The Global Toy Market in 2026: Three Divergences That Matter More Than the Headline Number
The global toy industry is growing faster than it has in years. It is also, simultaneously, becoming a harder business to make money in. Both statements are true, and the gap between them is the most important thing happening in the sector right now.
Circana's 2026 Global Toy Report put the industry at $123 billion in 2025, up 8% — a genuinely strong number for a mature consumer category, and roughly double the growth rate the sector had been conditioned to expect. The US has accelerated further into 2026, with dollar sales up 13% through April. On the surface, this is a boom.
Look underneath and the picture fractures. Growth is concentrated in a narrow band of categories, driven disproportionately by consumers who are not children, monetised increasingly outside the toy aisle, and priced up rather than sold up. Three divergences define the 2026 market, and none of them are visible in the headline figure.
Divergence One: The Growth Is Not Coming From Children
Children under 10 still account for more than 65% of global toy sales. That share is declining. The fastest-growing segment is recipients aged 15 and over, now approaching 20% of global sales, with spending that has more than doubled since 2020.
In the US, the effect is starker still. Circana attributes 35% of the entire industry's 2026 growth to consumers aged 18 and over. Female recipients generated more than half of all growth in the period.
This is not the "kidult" trend as the industry has been describing it for a decade — a pleasant incremental revenue line adjacent to the core business. It is now the growth engine, and it demands a fundamentally different operating model. The core competencies of the traditional toy business are seasonal forecasting, retailer relationship management, parent-facing marketing, and a Q4-weighted P&L. The competencies required to serve a 28-year-old Pokémon collector or a 35-year-old buying a LEGO Botanicals set are closer to those of a hobby publisher: release cadence, scarcity management, secondary-market awareness, community stewardship, and year-round revenue.
Most toy companies are organised for the first model and are trying to bolt the second onto it. The ones that have built the second natively — Pop Mart, Wizards of the Coast, arguably LEGO — are capturing a disproportionate share of the growth. That is not a coincidence, and it is not principally about product. It is about corporate architecture.
The strategic question for any toy business in 2026 is not "should we do more for adults." It is whether the organisation can run two commercially distinct businesses with different rhythms under one roof, or whether it will keep treating the faster-growing one as a line extension.
Divergence Two: The Profit Is Leaving The Toy Aisle
This is the divergence the industry is least comfortable discussing openly, and the 2026 results make it hard to avoid.
Hasbro's second quarter, reported on 21 July, showed total revenue up 16%. Beneath that: Wizards of the Coast and Digital Gaming up 27%, with Magic: The Gathering passing $500 million in quarterly revenue for the first time in its thirty-plus year history, and the Marvel Super Heroes set becoming the fastest in the game's history to reach $300 million. That segment posted a 41% operating margin.
The Consumer Products segment — the part of Hasbro that actually makes toys — grew 5% and posted an operating loss of $15 million.
Set that against LEGO's full-year 2025: revenue up 12% to DKK 83.5 billion, consumer sales up 16%, operating profit up 18% to DKK 22.0 billion, and an operating margin that rose to 26.4% from 25.2%. LEGO is roughly a tenth of the entire global toy market on its own, growing at more than twice the market rate, and expanding margin while doing it.
And against Mattel's Q1 2026: net sales up 4% to $862 million, but gross margin down 450 basis points and an operating loss that widened to $103 million from $53 million. Mattel's Q2 numbers land on 4 August and will be the more meaningful read.
The pattern across all three is consistent. Physical toy manufacturing is a low-margin, high-working-capital, tariff-exposed, retailer-dominated business. Card games, digital extensions, and vertically integrated brand retail are none of those things. The economics have separated to the point where the toy line increasingly functions as the front end of an IP business monetised elsewhere — a customer acquisition and cultural-relevance mechanism rather than the profit centre.
That has real consequences for anyone running a toy-only business. If your competitors can afford to run the toy line at or near breakeven because the money is made in cards, licensing, digital, or theme parks, and you cannot, you are not competing on a level field. The pricing, promotional depth, and shelf investment they can sustain are structurally beyond you. This is the single most under-discussed competitive dynamic in the industry, and it explains a great deal of the pressure independent mid-market suppliers report but struggle to articulate.
Divergence Three: The Tariff Arbitrage Has Closed — After The Capital Was Committed
The most consequential legal event in the industry's recent history was brought by a toy company, and the industry has barely registered what it means.
On 20 February 2026, the US Supreme Court ruled 6-3 in Learning Resources, Inc. v. Trump that the International Emergency Economic Powers Act does not authorise the President to impose tariffs. Learning Resources is a family-owned educational toy business in Vernon Hills, Illinois, employing around 500 people across it and its sister company hand2mind. Its import duties had risen from roughly $2 million to $14 million in a single year. It cancelled a 600,000 square foot expansion and abandoned planned hiring. It sued in April 2025, and it won.
The ruling opened a refund pool estimated at $166–175 billion. It was a landmark for the rule of law and a vindication for a sector that had spent a year being treated as collateral damage.
It also changed the industry's economics far less than most people assume, and in a direction almost nobody has priced in.
Here is where the toy tariff position actually stands as of the end of July 2026. Section 301 List 4A, which carries a 7.5% duty on Chinese-origin toys, was never at issue and remains in force. The administration replaced IEEPA within four days using Section 122 of the Trade Act of 1974 — a 10% global surcharge effective 24 February. The Court of International Trade found that action unlawful on 7 May, but granted relief only to three named plaintiffs.
Section 122 then expired by operation of law on 24 July, exactly 150 days in, as the statute requires. And on 23 July, hours before that sunset, USTR's Section 301 forced-labour duties took effect: 10% or 12.5% across roughly 80 economies, covering approximately 99.4% of US imports. Unlike Section 122, they carry no statutory rate ceiling and no expiry date.
Toys entered the US duty-free before 2025. They now carry, from China, a 7.5% List 4A duty plus the new Section 301 layer. From Vietnam, Indonesia, India or Thailand: the new Section 301 layer alone.
Read that differential again. The gap between sourcing in China and sourcing in Southeast Asia is now roughly seven and a half percentage points.
Through 2025, that gap was enormous and terrifying. Chinese goods faced rates that peaked at 145% — Learning Resources described paying $44 in duty for every dollar it had paid before, and called it an effective embargo. Vietnam faced a threatened 46%. Against numbers like those, relocating production was not a strategic choice; it was survival. Mattel committed to cutting US imports from China below 15% of global production in 2026 and 10% the year after. Hasbro guided to around 30% of toy and game revenue sourced from China by 2026. Across the mid-market, hundreds of smaller suppliers made the same call, qualified new factories, paid for duplicate tooling, and absorbed the learning curve.
Much of that capital was committed against a price signal that no longer exists.
This is not an argument that diversification was wrong. Concentration risk in a single jurisdiction is real, the policy environment remains volatile, and Section 301 and 232 are now the durable authorities precisely because they survived judicial scrutiny. There is a legitimate insurance case for a multi-country footprint, and it has not weakened.
But insurance is not the case most boards were sold. They were sold cost avoidance, on the basis of a tariff differential that has since compressed by an order of magnitude. What those businesses now own is a higher-cost, less mature production base — thinner mould-making ecosystems, longer qualification cycles, less depth in hand-finishing and small-run assortment work, and a supplier network that cannot match Guangdong's density for complex, high-SKU, low-volume lines — without the duty saving that justified it.
The honest strategic framing for 2027 planning is this: treat the multi-country footprint as risk management and price it as an insurance premium, not as a cost programme that will pay for itself. Some of it will. A meaningful portion will not, and pretending otherwise will distort sourcing decisions for another two years. Vietnam and Indonesia remain genuinely competitive for high-volume, low-complexity, stable-SKU production — LEGO's Vietnamese plant is the proof case. They are not, yet, a like-for-like substitute for the Pearl River Delta on a broad assortment.
What Is Actually Selling, And What Isn't
The category narrative most of the industry is still repeating is around two years out of date. Circana's US data for early 2026 tells a different story:
Games and puzzles: +39%, driven overwhelmingly by Pokémon. The Pokémon Company posted approximately $3.3 billion in revenue and $752 million in net profit for the year ended February 2026, its best ever. Target alone guided to over $1 billion in trading card sales. Pokémon TCG Pocket reached 150 million downloads in its first year. This is the growth story of the moment, and it is a hobby-goods story wearing a toy industry label.
Explorative and other toys: +36%, on NeeDoh, squishy formats and MLB collectibles. Squishy toys delivered triple-digit dollar and unit growth.
Building sets: roughly +20%, remaining the most reliable structural category in the industry.
Plush: down slightly. After roughly doubling between 2019 and 2024, plush has stalled. The vinyl-plush collectible wave that drove much of that expansion has cooled from its peak.
Outdoor and sports toys: down around 9%, the largest decline of any supercategory. The post-pandemic active-play thesis has fully unwound. Any strategy deck still citing it as a tailwind needs rewriting.
One offsetting signal worth watching: soccer-related toy sales rose 160% by value from January to April across the twelve markets Circana tracks, on the 2026 FIFA World Cup. That is a genuine, if time-boxed, opportunity.
The Number Everyone Should Be Watching
US dollar sales are up 13%. Units are up 5%. Average selling price is up 7%.
More than half of the reported growth is price, not volume.
Some of that is healthy premiumisation — mid-tier and premium price bands are outperforming, consistent with what is happening across discretionary retail. Some of it is tariff pass-through arriving on shelf. The two are difficult to separate from the outside, and most companies are not eager to separate them internally either.
Either way, the pressure point is at the entry level, where price bands are visibly softening. Historically well over half of US toys sold under $20, and that opening price point is how the category recruits new users — the impulse buy, the pocket-money purchase, the first independent toy decision a child makes. If tariff costs continue to hollow out sub-$20, the industry loses its recruitment mechanism while its revenue line still looks healthy. ASP-led growth flatters the P&L and obscures unit erosion, and unit erosion among children is the leading indicator that matters most.
The Concentration Problem At The Top Of The Growth Curve
Pop Mart is the defining case study in both the opportunity and the risk of the collectibles model.
Its 2025 was extraordinary: revenue of RMB 37.1 billion, around $5.4 billion, up 185%, with net profit up over 300%. Overseas markets contributed more than 40% of revenue. On revenue alone, Pop Mart is now comparable in scale to Mattel — a company that did not meaningfully exist as a global force five years ago.
The market's reaction to those results was to sell the stock 23% in a day. It has traded roughly 40% below its August peak. Q1 2026 revenue growth decelerated to 75–80%, still exceptional but a sharp step down, and Chairman Wang Ning has guided to "no less than 20%" for the full year. The company is building out theme parks, film, fashion collaborations and experiential retail — the Labubu bakery opened at Resorts World Sentosa on 30 July — in an explicit attempt to convert a character into a durable franchise rather than a cycle.
The structural lesson generalises well beyond Pop Mart. Blind-box and collectible economics produce spectacular operating leverage on the way up because scarcity, resale value and social velocity reinforce each other. The same mechanism runs in reverse. The secondary market is the real demand signal, and it turns before retail sell-through does — which means a collectibles business can look healthy on shipments for a full quarter after the underlying demand has broken. Anyone building a collectibles line should be watching resale price indices with the same seriousness they watch POS data.
Where Discovery Actually Happens Now
The retail framing of "shrinking physical shelf, infinite digital shelf" is directionally right and analytically lazy. The more precise change is that discovery has moved entirely upstream of retail.
TikTok Shop has gone from nothing to 1% of total US retail sales and 3% of e-commerce in two years, with toys, hobbies and collectibles its third-largest category. But the transactional share understates the influence: the platform now sets demand for products that are then bought at Walmart, Target and Amazon. NeeDoh is the template — a product that scaled through ASMR and unboxing content, not through a buyer meeting.
The operational implication is severe and mostly unaddressed. Social trend cycles run in weeks. The traditional toy development calendar runs 12 to 18 months from concept to shelf. No amount of marketing sophistication closes that gap. The businesses winning on social-driven demand are the ones that have restructured for it: compressed tooling cycles, domestic or near-shore short-run capability, pre-committed factory capacity held open for fast-follow, and a merchandising relationship that allows in-season replenishment outside the standard planogram reset.
That is a supply chain question and a retailer relationship question dressed up as a marketing question. Most companies are still answering it with more content.
The Strategic Read
Five propositions for the next eighteen months, offered as falsifiable positions rather than platitudes:
Toy-only businesses will continue to lose relative margin. If the toy line is your only monetisation route, you are competing against businesses that can run theirs at breakeven. Either find an adjacent margin pool — cards, digital, experiential, direct-to-consumer — or accept structurally lower returns and optimise for cash rather than growth.
Sourcing diversification should be re-underwritten, not accelerated. The differential that justified the 2025 relocation wave has largely closed. Reprice the footprint as insurance, keep it, but stop funding further moves on a cost case that no longer holds.
Watch units, not dollars. Dollar growth is currently flattered by price. Unit trends among under-10s are the honest health metric for the category's long-term base.
Build for the over-15 consumer structurally, not tactically. That means release cadence, scarcity discipline, community management and year-round revenue — an operating model change, not an adult-collector SKU.
Assume tariff exposure is permanent and re-plumbed, not resolved. IEEPA is dead. Section 301 and 232 are alive, uncapped, and un-sunsetted. Build landed-cost models that assume a durable 10–20% duty layer and price accordingly.
The toy industry remains one of the most resilient consumer sectors in the world. It is also, in 2026, one of the most economically stratified. The growth is real. The question every operator should be asking is whether they are positioned in the part of the market that is capturing it, or the part that is funding it.
