1.1 Introduction
The tourism industry is one of the leading service industries and is considered a pivotal source of economic growth in the world economy. A developed and established tourism industry catalyses national and regional development, helps establish the foreign exchange rate, creates more employment opportunities, and contributes to social development that will benefit the local community and tourists (Sharif & Lonik,2014).
Therefore, in this introduction chapter, the researcher will succinctly explain the background context regarding the tourism industry in Malaysia and the current economy status. Then, several key research issues will be identified from the past journals as the formation of research gaps used in this study. Through the determination of research gaps, the research questions and objectives are then formulated to cope with the addressed research gaps. Furthermore, the contribution of how this research contributes to the society is presented in the section of the significance of the study.
1.2 Background of Study
Global tourism has experienced continued intense growth and diversification to become one of the fastest growing economic sectors in the world. It brings hope, prosperity, jobs, and well-being to so many lives all over the world. Earlier this year, the United Nations World Tourism Organisation reported the number of tourist arrivals globally totalled up to a staggering 1.4 billion, achieved two years ahead of its forecast. Figure 1.1 below demonstrates for the international tourist’s arrivals by the global region. It can be observed that the number of international tourist arrivals is increasing exponentially for each region, such as Africa, Middle East, Asia & Pacific, Americas and Europe, which begin as 0 in 1950 until 400 million tourists in 1990 and boosted up to 1.4 billion in the latest 2018. This directly indicates that the tourism sector is progressively gaining momentum in the international platform, which leads to higher vitality of roles being played by the tourism industry.
Figure 0.1 The number of tourists around the global globe
Source: (Roser, 2017)
Trends have also shown that Asia and the Pacific region experienced the highest increase in international arrivals of around 6.5%, compared to other regions (Mahathir, 2019). According to World Travel and Tourism Council (WTTC, 2019), the total contribution of Global Travel & Tourism to its gross domestic product (GDP) which including wider effects from investment, the supply chain and induced income impacts was USD 8.811 billion in 2018(10.4% of GDP) and is expected to grow by 3.6% to USD 9,126.7 billion (10.4% of GDP) in 2019. Other than that, the travel and tourism sector generated economy employment that comprised of around 3.2 million people (10.0% of total employment) in 2016 and approximately 3.5 million of employment in Malaysia’s tourism industry in 2018. This drives the tourism sector in Malaysia is the second largest contributor to foreign exchange earnings to the country after the manufacturing sector (WTTC, 2019). The figure below depicts the rising employment rate in the tourism industry of Malaysia from 2011 to 2018.
Figure 0.2 Employment rate in Malaysia Tourism industry (2011 – 2018)
Source: (Hirschmann, 2019)
The tourism industry is known as the travel industry and it is one of the fastest growing industries in the world. It is linked to the people travel to other locations which outside their usual environment for leisure, casual business or social purposes. Today, the tourism industry in Malaysia is developing and has become the world’s most attractive travel destinations. The tourism industry is vibrant and competitive, it requires the ability to constantly adapt to customer’s changing needs and desires, customer satisfaction, happiness safety and enjoyment, and this is the most particular for tourism enterprises. The various activities have contributed significantly to the country’s economy. Wherein, tourism is the second largest source income of foreign exchange (WTTC, 2019). Nevertheless, the tourism industry has a positive relationship on economy growth. The industry plays a vital role in attracting investment, creating new jobs and increasing foreign exchange.
Traditionally, economic growth (development) has been linked to growth in the agricultural and manufacturing sectors as well as the influx of foreign capital. Whilst, the role of tourism in economic growth has often been downplayed and regarded as a non-growth oriented sector, hence attracting little attention of both economists and policymakers (Papatheodorou, 1999). Today, tourism has become one of the rapidly growing services sectors of the world. This has prompted the Malaysian government to set tourism as a key sector to propel Malaysia’s long-term economic growth. Specifically, the 10th Malaysia Plan (2011-2015) has identified the tourism sector as one of the National Key Economic Areas (NKEAs) for transforming Malaysia into a high income nation by 2020.
In line with the above objectives Ministry Tourism, Arts and culture (MOTAC) has clearly outlined to promote Malaysia various activities as an outstanding tourist destination. They aim to showcase Malaysia’s unique wonders, attractions and cultures, develop domestic tourism and enhance Malaysia’s share of the market for meetings, incentives, conventions and exhibitions (MICE). Their ultimate goal is to increase the number of foreign tourists to Malaysia, extend the average length of their stay and increase Malaysia’s tourism revenue (Tourism Malaysia Official Website, 2010).
The Malaysian tourism industry is the third largest contributor to economy in 2017 (Tourism sector to remain third largest contributor to economy, 2017). This is one of the industries the government emphasizes in the Economic Transformation Programme (ETP) to strengthen and diversify economic risk. According to World Travel and Tourism Council (WTTC, 2019), the total contribution of Malaysian Travel & tourism to its gross domestic product (GDP) was 13.3% in 2018. The total contribution of Malaysian Travel & tourism has grown by 2.6% of GDP in 2017 (Tourism Malaysia, 2019). Malaysian tourism is the third largest foreign income earner after manufacturing and palm oil industry in Malaysia which accounting for over 7.0% over the country economy in 2018 (Misachi, 2017). Other than that, the travel and tourism sector generated RM 190.3 billion in economy GDP (equivalent to 13.3 of total GDP), export revenue, international visitor representing RM 79.4 billion (equivalent to 7.9% of total exports), economy employment representing 1,766,700 jobs (11.9% of total employment) and it making tourism in Malaysia is the second largest contributor of foreign exchange earnings to the country after the manufacturing sector (WTTC, 2019).
Malaysia received 13.35 million international tourists and recorded a 6.8% growth in tourist receipts, contributing RM 41.69 billion to the country’s revenue during the first half of 2019. The first half of 2018 saw 12,730,368 visitors who brought in contributions totalling RM39 billion. Over the first six months of this year, tourism performance grew in terms of per capita expenditure, rising by 1.9% to RM3,121.6 while the average length of stay climbed by 0.4 nights to 6.2 nights. Asean arrivals continued to dominate the share of tourist arrivals to Malaysia with a 70% contribution (Bernama, 2019).
The first-quarter performance of 2019 showed growth in tourism revenue which reached RM21.4 billion, a surge of 16.9% compared to previous year. During 2019, it have 25.8 million tourists visited Malaysia and Malaysia secured 15th place among countries with the highest number of tourist arrivals. Furthermore, there are 78.2 million domestic tourists were recorded in 2018 with a growth of 10.9%, while domestic tourism expenditure has growth of 11.7% with RM60.4 billion (PM, 2019).
The total tourist receipts earned by the government in 2018 was RM84.1 billion and expected to grow until RM92.2 billion in 2019(Tourism Malaysia, 2019). Tourist receipts and capital investment in the tourism industry are important components of enhancing sustainable growth in the tourism industry. This is because tourist receipts represent tourists’ direct expenditures during their trips in the destination country. The tourism receipts constitute the revenue earned by the tourism industry and contribute significantly to the national income. The revenue earned by the government can be invested in the tourism industry or other industries to boost the country’s economy in order to achieve its target as a high income economy status in 2020. Hence, a variety of tourism packages has been launched and promoted to attract more local and international tourists.
The purpose of capital investment in the tourism industry is to further the business objectives and such investment is expected to be productive across time. The tourism industry can become more competitive with greater capital investment. This industry involves many supporting industries in the value chain, such as transportation, accommodation, food and beverage, telecommunication, and recreation, and thus it can have a series benefits for the local economy, including economic growth and employment opportunities. Tourist receipts and tourist arrivals are important for tourism industry growth, but the most important impact of the Malaysian tourism industry is on the sustainable development of the economy and society (Mohsen, 2015b).
1.3 Problem Statement
Tourism means to travel to a particular place for recreation and leisure (Erick, 2016). Tourism involves a temporary movement of people from one location to another for leisure and economic purposes which create a relationship between the people, places and the landscape (physical, social and culture). The World Tourism Organization defines tourism more generally, in terms which go “beyond the common perception of tourism as being limited to holiday activity only”, as people “traveling to and staying in places outside their usual environment for not more than one consecutive year for leisure and not less than 24 hours, business and other purposes” (World Tourism Organization, 1995).
At present, tourism has become a popular global leisure activity and a booming global industry. There are many countries like Greece, Thailand, Bahamas, Fiji, Maldives, etc. where economy largely depends on tourism. In such countries, if any disturbance occurs or anything that discourages the tourists from visiting these countries occur, the entire economy of the country gets affected. The undesired occurrences of such kind of interruption could result in huge issues that directly raise big concerns to the economic growth in the nation, especially Malaysia that plays its role as the top renowned country for that actively engaged in tourism.
However, Malaysia once had to deal with large challenge in two decades ago, in which the total number of tourists from all countries to Malaysia fell by 20.5% in 2003 to 10.57 million compared to 13.31 million the year of 2002, according to Tourism Malaysia figures. Throughout 2003, tourist arrivals from China also fell by 37.3% to 422,624 compared to 674,056 in 2002. Apart from a drop in arrivals, Malaysia’s tourism sector also suffered a drop in receipts. In 2003, the country’s total tourism receipts were RM21.29bil, a drop of 17.4% compared to the RM25.78bil in 2002 (The Star, 2020). In essence, there were several numbers of crucial factors identified by the previous scholars in pertaining to their dominant impacts towards the changes happen to the economic growth (GDP rate), namely tourism receipt, tourism arrival, real exchange rate and employment rate.
First and foremost, the tourism receipt plays a vital role in affecting the development of economy in each nation, as measured by Gross Domestic Product (GDP) rate. According to the study of Aleemi & Qureshi, (2015), the scholars has proven that the amount of consumer spending by the tourists whilst they are visiting the Pakistan country within 1981 to 2013 have largely dedicated to at least 24% GDP rate to the country economy. Moreover, study conducted to investigate the tourism industry of Tunisia also demonstrated that the revenues from the tourism countries had influenced their GDP rate and employment rate as a nature (Mahfoudh & Amar, 2016). In addition, another news published in “The Star Publication” in the latest 2019 also addressed that the tourism receipts recorded a growth rate of 6.8%, which the drives up to a substantial amount of around RM42 billion as the nation’s revenue to boost up Malaysia’s overall economy (Wei, 2019). Hence, this indirectly constitutes tourism receipt as an essential issue to be emphasized concerning to the economic growth, such as Malaysia in this study.
In addition, the tourism arrivals are also an important issue to address in tourism sector when to be associated with the economic growth rate. For instance, in 2018, France is regarded as the most visited country in the world witnessing 89.4 million tourists annually followed by Spain with 82.8 million annual tourists and the United States with 79.6 million annual tourists visiting the country (World Tourism Organization, 2019). In the meanwhile, they were successfully recording growth rate of 1.7% in 2018 and this lets France to ranks the seventh across the worldwide countries, in term of GDP rate (Countryeconomy, 2019). Apart from that, there were also 1.401 billion international tourist arrivals worldwide in 2018, with a growth of 5.4% as compared to 2017. Europe led the growth in absolute terms, having a 5.5% growth in international tourist arrivals in 2018 which reaching of 710 million. Following by it, Asia-Pacific have an increase of 7.3% over 2017 with a 347 million of international tourist arrivals and for Americas, there were 215.7 million international tourist arrivals which an increase of 2.3% in 2018 (World Tourism Organization, 2019). For example, Malaysia is also one of the countries enjoying the benefits brought by the tourism industry, in which the economic growth of Malaysia experiences a share of approximately 15% as compared to the previous years in 2018 (Bernama, 2018). As a result, this actually proves that the number of tourists arrivals can affect ongoing economic growth in Malaysia.
Developing countries like Malaysia are also gradually emerging as a global tourist destination. In Malaysia, tourism has established itself as a huge service industry in the country. As any disturbance occurs, tourism industry will be affected. Thus, the third issue to mention for tourism development would be the real exchange rate. The real exchange rate in Malaysia faces volatility over the past decades, due to occurrence of some unexpected disturbance issues that had led to negative influence to the development of tourism industry. A number of fatal incidents at Malaysia’s coastal lines in recent years have increased the security and safety concerns among tourists. This is mainly affected by east coast of Sabah which is the frequent kidnapping issues. Five cases have taken place between November 2013 and July 2014, which resulting in two deaths and six hostages, including tourists, hotel workers, polices, and also fish farm managers (The Star, 2014). Although there is no empirical relationship between the fluctuations of real exchange rate and the terrorism risk, however, the hotel industry in Sabah east coast area experienced a 35% drop in business. Beside than concerns for safety and security issues, effect on health issues will also hit the country’s tourism industry badly. In lately years, Malaysia facing a few unfortunate event that gave impact to Malaysia image as safe and secure destination to visit. From the several event, health issues become one of main concern. Started with out-break of dengue fever struck Penang in mid-1997, outbreak of Coxsackie B virus in Sabah and Sarawak, cholera epidemic, SARS and bird flu in this region give the big impact to tourism industry in Malaysia. This can be evidenced in which the real exchange rate of Ringgit Malaysia drops sharply in 2018 (as shown in figure 1.3), which then leads to declination of economic growth rate in Malaysia to record the lowest 4.4% GDP rate in 2018 (as shown in figure 1.4) over past decade (Ying, 2018).
Figure 0.3 Ringgit Malaysia Exchange rate from 2012 to 2020
Source: (Trading Economics, 2020c)
Figure 0.4 Expected GDP growth rate of Malaysia (2014 – 2024)
Source: (Plecher, 2019)
Besides that, the subsequent issue to be highlighted is the employment rate, that partly contributed by the tourism industry. According to Aynalem & Birhanu, (2016), the continuous development of tourism industry have indirectly opened up to thousand millions of career opportunities to the inhabitants in each nation. Tourism acts as the booster of career development and economic affluence, which occupy for at least 10% of the available jobs across global as well as occupied for at least 20% in 2017 (World Travel & Tourism Council, 2018). In addition, the statistics has shown that there were more than 13 million of people were recruited in the European Union Tourism industry in 2016. This shows that the substantial demands of labour forces that requires only low entry in the tourism industry and thus the economic growth was shown to be significantly rebound in EU in 2016 (Worldbank, 2020). As the employment rate was positively facilitated by the development of tourism, this means that the people’s spending powers are indirectly promoted due to higher household income level, which then helps to promote economic growth towards the nation. As the consequences, employment rate contributed by the tourism industry is considered as one of the most essential issue that should be constantly taken into deliberate consideration.
In fact, after the financial crisis in 1997, tourism Malaysia tried to increase the tourist arrival and receipts, therefore, in 1999 the financial condition in Asia are growing slowly, Malaysia tourism board come out one campaign “Malaysia Truly Asia”, is promote Malaysia to the world and attract more foreign tourists and encourage local people travel within Malaysia, promote inbound tourism and domestic tourism (Ministry of Tourism, Arts and Culture, Malaysia, 2020). On the other hand, global economy stability will also affect the country’s tourism industry. For instance, in year 1997, global economic crisis affected all the industry and tourism was regarded as one of the victims. During this period, tourist arrival in Malaysia reached 5.8 million (CEIC, 2018). The tourism industry will have affected by the economic problem, once the economic conditions was bad, many tourism industries like hotel, travel agency and restaurant, they will cut cost for example, cut down manpower, some of the hotel and travel agency’s private bus or car driver, tour leader and staff for cost saving, therefore in that financial crisis years many people losing their job.
In overall, different kinds of economic problems can unfavourably occur if the development of tourism industry in a country, such as Malaysia is affected by various determinants stated in this study. Thus, it can be concluded that tourism receipt, tourism arrival, real exchange rate and employment rate are regarded as among the main causes driving different kinds of variations towards the economic growth in each country, especially Malaysia. The tourism industry in the worldwide, including Malaysia is inextricably coping with one of the biggest threat that challenges the global tourism industry, which is the Covid-19 virus that exploded in the latest November 2019. The Covid-19 pandemic not only significantly reduces the number of inbound tourists and leads to the declination of tourism revenues to the global, but also it could potentially lead to severe economic crisis in cutting off at least 50 million jobs as well as the continuous depreciation of various countries’ real exchange rate, at the time when the outbreak is over in the future (Faus, 2020). Malaysia, serves as one of the most influential tourism nation in the worldwide, will definitely face an extremely hard challenge in dealing with the COVID-19 issues (Yunus, 2020). Therefore, this study is very important to be conducted so as to review how will the chosen factors, such as tourism receipt, tourism arrival, real exchange rate and employment rate could lead to impacts to the economic growth in Malaysia, as they are studied as a whole.
1.4 Research Questions
Research questions refers to the enquiry regarding to a research in which the academician will strive to figure out throughout the conduct of a research (Mattick & Johnston, 2018). It is a necessary component that plays vital role for leading a more comprehensive concept in articulating the overall research design, as well as it might influence the ways scholars determine the appropriate sampling plan for actual implementation of research during the FYP2 phase. Thus, a course of key questions is identified and listed as below:
- Does tourism receipt affect the economic growth in Malaysia from year 1998 to year 2018?
- Does real exchange rate affect the economic growth in Malaysia from year 1998 to year 2018?
- Does tourism arrival affect the economic growth in Malaysia from year 1998 to year 2018?
- Does employment rate affect the economic growth in Malaysia from year 1998 to year 2018?
1.5 Research Objectives
Research objectives refers to the to-be-achieved goals determined by scholar during a research, which generally can be categorized into main objectives and specific objectives (Wanjohi & Lee, 2009). The main purpose of conducting this study is mainly to investigate into the contribution of tourism development to the economic growth in Malaysia. In addition, there are four different specific research objectives demonstrated as followed:
- To investigate the relationship between the tourism receipt and the economic growth in Malaysia from year 1998 to year 2018.
- To investigate the relationship between the tourism arrival and the economic growth in Malaysia from year 1998 to year 2018.
- To investigate the relationship between the real exchange rate and the economic growth in Malaysia from year 1998 to year 2018.
- To investigate the relationship between the employment rate and the economic growth in Malaysia from year 1998 to year 2018.
1.6 Significance of Study
According to the study of Sarantopoulos & Demetris, (2015), the scholar emphasized that ‘economic growth’ as a key predictor to measure the ongoing development of the overall economy for a nation, especially Malaysia in this study. The term, economic growth signifies for the degree at which a large deal of benefits, especially revenues and profits would be potentially driven by the facilitation of economy to a nation within a specific duration of period (López & Arreola, 2019). This is supported by Meyer & Ferreira,(2019) that articulated that all the activities involved in the present economy system could be deemed as a construction of creation outcome, feasibly gauged by Gross Domestic Product (GDP) as the key predictor. In spite of experiencing immediate economic expansion across numerous decades, Malaysia is currently still not being regarded as a well-developed nation, as compared to others. For instance, this can be proven by observing that Malaysia is still not possessing the similar economic capability as contrast to other developed nations, in term of GDP, per capita income, extent of machination and quality of living standard (Puah, Jong, & Ayob, 2018). Thus, through the efforts made in boosting the tourism sector in Malaysia, there are a number of advantages that can be dedicated to several parties involved in this study.
First and foremost, the findings of this study can significantly contribute to the government and legal authorities or department that are playing their roles in managing the tourism industry in Malaysia. By knowing that tourism can be normally offer a whole country GDP rate that is ranged from at least 5 to 10%, there should be a myriad of measures enacted by our government after realizing the extent of how tourism sector is essential in driving huge impacts to the ceaseless growth of economy in Malaysia. Through developing a more inclusive insight towards the circumstances happen in Malaysia’s tourism sector, the government is able to enact more ideal and appropriate laws and regulations to impose stricter control over the insecurities and unknown impacts driven by the rapid changes of tourism industry in Malaysia. Moreover, the legal authorities related to the tourism sector can also constantly draft, design and structure the latest and most updated internal policies in order to deal with the varying changes happen to the tourism sector from time to time.
Secondly, this research will aim to provide a lot of discussions regarding to the background context of the current circumstances happen in the global tourism as well as the local tourism in Malaysia. Thus, the acquisition of the findings in this study is able to serve as a guidance to assist the travel agency or tourist agency. Through grasping more in-depth knowledge towards the overall tourism industry, the tourist agency is perceived to formulate more intelligent and wiser decisions in determining an effective marketing strategic plan to boost up the company growth in either short or long-term consideration. For instance, the findings obtained in this study would be mainly interpreted from the tourism data ranged from 1998 to 2018, whereby this allows the travel agency to predict the future growth trend of Malaysia tourism sector via observing the changes over two decades, by knowing how the overall tourism sector varied in relation to the several factors addressed in this study, namely tourism receipt, tourism arrival, real exchange rate and employment rate.
Apart from that, the results obtained from this research also dedicates meaningfully to the tourist itself, or so-called travellers, which incorporates for both the foreign travellers and local travellers. On the one hand, through the perceptions provided towards how the flourishment of tourism sector can lead to the rising growth of a nation’s economy, this actually indirectly boosts up the consciousness among the society regarding to their importance of roles played being a tourist, especially the hodophile (crazy travel lovers). On the other hand, the travellers are also regarded as one of the most benefited party in this study, as they are potentially having large opportunity to enjoy a better tourism experience from Malaysia, due to more ideal, wide-ranging and all-inclusive enforcement of nation policies that shapes the fundamentals of tourism sector.
1.7 Definition of Key Words
Various scholars could have raised different views in defining the variables used in the academic research. In addition, the occurrences of faulty perception towards the operational definitions used in the research could lead to unnecessary misunderstandings among both the readers and scholars. Therefore, this section aims to provide fundamental explanation from one of the previous scholars, in order to provide full insight to the readers regarding to the adoption of the chosen independent variables and dependent variable in this study. Through the support of past journals as evidence, the variables such as Economic Growth (DV), Tourism Receipt (IV), Tourism Arrival (IV), Real Exchange Rate (IV) and Employment Rate (IV) are explained in the table 3.1 below.
Table 0.1 Definition of Key Words
No. | Variable | Description given by past scholars | Sources of Journals |
1. | Economic Growth (DV) | A rising in the quantity of products and services generated for each of the individual population across a specific duration of period | (Bal, 2016) |
2. | Tourism Receipt (IV) | Disbursements contributed by the incoming tourists or travellers, which encompass the outflows of cash to the countrywide transporters for either local or global carriers | (Gramatnikovski & Milenkovski, 2016) |
3. | Tourism Arrival (IV) | Total number of ‘visits’ or ‘arrivals’ by the tourists to a particular nation within a specific time of period, regardless of different or same individuals | (Andres & Cheok, 2016) |
4. | Real Exchange Rate (IV) | The ratio of an overseas product price level over the local product price level, which is multiplied via the nominal exchange rate | (Barbosa & Jayme, 2018) |
5. | Employment Rate (IV) | The evaluation of the level to which existing workforces (those who are able to work) are hired and adopted | (Sadiku & Ibraimi, 2015) |
1.8 Chapter Layout
This report is mainly categorized into two primary phases, namely FYP phase one and FYP phase two. There are an accumulated of three chapters covered by the first phase of FYP as an overall, incorporates chapter one, chapter two and chapter three, whereas the second phase of FYP encompasses chapter four and chapter five.
In the first chapter, the topic regarding to the tourism industry in Malaysia is briefly explained in term of its background context, so that to allow the readers to grasp basic understandings towards the latest circumstances of tourism industry happens in Malaysia. In addition, the current issues exist in the tourism industry will be identified as to point out the research gaps that are essential in this study, which the leads to the formation of research questions and research objectives in the subsequent sections. The significance of study is also justified in relation to how this research contributes to several parties, such as government, tourism sector and the tourists itself.
While in the second chapter, the literature reviews are focused on the exploration into the previous findings discovered by the past scholars in each study. The chapter two covers the description of dependent variable, economic growth as well as the explanation of the independent variables, such as tourism receipt, tourism arrival, real exchange rate and employment rate. All the things will also be depicted in an overall view through the illustration of theoretical framework for quick-understanding among the readers.
Afterwards, chapter three is one of the most essential topic in the research, as it entails the research methodology that comprehensively informs the list of tools, equipment and procedures to be taken into account during the actual implementation of analytical processes. The important things to mention in this chapter includes sources of data, research design, sampling size, target respondents as well as the sampling techniques used in the study. Also, the processes to collect data is also an inextricably essential section to consider, as this affects the accuracy of the findings at the end of research.
1.9 Conclusion of Chapter
Above all, the tourism industry in Malaysia is showing a positive growth, in which it appears to have large correlation with the positive economic development in Malaysia. Some of the latest articles and news have shown that a generous amount of benefits and revenues were brought to the Malaysia government due to the rising of tourism industry in stimulating the flow of money in the global. The assessment into the existing literature reviews of the tourism industry, the researcher has also identified the important issues in tourism industry that might potentially affect the economy growth, namely tourism receipt, tourism arrival, real exchange rate and employment rate, and thus selected as the independent variables. In the next chapter, the researcher will further put more emphasis on more in-depth discussions on the chosen independent variables and dependent variable.
Chapter Two LITERATURE REVIEW
2.1 Introduction
Chapter two would aim to present for the discussions of findings obtained from the former scholars on the topic regarding to the tourism industry and its impacts on economy growth. This chapter initially begins by addressing for the dependent variable, in which relates to how the prior academicians define, explain and describe for the term “economy growth’ in each of their own studies. Then, the subsequent section is proceeded with the clarification of chosen independent variables in this study, that incorporates tourism receipt, tourism arrival, real exchange rate and employment rate. Lastly, a theoretical framework will be structured as well to delineate the overall view of the relationship between the independent variables and dependent variable.
2.2 Dependent Variable (DV)
The dependent variable is a kind of variable that is generally relies on the changes happen to other variables. The dependent variable that has been chosen in this research is economic growth (GDP). In the following section 2.2.1, the economic growth in various studies would be further discussed.
2.2.1 Economic Growth (GDP) and Global Tourism Industry
Economic growth is the growth of the production of goods and services during a particular period. Although the term is often used to discuss short-term economic performance, in the context of economic theory, it usually refers to the growth of long-term wealth. Economic growth is usually distinguished from economic development, and the latter term is limited to economies that are close to survival. The process by which a country’s wealth grows over time.
To measure economic growth, gross domestic product (GDP) is the best way to measure economic growth. GDP measures final production and does not include parts manufactured for the product. It also includes exports because they are produced in the country and imports are subtracted from economic growth. Generally, the World Bank will use gross national income instead of GDP to measure growth. It includes return of citizens working overseas. In Malaysia, GDP represents a measure of national income and output for a given country’s economy.
Chiu and Yeh (2017) reported that the tourism industry has a comparative advantage and can stimulate overall economic growth. Conversely, if tourism has a comparative advantage, the development of tourism may not stimulate economic growth. They further stated that more tourist receipts and arrivals may not necessarily lead to higher economic growth if the tourism industry faces a trade deficit and comparative disadvantage. This may be reason why tourist do not lead to economic growth in the certain countries.
According to Webster and Ivanov (2014) the tourism contribution to a country’s GDP can be measured using growth decomposition. Their study found that tourism’s share of a country’s GDP has a positive impact on a country’s economic growth. This shows that tourism specialization can enhance the impact of tourism on economic growth. Unfortunately, in their research, there is no evidence of the impact of tourism GDP on tourism’s contribution to economic growth. According to Bojanic and Lo (2016), tourism dependence has a moderating effect on the tourism–growth.
Chang, Khamkaew, and McAleer (2012) consider three threshold variables: the degree of trade openness, the share of investment in GDP, and the percentage of government consumption expenditure in GDP. When under the lower level of regime, they found that tourism development has a greater impact on economic growth, while under the upper level of regime, the relationship between tourism and economic growth is lower or insignificant.
2.2.2 Economic Growth (GDP) and Tourism industry in Malaysia
In this financial crisis occurred in 1997, Malaysian economy has been seriously damaged by the turmoil in the financial markets. The gross domestic product (GDP) contracted by 6.7% in 1998 and poised to register a growth of 1% in 1999 with a significant withdrawal of capital from Malaysia (Bank Negara Malaysia, 1999). In order to keep the country’s economy from degenerating further, Malaysian government announced capital control in September 1998. These controls cover mainly two aspects of the country’s financial system, which the currency exchange and stock exchange.
At the beginning of September 1998, Malaysia legally imposed capital controls in the country and pegged the Malaysian ringgit at a fixed exchange rate of RM3.80 to US$1.00. The main reasons were to protect the economy from global turbulence by making it impossible for speculators to have access to funds, to create a stable currency exchange rate and insulate the domestic economy from adverse global developments (Bank Negara Malaysia, 1999). Before the implementation of the capital controls in September 1998, the tourism industry was experiencing a serious decline in growth rates compared to earlier years. In 1994, tourism contribute 4.6% of Malaysian domestic product and tourist earnings that year had reached to RM9bil compared with RM4.5bil in 1990 (Kwan, 1999).
During the year 2003, Malaysia imposed a week-long travel restriction on travellers from mainland China, Hong Kong, Taiwan, as well as Vietnam, among others, following the outbreak of another coronavirus – Severe Acute Respiratory Syndrome (SARS). The travel restriction, which was announced by the government on April 10, 2003, was to stem the spread of SARS, which like the current coronavirus, also originated from China. The impact of the 2003 SARS outbreak on Malaysia’s tourism industry which the country’s third biggest foreign income earner behind manufacturing and palm oil. This actually leads to negative influence towards tourism industry in Malaysia and thus the overall economic growth of Malaysia in a continuous period of time.
Also, the latest issue, which is the Covid-19 virus that exploded in the latest November 2019 is posing a great challenge to the development of economic growth in each country, as the global tourism industry is critically being affected. The Covid-19 pandemic not only significantly reduces the number of inbound tourists and leads to the declination of tourism revenues to the global, but also it could potentially lead to severe economic crisis in cutting off at least 50 million jobs as well as the continuous depreciation of various countries’ real exchange rate, at the time when the outbreak is over in the future (Faus, 2020). Malaysia, serves as one of the most influential tourism nation in the worldwide, will definitely face an extremely hard challenge in dealing with the COVID-19 issues (Yunus, 2020).
2.3 Independent Variables (IV)
There are altogether a total of four different indicators selected as the independent variables in this study. They incorporate tourism receipt, tourism arrival, real exchange rate and employment rate. Each of the independent variable will be discussed in detail from section 2.3.1 to 2.3.4, via the reference of past journals and theses.
2.3.1 Tourism Receipt (TR)
According to World Trade Organization, tourism receipt is defined as an expenditure by destination country from inbound visitors which include payments to national carriers for international transport (Tilastokeskus, 2018). In short, the receipts of tourism can just be simply apprehended as the revenues gained by the tourism industry from the tourists and visitors as they perform spending during their travel in a country. For instance, the receipts can incorporate any kind of pre-payment conducted to the products and services accepted in the purchaser of nation. In certain nations, receipts were actually excluded for the passenger transport goods. It is important for identifying a country’s degree of tourism specialization. The relationship between tourism receipts and GDP has recently become the major focus of some recent research in tourism economics. According to the study conducted by Yalçinkaya & Daştan, (2018), it was found that international tourism receipts have positive influence towards both towards both the developing and developed countries.
Figure 2.1 Tourism Receipts in Malaysia History (1998-2018)
Source: (Trading Economics, 2020a)
Figure 2.1 above demonstrates for the 20-years period of tourism receipts data ranged within 1998 to 2018 for Malaysia. According to Tourism Malaysia official website, tourism industry effects positively on the economy. In addition, tourism sector is one of the main contributor in foreign exchange earnings, new investments and employment opportunities. Based on Boga & Erkisi, (2019) that studied for 483 observations within 1995 to 2017 from at least 21 Asia Pacific Economics, short-term relationship and long-term relationship between the tourism receipts and economic growths were explored using different techniques. On the one hand, the short-term causality was supported by Dumitrescu and Hurlin VAR Panel Causality Test, and eventually the bi-directional bond is found between tourism receipt and economic growth in short-term. On the other hand, Westerlund ECM Panel Co-Integration Test was conducted to test the long-term causality, and there was positive influence imposed by tourism receipt on economic growth in long-term consideration. In short, tourism receipt was found to significantly influence the economic growth in both short-term and long-term consideration.
Figure 2.2 Asia’s most visited countries
Source: (Brick, 2018)
Figure 2.2 demonstrates for the Asia’s most visited counties, that include China, Thailand, Malaysia, Japan, Indian, South Korea, Singapore, Vietnam, Philippines, Cambodia, Laos, Sri Lanka, Nepal, Mongolia, Brunei and Bhutan. According to the report by World Tourism Organization (WTO) that addressed in “The Star” media publication, the latest statistics shows that Malaysia places the top 3 in the ranking of counties among South East Asian region either by international tourism arrivals in year 2017 (Brick, 2018). The successful captivation of abundance number of tourists to visit Malaysia has triggered higher tourism receipts, as the chance at which the tourists to spend in tourism industry is increasing as well. As a result, the tourism receipts gained by Malaysia tourism industry contributed approximately 15% to facilitate the overall economy development in the year 2017 (Bernama, 2018). According to Shakouri & Yazdi, (2017) in the study towards the tourism sector in Iran, it was articulated in the study that tourism receipt plays a paramount role in controlling the growth of economy in the country, through analysis of data from 1995 to 2014.
Moreover, from the previous statistics, it was also shown that international tourism arrivals in Malaysia increased from 7.9 million in 1999 to 23.6 million in 2009, representing an average annual growth about 15.7 million and the growth of tourist receipts has been even more spectacular, rising from RM 12.3 billion in 1999 to RM 53.4 billion in 2009 with an annual growth about RM 41.1 billion (Malaysia Tourism Statistics, 2010). On the other hand, the latest statistics in 2019 reported that the huge contribution role played by the tourism receipt in Malaysia, in which it was excitedly recorded an breaking high of around RM84 billion in 2018, and this indicates for around 2% of growth rate as compared to the previous years (Hwang, 2019). The main reason that the tourism receipt was showing an increase in 2018, could be possibly because the slightly depreciation of Ringgit Malaysia at that year, which indirectly promotes more tourists to spend more willingly in the tourism industry, especially the tourists originated from China. According to the study of Gramatnikovski & Milenkovski, (2016) that mainly investigated on the case of republic of Macedonia Tourism industry, via the data ranged from 2005 to 2014, has proven that the international tourism receipt can significantly influence the overall GDP rate for the country.
On the other hand, another news published in New Strait Times media publication also forecasted that a total of RM92 billion tourism receipts will be received in 2019, due to the efforts made by Malaysia government in organizing Malaysia Truly Asia” promotional events with the aim to attract visitors mainly from China and India (Ching, 2019). Thus, revenue acquired from the tourism sector has surprisingly dedicated to at least 15% of the overall Malaysia’s economic growth, as the tourism receipt has driven the services export, present account and overseas exchange reserves. This indirectly means that the tourism receipt can positively influence the economic growth of Malaysia, from the statistics shown. This can be supported by the study of Aleemi & Qureshi, (2015) that examined into the tourism sector in Pakistan, via the data ranged from 1981 to 2013, in which the findings revealed by the scholar proved that there was significant and positive relationship that exists between the tourism receipt and economic growth.
Subsequently, the study of Ivanov and Webster (2014) adopted the growth decomposition methodology to measure the contribution of tourism receipt to economic growth. Their study found a positive relationship to the country’s economic growth, suggesting that tourism specialization can enhance the effect of tourism on economic growth. However, there was no evidence of the impact of tourism GDP on tourism’s contribution to economic growth in their study. Furthermore, another study that was also conducted in the year 2014 towards the tourism industry in Singapore through using the data from 1980 to 2009, also further observed that tourism receipt has significantly and positively driven the increased of GDP rate across the two decades (Lean & Chong, 2014).
2.3.2 Tourism Arrival (TA)
Tourism arrival can be defined as the total or accumulated quantity of tourists and travellers who make their visit towards a particular nation during a specific year (Andres & Cheok, 2016). In addition, Du & Lew, (2014) also described tourism arrival as number of tourists who travel to a country other than that in which they have their usual residence, but outside their usual environment, for a period not exceeding 12 months and whose main purpose in visiting is other than an activity remunerated from within the country visited. In general, the number of tourism arrivals are not measured solely depending on the number of individuals arrive at a country, but it shall be measured towards the same individual who is shown to visit a particular nation for numerous times within a specific duration of period. This indicates that every time the same visitor makes a new visit to a particular country, the new visit would be counted as new arrival. Based on the study performed by Podhorodecka, (2014), the scholar has stipulated that tourism arrival as one of the crucial indicators that affects the economic growth in Cayman Island, in which the day-to-day tourists arrival has positively and significantly affect its country economic growth from 1983 to 2011.
Figure 2.3 Tourists Arrival in Malaysia (1996 – 2018)
Source: (Trading Economics, 2020b)
Figure 2.3 above depicts for the tourists arrival in Malaysia within 1996 to 2018. The blue line in the graph indicates for the Malaysia Tourist Arrivals, while the dotted line implies for Malaysia GDP rate. From the figure above, the statistics has demonstrated that the tourism industry in Malaysia is gradually experiencing huge transformation across the 20 years, in which it plays more and more vital role in affecting the country in many aspects, especially economy. For example, the statistics above shows that the number of tourists arrival to Malaysia was recorded to roughly 10 million only in 2000, but they are constantly growing each year to eventually reach at least 25 million tourists arrival in 2018. Also, it can also be readily observed that the GDP rate almost complies with the changes of tourist arrivals, whereby the GDP rate is shown to rise exponentially within the year 2000 to 2014, as the number of tourists arrivals to Malaysia is increasing similarly as well. In addition, the statistics addressed by the New Strait Times also mentioned that Malaysia observed as positive growth in the tourism industry, which recorded over 20 million tourists in just the first 9 months of 2019 (Abas, 2019). This can be supported by the study of Catudan, (2016) towards Philippines tourism industry, with the adoption of secondary data from 1999 to 2008, the findings also revealed that tourists arrivals was among the predictor variables that raise positive relationship with the GDP growth rate for the Philippines.
Afterwards, in 1999, Malaysia tourism board launched a campaign called “Malaysia Truly Asia”, which succeeded in bringing in over 7.9 million of tourists into Malaysia and receipts of RM 12.3 billion revenue. Since then it was evident that the revenue and tourist arrival kept increasing year by year. The increase of revenues was then found to indirectly boost up the GDP rate in Malaysia. Also, in the latest 2019, the event regarding to the “Visit Truly Asia Malaysia 2020 campaign” was again intended to be relaunched as an approach to attract more number of tourists arrivals, which then encourage higher consumer spending in the tourism industry to promote economic growth (Euronews, 2019). The latest statistics in 2019 reported by TTR weekly media publication supported that the Malaysia’s visitors count up by around 5% and equivalent to 13.35 million and thus leads to a contribution of RM41.70 billion as the tourism receipts (7% of improvement as compared to 2018) (Weekly, 2019). Based on the study conducted by Ren & Can, (2019), the scholar also indirectly reveals that more number of tourists arrival are led by higher quality of tourism environment and services provided to them, and thus they are the key drivers of economic development for the 8 involved Mediterranean countries, due to higher number of consumer spending chances.
Apart from that, Carla and Paolo (2012) have investigated long-run, short-run, and contemporaneous relationships across international per capita tourism arrivals, real GDP, and the share of total commercial transactions on GDP for the Italian economy. The co-integration analysis gives evidence of two independent long-run relationships across the variables according to which real GDP and international arrivals on the one side, and total trade and international arrivals on the other, emerge as complementary activities.
Massidda and Mattana (2013) employed the structural vector error correction model and find bidirectional causality between international tourism arrivals and real GDP in the long run. They found international tourism arrivals have no impact on real GDP in the short run. Additionally, a study by Aslan (2014) and Seghir, Mostefa, and Abbes (2015) further showed evidence of bidirectional causality between tourism and economic growth in a panel of countries. the Mediterranean region, Tugcu (2014) found a feedback effect between tourism receipts and economic growth in European countries and unidirectional causality from tourism receipts to economic growth in Asian and African countries.
2.3.3 Real Exchange Rate (RER)
Real exchange rate can be defined as the ratio of an overseas product price level over the local product price level, which is multiplied via the nominal exchange rate (Barbosa & Jayme, 2018). It is generally adopted to gauge the comparison cost between the foreign products and local products. Conversely, the real exchange rate can also be expressed as the ratio of the price level abroad and the domestic price level. It allows us to know how many goods and services in the domestic country can be exchanged for foreign goods and services (“SparkNotes: International Trade: Exchange Rates”, 2020). We can expect healthy exports and a trade surplus when the domestic price is lower. However, if the prices in the domestic country are higher, that means we can expert sluggish exports and a trade deficit. Mathematically, the real exchange rate = (Nominal exchange rate X domestic price) / (foreign price).
Regarding to the linkages between the real exchange rate (RER) and growth, the existing research, Rodrik, (2008) has found a positive relationship between RER undervaluation and economic growth. Different reasons have been proposed for this connection, but they all suggest that it should be in developing countries. Besides, Rodrik (2008) analysed and found evidence that the relationship between RER and economic growth in developing countries is more general. The fluctuation of the RER may have negative or positive effects on growth when around its equilibrium level. Normally, researchers will use the different terms to indicate changes in RER such as exchange rate misalignment, exchange rate disequilibrium and exchange rate uncertainty to explore the equilibrium exchange rate. Exchange rate misalignment is defined as the deviation of the RER from its equilibrium value (Ehigiamusoe & Lean, 2019). According to Vieira (2013), there are some existing research shows that a highly fluctuating exchange rate can negatively affect economic growth and moderately exchange rate fluctuations will have a positive impact on growth.
Exchange rate undervaluation means that the domestic country’s currency is lower than 100 or seriously depreciated. However, exchange rate overvaluation is that the exchange rate of one currency is too high and domestic producers are not competitiveness. Nonetheless, according to Habib et al. (2017), most of the empirical work between weak real exchange rates and economic growth by existing researchers have shown a positive relationship. Therefore, we can know that undervalued exchange rates are subsidies for a more efficient trade sector.
Yet, Glüzmann et al. (2012) holds different views on the impact of undervalued exchange rates on the different components of GDP. Their results indicate that export sectors will not affect by currency undervaluation in developing countries. On the other hand, currency undervaluation will boost domestic savings, investment and employment growth.
According to the study of Selimi, (2017), it is demonstrated that real exchange rate can positively influence the economic growth of Macedonia using the data ranged from 1998 to 2015. Apart from that, another study performed by Huong, (2019) in his assessment towards the Vietnam using the data from 2007 to 2017 has proven that real exchange rate has positive impacts triggering the changes of economic growth.
Subsequently, the study by Korkmaz, (2016) also evidence that there is positive causality between the real exchange rate and economic growth for the 9 countries he has investigated in the European Union countries. This is supported by the study conducted by King’ola, 2018) towards Kenya that articulated that there is strongly correlation between real exchange rate and economic growth through its findings. However, the study of Barbosa & Jayme, (2018) that researched on the developing countries disputed that competitive real exchange rate can significantly promote for economic growth. While Habib & Mileva, (2016) that examined into more than 150 nations found that a real devaluation of currency exchange rate significantly boosts up the annual real GDP growth, within the developing countries.
2.3.4 Employment Rate (ER)
Tourism is one of the labor-intensive industry and has been criticized and complimented for its impact on employment. The focus on employment in the tourism industries is further justified by the fact that tourism industries have matured into a major consumer market experiencing increasing global and national competition, market turbulence and changes in consumer demand. These changes are deserving of attention, not only to understand the quality of the tourism products and services, but also to understand the quality in human resources which one of the major assets of the tourism industries. According to the study of Dayananda, (2014), tourism plays a vital role in increasing the employment rate, due to the abundance number of career opportunities created by the tourism sector, either directly or indirectly.
Based on Aynalem & Birhanu, (2016) on their study towards the employment opportunities led by Tourism sector, the scholars mentioned that the international tourism sector has led to an increase of 9.5% GDP growth rate in the international economy, through the labours forces career opportunities it provides to the people worldwide. Another study by Rokanuzzaman, (2017) further addressed that tourism sector as a fundamentally labour-intensive industry, and thus it possesses with large potential as the primary source of employment opportunities to aid in coping with low unemployment issues concerning each country. For instance, a report by ILO Global Dialogue Forum has expected the global tourism to offer for at least 300 million job opportunities by 2019, which makes tourism to gain a high momentum in the international visions of each nation.
Furthermore, the recent statistic in 2017 also demonstrates that tourism occupying for at least 20% of the all the new jobs introduced globally as well as 10% of the total jobs available in the global (World Travel & Tourism Council, 2018). For instance, tourism helped Maldives to create a lot of work opportunities, as tourism has accounted for approximately 40% of the overall employment to aid the inhabitants to continue survive the small island country. Also, another instance is the Bhutan, which enjoys more than 30,000 jobs driven by the progressive flourishment of tourism industry over the recent years since 2010. Besides that, the jobs offered by the tourism industry are generally low-entry in skills requirements, whilst in gigantic volume. For instance, those jobs that are directly or indirectly associated with the tourism sector are comprised of tour guides, drivers, planters, hotel service staffs, laundry staffs, restaurant waiters and so on (Staniūtė, 2018). As such, the travel and tourism sector is able to act as a substitution to replace the illegal works with authentic source of revenues, by providing them an alternative income option (World Travel & Tourism Council, 2017).
According to the report from World Tourism Organization (World Tourism Organization, 2001), the Least Developed Countries (LDCs) have been officially designated by the United Nations since 1971 as a category of countries suffering from structural difficulty in their socio-economic development and regarded by the international community as deserving special treatment in support of their efforts to overcome these difficulties. Although only 0.5% of the world’s exports of services originate from LDCs, international services are an important part of the economies of these countries. In 1998, services accounted for 20% of the total exports of goods and services of the LDCs. However, in 13 of the 49 LDCs, services export receipts exceeded merchandise export receipts and in all but three of those, the share of tourism services exports in total foreign exchange earnings was more than twice greater than the share of merchandise exports.
The share of the LDCs in the world’s exports of international tourism services was 0.6% in 1988 (with 2.4 million international tourist arrivals) and 0.8% (Belloumi, 2010) in 1998 (5.1 million). Throughout the 1990s, tourist flows toward the LDCs increased more rapidly than tourist inflows to the rest of the world. This growth was particularly strong in seven countries (Cambodia, Mali, Lao People’s Democratic Republic, Myanmar, Samoa, Uganda, United Republic of Tanzania), which hosted over 1.2 million visitors in 1998, in comparison with 0.4 million in 1992. During the 1990s, tourism growth was much slower in several LDCs, while a decrease was observed in a number of countries that suffered socio-political and economic instability.
The International Labour Organization (ILO) estimates that tourism generated 253 million jobs worldwide in 2010. Tourism’s indirect contribution to job creation confirms its importance for employment in the tourism supply chain. Tourism also has qualitative impacts by encouraging the creation of jobs for young people. In many countries, however, especially in Europe, a large proportion of these are seasonal jobs that can be secured in the long term only by improving the level of qualifications so that young people can become multi-skilled. The majority of the jobs created in the tourism industry are for young people under the age of 25, who account for about half of all tourism jobs (Goldin, 2010). In addition, most of them are jobs for women. This feature of employment in the tourism sector underlines the importance of continuous vocational training programmes for young people in order to secure long-term jobs.
2.4 Theoretical Framework of Study
Figure 2.4 Adapted Theoretical Framework
Sources: (Andres & Cheok, 2016; Bal, 2016; Sadiku & Ibraimi, 2015)
Figure 2.1 above demonstrates about the adopted theoretical framework in this research as the overall. The research is performed to investigate into the extent to which the tourism can contribute to the development of economic growth in Malaysia, and thus two primary variables, namely independent variables and dependent variable are identified separately in the theoretical framework. The dependent variable, as shown in the theoretical framework is identified as “Economic Growth in Malaysia”, whereas the independent variables incorporate tourism receipt, tourism arrival, real exchange rate and employment rate. From the theoretical framework, it can also be observed that hypothesis has been made to assume the relationship between each independent variable and dependent variable, such as H1, H2, H3 and H4.
2.5 Conclusion
As the summary, this chapter aims to provide all-inclusive reviews into the literature papers pertaining to the field of tourism industry and its relationship with economic growth, such as journals, dissertations, theses and articles published online in the internet platforms. This chapter has chiefly delved into the investigation towards the types of relationship links between the independent variables (tourism receipt, tourism arrival, real exchange rate and employment rate) and dependent variable (economic growth in Malaysia). However, the assessment into various journals clarified that different scholars revealed various findings in their research towards the influence of tourism industry towards the economic growth, possibly because of the variation of cultural context, research scope, research sample and countries. For instance, there are existing literature papers there discussing about the influence of tourism industry and its impacts on economy growth, such as Pakistan, Australia and United States, but showing different research methodologies in each publication. Thus, in the subsequent chapter, the researcher will further identify the appropriate methodology used in the study, so that with the purpose to discover how the overall development of tourism industry can dedicate its influence towards the growth of economy in Malaysia.
Chapter Three RESEARCH METHODOLOGY
3.1 Introduction
Chapter three is essentially served to offer comprehensive understanding towards the compulsory methodologies that will be adopted in the actual implementation of the subsequent analytic phases. This chapter will first aim to present about the research hypotheses development that encompasses the independent variables and dependent variable. Afterwards, the researcher summarizes the operational definition in a table form, with reference to the previous scholars. In addition, the research design and sampling plan would be constructed in a logical manner so that the researcher will perform analytical investigation to the data with full insight. While the data collection method, research procedures and data analysis techniques will be further described in each of its section separately.
3.2 Development of Research Hypotheses
There are four different hypotheses developed in the study, which incorporates H1, H2, H3 and H4. H1 represents the assumption of relationship between tourism receipt and economic growth in Malaysia, which presented in section 3.2.1. Then, H2 indicates the relationship between tourism arrival and economic growth in Malaysia, which demonstrated in section 3.2.2. Afterwards, H3 explains for the relationship between the real exchange rate and economic growth in Malaysia in section 3.2.3, while H4 describes for the relationship between the employment rate and economic growth in Malaysia in section 3.2.4.
3.2.1 Tourism Receipt (TR)
Tourism receipt is defined as the disbursements contributed by the incoming tourists or travellers, which encompass the outflows of cash to the countrywide transporters for either local or global carriers (Gramatnikovski & Milenkovski, 2016). According to the study conducted towards the tourism sector in Pakistan in 2015 that adopts data from 1981 to 2013, the scholar found that there is significant and positive relationship that links between the tourism receipt and economic growth of Malaysia (Aleemi & Qureshi, 2015). In addition, the assessment into the tourism and economic growth in Singapore also showed that tourism has a significant and positive influence towards the economic growth in Singapore in long-term evaluation (Lean & Chong, 2014).
H1: There is a significant and positive relationship between tourism receipt and economic growth of Malaysia
3.2.2 Tourism Arrival (TA)
Tourism arrival can just be briefly defined as the total number of ‘visits’ or ‘arrivals’ by the tourists to a particular nation within a specific time of period, regardless of different or same individuals (Andres & Cheok, 2016). The study of Podhorodecka, (2014) that investigated on the tourism sector of Cayman Islands via the data ranged from 1983 to 2011 have shown that the correlation between the tourism arrival and GDP rate (indicator of economic growth rate) is significant and strongly positive. Besides that, another study conducted by Catudan, (2016) also advocates that the rising in the quantity of traveller’s arrivals to a nation provides the chances for encouraging economic development at the time when it is regarded as a developing country, but it imposed no influence if the nations are deemed as properly-developed.
H2: There is a significant and positive relationship between tourism arrival and economic growth of Malaysia
3.2.3 Real Exchange Rate (RER)
Real exchange rate can be defined as the ratio of an overseas product price level over the local product price level, which is multiplied via the nominal exchange rate (Barbosa & Jayme, 2018). It is generally adopted to gauge the comparison cost between the foreign products and local products. According to the study of Selimi, (2017), it is demonstrated that real exchange rate can positively influence the economic growth of Macedonia using the data ranged from 1998 to 2015. Apart from that, another study performed by Huong, (2019) in his assessment towards the Vietnam using the data from 2007 to 2017 has proven that real exchange rate has positive impacts triggering the changes of economic growth.
H3: There is a significant and positive relationship between real exchange rate and economic growth of Malaysia
3.2.4 Employment Rate (ER)
Employment rate is apprehended as the evaluation of the level to which existing workforces (those who are able to work) are hired and adopted (Sadiku & Ibraimi, 2015). The findings in the study of Soylu & Çakmak, (2018) has shown that there is positive correlation links between the employment rate and economic growth. In addition, Mandel, (2019) in his study also advocates that there is significant and positive relationship between the employment rate and economic growth, as there are more people getting recruited and possessing stable income, and thus stimulate more consumer spending power in the existing market.
H4: There is a significant and positive relationship between employment rate and economic growth of Malaysia
3.3 Model Specification
A plenty of conceptual research frameworks and models had been studied and compared, through assessing to the previous journals, dissertations and publications articles, this research attempts to reformulate a model that was previously adopted by (Bal & Akca, 2010). Therefore, a new model is developed as below to describe how each variable should be defined for analysis purpose in the study.
Table 3.1 Variables, Description, Regression Function
Variables | Stands for | Description |
Economic Growth | Yi | GDP rate (in MYR currency), from 1998 to 2018 |
Tourism Receipt | TR | Income rate for Tourism (in MYR currency), from 1998 to 2018 |
Tourism Arrival | TA | The number of tourists in each annual year, from 1998 to 2018 |
Real Exchange rate | RER | MYR to USD-based, from 1998 to 2018 |
Employment Rate | ER | Real money supply, foreign exchange, production rate, from 1998 to 2018 |
According to the table shown above, Yi indicates for economic growth, TR denotes for Tourism Receipt, Real Exchange Rate means RER while Employment rate is expressed as ER. Through integrated-approach, a multiple linear regression model would be used with the aim to demonstrate the actual ways for which the predictor variables (tourism receipt, tourism arrival, real exchange rate and employment rate) can influence the dependent variable (economic growth). The equation is as listed below:
Whereby:
- refers to the GDP rate (to define economic growth)
- is the y-intercept
- refer to the regression coefficients that indicate the change in y-relative to a one-unit change in TR, TA, RER and ER respectively
- refers to the model’s random error term
According to the equation shown above, can simply mean a ‘constant’ while indicates for the random error term by model that might not incorporated in the deterministic portion. Despite the fact that the direction of results acquired from the conceptual descriptions from the prior researches, we expect that tourism receipt can boost up economic growth (represented by “GDP rate”, in which . On the other hand, another issue is the influence of tourism arrival on economic growth. Conceptually, the tourism arrival is greatly associated with the tourism receipt, in which it is further anticipated that the tourism arrival can significantly lead to rising of GDP rate, economic growth . In addition, the next case is the effects of real exchange rates on economic growth of Malaysia. In theoretical ways of thinking, provided the foreign exchange gaining from export falls, the country’s currency (MYR at this instance) will affect the economic growth positively (Bal & Akca, 2010). Thus, provided there were an increase in export happens continuously, we expect the occurrence of opposite consequences. In fact, looking through the literature reviews in the past, the influence of exchange rates on economic growth is quite controversial, as it might be varied for developing nations such as Turkey. The main reason here is because in light of the conditions where the corresponding prices are in advance of the developing nations, then there might be additional pre-requisites shall be fulfilled so as to facilitate the net import for the sake of positive economic growth. Thus, is not certain for its sign. Apart from that, the fourth case is the employment rate on its impacts towards the GDP rate, economic growth. Conceptually, employment rate can influence GDP in two primary ways. Firstly, the employment rate increases to cause the rising of inflation, and thus to drive for higher price level in the market, which indirectly leads to changes in real money supply and the foreign exchanges (Mosikari, 2013). Also, employment rate can also measure via the production rate, in which the increase of workers will boost up production rate and leads to greater income level and thus indirectly greater consumption rate in triggering higher GDP rate in the economy. Thus, the sign for is not certain as well.
3.4 Research Design
A research design is defined as a methodical and organized initiative for which a scholar adopts to perform an empirical research (Broadhurst & Holt, 2012). The primary goal to establish a research design in the study is mainly to enable the researcher to make sure that the information collected allows him or her to efficaciously stipulate the research issue as unmistakeably, definitely and clearly as possible (Abutabenjeh & Jaradat, 2018). In common the research design shall be observed from two primary standpoints, namely quantitative research design and qualitative research design. In this study, the quantitative research design is adopted as main research design, in order to investigate into the relationship between different variables through the use of ‘numerical input of data or statistics to act in explanation for the analysis of findings. There are also four categories can be classified by quantitative research design, include descriptive, correlational, experimental and quasi-experimental, but the only adopted category is the correlational research design, with the aims to explore whether two or more variables are linked to each other in certain ways, via the use of statistical analytics and observation of the variables (Boeren, 2018).
3.5 Sampling Plan
3.5.1 Target Population and Sampling Size
Figure 3.1 Estimated population size for Malaysia (2010- 2019)
Source: (Mahidin, 2019)
Figure 3.1 above demonstrates for the estimated population size for Malaysia ranged from 2010 to 2019. The total number of populations in Malaysia at the latest 2019 is predicted at 32.6 million, in which mainly consists of approximately 90% of citizens and 10% of non-citizens in the overall population statistics (Mahidin, 2019). On the other hand, the world population is currently shown at comprised of around 7.8 billion residents in the latest 2020 (Chamie, 2020). Since the goals of this study is to probe into the contribution of tourism development to economic growth in Malaysia, therefore, all the travellers, tourists or visitors, either the foreign or local tourists are to be taken into account, as long as they constitute as among the tourists that visit Malaysia over the past 20 years, in which the data would be ranged within 1998 to 2018. This research aims to collect secondary data instead of primary data, and thus the existing tourism data that incorporates under the ‘administrative data’ will be gathered from the online sources, such as nation statistic, government official portal, publication as well as the findings from other scholars. For instance, the types of data that should be emphasized include ‘Tourism receipts (1998 – 2018)”, “Tourism Arrivals (1998 -2018)”, “Real Exchange Rate (1998- 2018)”, “Employment rate (1998-2018)” as well as “GDP rate (1998-2018)” that has to specifically associates to Malaysia tourism industry and government.
3.5.2 Sampling Technique
There are a multitude of open source statistical analytical software available in the online marketplace. For instance, these include JASP, Jamovi, Gnu PSPP, Sofa Statistics, ASCEND, Gretl, RStudio, BlueSky Statistics, Deducer, SPSS and E-view (Zdenka & Petr, 2011). Among the available software, E-view is chosen as the main statistical software in this research, as it is generally adopted for dealing with the time-series oriented analysis for the economic studies (Mujawar & Joshi, 2015).
E-view is formulated by Quantitative Micro Software (QMS), being a pillar of IHS in the latest days. Its features is that it is capable to be implemented for common mathematical investigation and very appropriate for econometric analyses (Mujawar & Joshi, 2015). It provides user-friendly interface, such as Windows Graphical User Interface, that enables for much easier navigation, which creates flexibility to get immediate adaptation even begins to use as a beginner. It also contains the integration of spreadsheet and relational database technology with the conventional functions that can be similarly accesses in the analytical software as well. There are a number of benefits provided by the E-views software as compared to others, such as importing datasheets based (allows the direct import of multiple data formats), unique output-based of data findings (store database in a group format) and feature-based advantages (useful for time-series analysis) (Narang, 2019).
3.6 Sources of Data and Collection Methods
In term of the approaches to acquire data for the accomplishment of a study towards the tourism industry in Malaysia, the sources of data can be categorized into two different types, namely primary data and secondary data. There are both pros and cons for adopting either primary data or secondary data. In this study, the secondary data would be taken into account as the primary source of data in data collection for a plenty of reasons. In the following subsections 3.6.1 and subsections 3.6.2, the assessment into the primary data and secondary data will be briefly performed and thus discussions will be provided to explain the reasons that the secondary data is chosen over the primary data as the data source in this study.
3.6.1 Primary Data
Primary data refers to the latest acquired data from a bunch of sample population by an academician, which known as first-hand resources (Ajayi, 2017). It is generally conducted via methods such as surveys, interviews or experiments. In order to obtain primary data, the researcher will have to do his job to collect data directly from the topics they are of absorbed in. In fact, primary data has its own strength in which it allows the scholars to obtain any kind of info that is considered most recent and updated for them, which actually could be an ideal choice to fit in particular purpose in most researches. However, the limitation of primary data is mainly concerned to the amount of time consumed for collecting the data. In most circumstances, a large data set would potentially take up a substantial amount of time, such as few months in order to complete the entire data collection processes for primary data acquisition (Ajayi, 2017). Also, primary data seems to more suitable to be used when the research is structured to examine the perceptions from the respondents, which indicates that the data obtained are active responses in the presence of the respondent’s awareness (Lowry, 2015). Hence, primary data is not selected in this study, as the purpose of this study is mainly to emphasize on analyzing more on the passive data instead of active data contributed by the tourism industry in an indirect manner (Lowry, 2015).
3.6.2 Secondary Data
Secondary data is defined as those types of data that have been fully collected by someone else (Kabir, 2016). For instance, these include information such as the sources from the other researchers, global worldbank database or government official published data, details from the national population census, official records archives and so on. One of the most suitable type of secondary data to be adopted in this study would be the administrative data, in which it is the data that is attained regularly as part of the daily management of a company, organization or government (Kabir, 2016). In this research, the administrative data that will be taken into account is originated from the government, as it involves the economy data related to the tourism sector in the whole country. The strength of the secondary data to be adopted in this study is that it is commonly published online for high accessibility, which mostly free or relatively low-cost to obtain (Ajayi, 2017). Also, this research tends to collect an enormous sample size of data within 20 years, ranged from 1998 to 2018, which makes administrative data the most ideal choice, as the required data in this study has to undergone a relatively long period of collection process. Hence, this research adopts the use of secondary data, as the existing secondary data enables the scholars to identify the changes of data over time, and ease out the analytical process.
3.7 Research Procedures
This section will briefly explain the research procedures to be conducted in an overall. First and foremost, the researcher will initially tend to identify the research gaps exist in the literature papers regarding to the development of tourism industry towards the Malaysia’s economic growth, and thus formulate research objectives relevant on the research gaps. Subsequently, the second step is to identify the potential independent variables and dependent variable for further investigation in the entire research. Then, the next step is proceeded by drafting a detail methodology plan, prior to the execution of data collection processes. When all the preparations have been done and proceed to data collection, the data will be acquired from the published sources open to the public that exist in the government database, such as Statista, Malaysia Government Officials, MAMPU, Department of statistics Malaysia Official Portal and so on.
However, all the data collected will go through data pre-processing procedures, such as data cleaning and data filtration to cut off the redundant data and invalid data contained in the unorganized data obtained directly from various open sources. Afterwards, the researcher will analyze the data through several statistical techniques, via the use of e-view analytical machine. Lastly, observation will be conducted to evaluate the demonstrated findings from the data, and be further used to support or reject the assumption of hypotheses. The table 3.2 below depicts for the overall logical flow of the research procedures for the entire research.
Table 3.2 Overall Logical Flow of Research Procedures
No. of Steps | Description of Tasks and Activities |
1. | · Investigation of research gaps
· Formulation of research objectives |
2. | · Identification of potential independent variables (IV) and dependent variable (DV) |
3. | · Enactment of Research methodology plan (sampling design, sampling plan) |
4. | · Acquisition of data from open and published resources |
5. | · Data Pre-processing to increase data accuracy before use |
6. | · Analysis of data via e-view analytical machine |
7. | · Assessment, evaluation and discussion of findings
· Validation of research hypotheses |
3.8 Data Analysis Techniques
The data analysis refers to the process of examining, filtering, modifying and structuring the data with the motives to explore meaningful information concealed in the data. There would be a number of data analysis techniques that will be further adopted in subsequent phase of implementation discussed in the subdivided sections as shown below. These incorporate F-test, T-test, Normality Test, Pearson Correlation and Multiple Linear Regression Test.
3.8.1 F-test
An F-test denotes for any analytical test that deals with F-distribution. The purpose of conducting F-test is generally to perform comparison and evaluation between the difference of standard deviation between two different variables (Will Kenton, 2019). Within the F-distribution, there is two types of possibilities, which incorporates null hypothesis, H0 and alternate hypothesis, Ha. F-test is regarded as a model where it is used to evaluate mathematical frameworks which has been tailored to a specific data set so as to identity the statistical framework that most ideally matches the desired sample population. Majority of the F-tests are produced by taking into account the disintegration of changeability in a data set demonstrated as “Sum of Squares” (Will Kenton, 2019).
In addition, the ratio of the sum of the squares contained within F-test also reveals for the unusual and strange properties of unpredictability (Goldstein, 2013). Under the F-distribution, the null hypothesis, H0 is to be determined to be accepted or rejected for its assumption, which fully dependent on the coefficients value represented by the sum of squares (Goldstein, 2013). There are conditions that should be fulfilled in order to portray the variables falling under the null hypothesis, H0. For instance, there should be mathematically independent properties is shown towards the sum of squares, whereby each of the components contained inside shall compromise with scaled x^2 distribution (means the data shall be independent to be normally distributed with a general variance). In particular conditions, the null hypothesis can be directly rejected, provided the F-statistics value (F-critical value) is lower than the F-value in a test. The figure 3.2 below will briefly illustrate for the mathematical formula that can be used to compute F-tests.
Source: (Newman, 2019)
However, there shall be no simple and single formula for F-test, in which null hypothesis, H0 and alternate hypothesis, Ha are both required to be pre-defined prior to begin an F-test. The figure 3.3 below further describes the types of possibilities that alternate hypothesis can be defined.
- HO (Null Hypothesis): Variance of 1^{st }data sat = Variance of a 2^{nd} data set
- Ha: Variance of 1^{st} data set < Variance of 2^{nd} data set (for a lower one-tailed test)
- Ha: Variance of 1^{st} data set > Variance of a 2^{nd} data set (for an upper one-tailed test)
- Ha: Variance of 1^{st} data set ≠ Variance of a 2^{nd} data set (for a two-tailed test)
Figure 3.3 Null hypothesis and Alternate hypothesis
Apart from that, H0 can denote for the non-appearance of significance, while H1 is in the total opposite of H0, that indicates for the existence of significance. For instance, if the F-value is found out to be larger than the F-critical value, the hypothesis shall not be accepted due to its value does not fall under the significance area. The significance area can be represented by p-value, in which the null hypothesis would be accepted provided the p-value is larger than the significance level. It is solely dependent on the “p-value’ to whether clarify the significance of a framework used in the research.
3.8.2 T-Test
The T-test is a deductive analytical inspection in which adopted to evaluate the comparison of difference for the ‘mean’ value between two different sample populations (Will Kenton, 2020). In short, it means that t-test is used to determine if the ‘central tendency’ value of two different sets of data are arithmetically and significantly varied from each other. There are altogether 3 categories of t-test are available, namely one-sample t-test, unpaired two sample t-test and paired t-test (Walker, 1995). The first type of t-test, one-sample t-test is primarily adopted to assess the average value of a sample data with a hypothetical average value. The second one is generally implemented to determine the variation of central tendency value between two independent samples, while the last one is adopted for comparison of the average value between two associated groups of samples.
A T-test can be measured using the predictor, T-score, that is recognized as the ration between the variation “between” 2 samples populations and the variation “within” 2 samples population (Al-achi, 2019). The increase of T-score indicates for larger variation between 2 sample groups, while the decrease of T-score denotes for higher similarities between 2 sample groups. For instance, a T-score of 4 describes that the sample population are 4 times as varied from each other. Thus, a bigger t-score coefficient obtained during the analytical test generally guarantee more predictable and quotable outcome.
Besides that, each t-value contains a significance value (so-called p-value) in further explaining its presence. The significance value here can be just understood as the possibility for which the findings derived from the adopted sample data happened by percentage. The significance value here shall only range from 0% to 100% (Al-achi, 2019). Importantly to mention, a high significance value here is not good, but instead a low significance value is more favoured, as it specifies that the data does not readily occur by chance. As long as the significance-value (p-value) is less than or equal to 0.05, the data can be significantly accepted and declared as valid.
T-test employs ‘central tendency’ and ‘standard deviation’ of two different samples to contrast each other. Thus, the figure 3.4 below demonstrates for the formula of T-test.
, where
- = mean of first set of values
- = mean of second set of values
- = standard deviation of first set of values
- = standard deviation of second set of values
- = total number of values in first set
- = total number of values in second set
3.8.3 Normality Test
The occurrence of errors in performing analytical test is regarded as an extremely ordinary event throughout the process of each research. A normality test refers to an arithmetical technique in data science that is commonly executed to identify if the investigated sample of data in the analytical test complies with the rules of standard normal distribution (Carlo, 2013). There are abundance number of statistical tests, such as t-test, F-test, Pearson Correlation and Multiple Regression test that are fully dependent on the supposition that the sample data obeys the rules of standard normal distribution. The figure 3.5 below demonstrates for the formula for skewness and kurtosis that suits the normality testing.
Figure 3.5 Formula for Normality Testing
In general, the normality test can be measured via two main parameters, namely skewness and kurtosis (Das, 2016). Two of the most renowned technique to deal with normality testing are the Omnibus K-squared and Jarque-Bera tests that adopts skewness and kurtosis as well. Skewness here refers to a moment-based prediction, as it employs the anticipated value of the 3^{rd} power a random variable. It allows the researcher to examine through the degree of the deviation of the ‘overall shape’ of an actual distribution from the expected shape of standardized normal distribution (Das, 2016). It is totally acceptable to be shown in either positive or negative value. There is an extremely rare condition when the data is perfectly normally distributed, the skewness value is theoretically assumed at 0 and to indicate for fully symmetric data of findings (Bai & Ng, 2005). On the other hand, kurtosis designates for the extent to which the tail shape of the graph is different as compared to the expected tail shape of the normal distribution graph. The kurtosis value is expected to portray at exactly 3, when the graph is perfectly normally distributed. In overall, skewness emphasizes on the bell shape, while kurtosis keens on the tails shape (Bai & Ng, 2005).
Two hypotheses can be made prior to conduct the normality testing. These include Null hypothesis (H0) and Alternate hypothesis (Ha). H0 will assume that the data is normally distributed, while Ha is the opposite of H0, which assumes non-normal distribution of data.
3.8.4 Pearson Correlation Test
Pearson test refers to a kind of statistical method to look into the link of bond exist between two quantitative, continuous variables, that either suits for analysing either primary data or secondary data (Gogtay & Thatte, 2017). For instance, the tourism arrival and economic growth (GDP rate) in this study. This test mainly aims to explore the types of correlation between two different variables, which then can be measured via two parameters, namely p-value and r-value. On the one hand, r-value stands for Pearson Correlation Coefficient, refers to the prediction of how strong for two different variables can be associated with each other. On the other hand, p-value refers to the significance value, which enables the researcher to determine if the correlation between the two variables can be clarified as either significant or insignificant.
In Pearson Correlation, the most primary step refers to the sketching of a scatter-plot for the involved variables, with the purpose to examine for the linearity (Kumar & Chong, 2018). For instance, the r-value shall not be computed provided there were not linear relationship found between each other. One unique thing about Pearson Correlation is that the data on the graph-axis can be randomly plotted at any point, regardless of following the rules of x-axis or y-axis. Nevertheless, in order to create clearer sense of understanding towards the data for the readers, the predictor variable in Pearson test follows the general rule to be commonly located horizontally on the axis (x-axis), while the outcome variable is positioned vertically on the axis (y-axis). The figure 3.6 below demonstrates for the formula for Pearson Correlation.
Figure 3.6 Formula for Pearson Correlation
As the scattered plot of points are shown more closer to the linear line in the mid-point of graph, it means that the two different variables are portraying stronger bond of correlation between each other (Hauke & Kossowski, 2011). Positive correlation derives a meaning that both variables are changing positively together, while negative correlation is vice versa. Three types of extreme conditions can happen, for which the r-value shall only ranged from negative one (-1) to positive one (+1), specifically r = -1 (first condition), r=0 (second condition) and r = 1 (third condition). The first condition denotes that a perfectly occurred negative slope of straight line, while the third condition specifies that a perfectly occurred positive slope of straight line (Hauke & Kossowski, 2011). The second condition means that there is completely no linear correlation between the variables. The figure 3.7 below demonstrates the graphs shown for three types of special conditions for Pearson correlation.
Figure 3.7 Special Conditions for Pearson Correlation
Source: (Wong, 2020)
3.8.5 Multiple Linear Regression Test
Sometimes known also as multiple regression, is a kind of arithmetical method that adopts numerous continuous variables to measure and estimate the findings of an outcome variable (Fletcher, 2009). The purpose of conducting this test is to mainly structure and examine whether there was any relationship between the predictor variables and outcome variables for which how they are linked to each other linearly. Fundamentally, multiple linear regression test is actually an augmented version of Ordinary Least-Squares (OLS), whilst encompassing at least two and above for the independent variables. The figure 3.8 below delineates for the formula that describes multiple linear regression.
Whereby:
- i = n observations
- = dependent variable
- = explanatory variable
- y-intercept (constant term)
- slope coefficients for each explanatory variable
- refers to the model’s random error term (also known as residuals)
Figure 3.8 Formula for Multiple Regression Test
A linear regression is only applicable when the study involves at least two continuous variables, namely one predictor variable and one outcome variable (Koloğlu & Birinci, 2018). Thus, in ‘multiple’ linear regression test, despite the presence of only one outcome variable, but there could be a plenty of predictor variables there involved in the study. R-squared value in the multiple regression test, known as coefficient of determination as well, is a mathematical dimension employed to describe the extent of distinction in findings is able to be supported by the difference exist in the predictor variables. R-square will tend to boost up as the number of independent variables are facilitated up in the regression framework, regardless of the types of significance relationship between the newly added independent variables and the outcome variable. It is also not feasible to determine the inclusion and exclusion for the choices of independent variables, whereby it ranges from zero to one; the ‘zero’ coefficient means that the response variable is unpredictable by any predictors, but ‘one’ coefficient denotes that the response variable is predictable in the absence of error occurrences from the predictors.
Besides that, multiple regression test can be mainly measured through significance value and beta value as well. The significance value is known as p-value, while the beta value is the standardized beta of coefficients. There are two types of hypotheses can be formulated towards multiple linear regression test as well, namely the null hypothesis (H0) and alternate hypothesis (Ha). The p-value aims to investigate into the null hypothesis in which assuming that there is no relationship between the independent variables and dependent variable. This will constitute as lack of adequate evidence to prove that there is ‘impacts’ by the predictors towards the response variable at the population level. Provided the significance level is higher than the coefficient value of p-value, this indicates that the p-value is significant, and further supports that the adopted data of sample offers sufficient facts to directly reject H0 for the entire sample of data. On the other hand, the beta-value in the regression test implies for the rate at which the predictors are able to impose influence towards the response variable, when there are variations happen to the predictors. Also, the beta value can contain both positive and negative sign, ranges from -1 to +1. The value below 0 and within -1 denotes for the negative association between the independent variable and dependent variable, while the value more than 0 and within 1 signifies the positive association between each other.
3.9 Conclusion
In overall, the chapter three is playing an essential role in the research during the first phase of research implementation, as there have been a myriad of important research components and procedures pre-identified as the applicable methodology prior to official execution of data analysis processes. The relationship between the independent variables and dependent variable, have been explored through the findings found in various dissertations, and thus assumption are made to assume the possible relationship via the hypotheses made, such as tourism receipt (IV), tourism arrival (IV), real exchange rate (IV), employment rate (IV) and Malaysia’s economic growth (DV). In addition, this chapter also clearly explains that the researcher will take into account the secondary data as the chief data sources, in which the data will be ranged from 1998 to 2018. Moreover, all the data analysis techniques would be conducted using a popular statistical machine, e-view.
CHAPTER 4: Data Analysis and discussions
4.0 Introduction
The relationship between tourism and economic growth in Malaysia demands econometric method that can capture all the objectives outlines in this study. This section focuses on the econometric analyses and interpretation of results. The econometric literatures demand for both formal and informal test for the series before the analysis in which some of the pre-test carried out in this chapter were descriptive statistic, and unit root test.
Furthermore, the study adopted multivariate regression using Ordinary Least Square (OLS) to test the hypotheses postulated in this study. The chapter projected the nature of relationship between the response variable and exogenous variables with empirical and theoretical supports from the existing studies.
4.1 Descriptive Statistic.
Descriptive statistic is one of the most important pre-tests needed to be carried out in an analysis in order to know the behavior of each series. It helps to determine the values of central tendency such as mean, mode, median and the measure of dispersion from the average such as minimum, maximum, standard deviation and other test of normality.
Table 4.1 Descriptive Statistic
logTA | logER | logY_{i} | logtR | logRER | |
Mean | 19.43143 | 96.68857 | 210258.7 | 47338.1 | 96.04383 |
Median | 22.06 | 96.66 | 202257.6 | 53400 | 97.80135 |
Maximum | 27.44 | 97.12 | 358581.9 | 84100 | 103.462 |
Minimum | 5.56 | 96.31 | 72167.5 | 8600 | 85.1425 |
Std. Dev. | 7.049929 | 0.221975 | 100112 | 24094.47 | 5.005958 |
Skewness | -0.55926 | 0.209741 | 0.0029 | -0.0328 | -0.75048 |
Kurtosis | 1.879668 | 2.057916 | 1.425658 | 1.745307 | 2.814216 |
Jarque-Bera | 2.192963 | 0.930551 | 2.168762 | 1.381239 | 2.001484 |
Probability | 0.334044 | 0.627962 | 0.338111 | 0.501265 | 0.367606 |
Sum | 408.06 | 2030.46 | 4415433 | 994100 | 2016.92 |
Sum Sq. Dev. | 994.0301 | 0.985457 | 2.00E+11 | 1.16E+10 | 501.1924 |
Observations | 21 | 21 | 21 | 21 | 21 |
Source: Authors’ Computation from Eviews10, 2020
Table 4.1 shows the results of descriptive statistic for each variable in the analysis. It showed that all the series display a high level of consistency as their mean and median values fall within the maximum and minimum values of the series. Also, the standard deviation of the data series is very low which implies that the deviation of actual data from its mean value is very small. For a further test of normality, we can test whether the mean and median of the distribution are nearly equal, whether the skewness is approximately zero, and whether the kurtosis is close to 3 except that of Arrivals that showed platykurtic curve.
A more formal test of normality is the one given by the Jarque-Bera (JB) statistic. The Jarque-Bera statistic follows a chi-square distribution with 2 degree of freedom, all of the data series used in the study rejects the assumption of non-normal distribution due to the high p-value of JB thus reject the alternative hypothesis of non-normality and thus concluded that they are normally distributed which gives justification for further econometric analysis . In order to capture the objectives outlined in this study, tables 4.2 is presented below.
4.1.1 Unit Root Test
Determination of long run relationship among series in time series econometric analysis requires formal pretest called unit root test in order to determine the stationary of each series in the model. This enables researcher to know the best econometric method that can capture the objectives outlined in the study. For this reason, this study adopted conventional Augmented Dickey Fuller tests and Phillip perron as robustness check for the stationarity. Since most time series data tends to exhibit non-stationary at level because of high volatility, the study intended to test further at first difference for all the series in the model to avoid spurious results. The results of unit root tests from ADF and PP are presented in table 4.2.1and 4.2.2 respectively.
TABLE 4.2.1: Augmented Dickey Fuller Unit Root Test Result
ADF Results | |||||
At Level | At First Difference | Decision | |||
Variables | T-statistic | P-value | t-statistic | P-value | I(d) |
logY_{i} | -1.4806 | 0.5224 | 4.2243 | 0.0044*** | I(1) |
logTA | -1.3044 | 0.8568 | -6.0105 | 0.0006*** | I(1) |
logTR | -3.0501 | 0.1441 | -4.9202 | 0.0048*** | I(1) |
logRER | -3.0363 | 0.1486 | -3.0443 | 0.0487*** | I(1) |
logER | -3.0121 | 0.1533 | -5.7384 | 0.0010*** | I(1) |
*, ** and *** denotes 10%, 5% and 1% level of significance respectively
Source: Author’s Computation from Eviews 10, 2020
The results shown in table 4.2.1 denotes the unit root test from Augumented Dickey Fuller results at both level and first difference.
The result indicated that all variable in the model are not stationary at level, because the p-value for each series is higher than the 5% chosen level of significance, thus all the alternative hypotheses or the series were rejected at level. This led to further testing at first difference where all the series became stationary, thus, the null hypotheses of unit root were rejected against the alternative hypotheses. To further check for the robustness of the result obtained from ADF, this study subjected all the series to another test using Philip perron and its results are shown in table 4.3
TABLE 4.2.2: Philip Perron Unit Root Test Result
PP Results | |||||
At Level | At First Difference | Decision | |||
Variables | T-statistic | P-value | t-statistic | P-value | I(d) |
logY_{i} | -1.5687 | 0.4794 | 4.2243 | 0.0044*** | I(1) |
logTA | -0.9938 | 0.9222 | -6.9500 | 0.0001*** | I(1) |
logTR | -2.9575 | 0.9982 | -6.5475 | 0.0002*** | I(1) |
logRER | 1.9245 | 0.6047 | -2.8548 | 0.0696* | I(1) |
logER | 3.0246 | 0.1502 | -6.7135 | 0.0002** | I(1) |
*, ** and *** denotes 10%, 5% and 1% level of significance respectively
Source : Author’s Computation from Eviews 10, 2020
Table 4.2.2 shows the results of unit root test alternatively obtained from Philip perron method of stationarity. The result confirmed non-stationarity of all the series at level. Having found that all variable were not stationary at level, the next step is to difference all the variables once in order to perform stationarity test at first difference for each variable. The test confirmed that all variables were stationary at first difference, which is also the same with the result obtained from ADF test. The next objective is to examine the type of relationship that existed between the dependent and all the independent variables. This will be achieved by adopting Ordinary least square regression method.
4.1.2 Ordinary Last Square
Ordinary least square is a regression method used in econometric analysis in predicting the nature and parameter among variables. This method of analysis was first adopted by Francis Galton in his empirical study to predict the average height of children based on their parental height. When a study establish causal relationship between two variables, it is regarded as simple or single variable regression. This implies that the study consider one variable called independent variable which can influence the behavior of another variable known as dependent variable. In reality no single variable can completely explain the variation in another variable, hence a need to apply multiple regression that incorporates more independent variables. The multiple regression can be model in the following form;
Where;
Y is the dependent variable
X_{1} to X_{k} are the independent variables
The multiple regression model of this study is thus formulated as follow;
Where;
Y_{i}= Gross Domestic Product as proxy for economic growth
TA= total number of tourists arrivals per year
TR = Revenue generated from tourism industry per year
RER = Real exchange rate
ER= Rate of Employment in Malaysia
log = logathim of the variables
Table 4.3.1 Ordinary Least Square result
Dependent Variable: logY_{i} | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 15:44 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -86.41977 | 24.95316 | -3.463280 | 0.0032 |
logTA | -0.011041 | 0.463972 | -0.023797 | 0.9813 |
logER | 49.40157 | 12.77377 | 3.867424 | 0.0014 |
logTR | 0.746605 | 0.339654 | 2.198133 | 0.0430 |
logRER | -0.795483 | 0.721343 | -1.102781 | 0.2864 |
R-squared | 0.960218 | Mean dependent var | 11.26646 | |
Adjusted R-squared | 0.950272 | S.D. dependent var | 0.237344 | |
S.E. of regression | 0.052927 | Akaike info criterion | -2.835547 | |
Sum squared resid | 0.044820 | Schwarz criterion | -2.586851 | |
Log likelihood | 34.77324 | Hannan-Quinn criter. | -2.781573 | |
F-statistic | 96.54712 | Durbin-Watson stat | 1.432460 | |
Prob(F-statistic) | 0.000000 | |||
Econometric model (please refer to Table 4.3.1)
SE = 24.95316 0.463972 12.77377 0.339654 0.721343
t-statistic = -3.463280 -0.023797 3.867424 2.198133 1.102781
p-value = 0.0032 0.9813 0.0014 0.0430 0.2864
F-stat = 96.54712
Prob(F-stat) = 0.000000
R^{2 }=0.960218 and = 0.950272
Interpretation
= holding receipts, arrival, real exchange rate and employment rate constant, the estimated value of Malaysia GDP -86.42.
= one percent increase in receipts generated from tourism holding arrival, real exchange rate and employment constant will result to an expected average increase of 74.6 percent of the total GDP in Malaysia, ceteris paribus
= one percent increase in employment rate will result to an expected average increase of unit in Malaysia’s GDP, holding Arrival, Receipts and real exchange rate constant.
= one percent increase in arrival into Malaysia will result to an expected decrease of 1 percent of the total GDP in Malaysia, holding receipts, real exchange rate and employment rate constant.
= one percent increase in real exchange rate holding Receipts, Arrival and employment rate constant will result to an expected average decrease of 79.5 percent of GDP in Malaysia.
4.1.3 Test of Hypotheses
H_{0}: There is no significant and positive relationship between tourism receipt and economic growth of Malaysia
H_{1}: There is a significant and positive relationship between tourism receipt and economic growth of Malaysia
H_{0}: There is no significant and positive relationship between tourism arrival and economic growth of Malaysia
H_{2}: There is a significant and positive relationship between tourism arrival and economic growth of Malaysia
H_{0}: There is no significant and positive relationship between real exchange rate and economic growth of Malaysia
H_{3}: There is a significant and positive relationship between real exchange rate and economic growth of Malaysia
H_{0}: There is no significant and positive relationship between employment rate and economic growth of Malaysia
H_{4}: There is a significant and positive relationship between employment rate and economic growth of Malaysia
t_{cal} =
is the coefficient of parameters
is the standard error of parameter’s coefficient
Table 4.3.2 hypotheses
Parameters | Null hypotheses | Alternative hypotheses | T statistics
t = |
α @ 5% level of significance | Reject H_{0 }if t statistics is greater than 5% level of significance |
logTA | ≤ 0 | ≥ 0 | -0.023797 | 1.96 | Do not reject null hypotheses |
logER | ≤ 0 | ≥ 0 | 3.867424 | 1.96 | Reject null hypotheses |
logTR | ≤ 0 | ≥ 0 | 2.198133 | 1.96 | Reject null hypotheses |
logRER | ≤ 0 | ≥ 0 | -1.102781 | 1.96 | Do not reject null hypotheses |
Decision rule: reject null hypotheses if t statistics in absolute value is greater than 5% confidence level i.e 1.96 otherwise, do not reject null hypotheses.
A comprehensive analysis of this table shall be detailed in major findings in the next chapter.
The result presented in table 4.1 shows that the coefficient of determination (R^{2}) is 0.960218 implying that 96% of variation in Malaysia’s GDP is explained by TA, TR, RER and ER. Although R^{2} gives us the background information of the total variation in Y_{i} that was explained by the independent variables used in this study, it is important to check the result of F-statistic that measure the overall significance of the model. The null hypothesis states that TA, TR, RER and ER do not jointly have significant effect on GDP in Malaysia. Since the p-value from the result is 0.000000 less than 0.005 level of significance, the null hypothesis of no significance was rejected and concluded that TA, TR, RER and ER have significant effect on economic growth. The result identifies each variable in the model as of one of the predicators of economic growth in Malaysia, but the value of R^{2} is too high thus necessitated the test for the presence of multicollinearity in the model using variance inflation factor (VIF).
4.2 MULTICOLLINEARITY TEST (TABLE III-VI)
TABLE III
Dependent Variable: logTA | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:34 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 3.563098 | 13.01531 | 0.273762 | 0.7876 |
logER | -2.643807 | 6.646475 | -0.397776 | 0.6957 |
logTR | 0.724625 | 0.025237 | 28.71305 | 0.0000 |
logRER | 0.896464 | 0.308076 | 2.909879 | 0.0098 |
R-squared | 0.983609 | Mean dependent var | 1.251651 | |
Adjusted R-squared | 0.980717 | S.D. dependent var | 0.199238 | |
S.E. of regression | 0.027667 | Akaike info criterion | -4.167510 | |
Sum squared resid | 0.013013 | Schwarz criterion | -3.968553 | |
Log likelihood | 47.75885 | Hannan-Quinn criter. | -4.124331 | |
F-statistic | 340.0555 | Durbin-Watson stat | 1.379822 | |
Prob(F-statistic) | 0.000000 | |||
TABLE IV
Dependent Variable: logER | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:35 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 1.950535 | 0.025960 | 75.13714 | 0.0000 |
logTA | -0.003488 | 0.008769 | -0.397776 | 0.6957 |
logTR | 0.003716 | 0.006386 | 0.581866 | 0.5683 |
logRER | 0.016778 | 0.013078 | 1.282984 | 0.2167 |
R-squared | 0.136102 | Mean dependent var | 1.985374 | |
Adjusted R-squared | -0.016350 | S.D. dependent var | 0.000997 | |
S.E. of regression | 0.001005 | Akaike info criterion | -10.79816 | |
Sum squared resid | 1.72E-05 | Schwarz criterion | -10.59920 | |
Log likelihood | 117.3807 | Hannan-Quinn criter. | -10.75498 | |
F-statistic | 0.892752 | Durbin-Watson stat | 1.310494 | |
Prob(F-statistic) | 0.464965 | |||
TABLE V
Dependent Variable: logTR | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:37 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -7.891503 | 17.71512 | -0.445467 | 0.6616 |
logER | 5.255315 | 9.031824 | 0.581866 | 0.5683 |
logTA | 1.352143 | 0.047092 | 28.71305 | 0.0000 |
LogRER | -1.328396 | 0.401886 | -3.305406 | 0.0042 |
R-squared | 0.984969 | Mean dependent var | 1.601938 | |
Adjusted R-squared | 0.982317 | S.D. dependent var | 0.284208 | |
S.E. of regression | 0.037793 | Akaike info criterion | -3.543719 | |
Sum squared resid | 0.024282 | Schwarz criterion | -3.344762 | |
Log likelihood | 41.20905 | Hannan-Quinn criter. | -3.500540 | |
F-statistic | 371.3401 | Durbin-Watson stat | 1.404582 | |
Prob(F-statistic) | 0.000000 | |||
TABLE VI
Dependent Variable: logRER | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:37 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -8.456451 | 8.135396 | -1.039464 | 0.3131 |
logTR | -0.294522 | 0.089103 | -3.305406 | 0.0042 |
logTA | 0.370879 | 0.127455 | 2.909879 | 0.0098 |
logER | 5.261445 | 4.100944 | 1.282984 | 0.2167 |
R-squared | 0.497107 | Mean dependent var | 1.981891 | |
Adjusted R-squared | 0.408362 | S.D. dependent var | 0.023136 | |
S.E. of regression | 0.017796 | Akaike info criterion | -5.050092 | |
Sum squared resid | 0.005384 | Schwarz criterion | -4.851136 | |
Log likelihood | 57.02597 | Hannan-Quinn criter. | -5.006913 | |
F-statistic | 5.601480 | Durbin-Watson stat | 1.195640 | |
Prob(F-statistic) | 0.007378 | |||
VIF = = = = 66.5292
VIF = = = = 61.009
VIF = = _{ }= = 1.9885
VIF = = = = 1.1575
The results of variance inflation factor calculated by using the R^{2} obtained from the individual multicollinearity test shown in table iii to vi revealed the presence of Multicollinearity as the value of VIF of two independent variables (Receipts and Arrival) were greater than 10. Since the two other independent variables’ VIF were less than 10, we thus concluded that the problem of multicollinearity in the model was not that serious and we can proceed to another test.
To find out whether the residual variance has remained constant through the whole process we need to run a formal test known as the autoregressive conditioned heteroskedasticity ARCH test.
H_{0}: Residuals are Homoscedastic
H_{1}: Residuals are Heteroscedastic
Decision rule: Reject the null hypothesis of residuals are homoscedastic if the Probability of F-statistic and Obs*R^{2} are less than 0.05 level of significance.
F-statistic P-value = 0.4179
P-value of Obs*R-squared = 0.3911
Decision: Do not reject null hypothesis of homoscedasticity as the p-value of both F-stat and Chi-squared were greater than 0.05 level of significance.
Conclusion: There is no sufficient evidence for the presence of problem of heteroscedasticity in the model, thus we concluded that residuals are homoscedastic.
Autocorrelation occurs when there is a correlation between two consecutive observations of the residuals. This is a common problem when time series data is used and a correlogram can be generated to test for autocorrelation using Q-stat for decision.
4.3.4 Autocorrelation
Table 4.3.4
Date: 07/16/20 Time: 15:52 | ||||||
Sample: 1998 2018 | ||||||
Included observations: 21 | ||||||
Autocorrelation | Partial Correlation | AC | PAC | Q-Stat | Prob | |
. |**. | | . |**. | | 1 | 0.230 | 0.230 | 1.2766 | 0.259 |
. *| . | | . *| . | | 2 | -0.066 | -0.126 | 1.3890 | 0.499 |
.**| . | | . *| . | | 3 | -0.219 | -0.185 | 2.6754 | 0.444 |
. *| . | | . *| . | | 4 | -0.181 | -0.103 | 3.6107 | 0.461 |
. |* . | | . |* . | | 5 | 0.102 | 0.152 | 3.9223 | 0.561 |
Table 4.3.5
p | AC | PAC | Q-Stat | Prob | Conclusion |
1 | 0.23 | 0.23 | 1.2766 | 0.259 | No Autocorrelation |
2 | -0.066 | -0.126 | 1.389 | 0.499 | No Autocorrelation |
3 | -0.219 | -0.185 | 2.6754 | 0.444 | No Autocorrelation |
4 | -0.181 | -0.103 | 3.6107 | 0.461 | No Autocorrelation |
5 | 0.102 | 0.152 | 3.9223 | 0.561 | No Autocorrelation |
H_{0}: No serial correlation in the model
H_{1}: There is serial correlation
= 5% or 0.05
Decision Rule: Reject null hypothesis of non-serial correlation if the p-value of Q-statistic is less than 0.05 level of significance.
Decision Making: Do not reject H_{0} of no problem of serial correlation as the p-value from the result is greater than 0.05 chosen level of significance.
Conclusion: Based on the evidence from the result we concluded that there is no problem of autocorrelation as there is no sufficient evidence to support the problem of autocorrelation.
The essence of this test is to find out if there is a linear relationship between the dependent variable (GDP) and the independent variables (TA, TR, RER and ER) We adopted Ramsey Reset test of linearity to examine whether the model was correctly specified.
H_{0}: Model is wrongly specified
H_{1}: Model is correctly specified
Table 4.3.5 Ramsey RESET Test
Ramsey RESET Test | ||||
Equation: UNTITLED | ||||
Specification: logY_{i} C logTA logER logTR | ||||
LOGREXCR | ||||
Omitted Variables: Squares of fitted values | ||||
Value | df | Probability | ||
t-statistic | 2.601705 | 15 | 0.0200 | |
F-statistic | 6.768867 | (1, 15) | 0.0200 | |
Likelihood ratio | 7.821043 | 1 | 0.0052 | |
F-test summary: | ||||
Sum of Sq. | df | Mean Squares | ||
Test SSR | 0.013937 | 1 | 0.013937 | |
Restricted SSR | 0.044820 | 16 | 0.002801 | |
Unrestricted SSR | 0.030884 | 15 | 0.002059 | |
LR test summary: | ||||
Value | ||||
Restricted LogL | 34.77324 | |||
Unrestricted LogL | 38.68376 | |||
Unrestricted Test Equation: | ||||
Dependent Variable: logY_{i} | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:16 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 1873.726 | 753.7127 | 2.485995 | 0.0252 |
logTA | 0.965080 | 0.546796 | 1.764972 | 0.0979 |
logER | -1002.092 | 404.3045 | -2.478558 | 0.0256 |
logTR | -15.27961 | 6.166776 | -2.477731 | 0.0256 |
logRER | 16.28890 | 6.595674 | 2.469634 | 0.0260 |
FITTED^2 | 0.929677 | 0.357334 | 2.601702 | 0.0200 |
R-squared | 0.972588 | Mean dependent var | 11.26646 | |
Adjusted R-squared | 0.963450 | S.D. dependent var | 0.237344 | |
S.E. of regression | 0.045375 | Akaike info criterion | -3.112739 | |
Sum squared resid | 0.030884 | Schwarz criterion | -2.814304 | |
Log likelihood | 38.68376 | Hannan-Quinn criter. | -3.047971 | |
F-statistic | 106.4398 | Durbin-Watson stat | 1.787837 | |
Prob(F-statistic) | 0.000000 | |||
Decision rule: Reject the null hypothesis that the model is wronglyly specified if the p-value of t-stat. or F-stat. or Likelihood ratio test is greater than 0.05 level of significance.
P-value for T-stat. = 0.0200
P-value for T-stat. = 0.0200
P-value for Likelihood ratio test. = 0.0052
Decision making : The null hypothesis of for non-linearity was not rejected since the test statistic (t-statistic and F-statistic and likelihood ratio test) are statistically significant, that is the p-value from the results are lower than 0.05 level of significance.
Conclusion: the evidence base on our results indicated that the model is not linear and concluded that it may be wrongly specified. This result may be due to limited number of observations used in this study, thus recommend other researcher to increase their scope of study, since the model satisfied most of OLS assumptions.
Stability test is carried out to determine the appropriateness and stability of model for long run forecasting. This study used CUSUM residual test whether the model is suitable for making long run decision.
H_{0}: The model is not stable
H_{1} : The model is stable
Decision rule: Reject null hypothesis of stability if the plot of CUSUM for the model is not with the five percent critical bound, otherwise don’t reject H_{0. }
Figure 4.2 plot of CUSUM
Decision making: Based on the evidence from CUSUM the plots of CUSUM and CUSUMSQ statistics fall within the critical 5% bound for the period under consideration, thus we reject null hypothesis of non-stability.
Conclusion: Based on the evidence from CUSUM test, the model does not suffered structural break and thus concluded that estimated coefficients are stable are fit for long run decision making.
4.3 Conclusion
This chapter presents the estimated multiple regression results and interpretations using Eviews 10 software. The chapter begins with aprori expectation based on economic theory and existing literature on the nature of relationship between Y_{i}, TA, TR, RER and ER in Malaysia. The findings revealed that all the independent coefficients were correctly sign except for arrival which was wrongly signed. Although, its effect on economic growth of Malaysia was supported by existing studies on which will better be explained in the next chapter. The result of coefficient of determination(R^{2}) indicated that about 96 percent variation in economic growth proxied byY_{i} was explained by TA, TR, RER and ER while the F-statistic which measures the overall significance of the model reveals that TA, TR, RER and ER have significant effect on economic growth of Malaysia. The model was subjected to important assumptions of the Ordinary Least Square to validate whether the results obtained were good enough for decision making and it was concluded that the estimated model of economic growth of Malaysia is not spurious.
Chapter 5: Discussion, Conclusion and Implication
5.0 Introduction
This chapter presents the statistical analysis from the previous chapter of this study with the detail explanation of relationship that exist between Y_{i} and independent variables (TA, TR, RER and ER) in Malaysia. The results from the findings were analyzed critically in relation to a prior expectation, theoretical expectation and existing empirical studies. This was done so as to form a solid bedrock for proper policy recommendations that can be adopted as a guide for sustainable growth in Malaysia and the World at large. This section also present the implications of this research work with its contribution to existing literature in the field of economics generally and sustainable growth in particular. The limitations encountered in this research work was also presented as a research gap for new researcher to explore.
5.1 Summary
The study examines the significant effect of tourism income on economic growth in Malaysia using time series data of twenty-one years. The data was sourced from world development indicators on the various variables adopted in this study. The variables adopted in this study shown high volatility and unpredictable from the correllagram stationarity test at level, therefore, they were subjected to formal econometric testing on unit root using Philip Perron and Augmented Dickey Fuller test. The results from both methods at level indicated non-stationary of the series and to avoid obtaining spurious results, it was further subjected to testing by differencing the series where they all became stationary.
5.2 Discussions
Table 5.1: The expected and statistical results
Test | Result | Expected Relation | Major Findings |
Y_{i} and TR | Positive and Significant | Positive | Yalçinkaya & Daştan, (2018), Tugcu (2014), (Lean & Chong, 2014). |
Y_{i} and TA | Negative but insignificant | Positive | Podhorodecka, (2014), Massidda and Mattana (2013) |
Y_{i} and RER | Negative Not Significant | Negative | (Vieria & McDonald, 2016), (Habib, Mileva, & Stracca, 2016)
(Selimi & Selimi, 2017) |
Y_{i} and ER | Positive Significant | Positive | (Aynalem, Kassengn, & Sewnet, 2016), (Rokanuzzaman, 2019) (World Travel & Tourism Council, 2017). |
5.2 Major Findings
5.2.1 Arrival
The coefficient obtained from the regression results indicated that tourist’s arrivals into Malaysia has negative but insignificant effect on economic growth in Malaysia. This finding contradicts the findings of Podhorodecka, (2014) who found tourism arrivals as of the major determinants of economic growth of Cayman Island using time series econometric method with data spanning from 1983 to 2011. Although the insignificancy of the variable which lead to rejection of first hypothesis postulated in this study is consistent with the existing findings of (Massidda, 2013) who adopted Structural vector auto regressive model to capture both the long run and short run effect, the short run shows that arrivals has no significant impact on economic growth in the short run. Although this finding was contrary to apriori expectation that increase in the number of new arrival should shift the aggregate demand forward as a result of increase in consumer spending and quality business environment that can influence the new potential investors. The justification for negative effect of tourist arrival may be due to high supply of skilled labors from new arrivals competing with the local labor for employment and the insignificance may be due to institutional arrangement or the stage of development of the sector.
5.2.2 Employment Rate
The result from this finding reveals that employment rate has positive and significant effect on economic growth in Malaysia. The reason for positive relationship between Y_{i} and ER was supported by existing studies of the following scholars (Staniūtė, 2018) and World Travel & Tourism Council, (2017) with emphasis that increase in employment opportunities reduces social vices and create investment in an enabling environment with spilliover effect on the other macroeconomic variables in the economy. Belloumi (2010) who examines the relationship between tourism receipts, real effective exchange rate and economic growth concluded with positive relationship following the reason that the quantity of resources does not determine the level of development but the effective utilization and proper channelling determine how country will grow. The finding correlate with expected research result with the reason that most of jobs created through the development of tourism industry employ young people who fall within the age bracket and will act as stimulants to various sector of the economy because of efficiency. Also, the positive relationship between Y_{i} and ER implies that increase in the level of employment will reduces government expenditure on unemployment benefits and increase in foreign direct investment and all this will result to economic growth. Low level of unemployment strengthens labour union for proper negotiation between the parties involved in the market, reduction in hiring cost and equity in the labor market and this will create stability in policies that can increases the value of output in the country.
5.2.3 Receipts from Tourism Industry
The coefficient of Receipts from the result in the previous chapter indicated the existence of positive and significant relationship between receipts gained from tourism industry and economic growth in Malaysia. The finding of positive relationship between receipts and GDP is consistent with the findings of (Can, 2014), Lean & Chong (2014) ; Yalçinkaya & Daştan (2018), According to the study conducted by Yalçinkaya & Daştan, (2018), it was revealed that international tourism receipts have positive influence towards the growth of both developing and developed countries. Also, The finding of Boga & Erkisi, (2019) from the panel data regression of 483 observations within 1995 to 2017 from 21 Asia Pacific Economics confirmed that tourism sector is one of the predicators of economic growth as it constitute major part of foreign earning, employment of labor and new investment in the economy.
The previous statistic published by Malaysia data base revealed that revenue acquired from the tourism sector has surprisingly been dedicated to at least 15% of the overall Malaysia’s economic growth, as the tourism receipt acted as the catalyst to the exchange reserves, services export and balance of payment position of the country. Conclusion on the reason from positive relationship between Receipts generated from tourism industry on how it can influence economic growth positively is justified from the point of view that development of tourism sector attracts foreign direct investments with pressure on demand for domestic currency, this will not only boost investment level, employment opportunities and foreign capital inflows but also stimulate economic growth in a great measure.
5.2.4 Real Exchange Rate
The coefficient of real exchange rate indicated that it has negative but insignificant effect on economic growth in Malaysia. This finding contradicts the expected result and existing studies of Vieira (2013), Habib et al. (2017). Selimi, (2017) who found positive relationship between real exchange rate and economic growth. However, the study of Selimi, (2017) correlated with this findings. The negative effect implies that depreciation of Malaysia Ringgit will increase the cost of foreign input and thus affect the industries that are highly dependent on foreign inputs for survival and this will reduce the contribution of industrial sector to total output which is used as proxy for economic growth. The study conducted by (Rodrik, 2008) to determine the connection between exchange rate and economic growth justify this result with the reason that fluctuations in real exchange rate can cause disequilibrium which affect the confidence of foreign investors in the economy thus will result to negative effect on economic growth . This insignificance may be due to fluctuation or institutional arrangement in the country.
5.3 Policy Implications of the Study
Tourism industry in Malaysia is one of the fastest growing industrial sector in the country in terms of international receipts generated from the sector and new tourism arrival to harness resources in the country. Although, some researchers has focused on its significant effect on economic growth of Malaysia, while the larger proportion of the few that has researched on this topic only adopt tourism industrial sector output as a control variable. This effect does not capture the full description of what tourism industry and its contribution to general welfare and economic growth entails. In attempt to fill the gap left behind by existing studies, this study was conducted with the aim to reveal the effect of tourism industry on economic growth in Malaysia. The findings have not suggested just the expected outcome.
The finding reveals that expansion in tourism industry which generate higher revenue is associated with attraction of new potential investors both from abroad and within the economy through the creation of enabling business environments. Thus, it will induce positive relationship between receipts and economic growth. This finding supports existing empirical studies which implyies that an effective and efficient utilization of revenue generated from tourism industry will stimulate other macroeconomic variables such as employment opportunities, domestic saving, foreign reserve earnings etc and thus affect growth positively. This further implies that for Malaysia to experience sustainable economic growth, effective and efficient utilization of receipts generated from tourism industry and economic policies that can stimulate the rapid growth and development of tourism sector are non-negotiable.
The study supports the existing studies on the positive relationship between employment rate and economic growth with implication that higher employment rate in the country will shift the demand curve forward through reduction in government expenditures on idle economic resources such as unemployment benefits, security cost, and borrowing. Empirical results suggest the importance of correct policies management of revenue and employment of labor. The findings reveal that employment of human and other economic resources can possibly generate spillover effect through creation of a stable macroeconomic environment with greater effect on economic growth.
5.4 Limitations of the Study
This first limitation of this study is the failure of the assumption of multicollinearity between exogenous variables used for this study. The null hypothesis of no multicollinearity was rejected with conclusion that receipts and arrival are not directly proportional to GDP. However, the effect of the problem was not that serious on the overall model but has more effect on individual variable significance. This might have effect on some necessary policies that needed for long term growth.
The model also suffers the problem of linearity. The study used the Ramsey Reset linearity test to examine the correctness of the model and it was discovered that the model should not be expressed in linear form. This may be due to limited number of observations and omission of some important independent variable.
The scope of this study is limited which has resulted to failure of some assumptions of ordinary least square. This also has created impossibility of using some advanced econometric method like autogressive distributed lag modeling which helps to capture and solve the problem of endogeneity. Also, this has led to loss of degree of freedom.
The number of independent variables used in the model does not fully capture the model of economic growth, thus there is possibilities of inaccuracies in some of our decision.
5.5. Implications for Further Research
For further studies, model’s improvements should be considered, including additional variables or exchanging the existing ones. Such studies should increase the scope of the study either by increasing the number years of study or using panel data in order not to lose degree of freedom for best linear unbias estimate of coefficients. Further analysis could be intended to capture not only the direct effect of tourism industry on economic growth of Malaysia but also the indirect effect that is probably more realistic. For this purpose, the identification of macroeconomic variables through which tourism sectors affect economic growth indirectly could be performed using other advanced econometric method that can solve the problem of endogeneity in the model
5.5 Conclusion
Existing literature from the study of (Sharif & Lonik, 2014) regarded tourism industry as one of the leading service industries and is considered as one of the key predicators of economic growth in the world economy. A well-structured institutional framework and developed tourism industry act as economic catalyst for national and regional development through reduction of income inequality and increase foreign earnings, employment opportunities and contribute to social-economic growth of skilled labour and semi-skilled labor in the industry.
This research analyzed the significant effect of receipts from tourism industry, tourism arrivals, real exchange rate and employment rate of Malaysia.
The time series of twenty one years starting from 1998 to 2018 sourced from world development indicators (WDI) was used. The study adopted multiple linear regression called ordinary least square to esterblish the nature of relationship between the dependent variables (TR, TA, RER and ER) and the dependent variable (Y_{i}) using the log transformation of data for easy interpretations. The apriori expectation follows from existing empirical studies, this study assumes that Receipts, Arrival and Employment rate will have a positive and significant effect on economic growth proxy by Y_{i}.
The main objective of this study is to examine the significant effect of tourism industry and channels of transmission. Alongside the main aim of the study, other objectives were outlined for easy test of hypotheses. The study formulated four hypotheses in line with the objectives of this research in which two were rejected and while other two were accepted. The findings reveal that Receipt and Arrivals has positive and significant effect on economic growth in Malaysia with the following conclusions.
5.3 Recommendations
This study thus recommended the following policies based on the result from the findings
- The policy makers need to stabilize and diversify the employment of labor especially to the real sector of the economy than the administrative and other unproductive sector of the economy for better stimulation and effect on the growth of the economy in Malaysia. This can be achieved through the adoption of economic policies that are favorable to both labor and employers in terms of wages and other economic benefits and cost of employment, this will help to create equilibrium in the labor market.
- The availability of resources in any economy does not guarantee development but optimal utilization of resources does, therefore, the revenue generated from tourism industry should be monitored for effective utilization in the real sector of the economy that can generate addition value to the country as one of the key predicators of economic growth in Malaysia to avoid diversion through budgeting to elephant projects or unproductive expenditure. This will further guarantee development of infrastructural development in the country.
- The need for country to experience sustainable growth in the long run demands the effort of the government and other employers of labor to ensure strict monitoring on the immigrants with competitive skills entering the country as new tourist arrivals so as not to affect the growth and development of the country . But the implementation of economic policies that can scrutinize and allows complimentary skills immigrants into the economy will have significant effect on the tourism industry in particular and economy generally.
- The coefficient of receipts from tourism sector is positive and highly significant with implication that it acts as one of the major catalyst of economic growth in Malaysia. The study recommended that the revenue generated from tourism should be used for capital investment especially on human capital that can yield greater returns for both immediate and future generation for economic growth and development. Also building of vocational centers where semi-skilled labor can sharpen their skills for greater efficiency will induce the speed of growth in output.
APPENDICES
TABLE I: Multiple Regression Using Orinary Least Sqaure
Dependent Variable: LOGYI | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 15:44 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -86.41977 | 24.95316 | -3.463280 | 0.0032 |
LOGTA | -0.011041 | 0.463972 | -0.023797 | 0.9813 |
LogER | 49.40157 | 12.77377 | 3.867424 | 0.0014 |
logTR | 0.746605 | 0.339654 | 2.198133 | 0.0430 |
logRER | -0.795483 | 0.721343 | -1.102781 | 0.2864 |
R-squared | 0.960218 | Mean dependent var | 11.26646 | |
Adjusted R-squared | 0.950272 | S.D. dependent var | 0.237344 | |
S.E. of regression | 0.052927 | Akaike info criterion | -2.835547 | |
Sum squared resid | 0.044820 | Schwarz criterion | -2.586851 | |
Log likelihood | 34.77324 | Hannan-Quinn criter. | -2.781573 | |
F-statistic | 96.54712 | Durbin-Watson stat | 1.432460 | |
Prob(F-statistic) | 0.000000 | |||
TABLE II: Normality test
TABLE III –VI : MULTICOLLINEARITY TEST
TABLE III
Dependent Variable: logTA | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:34 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 3.563098 | 13.01531 | 0.273762 | 0.7876 |
logER | -2.643807 | 6.646475 | -0.397776 | 0.6957 |
logTR | 0.724625 | 0.025237 | 28.71305 | 0.0000 |
logRER | 0.896464 | 0.308076 | 2.909879 | 0.0098 |
R-squared | 0.983609 | Mean dependent var | 1.251651 | |
Adjusted R-squared | 0.980717 | S.D. dependent var | 0.199238 | |
S.E. of regression | 0.027667 | Akaike info criterion | -4.167510 | |
Sum squared resid | 0.013013 | Schwarz criterion | -3.968553 | |
Log likelihood | 47.75885 | Hannan-Quinn criter. | -4.124331 | |
F-statistic | 340.0555 | Durbin-Watson stat | 1.379822 | |
Prob(F-statistic) | 0.000000 | |||
TABLE IV
Dependent Variable: LOGER | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:35 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 1.950535 | 0.025960 | 75.13714 | 0.0000 |
LOGTA | -0.003488 | 0.008769 | -0.397776 | 0.6957 |
LOGRECEIPTS | 0.003716 | 0.006386 | 0.581866 | 0.5683 |
LOGREXCR | 0.016778 | 0.013078 | 1.282984 | 0.2167 |
R-squared | 0.136102 | Mean dependent var | 1.985374 | |
Adjusted R-squared | -0.016350 | S.D. dependent var | 0.000997 | |
S.E. of regression | 0.001005 | Akaike info criterion | -10.79816 | |
Sum squared resid | 1.72E-05 | Schwarz criterion | -10.59920 | |
Log likelihood | 117.3807 | Hannan-Quinn criter. | -10.75498 | |
F-statistic | 0.892752 | Durbin-Watson stat | 1.310494 | |
Prob(F-statistic) | 0.464965 | |||
TABLE V
Dependent Variable: LOGRECEIPTS | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:37 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -7.891503 | 17.71512 | -0.445467 | 0.6616 |
LOGER | 5.255315 | 9.031824 | 0.581866 | 0.5683 |
LOGTA | 1.352143 | 0.047092 | 28.71305 | 0.0000 |
LOGREXCR | -1.328396 | 0.401886 | -3.305406 | 0.0042 |
R-squared | 0.984969 | Mean dependent var | 1.601938 | |
Adjusted R-squared | 0.982317 | S.D. dependent var | 0.284208 | |
S.E. of regression | 0.037793 | Akaike info criterion | -3.543719 | |
Sum squared resid | 0.024282 | Schwarz criterion | -3.344762 | |
Log likelihood | 41.20905 | Hannan-Quinn criter. | -3.500540 | |
F-statistic | 371.3401 | Durbin-Watson stat | 1.404582 | |
Prob(F-statistic) | 0.000000 | |||
TABLE VI
Dependent Variable: LOGREXCR | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:37 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | -8.456451 | 8.135396 | -1.039464 | 0.3131 |
LOGRECEIPTS | -0.294522 | 0.089103 | -3.305406 | 0.0042 |
LOGTA | 0.370879 | 0.127455 | 2.909879 | 0.0098 |
LOGER | 5.261445 | 4.100944 | 1.282984 | 0.2167 |
R-squared | 0.497107 | Mean dependent var | 1.981891 | |
Adjusted R-squared | 0.408362 | S.D. dependent var | 0.023136 | |
S.E. of regression | 0.017796 | Akaike info criterion | -5.050092 | |
Sum squared resid | 0.005384 | Schwarz criterion | -4.851136 | |
Log likelihood | 57.02597 | Hannan-Quinn criter. | -5.006913 | |
F-statistic | 5.601480 | Durbin-Watson stat | 1.195640 | |
Prob(F-statistic) | 0.007378 | |||
TABLE VII: HETEROSKEDASTICITY TEST
Heteroskedasticity Test: ARCH | ||||
F-statistic | 0.687352 | Prob. F(1,18) | 0.4179 | |
Obs*R-squared | 0.735633 | Prob. Chi-Square(1) | 0.3911 | |
Test Equation: | ||||
Dependent Variable: RESID^2 | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 15:57 | ||||
Sample (adjusted): 1999 2018 | ||||
Included observations: 20 after adjustments | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 0.001637 | 0.000680 | 2.405414 | 0.0271 |
RESID^2(-1) | 0.189871 | 0.229017 | 0.829067 | 0.4179 |
R-squared | 0.036782 | Mean dependent var | 0.002052 | |
Adjusted R-squared | -0.016730 | S.D. dependent var | 0.002040 | |
S.E. of regression | 0.002057 | Akaike info criterion | -9.440079 | |
Sum squared resid | 7.62E-05 | Schwarz criterion | -9.340506 | |
Log likelihood | 96.40079 | Hannan-Quinn criter. | -9.420641 | |
F-statistic | 0.687352 | Durbin-Watson stat | 1.969443 | |
Prob(F-statistic) | 0.417926 | |||
TABLE VIII : Autocorrelation Test
Date: 07/16/20 Time: 15:52 | ||||||
Sample: 1998 2018 | ||||||
Included observations: 21 | ||||||
Autocorrelation | Partial Correlation | AC | PAC | Q-Stat | Prob | |
. |**. | | . |**. | | 1 | 0.230 | 0.230 | 1.2766 | 0.259 |
. *| . | | . *| . | | 2 | -0.066 | -0.126 | 1.3890 | 0.499 |
.**| . | | . *| . | | 3 | -0.219 | -0.185 | 2.6754 | 0.444 |
. *| . | | . *| . | | 4 | -0.181 | -0.103 | 3.6107 | 0.461 |
. |* . | | . |* . | | 5 | 0.102 | 0.152 | 3.9223 | 0.561 |
TABLE IX : SERIAL CORRELATION
Breusch-Godfrey Serial Correlation LM Test: | ||||
F-statistic | 0.696550 | Prob. F(2,14) | 0.5148 | |
Obs*R-squared | 1.900532 | Prob. Chi-Square(2) | 0.3866 | |
Test Equation: | ||||
Dependent Variable: RESID | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 15:54 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Presample missing value lagged residuals set to zero. | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 16.28861 | 42.41035 | 0.384072 | 0.7067 |
LOGTA | -0.301656 | 0.578397 | -0.521538 | 0.6101 |
LOGER | -8.541245 | 21.73062 | -0.393051 | 0.7002 |
LOGRECEIPTS | 0.225755 | 0.427058 | 0.528630 | 0.6053 |
LOGREXCR | 0.345854 | 0.819175 | 0.422197 | 0.6793 |
RESID(-1) | 0.385318 | 0.344900 | 1.117190 | 0.2827 |
RESID(-2) | -0.028759 | 0.385456 | -0.074610 | 0.9416 |
R-squared | 0.090502 | Mean dependent var | -8.88E-16 | |
Adjusted R-squared | -0.299284 | S.D. dependent var | 0.047339 | |
S.E. of regression | 0.053960 | Akaike info criterion | -2.739933 | |
Sum squared resid | 0.040764 | Schwarz criterion | -2.391758 | |
Log likelihood | 35.76929 | Hannan-Quinn criter. | -2.664370 | |
F-statistic | 0.232183 | Durbin-Watson stat | 1.930072 | |
Prob(F-statistic) | 0.958938 | |||
TABLE X : SPECIFICATION BIAS (LINEARITY TEST)
Ramsey RESET Test | ||||
Equation: UNTITLED | ||||
Specification: LOGYI C LOGTA LOGER LOGRECEIPTS | ||||
LOGREXCR | ||||
Omitted Variables: Squares of fitted values | ||||
Value | df | Probability | ||
t-statistic | 2.601705 | 15 | 0.0200 | |
F-statistic | 6.768867 | (1, 15) | 0.0200 | |
Likelihood ratio | 7.821043 | 1 | 0.0052 | |
F-test summary: | ||||
Sum of Sq. | df | Mean Squares | ||
Test SSR | 0.013937 | 1 | 0.013937 | |
Restricted SSR | 0.044820 | 16 | 0.002801 | |
Unrestricted SSR | 0.030884 | 15 | 0.002059 | |
LR test summary: | ||||
Value | ||||
Restricted LogL | 34.77324 | |||
Unrestricted LogL | 38.68376 | |||
Unrestricted Test Equation: | ||||
Dependent Variable: LOGYI | ||||
Method: Least Squares | ||||
Date: 07/16/20 Time: 16:16 | ||||
Sample: 1998 2018 | ||||
Included observations: 21 | ||||
Variable | Coefficient | Std. Error | t-Statistic | Prob. |
C | 1873.726 | 753.7127 | 2.485995 | 0.0252 |
LOGTA | 0.965080 | 0.546796 | 1.764972 | 0.0979 |
LOGER | -1002.092 | 404.3045 | -2.478558 | 0.0256 |
LOGRECEIPTS | -15.27961 | 6.166776 | -2.477731 | 0.0256 |
LOGREXCR | 16.28890 | 6.595674 | 2.469634 | 0.0260 |
FITTED^2 | 0.929677 | 0.357334 | 2.601702 | 0.0200 |
R-squared | 0.972588 | Mean dependent var | 11.26646 | |
Adjusted R-squared | 0.963450 | S.D. dependent var | 0.237344 | |
S.E. of regression | 0.045375 | Akaike info criterion | -3.112739 | |
Sum squared resid | 0.030884 | Schwarz criterion | -2.814304 | |
Log likelihood | 38.68376 | Hannan-Quinn criter. | -3.047971 | |
F-statistic | 106.4398 | Durbin-Watson stat | 1.787837 | |
Prob(F-statistic) | 0.000000 | |||