Text size
Drive
6 min readText size
by Toby Hagon
Published
Text size
Never before has fuel use been so front and centre for motorists.
Record high prices are incentive enough to make drivers think harder about how much of the liquid fuel their car is consuming.
But working out exactly how much fuel your car will consume isn’t always as simple as looking at the sticker on the windscreen.
The best way to find that out is to test drive the car exactly as you’d normally use it, but that’s not always possible. The next best option is the Australian Automobile Association’s Real World Testing program, which so far has tested 154 cars on Australian roads. As expected, there were some differences from the mandated lab tests.
First, some background.
Like washing machines and fridges, the ‘energy use’ of a new car must be displayed on a government-supplied label stuck to the windscreen.
The figures are also published on the government’s Green Vehicle Guide website.
Those official fuel consumption figures are arrived at following a laboratory test carried out under Australian Design Rule (ADR) 81/02 which is based on a test known as the New European Drive Cycle (NEDC). It applies to most light vehicles but excludes trucks, buses and larger utes, such as those sourced from the US.
The test involves placing vehicles on a rolling-road dynamometer and driving through a standardised 11-kilometre cycle lasting almost 20 minutes, covering both stop-start urban driving and higher-speed conditions, referred to as extra-urban.
The test is conducted in controlled conditions, with no traffic, weather, or accessory loads (the air conditioning is not on, for example), and uses a standardised reference fuel.
The idea is not necessarily to perfectly replicate real-world driving, but to create a consistent benchmark that allows consumers to compare vehicles under identical conditions.
“From vehicle A to vehicle B, obviously you need a standard to test to,” says Michael Barber, an engineer with Multimatic who previously worked at Holden.
“You need a yardstick.”
That discrepancy between the official figure and real-world fuel use does not mean the laboratory test is flawed.
Rather, it highlights that the priority of the lab test is experimental control and repeatability, wherever in the world the test is done. Real world testing may generate results that are closer to what drivers may experience in the real world, but they do not have laboratory levels of repeatability.
And both tests highlight the challenge of reducing the enormous variability of real-world driving into a single number.
While the lab test is highly repeatable, lab conditions are far removed from the way most Australians actually use their cars.
As mentioned above, the ADR 81/02 test is conducted indoors, on a rolling-road dynamometer, with no traffic, weather, hills or accessory loads. The air conditioning is switched off and the steering system not subjected to the resistance encountered when physically turning tyres on the road surface. Plus, the car follows a tightly prescribed driving cycle that often uses very gentle throttle application.
How a vehicle is driven during that cycle can have a significant impact on the result.
“Gentle acceleration is a huge enabler to getting a good economy number,” Barber says. “In the urban cycle, it’s all about acceleration. On the open road, it’s all about speed due to the wind resistance.”
Those lighter-load laboratory conditions can also mask some of the compromises manufacturers make when engineering modern turbocharged engines.
Barber says some turbocharger materials used to reduce manufacturing costs cannot tolerate the same temperatures as more expensive alternatives, requiring additional fuel under heavy load to help cool components and protect durability.
“Turbo turbine material has temperature limits,” he says. “A cheaper material has a lower temp limit, so you need to put more fuel in to reduce exhaust gas temperatures.
“Terrible for economy under peak load but you can’t have turbine blades disintegrating, so you put more fuel in.”
Because official fuel economy tests rarely subject engines to sustained heavy loads or aggressive acceleration, that additional fuel use may not appear prominently in laboratory figures, even if owners experience it in everyday driving.
That can favour vehicles calibrated to perform efficiently under those specific conditions, even if owners drive them very differently in everyday use.
“For the real world, driving style has so much to do with it,” says Barber. “It’s a massive influence - how much people step into the accelerator as they drive away from the lights.”
Modern vehicles are heavily optimised to perform as efficiently as possible under official testing procedures.
“A lot of this is just good, honest engineering, trying to do the best you can,” he says. “And there are a few little sneaky things that get done as well.”
Some are relatively minor. During laboratory testing, for example, the 12-volt battery can be fully charged beforehand, reducing the amount of work the alternator needs to do – in turn drawing minute amounts of energy from the engine during the test cycle.
“The fact that the 12-volt battery was float-charged and always full of juice means that it would be doing a bit less work when you’re running your test,” Barber says.
Even small changes can influence the final result, particularly when manufacturers are chasing improvements measured in tenths of a litre per 100 kilometres.
Yet despite the gap between official and real-world results, the laboratory test still serves an important purpose: giving consumers a consistent baseline for comparing one vehicle against another.
It was that gap between laboratory testing and everyday driving – as well as Volkswagen’s highly-publicised emissions cheating scandal - that prompted the Australian Automobile Association (AAA) to launch its Real-World Testing Program in late 2023.
Backed by federal government funding, the program independently tests vehicles on public roads rather than in a laboratory, measuring fuel use and emissions in conditions designed to better reflect how Australians drive.
The tests are conducted on a fixed 93km route around Geelong in Victoria, incorporating urban traffic, rural roads and freeway driving. Vehicles are tested under tightly controlled protocols, with the AAA aiming to strike a balance between real-world conditions and repeatability.
Since the program began, the AAA has tested 154 petrol, diesel and hybrid vehicles, finding that 77 per cent used more fuel than indicated on their mandatory windscreen labels. Some vehicles exceeded their official claim by more than 30 per cent.
AAA managing director Michael Bradley says the program is not designed to replace laboratory testing, but to provide motorists with an additional layer of information.
“Now more than ever, it’s critical that Australian consumers know exactly how much fuel a car will actually use, how much they will cost to run, or how far they will go on a single charge,” Bradley said.
The organisation says the testing also reflects growing scrutiny of laboratory-based fuel and emissions claims following the Volkswagen “dieselgate” scandal, which exposed how some manufacturers engineered vehicles specifically to perform well under official test conditions.
“The Real-World Testing Program was created in response to the Volkswagen scandal,” Bradley said recently, adding that the AAA wanted to ensure emissions reductions were occurring “in the real-world, not just in the lab”.
Importantly, the AAA’s testing has not found that every vehicle performs worse than its official figure.
Some vehicles have consumed less fuel on the road than in laboratory testing, reinforcing the idea that there is no simple “fudge factor” consumers can apply across the board.
The growing discrepancy between laboratory and on-road testing is one reason many overseas markets - particularly Europe - have shifted away from the older NEDC regime that has lingered in Australia.
There’s one type of car the AAA hasn’t tested yet: plug-in hybrid electric vehicles, or PHEVs.
Like a hybrid, they can use a petrol engine and electric motor/s at the same time, but they also have larger battery packs that allow electric-only driving, sometimes for more than 100km.
The real-world fuel consumption of PHEVs can differ dramatically from their laboratory fuel consumption figure because their EV component generally performs extremely well in the short laboratory test, allowing the car to rely heavily on electric power.
That doesn’t necessarily make the official figure misleading — it simply reflects how dramatically fuel use can vary depending on how far the vehicle is driven and how often it is charged.
Often, the lab test result for a PHEV is less than 2.0 litres per 100km due to its use of predominantly electric power throughout the lab test cycle.
But on longer trips - when the car is running mainly as a hybrid - it will use many multiples of its official fuel figure.
The discrepancy has prompted some car makers to quote two figures in their marketing material: the official fuel figure and a more realistic figure that indicates how much fuel you can expect to use once the car is running in hybrid mode.
Australia’s fuel consumption labels are finally set for a major overhaul.
Late in 2025, the federal government confirmed it would transition from the older NEDC-based testing regime to the newer Worldwide Harmonised Light Vehicle Test Procedure, or WLTP.
The updated labelling rules will apply to newly released vehicles from July 1, 2026, before being phased in across the broader market over the following two years.
Already used across Europe and many other global markets, WLTP is widely regarded as being more representative of real-world driving than the older NEDC standard still used under Australia’s ADR 81/02 regulations.
While still conducted in a laboratory on a rolling-road dynamometer, the WLTP test is longer and more demanding, incorporating a wider range of speeds, heavier acceleration and more realistic driving conditions.
In most cases, that results in higher - and generally more realistic - fuel consumption figures.
However, you don’t have to look far to find European drivers suggesting even the WLTP figures aren’t representative of their driving.
Also, the transition period is likely to create fresh confusion for consumers.
Until 2028, some vehicles sold in Australia will continue using the older NEDC-based figures, while others will display WLTP results.
That means two similar vehicles could appear to have very different fuel consumption figures simply because they were tested against different drive cycles.
Rather than the apples-versus-apples comparison consumers have now - however imperfect some regard it - the market is temporarily heading toward an apples-versus-oranges situation.
The government acknowledged the issue in its consultation papers, noting: “This will mean some new vehicles sold from 2025 to 2028 will have NEDC figures, and some will have WLTP figures.”
The discrepancy is significant enough that the government has introduced conversion factors under the New Vehicle Efficiency Standard to ensure manufacturers are treated fairly while the two systems operate side-by-side.
So what should buyers make of the growing gap between official fuel figures and the AAA’s real-world testing?
The key is understanding that the two tests are measuring different things.
The regulation ADR 81 fuel figure is best viewed as a controlled benchmark.
Because every manufacturer must follow the same laboratory procedure, it allows consumers to directly compare one vehicle against another under identical conditions.
That consistency is valuable, particularly when buyers are cross-shopping competitor vehicles.
A car claiming, say, 6.0L/100km on the official test will generally use less fuel than one claiming 9.0L/100km, even if most owners never exactly match either figure in everyday driving.
The AAA’s testing, meanwhile, is designed to provide additional context around what owners are more likely to experience on Australian roads.
Neither figure should be viewed as a guarantee.
Real-world fuel use can vary dramatically depending on traffic, temperature, terrain, driving style, load and even how frequently a car is used for short trips.
And ultimately the person behind the wheel is likely to have the biggest impact.
That variability is particularly important when comparing vehicles with very different powertrains.
Hybrid vehicles, for example, often perform exceptionally well in stop-start urban driving, while some turbocharged petrol engines can consume significantly more fuel when driven hard or carrying heavy loads.
Plug-in hybrids can vary even more dramatically depending on whether owners regularly recharge them or rely mainly on the petrol engine.
The best approach for consumers is to use the official fuel label as a comparison tool, then supplement it - where possible - with real-world data such as the AAA program, owner forums and independent road tests.
Ultimately, no single test can perfectly predict what every Australian motorist will experience behind the wheel.
