11-12-2012, 02:08 PM
OPTIMAL FINANCIAL KEY PERFORMANCE INDICATORS: EVIDENCE FROM THE AIRLINE INDUSTRY
OPTIMAL FINANCIAL KEY.pdf (Size: 495.66 KB / Downloads: 36)
ABSTRACT
Selecting relevant Key Performance Indicators involves an assessment of both cost- and revenue-driven
measures. Cost driven allocation usually predominates, due primarily to a traditional accounting mindset
coupled with the need for cost savings in the current economic environment. Using data from the airline
industry in all of the major markets in the world, this paper demonstrates that revenue- or profit-driven
KPIs, consistently applied, will more likely lead to better financial performance than ‘flying’ the business
based on cost-driven metrics or those representing a mixture of revenue target and cost-driven metrics.
Specifically it examines the effectiveness of models that characterize performance based on two performance
indicators, in particular – seats and passenger-kilometers. We document strong evidence indicating
that Operating Profit per Passenger or per Passenger-Kilometer is the most significant variable when
it comes to explaining the variation in airline profitability. Our conclusion is that despite the traditional
belief that measuring performance per seat is only appropriate for point-to-point destination services,
typically provided by Low Cost Carriers, the same model also fits Full Service Network Carriers and
thus, can be used by them as a meaningful tool for financial targeting and strategic decision-making.
INTRODUCTION
espite airline industry growth over recent decades, the majority of airline businesses remain consistently
unprofitable over entire business cycles. This paper is an empirical study that attempts to
distinguish between cost driven and revenue driving financial performance indicators that may
better help us predicate an airline’s financial performance. Our main assumption underlines the impact of
using two different Key Performance Indicators (KPIs) models. We examine the effectiveness of models
characterizing performance based on two activity drivers – seats or passengers (revenue driving) and passenger-
kilometers (cost driven). It has been traditionally considered that measuring performance by seats
is only appropriate for point-to-point destination services, typically provided by Low Cost Carriers and
not relevant for Full Service Carriers.
Our key findings indicate that a performance model based on kilometers fits the industry slightly better
than the one based on passengers (seats). Furthermore, we find strong evidence indicating that Operating
Profit per Passenger or per Passenger-Kilometer is the most significant variable explaining airline profitability.
In spite of classical beliefs, we found it is more meaningful than revenue, cost and load factor traditionally
used by the industry. We also found the relationship between profit margin and seats-based model
is strong enough for both classes – LCCs and Full Service network carriers. Therefore, we arrive at the
conclusion that Operating Profit per Seat can be successfully used for targeting the financial performance
of Full Service Network Carriers. The remainder of the paper is organized as follows. Section 2 provides
a summary of the relevant literature. Section 3 is a description of data and methodology presented. Section
4 discusses the results of the analysis while Section 5 contains concluding comments.
LITERATURE REVIEW
According to Doganis (1985), the profitability of an airline depends on the interplay of three variables,
unit costs, unit revenues or yields and load factors achieved. Airline managers must adjust costs, fares
and load factors to produce profitable combinations. He characterizes the industry by short-run marginal
costs, which are close to zero. Marginal cost of carrying an extra passenger on a flight, which is due to
leave with empty seats, is no more than a cost of additional meal, an airport passenger charge, the cost of
ground handling and a few pounds of fuel burnt as a result of extra weight. The run of these costs is short,
because if the seats remain unsold, these seats flown or seat-kilometers produced will be lost forever.
Therefore, he suggests maximizing revenues and load factors.
For passenger airlines, the average revenue per output sold is called Yield and measures average revenue
per passenger, per passenger kilometer, per passenger ton kilometer performed. Thus, he reasons the existence
of low cost carriers, stating that by combining passenger yields with low cost and relatively high
load factors one can achieve profitability. He also demonstrates that low cost itself does not provide big
margins interacting with low revenues, nor does high cost necessarily mean low profits if the revenues are
high enough. Doganis concludes that airlines deciding on their pricing strategy, and working out various
tariffs, must balance and assess all these factors, which transform the various fares into average yield. He
states that it is the yield in conjunction with the achieved load factor and the unit costs, which will determine
whether an airline’s revenue and financial targets can be met. To assure such process airlines apply
revenue management process, underlying revenue management systems.
DATA AND METHODOLOGY
We were able to collect operational statistics (number of passengers, number of kilometers offered/flown,
employees, aircrafts, etc.) from open sources, as for each specific company so for the industry. The classification
of airlines in this paper will follow a model used by the Research Unit of German Aerospace Institute
(DLR), thus, we distinguished airline companies by those of (abbreviation in brackets): Full Service
Network Carriers (“FSNCs”), Low Cost Carriers (“LCCs”), and Regional Carriers (“Regionals”)
Holiday / Charter Carriers (“Charters”).
Full Service Network Carriers are scheduled airlines with a business model that focuses on providing a
diverse and extensive service. These are typically international operating companies with a networkoriented
system (normally with one or more hubs), covering a wide geographical area and providing
transportation in several different classes. Low Cost Carriers category comprises those airlines that offer
low prices for the majority of flights and which mainly operate on short and medium-distance routes with
low overheads and a relatively high load factor; these airlines use a no-frills business model. We will not
work with Regionals or Charters, because their market influence is insufficient.
RESULTS
This section describes results of analysis and empirical testing. The sample includes 5 years data for 27
companies, i.e. 135 cuts on an annual basis. The analysis overlooks all variables and ratios used in both
Kilometers and Passenger modes and includes Minimum, Maximum, Mean, Standard Deviation, Skewness
and Kurtosis. We ran the correlation analysis for both models and three data clusters (Region, Model,
Performance). Analysis of separate clusters uncovered interesting facts about the airline business in
different classes and continents, relevant to cost-driven and revenue driving metrics. For the empirical
testing of two KPIs models, we applied regression analysis.
CONCLUDING COMMENTS
In attempting an empirical study identifying a workable model for predicating airline financial performance,
this paper reviewed commonly used metrics in the airline industry and in particular examined the
effectiveness of models that characterize performance based on two activity drivers – passengers and kilometers,
revenue drivers being passenger based, and cost drivers, being kilometer based. The study covered
27 top carriers over a 5-year period. The data was initially clustered according to airline type, region
of origin and operation, high or low financial performance, and then, analyzed in terms of peculiar properties
followed by a correlation analysis for three data clusters. Participating variables were checked for
multicolinearity, and variables strongly correlating with the dependent variable were excluded. 12 multiple
regressions were run on each data cluster with two different dependent variables such as Operating
margin percentage and Return on Assets percentage.