Looking forward, Hopper anticipates its Consumer Airfare Index will bottom out in January before rising again in anticipation of spring:
Figure 1: Actual average domestic consumer airfare prices through September 2015 (solid line), with six-month forward forecast price levels (dashed), comparing current and projected prices.
What informs our prediction for the future? Typically, after prices drop sharply at the end of the summer, they begin to stabilize as holiday travel demand replaces summer demand. By December, most holiday tickets are purchased, causing prices to decline again in the new year. However, so far this fall, unlike in previous years, prices have not stabilized and instead have continued to tumble. We believe this pattern will slow as we reach the end of the year.
While it is broadly expected that prices in the fall will be lower than in the summer, it's noteworthy that the typical fare was 18% lower than in September 2014, when it was measured at $259. Prices this year are well below last year's levels, and we expect consumers to continue enjoying better deals for the foreseeable future.
Prices this year have been significantly lower than last year, and predictions indicate that they will continue to stay low through the winter. Lower fuel costs are one factor we've consistently cited as a cause of this year's lower prices and we wanted to dig in deeper and explore to what extent this is driving the great prices consumers are enjoying. While the plunge in fuel prices early this year has not been completely reflected in lower ticket prices, it did coincide with the pattern of lower airfare we've observed this year:
Figure 2: Our airfare index (left scale) compared to oil spot prices (right scale), showing how lower airfare this year coincides with decreased oil prices
It's evident that the seasonality in oil prices is broadly similar to the seasonality in airfare, with both rising at the same time this summer. While it's no surprise that oil prices rose this summer, it's difficult to disentangle the seasonal effect from the expectation that prices would eventually return to normal. After decreasing again in August and staying low in September, however, it's clear that low oil prices are, for now, here to stay.
So, how did this pattern affect our index this month? Here's how we estimate different factors contributed to the 9% drop:
Oil prices accounted for a 3.25% drop in airfare in September, but how does this compare historically? Here's a look at this contribution over time:
Figure 3: Contribution of oil prices to change in airfare
Historically, oil prices have contributed relatively little changes in airfare, but over the past year it's influence has grown, pushing down airfare the last two months and earlier this year but pushing it up this summer. The good news here is that we are seeing airlines pass on some of the fuel savings to consumers.
We predict prices nationwide will decrease slightly in October, though some airports will experience price increases. Here are the airports we predict will see the biggest average increase in the price of departing domestic flights in the month of October:
On the other hand, some airports will experience more substantial decreases in price. Here are the airports we predict will see the biggest average decrease in the price of departing domestic flights in the month of October:
The map below shows projected prices in October for flights departing from each state, illustrating how prices can vary across the country:
The more sparsely-populated states with fewer airports have higher prices. Wyoming tops the list with average prices over $400, while Massachusetts, Illinois, and Florida are at the bottom with prices under $200. Smaller states also tend to have more volatile prices, evidenced by our October projections for airfare in each state. Prices are predicted to increase the most in New Mexico (over 7%) and decrease the most in Connecticut (over 5%).
While the figures above show how much it costs to fly from a given state, it's also interesting to see how much it costs to fly to a given state:
Figure 5: Average airfare for flights arriving in each state, the opposite of Figure 5. The pattern here is similar: it's more expensive to get to more remote states with smaller and fewer airports. To look at average prices for flights from a specific airport, use our interactive consumer airfare index tool at http://www.hopper.com/flights/tools/priceindex/index.html
Our Consumer Airfare Index combines search data for every origin and destination in the United States, providing a near real-time estimate of overall airfare prices - unlike other comparable indices that can lag by several months.
Our Consumer Airfare Index represents the price of tickets available for purchase in a given month, not necessarily for travel in that month. Since travel prices are represented in both time dimensions -- time of purchase and time of travel -- it can be difficult to interpret price dynamics. We use date of purchase because it reflects the price consumers are paying at a given point in time, and we report it alongside the typical advance purchase date to give an idea of how these prices translate into travel dates.
Other indices simply take the average of all fares to represent overall price which skews the results toward expensive fares and can give an unrealistic impression of the true cost of flying. We instead use what we consider to be a “good deal" for each route to reflect what consumers should reasonably expect to pay.
Since our index is constructed and forecasted at the origin-destination level, we can also provide comparable estimates for any combination of routes and extract insights on pricing not only across time, but also across different markets. We use monthly passenger data from the Bureau of Transportation Statistics to ensure that each domestic route is properly represented in the final index based on its share of total passengers.
When predicting future prices, we also consider a few key features of airline pricing. First, prices within a given route will fluctuate with the number of passengers.
Second, prices change predictably with the seasons, especially during the peaks of summer and holiday travel. Of course, much of this variation has to do with increased demand - but in peak travel seasons, airlines can raise prices not only because there are more people interested in traveling, but also because the average traveler is willing to pay more for their summer vacation or trip home for the holidays.
Finally, changes in prices may persist, especially if there are underlying conditions pushing prices up or down, as these effects may be spread over several months. Conversely, the opposite may be true - after a big price increase or drop, fares are more likely to change in the opposite direction in future months. Since dynamics like these and the above aren't always consistent, we evaluate future prices at the origin-destination level to capture the unique properties of pricing for different routes.
Of course, predicting the future is no easy task, and many factors that influence pricing are simply unforeseeable. However, by exploiting the factors that are predictable, like trends in passenger distribution, seasonal variation, and recent price activity, it's possible to extract insights about the near future of pricing.
This is the sixth month we've published a forecast - allowing us to track our current estimates against what we've predicted previously.
Our index came in significantly lower than our forecast due to an unusually dramatic price decrease in September.
This has caused us to similarly revise down our projection in price levels, though the growth rates remain broadly the same. It's possible, however, that prices may rebound more strongly than anticipated in reaction to the particularly low prices in September.