The difficulty to find the appropriate award pricing strategy

10 April 2019

With United announcing the removal of award charts in favour of a dynamic award pricing, many fear hidden increases of award prices. The more general question is though to what degree price increases in loyalty programs are justified at all.

While it is obviously too early to tell what impact United’s new pricing strategy will really have on concrete award levels in its MileagePlus post 15 November, it is safe to predict that it will follow the general global trend and result in a more or less important price increases especially in premium cabins. Virtually all Frequent Flyer Programs having revised their award charts as of lately have followed that pattern. Singapore Airlines, Thai Airways, South African or SAS are just a few examples.

Whether fixed or dynamic award tables apply – both in the airline and hospitality industry -, price increases are generally explained by increased retail prices. That sounds like a logical statement in first place few would question. But it is actually a wrong argument in many cases. It all depends on what basis accrual and redemption takes place.

There are three cases to distinguish:

1. Accrual and redemption on a distance basis:
This is the most traditional form of Frequent Flyer Programs and still the one dominating by far at a global scale: You fly one mile and you earn one mile in your account, with a multiplicator applied in function of your service/booking class. Awards are offered based on a fixed award table.

With increasing flight prices, this means that you have actually to spend more money to get the same number of miles. If a full fare Economy Class roundtrip ticket between London and New York used to be 2,000 USD and runs now at 2,500 USD due to inflation, higher fuel costs etc., you spend more money, but still get the same number of miles.

Increasing award levels is hence not justified since the price increase already reflects in the higher amount of money spent to accrue these miles.

It is also worth noting that revenue-based programs (cf. below) fall in this category if you use them on partner airlines, e.g. an American AAdvantage member accruing on British Airways, where a corresponding distance logic continues to apply.

The issue here is that this reasoning applies to an airline guest – but the contrary applies to a credit card holder, where the number of accrued miles increases only due to inflation. If you charged 20,000 USD to your credit card ten years ago – earning 20,000 miles -, the same purchases would equal 24,380 USD/miles today, based on a 2% annual inflation rate. If you look at inflation rates in certain markets, which may go up to 10% p.a., you easily understand the issue.

With non-air credits representing a share of some 50% of mileage accruals in many programs, price increases are indeed more justified (or even necessary) from that perspective – but they do penalise the actual airline customers and are therefore not really perceived as fair treatment by those members, which should be considered the truly best customers of the airline – and by extension of the loyalty program.

2. Accrual on a revenue basis, redemption on a fixed basis
Most legacy programs having moved to a revenue basis, but maintaining a traditional award chart, would be in this category as well as the majority of major hotel programs – even if they apply some opaque „dynamic“ pricing, which is though not directly correlated to the retail price (or at least not recognisable as such).

Once the initial calibration between a revenue-based earn rate and the appropriate redemption rates has been made correctly – where most programs would fail as they only introduce a new accrual rate without touching the redemption part -, a reasonable price increase of redemption prices in line with the evolution of retail prices seems justified if you look at it from a transaction perspective: For the same service, members pay more and earn hence more points. Without increasing award thresholds, they would ultimately need a lower number of transactions to achieve any given award.

The challenge is here more on the psychological side since members don’t necessarily realise that they really earn more (or don’t want to be reminded that they keep on paying more each year for the same product) while they are faced with directly visible and perceived price increases on the redemption side. This basically shows the limitations of this in-between model, although it gains in popularity across the globe.

3. Accrual and redemption on a revenue basis
Many low cost airlines operate on this basis, where you earn a % of the ticket price in points and each point has a clearly defined cash value. Accor is the biggest hotel program operating on such a basis as well.

This is obviously the most balanced system where rates are regulated automatically. If points are not used over a longer period, they face the risk of devaluation, but that process is clear to members and doesn’t surprise them – unlike if you have to announce price increases in a traditional award table. You are obviously well advised not to touch the redemption value per point (as Southwest was obliged to do some time ago), but to do the initial projections correctly.

This is probably the cleanest model, but not the most popular one among members as there are hardly any possibilities to „play“ the system and to increase the value of points like that. So, if the upfront discount is at a rather low end of 2 or 3%, it is definitely more difficult to create interest in the program among members since the return is somewhat limited, especially compared to the 10% or more you could achieve with traditional mileage-based programs.

In whatever of these three cases your program situates, it is important to understand the balance between accrual and redemption. Changing only one side of the equation will hardly ever work in the best interest of the program. Paying the required attention to such an exercise, while looking both at the economics of your program and the competitive landscape, is definitely required. If you, as a program manager, are fully satisfied with the outcome, it is highly likely that you did something wrong since such a balance act will automatically need to result in some form of compromise in almost all cases.