Note – Please see article for a more updated worst case bound/
While it may not be fully known the impact this virus will have on air travel and subsequently pilot demand , we do have enough information to construct an idea of what the impact may be.
Two epidemic models were shared at Nature.com
The best case model from the article estimates a million people infected in China with a peak in late Feb early March.
The worst case model in the article estimates a peak between late March and May with a total of 550-600 million infected in China. The models assume a large number of the infections are asymptomatic.
While the market realist charts (Above) only reflect the RPKs change for the SARS epidemic it is still instructive, as it shows what happens to revenue markets directly and indirectly affected by an epidemic outbreak.
The upper graph reflects the RPK response to markets directly involved in an epidemic, while the lower model shows an indirect effect since SARS never made it in epidemic form to the US during that outbreak.
This exercise will use the best and worst case model from nature.com and overlay the V type direct and indirect RPK impact from the market realist chart to attempt to model a pilot demand forecast given these scenarios.
The article assumes a 33% drop in RPK traffic at the peak for directly affected markets and a 24% drop in RPK traffic at the peak for indirectly affected markets.
The models we are building assume best case, a peak in March with a 6 month V and a worst case 10 month V with a center peak/dip in May.
It’s important to note that airlines will not directly cut capacity to match the drop in passenger demand. They may reduce capacity in proportional ways but it won’t be one for one. Some important elements need to be addressed before we continue an attempt at applying this to the model.
- A pandemic is a relatively short term anomaly in an airlines planning cycle.
- If hypothetically an airline needed to reduce staffing by 13% the spool down time, the associated training churn associated with displacements and the spool up (2-3 months) for pilot training would necessitate a longer time horizon than this models best and worst case scenarios to make it economical.
- Most airlines usually build staffing models around their peak season in the next 12 months.
- We appear to be 2 months into the drop further shortening the best and worst case scenarios.
- Because the airlines are relatively healthy and have shed much of their legacy costs they have more breathing room to absorb the impact.
- In a recession the time horizons would change and that outcome is explored at the end of the article.
So moving forward with the model let’s assume the summer peak no longer exists for pilot demand planning, although in the best case scenario the end of summer could still be a player. The next peak period would be the Holiday period 2020. Based on the SARS RPK recovery (barring a recession) the model does recover and we could estimate an previously planned growth of 3-4% compared to previous holiday traffic by the end of the year.
With this optimistic model hiring could be slowed as long as time remained in the year for an airline to achieve targets by the Holiday season. Keep in mind AA and UA both have 737 Max’s (24 and 14 out respectively). It’s hard to know if they’ve already staffed for the loss in capacity (2.5% and 1.7% respectively). If they have this could cushion reductions in total capacity, and in a strange twist of fate become an advantage.
Let’s build the worst case model and then build something somewhere in the middle. Based on this recessionary report by IATA from 2010. https://www.iata.org/en/pressroom/pr/2010-01-27-01/.
A passenger demand reduction of 3.5% was measured annually YoY from 2008-2009, the largest post war decline. This is likely over-aggressive in an attempt to model a recessionary environment but nevertheless provides a good bound in our model. Assuming a reduction in Holiday traffic by 3.5% for 2020 vs 2019 caused by a recessionary environment the model looks something as follows.
As can be seen the mandatory retirements dampen the effects of a recessionary environment. The “Max factor” simply reflects the possibility that capacity reduction in this scenario could be dampened by the already parked Max fleet.
Other important elements to consider when interpreting this data. https://thediplomat.com/2020/02/major-us-airlines-suspend-flights-between-us-china/
- United has 12% of its revenue tied to the Asia market
- Delta has 6% of its total revenue tied to Asia
- American rounds out one of the least exposed legacy’s at 4%.
The above models are built off of equal exposure, so UA’s could be slightly worse and AA slightly better. It’s difficult to predict how they will be effected differently and in the end if COVID-19 becomes pandemic the effect might be very similar as many geographical areas could be affected.
Finally let’s plot something in the middle. IATA has previously reported (prior to COVID-19) that they were expecting an industry wide 4.1% growth. Now as of the 20th of Feb they expect a 4.7% reduction in global air traffic bringing the loss closer to a .6% contraction. Probably a pretty accurate estimate given the previous trajectory.
Due to the retirements and based on IATA’s projection there is still a healthy demand for hiring this year. What can we take away from some of these models? While no one knows entirely how the COVID-19 epidemic will play out. The models provide some reasonable bounds based on previous market and pandemic experience .
The good news is due to the high numbers of retirements a furlough of any kind is highly unlikely. On the contrary airlines will still need to hire pilots throughout the next year depending on how the impact of this epidemic is contained, and their unique exposure to the market changes.
It is possible in an effort to reduce cost throughout the next 6-10 months that airlines may slow hiring, offering various leaves (with flexibility to reach holiday staffing targets). This will allow airlines to more closely relate cost to revenue during the epidemic V drop discussed at the beginning of the article, while not incurring large training costs associated with displacements. Some legacy’s may not need any of these tools as the number of retirements remains so high. Early retirements will likely slow with the drop in the market, making mandatory retirements a more reliable number in forecasts.
In every problem there usually lies an opportunity if airlines are smart. Air travel has a very good chance of rebounding, and what airlines do during the 6-10 months to make the most of their unique opportunities may bring important results.
- With so many concerns about pilot supply at the regionals this brief slowdown in hiring provides an opportunity to catch up in training and recruitment and further “sharpen the saw” in the pilot pipeline.
- The breather may provide an opportunity for UA and AA to catch up to Delta in performance by reducing scheduling pressure over this next summer as market pressures relax.
- This may provide some needed slow time to schedule training for the Max, and further complete important IT upgrades/airport capital projects etc to be better equipped for the market to recover.
As a final thought, and even despite the best efforts to predict the future, its always a good time to live within ones means, have an emergency fund and try to save for a 3-6 month rainy day.