Agricultural Land and the Law of Valuation: From Estate Duty to SARFAESI

This article examines the legal framework governing the valuation of agricultural and plantation land under the SARFAESI Act, 2002, tracing the evolution of specialist valuer registers from estate duty and wealth-tax laws to the present. It highlights gaps in the company-law valuation regime and proposes reforms to ensure that agricultural and plantation assets are valued by appropriately qualified professionals.

An analysis of the framework governing the competence of the valuer when agricultural and plantation land is sold under the Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Act, 2002 (SARFAESI Act) — tracing the specialist valuation registers Indian revenue law has maintained since 1968, their omission from the company-law valuation regime of 2017, and the modest repairs the existing statute book already makes possible.

The weight of one opinion

The issue is simple, and it has never been squarely answered for enforcement: who is competent to value land that grows something when a bank sells it under the SARFAESI Act? The Act permits a secured creditor — a lender holding security, typically a mortgage — to sell the secured asset without the intervention of a court.

When a bank enforces a mortgage under the SARFAESI Act, the valuation on which it acts is a professional opinion, not a decree. Before selling immovable property, the authorised officer (the bank officer who conducts the sale) must obtain a valuation from an “approved valuer”, and must then fix the reserve price — the minimum bid at which the property may be sold — in consultation with the secured creditor.¹ Around that figure sit the bank’s own credit committee, its policies and its standard operating procedures. Yet the opinion carries more weight than its legal status suggests. It is the only expert input the process requires from outside the creditor’s own establishment. No court or Tribunal examines the figure before the hammer falls; the borrower usually first sees it in the sale notice. Everything downstream — reserve price, auction, confirmation, and the shortfall the borrower still owes — is arithmetic performed upon it. The Supreme Court has therefore insisted that the secured creditor must act in good faith and make every endeavour to obtain the maximum price for the secured asset.²

The Act itself steps back from farmland: security interests created in agricultural land are outside it altogether.³ What reaches sale is land at the margins of that exclusion — plantations and garden lands, which High Courts in some States have held to be outside “agricultural land” for this purpose and in others not⁴, and land whose agricultural character is disputed on the record. For precisely this land, the number is hard to get right, easy to get wrong, and easiest of all to get wrong quietly. That makes it worth asking a question the Indian statute book has answered several times, in several incompatible ways.

The gap in the literature

Scholarship on the SARFAESI Act has concentrated on the definition of “agricultural land” in Section 31(i), on the limits of judicial review of enforcement measures, and on the balance between creditor rights and borrower protection. The competence of the valuer who produces the figure on which everything turns has received almost no sustained attention. This article seeks to fill that gap. It traces the statutory history of specialist valuation registers from the estate-duty era to the present, shows how the company-law valuation regime dispensed with asset-class distinctions without discussion, and maps the practical failures that follow when those distinctions are ignored in enforcement.

Land that grows is a different asset

An acre under rubber is not an acre of ground with trees standing on it. Its worth lies in what it will yield: the age of the stand and its place in the tapping cycle, the cost of replanting and the unproductive years that follow, irrigation and processing infrastructure, the labour economics of the district, the price cycle of the crop, and the timber value of the trees themselves. Valuing it is an exercise in capitalising income, not in comparing plot rates. Guidance values fixed for stamp duty — the minimum rates at which the State values land for the purpose of duty, known variously as guideline value or circle rate — are framed as per-acre slabs and capture little of this in either direction.

Legal overlays complicate matters further: land-reform ceilings and plantation exemptions, restrictions on conversion and on who may buy, and tenancy protections. All of these compress or fragment the market against which any “market value” must be tested.

Classification is contested before valuation even begins. The Revenue record may describe a survey number as agricultural while the ground has long carried something else, or the reverse. The Supreme Court has held that the character of the land is to be assessed with reference to its actual use and purpose, not merely the entry in the Revenue records⁵, and that the burden of establishing that the property is agricultural land, for the purpose of the statutory exclusion, lies on the party claiming the benefit.⁶ The materials for the judgment — entries, deeds, an inspection — vary in quality. The question sits where revenue law, land-reform law and fact meet. It is a genuinely difficult call, made early, and drastic in consequence: it decides whether the Act applies at all and, as the sections below show, it shapes who values the land.

Seven decades of specialist registers

Indian Revenue law answered the competence question early, and answered it with specialisation. The Estate Duty Act, 1953 required the Central Government to appoint “qualified persons” as valuers within a year of its commencement.⁷ Because the provision did not use the word “prescribed”, the qualifications came by executive notification rather than by rule — a habit of subordinate law-making that has marked the field ever since.

The 1968 Notification organised valuers under eleven heads and already fenced ordinary immovable property off from land that grows: the general category was “immovable property (other than agricultural lands and coffee estates)”, and seven years’ practice was demanded.⁸ The 1975 Notification, superseding everything before it, refined this into ten categories—agricultural lands; coffee, tea, rubber and cardamom plantations; forests; mines and quarries (each a separate head with its own qualifications), besides immovable property, machinery and plant, shares and business assets, jewellery, works of art, and life interests.⁹ It cut the experience requirement from seven years to five, but took back with one hand what it gave with the other: sufficient experience in the specified field became a distinct condition, full-time practice became the rule, and every valuer had to declare that he had no interest in the asset he valued.

Wealth tax borrowed before it built. From 1957 to 1972 there was no separate wealth-tax register; the estate-duty panel did double duty.¹⁰ In 1972 Parliament created the wealth-tax register proper, and the rules built the class walls in from the first day: the general valuer was defined, on the face of the rule and of every certificate issued under it, as a valuer of “immovable property (other than agricultural lands, plantations, forests, mines and quarries)”, while separate sub-rules with separate qualifications — agriculture graduates, plantation experience, forestry, mining — governed the excluded classes.¹¹ In 1988 the experience threshold doubled from five years to ten, and a minimum gross-receipts test entered the practice route; valuers already on the register were grandfathered, on conditions of continued standing.¹²

Nor were these registers appendages of taxes that spared farms. Estate duty reached agricultural land in the States that consented to the levy, and did so until Parliament switched the levy off agricultural land in 1984, months before the duty itself ended in 1985.¹³ Wealth-tax excluded agricultural land at its birth in 1957, brought it in from 1970 — Parliament’s competence to do so was tested and upheld by the Supreme Court¹⁴ — and let it go only with the 1993 shift to a short list of specified assets.¹⁵ Income tax exempts agricultural income and leaves rural agricultural land outside the definition of a capital asset, but since 1970 it has treated agricultural land within specified urban limits as a capital asset whose gains must be measured.¹⁶ The specialist heads were built while these levies were alive — 1968, 1972 and 1975 all fall within that window — and income-tax work still needs them.

Then the tax died and the register did not. The wealth-tax levy ended on 1 April 2016¹⁷, but the register survived because other statutes and institutions had borrowed it — banks for empanelment, other proceedings for approved valuers. Parliament has now re-enacted the whole scheme: the Income-tax Act, 2025 carries the registration provision forward, and the Income-tax Rules, 2026 transcribe the old rule with its class walls intact.¹⁸ Whatever else may be said of this lineage, its premise held steady: competence in valuation is asset-specific, and a person qualified to value a house is not, by that fact alone, qualified to value a rubber estate.

The premise abandoned

The company-law valuer register set the premise aside without discussing it. The Companies Act, 2013 created a new class of registered valuers¹⁹, and the 2017 Rules under it recognise exactly three asset classes: land and building; plant and machinery; and securities or financial assets.²⁰ “Land and building” swallows everything from a city flat to a plantation. Entry runs through civil engineering, architecture and planning; nothing in the eligibility conditions or the examination syllabus touches agronomy, plantation economics or forestry. There is no agricultural class, so no one can register for one — and no one can be disciplined for practising outside it.

The abandonment is the stranger because the State, acting for itself, still insists on the old premise. When land is compulsorily acquired, the acquisition statute directs that trees and plants be valued with the help of experienced persons in the fields of agriculture, forestry, horticulture or sericulture, and standing crops with agricultural expertise.²¹ When the State takes a plantation, agricultural competence is mandatory; when the same plantation is sold in enforcement, none is required.

The enforcement statutes are wired to these registers unevenly. The debt-recovery legislation of 1993 names no register at all.²² The Security Interest (Enforcement) Rules, 2002 originally defined an “approved valuer” simply as a valuer approved by the board of the secured creditor; in 2007 the definition was substituted to require registration under Section 34-AB, Wealth-tax Act, 1957 together with board approval — so secured enforcement points at the register that preserves the specialist classes, a register that continues in force and is now carried forward verbatim into the Income-tax Rules, 2026.²³ The insolvency framework draws instead on the company-law register, which has none.²⁴ Banking regulation keeps a foot in the older scheme too: as late as November 2025, Reserve Bank’s credit-risk directions for co-operative banks still point bank boards to the wealth-tax qualifications when constituting valuer panels.²⁵ And the corporate-law amendments now before Parliament reorganise the governance of the company-law register — registration, valuers’ organisations, discipline — without revisiting what its members are competent to value.²⁶

One reason the change passed unnoticed is that there was no one to notice it. Valuers have worked alongside the Revenue Authorities since the 1950s, yet the profession never acquired a statutory professional body of the standing that chartered accountants or company secretaries possess. When the 2017 Rules dispensed with the specialist land classes, no such body existed with the authority to object.

How enforcement actually works

Set the architecture against practice, and five failure points emerge.

First, one asset, two classifications — Registration under Section 34-AB is class-specific, and the general certificate states on its face that it covers immovable property other than agricultural lands, plantations, forests, mines and quarries.²⁷ Bank empanelment is likewise asset-class specific: panels are constituted class by class, and a valuer is empanelled against the class of his registration. The difficulty lies in the classifications the two registers supply. The tax register carves land that grows into separate heads; the company-law register places every parcel, from a city flat to a rubber estate, in the single class of “land and building”. A panel drawn on the first classification carries the agricultural head; a panel drawn on the second has no such head to carry. Whether a certificate can be read against the security therefore depends on which register’s classification the panel follows — a question the two sets of rules have never been made to answer together.

Second, the bare-land answer — Where the mismatch between certificate and asset is noticed, a familiar answer appears in valuation reports: the estate was inspected and found without a productive stand; replanting would be required; the land was therefore valued as bare dry land. Yet whether land under rubber is a plantation past its prime or bare land with residual trees is itself a plantation judgment — yield read against the age of the stand, the cost of replanting and the immature years that follow, the worth of standing timber, shade and irrigation. The older scheme reserved exactly that judgment for the specialist class. And the stakes are doubled because States differ on whether plantations are “agricultural land” for the Act at all: the same finding can bear on whether the Act applies as well as on what the land is worth.

Third, two valuations, one borrower — The same land often carries two professional opinions in the same lender’s records: one obtained at origination, another at enforcement, years apart and frequently far apart in amount. Both are commissioned from the lender’s panel; the borrower selects neither. Whatever explains the distance — market movement, method, the condition of the estate — the difference is absorbed by the borrower and the guarantor, because the sale may lawfully close at the reserve price while the unpaid balance remains recoverable.

Fourth, the State’s own number contradicts the sale — Guidance values exist so that no private sale escapes stamp duty on the State’s estimate of worth. Income tax law goes further, deeming the guidance value to be the consideration when private parties transact below it.²⁸ Yet an enforcement sale may fix a reserve price at a fraction of the guidance value of the very same survey numbers, and the law requires neither a floor nor recorded reasons. The result is a public record in conflict with itself: the registering authority collects duty on its own higher figure, while the enforcement file certifies that the land is worth far less. Whatever the true value, both numbers cannot be accurate descriptions of one asset on one date.

Fifth, discipline follows the register — The wealth-tax scheme carried its own policing — a bar on salaried employment, restrictions on practice, removal from the register for misconduct. After the levy was withdrawn in 2016 the occasions to administer that discipline grew fewer, but the register and its conditions remained in force. With the Income-tax Act, 2025 and the Income-tax Rules, 2026 re-enacting the scheme, conditions of practice included, the machinery is activated once more. The company-law regime has an active disciplinary apparatus but no agricultural class to enforce: a land-and-building valuer who values a plantation breaches no condition of that registration, because the registration never drew the distinction. Discipline, in short, runs where the register runs; the class the company-law register never drew is the one gap the revived machinery cannot reach.

Possible objections

Three objections deserve serious attention. First, it may be said that recovery must remain swift, and that adding specialist filters will slow auctions and depress recoveries. The answer is that the specialist classes already exist, and the register that carries them is already named in the Security Interest (Enforcement) Rules, 2002; what remains is for the rules to provide, in terms, that the certificate be read against the security.

Second, it may be argued that modern valuation is at bottom a financial exercise, so that a competent land-and-building valuer can pick up plantation economics on the job: yield data, on this view, can be discounted like any other cash flow. The argument deserves a charitable answer, and a firm one. A civil engineer understands zoning and construction costs; but whether a stand will repay tapping for another decade, what replanting will cost, and what the immature years will forgo are agronomic judgments before they are financial ones. Income capitalisation for a rubber stand is not a variant of the comparable-sales method used for urban plots, and the older statute book recognised as much for seven decades.

Third, some will say that the borrower can always challenge the valuation after the sale. That is true in form, but the challenge is costly, delayed, and framed against a completed auction; the asymmetry of information and resources is structural. A safeguard that operates only after the sale is an imperfect substitute for competence before it.

The repair

None of this needs new theory; the law’s own materials suffice.

First, the company-law register — the one insolvency proceedings draw on — should carry asset-class granularity for land that grows: agricultural land and plantations as distinct classes with distinct entry qualifications. The templates have existed since 1968 and 1975, and survive, transcribed, in the Income-tax Rules, 2026.

Second, assignment should be bound to class where the class already exists: an authorised officer selling land that the revenue records or the mortgage deed describe as agricultural or plantation should be required to engage a valuer registered for that class, and to state the classification of the land before the valuation is commissioned. The classification cannot be an output of the valuation it governs.

Conclusion

The position, then, is this. For seven decades the statute book has held to a single premise: land that grows is a different asset, and valuing it calls for a different competence. That premise was written into the estate-duty notifications of 1968 and 1975, built into the wealth-tax register in 1972, and carried forward intact into the Income-tax Rules, 2026. The specialist classes are not a relic; they are current law, and the enforcement rules themselves name the register that keeps them. What the last decade added is a second register that classifies land without the distinction, so that either classification may govern a valuation. The repair asks for nothing new. It asks only that the two classifications be brought to agree, so that the competence the law has maintained for land that grows since 1968 travels with the land wherever the land is valued.