Research Framework

By adopting the LPI measurement framework from World Bank LPI report where the components of LPI measurement were chosen based on recent theoretical, empirical research and on the practical experience of logistics professionals involved in international freight forwarding (Arvis, et al. 2012). Korpela and Tuominen (1996) stated that benchmarking company logistics operations is based on five logistic critical success factors such as Reliability, Flexibility, Lead time, cost effectiveness, and value added. Goh and Ling (2003) also describe in their research that transportation, telecommunication infrastructure, customs regulation, and warehousing are vital aspect to logistics development in China.

The figure below maps the six LPI indicators into two main categories namely input and outcomes category. The first category called input category is the area for policy regulations indicating main inputs to the supply chain which consist of customs, infrastructure and quality of logistics services. The second category is a service delivery performance to indicate the outcomes which is represented by the timeliness, international shipments, tracking and tracing. This represents the three main indicators for logistics performance which is time, cost, and reliability.

In this study, first, all the indicator from three influencing factors that already filtered based on theory, previous study, and also the completeness of the dataset are used to predict six LPI components (customs, infrastructure, and logistics competence, international shipping, timeliness, and tracking and tracing), beside to predict the value of LPI, this also to analyze

the relation between the indicators from influencing factors to LPI components. After all of the indicators for all LPI component identified, and also the relation between indicators and LPI components, then a study case analysis were conduct to further analyze the effect of each indicators to country logistics performance and to improve the logistics quality based on the indicators.

3.2 Factors identification

The factors that indicates has relation to logistics performance by referring to any related literature from previous study are proposed. All Factors and their reference in detail are shown in Table 2 then Table 3 describes the indicators that are selected in the model.

Table 2 Factors and indicators From Study Literature

Factors Possible set of Indicators Reference Economic -GDP per capita

-Export value Prater (2002), Kollurua and Ponnamb (2009)

Infrastructure -Road density -Road fatality Ling (2003), Hausman et al.

(2005), Arvis et al. (2010), Zhang and Figliozi (2010), Vijayaraghavan (2007)

The factors above are the indicators from the previous study that used to predicts the countries logistics. But, several of these indicators turned out to be highly correlated with one another and some of them have very sparse data. Based on a preliminary assessment of the indicators, a smaller subset was selected for inclusion in the analysis and in line with the SUR

model framework. These indicators from three factors that are used as independent variables are rationally related to the logistics in a country. Further discussion on the rational and reasons of these factors chosen are described below:

Table 3 Indicators that included into the model

Independent Variables CS IF LC IS TT TM

Economic

GDP per capita √ √ √

Cost to export √ √ √ √

Time to export √ √ √ √

Transport services export

value √ √ √ √

Social & Government

Document to export √ √

Unemployment number √

Control of corruption √ Custom clearance time

without physical inspect √

Infrastructure

Fixed broadband internet subs √

Road fatality √

Internet users/100 people √ √ √ √

Air transport freight √ √ √

Quality of port √ √ √ √

Mobile phone subscriptions √

Note: CS = LPI customs; IF = LPI infrastructure; LC = LPI logistic competence; IS = LPI International Shipment; TT= LPI tracking & Tracing; TM = LPI Timeliness

Comprehensive description on the motive behind choosing these indicators and their relation to logistics performance are explained below:

Economic Factors

• GDP per capita

In Gunner, et al. (2012) and Puertas, et al. (2013) study, GDP is used as a parameter in some study about Logistics. Gunner, et al. (2012) shows that GDP as one of the components of economic factor is significantly correlated with logistics performance

even though the coefficient of correlation is small. Puertas, et al. (2013) use GDP as one of the parameters along with LPI and others as factors affecting a countries trade flows. The rational reason to explain this variable is GDP which represent the investment, trade flow and consumption of the country. High GDP is as a result of some factor such as high investment, high consumption, high government spending and high surplus trade balance. The first and fourth factors are related to the country logistics. The investment on logistic sector will improve the countries logistics performance while the high value of trade between countries indicates the goodness of logistics quality of a country

• Time to Export

Time to export is the time needed to export goods measure in days. The time calculation for a procedure starts from the moment it is initiated and runs until it is completed. The relation between this variable to logistics performance is clear if a country have less time to export value. This means that the ease and the efficiency of that country are better and vice versa. Arvis et al. (2010) shows that time is one of the key element to measure logistics performance by proposing timeliness as the measurement indicator of service time.

• Cost to Export per Container

Cost to export represents all the fees associated with completing the procedures to export or import the goods. These include costs for documents, administrative fees for customs clearance, technical control, customs broker fees, terminal handling charges and inland transport. Hausman et al. (2005) states that the outcome on the use of cost as an indicator to measure logistics index shows the relations between cost and logistics performance appearing to be negative. This means that the lower the cost to export the more efficient the logistics performance.

• Transport service export value

This indicator represents the transport service exported in a country, reflects on how the logistics company works in the transportation services. Lai, et al (2002) measure the supply chain performance in transport logistics which focus on the efficiency and effectiveness of logistics or transport service providers. The logistics performance will be positive when this indicator has higher export value and high LPI score as well.

Social Factors

• Control of Corruption

Gunner, et al. (2012) shows that political risk being one of the indicators to represent social factor has a strong correlation to logistics performance which is represented by LPI. This study shows that not only economic and infrastructure factor affects countries LPI but also the social factor. Control of corruption is one of the indicators of political risk index. As a rational reason, the control of corruption in a country represents how well conducive the business environment and the ease of business regulation in that country would become. A good score of this indicator means the government supports it and therefore will ease the smooth running of an efficient logistics activity.

• Document to export

This indicator shows that all the number of documents required per shipment to export goods are recorded. This indicator reflect the ease of regulation and also the efficiency of customs in that country. The more the number of document required for export, the more inefficient the logistics performance become thus the relation between these indicators with logistics performance becomes negative. Arvis et al. (2010) in their research used this indicator as one of the assessment criteria to measure the effectiveness of customs policy in a country.

• Unemployment number of total labor

This indicator represents the number of unemployment compared to total labor in a country. The number of unemployment might be represented as the socio - economic condition of a country. The higher number of unemployment indicates that the business environment is not conducive because people will tend to get a job from anywhere especially in the informal sector and this can trigger illegal charges in the business activity in terms of logistics and as well might increase the cost. Kaliba et al.

(2012) show that unconducive macro business environment acts as a negative input to the business process and logistics is one of the main activity in doing business.

• Customs clearance time without physical inspection

This indicator measures the average time to pass goods through customs so that they can enter or leave the country. A document given by customs to a shipper shows that the customs duty has been paid and the goods can be shipped without physically inspecting the goods. Arvis et al. (2010) in their research used this indicator as one of the assessment criteria to customs efficiency and logistics service time.

Infrastructure Factors

• Road fatalities

The road fatalities is one of the indicators for logistics infrastructure as it shows the number of road accidents which ends up with fatal injury or death of the victims. Even though the number only represents a small percentage of total road accident, but yet still this data can be used as surrogate to represent that. When accident happens, it will cause congestion in the road where the congestion will extend the time and cost of travel thereby affecting the countries LPI as a domino effect. Therefore, high rate of road fatalities in a country shows that it has a poor infrastructure and regulation, hence leads to congestion, uncertainty and an increase of travel time. It is also mentioned by Goh and Ang, (2000) who made a study about the reality of logistics in Indochina area that for least developed and developing country, one of the greatest obstacle to logistics growth is the poor state of their basic infrastructure. There are several aspects which represent logistics infrastructure like airport, port, and road transportation.

• Quality of port

Quality of port represents the country’s port infrastructure quality level and it is clear that the quality of port infrastructure indication is determined by its logistics performance. Port as one of a component of logistics infrastructure as Goh and Ang (2000) mentioned in their research that it is a vital aspect to logistics developments. As for the least and developing countries, their main obstacles to logistic developments is the basic infrastructure. The rationale behind this reasoning is the about 90% shipment of goods in the world uses ships and therefore the better the quality of the port infrastructure, the more efficient and faster the logistics activity becomes. The quality of Port infrastructure therefore is an important entity for a country if it wants to improve its logistic performance.

• Air transport freight

Based on Goh and Ling (2003) who describe that air transport is one of the factors to country logistics development. This indicator shows the volume of freight, express, and diplomatic bags carried on each flight stage (operation of an aircraft from takeoff to its next landing), measured in metric tons by kilometers traveled. It reflects the air transportation of a country and the higher number of this indicator shows that the air transport performance of that country is good.

• Internet users/100 people

In this information and technology era, the role of internet becomes vital for economic development in general and logistics performance in particular in terms of global supply chain where the logistics activity is done and connected from different countries around the world. The indicator reflects the quality of telecommunication infrastructure. Higher ratio of internet users in one country is an indication of a good telecommunication infrastructure and logistics performance. Goh and Ling (2003) also describe that telecommunication infrastructure has a role to play in logistics developments in China.

• Fixed broadband internet subscriber

Similar with internet user, this indicators also represents the quality of telecommunication infrastructure but tends to be more specific to fixed broadband service for logistics activity especially in global supply chain for connectivity purpose.

It is necessary to have a broadband connection to build a reliable connectivity. Based on Goh and Ang (2000) telecommunication infrastructure is necessary to logistics development in developing country in south china region. The relation between this indicator and the logistics performance is positive meaning more number of this indicators will affect the increment of logistics performance.

• Mobile telephone subscriptions/100ppl

This indicator also reflects the telecommunication infrastructure by supporting logistics activities especially in global supply chain in order to have a reliable connectivity. Both studies from Goh and Ang (2000) and Goh and Ling (2003) emphasize the role of telecommunication infrastructure in logistics sector development.

The relation between this indicator and the logistics performance is positive, meaning the more number of this indicators will affect the increment of logistics performance.

3.3 Data Collection

The data used in this study is a secondary data and obtained indirectly through an intermediary medium. Secondary data is usually in the form of historical reports that have been arranged in the archives as (documentary data) either published or unpublished. This study only uses secondary data which can be either the index or the value of certain aspects that describes the condition of a country.

In order to build a comprehensive dataset to identify the factors that will be used to predict the LPI score would require a data collection of several sources such as:

 LPI score from year 2007, 2010, 2012, and 2014

 World development index (WDI) developed by the World Bank

− Economic sector (GDP, Growth rate, Export-Import volume, etc)

− Infrastructure sector (Road density, Road fatalities, internet users, quality of port, etc.)

− Private sector (lead-time to export/import, number of document, number of tax payment, etc.)

 Political Risks Index (PRS)

 Global Competitiveness Index (GCI)

Figure 2 Country included in this study

The population in this study is the country which has LPI score and the total population consist of around 150 countries (see Figure 2). The picture of figure 2 below depicts the countries included in the research marked in area as red. Data is key to the determine the inclusion of a country in the study and for most of the African countries and the underdeveloped ones are being excluded due to lack of the availability of data. This also applies to countries in conflicts in order to avoid the outlier dataset being included in the model. As a result, a complete dataset consisting of selected 42 countries representing developing and developed countries from each continent is proposed.

CHAPTER 4. ESTIMATION RESULT AND DATA ANALYSIS

4.1 LPI Component Prediction

As proposed in the previous chapter that once the indicators have been identified, and the model to predict LPI component are conducted, the result of this SUR model is showed in Table 3. Each equation in this model have 168 observation from 42 country sample multiply with four years of observation data. The table below shows that almost all of the indicators are statistically significant to their dependent variables, the un-significant indicators are document to export on LPI customs, transport service export value on LPI logistic competence, and mobile phone subscription on LPI tracking and tracing.

There is some assumption need to fulfill when apply SUR model, one of the assumption is, there should be no autocorrelation in the equation, therefore the Durbin-Watson test is conduct and the result shows there is no autocorrelation in the equations Taking from Durbin Watsons test results which was conducted earlier on reveals the nonexistence of autocorrelation in the equation. This is because the D-W value for LPI customs, LPI Infrastructure and LPI logistics competence are 2.135, 2.025, and 1.903 respectively. The value of upper bond (du) and lower bond (dl) for n=168 and k=6 are 1.550 and 1.803 respectively. Therefore, there is no autocorrelation in these three models and the correlation residuals for each models also shows that error correlation between equations are exist.

GDP per capita and cost to export are two indicator that includes in all of three LPI input category, GDP per capita have a quite constant coefficient for each dependent variables which means that the effect of this indicator is slightly the same to LPI input category components. Meanwhile, cost to export have a quite different coefficient value as shown in Table 3. This indicator affecting LPI customs more than the other dependent variables, with coefficient (-0.492), it indicates the cost reduction or improvement on policy might affect more to LPI customs, than infrastructure and logistics competence.

Table 4 SUR model result for LPI component

Control of corruption 0.545***

(0.115) Note: Standard error are shown in parentheses.

LPI custom: observation = 168; log likelihood = -4.919 ; D-W = 2.135;

LPI Infrastructure: observation = 168; log likelihood = 1.6821; D-W = 2.025.

LPI logistics comp: observation = 168; log likelihood = 7.846; D-W = 1.903.

LPI int. shipping: observation = 168; log likelihood = 21.662; ; D-W = 1.7367 *** p < 0.01 LPI tracking: observation = 168; log likelihood = -5.9088; ; D-W = 1.8018 ** p < 0.05 LPI timeliness: observation = 168; log likelihood = 9.5920; D-W = 1.7645; * p < 0.1

Transport service export value and internet users are two indicator that includes in all of three LPI outcomes category, it is can be explained because LPI outcomes category represents logistics service performance and transport service export is the indicator that reflect the logistics service in that country because the main activity of logistics is transferring goods. This indicators is affecting each LPI component in positive sign and tend to have same coefficient. While internet users is affecting LPI tracking more than the other LPI component, this explain that the better internet network coverage will help country to increase their logistics quality in terms of tracking and tracing quality.

The comprehensive analysis of each LPI component and how the indicators affecting it is described below, while the detail information about the coefficient and standard error for each indicator can be seen on Table 4. The descriptions are:

• LPI customs

Indicators that affecting this LPI component are GDP per capita, time to export, quality of port infrastructure, and control of corruption. GDP per capita represent the average income. In terms of customs quality, high GDP shows the goodness of the trade flow in the country supported by an efficient custom policy. Time to export indicator measures the time from which an export procedure starts from the moment it is initiated and runs until it is completed including the time related to customs procedures. The indicator might as well reflects the efficiency of customs procedures.

Quality of port deals with the port infrastructure quality which is also necessary to support the customs regulation and a well develop infrastructure will enhance the possibility of applying an efficient and flexible custom regulations. Cost to export indicator (cost in the indicator) includes all of the cost related to custom administration and therefore time and cost are key factors that represents the customs efficiency. For social aspect, control of corruption is also highly related. In many developing countries and for which their customs efficiency is hampered by widespread corruption thereby creating a major obstacle to trade expansion and logistics performance.

• LPI Infrastructure,

The indicators that affecting it are road fatalities, GDP per capita, quality of port, cost to export, internet users/100 people, and air transport freight. Road fatalities indicator represents the land transport infrastructure quality. Quality of port infrastructure is measure to logistics infrastructure quality for sea port where most of the international trading activities are conducted, while air transport freight might reflect the air transport infrastructure quality, and Internet users/100 people shows the information and communication technology