Airline Industry Default History Through 2024

Before putting 2024’s default experience in its historical context the table below shows the final list of airline failures we have been able to identify for the year. We have excluded CSA, OTT Airlines and Vistara as these airlines ceased operations due to merger and Air Malta – the latter ceased operations but was immediately succeeded by KM Malta Airlines, also owned by the government of Malta and operating at least some of the aircraft previously operated by Air Malta. We classify this as an orderly wind-down rather than a default.

 

 

The map below shows these defaults were dispersed geographically but defaults by numbers of affected aircraft were highly concentrated in the Americas, accounting for 379 out of 461aircraft (82%). The big driver of defaults in North America was the difficult business environment for low-cost airlines and in Latin America it was the hangover from the airline defaults that occurred in the pandemic. The airlines that restructured through bankruptcy at this time obtained a sufficient competitive advantage that those which did not ultimately have had to follow.

 

 

Sirius produces proprietary estimates of global airline defaults because we think there are important issues around credit risk that arise at the industry level as well as at the level of the individual airline, and we cannot find any alternative source with the coverage we would like. The appendix to this report gives a detailed description of our methodology – probably the most important point to bear in mind is that we exclude small airlines for the sake of practicality so that only 8 of the 19 airlines listed in the table above are included.
The rating agencies produce historical statistics by industry, but their universe of rated airlines is quite small (typically 25-30 airlines at any given time) as they only cover airlines that make use of debt capital markets funding.

 

 

 

Like 2023, 2024 saw a more normal level of defaults following the shocks of the pandemic in 2020[1] and Russia’s expropriation of lessor-owned aircraft in 2022. 2024’s default rate by airline was 3.7% compared to an average of 3.4% p.a. from 1990 and the default rate by aircraft was 2.0% compared to an average 2.9%. This means that the average defaulter had a smaller than normal fleet which is good news for financiers who had to remarket aircraft. The historic correlation between these two default rates is quite low at 0.36, reflecting the divergence caused spikes in defaults by aircraft when very large airlines such as the US majors enter bankruptcy. The analysis presented below focuses on defaults by airline to make for a more digestible read and the pattern of outcomes is broadly like that for defaults by aircraft.
One notable feature of our estimated default rate is that it is not highly correlated with traffic growth. The chart below compares the default rate by airlines with traffic growth for every year where defaults were above average with the highest default rates on the left. Traffic declined in only 3 of the twelve years concerned.

 

Another significant and related result is that default rates by region are not strongly correlated as can be seen in the table below. This should not really be surprising as analysis of traffic growth and airline profits by region shows a similar pattern.

 


 

We believe this lack of correlation is a key explanation of how most aircraft lessors are financially robust across the cycle despite serving a single relatively risky industry. 2024’s experience in terms of defaults by aircraft is an example how defaults are often skewed by region.
We also try to see what impact airline size has had on default rates. To do this we divide airlines into quintiles based on size as measured by number of seats, with each quintile representing roughly the same number of airlines. The chart below shows the population of airlines by quintile which has been quite stable since the late 2000s. During this period, bigger airlines have grown faster than smaller airlines resulting in a more concentrated industry.

 

 

The default rates by quintile as presented in the chart below are visually confusing but the numbers in the following table help to put some structure around the results.

 


The table shows that the biggest airlines have the lowest and least volatile default level – volatility matters because it is harder to manage a default rate that is “spiky”. Once we go to the other quintiles there is no observable relationship between size and default rate, but there is a definite upward trend in volatility as airlines become smaller. Overall, our analysis suggest size has a limited influence on creditworthiness, and it would be wrong to assume that the lower default rate for larger airlines makes them “investment-grade” by a flawed analogy with credit ratings. The airlines in Quintile 1 have a default rate of 1.7% vs 3.4% for all airlines – for purposes of comparison the average default rate for S&P investment-grade bond issuers was 0.1% p.a. vs 3.9% p.a. for speculative-grade for the same period.
It’s also worth considering whether a broader data set might have implications for how the rating agencies look at the credit status of the airline industry. Although the universe of airlines with public ratings is small, the rating agencies conduct a lot of private ratings as part of rating aircraft ABS transactions. Although these private ratings remain confidential, general feedback is typically that average credit ratings for the airlines in these portfolios is “high B/low BB” i.e. middling speculative grade.

 

 

A comparison of our estimated airline default rate with the default rate for all bond issuers rated speculative grade by S&P shows airlines with both a lower default rate (3.4% p.a. vs 3.9% p.a.) and lower volatility (1.7% standard deviation vs 2.7%). It looks unlikely that the rating agencies are being overly optimistic about airline credit quality, but the most striking difference arises in volatility where it seems the benefit of geographical diversification more than offsets the benefit of industry diversification (we assume that the S&P issuer universe is concentrated in the US).
Qualifications and Conclusions
One should bear in mind that the nature of a lessor’s credit exposure to an airline is different to that of a bondholder because the lessor typically has the benefit of asset ownership and a lease security package that may include cash security deposits, letters of credit and other features. This has allowed lessors to grant significant forbearance to airlines at various other times as well during the pandemic.
Despite these limitations on making fair comparisons between our airline default estimates and historic bond defaults we believe these results are striking none the less, particularly on the positive benefits of geographic diversification, the limitations of size effects and the lower volatility of airline defaults compared to speculative grade bond issuers.

 

Regular Topics

Macro-Economic Background

Despite the “trade wars” initiated by the US this year the IMF has not materially changed its World GDP forecast in its latest World Economic Outlook published in April, and its first forecast for 2030 shows a similar level of growth to prior years.

 

 

Economic growth is a key driver of long-term growth of air travel. However, since early 2020 its impact has been overshadowed by the fall and recovery in traffic associated with the pandemic. In time the influence of overall economic conditions on air travel is likely to reassert itself, but industry forecasts published by Airbus, Boeing and IATA assume much higher rates of traffic growth than GDP growth over the rest of the 2020s as the former catches up to its long-term trend (see our Q1 2024 Industry Update for a more detailed discussion).

 

 

Two key macro variables have been moving in favour of the airline industry in recent months. The cost of jet fuel has fallen below $2.00 per gallon, mainly due to a fall in the price of crude oil as the crack spread remains quite high at c. $17 per barrel. IATA estimates that fuel accounted for 29% of total industry costs in 2024, and the current prices is 15% below the average for 2024 so this will be a very welcome change if sustained. There will probably be a lag for many airlines in seeing the impact of cheaper fuel in their financial results as they will have hedged their costs forward for periods of up to around a year. The US Dollar has also weakened which also helps airlines outside the US for dollar-denominated costs such as fuel, aircraft rents and aircraft spares.
Another indicator that is potentially important to aircraft investors is the breakeven inflation rate on US Treasury Inflation-Protected Securities (TIPS). This indicator measures inflation expectations and it matters because used aircraft values are strongly influenced by the cost of new aircraft and over time this cost is linked to US Dollar inflation. In the short term this linkage is driven by escalation clauses in aircraft purchase contracts and in the long term by the general input cost environment for the aircraft manufacturers. The chart below compares the breakeven rate for 10-year and 5-year TIPS.

 

 

Although medium or long-term inflation expectations have never gone higher than 3.5%, actual inflation experience has been much higher in the last few years. This has led to higher appraised values for new aircraft. If tariffs are applied to aircraft and/or aircraft components tis is likely to increase the cost of new aircraft.

 

 

Traffic and Aircraft Demand

 

After a very strong January traffic global growth became relatively weaker, although RPKs[2] for the first quarter overall were up 5.3% and a slightly lower increase in ASKs[3] allowed for a modest improvement in load factor. The key growth driver was international traffic in Asia-Pacific which has lagged other markets in its post-pandemic recovery. This was significantly offset by weakness in both domestic and international North American traffic where volatile economic policy in the US appears to have had a negative impact on consumer confidence.

 




 

Although some short-haul aircraft serve international routes nearly all long-haul aircraft do so, and this is reflected in the relative demand for single-aisle (narrowbody) and twin-aisle (widebody) aircraft. Aircraft demand can be measured in terms of aircraft in service and ASKs, the standard measure of aircraft capacity deployed by airlines which indicates how intensively aircraft are being flown. Single aisle aircraft demand on both metrics is higher so far in 2025 than in 2019 whereas twin-aisle aircraft are in line.
The softer recovery for twin-aisle aircraft is mainly due to weak traffic to and from, and within the Asia-Pacific region. The figures by region in the tables above are based on airline domicile, so weak Europe to Asia traffic reduces recorded international RPKs in other regions. This short-term effect is accompanied by a very gradual long-term increase in single-aisle aircrafts’ share of global airline capacity at the expense of twin-aisle aircraft, which is caused by better operator economics and an increase in the number of markets where single-aisle aircraft can be deployed because of the greater range of new technology aircraft such as the A320 Neo and the B737 Max.

 


 

Full recovery has yet to be achieved for twin-aisle aircraft, mainly due to weak traffic to and from, and within the Asia-Pacific region. The figures by region in the tables above are based on airline domicile, so weak Europe to Asia traffic reduces recorded international RPKs in other regions. Twin-aisle aircraft in service has shown a greater improvement relative to 2019 than ASKs which suggests that aircraft are being returned to service with lower utilisation in anticipation of continued recovery.

 

 

New Aircraft Supply

 

 

Airbus delivered slightly fewer aircraft in Q1 2025 than in Q1 2024 mainly due to continuing supply chain constraints which were aggravated by the acceleration of some CFM International LEAP engine deliveries into Q4 2024 to meet last year’s delivery target. According to Cirium there is evidence of an increased level of aircraft rolled out of final assembly, but which have not yet flown which indicates that Airbus has a reasonable prospect of meeting its 2025 delivery target of 820 aircraft subject to resolving its supplier issues.

 

 

There has been no change in Airbus’s production plans (the current production figures in the table above include some external estimates as well as official Airbus guidance). Where Q1 actual deliveries are lower than current production rates this is most likely due to the typical fall back after the “Q4 rush”.

 

 

Boeing’s deliveries increased significantly year-on-year although at least some of these were accounted for by reductions in the number of aircraft in inventory. Boeing disclosed a drop of 20 in its B737 inventory in Q1 which we have deducted from Q1 deliveries to arrive at our estimated monthly actual production rate of 28.3. However, our approach relies on incomplete information as the inventory figures quoted by Boeing are in respect of aircraft produced prior to 2023 and there could well be other aircraft in inventory manufactured later. We have made another estimate for the B787 where the Q1 inventory reduction was 5 aircraft although this is subject to the same qualifications as for the B737.
When reporting its Q1 results in April Boeing’s CEO Kelly Ortberg said that the B737 production rate is up to 31 per month and they are working towards the FAA-mandated maximum production level of 38 per month with a medium-term target of 50. The production target for the B787 is an increase from 5 to 7 by the end of 2025 with a further increase to 10 in 2026.
One additional complication for Boeing is the status of aircraft contracted for sale to Chinese airlines who have refused delivery because of tariffs. Boeing said in April that they have 50 B737s and 4 B787s scheduled for delivery to Chinese airlines in 2025. Where an aircraft is not completed there should be relatively little downside for Boeing because other customers will likely take them given the large backlog for both aircraft types. Where the aircraft are completed there would be significant expense and delay to change to a different customer’s specification, but this can be done. Obviously, the situation around tariffs is subject to a lot of uncertainty and this problem may be resolved through political negotiations.

 

 

The broader impact of actual and potential tariffs on aircraft manufacturers is very hard to envisage at the time of writing as the global nature of their supply chains creates a lot of complexity compared to e.g., cars. Already certain exemptions for major components have been announced such as the Chinese government’s exemption for CFM engines supplied to COMAC.

 

 


Airline Industry Financial Performance

 

Airline stocks fell more heavily than the overall market in April, which is not surprising as the NYSE Arca Global Airline Index is heavily weighted towards US airlines and the latter have published several profit warnings on the back of weak travel demand.
There had been very few airline failures so far in 2025 until Azul filed for bankruptcy last month. Of the two largest bankruptcies in 2024 Spirit exited from bankruptcy in March and Gol will likely follow in June. Azul’s bankruptcy is another example of how difficult it is to avoid this procedure if all one’s competitors have availed of it (see above). Our historical analysis suggests that there will be more financial distress over the rest of 2025 than has occurred to date.

 

Appendix – Airline Industry Default Study Methodology

Our study has some limitations mainly driven by practicality. We only cover all airlines that have operated more than 5 passenger jet aircraft at any year end since 1970 according to Cirium Fleets Analyser. This reduces the number of airlines we need to research from 2,914 with no size cut-off to 1,163, but we still cover 98% of the passenger jet fleet by units.
We also limit our statistical analysis in time to start in 1990 because certain key characteristics of the industry changed around this time, notably deregulation and the rise of aircraft leasing as a source of capital. Going back to 1970 would lower our estimated average annual default rate but we believe that conditions prior to 1990 were not comparable.
We due diligence the history each airline based on public sources and record defaults captured by this research. We also add defaults for airlines in emerging market countries that experience sudden fleet reduction as the events caught by this rule may well represent informal defaults that occur without court proceedings. Our final output is adjusted defaults which combines both research-based and rules-based defaults after eliminating mergers and double-counts. The number of adjusted defaults is c.50% higher than for research-based defaults alone. We measure adjusted defaults based on the number of airlines and the number of affected aircraft. Our airline fleets data source is Cirium Fleets Analyzer.

  • Airline population limited to airlines with 6 or more aircraft at year end during study period – There are 1,163 airlines that had a fleet of 6 or more aircraft out of 2,914 total airlines
  • Historic default events are Research based, or Rules based.
    • Research based events are derived from study of relevant academic/government publications and online resources (Airlines for America, Wikipedia)
      • Single airlines can default on multiple occasions
      • Cessation/suspension of operations assumed to result in default unless there is clear evidence that this was not the case
      • Every single airline with its own licence (“AOC”) counts as a default e.g., both the LATAM and Avianca groups filed for bankruptcy in 2020, and as a result each of their subsidiaries, which also had distinct licences, experienced a default and is counted separately
  • A Rules based event only occurs for airlines based in an emerging market country (per World Bank classification) where default processes are assumed to be more informal. A Rules based event occurs if:
    • An airline with 15 or more aircraft has a year-on-year fleet reduction of 20% or more, or
    • An airline with less than 15 aircraft has a year-on-year fleet reduction of 3 or more aircraft
    • Some events are eliminated if research shows there clearly was no default i.e. Copa Airlines Colombia
  • An airline cannot default two years in a row e.g. bankruptcy shortly followed by cessation of operations in the following year is treated as a single default
  • Adjusted Defaults by Airline are the combination of Research Based and Rules Based events which are filtered to eliminate mergers and duplication of events.
  • Adjusted Defaults by Aircraft take the fleet size the year before the defaults occurs, not the max fleet size for the airline.

 

 

[1] Although 2020 was a record year for defaults there were also significant rent deferrals granted by lessors to airlines which can reasonably be considered as additional payment delinquency that is not captured by our methodology. As most of these deferred amounts have been repaid this was a strong case of enlightened self-interest on the part of the lessors, although this is clearer with hindsight than it was at the time.
[2] RPKs is the acronym for revenue passenger kilometres, which is the product of the number of paying passengers times distance flown.
[3] ASKs is the acronym for available seat kilometres, which is the product of the number of available seats flown times distance flown.

Downloadable Version

 

Disclaimer

This Presentation has been made to you solely for general information purposes and is not intended to provide, and should not be relied upon for legal, tax, accounting, investment, or financial advice. This Presentation is not a sales material and does not constitute or form any part of any offer, invitation or recommendation to the recipient, its affiliates or any other person to underwrite, sell or purchase securities, assets or any other product, nor shall it or any part of it form the basis of, or be relied upon, in any way in connection with any contract or transaction decision relating to any securities, assets or any other product. None of Sirius, its affiliates or shareholders shall have any responsibility or liability to the recipient, its affiliates, shareholders or any third party in relation to this Presentation or any other document or materials prepared by Sirius or its affiliates, officers, directors, employees, advisers, or agents. Sirius and its affiliates, officers, directors, employees, advisers, and agents have taken all reasonable care to ensure that the information contained in this Presentation is accurate. Neither Sirius nor any of its affiliates, officers, directors, employees, advisors, or agents has any obligation to update this Presentation. Under no circumstances should the delivery of this Presentation, irrespective of when it is made, create an implication that there has been no change in the affairs of the entities that are the subject of this Presentation. This Presentation may be updated and amended by a supplement and, where such supplement is prepared, this Presentation will be read and construed with such supplement. The statements herein which contain such terms as “may”, “will”, “should”, “expect”, “anticipate”, “estimate”, “intend”, “continue” or “believe” or the negatives thereof or other variations thereon or comparable terminology are forward-looking statements and not historical facts. No representation or warranty, express or implied, is made as to the fairness, accuracy, or completeness of such statements, estimates and projections. The recipient should not place reliance on any forward-looking statements. Neither Sirius nor its affiliates undertake any obligation to update or revise the forward-looking statements contained in this Presentation to reflect events or circumstances occurring after the date of this Presentation or to reflect the occurrence of anticipated events. The information set out in this Presentation has been prepared by Sirius based upon various methodologies and calculations which it believes to be reasonable and appropriate. Past performance cannot be a guide to future performance. In preparing this Presentation, Sirius has relied upon and assumed, without independent verification, the accuracy and completeness of all information available from public sources or which was provided to it or otherwise reviewed by it. This Presentation supersedes and replaces any other information provided by Sirius or its affiliates, officers, directors, employees, advisers, or agents in respect of the content of the Presentation. No information or advice contained in this Presentation shall constitute advice to an existing or prospective investor in respect of his personal position. None of Sirius, its affiliates, or its affiliates’ officers, directors, employees or advisers, connected persons or any other person accepts any liability whatsoever for any loss howsoever arising, directly or indirectly, from this Presentation or its contents.