Two Banks, Two Different Legal Opinions on the Same Property: Which One Do I Trust?

You applied to two banks for the same home loan, on the same flat, with the same set of documents. One bank's legal opinion says the title is clear and marketable. The other's says the title is not clear. Nothing about the property changed between the two. So which opinion do you trust, and how do you tell which one is actually right?

In short (2026)

Neither opinion is automatically correct, and the disagreement is not proof that one advocate is careless. A bank's legal opinion is one empanelled advocate's professional judgment on a presumptive title, not a certificate of title guaranteed by the government. India registers deeds, not ownership, so two competent advocates can honestly reach different conclusions on the same property.

Before you pick a side, settle three things:

1. Is the bank even objecting to the title, or just declining the loan? Banks run a separate legal check and a separate technical or valuation check, and they decline just as often on valuation, property age, a negative-area policy or loan-to-value as they do on a genuine title defect. A lending "no" is not always a title "no".
2. Which opinion searched more, and stated its finding more precisely? Trust the one that searched the longer period, that names the specific defect with the register or section it sits in, and that actually checked the two things a clean Encumbrance Certificate (EC) can hide: pending court cases and an equitable mortgage.
3. Get one independent, standardised title search that reproduces the whole process and states the defect precisely, so you stop trusting a coin toss between two bank formats.

The rest of this page shows you how to do all three.

Why did two banks reach opposite conclusions on the same title?

Start with the thing almost no builder blog will tell you: in India, registering a sale deed does not, by itself, confer or guarantee ownership. The Registration Act, 1908 makes registration compulsory for a sale of immovable property (Section 17) and limits what an unregistered document can prove (Section 49), and the Transfer of Property Act, 1882 requires a registered instrument for a sale (Section 54). But the register records that a transaction happened. It does not certify that the seller actually owned what he sold.

The Supreme Court put it plainly in Suraj Lamp & Industries v. State of Haryana and again, recently, in Mahnoor Fatima Imran v. Visweswara Infrastructure (2025): registration "gives notice to the world that such a document has been executed" but is "not to confer an unimpeachable validity on all such registered documents." Property registration alone does not confer ownership, and a buyer "must verify the entire chain of ownership, not merely rely on registered documents."

This is what lawyers mean by a presumptive title system. Unlike a Torrens system, where the state maintains a single register and guarantees the title on it, India puts the burden of proving title on the buyer and the buyer's lender. There is no government guarantee to appeal to.

That single fact explains the disagreement. A legal opinion is not a machine reading of a guaranteed register. It is an advocate's opinion on the risk that the seller's title is defective, formed from the documents that advocate examined and the searches that advocate ran. There is no national standard that tells every advocate what period to search, which registers to pull, or how much doubt is fatal. Each bank empanels its own advocates, each advocate uses their own format and their own risk appetite, and the same file can therefore produce a "clear" from one and a "not clear" from another. The disagreement is a symptom of a system with no common standard, not evidence that one advocate is wrong.

Is the bank even saying the title is bad, or just declining the loan?

This is where most people misread the situation, so resolve it first. When a bank evaluates a property for a home loan, it runs two parallel processes:

• Legal verification, done by the empanelled advocate, checks the chain of ownership over the past 13 to 30 years, the sale deed and prior link deeds, the approved building plan, the no-objection certificate, the occupancy certificate and the EC.
• Technical verification, done by the bank's approved valuer or civil engineer, is an on-site check of measurements against the sanctioned plan, construction quality, layout violations and the property's market value.

A bank can decline the loan on the technical or the commercial side without there being any title defect at all. Banks maintain "negative areas" they avoid for poor resale value or local disputes. Older buildings, roughly 30 years and above, have lower residual life and get shorter tenures or a lower valuation. An illegal extension can be deducted from the valuation or sink the whole file. And because banks typically fund only 75 to 90 percent of the assessed value rather than your asking price, a low valuation can read like a rejection.

So before you agonise over two "legal opinions," confirm they are both actually about the title. Ask each bank, in writing, for the specific ground: is it the legal opinion, the technical valuation, or a credit-policy call? Very often the property is fine, one bank simply does not fund that building type, that locality or that loan-to-value, and there is no title problem to solve at all.

One bank searched 13 years, the other 30: does the depth explain the gap?

If both objections really are about title, the next likely cause is scope. The conservative practice for a title search is an unbroken 30-year chain of ownership. Many banks historically accepted a 13-year search as a shortcut, and a 30-year search costs more and takes longer, which is exactly why bank policies differ.

Depth changes the answer. A 13-year search that begins in, say, 2013 will read perfectly clean if the defect sits in a 2006 transaction, a mortgage that was never formally released, a partition that left out a coparcener, or a break in the chain. A 30-year search reaches back far enough to catch it. So an opinion that says "clear" over 13 years and one that says "not clear" over 30 years can both be honest and still contradict each other. The longer search is the more reliable one.

Two related scope differences do the same thing:

• Which documents each advocate saw. One may have examined the full mother-deed chain, the other only the seller's current sale deed. Less paper, less risk visible.
• Which survey number the EC was pulled on. An EC is only as good as the survey or property number it was searched against. A NIL EC on the wrong survey number reads clean because it is a search of the wrong parcel. If the two banks searched different numbers, they were effectively searching two different properties.

When you compare the opinions, compare their scope first: the period searched, the documents listed as examined, and the exact survey or property number each ran the EC on.

The EC was clean for both, so why did one still flag a problem?

Because two of the most serious title risks in India do not show up on an Encumbrance Certificate at all, and whether an advocate catches them depends on whether that advocate bothered to look elsewhere.

• A pending court case. A suit affecting the property, a partition claim, an injunction or an ownership dispute, is never registered in the sub-registrar's records, so it can never appear on the EC. Under Section 52 of the Transfer of Property Act, the doctrine of lis pendens means that if you buy a property while a suit about it is pending, you are bound by the outcome of that suit. The only way to catch it is a litigation search across the relevant courts by party name and by village or survey number. One advocate ran that eCourts search and flagged the suit. The other relied on the clean EC and did not.
• An equitable mortgage. When a borrower deposits the original title deeds with a bank to create a mortgage (Section 58(f) of the Transfer of Property Act), that mortgage often does not appear on the EC. It is meant to be filed with CERSAI, the central charge registry, which anyone can search for about ten rupees, both by the asset and by the borrower's PAN. An advocate who searched CERSAI would see a live charge that a clean EC completely hides.

So "both ECs were clean" does not settle the argument. It usually means one advocate did the extra searches, eCourts and CERSAI, and the other did not. The opinion that checked more is the one to trust.

How do I actually decide which opinion to trust?

Stop comparing the verdicts and start comparing the work behind them. Ask each bank for the legal opinion or its scope, and reconcile the two on five points:

1. Search period. 30 years beats 13. A longer clean search is stronger evidence than a shorter one.
2. A named defect beats a vague verdict. "Title not clear" tells you nothing. "A partition suit, O.S. 123 of 2019, is pending in the civil court, so Section 52 lis pendens attaches" tells you exactly what is wrong and whether it is fixable. Trust the opinion that names the register, the document or the section, not the one that gives a one-line conclusion.
3. Registers actually checked. Did each advocate go beyond the EC to eCourts and CERSAI, and did they check the survey number on the deed against the revenue record? An opinion is only as good as the databases behind it.
4. Clear, clear with conditions, or not clear. Advocates often certify a title as acceptable "subject to conditions," for example, produce the release of an old mortgage, or obtain a specific no-objection. A conditional clearance is not the same as a rejection. Read whether the second bank raised a genuine defect or simply a condition the first bank folded into its sanction.
5. The date of each opinion. If the two opinions were issued weeks or months apart, a new charge, a new suit or a fresh attachment could have arisen in between. The later opinion may simply be more current.

Work through those five, and in most cases you will find the two opinions are not really in conflict. One searched deeper, or checked a register the other skipped, or is objecting to valuation rather than title, or attached a condition the other bank quietly accepted.

The permanent fix: one standardised title search, not two conflicting opinions

The reason you are stuck choosing between two verdicts is that each bank's opinion is written for that bank, in that advocate's format, to that bank's risk appetite. Neither was written to be compared with the other. That is the gap.

The way out is a single, independent title search that does not belong to either bank: a full 30-year chain of ownership, the EC pulled on the correct survey number, the revenue records, a CERSAI charge search, an eCourts litigation search and the prohibited-property and 22A checks, all in one place, ending in a standardised rating that names each defect and grades the risk. Instead of "clear" from one advocate and "not clear" from another, you get one comparable report that reproduces the whole process a bank advocate follows and states, precisely, what is wrong and whether it can be fixed.

That is exactly what a LegiScore report is built to do. It runs the same searches a bank's panel advocate runs, across every government database, and returns a single standardised title rating with each risk named and sourced, so you are not left trusting whichever of two bank formats happened to be more or less thorough.

If two banks have handed you two different answers, get one independent report and settle it on the facts.

Get an independent standardised title search report

Frequently asked questions

Can two legal opinions on the same property both be correct? Yes. India uses a presumptive-title system, so a legal opinion is a professional judgment on the risk of a defect, not a reading of a guaranteed register. Two competent advocates who search different periods, examine different documents or apply a different risk threshold can both be reasonable and still disagree. The way to break the tie is to compare their scope and the precision of the defect each names, not the one-line verdict.

One bank approved and another rejected. Does that mean the title is clean? Not necessarily. First check whether the rejection is even about the title. Banks run a separate technical and valuation check and decline on property age, negative-area policy, illegal construction or loan-to-value, none of which is a title defect. If the objection is genuinely a title objection, the safer reading is that the rejecting bank searched deeper or checked a register the approving bank skipped, not that the approval clears the property.

To be safe, should I just go with the bank that approved the loan? No. The approving bank protects its loan, not your ownership. If the other bank flagged a real, named defect, for example a pending suit or an unreleased mortgage, taking the loan does not cure it. You would own a property with the defect and a loan on top of it. Resolve the specific objection first, then choose your lender on rate and terms.

Can I use one bank's legal opinion at the other bank? Usually not. Most banks insist on an opinion from their own empanelled advocate and will not accept a report addressed to a different lender. That is precisely why a property gets re-examined at each bank and why the answers diverge. An independent title search does not replace the bank's own opinion, but it lets you go into each application already knowing what any competent advocate will find.

How far back should a title search go, 13 years or 30 years? Thirty years is the conservative standard and the one to rely on when the answers conflict. A 13-year search is a shortcut that can read clean simply because the defect lies further back in the chain. If one opinion is based on 13 years and the other on 30, the 30-year search is the more reliable of the two.