The Turbine Safety Gap the Headline Numbers Hide

Part 91 vs. Part 135: private aviation's safety statistics blend pistons and helicopters into the numbers. Broken out, turbine charter jets fly at near-airline safety levels.

By Noah Siebert · Commercial Pilot, Part 135 · Founder, VeraJets · July 2026

Ask how safe private flying is and you'll get an answer about a fleet you're not flying on.

The number that circulates — general aviation is roughly 25 times riskier than the airlines — is technically true and practically useless. "General aviation" is a statistical bucket dominated by decades-old piston singles flown by weekend pilots. The Part 135 charter average is a different kind of blend: it mixes professionally flown jets with air-tour helicopters, Alaska bush operations, and helicopter air ambulances. Neither average describes the aircraft that actually matter to an owner or charter client: turbine-powered, fixed-wing airplanes.

Strip the blends away and something surprising emerges. The familiar ranking — private flying risky, commercial flying safe — doesn't just shrink. Within turbine fixed-wing aviation, it tells you exactly where the risk lives, and it isn't where the headlines suggest.

No agency publishes accident rates for turbine fixed-wing aircraft split by operating rule. The pieces exist — NTSB accident records, industry fatal-accident tallies, FAA flight-hour surveys — but nobody routinely assembles them. So I assembled them. Here's the data, the math, and what it means.

Why the averages mislead

Two facts break the headline numbers.

First, the general aviation average is a piston statistic. Of the roughly 29 million GA hours flown in 2024, piston airplanes flew about 16 million. The overall GA fatal accident rate was 0.65 per 100,000 flight hours in 2023 — about one fatal accident every 154,000 hours. That figure has almost nothing to do with a professionally maintained Citation or PC-12.

Second, the Part 135 average is substantially a helicopter statistic. In 2019 — a fairly typical year — seven of the fatal Part 135 accidents were helicopters, four were in Alaska, two were in Hawaii, and two were servicing oil platforms. Helicopters run roughly 30% higher accident rates than fixed-wing aircraft overall, and helicopter air ambulance operations average about two fatal inadvertent-IMC accidents per year, every year. Roll all of that into one "charter" number and you've buried the fixed-wing turbine signal completely.

What's actually published

For Part 91 turbine airplanes, published rates exist. AOPA's Air Safety Institute analyzed the 2005–2014 decade of noncommercial Part 91 flying and found turboprops at 1.54 total accidents per 100,000 hours (0.55 fatal) and jets at 0.51 total (0.09 fatal) — roughly a quarter of the overall noncommercial fixed-wing accident rate for turboprops, and far lower still for jets. Peer-reviewed work agrees on the trajectory: Boyd and Stolzer's study in Accident Analysis & Prevention tracked Part 91 turbine airplanes from 1989 to 2013 and found the accident rate fell by more than half over the period.

For Part 135 turbine fixed-wing aircraft, no equivalent published rate exists. The NTSB publishes blended Part 135 rates (helicopters and pistons included); the trade press publishes turbine accident counts by operating rule but not rates. To get a rate you need two things: a clean numerator and a defensible denominator.

The numerator: eleven years of fatal accidents by rule

Aviation International News maintains the most careful public tally of U.S.-registered turbine business aircraft accidents by operating rule. For 2010 through 2020:

Read that table slowly. Over eleven years, U.S. charter jets — every legitimate Part 135 jet operation in the country, combined — had three fatal accidents. Private Part 91 jets had thirty-two. Turboprops show the same shape: fifteen fatal charter accidents against eighty-two under Part 91. And fractional programs operating under Part 91K went the entire decade without a single fatal accident.

Counts alone don't settle it, though. If Part 135 simply flew far fewer hours, low counts would be expected. That's the denominator question — and it's where this analysis does the work the headlines never do.

The denominator: the FAA's own hour matrix

The FAA's General Aviation and Part 135 Activity Survey tracks hours flown by aircraft type and by use. The agency's economic-analysis group compiles it into hours by operating rule. For 2022, the matrix looks like this:

And by aircraft type, the air-taxi (on-demand Part 135) share:

Two details matter. "Air taxi" is not all of Part 135 — the survey separately tracks air tours (0.18M hours) and air medical (0.91M hours) inside the 4.41M Part 135 total. And the FAA runs an independent cross-check against its Operating Specification records: roughly 4,200 Part 135-certificated jets and 2,200 turboprops, with estimated air-taxi hours that bracket the survey figures. The government itself flags "clear uncertainty" in on-demand hour estimates, which is why everything below is presented as a range, not a point.

One finding from this matrix surprised me: jets carry the higher charter share. About 40% of all U.S. business-jet hours fly under Part 135, versus about 22% for turboprops. Turboprops are predominantly an owner-flown, Part 91 fleet.

The result

Divide the eleven-year fatal counts by eleven years of hours, with the Part 135 shares anchored to the FAA matrix and flexed generously to cover uncertainty (jet charter share 30–42%, turboprop 20–30%, agricultural turbine hours removed from the Part 91 pool). These are estimates — clearly labeled as such — with 95% confidence intervals on the accident counts:

Three checks say these numbers can be trusted as estimates. The method reproduces the published figures where they exist: the derived Part 91 turboprop rate of 0.54 lands almost exactly on AOPA's independently published 0.55, and the derived Part 91 jet rate brackets AOPA's 0.09 from the adjacent decade. The statistical confidence intervals for Part 91 and Part 135 don't overlap for either aircraft type — the gap is not an artifact of small numbers. And the main biases run against the finding, not for it: air-medical fixed-wing accidents sit in the Part 135 accident count while most medevac hours sit outside the denominator, which overstates the Part 135 rate. The real gap is likely wider than shown.

In plain terms: a chartered jet under Part 135 suffered a fatal accident roughly once per five million flight hours. That's about three times the airline rate (roughly once per sixteen million hours) — and about thirty times better than the general aviation average everyone quotes. On the safety spectrum, a legitimate charter jet sits next to the airlines, not next to "private planes."

Why the gap exists

Part 135 isn't a paperwork distinction. It's a filter. A charter certificate requires vetted training programs, operational control and dispatch, duty and rest limits, drug and alcohol testing, minimum experience, higher weather minimums, and — increasingly — formal safety management systems. Most passenger charter flying is two-pilot flying.

Part 91 turbine flying contains two very different populations sharing one rulebook: professionally crewed corporate flight departments, and owner-pilots flying their own turbine aircraft single-pilot. The second group is where turbine risk concentrates. Breiling's long-running analyses found turbine aircraft certified for single-pilot operation had 3.4 times the accident rate and roughly 13 times the fatal accident rate of aircraft requiring two crew. AOPA's decade data shows the mechanism: when a single-pilot jet crashed, someone died 31% of the time; with two pilots up front, 12%. Nearly four out of five Part 91 turboprop accidents involved a single pilot.

That also explains why the turboprop gap (2.7x) is smaller than the jet gap (6x): the Part 91 turboprop column is where the owner-flown single-pilot population is largest.

The nuances worth knowing

The data holds a few findings that cut against easy narratives, and they're worth stating plainly.

A privately flown jet is not dangerous. At roughly 0.11 fatal per 100,000 hours, Part 91 jets — most of them professionally crewed corporate aircraft — outperform even chartered turboprops (~0.20). The machine matters and the crew matters more than the certificate alone.

Fractional ownership has the cleanest record in the dataset: zero fatal accidents across the entire 2010–2020 decade under Part 91K, a streak that has continued since.

And the recent years confirm the pattern rather than softening it. In 2023, U.S. Part 135 jet operations recorded zero fatal accidents. Across 2023 and the first nine months of 2024, eleven of twelve fatal U.S. turboprop accidents were Part 91 flights. Business-jet fatalities industry-wide fell to 21 in 2024 from 77 in 2019.

What this doesn't say

Honesty about the limits: the Part 135 rates above are derived estimates, not government-published statistics, because the government doesn't publish this cut. The flight-hour survey is voluntary and sample-based. The accident tallies come from AIN's business-aviation scope, which excludes some turbine operations (cargo feeders, agricultural turbines, government aircraft). Rule classification is fuzzy at the edges — a charter operator's empty repositioning leg flies under Part 91, and its accidents land in the Part 91 column. None of these caveats flips the ranking; every sensitivity run keeps Part 135 turbine fixed-wing several multiples safer than Part 91. But ranges are ranges, and I'd rather you see them than trust a false point estimate.

What to do with this

If you charter: the certificate is the filter that produces these numbers, so verify it. Ask for the operator's Part 135 certificate number and confirm the aircraft is on their operations specifications — not just "operated by" a name on a website. Third-party audits (ARGUS, Wyvern) exist because this verification matters.

If you own a turbine aircraft and fly it yourself: the single-pilot numbers above are the most important table in this article. The gap between owner-flown and professionally operated turbine aircraft is not about the airplane — it's about the operational system around it: two qualified pilots, recurrent training, dispatch, duty limits, and someone empowered to say no to a marginal flight. Professional management puts that system on your aircraft without you giving up the aircraft.

That system is the business we're in. But you don't have to take a broker's word for any of this — the sources are below. Run the numbers yourself.