Why this now: Retail IPO “access” keeps growing, but fills still feel tiny. A recent academic study puts real numbers around it so you can set expectations before you tap “request.”
Key takeaways
- Across U.S. IPOs with retail access, ~0.3%–8% of shares typically go to retail; median ≈2% (May 2025 study).
- The headline exception was Robinhood’s own 2021 IPO, which reserved ~20%–35% for its users—far above normal.
- Small retail slices + oversubscription = many zero or tiny fills, even when you do everything right.
- Returns for “retail-access IPOs” lagged peers on average in year one—access ≠ edge.
- Practical takeaway: treat access as lottery tickets, line up multiple channels (DSP, different brokers), and plan for open-market buys.
First, the basics: how retail gets in
Retail allocations happen mainly via two pathways:
- Broker-led “IPO Access.” Apps or brokers collect Conditional Offers to Buy (COBs) from eligible customers and pass them to the underwriters. If the broker receives shares, it allocates them—often randomly—among its requesters.
- Directed Share Programs (DSPs). The issuer sets aside shares for a defined group (employees, customers, partners). If invited, you follow a parallel flow that’s separate from the broker’s pool.
Think of these as two doors into the same offering. Door A (broker access) depends on how many shares your broker receives. Door B (DSP) depends on whether you’re in the issuer’s invited group.
Jargon buster
COB (Conditional Offer to Buy): A request to buy at the IPO price—not a guarantee.
DSP (Directed Share Program): Issuer-run, invitation-only set-aside.
What the data actually shows (ranges, medians, and the outlier)
A peer-reviewed working paper (revised May 21, 2025) looked across recent U.S. IPOs that offered retail access at the offer price. Excluding Robinhood’s own IPO, retail received roughly 0.3% to 8% of shares, with a median near 2%. That’s your base case for “normal” allocations.
Why the big gap? Because the broker’s slice of the syndicate is usually small, and the retail portion is smaller still after institutions, anchors, and other reserved groups get their allocations.
The one giant exception you’ve heard about: Robinhood (2021). Its S-1 and news coverage confirmed a plan to reserve ~20%–35% of its IPO for users. That was an issuer choice designed to showcase retail inclusion, not a new norm.
So why do fills feel so small? The mechanics behind “0 shares”
Even with a valid request, zero or tiny fills are common:
- Oversubscription math: If an IPO is 20×+ subscribed and retail only gets ~2% of shares, most requests won’t clear.
- Broker’s upstream allocation: Your broker can only allocate what it receives. A small upstream slice means small downstream fills.
- Eligibility and timing: Missed confirmations, insufficient buying power at allocation, or issuer-specific eligibility rules can move you out of the pool.
- DSP vs broker pool: If an issuer runs a DSP, those shares are separate; they don’t increase what’s in the broker’s lottery.
Set your mental model to “lottery ticket” for hot deals. You’re optimizing odds at the margin, not controlling the outcome.
Performance: access ≠ edge
That same 2025 study also found that retail-access IPOs underperformed other IPOs by about 20 percentage points in year one, on average. Two drivers stood out:
- Adverse selection: Issuers steering shares to retail often priced more aggressively relative to fundamentals.
- Attention dynamics: Higher pre-IPO search interest and heavier post-IPO fractional trading suggested attention-driven buying that later faded.
For investors, that means discipline matters. If you get filled, have a plan: thesis, time horizon, and risk limits. Don’t let the “win” of an allocation replace analysis.
Mini caselets: setting expectations before you click “request”
- Hot tech IPO, $1B float: Broker gets 2% of shares (hypothetical), then puts ~2% of float to retail access. If the deal is 30× subscribed, a large majority of requesters get 0 or single-digit shares.
- Issuer-run DSP: You’re invited as a customer or employee. Fill odds are much higher than broker pools, but still constrained by the DSP’s cap. Always check emails and portals early; these windows can be short.
Practical playbook:
- Use multiple doors: Broker access + watch for DSP invites.
- Fund accounts early: Ensure cash is settled before pricing.
- Pre-decide your exit: If you plan to flip, know your broker’s access penalties; if you plan to hold, know your valuation guardrails.
Bottom line
Across deals, retail typically gets low-single-digit percent of the float; ~2% median is the right baseline. Expect many zero or tiny fills in hot offerings—and remember that access alone hasn’t meant better performance on average.
One practical next step: For any must-own IPO, draft an open-market buy plan (price limits, size, time window) in case your allocation is small or zero.
Sources
- Gempesaw, Henry, Pisciotta, Xiao (SSRN, revised May 21, 2025) — “Retail IPO Access: High Hopes, Low Returns.” (allocation range ~0.3%–8%, median ≈2%; relative performance). SSRN
- Robinhood S-1 & contemporary coverage (July 2021) — Reserved ~20%–35% of its own IPO for users (exceptional, not a new norm). SECReuters+1
- Ritter, “IPO Statistics” (2025 update) — Market-wide IPO context and benchmarks (not retail-slice specific, but helpful background). Websites
Leave a Reply