Preparing Database for a Tourism CGE Model
A CGE model needs a large amount of data. What kinds of data are required? Where can we obtain these data? How are they related to the CGE model structure? This chapter provides answers to these questions. Using the tourism CGE model for Singapore, this c
- PDF / 292,782 Bytes
- 37 Pages / 419.527 x 595.276 pts Page_size
- 5 Downloads / 194 Views
The chapter is structured as follows: Section 6.1 generally describes the data requirement of the model while Sect. 6.2 addresses data availability and sources. In Sect. 6.3, we check the required data account by account so as to validate the feasibility of the model. In some model accounts, the treatment of the relevant data will be discussed in detail. Since the various behavioural parameters used throughout the model are vital to the functions and thus the simulation results, they are discussed specifically in Sect. 6.4.
6.1
Data Requirements
Based on the theoretical structure of the CGE model displayed in the previous chapter, the data required by the model are I–O data, coefficients, and parameters (such as various elasticities and shares). Among them, the I–O data is the major part of database.
© The Author(s) 2017 S. Meng, M. Siriwardana, Assessing the Economic Impact of Tourism, DOI 10.1007/978-3-319-40328-1_6
211
212
Assessing the Economic Impact of Tourism
6.1.1 I–O Data The I–O data needed in our model are displayed in an absorption matrix shown in Tables 6.1, 6.2, and 6.3. Table 6.1 is a schematic representation of the absorption matrix for our model, which is a variation of the standard database structure used in the ORANI-type models. Each row heading indicates different inputs (or costs) for various industries. The column headings show different agents involved in the economy, which includes I producers, I investors, H household groups, T tourism demands, one aggregate foreign demand, one government demand, and the changes in inventory. According to the number of commodities, sources, occupation types, and capital type, the first two columns of the table show the cost of each industry result from buying C*S kinds of commodities, paying C*S taxes, and employing O types of labour and 1 type capital. Table 6.1 requires considerable data. For example, the vector V1BAS, V2BAS, V1TAX, and V2TAX each consists C*S*I data. 2005 Singapore I–O tables have 136 kinds of commodities and industries. Considering that each commodity can be obtained locally or imported from overseas, each of the four vectors will contain 136*2*136 ¼ 36,992 data entries. In order to reduce the model size and emphasize the purpose of this study, we will aggregate relevant commodities and industries according to the characteristics of Singaporean economy and tourism industry. Since some industries produce more than one commodity (and one commodity may be produced by more than one industry), we need a Make matrix (see Table 6.2) to describe this information. Moreover, since import duties are assumed to be levied according to the commodity types instead of the categories of users, we need a tariff vector (shown in Table 6.3).
6.1.2 Other Data Besides the I–O data displayed above, economic data from various sources are also necessary. These data can be put into two categories. One is the data affecting the balance in the household account, investment account,
Sales by commodity
Source: Authors
Capital
Labour
Production taxes
P
Data Loading...
