Winner’s Curse in IPOs: Why You Get Full Allotment in the Issues You’d Rather Miss

Here’s a pattern almost every IPO applicant in India has noticed. The issue everyone is talking about, the one that’s subscribed a hundred times over, never seems to land in your demat account. The quiet one nobody wanted? You get it in full. That uncomfortable feeling has a name in finance: the winner’s curse in IPO allotments. It isn’t bad luck. It is built into how shares get rationed when some investors know more than others.

In this piece we’ll walk through where the idea came from, how the academic model applies it to new issues, a worked example with simple made-up numbers, and what it means for the way you apply in Indian IPOs. We’ll also look at the strongest criticism of the theory, because it’s not the whole story.

Where the winner’s curse comes from

The phrase started in oil, not stocks. According to Investopedia’s explainer, it was coined by three Atlantic Richfield engineers who saw that companies bidding for offshore oil drilling rights in the Gulf of Mexico were earning poor returns on what they won. Wikipedia’s winner’s curse article dates the first paper to 1971, by Capen, Clapp and Campbell in the Journal of Petroleum Technology.

The logic is neat. In a “common value” auction the thing being sold is worth roughly the same to everyone, but nobody knows that value for sure. Each bidder makes an estimate. Some guess high, some guess low. The person who wins is the one with the most optimistic guess. If the estimates are right on average, the winner has very likely overpaid.

Two details from that Wikipedia article matter for us. First, the curse gets worse as the number of bidders grows, because with more bidders it’s more likely that someone has badly overestimated the value. Second, careful bidders protect themselves by “shading” their bids, bidding below their raw estimate because they know that winning is itself bad news about the item’s value.

Winning is information. If you got the whole thing, ask yourself why nobody else wanted it.

How the winner’s curse in IPO allotment works

An IPO isn’t a classic auction. The price is fixed or set within a band, and when demand runs ahead of supply, shares are rationed. The twist is that rationing creates its own version of the curse.

The standard academic version comes from Kevin Rock’s 1986 paper. Jay Ritter of the University of Florida summarises it in his review Equilibrium in the Initial Public Offering Market (Annual Review of Financial Economics, Vol. 3, 2011). The setup goes like this:

• The company and its bankers don’t know the true value of the shares with certainty, but some investors do. Call them the informed investors.
• Everyone else is uninformed. They can’t tell a good issue from a bad one before applying.
• If demand is more than the shares on offer, shares are split pro rata. In Ritter’s example, if the issue is subscribed three times, every applicant gets one-third of what they asked for.
• The informed investors only put in orders when the offer price is at or below what they know the shares are worth.

Now follow it through. When an issue is cheap, the informed crowd piles in, the book is heavily oversubscribed, and the uninformed applicant gets only a slice. When an issue is expensive, the informed stay away, demand is thin, and the uninformed applicant gets everything they asked for. Ritter puts it plainly: uninformed investors “will receive all of the shares being sold when the offer is overpriced, but only some of the shares being sold when the offer is underpriced.”

That’s the curse. Your allotment is biggest exactly when the deal is worst. So, in Rock’s model, IPOs have to be underpriced on average to compensate the uninformed for this adverse selection. Without that discount they would stop applying, and issuers need their money.

A worked example with simple numbers

Let’s make this concrete. The numbers below are made up for illustration; they are not data from any real issue.

Say a retail investor applies for ₹1,00,000 worth of shares in each of two IPOs, both priced at ₹100.

• Issue A (the good one): informed investors spot that it’s worth more than the price and apply heavily. The issue is subscribed 10 times. Under pro rata rationing, our investor gets one-tenth of the application, ₹10,000 worth. The stock lists and settles at ₹120, a 20% gain on the issue price. Profit: ₹2,000.
• Issue B (the weak one): informed investors skip it. Demand is barely enough, so our investor gets the full ₹1,00,000 worth. The stock slides to ₹90, a 10% loss on the issue price. Loss: ₹10,000.

Look at what just happened. Each issue on its own had a fair shot at going either way, and averaged across the two, the issue prices moved +20% and -10%, which looks like a decent +5% average “IPO return”. Yet our investor is down ₹8,000 overall, because the losing issue came in at ten times the size of the winning one.

That gap between the average IPO return you read about and the return an ordinary applicant actually earns is the whole point. Headline studies measure returns per issue. Your account measures returns weighted by how much you were allotted. If allotment is biggest in the bad deals, those two numbers can point in opposite directions.

Now push it further. To break even in this example, Issue A would have to rise enough on ₹10,000 to cover a ₹10,000 loss on Issue B. That means a 100% listing gain on the good issue. Shrink the loss on the bad one, or increase the allotment on the good one, and the maths gets kinder. This is why, in Rock’s world, issuers have to leave a decent discount on the table across the board.

The average IPO can make money while the average small applicant loses it. The difference lies entirely in who gets how many shares.

How Indian allotment rules change the picture

Indian retail allotment doesn’t work exactly like the pro rata model in the paper. As Chittorgarh’s IPO subscription guide explains, allotment in an oversubscribed issue happens either on a proportionate basis or by lottery, depending on how heavily it is oversubscribed and on the investor category. If the number of applicants is lower than the maximum number of allottees, each applicant is assured one lot and the rest is shared out proportionately. If the number of applicants is higher, even one lot isn’t guaranteed and allotment goes by lottery.

Does a lottery kill the curse? Not really. It changes its shape. In a hot issue, your chance of getting even one lot falls as the applicant count climbs. In a cold issue where applicants are fewer than the maximum number of allottees, you’re assured a lot and you may get more on a proportionate basis. So the same asymmetry survives: the deals you’re most likely to get, and get more of, are the ones the market wanted least. Our explainers on how the basis of allotment works and what oversubscription actually means go into those rules in more detail.

There’s also a practical point. When retail applicants simply pick one lot in every issue, the lottery spreads hot-issue allotments thinly across a very large pool of applicants. The person who applies to everything ends up with a portfolio tilted towards whatever didn’t fill up. That’s the curse in its everyday Indian form.

The strongest criticism of the theory

We shouldn’t treat Rock’s model as settled fact. Ritter himself, in the same review, raises two serious problems with it.

The first is about cause and effect. He notes that many studies find evidence that fits the adverse selection story, but almost none ask which way the causality runs. With free entry, investors keep piling into IPOs as long as there’s money to be made, so the bigger the expected underpricing, the bigger the oversubscription. To an outside observer that looks like adverse selection driving underpricing, when it could just as easily be underpricing pulling in the crowd.

The second is about how deals are sold today. Ritter points out that Rock’s model assumes a fixed-price offer, where the price is set before demand is known and underwriters have no say in allocation. Many IPOs over the last two decades have used book building instead, where the price is set after institutional demand is gauged and underwriters have discretion over allocation. Given that, he argues, the levels of underpricing we actually see “seem to be far in excess of what could be explained by the winner’s curse problem.”

So the honest takeaway is this. The winner’s curse is a real mechanism, and you can feel it in your own allotment history. It just isn’t the only reason IPOs tend to list above their issue price. Our longer piece on why IPOs get underpriced covers the other explanations.

What the curse means when you apply

You can’t become an informed institutional investor overnight. But you can stop behaving like the purely uninformed applicant in Rock’s model. A few habits help.

Treat a full allotment as a signal, not a prize

If you got everything you applied for, the first question isn’t “how much will it list at?” It’s “why was demand so thin?” Go back and check how each investor category subscribed. Weak demand from qualified institutional buyers in particular is the closest thing a retail applicant has to seeing what the informed money thought.

Don’t apply blindly to every issue

The investor in our example lost money not because they were unlucky but because they applied the same amount everywhere. Applying mechanically to every IPO is exactly the behaviour the curse punishes. Read the offer document, look at valuation against listed peers, and skip issues you wouldn’t buy on the open market at that price.

Size your application to what you’d hold

Remember the asymmetry: bad issues fill your application in full. So never apply for more than you’d be comfortable owning outright at the issue price. If the answer for a particular company is “one lot at most”, apply for one lot.

Shade your expectations

The oil companies learned to bid below their raw estimate of value. The IPO equivalent is to discount the listing gains you hear about. Headline averages count each issue once, regardless of how much anyone got. Your own returns will be weighted by allotment, and in the issues you get plenty of, that weighting works against you.

Track your own allotment-weighted returns

Keep a simple sheet: issue, amount applied, amount allotted, listing-day value, value after a few months. Within a year or two you’ll see whether your IPO strategy is actually making money, or whether a few small wins are hiding bigger losses on full allotments. It’s the single most useful bit of record-keeping an IPO applicant can do. For the wider process, start with our guide to IPO investing in India.

A quick recap of the key ideas

• The winner’s curse began in oil-lease auctions: whoever wins with the most optimistic estimate tends to have overpaid.
• In IPOs it shows up through rationing. Informed investors crowd into good issues and avoid bad ones, so uninformed applicants get small allotments in good deals and full allotments in bad ones.
• Rock’s model says IPOs must be underpriced on average to keep uninformed investors coming back.
• Indian lottery and proportionate allotment rules change the mechanics but not the basic asymmetry.
• Ritter’s critique is a fair one: the curse probably explains some underpricing, not all of it.

The next time an issue fills your application in full, pause before celebrating. Sometimes it’s genuinely a good company the market overlooked. More often, it’s the market quietly telling you something.

Disclaimer: This article is for educational purposes only and is not investment advice. The worked example uses hypothetical numbers for illustration. IPO investments carry risk, including loss of capital. Please read the offer document and consult a SEBI-registered investment adviser before making investment decisions.