Amazon's Prime Air cargo operations were revealed in 2016 with announcements from Atlas Air and Air Transport Services Group (ATSG). Amazon reached an agreement with both companies to operate the flights on Amazon's behalf. It's possible that Amazon will get more involved in their cargo operations in the future as the company acquired warrants to purchase stakes in both ATSG and Atlas as part of the operating agreements. At the start of 2017 Amazon also announced plans for a cargo hub in Kentucky.
Over the past year both companies have been bringing on planes for their Amazon operations according to their financial disclosures. Air Transport Services Group began operating eight 767s for Amazon in Q2 2016 and has ramped up to 18 planes by the end of Q2 2017. Atlas Air's operations were slower to start and reached six 767s by Q2 2017. Overall, Amazon is targetting 20 planes per company with all 40 scheduled to be in service by the end of 2018.
The Department of Transportation provides freight statistics by airline. We can use these to calculate the total freight each airline carried per quarter. We also know when ATSG and Atlas started brining on airplanes on Amazon's behalf. So we can guesstimate the incremental volumes that are due to the Amazon agreements.
Note that this analysis only deals with US air freight carriers. It DOES NOT include freight carried by foreign carriers (for example Cathay Pacifc Cargo).
ATSG started operating on behalf of Amazon in April 2016, though Amazon was already an ATSG customer. As a reference point, Amazon accounted for 19% of ATSG's revenue in Q1 2016 and by Q2 2017 this had increased to 42%. Based on the company's financial disclosures, nearly all of the incremental planes that were coming online over this period were 767s for Amazon. So it seems reasonable to assume that any change in the company's freight volumes were largely due to Amazon's Prime Air (though it is possible that other contracts were seeing changes in volumes too). ATSG's total freight and assumed portion for Amazon's Prime Air are shown in the chart below.
Atlas Air started operating 767s on behalf of Amazon in August 2016. However, looking at the number of airframes brought online vs. the change in freight volumes from Q2 2016 until Q2 2017, it isn't as reasonable to assume that all the incremental volumes are solely due to Amazon. ATSG was operating 3x as many 767s as Atlas was for Amazon in Q2 2017, yet Atlas had more incremental volume since they started flying for Amazon than ATSG. So the Atlas freight volumes are likely being impacted by other areas of their business which fits with commentary from the company.
As a lower bound for Prime Air's market share we can use the estimated incremental volume for ATSG since nearly all of the new frames the company brough online were going to work for Prime Air. For an upper bound we can assume that all of Atlas Air's incremental volume since announcing the Amazon deal is due to that deal (this is obviously aggressive, but can still serve as a reference point). Combining the lower and upper bound we can estimate that Prime Air accounts for 1.5-4.0% of the overall air freight market as of Q2 2016.
How has that impacted the industry? It turns out the overall volume of air freight is growing, so Amazon's own Prime Air isn't necessarily hurting other carriers. The chart below plots the market's total freigh volumes, excluding the estimates for Prime Air, and you can see that it has continued to grow despite Amazon contracting 24 planes by the end of Q2 2017.
The analysis is based on data for US carriers and excludes foreign carriers (for example, Cathay Pacific Cargo). So it focuses on demostric freight operations. However the data probably captures flights from Asia that stopover in Alaska on their way to the US. So the overall market data probably overstates domestic volumes which in turn probably understates the market share of Prime Air.
It's possible that there are other significant drivers in ATSG's and Atlas' freight volumes. It could be that their other flights are operating at higher capacity and that the Amazon flights aren't carrying much freight. In fact, a Reuters report notes that the Amazon flights were often carrying freight that didn't weigh much, but was high in value.
If you have suggestions for other data sources that could be used to estimate Amazon's Prime Air freight volumes, please reach out to me below.
Using data from the DoT's Bureau of Transportation Statistics we can add a new twist to traditional airline route maps. With this data, we can produce route maps weighted by the number of passengers that fly each route to create a more informative visualization of airline route networks. The data also allows for viewing the routes by airport, in addition to by airline.
In 2016 there were over 390,000 taxi trips to O'Hare Airport from the Chicago area. After dropping off their passengers, not all drivers waited in the long queue to pickup a new passenger at the airport. This provides an opportunity to investigate whether waiting or leaving the airport is the best strategy to maximize revenue and tips.
Currency manipulation can be a complicated concept to grasp for those that haven't worked in finance or economics. Here we try and walk through the basics so that you can better understand the discussions that pop up in the press every so often. A few recent examples are used to illustrate the basics and give some background on the topic.