No country in the examined five-regulator set has implemented a universal child restraint mandate. The FAA’s most recent diversion estimate still leans on assumptions from 1995 and 2011, and the substitution effect has never been validated by an actual policy change.
A parent can legally hold a six-week-old on their lap for a nine-hour transatlantic flight, while the same family is expected to strap that child into a car seat for the drive to the airport. That gap exists because U.S. aviation regulators have spent thirty years worrying about the drive home.
The Federal Aviation Administration has never said lap-held infants are safe. It says a child restraint system is the safest option for a child under two. What stops a mandate is a substitution-effect model: require a seat, raise family fares, and some families decide to drive instead. Since driving is more dangerous than flying per mile, the model projects more road deaths than the mandate would prevent in the air.
But that trade-off has never been tested. No jurisdiction among the United States, Canada, the European Union, Australia, and New Zealand has actually made a separate child restraint seat mandatory for every infant. The central numbers still in use date from 1995 and 2011, long before modern vehicle safety systems, changed airfare structures, and new family travel patterns had fully matured.
Why the FAA child restraint seat mandate rests on an untested road-safety model
The most recent U.S. government cost-benefit analysis, a Department of Transportation update from December 29, 2011, concluded that requiring child restraints in Part 121 commercial operations would produce 72 additional transportation deaths over ten years and 115 over fifteen. The model assumed a 45 percent increase in family air-travel costs, between 6.4 billion and 10.2 billion diverted vehicle miles, and a road fatality rate of 11.3 deaths per billion vehicle miles traveled.
Those are projections, not observations. The DOT paper did not watch families switch modes after a real mandate. It built a counterfactual from fare elasticities and road-safety averages. Later FAA circulars have addressed restraint design and aircraft compatibility, but none has revisited the air-to-road arithmetic in a public rulemaking. The official case against a child restraint seat policy still runs on inputs that are decades old.
The National Transportation Safety Board has never accepted that logic as final. Its Safety Recommendation A-95-51, first issued in 1995, told the FAA to revise 14 CFR Parts 91, 121, and 135 so that every occupant — including infants and small children — would be restrained in a size-appropriate way for taxi, takeoff, landing, and turbulence. A companion rule published by the U.S. Government Publishing Office on June 4, 1996, banned several booster and harness devices during ground movement, takeoff, and landing, but kept the under-two lap-child exemption intact.
Lap child safety rules in five countries still stop short of a full mandate
At the policy level, the pattern is consistent. According to Transport Canada, no country in the examined set has implemented a universal requirement that every infant occupy a dedicated child restraint seat. Child restraint systems have been permitted on Canadian commercial flights since 1990, but they remain optional rather than required. In the European Union, EASA guidance lets operators choose between an approved child seat and a lap-held infant with a supplementary loop belt up to 24 months. Australia’s CASA warns that the adult seatbelt must not be fastened around both adult and infant. New Zealand’s CAA simply notes that airlines accept supplementary loop belts for toddlers up to age two.
The table below compresses the core estimates and the gap between guidance and mandate.
| Jurisdiction or model | Year of key document | Core estimate or policy position | Current status of rule | Was a full CRS mandate implemented? |
|---|---|---|---|---|
| United States — FAA/DOT diversion analysis | 2011 | 72 extra deaths over 10 years; 115 over 15 | Infants under two may be held on an adult’s lap | No — CRS voluntary |
| United States — NTSB Safety Recommendation A-95-51 | 2004 analysis; 1995 recommendation | Restrain all occupants, including infants, during taxi, takeoff, landing, turbulence | FAA did not adopt recommended mandate | No — recommendation not implemented |
| Canada — Transport Canada CRS mandate modeling | 2015–2017 | 45% fare increase; 164M extra vehicle km/year; at least 10 road deaths | Permits but does not require CRS; under review | No — under review |
| European Union — EASA infant guidance | 2008 law; guidance updated 2016–2023 | Infants under two secured in approved CRS or loop belt on lap | Loop belts and child seats remain accepted | No — no separate-seat mandate |
| Australia — CASA lap-infant restraint rules | CAO 20.16.3; AC 91-18 v1.1 (2021) | Lap infant must be restrained; adult seatbelt must not go around both | Allows loop belts and optional child restraints | No |
| New Zealand — CAA child-restraint guidance | 2020s web guidance | Infants up to two may sit on lap using supplementary loop belt | Lap infants with loop belts remain permitted | No |
| NHTSA — passenger car fatality rates | 1995 vs 2022 | 1995: 1.52 deaths per 100M VMT; 2022: 1.33 | Latest comparison points for road-risk models | Not applicable |
| Academic/clinical — Pediatric Emergency Care turbulence study | 2019 | Unrestrained lap children prone to injuries; most did not require diversion | Supports advocacy; no mandate | No |
| Source: U.S. Department of Transportation; National Transportation Safety Board; Transport Canada; European Union Aviation Safety Agency; Civil Aviation Safety Authority (Australia); Civil Aviation Authority of New Zealand; National Highway Traffic Safety Administration (NHTSA); Pediatric Emergency Care | ||||
Flight deals
most people never see
Our AI monitors 150+ airlines for pricing anomalies that traditional search engines miss. Air Traveler Club members save $650 per trip per person on average: see how it works.
Each deal saves 40–80% vs. regular fares:
How the child restraint seat policy’s substitution effect actually works
The swap is simple to describe. Require a seat for every infant, and a family flying with a child under two must buy an extra ticket. The price jump makes some share of families choose highways instead. The model then takes the added vehicle miles and multiplies them by a national road fatality rate.
Because commercial aviation is so safe per passenger mile, even a small diversion can tip the net result. The FAA’s earlier 1995 report to Congress projected five infant crash deaths prevented over a decade and 82 additional road deaths. The later 2011 update moved the totals but not the direction. A post-2015 equivalent does not exist; no later FAA, GAO, or congressional report found in this review resets the numbers.
What has changed is the underlying road, not the model. The arithmetic still applies road-risk assumptions that predate newer NHTSA data. That doesn’t make the projection false; it makes it unrefreshed.
What this means for you
For a traveler making a booking decision today, the rule is unchanged: a child who has not reached their second birthday does not need a separate seat, and the adult may choose to hold the infant. That is the FAA’s current position, not a loophole.
If a mandate ever moved from model to regulation, the immediate effect would be an extra seat charge for every infant. The 45 percent fare rise assumption is not a prediction for each family, but it captures the direction: some households would look at road miles and start comparing risks.
The road-risk half of the FAA’s trade-off has also aged. In 1995, passenger-car occupant deaths ran at 1.52 per 100 million vehicle miles; by 2022 the national rate had fallen to 1.33. Total passenger-vehicle occupant deaths in 2022 were still 25,420, so traffic remains deadly, but the per-mile risk used to justify the old model is not the same risk drivers face today. No updated FAA diversion model accounts for that shift.
Questions? Answers.
What is the FAA-approved child restraint seat for children?
The FAA does not require a separate child restraint system for children under two. Under present regulations, a child who has not reached their second birthday is not required to have a separate seat, and the accompanying adult may choose to hold the infant during flight.
Does my 2 year old need a harness to fly?
The sourced regulations describe an under-two exemption: a child who has not reached their second birthday may be held on an adult’s lap. Once a child reaches age two, that specific lap-child allowance no longer applies, though the material reviewed does not state a separate harness requirement for two-year-olds.
Why hasn’t the FAA made child restraint seats mandatory in U.S. commercial aircraft?
The FAA’s stated reason is diversion risk: requiring extra seats would raise family travel costs and push some travelers onto highways. DOT modeling projects net additional transportation deaths, which the FAA treats as worse than the infant lives a mandate would save. NTSB has recommended restraints, but the FAA has not adopted a mandate.
Do other regulators (Canada, the EU, Australia, New Zealand) also allow lap infants rather than requiring a dedicated child restraint seat?
Yes. Canada permits child restraint systems but does not require them. The EU allows an approved child seat or a lap-held infant with a supplementary loop belt up to 24 months. Australia requires a lap-held infant to be restrained but not around the adult, and New Zealand permits lap infants with a supplementary loop belt.
