Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Standard view
Full view
of .
Look up keyword
Like this
0 of .
Results for:
No results containing your search query
P. 1
51 Fortes Oliveira Catering Services

51 Fortes Oliveira Catering Services

Ratings: (0)|Views: 481|Likes:
Published by W.J. Zondag

More info:

Published by: W.J. Zondag on Dec 30, 2012
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





e-zine edition 51
 An Empirical Study of Ca- tering Costs in an Emer-ging Airline Market
This study presents an empirical model of airline catering costs for the Brazilian domesticairline industry since the late nineties. Airline passengers, used to the aviation’s golden years of high-quality on-board services, have faced severely reduced and, in some cases,charged catering. The study analyzes panel data from statistical yearbooks published by the Brazilian National Aviation Agency in order to investigate this issue. We employ a fixedeffects estimator of the determinants of catering costs. Important factors influencing ca- tering costs such as the airline business model, demand and input prices, are pinpointedand analyzed in light of the economic characteristics of air transportation. Our final re-sults show that the entry of a low cost carrier had a downwards impact in overall cateringcosts, producing a forty percent decrease on average.
Air transportation was in the past marked by excellent stan-dards regarding in
ight services, with catering being one of thebest attributes of airlines. Major carriers used to offer outstand-ing service beyond the average passenger expectation. Brazilwas not an exception, with legacy airlines such as VARIG andTRANSBRASIL, along with the then regional TAM, being themost noticeable in terms of catering standards. However, dueto the huge dif 
culties faced by the Brazilian air transportationsector over the past decade – with the 9/11 crisis, the bankruptcyof major airlines, the Brazilian deregulation act, among others –,cuts in catering costs has become a commonly observed practicein the market. Aware of need of lowering costs to attract newclients, Low Cost Carriers (LCC) emerged in the Brazilian trans-portation market – a trend also observed all over the world. Oneof the most notable characteristics of the business model of theseairlines is to suppress items considered super
uous in the air-line service – the “no frills” classic approach of LCCs. The
rstBrazilian (and Latin American) LCC was GOL Airlines, whichentered the domestic market in 2001. GOL Airlines’ low-costphilosophy was completely opposed to VARIG, which was themajor airline by that time and considered by many, until nowa-days, as the best standard ever with respect to in-
ight servicesquality.According to International Air Transport Association (IATA),catering costs normally account to approximately 2 or 3 % of anairline’s total cost. Besides the production costs, other expensesstrictly related to catering are those involving regulations andstandards of aeronautical legislation and logistics costs. Addi-tionally, each kilogram associated with meals means extra avia-tion fuel burnt. Therefore, the elimination this “extra weight”may make the difference between pro
t and loss for the operator.Recently, one alternative to the complete elimination of in-
ightcatering has been the practice of charging an extra fee, i.e., acharged catering. Table 1 shows how prices vary for extra itemsin some airlines over the world.
Table 1:
Price comparison for products sold inside by main low-cost airlines.
Values in USS
GOL Webjet Spirit Air Asia Ryanair
Country Brazil Brazil UnitedStatesAsianCountriesEuropeanCountriesBeverage 1.71 unavail-able5.29 2.7 1.6 (150ml)Mineral Water (300ml) 1.71 unavail-able5.29 2.7 2.43 (150ml)3.39 (500ml)Beer 5.71 unavail-able10.54 unavail-able5.22Sandwich unavail-able6.85 unavail-able3.8 6.57Hot Meal 8.57 unavail-ableunavail-able4.9 unavail-ableCombo(sandwich or hotmeal + non alcoholicbeverage + candy)8.57 10.28 unavail-able10.3 10.6Salted Peanuts 1.71 1.71 unavail-able2.17 2.61
 Adapted from site
by: João Luiz de Castro Fortes and Alessandro V. M. Oliveira
2According to An e Noh (2009), passengers tend to be more sen-sible to the quality of services as it interacts directly with them.Moreover, from the passenger perspective, overall the airlineimage may be directly associated to the quality of in-
ight ser-vices. Although the de
nition of quality may be subjective inmost cases, one important characteristic perceived by the pas-senger is the standard of meals served by the airlines.This study brie
y evaluates the conditions of catering in theBrazilian airline industry. We use panel data collected from theStatistical Yearbooks of the National Aviation Agency (ANAC),overing the period of 1997 to 2008. We use unit cost of cateringby airline as a proxy for assessing “catering quality”. Unit costsare total catering costs divided by total enplanements. We employan econometric model of unit catering cost to uncover some of its most important drivers. We also analyze how these expensesevolved within a ten-year period along. Others metrics for cater-ing quality, such as passenger opinions on airline catering qual-ity could also be obtained from surveys, and would be probablymore adequate for this kind of analysis. However, this kind of information is typically hard to get from airlines and would notover a long period suitable for econometric investigation.
Data Analysis
The data used in the modeling of problems were obtained fromthe Statistical Yearbook of Air Transport, Volume II, publishedby the airline regulator ANAC. The yearbook comprehends 1997to 2008 data from all regular air transport Brazilian airlines.Some important facts occurred during these years in Brazilianivil aviation and have to be brought up to the reader for a fullunderstanding of possible data oscillations. The
rst main fact isthe deregulation act stated in 2001 and represented a regulatorylandmark for the aviation sector. This regulatory reform createda harsher competition among airlines. Some of them could notope up with the new environment and exited the market, whilenew entrants emerged in the air transport market with new ideasand business models. Another fact that it is important to mentionis the bankruptcy of many airlines. Among the ones that entereda period of poor
nancial condition, VARIG was the most im-portant. As mentioned before, it was considered by passengersas a benchmark in terms of catering quality.
Catering Services Costs
rst analysis performed was the study of how airline aver-age catering expenses per passenger in a trip have evolved dur-ing the investigated years. Also, an analysis regarding cateringrevenue would be interesting for the full understanding of howthe quality of that service is varying. However, this data is notavailable due to the fact that charging for this kind of servicesnboard is a novelty in the Brazilian air transport industry. How-ever, in the near future it is expected that the ANAC will be moreinterested in studying this new phenomenon and will request air-lines to disclose this information. We strongly recommend theregulator to embrace this idea.Initially, Graph 1 shows that the observations are somewhatdisperse, meaning a high variability of expenses with cateringservice. However, it can be seen that the average value and itsvariability had been decreasing along the years. This fact can beaused by the entering of the low-cost philosophy in the air trans-port market along bankruptcy of the major high-cost airlines.
Graph 1:
 Heterogeneity of catering services expenses.
Another market dynamic that has to be evaluated here is thekind of equipment which the airline is operating. It is expectedthat the higher the number of carried passengers per
ight is, thelower the average catering expenses are. This fact is caused byeconomies of density that the airline may obtain from ful
llingits aircraft more ef 
ciently. Note that some of the catering costsmay be considered
xed, such as logistic costs and regulation-related costs. Contrary to this expectation, we have Graph 2which shows that the average catering cost of aircraft such asthe ATR-42 and the Fokker F-50 – both with less the a hundredseats, is almost the same as of an Airbus A320. We believe thatfurther investigation of the presence of economies of density incatering costs is necessary however.The type of 
ight may affect in-
ight services’ expenses aswell. Of course, we have noti
ed that transatlantic international
ights last longer and have more meals served domestic
ights.This fact makes the unit cost of catering in aircraft operating ininternational routes to be higher – for example, Boeing 747-300and DC-10 -, as we can see in Graph 2.
Graph 2:
 Heterogeneity of catering service expenses for different kind of aircrafts.
Graph 3 shows the representativeness of in-
ight expenses inairline total costs along the studied years. It is possible to see thediminishing of dispersion among airlines across the years, mean-ing that cost convergence has become a reality in the industry.
Graph 3:
 Impact of catering services expense in airlines’ total cost.
The Econometric Model
Initially, observing Graph 3, it is possible to see a trend of di-minishing cost for the airlines for the last years. Therefore, atrend variable (IND_T) was included in the empirical model.Along with that fact, the demand for air transport has been in-creasing in the last years. Many of the “new passengers” (pas-sengers who are using air transport for the
rst time) are strong-ly price-sensitive. For those passengers, although the qualityof onboard service is important, the ticket’s
nal price is themain attribute that determines his or her choice. Therefore, weexpect that along with the increasing demand in air transport,there will be a decrease inprices in order to makeit economically more at-tractive to this type of consumer. It is reasonabletherefore to assume thatone of the core variableswould be demand for airtransport. However, de-mand is a variable whichdepends on other factors,with gross domestic prod-uct (GDP) being one of the most important.Another variable of theempirical model is air-craft average stage length(AVST). We believe that longer trips are associated with highercatering costs. Also, we expect that the exchange rate (USDexpressed in local currency, BRL) would affect catering costs,particularly in international trips when international cateringcontracts may have values quoted in American dollars. In oth-er words,
uctuations in the exchange rate may affect directlythese costs. International trips have another fact to be observed:since they are typically longer and use wide-body aircrafts, thesecosts are higher in most cases. Moreover, with bigger aircrafts,airlines offer other classes than economy class, which increasescatering costs. Concluding: we expect that the type of segment(either international or domestic) will directly in
uence cateringcosts. We use a dummy variable SEGMENT1 to represent thekind of segment in the model.Finally, the entry of GOL in the market is essential in order toexplain the evolution of catering costs in Brazil. GOL’s entrynotably in
uenced other airlines to become more competitive. Adummy variable ENT_GOL represents this period. Ultimately,this dummy variable also accounts for the effects of airline de-regulation in Brazil.We have then,We employ a
xed-effects estimator to account for the strongheterogeneity observed across airlines/aircrafts/periods in thedata. Estimation results with airline
xed effects are presentedby Table 3. We have all variables statistically signi
cant at a10% level. Most of them are statistically signi
cant at 1% level.All coef 
cient signs corroborate theinitial hypothesis. We have consid-erable evidence that the increase of number of passengers onboard, theunceasing depreciation of BrazilianReal and the increasing average stagelength will result in increasing costs.On the other hand, the increase of GDP results in more people travel-ling and therefore a greater numberof passengers. This scenario wouldlead to a more competitive air trans-port market for airlines and, meaningthe necessity of cutting off cateringservices.Moreover, it is possible to identifythat GOL’s entry had a statisticallysigni
cant coef 
cient. The estimated impact amounts to approx-imately 40% on average. This means that its entry into the airtransport market (along with its low-cost-operation idea) causedthe lowering catering costs (coef 
cient’s negative sign) in a gen-eral way for all Brazilian airlines.
Table 3:
 Results for regression with
 xed effect of Airline
lnexpperpax Coef. Std. Err. T P>|t| [95% Conf. Interval]lnrpax
0.0554 0.0146 3.8000 0.0000 0.0269 0.0840
0.1534 0.0322 4.7600 0.0000 0.0902 0.2166
-0.8788 0.1391 -6.3200 0.0000 -1.1514 -0.6062
0.3001 0.0736 4.0800 0.0000 0.1559 0.4443
-0.4182 0.0961 -4.3500 0.0000 -0.6066 -0.2297
0.0155 0.0091 1.7000 0.0900 -0.0024 0.0334
0.0003 0.0000 12.7900 0.0000 0.0002 0.0003
Tuna Sandwich and Soft Drink on sale on board low cost Air Asia
www.airlinemeals.net Lachlan Burnet 
is the natural logarithm of average expense, inlocal currency (BRL), for each passenger for each trip,
is the trend variable,
is Gross Domestic Product, in billions of BRL,
is the natural logarithm of the number of revenue passen-gers,
the average trip length,
is the exchange American Dollar – Real (BRL/USD),
is the dummy variable dummy which means do-mestic or international trip,
is the dummy variable which represents the periodbefore and after GOL’s entry.

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->