What an Auction Headline Doesn't Tell You About Indian Art

Sometime in 2022, I started paying attention to auction results.

Not because I was flipping art. I'd been collecting for a while by then, mostly Indian Modernists (Husain, Souza, Raza, some younger contemporaries), and I was doing what many collectors do: buying work I loved without putting every purchase into a spreadsheet. Then I started pulling hammer prices from Pundole's and Saffronart and comparing similar works. I wanted to know what those prices could tell me, and what they left out.

A public sale is useful evidence, but it is not an investor return. Auction records show the works that reached the room and sold. They do not show what every owner paid, how long the work waited for a buyer, or what remained after fees, insurance, storage, and restoration. You can inspect individual results in Saffronart's artist records. What you cannot do responsibly is take one record sale, turn it into a market return, and compare it with a fund that publishes a continuous, net-of-fees series.

The problem is that the information a serious buyer needs is still difficult to assemble in one place.

Price histories exist, but they are uneven. The information a buyer needs is scattered across catalogues, gallery records, private conversations, condition reports, and ownership documents. A headline price is easy to find. A defensible comparison between two different works is much harder. Provenance and condition are still questions you have to investigate work by work.

This is not unique to India. Art markets everywhere have gaps between what is public and what a buyer needs to know. In India, that gap is one reason I think better records and a more usable secondary market matter.

What that means practically:

Prices are hard to compare. Similar works can be offered on very different terms in a private sale, an auction, or a gallery. A buyer who does the homework may find an opportunity. A seller without access to the right buyers may accept less than a stronger market could have offered.

Liquidity is occasion-dependent. You can own a genuinely valuable work and find it difficult to realise that value quickly. The exit options are: consign to a major auction house (takes months, takes commission, uncertain outcome), sell privately (requires network), or hold. There's no secondary market in the conventional sense.

The time horizon belongs in the price. An estimate is not cash in the bank. A house may not accept the work at the reserve an owner wants, and a private buyer may need months of conversation before a deal closes. If you know you will need the money on a particular date, that is part of the purchase decision, not an inconvenience to discover later.

Authenticity risk is real but overstated in the wrong direction. Everyone worries about fakes. The more common problem, in my experience, is works with unclear provenance that are genuine but unsaleable because you can't prove it. The documentation culture in the Indian art world is poor.

The return has to survive the full calculation. A work that appears to have doubled between two public sales may have produced a much smaller return for its owner. Buyer's premium, seller's commission, taxes where applicable, insurance, storage, shipping, and the time spent waiting for an exit all belong in the calculation. If those numbers are missing, I would describe the sale as an interesting price observation, not proof of performance.

What I would want on one page

Before calling a work an allocation, I would want its identity and condition documented, a chain of ownership I can follow, and comparable sales that match its medium, size, period, and quality. I would also want a low, middle, and high estimate for an eventual sale, with the costs of each route to market deducted. The lowest case is not there to scare me. It tells me whether the purchase still makes sense without a record-breaking exit.

The comparison set matters most. A small work on paper and a major canvas by the same artist are not interchangeable. Nor are a highly documented work and one whose history is mostly a seller's assurance. A single artist average is a convenient number and a poor way to price a specific object.

That is the kind of information I want Hacoco to make easier to assemble. Data does not remove judgment. It makes the basis for judgment visible enough that another serious buyer can inspect it.

I've spent the last couple of years building infrastructure around this: a data layer that tracks auction results across houses, an AI-driven valuation tool, a platform for secondary transactions in Indian art. Not because I think I can turn art into a stock, but because I think the gap between what this market is and what it could be is genuinely large, and someone has to close it.

The question for the next decade is whether the market can give careful buyers a fuller picture: comparable sales, documentation, realistic selling costs, and a candid view of how long an exit may take. Better information would help collectors who care about both the work and the price.

I think that work is worth doing. I still buy art because I want to live with it. If the numbers also make sense, that is a better purchase. If they only work after I borrow a record price from a different painting and ignore the cost of getting out, I should admit that I am buying for love.

The infrastructure is the bet.