keeping the operating asset in use while the claims against its owner are resolved, a compromise the Code’s text supports through the treatment of essential supplies and the moratorium’s limits.

The power sector wrote the stress chapter first and in the largest figures, and its episode supplies the interface’s cautionary tale about non-statutory fixes. When a generation of generation projects stalled on fuel supply, land and purchaser distress, the resolution pressure ran through the banking system, and the Reserve Bank’s circular of 12 February 2018, which imposed a strict timetable for resolving large stressed accounts and pushed defaulting power companies toward the insolvency process, was challenged by the affected producers and associations. In Dharani Sugars and Chemicals Limited v. Union of India, decided in 2019, the Supreme Court struck the circular down as beyond the Reserve Bank’s statutory power in the form in which it was issued, holding that a direction of that character had to be anchored to the authorisations the banking statute requires. The projects then travelled to the Code by the ordinary creditor route instead. The episode is this field’s standard demonstration that infrastructure stress is processed by general law or not at all: a regulator’s circular, a ministry’s committee and a scheme of restructuring each operate only within the power some statute confers, and the documents of the contract layer, however elaborately they provide for default, meet their limit where a general code or a constitutional court redraws the forum.

Renegotiation is the alternative to insolvency, and its law is thinner than its practice. Post-award renegotiation of concessions in stressed sectors, notably highways, proceeded for years through administrative devices rather than law: one-time fund infusions, premium deferral and restructuring for stressed toll projects, substitution of concessionaires with lender support, and the migration of new awards from the toll model, in which the concessionaire carries traffic risk, to annuity and hybrid annuity models, in which the authority carries most of it and the traffic risk that sank the earlier generation is removed from the private balance sheet altogether. The Kelkar committee’s recommendation of a standing renegotiation framework, made in 2015, addresses the gap directly: renegotiation conducted without a framework is renegotiation conducted by bargaining power, and its outcomes, whether generous or harsh, acquire no public discipline. Until a framework exists, the honest summary of this part of the field is that India resolves stressed infrastructure by contract surgery and insolvency process, under documents drafted for success, and that the scarcity of statutory law on failure is as much a feature of the field as the absence of a partnership statute it descends from.

The surgery has a documented history in the highways sector, and its instruments show a learning curve the field should record. When the toll concessions awarded in the build-out years began to fail on traffic, the first response was the rescheduling of the premium the concessionaires had bid: a committee chaired by C. Rangarajan framed a formula under which the premium payable to the Authority could be deferred and restructured for stressed projects against conditions, converting an immediate default into a stretched obligation. One-time fund infusion followed for languishing projects whose completion was near and whose failure would have wasted the sunk work, and the substitution route was used to replace concessionaires who could not continue with parties the lenders and the Authority could accept. Each device was administrative: a cabinet decision, a committee’s formula, a policy circular, applied project by project. Each worked, in the cases it reached, by re-cutting the original bargain in the direction the traffic figures required. And each confirmed the structural point this section exists to make, that in a field without a renegotiation law, renegotiation is an act of government discretion exercised under fiscal pressure, available in full measure to the projects large enough to command a committee and uncertainly, or not at all, to the small concessionaire whose failure is a local event. The Kelkar recommendation, read against this record, is not a plea for novelty; it is a plea that the discretion already being exercised be brought under published rules.

User charges: the fee, the tax and the toll

Almost every rupee the infrastructure sectors collect from the public arrives under one of two legal characters, and the character decides who may levy it, under what authority, and subject to what challenge. A tax is a compulsory exaction for public purposes, imposed by a legislature under a taxing entry, with no necessary correspondence between what the payer pays and any service the payer receives. A fee is a payment for a service rendered or a benefit conferred, levied under a regulatory or service power, and it carries the requirement the Supreme Court articulated in Commissioner, Hindu Religious Endowments, Madras v. Sri Lakshmindra Thirtha Swamiar of Sri Shirur Mutt, decided in 1954: that a fee, unlike a tax, rests on a quid pro quo, a special benefit to the payer as a class, with a broad correlation between the collections and the cost of the service, though no arithmetical equivalence between each payment and each benefit is required. User charges in infrastructure are fees in this sense when they are lawful at all: the toll on a national highway is the price of the use of a maintained road, levied by the Central Government under Section 7 of the National Highways Act, 1956 at the rates the fee rules prescribe and collected by the Authority or the concessionaire as its agent; the electricity tariff is a charge for supply determined by a commission against the cost of supply; the airport’s user development fee and the port’s vessel and cargo charges each answer, in their own statute, to the cost of the facilities used. The quid pro quo discipline is the user’s one general protection in the departmental sectors: a levy that outruns any correlation with the service, or that is imposed without the statutory authority the character requires, is open to challenge as a tax dressed as a fee, and the courts have struck levies on exactly that ground across the wider field of Indian fiscal law.

The same bill can lawfully carry both characters at once, and the electricity bill is the standing example. The tariff the consumer pays is a fee for supply, determined by the commission against the cost of serving. The electricity duty that many States levy on consumption or sale of electricity is a tax, imposed by State legislation under Entry 53 of the State List, which is taxes on the consumption or sale of electricity, and it is collected through the same bill by the distribution licensee as the State’s collection agent. Petrol and diesel carry the pattern into transport: the price at the pump includes excise and State value added tax imposed under taxing entries, which is one reason the petroleum regulator’s exclusion from retail price, noted earlier in this article, surprises only readers who expect a regulator to stand behind every figure on the bill. The distinction decides the remedy as well as the label. A fee is challenged for failing its quid pro quo or its statutory authority; a tax is challenged on the legislature’s competence and on the constitutional discipline of taxation, and confusing the two challenges is among the commonest errors in litigation over infrastructure charges.

The character question also explains the politics of the two models. In the regulator sectors, the fee is set by a body that must state its cost reasoning in an order, and the order can be appealed on its reasoning. In the departmental sectors, the fee is set by notification, the reasoning is the government’s, and the user’s protection is the constitutional discipline of the fee itself plus the writ court’s review of legality. A toll protest and a tariff appeal are therefore different legal events aimed at the same grievance, and a reader who understands why can read the field’s news correctly: the sectors generate organised tariff litigation where regulators exist, and episodic political agitation where they do not, because the law has given the first grievance a forum and the second only a street and a court of general jurisdiction.

The edges of the field: municipal services, pipelines and the unbuilt sectors

A field map owes the reader its own boundaries. Three territories adjoin the statute set this article has described and behave differently enough to be marked rather than absorbed. The first is municipal infrastructure: water supply, sewerage, drainage, solid waste and urban roads, which the Constitution assigns to the State List and, through the Twelfth Schedule inserted by the Seventy-fourth Amendment, to the municipalities as functions they may be endowed with. There is no central municipal services statute and no regulator; the sector’s law is State municipal legislation, State public health engineering departments and, in the larger cities, statutory water and sewerage boards on the departmental model, financed substantially by user charges whose legal character is municipal taxation and fee under State law. The second territory is the pipeline as a legal object in itself. A petroleum pipeline is authorised under the petroleum framework, its land is taken as a right of user under the Act of 1962 named earlier, its safety is governed by technical standards under the petroleum and explosives framework, and its crossings of highways, railways and rivers collect permissions from each network it passes under or over, so that a single linear asset assembles its legality from half a dozen regimes, none of which governs the whole. The third territory is the sector that was never separately built at all: logistics and warehousing, which operate under the general law of contract, warehousing legislation of limited scope, State industrial policy and the sector statutes of the networks they serve, without a statute of their own. A fourth territory belongs on the boundary for the opposite reason, because it tests the article’s claim from outside the brief’s eight sectors: telecommunications, which this series treats in its technology cluster, acquired its regulator, the Telecom Regulatory Authority of India under the Act of 1997, at the moment private entry into a departmental monopoly began, and then kept the department as licensor and as operator through its public companies alongside the regulator, producing the field’s most crowded institutional room: a licensor department, an operating public incumbent, private operators, a recommendatory and tariff regulator and a specialised appellate tribunal, all in one sector. The telecom room is untidy, and it is the strongest external confirmation of the map’s rule, because the regulator arrived there with the entrants, on schedule, and the frictions that followed have all been frictions between the regulator the entrants required and the department that never left. The map records these edges because the field’s central claim is a claim about statute-making, and the places where Parliament and the State legislatures have chosen not to make a sector statute at all, leaving a field to municipal law, to overlapping permissions or to contract alone, are data for the same argument: the legislature builds a framework where an operating structure demands one, and the demand has historically been made by the State’s own operating departments or by the private capital invited to replace them, and by nobody else.

The reform debate: three standing proposals

The field’s reform literature converges, across four decades, on three proposals that recur in committee after committee, and a guide should state each with its strongest objection, since all three are live and none is enacted. The first is a statutory regulator for the railways, recommended in substance by the Rakesh Mohan expert group of 2001, the National Transport Development Policy Committee and the Debroy committee of 2015, and approved in a diluted, advisory and executive form in 2017 without being brought into working existence. Its strongest form is the electricity argument transported: separate the pricing and access functions from the operator, give freight users and prospective operators a forum, and let an independent body publish the cross-subsidy’s size so that Parliament funds it knowingly or ends it. Its strongest objection is the one the departmental section recorded: a regulator over a single State operator, without separation of track from trains, regulates a monopoly it cannot discipline by competition and may merely relocate the politics of fares from a ministry that answers in Parliament to a commission that answers in an annual report.

The second proposal is a public private partnership statute, recommended for consideration by the Kelkar committee of 2015, which would define the concession, prescribe award discipline, create a standing renegotiation framework and an institutional home for the programme. Its strongest form is the rule of law argument: transactions of this scale and duration should rest on published law, with renegotiation conducted under principles a citizen can read, rather than on departmental documents revised by the party that also awards the contracts. Its strongest objection is the flexibility defence recorded in the contract section: the model documents have adapted across sectors and failures faster than any statute plausibly would, a partnership Act risks freezing one decade’s risk allocation into the next decade’s projects, and the renegotiation problem may be a problem of bargaining power and fiscal honesty that no statute can cure.

The third proposal is appellate consolidation: the periodic suggestion, acted on in part by the tribunal mergers of the Finance Act, 2017, that the sector tribunals be rationalised into fewer and stronger bodies, since small tribunals carry chronic vacancies, thin technical benches and uneven quality. Its strongest form is administrative: a tribunal that cannot sit cannot regulate, and the appellate stage is where the regulator model’s promise of expert, speedy review is most often broken in practice. Its strongest objection is the specialisation defence: tariff law for electricity, gas and airports shares a method but not a subject, and the mergers already made have placed airport appeals before a tribunal built for telecom and broadcasting without any visible gain in either field’s jurisprudence. The three proposals share a diagnostic core, and it is the note on which the analysis of the field properly ends before its verdict: each identifies a place where the governance model a sector happens to have produces a cost the sector’s own statute cannot reach, in pricing, in contracting or in appeal, and each has been pending long enough to suggest that the obstacle is not drafting but the distribution of advantage under the arrangements as they stand.

Reading the field: a navigation rule and a verdict

The working rule this article leaves is procedural. For any infrastructure question, ask three things in order. First, which List entry carries the sector, because that identifies the legislature and the government. Second, which governance model the sector statute adopts, because that identifies who sets the charge, who licenses the operator and where an appeal travels. Third, which layer the question inhabits, the statute, the land or the contract, because each layer has its own instruments and its own forum, and most errors in writing about this field come from answering a contract question out of the statute or a compensation question out of the general acquisition Act. The reader payloads follow the same order. The business reader gets the routing: the statute, the charge-setter and the forum for each sector, settled by the table. The aspirant gets the retrievable structure: electricity Concurrent at Entry 38, railways and atomic energy Union, mines split by the Entry 54 declaration, roads split by the Entry 23 declaration, and the regulator statutes of 2003, 2006 and 2008 against the departmental statutes of 1956, 1957, 1962 and 1989. The researcher gets the claim: that the presence of an independent economic regulator in Indian infrastructure tracks the withdrawal of the State from operating the service, and predicts the next regulator’s location more reliably than any argument from the importance of a sector, since it will appear, if it appears, in the sector where operations are next genuinely opened, and not in the sector whose users are merely most numerous.

The rule is worth one worked pass. Take a freight user who believes a railway goods classification overcharges his commodity. The List question answers itself: railways are a Union subject. The model question decides the case before it begins: the sector is departmental, so there is no tariff forum, no appeal against the classification to any tribunal, and the user’s routes are a representation inside the administration and a writ petition on legality, which is why railway freight grievances surface as trade association politics rather than as tariff litigation. Take, against that, a landowner whose field is notified for a national highway. The layer question governs: the procedure is the highways code of Sections 3A to 3J, the objection hearing is about the taking and not the price, the amount is determined by the competent authority with a reference to the government-appointed arbitrator, and the computation of the amount follows the 2013 Act’s First Schedule because the 2015 order says so, so the adviser’s first act is to separate the client’s procedural complaints, which belong to the sector statute, from the valuation case, which belongs to the compensation schedule. Take, finally, a concessionaire on a stressed toll road whose traffic has halved. The layer is contract, the counterparty is the highway authority, the documents decide what relief exists, the conciliation committee offers a negotiated door, arbitration offers an adjudicated one at the price the 2024 judgment has reset for appointment clauses, and behind both stands the insolvency code, in which the lenders’ substitution rights and the authority’s termination rights will be re-sorted by a tribunal that answers to none of the documents. Three questions, three different legal worlds, one field: that is the whole argument of this article in application.

The verdict on the field as a whole is therefore double. As an act of institutional design spread over seven decades, the statute book is more coherent than its reputation: each sector’s law fits its operating structure, the regulator statutes solved the credibility problem they were built for, and the departmental statutes preserve an integration their defenders can justify in the language of safety and social obligation. As a system for the user, it is indefensibly uneven: the citizen’s protection against an unjust charge depends on the accident of whether the State happened to exit the sector before the charge was imposed, the partnership economy rests on unpublished documents while its failures are resolved in public insolvency, and the land beneath every project is governed by a two-instrument settlement that almost no general account states correctly. The field is navigable, which is the most this article claims, once its organising distinction is named: find who operates the service, and the rest of the sector’s law explains itself.

What would change the map is equally statable, and the article ends by stating it, because a claim that cannot say what would falsify its prediction is a slogan. If train operations were separated from the track and opened to rival operators in substance rather than at the margins, the railway would acquire a regulator within a Parliament or two, because the electricity sequence would repeat: entrants would demand a neutral keeper of access and price, and the department could not supply one. If a major regulated sector were renationalised in operation, its commission would wither into an advisory office within a decade, whatever its statute said, because a regulator with one regulated entity that is also its owner has nothing to adjudicate. And if Parliament ever enacts a partnership statute with a renegotiation framework, the field’s centre of gravity would move from documents to law in a single commencement, and this article’s contract section would become a history of the interim. None of the three events is scheduled. The map as drawn is the map the reader will use, and its asymmetry, regulator where the State withdrew and department where it stayed, is the most durable single fact in Indian infrastructure law.

Frequently Asked Questions

Q: Which laws govern infrastructure in India?

Infrastructure in India is governed sector by sector, not by any single code. The principal statutes are the Electricity Act, 2003 for power, the Petroleum and Natural Gas Regulatory Board Act, 2006 for downstream oil and gas, the Airports Economic Regulatory Authority of India Act, 2008 for major airports, the Railways Act, 1989, the National Highways Act, 1956 with the National Highways Authority of India Act, 1988, the Major Port Authorities Act, 2021 for major ports, the Motor Vehicles Act, 1988 for road transport, the Mines and Minerals (Development and Regulation) Act, 1957 for mining, and the Atomic Energy Act, 1962. Across all of them run the general acquisition law of 2013, the arbitration law of 1996 and the Insolvency and Bankruptcy Code, 2016.

Q: Which infrastructure subjects are Union, State or Concurrent?

The Seventh Schedule scatters the field. Railways, declared national highways, major ports, airways, carriage by railway, sea or air, atomic energy, oilfields and, by declaration, mines and mineral development are Union subjects. Roads other than national highways and communications generally are State subjects, and State power over mines and industries is subject to the Union entries. Electricity is Concurrent, as are mechanically propelled vehicles, ports other than major ports, shipping on inland waterways other than declared national waterways, economic and social planning, and the acquisition and requisitioning of property. The placement of a sector predicts which legislature wrote its statute and which government administers it.

Q: Which infrastructure sectors have independent regulators?

Electricity has the fullest structure: the Central Electricity Regulatory Commission and the State Electricity Regulatory Commissions license operators and determine tariffs, with appeals to the Appellate Tribunal for Electricity. Downstream oil and gas has the Petroleum and Natural Gas Regulatory Board, which authorises pipelines and gas distribution networks and secures third party access, though it neither governs production nor sets retail fuel prices. Major airports have the Airports Economic Regulatory Authority of India for aeronautical tariffs above a passenger threshold. Railways, national highways, road transport and atomic energy have no independent economic regulator, and in ports the tariff authority that existed under the earlier statute was removed by the Major Port Authorities Act, 2021.

Q: How is land acquired for infrastructure projects?

Principally under the sector statutes rather than under the general acquisition Act of 2013. Land for a national highway is acquired under Sections 3A to 3J of the National Highways Act, 1956 through notification, hearing of objections, a declaration that vests the land in the Central Government, and determination of the amount by the competent authority, with arbitration under the Act. Land for special railway projects follows the chapter inserted into the Railways Act, 1989. The procedure of the 2013 Act, including social impact assessment, does not govern these acquisitions, but since the order of 2015 the compensation is computed on the 2013 Act’s First Schedule terms and rehabilitation and resettlement entitlements follow its Second and Third Schedules.

Q: Is there a law governing public private partnerships?

No. India has no statute that defines a public private partnership, prescribes how a concession is awarded, allocates its risks or governs its renegotiation. A concession agreement is a contract, resting on the general law of contract, on the sector statute under which the awarding authority holds the rights it concedes, and on the model concession agreements that governments publish administratively for each sector and project model. Those models carry the working rules on risk allocation, escrow, lender substitution and termination payments, and they bind as contract terms once incorporated into an executed concession, not as legislation. The committee chaired by Vijay Kelkar recommended in 2015 that dedicated legislation and a standing renegotiation framework be considered.

Q: Who resolves an infrastructure contract dispute?

The forum depends on the character of the dispute. A dispute under a concession, construction or supply contract travels the route the contract prescribes, which in this field ordinarily means arbitration under the Arbitration and Conciliation Act, 1996, with the courts in support and supervision. A dispute that the sector statute commits to a regulator, such as a dispute between a licensee and a generating company in electricity, belongs to the commission and then to the Appellate Tribunal for Electricity, and an arbitration clause cannot remove it to a private tribunal. Acquisition disputes go to the competent authority and the statutory arbitrator or court the sector statute names. Challenges to departmental charges go to the High Courts in writ jurisdiction.

Q: Which infrastructure statutes were recently replaced?

The pattern of replacement is continuous across the field. The Electricity Act, 2003 repealed the Indian Electricity Act, 1910, the Electricity (Supply) Act, 1948 and the Electricity Regulatory Commissions Act, 1998. The Railways Act, 1989 replaced the Indian Railways Act, 1890. The Major Port Authorities Act, 2021 replaced the Major Port Trusts Act, 1963 for the major ports it names. The Inland Vessels Act, 2021 replaced the Inland Vessels Act, 1917, and the National Waterways Act, 2016 consolidated the individually declared national waterways into a single declaration of 111. The Motor Vehicles Act, 1988 had earlier replaced the Act of 1939, and was itself substantially amended in 2019.

Q: How do user charges and tolls get fixed?

It depends on the sector’s governance model. Electricity tariffs are determined by the Central and State Regulatory Commissions in reasoned orders that can be appealed to the Appellate Tribunal for Electricity. Aeronautical charges at major airports are determined by the Airports Economic Regulatory Authority of India, and regulated gas transportation tariffs by the Petroleum and Natural Gas Regulatory Board. The toll for a national highway is levied by the Central Government under Section 7 of the National Highways Act, 1956 at rates prescribed by the fee rules of 2008, and railway fares and freight are set by the railway administration through government notification, so in the departmental sectors the charge is an act of the owning government rather than of an independent body.

Q: What is the difference between a fee and a tax in infrastructure law?

A tax is a compulsory exaction for public purposes, imposed under a taxing power, with no necessary connection between the payment and any service to the payer. A fee is a charge for a service rendered or a facility used, and it must rest on a quid pro quo. The Supreme Court in the Shirur Mutt case of 1954 held that a fee requires a special benefit to the payer as a class and a broad correlation between the collections and the cost of the service, though not an exact equivalence for each payer. Tolls, tariffs and port charges are defended as fees on this basis, and a levy that outruns any correlation with the service is open to challenge as a tax imposed without taxing authority.

Q: Can a signed concession agreement be renegotiated in India?

Yes, but no statute governs how. A concession binds as a contract, so its terms can be revised only as the contract permits or as the parties agree, and Indian law supplies no general doctrine entitling a concessionaire to revision when traffic, demand or policy disappoints the projections on which the bid was priced. Relief therefore comes from clauses drafted in advance, such as change in law and force majeure provisions, from negotiated restructuring of the kind applied to stressed highway projects through premium deferral and fund infusion, or from termination and re-award. The Kelkar committee of 2015 recommended a formal renegotiation framework precisely because revision by bargaining power alone produces outcomes that vary with the parties rather than with principle.

Q: What happens to a highway concession when the concessionaire enters insolvency?

The concessionaire is almost always a special purpose company whose principal asset is the concession itself. On default, a creditor or the company may commence the corporate insolvency resolution process under the Insolvency and Bankruptcy Code, 2016 before the National Company Law Tribunal, and a moratorium then restrains suits, enforcement and, in substance, steps that would strip the company of its assets while resolution is attempted. The highway authority’s termination rights under the concession, and the lenders’ substitution rights drafted into it, must operate inside that collective process, and the Code overrides inconsistent instruments. In practice the road continues to operate while the claims against its owner are resolved, by resolution, substitution or sale of the company as a going concern.

Q: Does the petroleum and natural gas regulator set the price of petrol and diesel?

No. The Petroleum and Natural Gas Regulatory Board, created by the Act of 2006, regulates the downstream infrastructure of the sector: it authorises pipelines and city gas distribution networks, declares pipelines to be common or contract carriers, secures third party access to them, registers entities and lays down technical and safety standards. The production of crude oil and natural gas is expressly outside the Act. Retail prices of petrol and diesel stand outside the Board altogether: they were administered by government for decades and moved to a market linked regime by executive decision, with the oil marketing companies revising prices, and no regulator’s tariff order stands behind the price at the pump.

Q: Who is responsible for railway safety in India?

Safety functions are divided inside the departmental structure. The railway administration itself operates the network and maintains the track, signalling and rolling stock under the Railways Act, 1989, and is responsible for the safe running of trains. The independent element is the Commissioner of Railway Safety, who inquires into serious railway accidents, inspects new lines before they open to traffic and reports to the Central Government. The Commissioner functions under the Railways Act but is administered through the Ministry of Civil Aviation, a deliberate separation intended to keep the investigation of the operator outside the operator’s own ministry. There is no statutory safety regulator for the railways comparable to the economic regulators of other sectors.

Q: How many national waterways has Parliament declared?

One hundred and eleven. Five waterways were declared national waterways individually under separate Acts over the decades, beginning with the Ganga stretch in 1986. The National Waterways Act, 2016 then declared 111 inland waterways to be national waterways in a single statute, including the five earlier declarations, and assigned their development and regulation to the Inland Waterways Authority of India, the statutory authority created in 1985. Declaration makes a waterway a Union responsibility for development for shipping and navigation; it does not itself dredge a channel or build a terminal, and the declared list is considerably longer than the list of waterways carrying regular commercial traffic.

Q: Does the Centre need a State’s consent to declare a national highway or acquire land for one?

No, at neither stage. A highway becomes a national highway by the Central Government’s declaration under the National Highways Act, 1956, which is the act that moves the road from the State List to Entry 23 of the Union List; States are consulted in practice and most declarations begin as State roads the Centre takes over, but State consent is not a statutory precondition. Acquisition then follows the same central hand: land for the declared highway is acquired by the Central Government under Sections 3A to 3J of the 1956 Act through its own competent authority, and the Railways Act, 1989 likewise empowers the Centre for special railway projects, although acquisition in general is a Concurrent subject. A State’s cooperation matters greatly in practice, for land records, policing and rehabilitation administration, and a State may legislate on acquisition within its competence, but the landowner’s remedies run to the amount payable, through the Act’s arbitrator and the courts, rather than to stopping the vesting.

Q: What does the Appellate Tribunal for Electricity do?

The Appellate Tribunal for Electricity is the appellate body the Electricity Act, 2003 constitutes for the power sector. It hears appeals against orders of the Central and State Electricity Regulatory Commissions and of the adjudicating officers under that Act, including tariff orders, licensing decisions and penalties, and it can confirm, modify or set aside the order appealed against. Parliament later made it the appellate forum for the Petroleum and Natural Gas Regulatory Board as well, so gas pipeline and distribution disputes travel to the same tribunal. Its decisions are appealable to the Supreme Court only on a substantial question of law, which makes the tribunal the effective final forum on most tariff and regulatory questions in the two sectors.

Q: Why do infrastructure contracts in India so often end in arbitration?

Because the documents are built that way and the alternatives are slow or unavailable. Concession agreements and works contracts in this field descend from standard conditions that have always routed disputes to arbitration, and the statutory forums regulate licensees and tariffs, not the run of contract claims about variations, delays, escalation and termination. A contractor faced with a departmental employer also prefers a tribunal to suing the government in courts carrying heavy dockets. The result is a very large population of arbitrations against public authorities, which is why the Supreme Court’s 2024 Constitution Bench ruling on unilateral appointment of arbitrators in public contracts matters so widely: it governs the clause that decides who will hear most of those disputes.

Q: What is the difference between a concession and a licence?

A licence is a permission granted by a regulator or authority under a statute, on the statute’s terms, to carry on a regulated activity; it creates no property in the licensee beyond the permission, and the grantor may vary its conditions within the statute. A concession is a contract by which a public authority transfers specified rights, typically to finance, build and operate an asset and to collect its revenues for a fixed period, in return for the concessionaire’s investment and performance obligations; the authority’s promises bind it as contract, and the concessionaire’s protections, including compensation on termination, are those the document states. Electricity distribution uses licences under the 2003 Act; highways, ports and airports use concessions, which is why their law is found in documents rather than in a regulator’s orders.

Q: What is a change in law clause in a concession agreement?

It is the clause that allocates the cost when the legal environment moves after the bid. A change in law clause typically defines the events that count, such as a new statute, an amendment, a change in a tax or a judicial or regulatory interpretation, that occurs after the bid date, and then prescribes the consequence: the concessionaire is restored, so far as money can do it, to the position it would have occupied had the change not occurred, by an adjustment of the concession period, the fee or a payment from the authority, subject to thresholds and to a duty to pass back gains when a change runs in the concessionaire’s favour. In a country without a partnership statute, this clause is the project’s principal private law protection against public law itself.

Q: Who governs the minor ports that the Major Port Authorities Act does not cover?

The States. Ports other than major ports are a Concurrent List subject, and in practice the maritime States govern their minor ports through State maritime boards, as in Gujarat, Maharashtra and Tamil Nadu, or through port departments of the State government where no board has been constituted. The State authority allots waterfront land, grants concessions and licences to private terminals and captive jetties, fixes the charges for the port’s services and administers the Indian Ports Act framework for conservancy and vessel matters within its limits. The Major Port Authorities Act, 2021 governs only the major ports it names, so a single coastline can hold a central port under that Act and State ports under entirely State made arrangements.