What is the relationship between the Malaysian aviation industry and the economic development of the country? : a case study based on AirAsia

(1)

What is the relationship between the Malaysian aviation industry and the economic

development of the country? A case study based on AirAsia.

Bachelor Thesis

Romy Horeman

Student number: 10751165

E-mail: romy.horeman@student.uva.nl

Supervisor: Stan Olijslagers

BSc: Economics & Finance

Faculty of economics and business 

(2)

Abstract

This paper examines the causal relationship between air transport and economic growth in Malaysia. Using a time period of 46 years (1970-2016), the Engle-Granger and Johansen co-integration test method are applied, followed by Granger long-run and Wald short-run causality tests. The results can not confirm a relationship and thus ends in interdependence between GDP and air passenger traffic. There is only a short-run causality from the aviation industry to the economic growth. In contrast to existing literature, there is no uni-directional or bi-directional relationship.

Statement of Originality

This document is written by Student Romy Horeman who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents. 

(3)

1. Introduction

Air transportation has appeared as a key facilitator of economic development and social change, because of its facilitation to the flow of people, goods, capital and information (Dai, Derudder, & Liu, 2018). This is especially accurate for South East Asia, which has been one of the most economically dynamic and strategically significant regions in the global economy. Over the last few years Asia has been the world’s most dominant region for air travel (Atkinson, & Palumbo, 2018). According to Atkinson and Palumbo, 35% of all passengers which were carried in 2016 by airlines globally flew on Asian carriers (2018). However, this market is still expanding. Growing incomes, a flourishing middle class and peaking working age populations has all translated into this rapid rise in the tendency to travel.

With the introduction of its low cost carriers in the beginning of 2001, AirAsia started as Malaysia’s second national carrier and had a huge contribution to this peak in Asia’s aviation market. The company started right away as a huge competition for the already existing airlines but most of all for first national carrier Malaysia Airlines. A competitor who had adapted the same strategy that was already developed by Southwest Airlines in the United States and used with great success by European companies, such as Ryan Air and Easy Jet (Ahmad, & Neal, 2006). The liberalization of the market done by the Association of South-East Asian Nations (ASEAN), the characteristics of Asia -like the large demographic area and the geography of South East Asia, with a big part of the continent divided by water- and the growing number of business travelers, created a nice business climate for low-cost carriers like AirAsia (Hanaoka et al. (2014). Ahmad and Neal (2006) explain that AirAsia had to compete against bus and train with its domestic and short flights, but also got more and more competition in the countries where it spread to. The Asian aviation market has become an incredibly competitive market in a really short period of time. However, with a domestic market share of 47% in Malaysia, the role of AirAsia for the economic growth may be significant (AirAsia Annual Report, 2016). This makes it quite interesting to observe what all of these developments did to the country where the initiator of this huge competition, AirAsia, initially operated from: Malaysia. This leads us to the question this thesis will answer:

What is the relationship between the Malaysian aviation industry and the economic development of the country? A case study based on AirAsia.

The question will be tested through four hypotheses, because of the unknown direction of the relationship between the aviation market and the economic growth of the country (Hakim, & Merkert, 2016). It is on the one hand possible that the economic growth stimulates the tendency to travel and therefore leads to more civil air transport, but on the other hand it could also be the other way around. Both of these directions will be tested by using the Granger causality model, together with the option of a two-way causality and no interdependency at all.

(5)

2. Literature review

It is well ratified in previous research that there exists a strong correlation between air traffic and economic growth, however, the direction of causality is not always clear (Green, 2007) and only a few causal analyses have been done on the transportation industry so far. Button et al (1999) observed the causality between airport traffic and employment using the Granger Causality test, and detected a significant effect of air traffic on the employment rates. Marazzo et al. (2010) studied the relationship between air passengers and demand and economic growth (GDP) in Brazil and found that GDP and air passenger growth are co-integrated. But closer to the context of Asia and by using the Augmented Dickey Fuller (ADF), Johansen co-integration tests, the Granger causality test and the Vector Error Correction Model (VECM), Chang and Chang (2009) found that there is a bi-directional causal relationship between Air cargo and economic growth in Taiwan. And finally, Hakim and Merkert (2016) have analyzed the direction of the relationship between the air passengers and the economic growth for whole South-East Asia based on panel data. In this analysis they found that there is only a uni-directional Granger causality running from economic growth (GDP) to air passenger traffic.

In the majority of previous research, the focus is on high income countries with often matured aviation market, and results in a bi-directional causality between air transport and economic growth. Middle income countries show, in contrast to these high income countries, strong causality from economic growth to air transport. There has not been any research on previous low income countries with high economic growth rates and also a strong growth of air transport demand. Therefore, in this paper the focus will be on Malaysia, a country which has gone through both huge economic growth rates and a rapid growing aviation industry. It will be a complementary research on the one of Hakim and Merkert (2016) where the same causal relationship will be tested applied on the situation of Malaysia. To understand the way in which the economic growth and the aviation industry can influence each other, there will be some background information in this chapter.

This section will start with the introduction of the low-cost carriers as a phenomena in other demographic areas and afterwards its arrival in Asia. Then the development of Malaysia will be explained subsequent with the way in which the aviation industry in Malaysia could have influenced the economic growth of the country. Then the rise of AirAsia will be discussed. Here the market conditions that created space for AirAsia to grow will be sketched together with AirAsia’s strategy to grow into this market. And finally, there will be some conclusions on previous research on this subject.

2.1 History of LCC’s

Cost minimization is the core business principle that drive the business of companies using the low-cost carriers. This business principle does not imply that the products of a low-cost airline are always the cheapest in the market or that it offers products with low quality. Low-cost as a business principle simply emphasized the need to keep operating costs low (Ahmad, & Neal, 2006).

(6)

The phenomena of low-cost carriers found its origin in the United States, where the entering of LCC’s all started with the Airline Deregulation Act of 1978. Before this act, the aviation market of the United States was mainly a duopoly and a oligopoly and there was no space for competition. Doganis (2007) explains that it was a case of a high level of exclusion of new airlines, the protection of relatively inefficient carriers, high costs for service and labour and a lack of price competition. In the years after this act a lot of new airlines entered the market and prices fell substantially on most routes, particularly on routes with long distances (Anderson, Bon, & Lakshmanan, 2005). A competitive market was created and Southwest Airlines acted most of all on this new situation. This company became America’s biggest provider of air transportation, and was seen as the principal driving force behind the large fundamental changes that occurred in the U.S. aviation market (Bennet, & Craun, 1993).

2.2 LCC’s in Asia

The Southwest effect flew over to Europe where companies like Ryan Air started with building the market for LCC’s, and finally also spread to Asia. Before 2002, there were no significant low-cost carriers operating in the Asia Pacific area. This delayed development in Asia was partly due to the perception that the low cost model adopted in the United States and Europe could not be replicated in Asia, because of the longer aircraft stage lengths, lack of secondary smaller airports and the regulatory restrictions preventing access to international airports (O’Connell, & Williams, 2005). Since the beginning of the year 2000 it proved possible due to several important reasons. First, the outbreak of the Asian financial crisis in 1997 created a demand for low-cost air travel for business travelers funded by cost-conscious finance departments of private firms (Condom, 2005). The huge decline in air travel caused by the Asian Financial Crisis also placed pressure on the governments in some Asian countries to open up both domestic and international aviation market for independent low-cost start-ups (Shuk-Ching Poon, & Waring, 2010). Second, restrictions on reciprocal air services agreements were recently removed by many national governments in Asia to boost further growth of trade and tourism in the region. The Association of South-East Asian Nations (ASEAN) has been working towards developing an ‘open skies policy’ targeted to give national carriers of the Association’s member countries unrestricted access between capital cities of the ASEAN pact since 2008. Deregulation of aviation rules has enabled many LCC’s to offer multi-country flying services between the areas (Baker, Field, & Ionides, 2005). Third, low-cost terminals such as those opened in Kuala Lumpur International Airport in 2005 and Singapore’s Changi Airport in 2006, have supported further development in LCC’s in the region. These airports are designed for the passengers to board and get off the plane rapidly and are not build for giving them more service then their primary needs. The LCC’s have therefore been able to keep the costs low and pass the savings onto the airlines using their services (Shuk-Ching Poon, & Waring, 2010).

Thereby, the growth of LCC’s is also made possible because of the huge population in ASEAN, China and India, which provided the needed demographic base to fuel further development of LCC’S in the region. This is even more empowered by an increase in the number of Asian impulse travelers who would like to fly to nearby holiday destinations. Additionally, Ahmad

(7)

and Neal (2006) argue that an increase in urbanization caused higher demand by growth of the cities and the population.

2.3 Development of Malaysia

In the last twenty years, the Malaysian economy has been transformed from a protected low income supplier of raw materials, to a middle income emerging multi-sector market economy (Yusof, Bhattasali, 2008). This economy is mostly driven by manufactured exports, electronics and semiconductors in particular, which constitutes around the 90% of the exports. With the New Economic Policy launched in 1970 the government’s commitment to the free market has been hedged by policies aimed at providing “constructive protection” for Islamic Malays against economic competition from other ethnic groups and foreign investors. During the Asian financial crisis in 1997, most of the major companies that the government had privatized, including Malaysian Airlines, had to be renationalized to prevent their collapse. A powerful recovery program mounted by the government that was showing positive results ran abruptly into the wall of the global slowdown in 2001. Yusof and Bhattasali (2008) showed in their report that in this year the foreign direct investment dropped almost 50% worldwide, the decline in Malaysia was even bigger with 85%. GDP dropped during this period to 0.7%, from its usual 7% to 9%. The government generally remains committed to a policy of free enterprise, despite the fact that it owns and operates the railway and the majority of the communications systems and has become increasingly involved in certain key industries.

2.4 The influence of the aviation industry on Malaysia

A report of Oxford Economics (2011) has taken care of giving the following explanations and numbers for the year 2009 and the years before. It starts with explaining that the air transport to, from and within Malaysia have contributed to three types of economic benefit. The focus is here most of all on the ‘economic footprint’ of the industry, which measures its contribution to GDP, jobs and tax revenues generated by the industry and its supply chain. However, the economic value created by the aviation sector is more than that. The main benefits are created for the customer who is using the air transport service. Additionally, the connections created between cities and markets serve as an important infrastructure. This generates benefits through enabling foreign direct investment, business clusters, specialization and other spill-over impacts on an economy’s productive capacity (Oxford Economics, 2011).

These improvements in connectivity came together with a steady fall in the cost of air transport services by around 1% a year over the past 40 years. Hence, it had contributed to the rapid expansion in the volume of trade seen over this period. A report by Oxford Economics (2011, p.10) explains that: “air transport has steadily become more competitive relative to other modes of transport. For example, it is estimated that its relative cost has been falling by around 2.5% a year since the 1990s. As its relative cost has fallen, air shipments have become increasingly important for international trade”.

In numbers, the aviation market has made a significant difference for the Malaysian economy. In the year 2009, calculated was that this market has contributed MYR 7.3 billion (1.1%)

(8)

to the Malaysian GDP of that year. Besides, the aviation sector have supported 102,000 jobs in Malaysia, which were either direct, or indirect through the supply chain or tourism effects. Finally, the aviation industry has also contributed to the public finance, in numbers of MYR 564 million in tax revenues, including income tax receipts from employees, social security contributions and corporation tax levied on profits (Oxford Economics, 2011).

2.5 The rise of AirAsia

With the created market opportunities in Malaysia and its economic development, there was space for AirAsia to rise as a low-cost carrier in this demographic area. In this chapter, the rise of this company will be explained together with the strategy it choose to become the company they are today.

AirAsia was incorporated in 1993 and became Malaysia’s second national carrier and commenced full-service domestic operations with two Boeing 737-300s. It initially flew from Kuala Lumpur to four destinations whereof three in the east part of the country and one in the west. In 2001, Tune Air acquired AirAsia for 1 ringgit (0.26 USD) and assumed its debt of 40 million ringgit (10.5 million USD), because of its malfunction in the years before. The management team, headed by Tony Fernandes as CEO, transformed AirAsia into a successful low-cost carrier and expanded within two years to fifteen aircrafts. By June 30th, AirAsia achieved a net profit of RM19.1 million (USD 5.0 million) despite of its limited operational and financial resources. Since the Arrival of Fernandes and his team, AirAsia had built a brand and became a household name, in not only Malaysia, but also in Singapore, Thailand and Indonesia (Ahmad, & Neal, 2006).

AirAsia targeted destinations within three-and-a-half hour flight time from its hubs, which already covered the whole of South-East Asia. Estimated was that this would enable approximately 500 million people to travel through these destinations (Ahmand, & Neal, 2006). The reason for these short flight-time air routes was that this would give AirAsia the ability to optimize the utilization of its aircrafts and other ground support assets. With this, AirAsia optimized both the frequency and turnaround time (time between arrival and departure). Ahmad and Neal (2006) summed up the six business strengths which underlie AirAsia’s succes:

• Single-class, No frill Service: There is only one class in all AirAsia flights, and the service does not provide free inflight meals, in-flight entertainment, airport lounges or other amenities.

• High Aircraft Utilization and Efficient Operations: AirAsia maximized the utilization of the Boeing 737-300 by adding extra seats, operating on a longer working day and it maintained a low turnaround time.

• Low Fixed Costs: AirAsia negotiated and obtained lower lease charges for its aircrafts, lower rates for longterm maintenance contracts, lower rates on its insurance fees and lower airport fees, because of its high safety and maintenance standards.

• Low production/distribution costs: AirAsia did not issue tickets, and that helped save administrative costs and related expenses. People got a discount on buying tickets on the internet instead of buying them through a call centre or by text.

• Using only one single Aircraft Type: This reduced the amounts needed for spare parts inventory requirements and helped the company to increase cost savings.

(9)

• Maximizing the benefits of Regional Media Coverage: AirAsia’s success in South-east Asia attracted publicity and the groups used these opportunities to promote and increase its brand awareness without additional promotional costs.

The rise of AirAsia is also made possible by the malfunction of Malaysia’s first national carrier Malaysian Airline System (MAS). MAS started with a vision to become “An Airline of Excellence”, offering the very best to its passengers in terms of safety, comfort, service and punctuality (Zaid, 1994). However, in the period between 1998 and 2004, the airlines unit cost differential was very significant and was due to its excessive labour force, its poor productivity, low aircraft utilization, unprofitable domestic routes and the limitations of intra-Asian bilateral (O’Connel, & Williams, 2005). AirAsia responded to this with its totally different strategy and won in this way the trust of the price-sensitive Asian citizens.

3. Method

The impact of the aviation industry in Malaysia on the economic growth will be tested using annual data over the period from 1970 until 2016. This will be done annually because the data for aviation industry is only reported once a year. It will start in the beginning of 1970 because this is the year in which the Malaysian airlines started reporting their financial reports for investors. Here, the situation before the arrival of AirAsia can be observed with the existing Airline in the market, and then the change in the period after. The period ends with AirAsia being in a stable place in the Malaysian market. The focus will be on the total Malaysian aviation industry because there is probably no significant relationship between the whole economy and just one company. With this data, conclusions can be made about the influence that AirAsia had in this effect.

The variable that will be used to indicate the impact of the Malaysian aviation industry will be the amount of passengers that are carried by the company during this period. This variable measures the amount of passengers that have been on domestic or international flights in and out of the country by the national airlines. If AirAsia and the whole Malaysian aviation industry show the same trend line, it can carefully be assumed that the statistical results will hold for both the company as the whole country. To measure economic growth, one of the variables which are mentioned before as indicators of the ‘economic footprint’ will be used, namely the contribution to GDP. Based on previous research done by Merkert and Hakim (2016), the hypothesis will be that there is a uni-directional relationship from GDP to the amount of passengers carried.

In this paper, a series of econometric models are used to test the stationarity of the variables and the co-integration and causality between civil aviation and economic growth in Malaysia. Here, the causal relationship can be uni-directional, bi-directional or there is a possibility of no interdependency between the variables, which is the reason that the simple OLS regression is not suitable in this case. Therefore, the Granger Causality Framework will be employed just like Hakim and Merkert (2016) did. For both the amount of passengers and the change in GDP, a natural logarithm will be used to ease the interpretation of coefficients.

To do this analysis without getting into misleading conclusions, this paper will follow a three-step Granger causality test which is adopted from Hakim and Merkert (2016). First, the unit root

(10)

test is applied, which can confirm that the future and the present are caused by the past and not in a different order (Granger,1969). If this is not the case then the data is non-stationary and the results can be spurious. The second step contains the test for co-integration, where co-integration will be examined based on the fact that the variables are integrated of order at most 1. This co-integration is required for doing the Granger Causality test. Finally the Granger Causality Test based on the Vector Error Correction Model (VECM) will be examined to test the direction of the relationship and to test whether this relationship is in the long-run or short-run.

3.1 Unit root test

So the first step is to make sure that the variables are showing stationarity. It is imported that the variables are stationary because otherwise it means that their means and variances alter due to change of time (Heij et al., 2004). You can test this stationarity based on the normal Dickey-Fuller Test, the Augmented Dickey-Fuller (ADF) Test and the Phillips-Perron Test. The Augmented Dickey-Fuller test includes, in contrast to the normal Dickey Fuller, also the lagged values of the dependent variable. The ADF test can therefore handle more complex models and is also more powerful. The Philip Perron test can be viewed as Dickey-Fuller statistics that have been made robust to serial correlation by using Newey-West (1987) correction for serial correlation. In addition, a series denoted by I(d) indicates that the series is integrated of order d and means that a non-stationary series has to be differenced d times to be non-stationary. The autoregressive specification used in this paper can be expressed as:

∆GDPt=αGDPt-1+PCtδ+ ∈t (1)

Where t=1,2,…,N represents time periods, GDPt represents the economic growth for Malaysia,

including any fixed effect of individual trends; PCt represents the amount of passengers carrier by

Malaysian airlines, α refers to the autoregressive coefficients and ∈t represents the error terms

which are assumed to be mutually independent. The null hypothesis in this test is that the variables contains a unit root, whereas the alternate means that the variable is generated by a stationary process.

3.2 Co-integration test

The series of economic growth and air transport data are thus co-integrated when all of the series are integrated with the same order. The test in this paper is based on Oh (2005), who examined the co-integration by using the Engle-Granger two-stage approach (Engle & Granger, 1987). According to Granger (1981), co-integration means that the non-stationary variables are integrated in the same order with the residuals being stationary. If there is co-integration between two variables, then there is a long-run effect that prevents the two variables from drifting away from each other and there exists a force to converge into long-run equilibrium. The Engle-Granger two-stage method is performed by two equations:

(11)

∆PCt=ß0+ß1GDPt+ ∈t or (2)

∆GDPt=δ0+δ1PCt+ ∈t (3)

Where t=1,2,…,N refers to the time period. The estimated residuals represent deviations from the long-run relationship and are denoted by ∈t. Here, the null hypothesis is that there is no

co-integration in which case the residuals ∈t are integrated with order one I(1) distribution. To check

whether ∈t follows an I(1) distribution or not, the following regression analysis is used:

∈

t

=∈

t-1

+µ

t

(4)

Besides, also the Johansen Integration Test will be employed, by using the VECM, to obtain more convincing results. This test will include two different likelihood ratio (LR) tests and defines a number of co-integrating equations given any normalization applied (Baker et al., 2015)

3.3 The Granger causality test

When the variables are stationary, the Granger causality test can be used to analyze the direction of the relationship between the amount of passengers carried (PCt) and the GDPt. This model is

based on the VECM model by which the long-run and short-run causality can be tested. The Causality Framework indicates that if there is a significant effect of lagged values of a variable X in a regression model to Y that not only depends on X, but also on its own lagged values Yt-1 of Yt-2

etc. then it can be said that X Granger causes Y and potentially Y all Granger causing changes in X (e.g., Gujarati, 2004). To test the Granger Causality, Y will be regressed on its own lagged values and on lagged values of X. Here, the null hypothesis is that the estimated coefficients on the lagged values of X are jointly zero, and x does thus not Granger cause Y. This is why the following formulas can be formulated:

∆lnPC

t

= α

t

+ ß

t

ECT

t-1

+γ

t

∆lnPC

t-1

+ δ

t

∆lnGDP

t-1

+ ∈

t

(5)

∆lnGDP

t

= α

t

+ ß

t

ECT

t-1

+ γ

t

∆lnGDP

t-1

+δ

t

∆lnPC

t-1

+ ∈

t

(6)

where ∆PCt denotes the first difference in PCt which captures their short-run disturbances within

t=1,2,…,T periods. ∈t is the white noise error term and ECT is the error correction term (ECT) that

is resultant from the long-run co-integration and evaluates the extent of the past disequilibrium. The coefficient of ECT determines the deviation of the dependent variables from the long-run equilibrium.

With the VECM you can test for both long-run and short-run causality. The long-run causality is stated by the coefficient of ECT and shows the speed to adjustment. The ECT is the one period lagged value of the error term and its significance suggests that the long-run equilibrium relationship is driving the dependent variable. The expected value of

ß

t should be significant and a negative number, whereas its absolute value indicates how quickly the equilibrium is restored.

(12)

The short-run causality is here tested by evaluating the combined significance of the coefficients of the independent variables of

γ

t and

δ

t. This is analyzed by using the Chi-squared Wald Test. If the estimated coefficients on lagged values of PC are statistically significant, then PC is Granger causing GDP in the short-run and otherwise is does not.

4. Data & Results

In this section the data and results will be showed for the different statistical tests. It will start with the demonstration of the changes in the economic growth of Malaysia and its aviation industry. First, in graph 1 the GDP growth rate is set against time, which shows some huge falls in growth in the years 1975, 1985, 1998 and 2009. During these falls the following events happened: the oil crisis, the global economic recession, the asian financial crisis and the world trade recession, respectively. As you can see, the GDP growth was except for these events relatively high during the period 1962 to 2017. Which indicates that Malaysia has gone through a huge economic development.

&

Source: Data Worldbank

&

Source: Data World bank

Graph 1: GDP growth of Malaysia

Gr owth % -9 -6 -3 0 3 6 9 12 1964 1971 1978 1985 1992 1999 2006 2013 Malaysia

Graph 2: Trend analysis between GDP and civil air transport in Malaysia Passengers carried 0 15.000.000 30.000.000 45.000.000 60.000.000 GDP in US dollars 0 100.000.000.000 200.000.000.000 300.000.000.000 400.000.000.000 19601965 1970 1975 1980 1985 1990 1995 2000 2005 20102015 GDP Malaysia

(13)

Second, the data for the economic growth of Malaysia and the Malaysian air transport are shown in graph 2. You can see that both slopes are moving upwards which means that both the GDP and the aviation industry have been following the same trend. For both the GDP and the Malaysian aviation industry, there are some falls in with the Asian Financial Crisis around 2000 and the global financial crisis around 2008.

Third, there will be a closer look on the Malaysian aviation industry, where also AirAsia and Malaysian Airline System are shown against the total Malaysian aviation industry. This data is regressed in graph 3 below. You can say that here, Malaysian Airline System moved along a quite horizontal trend line, whereas AirAsia moved in quite the same direction as the whole country Malaysia as they both show an upward trend.

&

Source: Data World bank, Annual Reports of AirAsia and Chart Nexus.

Because AirAsia follows the same upward trend as the aviation industry of the whole country it might be assumed that this company has a great influence on the total market. In addition, AirAsia had a marketshare of 47% percent during the last couple of years which confirms it even more. Therefore it might carefully be assumed that when the GDP of Malaysia and its domestic aviation market show a significant relationship, the same holds for AirAsia.

Then the results of the statistical tests will be showed. First, the descriptive statistics are shown in table 1. Here, both the normal Gross Domestic Product (GDP) and the Passengers Carried (PC), as the natural logarithm of these variables are used.

Graph 3: Amount of passengers carried

0 15.000.000 30.000.000 45.000.000 60.000.000 2000 2002 2004 2006 2008 2010 2012 2014 2016 Passengers carried Malaysia Passengers carried MAS Passengers carried AirAsia

(14)

For the unit root test, like said before, the Dickey Fuller Test, the Augmented Dickey Fuller Test and the Phillips Perron are used to test for stationarity. The results are shown below. If the null hypotheses are rejected, none of the variables follow a unit root process which means that they are stationary.

Note: * p<0.10, **p<0.05 and ***p<0.01, these represent significant p values. Here, for both tests, 2 lags are chosen. Proved is that for another amount of lags the same result will appear.

The results of this test conclude that in the choice between the normal variables and its natural logarithms, the natural logarithm would be the right pick. The significance is for both variables higher if you take their delta, which means that they show more stationarity. Therefore, for both the co-integration test and the Granger Causality Test the first differences should be used.

The Co-integration test will be, like said, done based on the two-step residual-based method, where first the Engle-Granger Test is used and then the Johansen Test. The results of the Engle-Granger test are shown in appendix (B). Here, as you can see, the residuals follow a random walk which means that the correlation of the residuals is minimized by using this test. With the confirmation of these test the ADF test can be done on the residuals. The right amount of lags for this test is chosen based on Akaike’s Information Criterion (Akaike, 2011) and is shown in appendix C. Based on this test the right amount of lags is one. In table 3 you can see the results from this test on unit root of the residual. In this test, no co-integration is the null hypothesis, in which case the residuals ∈t are integrated with order one (1) distribution.

Table 1: Descriptive statistics

Variable Obs Mean St. Dev. Min Max

ln_GDPMal 47 24.71799 1.246308 22.07501 26.5465

ln_PCMal 47 16.0801 1.086474 13.52636 17.80111

GDPMal 47 1.00e+11 1.02e+11 3.86e+09 3.38e+11

PCMal 47 1.54e+07 1.42e+07 748900 5.38e+07

Table 2: Unit root test by DF, ADF and Phillip Perron

Dickey Fuller test Augmented

Dickey-Fuller Test Phillip Perron ln_GDPMal -2,475 -1.393 -2.481 ln_PCMal -3.358* -4.479*** -3.372* dln_GDPMal -16.052*** -3.613** -5.863*** dln_PCMal -5.470*** -4.569*** -5.434***

(15)

Note: Lag length is based on AIC ( Akaike’s information criterion) * p<0.10, **p<0.05 and ***p<0.01, these represent significant p values.

Based on the ADF test on residuals it can be concluded that the variables for GDP and Passengers Carried show co-integration. For the Johansen test that is shown in table 4, both null-hypotheses can be rejected that i) there is no co-integration at all and ii) there is only one or less co-integrating equation based on the fact that their trace statistic is higher than their critical value.

Note: *p<0.05, this represent a significant p value

Then the Vector Error Correction Model (VECM) is applied, this makes it possible to either test for short-run or the long-run causality between the variables. The results of these tests are shown in table 5. From these results, it can be concluded that there is no long-run causality between the variables GDP and passengers carried and only Passenger Carried influences GDP in the short run.

* p<0.10, **p<0.05 and ***p<0.01, these represent significant p values.

Finally, the results for the Granger Causality Test are also shown in table 5. To test this causality, there will be two null hypotheses to reject, which are i) GDPt does not Granger cause PCt and ii)

PCt does not Granger cause GDPt. Based on the value of the F-statistic and its corresponding

p-values, the null hypothesis can in both cases not be rejected. This means that there is no significant relationship between PCt and GDPt. This will hold too for the influence of AirAsia on

GDP and the other way around. It might be interesting to test wether this is the same for the period before AirAsia’s arrival and the period after. The results are shown in Appendix F, from which it can be concluded that the same results hold for the period separately.

Table 3: Engle-Granger Test on Residuals

Residual ADF

F-statistic -4.938***

Table 4: Johansen test for co-integration

rank Trace statistic 5% critical value

0 44.0586* 15.41

1 20.2242* 3.76

2 -

-Table 5: Granger Causality Test and the test on long- and short-run causality

Granger Causality Test (F-Statistic)

Long-run Causality (ECT) Short-run Causality (Chi square statistic)

GDP to PC .18092 -3.48 6.745394

(16)

5. Conclusion

With the huge impact of low-cost carriers in the aviation industry of today, the consequences for the population has been enormous. People are enabled to fly wherever they want for a price that has only declined during the past 20 years. After the introduction of the low-cost carriers in the United States and Europe it eventually also flew over to Asia in the beginning of the twenty first century with the arrival of AirAsia. The effect of these low-cost carriers has, most of all, been really serious in South East Asia, where it has made a significant difference for its inhabitants. The aviation industry could have influenced the economic development through an improved infrastructure, more job opportunities and more income from taxes. In this paper the relationship of this economic development and increasing aviation industry are examined, where the focus is on the direction of the relationship between the amount of passengers carried and the GDP. The analysis is based on the country where the initiator of the low-cost carriers in Asia is coming from: Malaysia. The analysis of this paper will give an answer to the question about the direction of the relationship between the aviation industry of Malaysia and its economic development.

To test the direction of this relationship the unit root test, the co-integration test and eventually the Granger Causality Test are employed. The results from the unit root test show that the first differences should be chosen, based on their stationarity. Therefore, with the next tests the first differences will be used. The results of the co-integration test show that the variables were integrated with the same order, which confirms the co-integration between the two variables GDP and passengers carried. This co-integration is also required to move on with the causality tests. Finally the long-run and short-run causality between the variables is tested. This resulted in a rejection for no existence of a short-run causality from the amount of passengers carried to the GDP. This means that the aviation industry of Malaysia only influences the economic growth in the short run. However, there has not been a significant long-run relationship between the two. The last test, which was the Granger Causality test, has resulted in the conclusion of no interdependence between the GDP and the amount of passengers carried. The tests shows, in contrast to previous research, no significance. Therefore, the same holds for the company of AirAsia which has said to follow the same trend line as the whole aviation industry of Malaysia. The conclusion will be that the hypothesis of causality between the aviation industry and the GDP will be rejected. There is only a short-run causality on economic growth but this effect will not last. This results in a conflicting opinion on how to react on these findings. From previous research, policy makers are advised to always take into account the aviation market of a certain area. This papers concludes that there is no significant effect between the two variables, which means that the effect of the aviation industry is not everywhere the same. Therefore, tourist-attracting policies and improvements in the infrastructure by air may not be fully effective in that economic expansion leads to growth in the aviation sector or the other way around. However, because this paper is only examined on one country, there can not be any forecasting conclusions for others.

(17)

6. Discussion

The results of this paper together with previous research rises a lot of questions. With the hypothesis of a uni-directional causality, the findings are slightly odd. Thereby, in the literature and in previous reports on the Malaysian economy the aviation industry also seemed to have influence on the economic growth. In the end there are also some limitations on this research and there will be some recommendations on further research.

Since it is well known that air transport and economic growth are closely tied, it is rational to believe that economic growth is strongly affected by air transport. However, this is not conclusion of this paper. These results point to several research directions in the future. First, a simple bivariate VAR model was used in this study. Important variables such as exchange rates which play a critical role in model specifications might not be completely considered. This can be improved by employing a multivariate approach of multivariate co-integration (Johansen 1998), including important variables such as income, exchange rates and international trade. And second, for further research it might be helpful to analyze other countries with the same characteristics and confirm that it shows the same interdependence between the two variables. Malaysia may be an outlier, and can therefore show different results than the overall aviation industry. Another limitation of this analysis is the fact that aviation patters could change in the future and may grow slower as the development level decreases.

(18)

7. References

Ahmad, R., & Neal, M. (2006). AirAsia: The Sky's the Limit. Asian Journal of Management Cases, 3(1), 25-50.

Akaike, H. (2011). Akaike’s information criterion. In International encyclopedia of statistical science (pp. 25-25). Springer Berlin Heidelberg.

AirAsia Annual Report. (2016) Retrieved from: https://www.airasia.com/cdn/docs/common-docs/ investor- relations/airasia-annual-report-2016.pdf

Atkinson, S., & Palumbo, D. (2018, February 6). Singapore Airshow: Asia aviation in five charts. BBC news. Retrieved from: http://www.bbc.com/news/business-42855503

Baker, C., Field, D., and Ionides, N. (2005), ‘Global Reach’, Airline Business, May 21, 5, 60–65.

Baker, D., Merkert, R., & Kamruzzaman, M. (2015). Regional aviation and economic growth: cointegration and causality analysis in Australia. Journal of Transport Geography, 43, 140-150.

Bennet, R., & Craun, J. (1993). The airline deregulation evolution continues: The Southwest effect. Washington, DC: Office of Aviation Analysis, U.S. Department of Transportation.

Button, K., Lall, S., Stough, R., & Trice, M. (1999). High-technology employment and hub airports. Journal of Air Transport Management, 5(1), 53-59.

Condom, P. (2005), ‘Low-profit Carriers’, Interavia Business & Technology, 676.

Chang, Y. H., & Chang, Y. W. (2009). Air cargo expansion and economic growth: Finding the empirical link. Journal of Air Transport Management, 15(5), 264-265.

Dai, L., Derudder, B., & Liu, X. (2018). The evolving structure of the Southeast Asian air transport network through the lens of complex networks, 1979–2012. Journal of Transport

Geography, 68, 67-77.

Doganis, R. (2013). Flying off course: The economics of international airlines. Routledge.

Economic Benefits from Air Transport in Malaysia. (2011) Oxford Economics. Retrieved from https://www.iata.org/policy/Documents/Benefits-of-Aviation-Malaysia-2011.pdf

Engle, R. F., & Granger, C. W. (1987). Co-integration and error correction: representation,

estimation, and testing. Econometrica: journal of the Econometric Society, 251-276.

Forecasting, O. E. (2006). The economic contribution of the aviation industry in the UK. London,

UK.

Gujarati, D. N. (2004). Basic econometrics.: Student solutions manual for use with Basic econometrics.

Granger, C. W. (1969). Investigating causal relations by econometric models and cross-spectral methods. Econometrica: Journal of the Econometric Society, 424-438.

Granger, C. W. (1981). Some properties of time series data and their use in econometric model specification. Journal of econometrics, 16(1), 121-130.

Green, R. K. (2007). Airports and economic development. Real estate economics, 35(1), 91-112.

Hakim, M. M., & Merkert, R. (2016). The causal relationship between air transport and economic growth: Empirical evidence from South Asia. Journal of Transport Geography, 56, 120-127.

(19)

Hanaoka, S., Takebayashi, M., Ishikura, T., & Saraswati, B. (2014). Low-cost carriers versus full service carriers in ASEAN: The impact of liberalization policy on competition. Journal of Air

Transport Management, 40, 96-105.

Heij, C., De Boer, P., Franses, P. H., Kloek, T., & Van Dijk, H. K. (2004). Econometric methods with applications in business and economics. OUP Oxford.

Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of economic dynamics and control, 12(2-3), 231-254.

Marazzo, M., Scherre, R., & Fernandes, E. (2010). Air transport demand and economic growth in Brazil: A time series analysis. Transportation Research Part E: Logistics and Transportation Review, 46(2), 261-269.

Newey, W. K., & West, K. D. (1987). Hypothesis testing with efficient method of moments estimation. International Economic Review, 777-787.

O’Connell, J. F., & Williams, G. (2005). Passengers’ perceptions of low cost airlines and full service carriers: A case study involving Ryanair, Aer Lingus, Air Asia and Malaysia Airlines. Journal

of Air Transport Management, 11(4), 259-272.

Oh, C. O. (2005). The contribution of tourism development to economic growth in the Korean economy. Tourism management, 26(1), 39-44.

Phillips, P. C., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335-346.

Shuk-Ching Poon, T., & Waring, P. (2010). The lowest of low-cost carriers: the case of AirAsia. The International Journal of Human Resource Management, 21(2), 197-213.

Yusof, Z. A., & Bhattasali, D. (2008). Economic growth and development in Malaysia: policy making

and leadership. Washington, DC: International Bank for Reconstruction and Development/The

World Bank.

Zaid, A.M. (1994) "Malaysia Airlines’ Corporate Vision and Service Quality Strategy", Managing Service Quality: An International Journal, Vol. 4 Issue: 6, pp.11-15 

(20)

7. Appendix

Appendix B: Engle-Granger Test on residuals

Note: Var25 is here the variable Years

Appendix A: Passengers carried by the examined companies and the whole country. Data from 2000-2016

Year Malaysia (country) Malaysian Airline System AirAsia

2000 16.560.793 - -2001 16.107.156 16.745.000 -2002 16.281.275 15.734.000 -2003 16.704.600 16.325.000 1.481.097 2004 19.226.564 15.375.000 2,838,822 2005 20.369.086 17.536.000 4,414,069 2006 17.833.364 15.466.000 5,719,411 2007 21.325.754 14.213.000 8,737,939 2008 22.420.870 13.760.000 11,808,058 2009 23.766.316 13.870.000 14,253,244 2010 34.239.014 15.708.000 16,054,738 2011 38.218.609 17.046.000 17,986,558 2012 39.165.195 16.651.000 19,678,576 2013 47.995.842 20.733.000 21,853,036 2014 49.673.884 - 22,138,796 2015 50.345.821 - 24,254,506 2016 53.817.353 - 26,410,922

(21)

Appendix C: Chosen amount of lags for Johansen Test

(22)

Appendix E: The Vector Error-Correction Model

Appendix F: Granger Causality Framework for diﬀerent time periods Granger Causality for the whole period 1970-2016

Equation Excluded F df df_r Prob > F

dln_GDP ln_PC .21916 2 39 0.8042

dln_GDP All .21916 2 39 0.8042

ln_PC ln_GDP .18092 2 39 0.8352

ln_PC All .18092 2 39 0.8352

Granger causality Wald test for the period 1970-1995

dln_GDP ln_PC .00813 2 17 0.9919

dln_GDP All .00813 2 17 0.9919

ln_PC ln_GDP .14474 2 17 0.8663

(23)

Granger causality Wald test for the period 1995-2016

dln_GDP ln_PC .38998 2 17 0.6830

dln_GDP All .38998 2 17 0.6830

ln_PC ln_GDP .54525 2 17 0.5895

What is the relationship between the Malaysian aviation industry and the economic development of the country? : a case study based on AirAsia