Bookbuilding: How Underwriters Price and Allocate IPO Shares Using Institutional Bids

The final price of an Initial Public Offering (IPO) is not dictated by a mathematical formula, but by a structured auction called bookbuilding, where institutional investors submit confidential bids across a predetermined price band. By aggregating these bids, underwriters determine the highest price at which all available shares can be sold, establishing the clearing price before retail trading begins.[2][3]

In the modern financial system, the vast majority of public offerings rely on the bookbuilding mechanism rather than fixed-price issues. According to PwC's analysis of structural IPO costs, underwriting fees typically consume 4% to 7% of gross proceeds. This fee compensates investment banks for managing the complex logistics of this exact price discovery process, balancing the issuer's need for capital against the market's willingness to pay.[6]

The process begins when a company files its preliminary prospectus, known in the US as an S-1 or a "red herring." As seen in standard SEC filings, such as a 2026 S-1/A amendment, this document contains comprehensive financial data but deliberately omits the final share price and exact number of shares offered. Instead, it provides a price range, such as $28 to $32 per share, setting the boundaries for the upcoming auction.[7]

Armed with the red herring, the issuing company's management and lead underwriters embark on a roadshow. They pitch the business to institutional investors—pension funds, mutual funds, and hedge funds. Fidelity Investments notes that retail investors are largely excluded from this phase, which is designed to gauge the appetite of buyers capable of absorbing millions of shares at once.[8]

During the roadshow, the "book" opens. Institutional investors submit non-binding bids specifying the number of shares they want and the price they are willing to pay within the established band. A fund might bid for 500,000 shares at $30, or submit a "strike bid," agreeing to purchase a set number of shares at whatever final price the underwriter determines.[2][3]

Oxford Academic research on IPO motives and mechanisms highlights that this bidding phase is an exercise in information extraction. The underwriter needs institutional investors to reveal their true valuation of the company. To incentivize honest bidding, underwriters implicitly promise to underprice the IPO slightly, leaving "money on the table" for these early backers to capture a first-day trading profit.[4]

As bids accumulate, the lead underwriter maintains a confidential ledger—the literal "book" in bookbuilding. If the book is "oversubscribed," meaning investors have requested more shares than are available, the underwriter has the leverage to price the offering at the top of the band or even raise the band entirely. Conversely, a sluggish book might force the price to the bottom of the range.[2]

Once the roadshow concludes, the book closes. The underwriter and the company's board analyze the demand curve to select the final issue price. This clearing price is typically set just below the maximum price the market can bear, ensuring all shares are sold while generating a first-day "pop" in the secondary market.[3][4]

Price is only half the equation; allocation is the other. Unlike a standard exchange where orders are filled strictly by price and time priority, IPO allocation is discretionary. The underwriter decides who gets shares and how many. Groww's 2024 analysis of the process notes that institutional buyers often receive the lion's share of the allocation, sometimes up to 50% or more of the total offering.[2]

Retail investors typically access IPOs through their brokerages, but they are at the mercy of the allocation cascade. The Securities and Exchange Board of India (SEBI), for example, mandates specific allocation buckets, requiring at least 35% of a book-built issue to be reserved for retail investors. However, in the US, no such strict retail quota exists by law.[1]

When a retail bucket is oversubscribed, shares are usually rationed on a pro-rata basis or through a lottery system. If a retail tranche is oversubscribed by 10 times, an investor who bid for 100 shares might only receive 10. SoFi's 2025 breakdown of the process emphasizes that retail investors must fund their accounts for the full bid amount, even though they are highly unlikely to receive their full requested allocation in a hot IPO.[3]

While bookbuilding dominates, it is not the only method. Emerald Publishing's comparison of IPO mechanisms contrasts bookbuilding with discriminatory price auctions and uniform price auctions. In a uniform price auction, all winning bidders pay the same clearing price, stripping the underwriter of allocation discretion. However, issuers overwhelmingly prefer bookbuilding because it allows them to curate a stable, long-term shareholder base rather than selling to short-term flippers.[5]

The structural cost of this price discovery is significant. Beyond the 4% to 7% underwriting spread, the deliberate underpricing of the shares represents a massive indirect cost to the issuing company. If a company prices 10 million shares at $30, and the stock opens on the exchange at $40, the company has effectively transferred $100 million in value to the institutional investors who participated in the bookbuilding phase.[4][6]

The bookbuilding process remains the central nervous system of public market debuts. It balances the issuer's need for capital certainty against the institutional investor's demand for a discount. Until alternative models like direct listings or Dutch auctions gain wider traction, the confidential ledger of the underwriter will continue to dictate the terms on which private companies enter the public domain.[9]