The sustainability benefits of hubs
Are hubs good or bad for emissions?
As sustainability has risen in importance for the airline industry, the emissions efficiency of connecting journeys compared to non-stop flights has become a higher profile issue.
Point to point specialists, like the low-cost carriers, expound the benefits of non-stop flights. They highlight that connecting journeys require one or more additional takeoffs - the most fuel intensive part of any flight. You can see that in the chart below, where I’ve plotted the per passenger emissions of an A320 operating on a range of sectors lengths. The data comes from IATA’s carbon calculator service. You can see that the line of best fit doesn’t cross the y-axis at zero, driven by the fuel cost for the takeoff.
The eagle-eyed will also have noted that I didn’t fit a straight line to the data. I actually fitted a quadratic line, as you can see from the equation which is shown on the chart above. The x2 term means that the incremental emissions for each km flown go up as distance increases. That is because of the higher initial fuel weight needed to support a longer flight. Fuel consumption in the early part of flight is higher, due to the need to lift and transport the fuel that will be needed later.
I’ve taken this fitted line and plotted emissions per km. As you would expect, you see a curve with emissions per km initially dropping rapidly as the fixed penalty of the takeoff is spread over more km. But that becomes a smaller and smaller part of the per km emissions and by about 2,000 km the curve flattens out and eventually starts to increase again as the x2 term becomes more and more important.
This non-linear behaviour means that there is a point of minimum emissions per km. In this example, it is at about 3,200 km. I wouldn’t put too much weight on the exact figure, as the IATA emissions data may not be up to the job of establishing the minimum with high precision. It will also vary by aircraft type. Newer, more fuel efficient aircraft have a smaller penalty for extended flight durations. The main point I want to make is that the penalty for splitting a journey into two can be much less than you might think, as long as your connecting point doesn’t involve a significant detour (“circuity”, as network planners would call it).
For the very longest journeys, it can even use less fuel to split your journey. For example, I estimate that Qantas’ record breaking 14,500 km Perth London flight would use 8% less fuel if it made a refuelling stop in Dubai.
But it is true that for most real world journeys, taking two flights will use more fuel than going non-stop, assuming all these flights are taken on the same aircraft type. Let’s use an example of a journey from Edinburgh to Istanbul. Fly direct on an A320 and you’ll have emissions of about 183 kg. Break the journey in Amsterdam, which adds only 20km (0.7%) to your journey, and your emissions would go up by around 14%. Going via LHR is a bit worse at 20%, since the circuity is greater (192 km / 7% of added distance). I think a 15-20% “one stop penalty” is a decent rule of thumb for low circuity connections.
At this point, you might be wondering how on earth I’m going to justify the title of this article. Even if a connecting itinerary is “only” 15-20% higher emissions, non-stop flights are still clearly the best for the environment, right?
The reason this isn’t quite right is that in making the comparisons, we made a big assumption - that the aircraft type would be the same in all cases. That is not going to be true in reality and it can make a big difference as we’ll see.
How emissions vary by aircraft type
Let us start by looking at some data. Again, I've used figures from IATA's CO2 Connect emissions calculator. The chart below shows what it gives for different aircraft types on Dublin to Paris. I've plotted the points against an "all economy" seat count, since there is a significant aircraft size effect on emissions per passenger.
I've fitted two curves to the data. The red line shows the previous generation aircraft types, whereas the blue line shows the new generation equipment. The new generation aircraft are around 20% more fuel efficient than the aircraft that they replace at the same size point.
I didn't include the A330-200 "outlier". Flying wide-body aircraft on short sectors like this is wildly fuel inefficient.
I will return to the question of the difference in fuel efficiency between aircraft generations a little later in this article. But for the moment, I’m going to focus on the aircraft size issue.
As before, I didn’t fit straight lines to the data. I used a “power curve”, where every given percentage increase in the aircraft size gives a fixed percentage reduction in the emissions per passenger. With a power factor of -0.551 for the new generation aircraft, you can see that this fits the data well. It tells us that Increasing the aircraft size by 50% reduces emissions per passenger by 20%.
Why is this relevant to the sustainability benefits of hubs? Concentrating demand is what hubs do. Rather than trying to operate point to point flights connecting every city pair to every other one, each city just needs to be connected to the hub to be connected to the world.
Let us consider the case of an airport located in a big city. To simplify things, we’re going to imagine that even without connecting passengers, there would be sufficient demand to support the operation of flights to many destinations. What is the impact of adding connecting passengers? There will of course be new destinations that become viable, but let’s ignore that for now. For destinations that would be served in either case, the existence of connecting passengers will push up the aircraft size that can be profitably operated. For typical hubs, connecting passengers will represent at least a third of demand. That means that average aircraft size can be pushed up by 50% or more.
We’ve already seen that a 50% increase in aircraft size should reduce emissions per passenger by around 20%. Another way of saying the same thing is that adding extra seats to an existing flight generates emissions at only 40% of the average. That’s more than enough to offset the emissions penalty of taking two flights rather than one.
Network dynamics
Of course, in the real world it is all a lot more complicated than this. Hub operators will not necessarily respond to increases in demand by flying bigger aircraft. Part of the additional demand will be met by adding frequencies. That means that some of the potential “aircraft size” benefits of concentrating traffic at a hub will therefore be realised in higher frequencies. That will benefit both point to point passengers and connecting passengers alike, with a better choice of flight timings.
Connecting passengers also cause new destinations to become viable, enabling passengers that previously had to make a connection to fly direct. The fact that connecting passengers also enable new non-stop routings is something that is often ignored by critics of hubs.
Diminishing returns and outliers
In practice, aircraft economics can be a little more messy than the nice smooth curves I showed earlier. Let’s look at a long-haul example, London to New York. For the new generation aircraft in blue, things are much as before. The scale factor is slightly lower at -0.398, which would give a 15% reduction in emissions per passenger for a 50% increase in aircraft size. Amongst the older generation aircraft, we don’t see any real benefit to size at all. I should say that the 777-300ER figure that comes out of the IATA emissions calculator looks a bit odd to me. Other sources suggest fuel burn per seat should be 6% below the smaller 777-200ER, not 2% above as shown here. But the real thing that jumps out is the green narrow-body outlier. Although IATA shows it as an A321-NEO, I’m assuming it is actually the Jet Blue A321-LR. Despite coming in at less than half the size of the A350-1000, it beats it in terms of emissions per passenger. Partly, this shows how much more economical a narrow-body is than a wide-body. It also shows the fuel efficiency benefit of flying an aircraft at the limits of its range, compared to types capable of much longer ranges, when compared on missions where this is extra capability is just wasted weight. You can see the same effect playing out for the more range-limited 787-10 and A330-300.
I can’t resist showing you another long-haul example, the Heathrow San Francisco route. Here you can see quite how bad the A380 is. Truly the exception that proves the rule that bigger is better when it comes to fuel efficiency. Not only does it have the worst fuel efficiency of any of the aircraft types shown here, but it is even more of an outlier compared to where it should be for its size.
Two ways to fill a big aircraft
An obvious counter-argument to the logic that hubs enable the operation of bigger aircraft is the fact that network airlines don’t actually do so, at least in the short-haul market. The average size of short-haul aircraft operated by the low-cost carriers is if anything higher than the network carriers.
Part of the explanation is that network airlines care more about frequency than the point to point specialists do. Business travellers value frequency whilst leisure travellers are more price focused. Achieving strong two-way connections at hubs requires multiple daily frequencies. These factors cause network planners at each kind of airline to make a different optimisation decision when balancing the revenue advantages of frequency against the cost advantages of larger aircraft.
The other way in which low cost carriers compensate for the lack of connecting traffic is by stimulating volumes with low prices. Both types of airlines end up in much the same place in terms of the aircraft types operated, but where network carriers fill their aircraft by consolidating demand from a broader market, low-cost carriers do it by stimulating demand.
Which is better from a sustainability perspective? Using price to incentivise passengers to take an indirect flight where in some cases they could otherwise fly direct can of course be criticised. As can stimulating people to fly who might otherwise have stayed at home or taken ground transport options. In both cases consumers win through lower prices. In both cases emissions increase.
The impact of innovation
We saw earlier that there has been about a 20% jump in fuel efficiency between generations of aircraft. The industry is counting on further improvements into the future to help deliver on sustainability targets. Where are those innovations likely to come from? And how will they impact on the relative merits of hubs versus point to point flights?
We’ve already seen the potential for the A321-LR and the A321-XLR to disrupt the aircraft size advantages that hubs bring for long-haul flights. I should point out that they will also bring new opportunities for hub operators to broaden their range of destinations and add frequency depth to their schedules, so it is not a one-way street as is often presented.
The other clear aircraft innovation trend I see is driven by the focus on sustainability. Wherever you stand on the optimism-skepticism spectrum regarding electric or hybrid-electric aircraft, it is clear that the first aircraft to emerge from this next wave of investment will come at the shorter range, smaller size end of the market.
New, more sustainable aircraft options at the small end of the market will probably favour point to point offerings, opening up markets that were previously too small to be served direct. But they will also be used at hubs, replacing turboprops and regional jets, helping long-haul operators feed their flights in a more cost efficient and more sustainable way.
The range limitations of electric and hybrid aircraft may mean they benefit hubs at the expense of point-to-point services. Taking two short electric aircraft hops could be much lower emissions than taking one longer conventionally powered flight.
The value of business model diversity
Despite the title of this article, I don’t believe you can say that hubs are either good or bad from a sustainability perspective, relative to point-to-point business models. There are good arguments to be made both ways and where the balance lands depends heavily on the specifics of the market in question and is likely to change over time too. As is often the case in aviation, the devil is in the detail.
What seems clear to me is that neither hubs nor point-to-point based business models are going away. Each brings benefits to consumers and each has a part to play in delivering an efficient air transport system.
That is likely to remain true, whether you judge efficiency by traditional economic indicators or by sustainability metrics.
