. gawk
33%;»?

, H14

 

..

m...» . .

.3. .53.“.
.5

. ‘5

L.» 3:
1.521 :NWPuqum...
t.» mam...”

p.>\
2 v3.3.

La.
P.

9 .

Lu. . 3d". .
a...

5M9.

P.

I 2...
1. .
1

 

.‘I‘n..|
I»... .4 PH}

.3. a

nil... 9:21
. ISP‘H-stq1
‘fi
" 3 ‘I‘

22.4
. :wzuvnim...
\ a... . .n. H.

 

. .

unhlfl
m.

1. 5.3qu
In“ .

i .

l‘\;'u"h‘v

 

K?» : n. r

 

-..,S,........m...fi.i;~. . , , . , .2, .
.3 ,7 :5. .. , . , . r.” Jfimfizvmww

ENE”

. _." ‘2’

13 LIBRARIES
E . MICHIGAN STATE UNIVERSITY
1?» ‘ . EAST LANSING, MICH 48824-1048

This is to certify that the
dissertation entitled

AN INPUT-OUTPUT MODEL WITH VARYING JOB AND
INCOME RATIOS FOR SERVICE INDUSTRIES: AN
APPLICATION OF THE “DOUBLING TOURIST ARRIVALS
PLAN" IN TAIWAN

presented by

Ya-Yen Sun

has been accepted towards fulfillment
of the requirements for the

 

 

 

PhD. degree in Park, Recreation, and Tourism
Resources
TQM/W
“’7 MajJr PrBfeSsfi’s Signature

 

@w/ 2.7! 2 0°?”

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

4-.-.---o-o-.-o-c-a-o-n-s-o—o-o-c-o-o-u-o-o-o-.-o-o-a-c-o-u-.-o-o-v-o-o-.-—-—~—--u-o-o-a-c-o-o-o-o—o—o-u-c-¢-o-—o-o--.-a-

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECAUID with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

@599 “03907

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:/ClRC/DateDue.indd-p.15

 

AN INPUT-OUTPUT MODEL WITH VARYING JOB AND INCOME
RATIOS FOR SERVICE INDUSTRIES: AN APPLICATION OF THE
“DOUBLING TOURIST ARRIVALS PLAN” IN TAIWAN

By

Ya-Yen Sun

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Park, Recreation, and Tourism Resources

2005

ABSTRACT

AN INPUT-OUTPUT MODEL WITH VARYING JOB AND INCOME RATIOS FOR
SERVICE INDUSTRIES: AN APPLICATION OF THE “DOUBLING TOURIST
ARRIVALS PLAN” IN TAIWAN
By

Ya-Yen Sun

An Input-Output model that allows the value added components to vary with
respect to capacity utilization is developed for service industries. The assumption of
constant ratios of jobs and income in the standard [-0 framework is tested by examining
firm-level time-series data of tourist hotels in Taiwan from 1999 to 2003. A national
input-output model with fixed and varying economic ratios and multipliers are
established to evaluate the economic impacts of the Taiwan tourism policy, “Doubling
Tourists Arrival‘s Plan” (DTAP). under three scenarios. Each scenario specifies distinct
international visitor volume and hotel capacity. simulating alternative conditions of
demand and supply of tourism activities in Taiwan.

Based on empirical data. jobs to sales ratios and personal income to sales ratios
are not constant with respect to capacity utilization. Hotel businesses with higher
occupancy rates or more rooms support fewer jobs per dollar of sales and allocate lower
percentages of revenue to wages and salaries. A five percent increase in occupancy rates,

for example, leads to a 7% decrease in jobs to sales ratios and 1.5% reduction in income

to sales ratios. Given a rise in occupancy rates, the proposed l-O model generates lower
type I and type 11 job and income multipliers.

Inbound tourism contributed 0.76% of Gross Domestic Product (GDP) and 1.10%
total employment in 2001 in direct effects. Projected changes in tourism demand and
hotel capacity in the DTAP policy would increase foreign receipts to Taiwan from the
current US$35 billion in 2001 to a range of US$52 billion to $6.5 billion by 2008. The
increased tourism would support 142 ~ 190 thousand jobs and $1.89 ~ $2.51 billion
personal income by 2008. The contribution of foreign receipts is expected to grow to
0.87% ~ 1.16% GDP by 2008. assuming a 3% national GDP growth rate in Taiwan.

Relaxing jobs and income ratios in the 1-0 model introduces a non-linearity
between impact estimates and visitor spending. Economic impacts are determined
simultaneously by final demand changes as well as the level of capacity utilization in the
accommodation sector. Across three scenarios, total projected job estimates for the hotel
sector differ from 12% to 30% between 1-0 models with fixed and varyingjobs to sales
ratios. Estimates for total personal income in the hotel sector, on the other hand, are more
stable with differences ranging from 2% to 7%.

The assumption of linearity and constant coefficient in the production system
generally leads to overestimation in the secondary effects, especially with respect to
income and employment. The proposed I-O model with value added components varying
with capacity utilization allows changing returns to scale. and provides more accurate

assessments Of jobs and personal income for service industries.

ACKNOWLEDGEMENTS

First and foremost. I would like to express my thanks and appreciation to my
Ph.D. advisor, Dr. Daniel Stynes. for his support and advice. In addition to his patience
and insights throughout the dissertation process. he has inspired me greatly on the
systematic approaches to and logical reasoning within research. He has also been a great
role model for me. as true scholars are.

I would also like to expand my appreciation to my committee members: Dr.
Dennis Propost for his continuous encouragement during my Ph.D. program; Dr. Bruce
Pigozzi for providing suggestions on the technical aspects Of the analysis and teaching
me to be confident; and Dr. Larry Leefcrs for providing viewpoints that go outside the
box. Each committee member provided valuable insights in shaping this manuscript.

1 would also like to thank the many friends I have made at Michigan State
University for their love. care and countless laughs that we shared together. Last, I would
like to express my gratitude to Ariel Rodriguez. little brother Andrew, and my parents for
their unconditional love, understanding. and supports. They accompany me throughout
these years. and their love sculpture part of who I am. Without them. and all the people

above, my graduate study at Michigan State University would not have been possible.

TABLE OF CONTENTS

 

 

 

ACKNOWLEDGEMENTS ........................................................................................... IV
LIST OF TABLES ....................................................................................................... VIII
LIST OF FIGURES ......................................................................................................... X
CHAPTER I INTRODUCTION ................................................................................... 1
Tourism Policy in Taiwan ............................................................................................ 1
Problem Statement ....... - - -_ -_ ........ 3
Conceptual Framework ................................................................................................ 4
Objectives .................................................................................................................... 10
Assumptions -- ...... - ......................... - - -- 13
Limitations ................................................................................................................... 14
Definition of Terms ..................................................................................................... 15
CHAPTER II LITERATURE REVIEW ........ . 17
Tourism Development in Taiwan .............................................................................. 17
Domestic and Inbound Tourism ................................................................................ 17
Doubling Tourist Arrivals Plan (DTAP) ................................................................... 23
Taiwan Tourism Economic Impact Studies .............................................................. 27
Tourism Economic Impact Estimation Model ......................................................... 30
Tourism Satellite Accounts ....................................................................................... 30
Tourism Multipliers .................................................................................................. 31
Computable General Equilibrium ............................................................................. 32
Input-Output Analysis ............................................................................................... 34
Operation of Input-Output Model ............................................................................. 35
Capacity Utilization and Tourism Economic Impact Analysis .............................. 38
Price Effects .............................................................................................................. 38
Economics of Scale and Substitutions ...................................................................... 40
Capacity Constraints ................................................................................................. 42
Summary ................................................................................................................... 43
CHAPTER III METHODS ........................................................................................... 45
Objective I ................................................................................................................... 46
Average Visitor Spending ......................................................................................... 46

Total Spending ofInternational Visitors ................................................................... 47

Taiwan Input-Output Model ..................................................................................... 48
Total Economic Impacts ........................................................................................... 49
Objective II .................................................................................................................. 50
Econometric Relationships ....................................................................................... 50
Predicted Economic Ratios and Multipliers ............................................................. 54
Objective III ................................................................................................................ 55
Ratio Prediction Model ............................................................................................. 57
Input-Output Model .................................................................................................. 59
CHAPTER IV RESULTS ............................................................................................. 61
Objective I ................................................................................................................... 61
Visitor Spending of Inbound Tourism ...................................................................... 62
Taiwan Input-Output Model ..................................................................................... 63
Economic Impacts of Inbound Tourism, 2001 ......................................................... 67
Objective II .................................................................................................................. 70
Job to Sales Ratio ...................................................................................................... 70
Personal Income to Sales Ratio ................................................................................. 7]
Predicted I-O Multipliers .......................................................................................... 73
Objective III ................................................................................................................ 76
Occupancy Rates by Scenario ................................................................................... 77
Predicted Economic Ratios and Multipliers By Scenario ......................................... 80
Inbound Tourism Visitor Spending .......................................................................... 82
Economic Impacts - Scenario 1 ................................................................................ 82
Economic Impacts - Scenario 2 ................................................................................ 84
Economic Impacts - Scenario 3 ................................................................................ 86
CHAPTER V CONCLUSION ..................................................................................... 89
Importance of Inbound Tourism to Taiwan’s Economy ......................................... 89
Changing Job and Income Ratios by Occupancy Rates and Hotel Scale .............. 90
Evaluation of the DTAP Policy .................................................................................. 93
Economic Impacts ..................................................................................................... 93
Implications ............................................................................................................... 94
Application of the Varying Ratios [-0 Model .......................................................... 96
Recommendations for Future Studies ....................................................................... 97

APPENDIX A. ALLOCATION OF VISITOR SPENDING TO [-0 SECTORS .. 102

APPENDIX B. DEFINITIONS OF TAIWAN I-O SECTORS ................................ 103

APPENDIX C. ECONOMIC RATIOS FOR TAIWAN I-O SECTORS, I999 ...... 105

vi

APPENDIX D. M ULTIPLIERS FOR TAIWAN I-O SECTORS, 1999 ................. 106

APPENDIX E. TRADE MARGIN, TRANSPORTATION MARGIN AND
REGIONAL PURCHASE COEFFICIENTS FOR MANUFACTURING SECTORS
......................................................................................................................................... 107

APPENDIX F. SCATTERPLOT OF LINEAR AND NATURAL LOG FOR JOBS
TO SALES RATIOS VERSUS OCCUPANCY RATES ........................................... 108

APPENDIX G. SCATTERPLOT OF PERSONAL INCOME TO SALES RATIOS
VERSUS OCCUPANCY RATES ................................................................................ 109

BIBLIOGRAPHY ......................................................................................................... 110

vii

LIST OF TABLES

Table I. An Example of the Cost Structure for a Hotel Room by Absolute Value and
Percent .................................................................................................................. 9
Table 2. Demand and Supply Level under the Three Scenarios ...................................... 12
Table 3. International Arrivals and Expenditures in Taiwan, 1960 - 2003 ...................... 18
Table 4. International Visitor Arrivals by Region and Country of Residence. 2000 - 2003
............................................................................................................................ 19
Table 5. Basic Statistics of Taiwan Domestic Tourism, 1997 - 2003 .............................. 20
Table 6. Basic Statistics of Tourist Hotels in Taiwan, 2002 ............................................ 21
Table 7. Basic Statistics of Tourist Hotel by Region in Taiwan, 2002 ............................ 23
Table 8. Intended Goal of Intemational Visitors to Taiwan. the DTAP Policy ............... 24
Table 9. Allocation of Current and Future Tourist Hotel Rooms by Region in Taiwan,
"Tourist Hotel Development Plan” .................................................................... 27
Table 10. Tourism Economic Impacts in Taiwan in 1984, 1986 and 1988 ..................... 28
Table l I. Comparison of Taiwan Tourism Satellite Accounts from Two Reports ......... 30
Table 12. Inbound Tourist Spending Profiles by Country OfResidence. 2000-01 .......... 63
Table 13. Visits and Total Spending By Country of Residence. 2001 ............................ 63
Table 14. Economic Ratios for Primary Tourism Sectors, 1999 ..................................... 64
Table 15. Multipliers for Primary Tourism Sectors, 1999 ............................................... 65
Table 16. Economic Ratios and Multipliers by Country of Residence ............................ 66
Table 17. Direct Effects of Inbound Tourism by Country of Residence. 2001 ............... 67
Table 18. Direct Effects of Inbound Tourism by Sector, 2001 ........................................ 68
Table 19. Economic Impacts of Inbound Tourism by Sector. 2001 ................................ 68
Table 20. Indirect and Induced Effects of Jobs and Personal Income. 2001 ................... 69
Table 21. Regression Statistics for the Log Jobs to Sales Ratio ...................................... 71
Table 22. Regression Statistics for the Income to Sales Ratio ........................................ 72
Table 23. Predicted Jobs and Income Ratios for Tourist Hotels ...................................... 73
Table 24. Predicted Jobs and Income Ratios for the Accommodation Sector ................. 74
Table 25. Multipliers Ofthe Hotel Sector Based on Different Occupancy Rates ............ 75

viii

Table 26. Economic Impacts of Inbound Tourism in 2001 with Fixed and Varying Ratios

............................................................................................................................ 76
Table 27. Demand and Supply Levels under the Three Scenarios .................................. 77
Table 28. Trips and Room Nights for Domestic Tourism. 2008 ..................................... 78
Table 29. Supply. Demand, and Average Occupancy Rates of Tourist Hotels ............... 79
Table 30. Predicted Hotel Economic Ratios and Multipliers under the Three Scenarios 81
Table 31. Total Visitor Spending under the Three Scenarios (USS million’s) ................ 82
Table 32. Economic Impacts of Scenario 1 using Fixed Ratios ...................................... 83

Table 33. Comparison of Economic Impacts between Fixed and Varying Ratios for
Scenario 1 ........................................................................................................... 84

Table 34. Economic Impacts of Scenario 2 using Fixed Ratios ...................................... 85

Table 35. Comparison of Economic Impacts between Fixed and Varying Ratios for
Scenario 2 ........................................................................................................... 85

Table 36. Economic Impacts Of Scenario 3 using Fixed Ratios ...................................... 86

Table 37. Comparison of Economic Impacts between Fixed and Varying Ratios for

Scenario 3 ........................................................................................................... 87
Table 38. Total Jobs and Ranking Using Varying Job Ratios ......................................... 88
Table 39. Total Personal Income and Ranking Using Varying Income Ratios ............... 88
Table 40. Economic Ratios for Tourist Hotels in Taiwan ............................................... 91

Table 41. Sensitivity Of Impact Estimation using Fixed [-0 Ratios by Returns to Scale 93

LIST OF FIGURES

Figure I. The Conceptual Framework Of Capacity Utilization and I-O Analysis ............. 6
Figure 2. Seasonal Distribution of Domestic and International Customers at Tourist
Hotels, 1999 - 2003 ............................................................................................ 22
Figure 3. Four Geographic Regions, Taiwan .................................................................... 51
Figure 4. Estimation Process for Objective III ................................................................ 56

Figure 5. Predicted Jobs and Income Ratios with 95% C onfidence Intervals for the
Accommodation Sector ...................................................................................... 74

Figure 6. Job Ratio. Income Ratio and Average Salary per Employee by Occupancy
Rates for Tourist Hotels in Taiwan .................................................................... 92

CHAPTER I INTRODUCTION

Tourism Policy in Taiwan

Taiwan has experienced a steady expansion in both international arrivals and
foreign tourism receipts since 1960. International visitors to Taiwan increased from 24
thousand person visits in 1960 to 2.8 million visits in 2001 (Taiwan Tourism Bureau,
2004). Concurrently. total foreign receipts increased from USS 1 million to USS 4.3
billion.

In 2002. the gm'ernment Of Taiwan announced the "Doubling Tourist Arrivals
Plan“ (DTAP) as one major investment project for the national long term development
plan (Council for Economic Planning and Development. 2002). Its goal is to increase
international pleasure visitors from 1 million yearly visits in 2001 to 2 million by 2008,
and overall international visitors including pleasure. business. study purposes, visiting
friends and relatives from 2.8 million in 2001 to 5 millionl person visits by 2008.

TO cope with rising demands on accommodation services. a sub-policy. the
“Tourist Hotel Development Plan”, was announced, which encouraged private
investments in lodging facilities by providing them with low interest financial assistance
(Taiwan Tourism Bureau, 2003‘). This sub-policy intends to increase the lodging capacity
of tourist hotels from 20.697 rooms in 2001 to 37.900 rooms by 2008. The overall public

investment for the "Doubling Tourist Arrivals Plan” is USS 2.5 billion.

 

' If excluding foreign laborers working in Taiwan, approximately 200.000 person visits per year,
the policy aims to generate 4.8 million international visitors by 2008.

The economic Objectives of the DTAP policy are to increase foreign receipts.
create job opportunities. and diversify the economy (Council for Economic Planning and
Development, 2002). It is therefore critical to measure, monitor. and forecast the
economic impacts of international visitor spending in Taiwan over the policy span.
Economic impact analysis can be used to evaluate and compare various policy
altematives. advise marketing strategies. generate constituencies’ support. and most
importantly, assess the relative contribution of tourism developments to the national
economy.

Input-Output (I-O) analysis has been used extensively for evaluating the
economic impacts oftourism since the 1970‘s (Archer. 1977. 1978. 1984; Fletcher,
1989). 1-0 analysis estimates the changes in economic activity within a region resulting
from a given level offinal demand changes. such as visitor spending or government
investment. Impacts are computed first by converting the final demand changes into
direct effects in jobs. personal income. tax revenue or value added using economic ratios
pertaining to the regional economy. Secondary effects are then computed by multiplying
the direct effects by the regional multipliers. which result from interindustry transactions.
Total job impacts. for example. are determined by three factors: jobs to sales ratios.
regional sales multipliers. and final demand changes. The first factor. the jobs to sales
ratio, represents the average ratio of labor inputs to production. regional sales multipliers
indicate the structure of the local economy. and final demand changes specify the total
amount of money injected.

Within the standard I-O framework, total impact is solely determined by final

demand changes (Armstrong 8; Taylor. 2000; Miller & Blair. 1985). Constant technical

IQ

coefficients and fixed economic ratios of labor cost, employment, and profitability are
applied when calculating total impacts regardless of the scale of demand changes as well
as the level of sector capacity. The dynamics of the supply function for a given sector and
its interaction with the cost function are not accurately reflected in the 1-0 model. A
service “firm operating at a 40% capacity. for example. is assumed to have the same cost

structure as one operating at an 80% capacity.

Problem Statement

The "Doubling Tourist Arrivals Plan" influences the tourism industry in Taiwan
from both supply and demand perspectives. Tourism consumption is expected to increase
by hosting more international visitors while the infrastructure and service supply is
expected to expand with financial assistance/investments from government and private
sectors. The future operational capacity of tourism-related industries, a function of both
demand and supply. is currently undetermined.

For service-oriented sectors, Bryden (1973) notes that input-output coefficients
and economic ratios are likely to be influenced by the utilization of operating capacity. A
concern is therefore raised for evaluating the economic impacts of the DTAP policy.
because the suitability of standard I—O assumptions- the fixed ratios of labor cost and
employment with respect to different capacity utilization, is challenged. Without
understanding changes in cost functions due to capacity utilization. the accuracy in

evaluating and forecasting economic impacts of the tourism policy is in question.

b.)

Conceptual Framework

One primary limitation of the I-O analysis is that constant technical coefficients
and fixed economic ratios of labor cost. employment. and profitability are applied when
calculating total impacts. For manufacturing sectors. constant economic ratios and
regional multipliers within a short term period may provide a satisfactory portrayal of the
real-world production function and offer impact estimates with acceptable errors (West.
1995). Tourism. however. is not a single acknowledged sector (Fletcher, 1989; Smith.
1994). It covers a combination of industries. most providing time perishable services,
such as lodging. entertainment. and transportation. These services have distinct
production and consumption characteristics. compared to manufactured goods.

The major characteristics of services are intangibility, simultaneity and
perishability. Intangibility indicates that services are not solid products, and they are
perishable because they cannot be stored for repeated use. Simultaneity means that the
production and the consumption of services occur at the same point in time and location.
Unlike manufactured goods. unused perishable products (e.g.. hotel rooms or airline
tickets) cannot be resold somewhere else and the associated sales are lost forever. High
fixed costs are generally incurred in the production process regardless of the demand
level (Kimes. 1989; Lewis & Chambers. 1989). The marginal cost of an additional unit is
relatively low unless at full capacity. The Operational objective. therefore. is to maximize
profits instead of minimizing costs (Berman. 1994).

One approach to generate maximum profits for perishable products is through
price discrimination in which the price is determined based on the level of demand and

unused capacity. Accommodations and air transportation generally employ this pricing

strategy (referred as yield management by the academicians). Due to high fixed capacity
and product perishability. the profit level of many service firms is greatly tied to capacity
utilization (Allen. 1988).

The other general characteristic of service industries is that they are labor and
capital intensive (Mullins. 1993). Due to limited productivity gains through technological
progress. accommodations and restaurants tend to make the best use of existing labor to
accommodate additional demands (I Iollenstein. 2001; Peneder. Kaniovski. & Dachs.
2003; Smeral. 2003).

Due to these fundamental characteristics - perishability. intangibility and
simultaneity. cost structures of perishable services are determined by the number Of units
consumed (Sold). instead ofthe number of units produced. This has led to a distinct
measurement for time—perishable services: capacity utilization (CU). the ratio of actual
used products to some measure of potential outputs (Nelson. 1989; Ng. Wirtz, & Lee.
1999). Based on Berndt and Morrison (1981). CU is an important factor in explaining
changes in the rate of investment. labor productivity and inflation for services.

A conceptual model connecting capacity utilization and the I-0 framework is
presented in Figure I. It hypothesizes that a shift in utilization for perishable services will
lead to economics ofscale in both labor and material inputs (Miller & Blair, 1985), price
changes in final output (Arenberg. 1991). and possible substitution between input factors,
especially between capital and labor (Krakover. 2000; West & Gamage. 2001). The

pricing strategy will modify the price ratio of intermediate inputs to final outputz; the

 

3 Price change may result in technical substitution between different intermediate inputs.
However. it has been argued that a major swing in relative price is required before substitution
takes place (Rose & Miernyk. I989).

economies of scale change the input ratio of labor and materials; and the substitution
pattern between labor and capital adjusts the average salary per employee. These factors
Simultaneously affect the I-O technical coefficients. jobs to sales ratios. and personal

income to sales ratio.

1 Time perishable
services

i Capacity utilization

- 2 22, __-v- .__2 2 2, - L___y-_2_-2._2_2_-
1 1 Substitution between
Economy of scales ' labor and capital

‘ i I t inputs

,22- 2_A 2 2-2-, - - A

t ‘ Intermediate input

Average salary per 1
per unit Ofoutput '

employee

1 Labor input per unit

intermediate inputs ‘ .
I of output

i . .
i Price ratio between
I and final outputs j

l
l
l
l
l

- /

“" T‘ — " #' - "~;:,:,-_' v i" "’ A i ~—,r

I-O technical t ‘ . i Personal income to .
- . . 1 Jobs to sales ratio ; , . i
coefircrents i ‘ sales ratio

Figure I. The Conceptual Framework of Capacity Utilization and I-O Analysis

As demonstrated in Equation 1. economies of scale in input materials and price

changes of the final product would modify the I-0 technical coefficients. Economies Of

scale in labor inputs (labor efficiency) and changes in the price of the final product would
adjust the jobs to sales ratios (Equation 2). Personal income to sales ratios on the other
hand are influenced concurrently by labor inputs. output price and average salary per

employee (Equation 3).

P- . . . . . . . ..
a -- = q-- * —'—= f (economy of scale In materials. price ratio of Inputs to final
Ij Ij P' .
J

product) (Equation 1)

Where a ij = technical input coefficient = output ofindustry i that is bought by
industry j to produce one unit of product j

qij = physical input coefficient = physical output of industry i required to produce a
physical unit output of industry j

P; = price of product i ;

Pi = price of product j

Total employees

 

Jobs to sales ratio =
Total sales

Total employees * 1

Total units sold Price of the final product

 

=f (labor productivity. price Ofthe final product) (Equation 2)

Total wages

 

Personal income to sales ratio =
Total sales

= Total employees * 1 * Average salary per employee

Total units sold Price of the final product

 

 

f(labor productivity. price of the final product. average salary per employee)

(Equation 3)

A hypothesized example is used in Table I to demonstrate changes in the cost
structure by allowing output price and labor ratio to vary. Labor ratio. in this example. is
defined as number of final products sold per employee. which can be regarded as a
measure of labor efficiency. The base year cost function of the accommodation sector is
presented first using the 1-0 coefficients from the 1999 Taiwan Input-Output table.
Assuming a room price of S100. thirty-three percent of the cost will go to intermediate
inputs. such as business services and manufacturing. 41% will be allocated to employee
compensation. 2% to imported products. and 24% to profits and taxes. In this instance.
the labor efficiency equates to the number of occupied rooms per worker. In the base year
(the year I-O table is compiled). one worker is hypothesized to support five occupied
rooms at an occupancy rate of 50%.

By allowing labor efficiency and output price to change along with occupancy
rates, the cost function and profitability. in absolute value and relative percent. are
different. For example. if labor efficiency is improved from serving 5 occupied rooms per
employee to 8 occupied rooms per employee. the relative ratio of labor cost is reduced
per room while the ratio of input materials remain unchanged. In this instance. type I
sales multipliers will remain unchanged from the base-year level but the jobs to sales
ratio will decrease. indicating that fewer jobs are supported with the same amount of

hotel sales. A change in room price. however. will affect the relative cost ratio for both

intermediate inputs and labor inputs. In this case. the elasticity of changes in 1-0

multipliers will be determined simultaneously by three factors: labor efficiency, the price

of input materials. and the price of final products. For example, if the average room price

rises from $100 to $150. assuming the input materials still cost $33 per room and one

employee supports 5 occupied rooms (unchanged from the base-year level). the absolute

monetary costs of intermediate inputs and labor are unaffected but the relative ratio of

costs is reduced. This will then lead to reduced type I sales multipliers and jobs

multipliers.

Table 1. An Example ofthe Cost Structure for a Hotel Room by Absolute Value and Percent

 

 

Absolute valuciper room (S) Percent
Change Change
of labor of labor
Change Change efficiency Change Change efficiency

 

 

 

 

 

Base oflabor of room and room Base of labor of room and room
Year efficiency price price year efficiency price price
Hypothesized parameters
Room price $100 $100 $150 $150 $100 $100 $150 $150
Occupied rooms! 5 8 5 8 5 8 5 8
employee
Intermediate inputs
Manufacturing 8 8 8 8 8% 8% 5% 5%
Electricity 6 6 6 6 6% 6% 4% 4%
Wholesale & retail 2 2 2 2 2% 2% 1% 1%
Business service Q E L8 L3 18% 18% 12% 12%
Subtotal 33 33 33 33 33% 33% 22% 22%
Primary inputs
Personal Income“ 41 26 4 l 26 41% 26% 27% 1 7%
Profits and taxb 24 39 74 89 2494) 399/0 499/0 600/0
Imported product 2 2 2 2 2% 2% 1% %
Subtotal .1 fl LE I|_7- fl m. M M
Total 100 100 1 5 l 50 100% 100% 100% I 00%

 

 

 

a . .
Assuming the average salary per employee IS fixed. When an employee can clean more room, for
example. the average wage cost per room is reduced (marginal labor cost per room is decreasing).

b . . . . . .
Profits and tax are computed by deducting the rooms sales With costs on IntermedIate Inputs. Imported

products and labor.

Based on this example. an increased room price and more efficient labor usage
result in a reduced input costs and subsequently, higher operating surplus. Allowing the
average salary per employee to vary creates a more complicated example. Any change in
the input factors will lead to modifications of the cost structure. The assumption of
Constant Returns to Scale (CRTS), which indicates fixed input costs. labor expenses. and
profitability in relation to sales volume. is not valid when output price. labor efficiency.
and average wages are allowed to vary.

Based on the proposed model in Figure I. shifts in capacity utilization will lead to
modifications oftechnical input coefficients and economic ratios for the value added
components. In this study. the primary focus is to relax the assumption of constant value
added components (the dynamics ofjob, income and profits) in the [-0 model while the

technical input coefficients remain constant.

Objectives

The accommodation sector in Taiwan is selected as the study subject to examine
the relationships between capacity utilization and two economic ratios (jobs to sales
ratios and personal income to sales ratios) due to the following reasons. First. the 'I‘aiwan
tourism policy directly influences both the supply (the expansion of tourist hotels) and
the demand (doubling international visitors) Of the accommodation sector. Hotel
occupancy rates as well as cost structures of the accommodation sector are expected to
vary, depending upon the realization of the proposed policy. Second, hotel rooms are a
good example of time perishable services and the accommodation sector is sensitive to

demand and supply changes. in terms of room price. labor use and the resulting

profitability. Third. the concept Ofcapacity utilization is well defined and measurable in

the accommodation sector. With data on total available and occupied rooms. capacity

utilization of the accommodation sector can be precisely measured in terms of occupancy

rates (Borooah. 1999).

II.

111.

Three study objectives are:
Estimate the economic impacts of inbound tourism on the Taiwan economy in
2001. before the implementation ofthe DTAP policy. in terms of sales. jobs and

personal income.

Test the stability ofjobs to sales ratios and income to sales ratios with respect to

occupancy rates in the accommodation sector

Estimate economic impacts of inbound tourism on the Taiwan economy for three
scenarios based on 1) standard I-O models with fixed job and income ratios. and
2) I-O models with varying jobs to sales ratios and income to sales ratios by
occupancy rates. The former I-O model is referred as “fixed ratios model" and the
latter as “varying ratios model" in the following.
Three scenarios to be evaluated are
Lodging capacity does not expand during the policy period. maintaining the
same capacity of 2001 as 20.697 rooms. Estimate the maximum economic
impacts in 2008 under a possible capacity constraint in the accommodation

SCCIOI‘.

IQ

Lodging capacity expands from the current 20.697 rooms to 37.900 rooms by

2008. based on the "Tourist Hotel Development Plan” (Taiwan Tourism

Bureau, 2003). and 75% ofthe intended visitation volume in the DTAP policy

is achieved (3.6 million of international visitors).

DJ

Lodging capacity expands from the current 20.500 rooms to 37,900 rooms by

2008. and 100% of the intended visitation volume in the DTAP policy is

achieved (4.8 million of international visitors).

Each scenario involves particular demand and supply conditions in the

accommodation sector (Table 2). Supply levels are measured by the number of total

available rooms in tourist hotels. Demand levels. on the other hand, are measured in

terms of the total tourist hotel rooms demanded from inbound and domestic tourists. The

domestic tourism volume is assumed constant (135 million person trips a year by 2008)

across the three scenarios (Taiwan Tourism Bureau. 2003). Maintaining constancy in

domestic tourism helps to simulate the changes in occupancy based on fluctuations of

inbound tourism and hotel capacity.

Table 2. Demand and Supply Level under the Three Scenarios

 

Domestic Tourism Inbound Tourism

 

 

Scenarios Tourist Hotel Capacity (person trips) (person trips)a
Base Year (2001) 20.697 rooms 97 million 2.6 million
Scenario 1 (2008) 20,697 rooms 135 million Maximum visitors (1’)
Scenario 2 (2008) 37,900 rooms 135 million 3.6 million
Scenario 3 (2008) 37.900 rooms 135 million 4.8 million

 

a Foreign laborers to Taiwan are excluded. which are approximately 20 thousand persons per year.

Assumptions

For objective I. three primary assumptions Of standard [-0 analysis are employed.
They are I) the output of each sector is produced with a unique set of inputs, 2) the
amount Of inputs required is solely determined by the level of output, and 3) there are no
capacity constraints in the production process (Miller & Blair, 1985; Otto & Johnson.
1993). In other words. constant cost structures during the policy evaluation period are
assumed. Fixed technical coefficients and economic ratios are employed.

In objective II. the assumption of constant job and income ratios for the
accommodation sector is tested using tourist hotel Operational data in Taiwan. Due to
limited data. it is assumed that small accommodation firms. such as Bed & Breakfast and
general hotels. behave in a Similar fashion as tourist hotels toward capacity utilization. In
1996, tourist hotels accounted for 4% of total accommodation entities, 70% of total
employees. 60% of total sales. and 55% of total personal income in the accommodation
sector (Directorate-General of Budget Accounting and Statistics. 2003). Tourist hotels are
assumed to be representative for the accommodation sector as they account for a majority
of hotel sales and workforce.

For objective III. the following assumptions are employed. First. technical input
coefficients for the accommodation sector are assumed constant across the policy period
from 2002 to 2008. This implies technologies. trade flows and price movements between
intermediate inputs and room prices3 are constant. This assumption is proposed because

I) the accommodation sector possesses a much slower rate of technological change than

 

P.. P..
3 iat time (t) = -—U— at time (t-tl). where as P, equates to the price ofproductj and PU equates to the price
.i j

ofinput i required to produce productj.

other service sectors (Hollenstein. 2001; Peneder et al.. 2003). and 2) only 2% of the
inputs in the Taiwan accommodation sector were imported based on the 1999 [-0 table.
The low level ofimported products and limited technological change are expected to
pose minimal influences on the 1-0 coefficients over time. Assuming constant technical
input coefficients for the accommodation sector. in other words. implies constant type I
sales multipliers across the three scenarios.

Second. for the rest of the economy (besides the accommodation sector), [-0
technical coefficients and economic ratios are assumed to be constant during the policy
period. Third. trip characteristics in visitors" length of stay. average spending. propensity
of hotel stays. and regional distribution in trips are assumed constant. The shift of
demands due to crowding or regional development will not be considered here. The only
factor influencing final demand changes (inbound tourism visitor spending) is the total
number of international arrivals. Last. domestic tourism volume is provided by the
Taiwan Tourism Bureau (2003). estimated to be 135 million person trips in 2008. The
possible fluctuation of domestic tourism volume due to competition or price inflation is

not discussed.

Limitations
Limitations of this study primarily rest on the assumptions it employs. The first
limitation is that the current model does not consider possible cost structure changes for
other service sectors at different capacities. Joint effects across sectors with respect to
capacity utilization are not modeled. Second. modifications of existing travel patterns. in

terms of length ofstay and accommodation types. are possible. Regional development.

the amalgamation of new attractions. quality of services. the newly constructed west-
coast Highway 3 and the hi gh-speed bullet train may influence current travel patterns in
Taiwan. Also. gasoline price. exchange rates, or effects of terrorism. weather or disease
outbreaks may also influence the volume of international arrivals to Taiwan. These
factors are currently not modeled. The third limitation lies in the assumption that small
establishments may behave in a similar fashion as tourist hotels with respect to capacity
utilization. Due to distinct organizational and cost structures. sensitivity and direction of

changes for small establishments may be different.

Definition of Terms

Capacity: Lovelock (1992) defined the capacity of service firms is “the highest quantity
of output possible in a given time period with a predefined level of staffing.
facilities and equipment"( p.26). In the context of the accommodation sector, the
capacity is the maximum rooms that are available for rent in the short-term.

Capacity utilization (CU): The ratio of actual used products to some measure of potential
output (Nelson, 1989; Ng et al.. 1999). Potential output can be referred to 1) the
maximum output that may be produced given a firm's short-run stock of capital.
or 2) the output based on the economic measurements using the long-term and
Short-term total cost curves (Nelson, 1989). In the context of the accommodation
sector. CU is referred as the occupancy rate (Borooah. 1999).

Returns of scale: The percentage increase in output by a one percent increase in all inputs

(Caves & Christensen. 1988).

Labor efficiency: In this study. labor efficiency is defined as number of products
consumed (sold) per employee. This indicator equates to the occupied rooms per
employee and is independent from room price.

Jobs to sales ratio: Number ofjobs supported per dollar sale.

Personal income to sales ratio: The ratio of wage. salary. proprietor‘s income and
employee benefits per dollar sale.

Type I multipliers: The sum Ofdirect and indirect effects divided by the direct sales
effect. The multiplier can be measured in terms of sales. jobs. personal income or
value added.

Type II multipliers: The sum Ofdirect. indirect, and induced effects divided by the direct
sales effect. Induced effect is resulted from the inclusion of household
consumption in the 1-0 model. The multiplier can be measured in terms ofsales.
jobs. personal income or value added.

Value added: The sum ofpcrsonal income plus rents. profits and indirect business taxes.

Tourist hotels: There are four hotel categories in Taiwan: the international tourist hotel.
standard tourist hotel. general hotel and Bed and Breakfast (B&B), based on
standards issued by Taiwan Tourism Bureau (2003). The first two categories are
referred to as "tourist hotels“. which are subject to governmental regulations on
its facilities and service procedures. serving mainly international visitors.

Tourists. travelers or visitors: In the context of this study, these terms are used
interchangeably. Tourists include travelers with business. leisure and other trip
purposes.

N.T.: New Taiwan Dollar. the currency of Taiwan.

CHAPTER II LITERATURE REVIEW

The literature review chapter includes three sections. The first section reviews the
current status of Taiwan domestic and inbound tourism, followed by the “Doubling
Tourist Arrivals Plan" and Taiwan tourism economic impact studies. The second section
compares the four primary tourism economic impact models with detailed specification
ofthe Input-Output (I-O) analysis. The third section discusses estimation errors in

tourism I-O models with respect to capacity utilization for service sectors.

Tourism Development in Taiwan

Domestic and Inbound Tourism

Inbound tourism. Following worldwide trends in tourism. Taiwan has experienced
strong growth since 1960 in both international arrivals and foreign exchange.
lntemational visitors to Taiwan increased from 24 thousand person visits in 1960 to 2.25
million visits in 2003. Concurrently, total foreign receipts increased from USS 1 million
to USS 2.98 billion (Table 3). Inbound tourism peaked in 2002, with approximately 3
million international arrivals and USS 4.6 billion foreign exchange receipts. The average
spending per visitor was USS 1.539 for a 7.5 night stay in Taiwan. Due tO the spread of
Severe Acute Respiratory Syndrome (SARS) and Avian influenza in Taiwan in 2003.
international arrivals decreased by 25% and foreign receipts were reduced by 35%
compared to 2002 (Taiwan Tourism Bureau. 2004). The following discussion of domestic
and inbound tourism is based on the 2002 statistics as the 2003 data may not represent the

general behaviors of tourists in Taiwan.

Table 3. International Arrivals and Expenditures in Taiwan, 1960 - 2003

 

 

International Total visitor Average spending Average spending

visitor arrivals expenditure (SUS per person per trip per person per day Length of stay
Year (000's) million dollars) (SUS dollars) (SUS dollars) (nights)
1960 24 1 63 25 2.50
1965 134 18 137 38 3.64
1970 472 82 173 36 4.86
1975 853 359 421 67 6.30
1980 1.393 988 709 105 6.78
1985 1.452 963 664 102 6.51
1990 1.934 1.740 1.346 187 7.20
1995 2.332 3.286 1.409 191 7.38
2000 2.624 3.738 1.425 193 7.40
2001 2.831 4.335 1.531 208 7.37
2002 2.978 4.584 1.539 204 7.54
2003 2.248 2.976 1.324 166 7.97

 

 

Source: Taiwan Tourism Bureau (2004)

In 2002, eighty percent of international visitors came from Asia. The top three
origin countries were Japan (34%). Hong Kong/Macao (15%) and United States (13%).
representing more than 60% of total arrivals (Table 4). Thirty-seven percent of
international visitors were leisure travelers, followed by business visitors (30%) and those
visiting friends and relatives (10%)4. In terms of lodging types. eighty-one percent of
international visitors stayed at hotels. Among those, 52% Stayed at international tourist
hotels, 8% at standard tourist hotels, and 40% at general hotels (Taiwan Tourism Bureau.
2003). Approximately one in every two international visitors stayed at tourist hotels in

Taiwan [81% * (52% + 8%)].

 

4 Fifteen percent of visitors did not identify their trip purposes to Taiwan (Taiwan Tourism
Bureau. 2003).

Table 4. International Visitor Arrivals by Region and Country of Residence, 2000 - 2003

 

 

 

 

Visitor Arrivals (000’_s) Percent
2000 2001 2002 2003 2000 2001 2002 2003
Region
Asia 1.985 2.224 2.331 1.768 76% 79% 78% 79%
America 410 402 437 3 15 16% 14% 15% 14%
Europe 161 I49 149 1 19 6% 5% 5% 5%
Oceania 3 8 3 8 4 1 32 1% 1% 1% 1%
Africa 9 9 9 8 0% 0% 0% 0%
Unstated 2 2 1.9 1 fl 0°19 ma 0%
Grand Total 2.624 2.831 2.978 2.248 100% 100% 100% 100%
Country
Japan 916 977 998 657 35% 35% 34% 29%
Hong Kong & Macao 361 435 457 323 1496 15% 15% 14%
United States 360 349 377 273 14% 12% 13% 12%
Korea 84 86 84 93 3% 3% 3% 4%
Singapore 95 99 1 l 1 79 4% 3% 4% 4%

 

Source: Taiwan Tourism Bureau (1999-2003)

Domestic Tourism. The domestic tourism market had increased from 72 million
person trips in 1997 to 106 million person trips in 2002, averaging 4.01 trips per person5
in 1997 to 5.62 trips per person in 2002 (Taiwan Tourism Bureau. 1998-2004). A trend of
increasing day trips and fewer overnight stays at hotels was observed (Table 5). This has
lead to shorter stays and lower average spending per person per trip over the past few
years. In 2002. sixty-three percent of domestic trips were day trips. 22% were two-day
overnight trips. and 10% were three—day overnight trips. Sixty-one percent of domestic
trips were pleasure trips and 23% were visiting friends and relatives (VFR). The average
expenditure per person per trip was NT$ 2.228. with food and transportation each
accounting for 25%. In 2002. total expenditures by domestic travelers were NT$ 236.8
billion or US$ 6.95 billion. In terms oftotal trip spending. the domestic tourism market in

Taiwan is approximately 1.5 times the inbound tourism market (US$ 4.5 billion in 2002).

 

5 Excluding children younger than 12

Table 5. Basic Statistics of Taiwan Domestic Tourism, 1997 - 2003

 

 

Domestic Travel 1997 1999 2001 2002 2003
Average trips per person per year 4.01 4.01 5.26 5.62 5.39
Total person trips (0005) 71.879 72.651 97.445 106.278 102.399
Distribution oftrip lengths
1 day 50% 59% 62% 63% 64%
2 days 26% 23% 22% 23% 22%
3 days 150/ 12% 10% 10% 9%
4 days or more 8% 6% 5% 5% 5%
Average length ofstay 2.0 1.8 1.7 1.7 1.7
Lodging types
Day trips 50° '0 5 8% 62% 64%
Hotel 25% 20° 6 19% 18%
VFR 1 8% 18% 16% 13%
Spending
Average spending per person per trip (NTS) 3.300 2.738 2.480 2.228 2.130
Total domestic travel spending (NTS million’s) 7.201 198.918 241.664 236.787 218.110
Spending percentage by sector
Transportation 24% 22% 24% 24% 22%
Lodging types 17% 18% 16% 18% 17%
Food 26% 26% 25% 25% 24%
Entertainment 1 1% 10% 10% 7% 7%
Shopping 1 7% 20% 20% 21% 22%
Others 5% 4% 6% 4% 8%

 

 

Source: Taiwan Tourism Bureau (1998-2004)

The lodging indusn‘y. Hotels in Taiwan are divided into four levels: international

tourist hotel. standard tourist hotel. general hotel and Bed and Breakfast (B&B). based on

standards issued by Taiwan Tourism Bureau (2003). The first two categories are referred

to as “tourist hotels". International tourist hotels are four or five star equivalent.

providing an array of services besides lodging. and supporting more rooms and

employees per establishment. Standard tourist hotels are two or three star equivalents

with a smaller operating scale. The average international tourist hotel has 292 rooms with

329 employees. Standard tourist hotel. on the other hand. has 132 rooms with 88

employees on average.

In 2002. there were 62 international tourist hotels and 16 standard tourist hotels in

Taiwan. providing 17.840 rooms and 2.713 rooms respectively (Taiwan Tourism Bureau.

1999-2003). The average occupancy rate was 61.3% and the average room price was
NT$2.907 in 2002 (Table 6). Forty-four percent of hotel revenue came from food and
beverage sales6 and 40% from room sales. Forty-two percent ofemployees worked in the
food and beverage division. 29% in the room division and 18% in the administration
division. In 2002, total sales of tourist hotels were NT$ 33.7 billion and total employment

was 20.879 jobs. indicating 0.62 jobs per million NTS dollars in sales.

Table 6. Basic Statistics of Tourist Hotels in Taiwan, 2002

 

 

 

 

 

 

Total
lntemational Standard (Tourist
tourist hotels tourist hotels hotels) Percent
Number of establislnnents 62 16 78
Room numbers 17.840 2.713 20.553
Occupancy Rate 62% 59% 61%
Average room rate (NTS) 3.025 2.101 2.907
Revenue (NTS million's)
Room sales 12.133 1.230 13.363 40%
Food and beverage 14.026 915 14.941 44%
Others $819 487 m 1_6%
Total 31.058 2.632 33.690 100%
Employee number
Rooms 5.319 634 5.953 29%
Restaurants 8.477 564 9.041 43%
Administration 3.291 364 3.655 18%
Others 1.93_6 2‘5 2.31 1 1%
Total 19.023 1.856 20.879 100%
Customer number (person nights 0005)
Domestic customers 2.402 337 2.739 43%
International customers i988 554 3.642 57%
Total 5.490 8 1 6.381 100%
Jobs/ million NT dollars in sales 0.61 0.71 0.62
Jobs/ million US dollars in salesa 21.18 24.38 21.43

 

 

a One US dollar equated to $34 Taiwan dollar in 2002 (OAN DA Corporation. 2004).
Source: Taiwan Tourism Bureau (1999-2003)

 

6 . . .
Mainly wedding banquets and conferences.

In 2002. tourist hotels reported 6.38 million person nights. Forty-three percent of
these were generated by domestic customers. followed by Japanese customers (25%) and
other Asian customers (1 1%). Demand for tourist hotels was fairly consistent across four
seasons (Figure 2) as high seasons for domestic and international travel were different. A
higher percentage of domestic customers were reported during the summer season of July
through September (31%) while higher volumes of intemational visitors were reported

during the first and fourth quarter of the year (26~27%).

 

 

 

 

 

 

 

 

 

 

 

[ OJan-Mar lApr-Jun AJul-Sep .Oct-Dec i
40%
35%
H 0/ . g
g 30.0 ‘ . . .
I- . A . ‘
:1: 25% —‘ I 1 x 3 t 3
' I I I
20% —
15%
Domestic Overseas North Japan Asia Europe Australia Other Total
customers Chinese America region customers
Region

 

 

 

Figure 2. Seasonal Distribution of Domestic and International Customers at Tourist Hotels. 1999 -
2003

Source: Taiwan Tourism Bureau (1999-2003)

. . . . 7 .
Twelve thousand tourist hotel rooms are located in northern T arwan . accounting
for 60% of the total tourist hotel capacity. followed by southern Taiwan (22%). eastern

Taiwan (10%) and central Taiwan (9%) (Table 7). Tourist hotels in northern Taiwan

 

7 Eighty-five percent oftourist hotels in Northern Taiwan are located in Taipei (the capital of Taiwan).

contribute 71% and 66% ofthe nation-wide tourist hotel sales and employees.

respectively. The average room price in the Taipei area was NT$ 3.158 in 2002 and the

average occupancy rate was 67%. representing 8.6% and 9.8% higher than the nation-

wide averages. Around three and half million customers were reported in the Taipei area.

which accounted for 57% of total customers in 2002. Tourist hotels in northern Taiwan

primarily serve international visitors. only 20% of their customers are domestic travelers.

In summary. hotels in northern Taiwan account for the majority of hotel capacity. have

higher occupancy rate and room price. and mainly serve international visitors.

Table 7. Basic Statistics 01‘ Tourist Hotel by Region in Taiwan, 2002

 

 

Region Northern Central Southern Eastern Total
Taiwan Taiwan Taiwan Taiwan
Number of rooms 12.250 1.748 4.551 2.005 20.553
Occupancy Rate 67% 52% 55% 49% 61%
Average room price (NTS) 3.158 2.649 2.333 2.515 2.907
Total room sales (NTS million's) 9.444 880 2.139 901 13.363
Total revenue (NTS million's) 24.224 2.292 5.503 1.671 33.690
Total employees 13.752 1.599 4.103 1.425 20.879
Number of customers in 2002 (000's)
Domestic 708 294 1.066 671 2.739
lntemational L215 112 4g 1_3_; 3.642
Total 3.623 472 1.484 802 6.381
Pet of international customers 80% 38% 28% 16% 57%

 

 

Source: Taiwan Tourism Bureau (1999-2003)

Doubling Tourist Arrivals Plan (DTAP)

The “Challenge 2008- National Development Plan” establishes the long-term

d€“V’1elopment goals and investment plans for Taiwan from 2002 to 2007. It includes three

gov emmental refomis on political. fiscal and financial aspects and ten industrial

23

development plans8 (Council for Economic Planning and Development. 2002). The main
objectives of this policy are to stimulate economic growth through public investments.
create employment by diversifying economic structures. and increase the quality of life
for people in Taiwan.

The “Doubling Tourist Arrivals Plan“ (DTAP) is one ofthe ten industry
development policies in. The goal of this tourism policy is to increase the international
pleasure visitors from 1 million yearly visits in 2001 to 2 million by 2008. and overall
international visitor from the current 2.6 million to 5 million yearly visits. The proposed
international visitor numbers by individual country from 2002 to 2007 are listed in Table
8. International visitors from Japan. Hong Kong/Macao are the two current major markets
for Taiwan. which are expected to increase the visitation number from a combination of
140 million person visits in 2002 to 260 million in 2008. representing more than 50% of

total international visitors.

Table 8. Intended Goal of International Visitors to Taiwan, the DTAP Policy

 

 

 

Hong New

Kong/ Singapore North Zealand/
Year Japan Macao Korea I’Malaysia America Europe Australia Others Total

Visitors (10.000)
2002 100 45 9 16 35 15 3 47 270
2003 110 50 10 18 36 16 3.5 56.5 300
2004 120 55 14 20 38 18 4 81 350
2005 140 63 18 23 4O 20 5 91 400
2006 160 70 22 26 42 24 6 100 450
2007 180 80 30 30 50 28 8 94 500
A nn ual 13% 12% 28% 13% 8% 13% 22% 16% 13%
.é’LO \Nth rate

 

Source: Council for Economic Planning and Development (2002)

r\
Te 11 key individual plans are e-Generation Manpower Cultivation Plan. Cultural and Creative
Ind.L_1 stry Development Plan. International Innovation and R& D Base Plan. Industrial Value
e ‘ ghtening Plan. e-Taiwan Construction Plan. Operations Headquarters Development Plan.
IS“'51“ld—Wide Trunk Transportation Construction Plan. Doubling Tourist Arrivals Plan. Water and
re§n Construction Plan. and New Home Community Development Plan.

24

Five development strategies are proposed in the DTAP policy (Council for

Economic Planning and Development. 2002).

10

(Reorganize existingfitour itineraries

Five tour itineraries. which mainly serve international visitors. would be reformed
through improving current transportation systems. landscape designs. events
organizations. lodging services. and managements. These five destinations are located
across Taiwan with two in central mountain regions. one in northern Taiwan (near the
capital. Taipei). one in eastern 'I‘aiwan. one in the southeast region.

Develop new tour packages and new destinations

Thirteen destinations and travel routes are proposed as the additional destinations for
international visitors. New tourism destinations would be incorporated into regional
development plans to establish and coordinate new infrastructures. transportation
systems. and management agencies. Each region of Taiwan is assigned one major
destination development plan with specific themes. such as cultural. flora. sports. and
natural resources.

Establish a tourist service network

 

This objective aims to improve the quality of travel experiences through improvement

in transportation. lodging facilities. package discounts. and information network.

Regional transportation system. such as “bus seeing system". and national rail tours
around the island would be established to increase the convenience of traveling.
Tourist information networks. such as itinerary booking systems. would be expanded

through online information and regional information centers in multiple languages.

to
Ur

 

4. Launch international tourism promotion campaigps
Marketing programs are implemented in target markets. The short-term and primary
goal focuses on markets in Asia while the long-term goal will focus on visitors from
Australia and Europe.

5. Develop the MICELh/leetings. Incentives. Conventions. and Exhibitions) market
This objective aims to increase facilities for international conventions and
exhibitions. offered financial incentives for hosting international meetings and
provided training and education opportunities for cultivating Professional Convention

Organizers (PCO).

In response to possible capacity constraints on lodging. the Taiwan govemment
also proposed to expand hotel capacity through the following two approaches. First.
general hotels would be financially supported to upgrade and renovate current facilities to
a higher standard (three stars or higher). Second. existing regulation regarding land use
and zoning would be modified to allow new establishments on agricultural lands (Taiwan
Tourism Bureau. 2003). An estimated 15.100 upgraded and newly constructed rooms will
be added. Eighty-live percent would be allocated to northern Taiwan. and 3%. 8%. 4% to
central. southern. and eastern Taiwan respectively. By 2008. a total of37.900 tourist
hotel rooms will be available and 70% of these will be located in northern Taiwan (Table
9).

The DTAP policy is funded by the central government (NT 75.71 billion / US$
2.16 billion). local governments (NT 0.15 billion). and special funds (NT $9.06 billion/

uss 0.26 billion).

Table 9. Allocation of Current and Future Tourist Hotel Rooms by Region in Taiwan, “Tourist

Hotel Development Plan”

 

 

 

Northern Central Southern Eastern
Taiwan Taiwan Taiwan Taiwan Total
Number of tourist hotel rooms a
Room available in 2002 13.100 2.200 4.800 2.700 22.800
Rooms that will be added w M IAOQ _60_0 mg
Total rooms available in 2008 26.000 2.600 6.000 3.300 37.900
Pct by regions
Room available in 2002 57% 10% 21% 12% 100%
Rooms that will be added 85% 3% 8% 4% 100%
Total rooms available in 2008 69% 7% 16% 9% 100%

 

 

Source: Taiwan Tourism Bureau (2003)

a . . . . . . . . .
These are the proposed lodging capacrtres by Taiwan Tourism Bureau. There rs slight differences between
the proposed volume and the reported hotel rooms for 2002.

Taiwan Tourism Economic Impact Studies

The first nation-wide tourism economic impact study for Taiwan was carried out
in 1991 using an Input-Output model (Wang & Wang. 1991). The study estimated the
total economic impacts of domestic tourism. inbound tourism. and government
investments in 1984. 1986. and 1988 using the 1984 and 1986 Taiwan National I-O
tablesg. Results of the study are summarized in Table 10.

From 1984 to 1988. total visitor spending ofinbound and domestic tourism
ranged from NT$ 72 billion to NT$ 125 billion. In 1988. tourism activities in Taiwan
supported 4.0% of the national total sales and 5.8% of the total personal income. The
aggregate type 11 sales multiplier ranged from 2.56 to 2.63 during 1984 to 1988. The
attribution of tourism activities. including government investments. to the total national

employment ranged from 4.75% to 6.81%.

 

Tourism economic rrnpacts m 1988 were estimated using the 1986 Tarwan l-O table.

27

Table 10. Tourism Economic Impacts in Taiwan in 1984, 1986 and 1988

 

 

 

Year 1984 1986 1988
Final demand (NTS billion's)
Domestic tourism 26.96 34.02 58.84
lntemational inbound tourism 45.14 48.76 66.46
Government investment 0_.36 9.39 94—5
Total 72.45 83.17 125.75
Total economic impacts (Direct. indirect and induced effects)
Sales (NTS billion's) 190.64 212.84 323.48
Employment (1000's) 371 367 552
Personal income (NTS billion's) 74.40 89.10 134.45
Pct oftourism economic impacts to the national economy
Sales 3 .2 7% 3.14% 3.99%
Employment 5.08% 4.75% 6.81%
Personal income 3.14% 3.05% 5.76%
Type 11 sales multipliers 2.63 2.56 2.57

 

 

Source: Wang and Wang (1991)

The World Travel & Tourism Council (W TTC ). an affiliated organization of

World Travel Organization (WTO). has provided tourism economic impact estimates for

all major countries in the world since 1990 using standard Tourism Satellite Accounts.

WTTC (2002) estimated that travel and tourism demand in Taiwan
domestic leisure travel. business travel. governmental expenditures

inbound tourism and other exports. was NT $ 754.2 billion in 2002

. which includes

. Tourism sales

. capital investment.

generated a total of 172.902 jobs or 1.8% of total employment in terms of direct effect;

585.587jobs including indirect effects. The travel and tourism economy contributed NT$

106.2 billion in Gross Domestic Product (GDP) in direct effects. corresponding to 1.1%

of national GDP.

In 2002. the Taiwan Tourism Bureau published the first government report of the

Tourism Satellite Accounts for 1996 (Taiwan Tourism Bureau. 2002). The study

estimates of tourism GDP and employment in Taiwan were substantial higher than the

WTTC report. The Taiwan Tourism Bureau figures should be more accurate because of
its greater use of local data sources and detailed methods provided.

This study covered tourism expenditures associated with hotel. food service.
ground transportation. aviation transportation. car rental. travel service. culture and
entertainment and retailing from inbound tourism. outbound tourism”). and domestic
tourism. Government and private investments in tourism-related activities were not
included in the report. Transactions that were associated with non-currency products or
services. such as barter transaction. production for own use. and counterpart of income in
kind (Taiwan Tourism Bureau. 2002. p.3-6) were also not included.

Using 1996 as the base year. total tourism expenditures in Taiwan were NT
$321.63 billion. including domestic tourism ofNT $166.05 billion. inbound tourism NT$
108.62 billion. and outbound tourism NT $46.96 billion (Table I 1). One third oftotal
expenditures went to transportation. followed by food (20.0%). shopping ($18.4%). and
accommodation (1 1.5%). The percent of tourism consumption to total supply are
computed by two systems: by products and by industry. Percentages by product indicate
the share of tourism consumption by individual product whereas percentages by industry
present the share of tourism consumption in relation to the total production by individual
industries. which may offer a bundle of tourism products. The percentages of tourism
sales to total supply by products of accommodations. aviation transportation. car rental.
travel service and food service were 0.94. 0.76. 0.85. 0.92. and 0.37. respectively. In
terms of percentage contribution oftourism sales to the total supply by industries on

accommodation. aviation transportation. car rental. travel service and food service were

 

10 . . . . . .
Only counting expenditures that were Incurred rn Taiwan before and after the trip abroad.

29

0.73. 0.5. 0.82. 0.9. and 0.35. respectively. Tourism value added was NT $197.74 billion.
corresponding to 2.58% of the total GDP in Taiwan. Tourism activities supported
221.974 fulltime equivalent jobs. with the majority in the food service. ground

transportation. hotel and retail industries.

Table 11. Comparison of Taiwan Tourism Satellite Accounts from Two Reports

 

 

 

World Travel World Travel

Taiwan Tourism Bureau & Tourism Councila & Tourism Council

Year 1996 2002 2012
GDP (NT billion's) $197.7 $106.2 $220.6
Employment (000's) 222.0b 172.9 192.8
Pct oftotal GDP 2.6% 1.1% 1.1%
Pet of total employment - 1.8% 1.7%

 

Government and prrvate Investments rn tourism-related actrvrtres are excluded so that the estimates are
comparable across two studies.

Full time equivalent.
Source: Taiwan Tourism Bureau (2002) and World Travel and Tourism Council (2002)

Tourism Economic Impact Estimation Model
There are four main evaluation tools for estimating tourism economic impacts:
Tourism Satellite Accounts (TSA). tourism multipliers. Input-Output (I-O) analysis. and
Computable General Equilibrium (CGE) models (C orcoran. Allcock. Frost. & Johnson.
1999; Song. 2002). Each analytical tool requires different data input and has distinct

advantages and limitations.

Tourism Satellite Accounts
Tourism Satellite Accounts are recommended by the World Travel Organization
(WTO) as a standard method for measuring the contribution of the tourism industry to

national Gross Domestic Product (GDP). final consumption. and gross fixed capital

30

formation (private investment) (Frechtling. 1999). This method mainly computes the
direct effects of total visitor spending. business investment. non-profit organizational
operations. and governmental expenditure. The underlying concept is to address the
balance between demand and supply of tourism commodities. the number of jobs
depending on tourism as well as the indirect taxes generated by the sales of goods and
services to tourists within the framework ofa national accounting system (Lapierre &
Hayes. 1994).

TSA is widely adopted at the national level and the results are comparable across
nations. consistent through time. and compatible with the standard measures of a national
economy. However. despite its wide—spread applications at the national level. drawbacks
ofthis method primarily rest on debates over definitions and quantifications oftourism
activities. goods. and services (Frechtling. 1999; Quevedo. 2002). Other criticisms
include the inability of TSA to monitor tourism activity within more narrowly defined

user segments. and also. most TSA models do not include indirect and induced effects.

Tourism Multipliers

Tourism multipliers. I—O analysis. and CGE share a common theoretical basis.
Each model examines the relationship between an injection of spending into an economy
and the resulting change in economic activities from an equilibrium perspective. The
difference between these three methods is mainly determined by the model specifications
in terms of endogenous variables and assumptions. The tourism multipliers method is the
first developed analytical tool and has not been used in recent applications (Archer.
1977). The primary technique to estimate the value of multipliers is through the

Keynesian approach. which takes into account investments in the local region and the

31

marginal propensity to consume locally produced goods (Archer. 1978; Armstrong &
Taylor. 2000; Wagner. 1997).

Limitations ofthe model arise from applying an aggregate multiplier to a region
without differentiating unique patterns of spending between visitor segments. such as
cruise passengers vs. backcountry campers. and ignoring the variation and magnitude of
leakages during each round of transactions (Archer & Fletcher. 1988). Although
improvements have been made by disaggregating multipliers to the sector level based on
individual export ratios. the most sophisticated economic base models only produce
multipliers for ten or fewer sectors (Krikelas. 1992). In terms of (practical applications.
tourism multipliers are overly simplified in terms of the use of aggregated multipliers.
and the modeldoes not provide the complete information by individual user groups for

policy making.

Computable General Equilibrium

Computable General Equilibrium (CGE) model is the most advanced tourism
economic impact model in terms ofthe number of variables and the comprehensiveness
of assumptions. C GE models employ tight theoretical structures. Households. producers.
importers. exporters. and distributors are modeled separately as individual agents whose
behaviors follow the rules of utility maximization and cost minimization (McDougall,
1995). In general. the multiplier effects estimated by C GE are more conservative than 1-0
models as the original I-O assumptions of infinite supply and constant price in
transactions are relaxed (Dwyer. Forsyth. Madden. & Spurr. 2000; Dwyer. F orsyth. &
Spurr. 2003; West. 1995). Additionally. C GE catches feedback effects. integrates

possible supply constraints on labor or capital. and considers opportunity costs of

resources across all sectors. These factors typically create negative impacts elsewhere in
the economy and are reflected in the CGE outputs (Dwyer. Forsyth. & Spurr. 2004). For
example. CGE model can simulate the shifting of labor from the manufacturing sectors to
the service sectors or changes of labor cost and price indexes between imported and
exported products. A ten percent increase in world demand for Australian tourism.
demonstrated through the C GE model. will lead to negative impacts on the output and
employment of mining and machinery sectors (Dwyer et al.. 2003). The reason for the
negative impacts is that a higher tourism demand would drive up the consumer price
index. exchange rates and labor cost. which in turn decrease the competitiveness of
exports and generate negative impacts on the export-oriented industries.

Although the use of C GE models has been advocated in the tourism evaluation
process (‘Dwyer et al.. 2003. 2004). real world examples are uncommon. The major
criticisms ofCGE are the technical complexity. complicated interpretation of results. and
the amount oftime and data required in the model building (I—lunn & Mangan. 1999;
Rose. 1995; West. 1995). Results ofC GE are sensitive to the assumptions and parameters
used because the system is composed of extensive inter-relationships between each sector
and intra-relationships within itself. Some parameters. lacking the support of historical
data. are adopted from best-guest values. studies conducted in other countries or outside
sources (Blake & Sinclair. 2002: Dixon. 1994; Rose. 1995; West. 1995). For example.
the production function of Constant Returns To Scale (CRTS) is generally assumed in the
C GE models. in which. the scale of economy is assumed to have no influence on its cost
and output structures (Blake. 2000; Dwyer et al.. 2004; Woollett. Townsend. & Watts.

2003). One approach to mitigate the concern over the accuracy of parameters in the C GE

33

models is to adopt sensitivity analysis on key parameters and to test the relative
influences on outputs from different parameter specifications.

The high demands ofdata input and complexity of computations make CGE
models somewhat impractical for tourism applications. Applications of tourism CGE
models are primarily at the national level. mainly in Australia. United Kingdom. Canada.
and United States (Dwyer et al.. 2003). Published examples of tourism CGE models are
Adams. Dixon. McDonald. Meagher and Parmenter (1994) for the Australian economy;
Zhou. Yanagida. Chakravorty and Leung (1994) for Hawaii; Blake (2000. 2001) for
Spain; Dwyer. Forsyth. Madden and Spurr (2000) for multi-regions in Australia; Cooper
and Wilson (2002) for United Kingdom. Blake and Sinclair (2002) for the United States.

and Sugiyarto. Blake and Sinclair (2003) for Indonesia.

Input-Output Analysis

Input-Output analysis has been used extensively since the 1970’s to evaluate the
economic impacts oftourism development (Archer. 1977. 1978. 1984; Fletcher. 1989).
Advantages of I-O analysis are 1) its flexibility to evaluate economic impacts from the
equilibrium perspective based on different temporal and spatial scales. 2) its ability to
monitor the economic impacts of individual groups who have distinct spending patterns.
3) the ease of interpretation and convenience of data availability. and 4) its ability to
assess the importance of tourism in terms of what proportion ofGDP is attributable to it
(Blake. 2000; Briassoulis. 1991; Fletcher. 1989). Applications of tourism I-O analysis
range from the evaluation of national tourism economic impact (Archer & Fletcher.

1996) to regional tourism development (Eriksen & Ahmt. 1999); and from year-round

34

recreation activity (Stynes & Sun. 2003) to short-term events or festivals (Upneja. Shafer.
Seo. & Yoon, 2001).

Like other economic models. [-0 analysis involves assumptions to simplify
calculations. These assumptions limit the I-O model to simulate the real world operation
in terms of the dynamics of prices or the negative effects on the rest ofthe economy. such
as labor shortages. Also. [-0 analysis is insufficient to model the more extensive
economic issues such as taxation. welfare distortions or exchange rate inflation
(Copeland. 1991). Arguments in adopting I-O models over C GE models are first that an
I-O model may produce similar results as the C GE when it is appropriately formulated
(McGregor. Swales. & Yin. 1996). McGregor et al. (1996) demonstrated that a detailed
calibrated I-O model. which incorporates the infinite elasticity of labor supply and
exogenously fixed interest rates and import prices. will produce similar outcomes as CGE
models in the long run. Furthermore. I-O models work well for small and open
economies in which the relative price is set by the outside world and the supply of labor
and resources are unlimited because of imported materials and immigration workers
(Hunn & Mangan. 1999: Rose. 1995: West. 1995). Blake (2000) claims that CGE is a
better tool for comparing “with/without“ or “before/after" scenarios. while fixed-price
input-output models are more appropriate for evaluate the relative contribution of tourism

activities to the overall economy.

Operation of Input-Output Model
[-0 analysis was originally proposed by Wassily Leontief (1936). The theoretical
background is based on the concept of the Marxian and Walrasian general equilibrium.

which states that all transactions in an economic system are balanced and can be

quantified. The transactions of"sell" and “buy” activities by industries in a defined
region during a given period oftime are presented in an I-O table in monetary or physical
units. The interdependence of industries is calculated from an 1-0 table. and the resulting
impacts on each sector can then be computed for a given final demand change.

Economic entities are divided into a number of sections in the [-0 table. including
industries (intermediate). final demand. and final payment. Businesses are aggregated
into a number of industries (sectors) where input (ingredients) and output (sales) flows
can be traced. Rules are imposed to compile the I-O table. First. the total output of each
industry is equal to the intermediate demand from other industries plus the final demand.
Similarly. the total input each industry relies on is equal to the intermediate inputs from
other industries plus final payments. Lastly. for each industry. total inputs equates to total
outputs. These relationships can be presented in a matrix format (Equation 4).

X = AX + Y . (Equation 4)

Where X = Total output

A = Technical input coefficients
Y = Final demand

Equation 4 could be further modified as follows

x —AX = Y
(I-A) x =v
x = (I-A)"Y = BY (Equation 5)

Where (l-A)‘l or B matrix is the Leontieflnverse Matrix

36

The Leontief Inverse Matrix in Equation 5 represents the interrelationships of
transaction flows between industries. When industry A produces one unit of output. it
will require inputs from industry B and C. which in turn stimulates sales in industry B
and C. This interdependence produces the secondary effects as money circulates in the
region. which can be calculated as the indirect and induced effects (type I and type II
multipliers)I '. The economic impacts resulting from changes in final demand are then
assessed by multiplying the Inverse Matrix by the changes in final demand (Equation 5).
Impacts are generally expressed in terms ofjobs. personal income and value added

(Equation 6-8).

Jobs = E BY (Equation 6)
Income = 1 BY (Equation 7)
Value added = 17" BY (Equation 8)

A A A

Where E . 1 . V respectively. are the diagonal matrix ofjob to sales ratio. income to

sales ratio. and value added to sales ratio.

Three principal assumptions employed in the 1-0 analysis are l) the output of
each sector is produced with a unique set of inputs. 2) the amount of inputs required is
solely determined by the level ofoutput. and 3) there are no capacity constraints in the
production process (Miller & Blair. 1985; Otto & Johnson. 1993). Based on these
assumptions. economic impacts are solely determined by final demand changes.

Multipliers and economic ratios of jobs. personal income and value added are assumed to

 

H Ifthe household consumption is included as an endogenous variable in the model. the
multipliers represent the direct. indirect and induced effects (Type II multipliers).

37

be constant. Tourism however experiences frequent fluctuation in supply and demand.
Standard I-O analysis. in this instance. may not be appropriate for evaluation. as it does
not differentiate cost structures for service firms. for example. operating at 40% capacity

versus those operating at the 80% capacity.

Capacity Utilization and Tourism Economic Impact Analysis
Factors that influence the production function at different level of operating
capacities include 1) the price effects regarding material cost. labor wages. and final
outputs. 2) economies of scale in factor inputs. 3) substitution between labor and capital.
and 4) capacity constraints in the production process (Armstrong & Taylor. 2000:
Bryden. 1973: Fletcher & Snee. 1989; Miller & Blair. 1985; Porter. 1999; West &
Gamagc. 1997). The following section describes how each of these factors influences the

sensitivity of tourism economic impact estimates with respect to capacity utilization.

Price Effects

The stability of the production function and I-O coefficients are mainly
determined by four factors: price changes. technological changes (TC). returns to scale,
and trade patterns (Rose & Miernyk. 1989; West. 1995). The influence of price
movements on the 1-0 structure can be directly evaluated through the price-flex model
proposed by Moses (1974). Moses‘ model indicated that outputs in the I-0 table are
generally expressed in value terms. which can be further distinguished in terms of price
and physical quantity. As demonstrated in Equation 9. each value coefficient in the 1-0
table is the underlying physical input coefficient multiplied by a relative price ratio of

input material and final product. Assuming the input coefficient for material i to product j

38

1.

is 0.3 for the base year. the coefficient will change to 0.25 (=0.3* l—l—) if price of productj
.3

increases 30% and the price of material i increases 10% from the base year point.

 

xi: 01"“ Pi
Xij = (Iij* Qj* Pr
X.. q.. Q . P. P
ii = U = U J 1 Z cIii * —1 (Equation 9)
X1 0ij ‘ Pi

Where

Xj = Output (sales) of sectorj

Xij = Output (sales) of sector i required to produce one unit output of product j

Q = Quantity of product j

P; = Price ofproduct i : Pi = Price of productj

ouj = Technical input coefficient = Output of industry i that is bought by industry j to
produce one unit of product j

qij = Physical input coefficient = Physical output ofindustry i required to produce a

physical unit output of industry j

Traditional I-O analysis assumes that the physical input coefficients (qij) remain
constant and price ratios between inputs and the final product are unchanged during the
policy period (Miller & Blair. 1985). This assumption is more realistic for manufacturing
sectors as the product price is generally determined by the outside world. and material

and labor are imported if domestic resources are constrained (Hunn & Mangan. I999:

39

 

Rose. 1995; West. 1995). However. time-perishable services. such as the lodging.
amusement. and transportation sectors. generally employ a strategy of price
discrimination to maximize profits. In this case. the output price is based on levels of
demand and unused capacity (Arenberg. 1991; Kimes. 1989). With a price discrimination
strategy. I-O technical coefficients will change primarily depending upon the sensitivity
of the price ratio between input material and the final product (Hudson & Jorgenson.
1974; Moses. 1974). A similar reasoning can also be applied to argue that the stability of
income to sales ratios and jobs to sales ratios in the 1-0 framework will be directly

influenced by the relative price change of final output.

Economies of Scale and Substitutions

Gosh (1958) points out that the matrix coefficients observed in the 1-0 table not
only express a snapshot of the production/allocation function of an individual industry at
the survey period. but also indicate how businesses perform at a given level of demand
and supply. When demand is shifting in greater extent. the producers of a specific sector
would then experience economics of scale (Miller & Blair. 1985). That is. as the price of
input factors are fixed. the input per dollar of output falls as the scale of production
increases (Lin & Lin. 2000). When economies of scales are experienced. [-0 technical
coefficients as well as job to sales ratio and personal income to sales ratios must be
modified to reflect changes in the input-output relationship.

[-0 technical coefficient. Bryden (1973) was the first study to indicate the
quantitative differences among hotel revenue and cost structures at different utilization
(occupancy) rates. He used financial data ofa single hotel in Antigua to simulate the

revenue and cost structures at a 45% occupancy rate versus a 65% occupancy rate. By

40

assuming a constant room price. the cost increased less than proportionately with revenue
when occupancies were up. This in part can be explained by economies of scale. as
marginal cost is lower than average cost for an additional unit sold.

Arranging the projected financial data of a hotel into a national accounting
framework. Bryden (1973) further demonstrated the differences of Gross Domestic
Product (value added) multipliers at 31%. 45%. 55%. and 65% occupancy rates for the
accommodation sector. By assuming constant price indices and fixed I-O coefficients for
the rest ofthe economy. the value added multipliers of the hotel sector decreased from
0.684 at a 31% occupancy rate to 0.644 at a 65% occupancy rate. The personal income
multipliers decreased from 0.490 at a 31% occupancy rate to 0.443 at 65%. and the gross
profit multipliers increased from 0.194 to 0.201. The decreasing value added multiplier
effect is mainly due to increasing import contents and decreasing labor costs as demand
rises. These results support Bryden‘s statement that “the assumptions regarding
utilization of capacity are likely to be important in determining the input-output
coefficients in the hotel industry" (p.126).

Jobs. Standard I-O models have generally been criticized for overestimation in
employment effects because they assumed existing labor was fully utilized so that an

increase in external demand would lead to a proportional increase in the level of
employment (Fletcher & Snee. 1989: Porter. 1999'. West & Gamagc. 2001; West. 1995).
However. in reality. labor is rarely at full utilization levels and employers will generally
attempt to increase output without corresponding proportional increases in employment
numbers. West (1995) claimed that “service—type industries are better able to support an

increase in tourist activity largely within existing resources. whereas manufacturing-type

41

industries. which have more rigid production structures. respond in a manner closer to
that of the Leontief production system“ (p.223). It can be inferred from West’s research
that, in service sectors. production costs do not rise in direct proportion to sales when
demand increases because certain levels of outputs can be accommodated by better use of
the existing work force. This also implies that the level of employee expansion would
depend upon the current level of utilization in existing workers.

Economies of scale are generally accompanied by substitution of primary factors
in service-type industries. especially between employment and capital. For example. a
greater labor productivity can be achieved through by subsidizing existing workers for
the extra working hours/efforts (Krakover. 2000; West & Gamage, 2001). In the example
of Grand Prix in Australia (Hatch. 1986). the study first used an 1-0 model to estimate
total economic impacts from visitor spending at a car-racing event and then surveyed
local businesses to validate the estimated results. The study revealed that no new
permanent jobs were generated in the transportation sector as a result of the Grand Prix.
and for some operations. the workload was rearranged so that even overtime pay was not
required (Hatch. 1986). Hatch (1986‘) concluded that profits and value added components

increased at higher utilization levels at the expense of employment.

Capacity Constraints

The third factor that explains the possible overestimation in the 1—0 model is the
supply constraints of domestic traded goods. Supply constraints or bottleneck conditions
generally result in increasing import requirements when businesses are operating at full
or near capacity. The problem was first raised by Bryden (1973). who indicated that the

domestic agriculture production in Antigua cannot satisfy the increasing demand from

tourism developments. and therefore substantial imported agriculture products were
observed. To account for the leakage effects in the I-O model. Bryden (1973) totally
removed the agriculture sector in the l-O technical coefficient matrix and included it in
the imports vector. The result was a lower sales multiplier due to higher import contents.
O‘Hagan and Mooney ( 1983) experimented with tourism multipliers based on
different supply constraints for multiple traded sectors in Ireland. The tourism income
multipliers decreased from the optimum 0.86 when all sectors were able to meet extra
demand to 0.51 when both agriculture and food productions were constrained by
domestic supply. By incorporating joint capacity constraints. the reduced income
multipliers. due to leakage effects. behaved in a non-linear manner. depending on the
nature of the linkages within the l-O table. A similar study was also found in Wanhill
(1988). who examined the relationship between leakage effects and multipliers when
multiple sectors were operating at different capacities. I-Ie simulated the income and
employment multipliers for Mauritius when the agriculture and manufacturing sectors
were assumed to operate at zero. 25%. 50%. 75%. and 100% of available capacity. The
overall income multiplier was 0.69 when both industries operated at zero available
capacity. increased to 0.78 at the 50% capacity. and to 0.96 when both operated at the

100% available capacities.

Summary

This chapter has reviewed four major tourism economic impact models and the
detailed methodological specification of the Input-Output analysis. In the standard I-O

model. fixed input costs. labor expenses. and profitability are applied in computing total

43

 

impacts regardless of the sales volume or operating capacities. This assumption has
resulted in the exclusion ofprice effects. economies of scale. substitution between input
factors and possible capacity constraints in the I-O framework. For all the cited
references in this section. Bryden‘s (1973) is the most comprehensive tourism economic
impact study that exploits the concept of capacity utilization within the 1-0 model. IIis
study considered the economies of scale for the accommodation sector at different
occupancy levels and supply constraints for agricultural products. Differences with
respect to the standard l—O model were demonstrated in terms of sales multipliers and
value added multipliers.

Although Bryden‘s study is pioneering. refinements are called for to address some
of his model limitations. The first limitation is that the model employs constant price
indexes for all input and output factors. Room prices. average personal wages. and
intermediate product price are assumed constant and are not influenced by the level of
capacity utilization. Second. his data are based on a single hotel and its validity to
represent the overall accommodation sector is in question. Third. the study only briefly
discusses the changes of sales multipliers and value added multipliers from the
perspective of [-0 technical coefficients. Stability of thejobs to sales ratio and personal
income to sales ratio are not explicitly discussed. Economies of scale in labor and

substitutions between labor and capital inputs are not included in the model.

44

lux

CHAPTER III METHODS

The methods chapter presents the data sources. treatments. and the estimation
process. Five secondary datasets are used in the analysis including. the “Inbound Traveler
Consumption and Trends Survey (.ITCTS)". the “Domestic Travel by Domestic Residents
Survey (DTDRS)"‘. Taiwan tourist hotel operation data. the 1999 Taiwan Input-Output
table. and the "Manpower Survey Statistics in Taiwan Area". The first two datasets are
used to estimate trip characteristics. spending averages. and the propensity of hotel stays
for inbound and domestic tourists. These parameters are required to compute total
inbound tourism visitor spending. and hotel rooms demanded from domestic and
international tourists.

Tourist hotel data is adopted to establish the econometric models for jobs to sales
ratio and income to sales ratio with respect to occupancy rates. These equations specify
the relationship between capacity utilization and cost structures in jobs and income. Last.
the I-O table and manpower survey are used to compile a national Input-Output model for
Taiwan. Multipliers with fixed economic ratios based on standard I-O assumptions are
first generated. Distinct sets of multipliers with varyingjob and income ratios are
computed separately by incorporating the predicted job and income ratios from
econometric equations. Comparison of impact estimates of the DTAP policy is calculated
using I-O models with fixed and varying ratios under three scenarios. Detailed methods

in estimating each objective are listed below.

45

Objective 1
The purpose ofobjective I is to estimate the economic impact ofinbound tourism
to Taiwan in 2001. The estimation process follows a standard Input-Output Analysis.
Final demand is calculated first by multiplying the average visitor spending with total
number of international arrivals. Economic impacts are then computed by applying total

visitor spending to a national Input-Output model of Taiwan.

Average Visitor Spending

Data sources. The "Inbound Traveler Consumption and Trends Survey (ITCTS)"
was used to estimate average visitor spending. ITC TS is periodically conducted by the
Taiwan Tourism Bureau to estimate spending profiles and trip characteristics for
international visitors to Taiwan. The survey is conducted monthly at international airports
in Taipei and Kaohsiung. collecting approximately 5.000 cases each year. The sample is
stratified in proportion to the overall visitors admitted to Taiwan each year by country of
residence. The database used in this analysis contains interviews conducted in 2000 and
2001. a total of 10.006 cases.

ITCTS measures visitor spending on a person trip basis. differentiating prepaid
expenses12 and expenses incurred in Taiwan. Visitors are asked to report an aggregate
prepaid total and individual expenses in Taiwan within six categories: hotel”. food.

. . . 14
transportation. entertamment. shopping . and other expenses.

 

‘7 .
“ Exclude airfares.
'3 Expenses related to lodging. food. laundry and personal services inside the hotel.

'4 For the 2001 survey. visitors were asked to report their shopping expenses in 9 categories.
including clothes.jewel. souvenir. cosmetics. local products. tobacco/alcohol. Chinese herbs. and

others.

46

 

Data treatment. Variables selected from the database to perform the spending
analysis were pre-paid expenses. expenses in Taiwan within six categories. length of stay
in Taiwan. party size. lodging types. country of residence. and trip purpose. Cases were
examined first to detect outliers and invalid cases. Cases were excluded from the
spending analysis if 1) all spending questions were skipped. 2) the length of stay was
longer than 60 days”. 3) party size was larger than 10 persons. or 4) no spending was
reported.

Expenses were initially reported in the currency of the visitors‘ origin and then
converted to US. dollars. Spending per party per night was estimated by multiplying the
per person per trip expense by the average party size and dividing by the average length

of stay. Separate spending profiles were estimated by country of residence.

Total Spending of International Visitors

Total visitor spending was computed by multiplying l) the average spending per
person per day. 2) the official length of stay published by Taiwan Tourism Bureau in
2000-01. and 3) total international arrivals in 2001 by country of residence. Total visitor
spending was allocated to six service sectors and ten manufacturing sectors based on the
percentages in Appendix A. This procedure bridges the visitor spending to sectors in the

Taiwan I-O table.

 

'5 This was the standard cut-off point in the official report by Taiwan Tourism Bureau (Taiwan
Tourism Bureau. 2002).

47

Taiwan Input-Output Model

The 1999 Taiwan I-O table (Directorate-General of Budget Accounting and
Statistics. 2003) was used to compile type I and type 11 sales multipliers. personal income
to sales ratios. and value added to sales ratios. The 1999 Taiwan I-O table has 160 sectors
that were further aggregated into 39 sectorsl6. including 18 service sectors and 14
manufacturing sectors. Type I and type 11 sales multipliers were calculated as specified in
Equation 10. Type 11 sales multipliers were calculated with the household sector

endogenously included in the table.

B = (I-A)" (Equation 10)
Where B = Leontief inverse matrix
A = Technical input coefficient of39 industries

1 = Identity matrix

Regional purchase coefficients (RPC)l7 were computed as one minus the
percentage of imported goods18 for each manufacturing sector. Trade margins and
transportation margins for goods were estimated using the 1996 producer price I-O table.
w’hich separates trade and transportation costs from production costs. Ratios were
computed by dividing the trade and transportation costs by the total domestic sales for

each manufacturing sector.

 

‘6 See Appendix B for sector definition and category.

'7 The proportion ofthe regional demand for a good or service that is fulfilled by regional
production.

'8 Sum of territorial import and non-territorial import.

48

 

The “Manpower Survey Statistics in Taiwan Area” was used to establish jobs to
sales ratios. The report includes sole prOprietors and jobs in the government and private
sectors (Directorate-General of Budget Accounting and Statistics. 2003). Jobs estimated
from the Manpower Survey Statistic were grouped into the 39 1-0 sectors“). The jobs to
sales ratio for each l-O sector was computed by dividing the jobs by NT$ sales (in
million‘s). The type I and type II multipliers ofjobs. personal income. and value added

were calculated as Equation 1 1-13.

Jobs = E BY (Equation 11)
Income = 1 BY (Equation 12)
Value added = 1; BY (Equation 13)

Where E . l .1" respectively. are the diagonal matrix of job to sales ratio. income to

sales ratio. and value added to sales ratio.

Total Economic Impacts
Total economic impacts were computed by applying total visitor spending to the
Taiwan I-O model. The calculation process was carried out in three steps. First. sales for

retail trade and domestic transportation were computed by multiplying total visitor

 

'9 The 1-0 table includes hotel restaurant sales within the “food and beverage” sector while the
Manpower Survey counts employees working in hotel restaurants within the “hotel services”
sector (Directorate-General of Budget Accounting and Statistics. 2003. 2003). For consistency
with the [-0 accounts. hotel employees working in the restaurant division were moved from the
hotel sector to the food and beverage sector in the Manpower Survey Statistics. The estimates of
restaurant employees in hotels were taken from data reported by tourist hotels in Taiwan. It is
assumed that regular hotels (small-scale establishments) do not provide food and beverage
services. and therefore no employees are reported in the restaurant division.

49

expenses in manufactured goods by the trade and domestic transportation margins.
respectively. These expenses were later individually applied to multipliers for the retail
trade and transportation sectors. Direct sales were computed by excluding imported
manufactured goods (Equation 14). Finally. total direct sales were multiplied by the type
I and type II multipliers to compute total sales. jobs. personal income and value added.
Economic impacts were calculated for 39 I-O sectors. Results were summed and
presented in 14 major categories to facilitate interpretation and presentation (Appendix
B).

Direct sales = Z Expenses of non-traded goods + Z (Expenses ofmanufactured goods

* regional purchase coefficients) (Equation 14)

Objective II
The purpose of objective II is to test the stability ofjobs to sales ratios and income
to sales ratios with respect to occupancy rates. Econometric relationships between these
two ratios and occupancy rates were established using time series data of tourist hotels in
Taiwan. Predicted job and income ratios were then applied to the Taiwan I-O model to
re-compute multipliers by different occupancy rates. These distinct sets of multipliers
were then multiplied with the 2001 inbound tourism visitor spending to re—compute total

impacts in Taiwan.

Econometric Relationships
Two panel data sets. provided by the Taiwan Tourism Bureau (1999-2003). were

employed in the estimation ofjobs to sales ratios and income to sales ratios. Econometric

50

models were constructed to determine the relationship between occupancy rates and (1)
jobs per NT$ million hotel sales. and (2) personal income per NT$ million hotel sales.
Geographic location. consumer price index. and hotel scales were included as control
variables in the regression equation to account for differences between hotels (Liu & Var,
1982; Pan). Locations were introduced as dummy variables representing northern
Taiwan, central Taiwan, southern Taiwan and eastern Taiwan (Figure 3). The consumer
price index (Directorate-General of Budget Accounting and Statistics, 2004) was
incorporated to capture changes in jobs to sales ratios and income to sales ratios due to
general price inflation. Hotel scale was measured as the number of rooms in each

establishment.

 

Tarp er

  

Northern Taiwan

 

 

 

Figure 3. Four Geographic Regions, Taiwan

51

Jobs to sales ratios. The econometric relationship between hotel occupancy rates
and jobs to sales ratios was estimated using the yearly firm level operational data for
tourist hotels in Taiwan from 1999 to 2003 (Taiwan Tourism Bureau, 1999-2003).
Variables included in the data set were the average room price. occupancy rate. total
available rooms. total employeeszo. and total revenuezl per month. The dataset included
4.751 cases of monthly data from 75 international tourist hotels and 20 tourist hotels over
the 5-year periodzz.

The data set was examined first for data consistency and accuracy. Cases with
incomplete information on key variables. such as sales and occupancy rate. were deleted.
Monthly hotel operational data were aggregated to compute yearly totals for each
individual establishment. Aggregating to a yearly basis controls for seasonal fluctuations
in employment and sales.

After aggregation. 361 observations representing 83 hotels over a 5 year period
were available covering average room price. occupancy rate. total employees and
revenues. As noted above. restaurant sales and employees within hotels were excluded

from the hotel sector to ensure that the computed jobs to sales ratios23 correspond to the

definitions ofthe hotel sector in the Taiwan I-O table.

 

20 Total jobs were differentiated by employees working at the room division. food and beverage
division. administration division. and others.

21 Total revenue included the sales in rooms, affiliated restaurants, and others.

22 International tourist hotels reported 3.638 cases and tourist hotels reported 1.1 13 cases during
1999 to 2003. Due to closures and renovations. not all hotels had complete monthly data for the
five year period.

33 Jobs to sales ratio = [(Employees from the room division and managers) / rooms sales (NT$
million‘s)]. Number of managers was computed by multiplying the number of employees
working in the administration division with a proportion. representing the room sales to total
hotel sales.

Ur
L)

 

Three dummy variables were included in the equation to represent the central.
southern. and eastern Taiwan (Equation 15). The dummy variable representing northern
Taiwan was excluded to avoid perfect multicollinearity. Regression analysis was
perfomied within the Statistical Package for Social Science (SPSS) and Stata using the
stepwise approach. The functional form was determined by comparing adjusted R-square
values across 1 1 functional forms (curve-estimation process in the SPSS program).
Multicollinearity and autocorrelation were examined during the analysis. Corrections

were performed when needed.

Jobs to sales ratio = on + f (occupancy rate. rooms per hotel. consumer price index,
central Taiwan. southern Taiwan. eastern Taiwan) + 80 (Equation 15)

where 0.018 the intercept and so is the error term with normal distribution

Income to sales ratio. The econometric relationship between hotel occupancy
rates and personal income to sales ratios was estimated using the yearly firm level
international tourist hotel financial data from 1999 to 2002 (Taiwan Tourism Bureau.
1999-2003). Income statements. balance sheets. and other related operating data are
reported yearly by all international tourist hotels in Taiwan. Cost information is itemized
by labor. material (food or laundry expenditure). energy (water. electricity and fuel) and
others (marketing. rent or business services). A total of22l cases (annual averages) from

59 hotels were reported over the 4-year span.

53

The same procedure of the regression analysis. as described above, was used to
analyze the relationship between personal income to sales ratios and occupancy rates

(Equation 16).

Income to sales ratio = a. + f(occupancy rate. rooms per hotel. consumer price index,
central Taiwan. southern Taiwan. eastern Taiwan) + 8. (Equation 16)

where or) is the intercept and e] is the error term with normal distribution

Predicted Economic Ratios and Multipliers

Economic ratios. The predicted jobs to sales ratios and income to sales ratios for
tourist hotels were estimated using econometric equations under different occupancy
rates. Proportional changes in these two economic ratios from the base point (year 1999)
were computed. and then applied to adjust the hotel job and income ratios in the 1999
Taiwan I-O table.

[-0 multipliers. To calculate multipliers with varying ratios. the first step was to
regenerate type 11 sales multipliers by adopting the predicted income to sales ratio in the
accommodation sector. Multipliers of 39 1-0 sectors were then recalculated by replacing
the fixed job and income coefficients with the predicted job and income figures under
different occupancy rates. These distinct sets of multipliers were applied to the 2001
inbound tourism visitor spending to recompute total impacts in Taiwan by occupancy

rates.

54

 

Objective III

The purpose ofobjective III is to compare economic impact estimates of the
DTAP policy under the three scenarios based on 1) standard I-O models with fixed
multipliers. and 2) I-O models with varyingjobs to sales ratios and personal income to
sales ratios with occupancy rates. Each scenario is defined by international visitor volume
and hotel lodging capacity to simulate altemative conditions of demand and supply of
tourism activities.

The estimation process involves two primary models as illustrated in Figure 4.
The ratio prediction model generates specific job and income ratios based on the
predicted occupancy rate. which is determined by volumes of inbound travel. domestic
travel and hotel capacity. Given the predefined demand and supply levels in each
scenario, average occupancy rates are calculated and then applied to the econometric
equations (objective II) to obtainjobs to sales ratios and income to sales ratios.

The input-output model computes total economic impacts. Final demand changes
are calculated by multiplying average spending by the number ofinternational visitors.
Total visitor spending is then applied to fixed and varying multipliers. Fixed multipliers
are computed using the Taiwan l-O model with averagejob and income coefficients. The
varying multipliers for each scenario are individually computed using the predicted job
and income ratios from the econometric equations.

Four major inputs are used in the estimation: the spending averages of
intemational visitors. the volume of inbound tourism. the volume of domestic tourism.
and tourist hotel capacity. In this analysis. average spending and the level of domestic

tourism are assumed constant across scenarios. The changing variables for each scenario

 

are the number of international arrivals and hotel capacity. A lodging capacity of 20.697
rooms in 2008 is assumed in scenario 1. The maximum volume of international tourists is
estimated based on hotel capacity. Scenarios 2 and 3 assume an increase in lodging
capacity to 37.900 rooms. and that international arrivals grow to 75% and 100%.

respectively. of the DTAP 4.8 million vitiation volume.

Input-Output Model ti redictio 0 e1
g‘*....;.’:;v.:;..\ H 7 T“ x "A CT T" . , i 7’ T
p . b . g Total numberof Total number of. .
of lnternatlonal . . . . ._ . . . ‘ Hotel capac1ty ;
. . lntemational v151tors . domestic v151tors i‘
v151tors . _ .
ILLLYLLLE ,_ iiiyi if x“ ./ \.-'

 

Final Demand:
Estimate total visitor spending for inbound
tourism under the three scenarios

Average Occupancy Rate:
Estimate average occupancy rates
under the three scenarios

 

‘Lassar‘fl- 2. . .. TEL L L

 

 

 

 

1 Fixed Multipliers: . Varying Multipliers: l Prédictefl Econ9ml° Rat195=
1 Estimate [—0 5 Estimate 1-0 1 1 Estlmate job and meome ratlos
. multipliers based on multipliers based on i <1 :‘ usmg the econometrlc equatlons
: fixed jobs and income 1 varyingjobs and 1 based on the computed occupancy
1 ratios 1 income ratios . rates
., -_jzfi_.éi .L.._ .3 _,_ _-_ _.L__ ..__.,_.___1
5 1

.-2-,-§,,-s . ,, j s, L
“ Economic Impacts: 5 Economic Impacts: 1

Estimate economic " ‘ Estimate economic i

impacts based on impacts based on :
1 fixed multipliers l varying multipliers i
L -_LLMW_.LL ‘ 1

 

Figure 4. Estimation Process for Objective III

56

For [-0 models with varying job and income ratios. the volume of inbound
tourism serves two purposes. It determines the size of total visitor spending and also
influences the job and income ratios. which are adjusted based on occupancy rates. In
other words. economic impacts are determined by final demand changes as well as the

level ofcapacity utilization in the accommodation sector.

Ratio Prediction Model

The ratio prediction model involves two steps. In the first step. average occupancy
rates are estimated for each scenario. The second step then applies the computed
occupancy rates to the econometric equations (established in objective II) to obtain the

jobs to sales ratios and income to sales ratios.

()ccupc'lncv rates under the three scenarios

The average occupancy rates for tourist hotels were determined from 1) the
demand level- number of domestic and international tourists. and 2) the supply level- the
hotel capacity. Occupancy rates are computed for four geographic regions in Taiwan to
take into account possible regional capacity constraint.

Domestic Tourism. Domestic tourism is assumed to grow from 97 million person
trips in 2001 to 135 million person trips by 2008 for adults and children older than 12
years (Taiwan Tourism Bureau. 2003). The same level of domestic tourism volume (135
million person trips) and trip allocations to the four regions is assumed in all 3 scenarios.

Trip characteristics ofdomestic travel were obtained from the “Domestic Travel by

57

 

 

Domestic Residents Survey” (DTDRS)24 (Taiwan Tourism Bureau. 1998-2004). These
variables included average propensity to stay at tourist hotels. regional trips distribution.
party size. and average length of stay. Using these parameters, 135 million person trips
were first allocated to the four regions and then converted to total room nights at tourist
hotelszs.

Inbound tourism. Rooms demanded by international visitors were computed by
multiplying the average hotel stay per visitor by the total number of international arrivals.
The average propensity to stay at tourist hotels. regional trip distribution. party size and
length of stay are assumed constant for all 3 scenarios. These parameters were computed
from the “Inbound Traveler Consumption and Trends Survey” by country of residence.

Hotel capacity. Scenario 1 assumes a capacity of 20.697 tourist hotel rooms in
2008 with 58% ofthe capacity in northern Taiwan. The anticipated growth in the volume
ofinternational visitors is subject to room capacity constraints. It is assumed that when a
region reaches the maximum capacity (100% occupancy). inbound tourism reaches the
upper limit and additional international visitors are not allowed. Under current use
patterns. northern Taiwan will bethe first region to reach full capacity in tourist hotels.
Additional trips will not be feasible if northern Taiwan remains the major destination and

point ofentry and exit for international visitors. The shifting of excess demand for tourist

 

24 DTDRS is conducted bi-annually to Taiwan residents for their trips within Taiwan. and is used
to estimate trip characteristics for domestic tourism in this study. The dataset was obtained from
the Taiwan Tourism Bureau. which collected 12,504 cases for the year of 2001.

25 Domestic demand for tourist hotels (room nights)
0 4 o o o o o s a
= total person trips * E (percent oftrlp dlstrlbutlon by reglons * percent of tourlst hotel
region 2 l
stays * length of stay/ party size)

58

hotels from northern Taiwan to other regions is not permitted in the model. Scenarios 2
and 3 assume an expanded tourist hotel capacity of 37,900 rooms in 2008 with 69% of
capacity in northern Taiwan.

Average occupancy rates. Average occupancy rates for the 3 scenarios are
computed by dividing the total tourist hotel rooms demanded from domestic travel and

inbound travel by capacity of tourist hotels.

Predicted economic ratios

The estimated occupancy rates from the previous section were applied to the
econometric equations (established in objective II) to predict jobs to sales ratios and
income to sales ratios for tourist hotels under each scenario. Proportional changes in these
two economic ratios from the baseline in 1999 were computed. and then applied to adjust

the hotel job and income ratios in the 1999 Taiwan l-O table.

Input-Output Model

Economic impacts of inbound tourism were calculated using the input-output
analysis. Visitor spending is applied to the national l—O model with fixed and varying
ratios. Total visitor spending by country of residence was computed by multiplying 1) the
average spending per person per day. 2) the official length of stay published by Taiwan
Tourism Bureau in 2000-01. and 3) total visitation volume specified in objective lll.
Total visitor spending was then allocated to six service sectors and ten manufacturing
sectors based on the percentages in Appendix A.

Multipliers with fixed job and income ratios were reported under objective I. To

calculate multipliers with varying coefficients. the predicted job and income ratios from

59

the ratio estimation model were applied to the Taiwan l-O model. Type 11 sales
multipliers were first recalculated by substituting the predicted income to sales ratio for
the accommodation sector. Multipliers for the 39 l-O sectors were then computed by
replacing the fixed job and income coefficients with the predicted figures. The last step

applied final demand changes to the [-0 models with fixed and varying multipliers.

60

CHAPTER IV RESULTS

This chapter presents the results for each of the three study objectives.
The economic impacts of international visitor spending in Taiwan in 2001 (objective I)
are presented first to establish the baseline for the DTAP policy. Econometric equations
in estimating jobs to sales ratios and income to sales ratios (objective II) are discussed.
These equations specify the relationship between capacity utilization (occupancy rate)
and economic ratios injobs and income for tourist hotels in Taiwan. Predicted economic
ratios and multipliers at different occupancy rates are then provided. The final section
introduces the economic impact estimates ofthe DTAP policy under the three scenarios
to simulate different conditions of demand and supply of tourism activities (Objective
III). For each scenario. economic impacts are estimated using 1) the fixed economic
ratios. and 2) the varying job and income ratios with respect to occupancy rates in the 1-0
model. The magnitude and direction of differences in impact estimates between the fixed

and varying l-O ratios are summarized at the end of the chapter.

Objective I
Estimate the econon-zic' impacts ofinbound tourism on the Taiwan economy in 2001,

be are the implementation ofthe "Doubling Tourist Arrivals Plan.

Economic impacts of inbound tourism are calculated using the Taiwan Input-
Output model. Total international visitor spending is presented first. followed by the

Taiwan l-O multipliers and economic impacts of inbound tourism in 2001.

6]

 

Visitor Spending of Inbound Tourism

Average Spending. Visitors from J apan, Hong Kong/Macao and the “other”
region have the highest spending per person per day in Taiwan. representing US$ 233,
$188. and $173 respectively26 (Table 12). On a party trip basis. visitors from the “other”
region, North America and Japan are the top three spenders. Visitors from North America
spend more than 45% of their trip expenses on accommodations. whereas visitors from
Hong Kong incur half of their expense on food and shopping. In general. visitors from
North America and Europe stay longer and incur more spending per visitor in Taiwan
than Asian visitors.

Total visitor spending. Based on 2.6 million international visitors27 with an
average spending ofUS$ 1.361 per person per trip in Taiwan. inbound tourists spent US$
3.5 billion in Taiwan in 2001 (Table 13). Thirty-one percent ofthis spending is
contributed by Japanese visitors. followed by “other“ regions (27%). and North American
visitors (17%). Excluding airfares. lodging expenses account for 41% oftotal trip

expenditures. expenses for food are 18%. and shopping represents 14%.

 

2" Spending averages reported in this study are different from the spending averages published by
the Taiwan Tourism Bureau due to exclusion of outliers and calculation formula. Taiwan Tourism
Bureau directly computes the daily spending average per person without weighting the case with
respect to length of stay. which leads to overestimation of short-term visitors” spending. In
addition. the spending analysis conducted by Taiwan Tourism Bureau included all cases, covering
outliers in expense or length of stay. The spending averages published by the Taiwan Tourism
Bureau are approximately 15% higher than the averages computed by the author.

27 Foreign labors (approximate 200.000) are excluded.

62

Table 12. Inbound Tourist Spending Profiles by Country of Residence, 2000-01

 

 

 

 

 

 

 

Hong New
Kong/ Singapore North Zealand/
Japan Macao Korea ."Malaysia America Europe Australia Others
Spending per person day (USS)
Hotel 79.5 53.1 46.7 55.2 70.2 63.1 53.1 62.7
Food 50.4 48.5 24.9 31.9 35.8 34.8 30.4 39.6
Transportation 186 10.7 11.6 11.1 9.7 10.5 9.2 12.4
Entertainment 22.6 16.5 11.2 11.8 15.2 10.6 10.7 16.8
Shopping 44.9 46.3 19.7 14.2 14.0 15.6 13.6 22.9
Others l]._5 fig 1_2_.1_ 1 1.5 M m 1&9 13;]
Total 233.6 188.2 126.2 135.6 155.1 149.7 127.8 173.1
Spending per person trip (USS)
Hotel 382.8 271.0 255.7 452.3 700.6 492.5 422.1 960.0
Food 242.6 247.7 136.2 261.3 358.0 271.7 242.1 605.7
Transportation 89.7 54.5 63.3 91.2 96.6 81.6 72.9 189.8
Entertainment 108.5 84.1 61.4 96.8 151.8 82.5 85.0 257.7
Shopping 216.3 236.3 107.5 116.2 139.5 121.6 108.0 350.1
Others 8.1.3. 62.6 6.63. 9.42.3. 1923. .l_|.7_..6_ 86.3. 2313.
Total 1,124.4 961.2 690.4 1,112.1 1,548.5 1.167.4 1.016.5 2,648.6
Length of stay 4.81 5.1 1 5.47 8.20 9.99 7.80 7.95 15.30
Partysize 1.26 1.15 1.19 1.12 1.12 1.10 1.17 1.14
Cases 3.280 1.014 361 301 990 1.156 162 666
95% confidence interval of daily spending per person (USS)
Upper bound 244.4 203.9 144.0 152.2 169.2 174.5 183.8 194.3
Lower bound 222.8 172.5 108.4 119.0 141.0 124.9 71.8 151.9
Table 13. Visits and Total Spending By Country of Residence, 2001
Hong New
Country of Kong.“ Singapore/ North Zealand/ Total/
residence Japan Macao Korea Malaysia America Europe Australia Others Average
Person trips (000's) 977 435 86 156 392 149 38 364 2.597
Pet of person trips 38% 17% 3% 6°26 15% 6% 1% 14% 100%
Average spending
per person trip $1.124 $961 $690 $1.112 $1.549 $1.167 $1.016 $2.649 $1.361
Total spending
(US million's) $1.098 $418 $59 $174 $607 $173 $39 $965 $3,534
PCI Of spending 3 10/0 1 20"?) 294) 50/0 1 70/0 50/0 lo/o 270/0 1009/0

 

Taiwan Input-Output Model

Economic ratios and multipliers for tourism related sectors are displayed in

Tables 14 and Table 15. Across all service sectors. food and beverage services have the

63

highest jobs to sales ratio (1.84 jobs per NT$ million sales), followed by retail trade
(1.38). and theatrical & arts (1.09). “Food and beverage services” and “wholesale trade
and retail trade” have the highest personal income to sales ratio. Approximately half of
the sales in these sectors go to employee compensation. With the exception of apparel
and wood products. manufacturing sectors employ fewer people per million in sales and

spend a lower percentage of their revenue on wages and other value added components.

Table 14. Economic Ratios for Primary Tourism Sectors, 1999

 

 

 

Income to sales Value added Jobs/ million JObS/ millior;
ratio to sales ratio NT$ sales US$ sales
Services
Hotel services 0.41 1 0.649 0.497 16.891
Food & beverage services 0.527 0.707 1.841 62.567
Transportation 0.362 0.560 0.523 17.774
Theatrical and other arts 0.330 0.567 1.091 37.078
Recreational & cultural services 0.376 0.693 0.729 24.775
Other personal services 0.407 0.579 0.876 29.771
Wholesale trade 0.496 0.687 0.480 16.313
Retail trade 0.485 0.738 1.383 47.002
lntemational trade 0.339 0.599 0.433 14.716
Manufacturing

Process foods & beverages 0.108 0.245 0.264 8.972
Tobacco 0.1 16 0.731 0.038 1.291
Textile mill products 0.149 0.241 0.269 9.142
Wearing apparel and accessories 0.243 0.300 0.886 30.1 1 1
Leather & leather products 0.155 0.215 0.653 22.192
Wood products 0.271 0.391 1.153 39.185
Paper & paper products 0.239 0.360 0.427 14.512
Chemical manufactures products 0.125 0.275 0.212 7.205
Non-metallic mineral products 0.202 0.369 0.318 10.807
Iron and steel products 0.143 0.246 0.384 13.050
Machinery & electronic products 0.136 0.256 0.264 8.972
Transport equipment 0.141 0.308 0.258 8.768
Other manufacturing products 0.218 0.315 0.318 10.807

 

Note. Economic ratios and multipliers for 39 sectors can be found in Appendix C and D.
a One US dollar equated to 33.99 New Taiwan dollars in 2001 (OAN DA Corporation. 2004).

Type I sales multipliers for service sectors generally fall between 1.4 and 1.5.

while type 11 sales multipliers average about 3.5 (Table 15). With the exception of

64

tobacco. type 1 sales multipliers for manufacturing sectors are greater than 1.5 with some
above 2.0. Type 11 sales multipliers for manufacturing sectors are slightly below service
sectors. The general pattern is for larger induced effects for service sectors and greater
indirect effects for manufacturing. The international trade. wholesale trade and retail
trade margins range from 9% to 23% for manufacturing sectors (Appendix E).
Transportation margins account for 1% to 4% of total sales. Regional purchase
coefficients vary substantially across manufacturing; from a high of 91 % in the textile

sector to a low of49% in “other manufactured products”.

Table 15. Multipliers for Primary Tourism Sectors, 1999

 

 

 

Type ljobs
Type 1 Type 11 to $NT Type 1 Type 1
sales sales million sales income to value added
Sector multipliers multipliers ratio sales ratio to sales ratio
Services
Hotel services 1.507 3.558 0.749 0.619 0.979
Food & beverage services 1.423 3.768 2.620 0.750 1.006
Transportation 1.449 3.271 0.758 0.525 0.81 1
Theatrical and other arts 1.564 3.449 1.706 0.516 0.887
Recreational & cultural services 1.427 3.261 1.040 0.537 0.989
Other personal services 1.573 3.680 1.378 0.640 0.91 1
Wholesale trade 1.421 3.761 0.682 0.705 0.977
Retail trade 1.363 3.531 1.885 0.661 1.006
lntemational trade 1.519 3.394 0.658 0.515 0.910
Manufacturing
Process foods & beverages 2.135 3.706 0.564 0.231 0.523
Tobacco 1.286 2.005 0.049 0.149 0.940
Textile mill products 2.156 3.567 0.580 0.321 0.519
Wearing apparel and accessories 2.135 3.962 1.891 0.519 0.641
Leather & leather products 1.560 2.635 1.019 0.242 0.336
Wood & wood products 1.535 3.1 18 1.770 0.416 0.600
Paper & paper products 1.744 3.412 0.745 0.417 0.628
Chemical manufactures products 1.674 2.679 0.355 0.209 0.460
Non-metallic mineral products 1.779 3.329 0.566 0.359 0.656
lron and steel products 2.028 3.454 0.779 0.290 0.498
Machinery & electronic products 1.633 2.700 0.431 0.222 0.417
Transport equipment 1.908 3.215 0.492 0.269 0.588
Other manufacturing products 1.831 3.356 0.582 0.399 0.577

 

65

Multipliers do not vary substantially between visitor groups by country of
residence. except for the jobs to sales ratios (Table 16). The total jobs per million US$
sales by country of residence range from 27 to 30. These variations depend on the relative

amounts spent on different items/services during the trip. Visitors from Hong Kong/

Macao support the most jobs per million dollar of spending. due to a high proportion of

spending on food/restaurants. a sector with the most jobs per million sales. On the other

hand. Korean visitors support the fewestjobs per million dollars in spending due to a

high proportion spending on accommodation. The accommodation sector supports 16.9

jobs per million sales. less than a third of the food & beveragejob ratio. Personal income

to sales ratios vary by plus or minus 1% from the average of 40% for different visitors by

country of residency. and value added to sales ratios vary by plus or minus 2% from the

average of 62%.

Table 16. Economic Ratios and Multipliers by Country of Residence

 

 

 

Hong New
Kong/ Singapore North Zealand/
Japan Macao Korea Malaysia America Europe Australia Others
Direct effects ratioa
Capture rate 0.95 0.93 0.96 0.98 0.98 0.97 0.97 0.97
Jobs/ US$ million sales 27.44 29.83 26.58 28.33 27.42 27.84 27.85 28.30
Personal lncome 0.39 0.39 0.39 0.41 0.41 0.41 0.41 0.40
Value Added 0.61 0.60 0.61 0.63 0.64 0.63 0.63 0.62
Type 1 multiplier
Sales 1.53 1.54 1.52 1.50 1.50 1.51 1.51 1.51
Jobs/ US$ million sales 36.15 38.95 35.06 36.15 35.07 35.79 35.80 36.45
Personal lncome 0.53 0.53 0.53 0.54 0.54 0.54 0.54 0.54
Value Added 0.88 0.88 0.89 0.90 0.90 0.90 0.90 0.89
Type 11 multiplier
Sales 3.49 3.50 3.49 3.52 3.53 3.53 3.52 3.51
Jobs/ US$ million sales 74.89 77.71 73.91 76.04 75.15 75.63 75.60 75.91
Personal lncome 1.11 1.11 1.12 1.14 1.15 1.14 1.14 1.13
Value Added 2.02 2.02 2.03 2.07 2.08 2.07 2.07 2.05

 

Note. 3 Ratio is expressed in terms ofeffects (jobs. income and value added) divided by direct sales.

66

Economic Impacts of Inbound Tourism, 2001

Direct sales to inbound tourists in 2001 were US$ 3.4 billion (Table 17).
Approximately 4% of visitor spending leaked out the Taiwan economy for purchases of
imported products. Total direct visitor spending supported 95 thousand jobs. US$ 1.4
billion in personal income. and US$ 2.1 billion in value added. Thirty percent ofjobs
were supported by Japanese visitors. followed by visitors from other regions (28%). and
North American visitors (17%).

Table 17. Direct Effects oflnbound Tourism by Country of Residence, 2001

 

 

 

Total spending Direct sales Jobs lncome Value added
Country of residence (US million's) (US million's) (000's) (US million's) (US million's)
Japan $1.098 $1,042 28.5 $407 $634
Hong Kong/Macao $418 $391 1 1.6 $152 $235
Korea $59 $57 1.5 $22 $35
Singapore/Malaysia $174 $170 4.8 $70 $108
North America $607 $596 16.3 $245 $379
Europe $173 $169 4.7 $69 $106
New Zealand/Australia $39 $38 1.1 $15 $24
Others $965 $_93_5 2.6.4 M $3;
Total/ Average - 3.534 $3.398 94.8 $1.356 $2.103

 

In 2001. forty-three percent of total tourism foreign receipts accrued to the hotel
sector. followed by food & beverage services (19%). and recreation and culture services
(10%) (Table 18). The majority of jobs were in food & beverage services (42%). Twenty-
six percent of jobs were in the hotel sector and 9% in recreation and culture services.

Through secondary effects. inbound tourism supported an additional US$ 2.0
billion in sales. 37 thousand jobs and US$ 535 million in personal income (Table 19).
The aggregate type 1 sales multiplier for tourist spending is 1.58. and the type 11 sales
multiplier is 3.59. The approximate ratios of direct effects. indirect effects and induced

effects are 2 : 1 : 3. indicating substantial induced effects across the economy. In terms of

67

direct and indirect effects. international visitors to Taiwan generated $5.3 billion sales

and 132.000 jobs in 2001. When induced effects are included. the impacts increase to

$12.2 billion sales and 267.000jobs.

Table 18. Direct Effects of Inbound Tourism by Sector, 2001

 

 

 

 

 

Personal Value Percentage

Sales Jobs lncome Added Value

(million's) (000.3) (million's) (million's) Sales Jobs lncome added

Hotel services $1.456 24.6 $598 $946 43% 26% 44% 45%
Food & beverage $642 40.2 $338 $454 19% 42% 25% 22%
Transportation $264 4.7 595 S I 47 8% 5% 7% 7%
Recreational service $336 8.3 $126 $233 10% 9% 9% 1 1%
Theatrical & arts $166 6.2 SSS 894 5% 6% 4% 4%
Other personal service 8166 5.0 $68 $96 5% 5% 5% 5%
Manufacturing $286 3.8 $42 $83 8% 4% 3% 4%
Retail trade $37 1.7 $18 $27 1% 2% 1% 1%
Wholesale $23 0.4 $1 1 $16 1% 0% 1% 1%
lntemational trade S_2; 0_,3 $7 513 13/1) 0% l_°/_o 1%
T0131 $3.398 95.2 $1.360 $2.110 100% 100% 10 % 100%

 

Table 19. Economic Impacts of Inbound Tourism by Sector, 2001

 

Direct 8; Indirect Effects

Direct. Indirect & Induced Effects

 

Value Value
Sales Jobs lncome Added Sales Jobs lncome Added
(million) (000‘s) (million) (million) (million) (0005) (million) (million)
Hotel services 1.458 24.6 599 947 1.462 24.7 600 950
Food & beverage 646 40.4 340 457 828 51.8 437 586
Transportation 313 5.6 113 175 568 10.1 205 318
Recreation service 576 17.0 206 370 691 20.4 248 446
Theatrical & ans 226 8.4 75 128 264 9.8 87 150
Personal services 232 6.9 95 135 548 16.3 223 317
Manufacturing 849 10.3 134 245 2.602 28.4 384 740
Retail trade 87 4.1 42 64 609 28.6 295 449
Wholesale 60 1.0 30 42 316 5.2 157 217
International trade 52 0.8 18 31 174 2.6 59 105
Financial & business 596 7.1 183 430 3.135 38.0 984 2.383
Agri., forest. fishery 76 3.9 24 37 472 25.5 144 227
Utilities 176 0.7 26 80 362 1.3 52 169
Construction 36 M 9 g 1E 411 3_9 51
Total 5 384 131.8 1.895 3 154 12.182 266.8 3.916 7.107
Type 1 multipliersa 1.58 38.80 0.56 0.93
Type 11 multipliersa 3.59 78.51 1.15 2.09

 

 

 

a Impacts (jobs. income and value added) divided by direct sales.

68

“Recreation & culture services”, “financial & business services” and

“manufacturing" received the largest indirect effects from international visitor spending

(Table 20). When induced effects are taken into account. “financial & business services”,

“retail trade”. and "agriculture. forest & fisheries” benefited the most in terms of job

creation. Due to limited interindustry linkages with other sectors. hotel services received

relatively small secondary effects. In 2001, inbound tourism indirectly supported 38,000

jobs in “financial & business services” (indirect + induced effects) followed by 26.900

jobs in “retail trade" and 25.500 jobs in "agriculture. forestry and fisheries”.

Table 20. Indirect and Induced Effects ofJobs and Personal Income, 2001

 

 

 

 

 

Jobs (0005) lncome1$ million) Ranking ofjobs Rankingofincome
Indirect + Indirect + Indirect + Indirect +
Indirect induced Indirect induced Indirect induced Indirect induced
effects effects effects effects effects effects effects effects

Recreation services 8.7 12.1 80 122 1 5 3 7
Financial & business 7.1 38.0 183 984 2 1 1 1
Manufacturing 6.5 24.6 92 342 3 4 2 2
Agri.. forest. fishery 3.9 25.5 24 144 4 3 6 6
Retail trade 2.3 26.9 24 277 5 2 6 3
Theatrical & arts 2.2 3.6 20 32 6 1 1 8 13
Personal services 2.0 1 1.4 27 155 7 7 4 4
Construction 1.0 4.1 9 39 8 10 12 12
Transportation 0.9 5.4 18 1 10 9 8 10 8
Utilities 0.7 1.3 26 52 10 13 5 10
Wholesale 0.6 4.8 19 146 1 1 9 9 5
lntemational trade 0.4 2.2 1 1 52 12 12 1 l 10
Food & beverage 0.3 1 1.7 2 99 13 6 13 9
Hotel services 0.0 0.1 1 2 14 14 14 14

 

 

 

In 2001. Taiwan reported NT$ 9.448 billion in Gross Domestic Product (GDP)

and 9.38 million total jobs (Ministry of Economic Affairs. 2005). Visitor spending of

inbound tourism in 2001 contributed 0.76% of national GDP and 1.01% oftotal jobs in

terms of direct effects: 2.56% of GDP and 2.84% of total jobs in terms of total effects.

69

Objective [1
Test the stability ofjob to sales ratios and personal income to sales ratios relative to

occupancy rates in the accommodation sector

Job to Sales Ratio

A log linear functional form best captures the relationship between hotel jobs to
sales ratios and occupancy rates. This functional form has higher adjusted R-square value
than others (Appendix F. Equation 17). No multicollinearity is detected among the

independent variables. and no heteroskedasticity is detected among the error terms.

In (jobs to sales ratio) = a + [31* occupancy rate + [33* hotel scale + [33* consumer

price index + [34* central Taiwan + [35* southern Taiwan + /)’(.* eastern Taiwan + error

term (0. (5 3 ) (Equation 17)

Occupancy rates and hotel scales (number of rooms) are significant in
determining jobs to sales ratios at the 95% confidence level. The consumer price index
and three region dummy variables are not significant (Table 21). Jobs to sales ratios are
negatively correlated with occupancy rates. A rise in occupancy rate leads to fewer jobs
per million sales (Equation 18). Due to the non—linear functional form. the percentage
changes injobs to sales ratio are not constant at different occupancy levels. A 5%
increase from 80% to 85% in occupancy. for example. will lead to a smaller reduction in

jobs to sales ratios than a 5% increase from 40% to 45%.

70

Table 21. Regression Statistics for the Log Jobs to Sales Ratio

 

 

Variables in the equation B Std. Error BETA T SIG T Tolerance
Constant 3.1342 0.0663 47.2566 0.0000

Occupancy Rate -1.4776 0.1144 -0.5503 -12.9182 0.0000 0.895
Rooms -0.0006 0.0001 -0.2058 -4.8305 0.0000 0.895

Excluded variables
Consumer price index
Central Taiwan
Southern Taiwan
Eastern Taiwan
Stepwise analysis is employed with 0.05 to enter and 0.10 to remove. Adjusted R-Squared = .415 : F-Value
= 128.820 (p value = 0.0000); Number of cases = 361

 

ln (jobs to sales ratio) = 3.1342 —1.4776*occupancy rate - 0.0006* rooms

(Equation 18)

Personal Income to Sales Ratio

Personal income to sales ratios are modeled as a linear function of occupancy
rates. the number of rooms per establishment. consumer price index. and regional dummy
variables. The linear functional form is applied because of its high adjusted R-square

values than other fomis (Appendix G. Equation 19).

Personal income to sales ratio = a + /il* occupancy rate + [33* hotel scale + [13*
consumer price index + [34* central Taiwan + [35* southem Taiwan + [36* eastern

Taiwan + error term (0. (53) (Equation 19)

No multicollinearity is observed in the independent variables but some level of
heteroskedasticity is discovered among the error terms. Variances of the coefficients are

re-calculated using the I-luber-White sandwich estimation. which corrects the bias of

71

heteroscedasticity. The correction provides larger variances (robust standard error) for the

constant and occupancy rates in the equation. but the recomputed standard error does not

change the significance of the independent variables.

Occupancy rates and hotel scales (number of rooms) are significant in

detemiining personal income to sales ratios at the 95% confidence level (Table 22).

Income to sales ratios and occupancy rates are negatively correlated. indicating that a rise

in the occupancy rate leads to a smaller percentage of sales going to employee

compensation. A one percent increase in occupancy rate would yield a decrease of 0.1 1 in

the income to sales ratio. cetcris paribus (Equation 20).

Table 22. Regression Statistics for the Income to Sales Ratio

 

 

Variables in the Robust

equation B Std. Error Std. Error BETA T SIG T Tolerance
Constant 0.4672 0.0246 0.0334 13.9760 0.0000
Occupancy Rate -0.1060 0.0410 0.0528 -0.1740 -2.0080 0.0445 0.895
Rooms -0.0002 0.0000 0.0000 -0.2868 -3.8810 0.0000 0.895

Variables excluded
Consumer price index
Central Taiwan
Southern Taiwan
Eastern Taiwan

 

 

Stepwise analysis is employed with 0.05 to enter and 0.10 to remove. Adjusted R-Squared = .1368: F-
Value = 18.0350 (p value = 0.0000): Number of cases = 216

Income to sales ratio = 0.4672 - 0.1060*occupancy rate - 0.0002 *rooms

(Equation 20)

The predicted job and income to sales ratios are estimated using Equation 18 and

20. With the average occupancy rate of 62% in 1999 and an average 253 rooms per hotel.

a five percent increase in occupancy rates would lead to a 7% decrease in job ratios. and a

1.5% reduction in income ratios (Table 23). A five percent decrease in occupancy rates,

on the other hand. would raise the job ratio by 8% and increase the income ratio by 1.5%.

Table 23. Predicted Jobs and Income Ratios for Tourist Hotels

 

Predicted jobs to

Predicted personal
income to sales

 

Occupancy rate sales ratio Pct change ratio Pct change
40% 0.9102 38.4% 0.372 6.7%
45% 0.8453 28.6% 0.366 5.2%
50% 0.7851 19.4% 0.361 3.7%
57% 0.7080 7.7% 0.354 1.5%
62% (base year) 0.6576 0.0% 0.348 0.0%
67% 0.6107 -7.1% 0.343 -1.5%
70% 0.5843 -11.1% 0.340 -2.5%
75% 0.5427 -l7.5% 0.334 -4.0%
80% 0.5040 -23.4% 0.329 -5.5%
85% 0.4681 -28.8% 0.324 -7.0%

 

 

Predicted l-O Multipliers

These predicted changes are applied to the hotel job and income ratios in the 1999

Taiwan l-O table. The results are displayed in Table 24. In 1999. when the average

occupancy rate was 62%. the jobs to sales ratio was 0.4972 per NT$ million sales. The

jobs to sales ratio would rise to 0.6681 if occupancy rates decrease to 42%. The ratio

would fall to 0.3436 if occupancy rates increase to 87%. The personal income to sales

ratio. on the other hand. is less responsive to changes in occupancy rates. ranging

between 0.38 and 0.44 over the same range of changes in occupancy (Figure 5).

73

Table 24. Predicted Jobs and Income Ratios for the Accommodation Sector

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Predicted jobs to Predicted personal income

Occupancy rate NT$ million sales ratio to sales ratio
42% 0.6681 0.4358
47% 0.6206 0.4295
52% 0.5764 0.4232
57% 0.5353 0.4169
62% (base year) 0.4972 0.4106
67% 0.4618 0.4043
72% 0.4289 0.3981
77% 0.3984 0.3918
82% 0.3700 0.3855
87% 0.3436 0.3792

0.85

0.75 ‘1' " Jobs to NT$ milllon sales

— Income to sales ratio
0.65 1' 1 ~
1 r
.2 1
‘55 055 1 i L r'
j , ,.
l ‘ . 1 ' f _
0.45 i 1""~ . 1 l r
. 1 1 T
t 1 l
0.35 1' - .l
0.25 4 l l I I
35% 45% 55% 65% 75% 85% 95%
Occupancy rate

 

 

 

Figure 5. Predicted Jobs and Income Ratios with 95% Confidence Intervals for the Accommodation

Sector

The predicted job and income ratios with respect to occupancy rates in Table 24
are applied to re-compute multipliers for the hotel sector in the Taiwan I-O model (Table
25). Type I sales multipliers remain constant with respect to occupancy rates. while type
11 sales multipliers vary by plus or minus 3% from the base year value when occupancy

rates fluctuate between 42% and 82%. Type 1 job multipliers for the hotel sector differ

74

substantially. ranging from 1.01 jobs per NT$ million hotel sales at 42% occupancy to
0.56 jobs per NT$ million hotel sales at 82% occupancy. The job ratio is 34% higher than
the base year job ratio when businesses operate at 42% occupancy. Using fixed ratios in
the standard l-O model will therefore underestimate hotel jobs by as much as 34% if
occupancy rate drops to 42% or will overestimate hotel jobs by 26% if occupancy
increases to 82%. The type 1 income multiplier is more stable with a maximum 6%

difference from the base year figure for occupancy rate between 42% and 82%.

Table 25. Multipliers of the Hotel Sector Based on Different Occupancy Rates

 

 

Type ljobs
Occupancy rate Type 1 sales Type 11 sales 7NT$ million sales Type 1 income
Multipliers
42% 1.51 3.65 1.01 0.66
52% 1.51 3.61 0.87 0.64
62% (base year) 1.51 3.56 0.75 0.62
72% 1.51 3.51 0.65 0.60
82% 1.51 3.46 0.56 0.58
Pct change
42% 0% 3% 34% 6%
52% 0% l % 1 6% 3%
62% (base year) 0% 0% 09/6 0%
72% 0% -1% -14% -3%
82% 0% -3% -26% -6%

 

 

To demonstrate the direction and magnitude of biases with fixed and varying l-O
ratios. the multipliers in Table 25 are applied to inbound tourism visitor spending in 2001
under 42%. 62% and 82% occupancy rates. Standard I-O models with fixed ratios yield
the same impact estimates regardless of capacity utilization. l-O models with varying job
and income ratios. in contrast. estimate more jobs and higher wages and salaries when
hotels operate at lower occupancies (Table 26). At an occupancy rate of 42%. the [-0

model with fixed ratios underestimates hotel jobs by 34% and personal income of hotel

75

employees by 6%. At an occupancy rate of 82%, the fixed ratios overestimate jobs by

26% and income by 4% for the hotel sector. Differences in the overall economy are 4%

injobs and 2% in income. As changes in the job ratios are nonlinear. the estimation error

is reduced when occupancy approaches higher capacity.

Table 26. Economic Impacts of Inbound Tourism in 2001 with Fixed and Varying Ratios

 

Direct. indirect &

Direct & indirect

 

 

Direct indirect &
induced income

 

 

Direct & indirect iobs induced iobs income (million‘s) (millions)

Occumncy All All All All
rate Hotel sector Hotel sector Hotel sector Hotel sector
[-0 models with fixed ratios

42% 24.644 131.819 24.705 266.770 $599 $1.895 $600 $3.916

62% 24.644 131.819 24.705 266.770 $599 $1.895 $600 $3,916

82% 24.644 131.819 24.705 266.770 $599 $1.895 $600 $3.916
l-O models with varyingjob and income ratios

42% 33.1 18 140.293 33.201 278.380 $636 $1.932 $637 $4.000

62% 24.644 131.819 24.705 266.770 $599 $1.895 $600 $3.916

82% 18.339 125.514 18.383 257.334 $562 $1.858 $564 $3.833
Pct error between fixed and varying ratiosa

42% 34% 6% 34% 4% 6% 2% % 2%

62% 0% 0% 0% 0% 0% 0% 0% 0%

82% -26° 0 -5% -26% -4% -6% —2% -6% -2%

 

 

 

a (Impacts with varying ratios 4 impacts with fixed ratios) t‘ impacts with fixed ratios

Objective III

Estimate economic impacts of inbound tourism on the Taiwan economy for three

scenarios based on 1) standard [-0 models with. fixed multipliers. and 2) [-0 models with

varying jobs to sales ratios and personal income to sales ratios by occupancy rates

76

Three scenarios are defined with alternative levels of demand and supply as
summarized in Table 27. Impacts for scenario 1 are estimated at the current hotel
capacity. of 20.697 rooms and the maximum international visitor volume that this
capacity can accommodate. Scenarios 2 and 3 assume an increase in lodging capacity to
37.900 rooms. and that international arrivals grow to 75% and 100%. respectively. of the
DTAP visitation target.

Economic impacts ofthe 3 scenarios are estimated using Input-Output models
with fixed and varyingjob and income ratios. Multipliers for 39 I-O sectors with fixed
ratios were reported in the previous section (Table 15). Multipliers with changing job and

income ratios with respect to occupancy rates are computed using equations 18 and 20.

Table 27. Demand and Supply Levels under the Three Scenarios

 

 

 

Domestic Tourism Inbound Tour isn;
Scenarios Tourist Hotel Capacity (person trips) (person trips)
Base Year (2001) 20.697 rooms 97 million 2.6 million
Scenario 1 (2008) 20.697 rooms 135 million Maximum visitors ('2)
Scenario 2 (2008) 37.900 rooms 135 million 3.6 million
Scenario 3 (2008) 37.900 rooms 135 million 4.8 million

 

a . . . .
F oreign labors to Talwan are excluded. which are approx1mate 20 thousand persons per year.

As hotels operate with distinct cost structures with respect to capacity utilization
(indicated in objective 11). the first step to compute varying l-O multipliers is to calculate

hotel occupancy rates under each scenario.

Occupancy Rates by Scenario
Three pieces of information are needed to compute the average occupancy rates of

tourist hotels in 2008: (1) rooms demanded from domestic tourists. (2) rooms demanded

77

from international visitors. and (3) the proposed lodging capacity under each scenario in
2008. For domestic travel. the same level of demand and regional distribution are
assumed across the three scenarios. llotel occupancy rates are therefore primarily
determined by the international visitor volume and lodging capacity of tourist hotels.

Domestic tour-isn't. Based on a projected volume of 135 million domestic person
trips (Taiwan Tourism Bureau. 2003). domestic tourism is estimated to support 2.02
million room nights at tourist hotels in 2008 (Table 28). The regional distribution of
domestic travel is assumed to be constant between year 2001 and 2008. estimated from
the “Domestic Travel by Domestic Residents Survey”. Forty percent of these room nights
are in southern Taiwan. 28% in northern Taiwan. and 21% in eastern Taiwan.

Table 28. Trips and Room Nights for Domestic Tourism, 2008

 

 

 

Northern Central Southem Eastern
Taiwan Taiwan Taiwan Taiwan Total
Person trips ofdomestic travel (million‘s) 48.02 34.24 44.08 8.66 135.00
Pet of person trips 36% 25% 33% 6% 100%
Room nights at tourist hotels (million‘s) 0.57 0.23 0.81 0.42 2.02
Pet of room nights 28% 1 1% 40% 21% 100%

 

Note. a Party size covers adults and children 12 or older.

Inbound Tourist-n. For every one thousand international visitors. 1.141 room
nights (party nights) are generated at tourist hotels. On average. 85% of these 1,141 room
nights are located in northern Taiwan. 9% in southern Taiwan. 4% in central Taiwan. and
2% in eastern Taiwan. Total room nights generated by international visitors are computed
by multiplying the average hotel room nights per visitor by the total volume of inbound
travel.

For scenario 1_. the maximum number of room nights that tourist hotels in

northern Taiwan can supply is 4.34 million (Table 29). Subtracting 566 thousand room

78

nights for use by domestic tourists leaves 3.78 million room nights available for

intemational visitors. Assuming tourist hotels in northern Taiwan operate at 100%

occupancy that no shifting of excess demand to other regions is permitted, the maximum

volume ofinternational visitors that can be accommodated is 3.89 million. This

corresponds to 81% of the DTAP visitation goal of 4.8 million international visitors.

Table 29. Supply, Demand, and Average Occupancy Rates of Tourist Hotels

 

 

 

 

 

 

 

 

 

Northern Central Southern Eastern Total/
Scenario Taiwan Taiwan Taiwan Taiwan Average
Base year, 2001
Supply - Tourist hotel rooms 1 1.906 1.726 4.618 2.447 20.697
Demand (000’s room nightsz’year)
Domestic tourism 409 164 584 303 1.458
Inbound tourism 2_._5_19 110 Zfi _7_3 311i
Total 2.988 274 85 376 4.496
Occupancy rate 70% 44% 51% 420/ 62%
Scenario 1, 2008
Supply - Tourist hotel rooms 1 1.906 1.726 4.618 2.447 20.697
Demand (000's room nights/year)
Domestic tourism 566 227 809 419 2.021
Inbound tourism m 162 Q; 1_0_7 4.4_48
Total 4.343 389 1.21 1 526 6.469
Occupancy rate 100% 62% 72% 59% 86%
Scenario 2, 2008
Supply - Tourist hotel rooms 26.000 2.600 6.000 3.300 37,900
Demand (000's room nights/year)
Domestic tourism 566 227 809 419 2.021
Inbound tourism M 150 3_72 99 1L8
Total 4.063 377 1.181 518 6.139
Occupancy rate 43% 40% 54% 430/ 44%
Scenario 3, 2008
Supply - Tourist hotel rooms 26.000 2.600 6.000 3.300 37.900
Demand (000's room nights/year)
Domestic tourism 566 227 809 419 2.021
Inbound tourism M 2_OQ 4% £2 5.491
Total 5.229 426 1.306 551 7.512
Occupancy rate 55% 45% 60% 46% 54%

 

 

79

Given the expanded hotel capacity of 37.900 rooms. scenarios 2 and 3 do not
experience a capacity constraint in accommodations. International visitors of 3.6 million
in scenario 2 purchase 4.12 million room nights at tourist hotels. lntemational visitors of
4.8 million in scenario 3 purchase 5.50 million room nights.

Average. occupancy rates. For scenario 1. the estimated average occupancy rate is
86% nationwide. Tourist hotels in northern Taiwan reach 100% occupancy while the
three other regions still have rooms available (Table 27). For scenario 2. the estimated
occupancy rate is 44% based on 3.6 million international visitors. Occupancy rates range
from 40% to 54% across the four geographic regions. Under scenario 3. international
visitor numbers are assumed to increase to 4.8 million. which raises the average
occupancy rate to 54%.

The “Tourist Hotel Development Plan” calls for significant expansion oflodging
capacity in northern Taiwan. from 1 1.900 rooms in 2001 to 26.000 rooms by 2008. The
increased capacity in northern Taiwan (scenario 2 and 3) relieves the lodging constraint
on tourism consumption. and room utilization is more balanced across the four
geographic regions. However. the overall occupancy rates in scenarios 2 & 3 are below
the base year occupancy (62%). The proposed increase in lodging capacity in the “Tourist
Hotel Development Plan" is greater than the projected increase in hotel demand from

international visitors in the DTAP.

Predicted Economic Ratios and Multipliers By Scenario
Armed with the average occupancy rates for each scenario. the next step is to

calculate economic ratios and multipliers at the predicted level of capacity utilization.

80

Distinct economic ratios and multipliers for the accommodation sector are computed by
inserting the occupancy rates from Table 29 into the two econometric equations. Scenario
1 has the highest occupancy rates. and thus the lowest predicted job and income to sales
ratios (Table 30). Scenarios 2 and 3. on the other hand, have job ratios between 0.65 and
0.55 and income to sales ratio around 0.42 to 0.43. Type 1 sales multipliers remain
constant across the three scenarios (1.507). Type 11 sales multipliers vary slightly with
changing occupancy rates due to small changes in personal income to sales ratios. The
type 11 sales multiplier for scenario 1 is 3% lower than the base year multiplier. while the
type 11 sales multipliers for scenarios 2 and 3 are higher by 2% and 1%. respectively.
The job multipliers are more sensitive to the changes in occupancy rates across
the three scenarios than the income multipliers. For type ljob multipliers. the percentage
change ranges from 12% in scenario 3 to 30 % in scenario 2. Income multipliers. on the

other hand. are more stable. with a maximum change of 7% across the three scenarios.

Table 30. Predicted Hotel Economic Ratios and Multipliers under the Three Scenarios

 

 

Type 1 Type II Type 1 Type 1

Occupancy Jobs to Income to sales sales job income

rate sales ratio sales ratio multiplier multiplier multiplier multiplier

Base year 6296 0.497 0.41 1 1.507 3.558 0.749 0.619

Scenario 1 86% 0.351 0.381 1.507 3.446 0.529 0.574

Scenario 2 44% 0.645 0.433 1.507 3.641 0.972 0.653

Scenario 3 54% 0.557 0.420 1.507 3.594 0.839 0.633
Pct change to the base year

Scenario 1 39% -29% -7% 0% -3% -29% -7%

Scenario 2 -29% 30% 5% 0% 2% 30% 5%

Scenario 3 -13% 12% 2% 0% 1% 12% 2%

 

 

81

Inbound Tourism Visitor Spending

Economic impacts of the three scenarios are calculated by applying inbound
visitor spending to the Taiwan l-O multipliers. Total visitor spending under the three
scenarios is estimated to be US$ 5.25 billion. US$ 4.86 billion and US$ 6.48 billion.
respectively (Table 31). Excluding airfares, lodging expenses account for 41% of total

trip expenditures. expenses for food are 18%. and shopping represents 15%.

Table 31. Total Visitor Spending under the Three Scenarios (US$ million’s)

 

 

Spending category Scenario 1 Scenario 2 Scenario 3 Pet
Lodging 2.135 1.977 2.636 41%
Food 951 881 1.174 18%
Transportation 386 357 477 7%
Entertainment 499 462 616 10%
Shopping 774 717 956 15%
Others £10 4_6_3 fl Log/9
Total 5.246 4.857 6.476 100%

 

 

Economic Impacts - Scenario 1
Lodging capacity maintains 20, 697 rooms. The maximum number ofinternational
visitors. under the hotel capacitj.’ constraints. is estimated to be 3.8 9 million.

Fixed [-0 ratios. The maximum number of international visitors. under the
lodging constraint in northern Taiwan. is 3.89 million. These visits are estimated to spend
US$ 5.24 billion in Taiwan. which supports US$ 5.04 billion of direct sales. 141
thousand direct jobs. and US$ 2.01 billion of direct personal income in Taiwan (Table
32). Indirect effects create an additional 55 thousand jobs and $796 million in personal
income. Induced effects generate an additional 200 thousand jobs and $2.995 million

personal income.

Table 32. Economic Impacts of Scenario 1 using Fixed Ratios

 

Direct Effects

Direct & Indirect

Direct. Indirect &

 

Effects Induced Effects

Sales Income Income Income

(S million) Jobs ($ million) Jobs ($ million) Jobs ($ million)

Hotel services 2.135 36.078 877 36.143 878 36.233 881
Food & beverage 951 59.522 501 59.898 504 76.833 647
Transportation 396 7.035 143 8.350 170 15.050 307
Recreational service 499 12.362 188 25.452 308 30.396 370
Theatrical & arts 250 9.277 83 12.597 112 14.688 131
Personal services 250 7.445 102 10.350 142 24.263 332
Manufacturing 434 5.807 64 15.486 201 42.236 571
Retail trade 56 2.645 27 6.121 63 42.463 438
Wholesale 34 563 17 1.480 45 7.667 233
lntemational trade 33 487 11 1.144 26 3.818 88
Financial & business 0 0 0 10.467 271 56.355 1.458
Agri.. forest fishery 0 0 0 5.970 37 37.839 214
Utilities 0 O 0 974 39 1.880 77
Construction 0 Q Q 1% 15 6._134 g
Total 5.039 141.221 2.014 195.909 2.810 395.856 5.805

 

 

Modified [-0 economic ratios. The recalculated direct effects on the

accommodation sector are 25 thousand jobs and $813 million of employee wages and

salaries using the predicted job and income ratios (Table 33). Hotel jobs estimated using

the predicted job ratio is 29% less than the value estimated using constant ratios in the

standard [-0 models. This is because the hotel jobs to sales ratio is lower at the 86%

occupancy rate. reflecting fewer jobs per millions in sales. Personal income is reduced by

7%. from $877 million to $813 million. Adjusting the economic ratios for the hotel sector

reduces the direct effects across the entire economy by 8% in jobs and 3% in employee

compensation. Total effects. including direct, indirect and induced effects, are reduced by

4% in jobs and 2% in personal income.

83

Table 33. Comparison of Economic Impacts between Fixed and Varying Ratios for Scenario 1

 

 

Direct Effects Direct & Indirect Effects Direct. Indirect 8: Induced
Effects
Personal Personal Personal
Income lncome Income

Jobs (8 million's) Jobs (S million's) Jobs ($ million's)

 

Fixed I-O ratios
Hotel sector 36.078 877 36.143 878 36.233 881

Total sectors 141.221 2.014 195.909 2.810 395.856 5.805
Varying I-O ratios

 

Hotel sector 25.445 813 25.491 815 25.553 817
Total sectors 130.588 1.950 185.256 2.746 379.779 5.660
Percent changeal
Hotel sector 2996 -7% -29% -7% -29% -7%
Total sectors -8% -3% -5% -2% -4% -2%

 

Note. 3 (effects with predicted economic ratios - effects with fixed ratios) / effects with fixed ratios

Economic Impacts - Scenario 2
Lodging capacity expands to 3 .7. 900 rooms by 2008. and international visitors increase
to 3.6 million.

Fixed [-0 ratios. Scenario 2 assumes a higher lodging capacity with fewer
international visitors. In scenario 2, 3.6 million international visitors spend US$4.87
billion in Taiwan. which support US$ 4.67 billion of direct sales. 131 thousand direct
jobs. and US$ 1.87 billion ofdirect personal income in Taiwan (Table 34). Indirect
effects create an additional 51 thousand jobs. and induced effects generate an additional
185 thousand jobs.

Varying [-0 ratios. The recalculated jobs in the accommodation sector are 43
thousand. 30% higher than the number calculated using fixed ratios (Table 35). Personal
income increases by eight percent for the accommodation sector. from $812 million to
$856 million. These differences arise because hotels operating at 44% occupancy support

higher jobs to sales ratio and higher income to sales ratio than the base year point.

84

Relaxing the economic ratios for the hotel sector increases the tourism impact estimates

across the entire economy by 8% in total jobs and 2% in personal income in terms of

direct effects; 4% in total jobs and 2% in personal income. in terms of total effects.

Table 34. Economic Impacts of Scenario 2 using Fixed Ratios

 

Direct Effects

Direct & Indirect

Direct. Indirect &

 

 

Effects Induced Effects

Sales lncome Income Income

($ million) Jobs ($ million) Jobs (S million) Jobs ($ million)

Hotel services 1.977 33.405 812 33.466 813 33.549 815
Food & beverage 881 55.113 464 55.461 467 71.142 599
Transportation 367 6.513 133 7.732 158 13.935 284
Recreational service 462 1 1.447 174 23.567 285 28.144 342
Theatrical & arts 232 8.590 77 1 1.664 104 13.600 121
Personal services 232 6.893 94 9.583 131 22.466 307
Manufacturing 402 5.377 60 14.339 186 39.108 529
Retail trade 52 2.449 25 5.668 58 39.317 406
Wholesale 32 521 16 1.370 42 7.099 216
International trade 31 451 10 1.059 24 3,535 81
Financial & business 0 0 0 9.691 251 52,181 1,350
Agri.. forest. fishery 0 0 0 5.527 34 35.036 199
Utilities 0 0 0 902 36 1.741 72
Construction 0 Q 0 LE] _1_3 M 3
Total 4.666 130.761 1.865 181.397 2.602 366,534 5,375

 

Table 35. Comparison of Economic Impacts between Fixed and Varying Ratios for Scenario 2

 

Direct Effects

Direct & Indirect Effects

Direct. Indirect & Induced

 

Effects
Personal Personal Personal
Income Income Income
Jobs ($ Million's) Jobs (8 Million's) Jobs ($ Million's)
Fixed l-O ratios
Hotel sector 33.405 812 33.466 813 33,549 815
Total sectors 130.761 1.865 181.397 2.602 366.534 5.375
Varying l-O ratios
Hotel sector 43.339 856 43.417 857 43,528 859
Total sectors 140.694 1.908 191.348 2.645 380.238 5.475
Percent changea
Hotel sector 30% 5% 30% 5% 30% 5%
Total sectors 8% 2% 5% 2% 4% 2%

 

 

Note. a (effects with predicted economic ratios - effects with fixed ratios) / effects with fixed ratios

85

Economic Impacts - Scenario 3

Lodging capacity expands to 3 7. 900 rooms and international visitors increase to 4.8

million.

Fixed [-0 ratios. In scenario 3. inbound tourism with 4.8 million visitors is

estimated to spend US$ 6.48 billion in Taiwan. which supports US$ 6.22 billion of direct

sales. 174 thousand direct jobs. and U S$ 2.49 billion of direct personal income in Taiwan

under scenario 3 (Table 36). Indirect effects create an additional 68 thousand jobs. and

induced effects generate an additional 247 thousand jobs.

Table 36. Economic Impacts of Scenario 3 using Fixed Ratios

 

Direct Effects

Direct & Indirect

Direct Indirect &

 

 

 

 

Effects Induced Effects

Sales Income Income Income

(5 million) Jobs (3 million) Jobs (3 million) Jobs (5 million)

Hotel services 2.636 44.540 1.082 44.621 1.084 44.732 1.087
Food & beverage 1.174 73.484 619 73.948 623 94.856 799
Transportation 489 8.685 177 10.309 210 18.580 378
Recreational service 616 15.262 232 31.422 380 37.526 456
Theatrical & arts 309 11.453 102 15.551 139 18.134 162
Personal services 309 9.191 126 12.777 175 29.955 410
Manufacturing 536 7.169 80 19.119 248 52.144 705
Retail trade 69 3.266 34 7.557 78 52.423 541
Wholesale 43 695 21 1.827 56 9 466 288
International trade 41 602 14 1.412 33 4 714 109
‘inancial & business 0 0 0 12.922 335 69 574 1 800
Agri.. forest. fishery 0 0 0 7.370 45 46 715 265
Utilities 0 0 0 1.203 48 2.321 95
Construction 9 Q 9 1.823 E 1.5—73 7_1
Total 6.222 174.347 2.486 241.863 3.469 488.711 7.166

 

Varying [-0 ratios. The recalculated jobs in the accommodation sector are 50

thousand. 12% higher than the value calculated using fixed ratios (Table 37). Personal

income increases by 2 percent for the accommodation sector. from $1.082 million to

$1,108 million. These differences arise because job to sales ratios are adjusted upward by

86

12% and income to sales ratios are upward by 2% to reflect that business operates at

lower occupancy rates supporting more jobs and personal income per unit sales. Relaxing
the economic ratios for the accommodation sector increases the tourism impact estimates
across the entire economy by 3% in total jobs and 1% in employee compensation in terms

of direct effects; 2% in total jobs and 1% in personal income in terms of total effects.

Table 37. Comparison of Economic Impacts between Fixed and Varying Ratios for Scenario 3

 

 

 

 

 

 

Direct Effects Direct & Indirect Effects Direct. Indirect &
Induced Effects
Personal Personal Personal
Income (8 Income (S Income (8
Jobs Million's) Jobs Million's) Jobs Million's)
Fixed I-O ratios
Hotel sector 44.540 1.082 44.621 1.084 44.732 1.087
Total SGCIOFS 174.347 2.486 241.863 3.469 488.71 I 7.166
Varying I-O ratios
Hotel sector 49.905 1.108 49.995 1.1 10 50.121 1.1 13
Total sectors 179.712 2.512 247.236 3.494 496.269 7,225
Percent changea
Hotel sector 12% 2% 12% 2% 12% 2%
Total sectors 3% 1% 2% 1% 2% 1%

 

Note. 3 (effects with predicted economic ratios - effects with fixed ratios) / effects with fixed ratios

Summwy. Jobs and income estimated based on the varying ratios are reported in
Table 38 and Table 39. Job ratios are more sensitive to changes in occupancy rates,
leading to greater variation in job estimates. In terms of total jobs. the hotel sector ranks
7th among the 14 sectors in scenario 1. 3rd in scenario 2. and 5th in scenario 3. Total
personal income is not as sensitive to changes in occupancy. The hotel sector ranks 2nd

among the 14 sectors across all 3 scenarios.

87

Table 38. Total Jobs and Ranking Using Varying Job Ratios

 

 

 

 

 

 

 

 

Jobs in total effects Ranking

Scenario 1 Scenario 2 Scenario 3 Scenario 1 Scenario 2 Scenario 3
Hotel services 25.553 43.528 50.121 7 3 5
Food & Beverage 76.376 71.457 95.039 1 l 1
Financial & Business 55.1 16 53.036 70.072 2 2 2
Manufacturing 41.514 39.606 52.434 3 5 4
Retail Trade 41.481 39.995 52.817 4 4 3
Agri.. forest. fishery 36.979 35.630 47.061 5 6 6
Recreational services 30.262 28.237 37.580 6 7 7
Personal services 23.888 22.726 30.106 8 8 8
Transportation 14.869 14.060 18.652 9 9 9
Theatrical & arts 14.632 13.639 18.157 10 10 10
Wholesale 7.500 7.215 9.533 11 11 11
Construction 6.008 5.766 7.623 12 12 12
lntemational Trade 3.746 3.585 4.743 13 13 13
Utilities 1.855. 1.758 2.331 14 14 14
Table 39. Total Personal Income and Ranking Using Varying lncome Ratios

Income in total effectsLUSS million‘s) Ranking

Scenario 1 Scenario 2 Scenario 3 Scenario 1 Scenario 2 Scenario 3
Hotel services 817 859 1.1 13 2 2 2
Financial & business 1.426 1.372 1.813 1 1 1
Food & beverage 643 602 800 3 3 3
Manufacturing 561 536 709 4 4 4
Retail trade 428 413 545 5 5 5
Recreational services 368 343 457 6 6 6
Personal services 327 31 1 412 7 7 7
Transportation 303 286 380 8 8 8
Wholesale 228 219 290 9 9 9
Agri.. forest. fishery 210 202 267 10 10 10
Theatrical & arts 130 122 162 1 1 1 1 I 1
lntemational trade 86 83 109 12 12 12
Utilities 76 72 96 13 13 13
Construction 57 54 72 14 14 14

 

 

88

CHAPTER V CONCLUSION

The purposes of this study were (I) to establish baseline economic impacts for
inbound tourism to Taiwan in 2001. (2) to examine the stability of jobs to sales ratios and
income to sales ratios for the accommodation sector based on different capacity
utilization. and (3) to estimate economic impacts ofthe “Doubling Tourist Arrival’s
Plan” using a national input-output model with fixed and varying value added
components.

The assumption of constant job and income ratios in the standard I-O framework
under varying capacity utilization was tested by examining hotel operational data. F irm-
level time-series data of tourist hotels in Taiwan from 1999 to 2003 were used to
establish econometric relationships between job and income ratios and occupancy rates.
The predicted ratios were incorporated into the Taiwan I-O model to generate distinct
sets of multipliers based on given occupancy rates. Economic impacts of the DTAP
policy were estimated under three scenarios. defined by the international visitor volume
and hotel lodging capacity. These scenarios simulate alternative conditions of tourism

demand and supply.

Importance of Inbound Tourism to Taiwan’s Economy
Inbound tourism impacts the Taiwan economy by generating foreign receipts.
creatingjob opportunities. and contributing to the balance of payments. In 2001, 2.6
million international visitors spent US$ 3.5 billion in Taiwan. which supported 95

thousand directjobs. and US$ 2.1 billion in value added. The sectors receiving the largest

89

direct effects in value added were hotel services. food & beverage. and recreational
services. Sectors receiving the largest secondary effects were financial & business.
manufacturing. and retail trade. Visitor spending of inbound tourism in 2001 contributed
0.76% of Gross Domestic Product (GDP) and 1.01% oftotal jobs in terms of direct
effects; 2.56% of GDP and 2.84% of total jobs in terms oftotal effects.

Tourism has significant indirect and induced effects across the economy. Input-
Output models are useful tools to estimate theses effects and demonstrate the complex
interrelationships among sectors. In 2001, every direct tourism job in Taiwan supported
1.80 secondary jobs. and each dollar of income in tourism businesses yielded an
additional $1.88 in secondary effects. Secondary effects. although a major part of
tourism‘s economic impacts in Taiwan. are harder to monitor in a timely manner and are
easily overlooked. For example. during the collapse of tourism activities in Taiwan due to
the spread of Severe Acute Respiratory Syndrome (SARS) and Avian influenza in 2003.
govemment subsidies (financing and tax breaks) were primarily directed to the
accommodation and air transportation sectors (Taiwan Tourism Bureau. 2003). Economic
losses in indirectly affected sectors were not recognized. Government tourism and
economic planning therefore requires a comprehensive approach that considers the

interrelationships among industries.

Changing Job and Income Ratios by Occupancy Rates and Hotel Scale
Jobs to sales ratios and personal income to sales ratios for the hotel sector are not
constant. They change with capacity utilization and hotel scale. Changes in jobs to sales

ratios are mainly explained by occupancy rates (Beta =-.5503) whereas changes in

90

income to sales ratios are more influenced by hotel scales (Beta = -.2868). Based on
empirical data. tourist hotels in Taiwan operate with increasing returns to scale”.
indicating decreasing marginal costs for selling additional rooms.

Job and income ratios are related to the average salary per employee. Given the
predicted job ratios and income ratios from econometric equations. the computed average
salary is positively correlated with occupancy rates29 (Table 40 & Figure 6). The average
wage and salary received per employee at tourist hotels rise from NT$ 0.42 million at a
42% occupancy rate to NT$0.71 million at a 87% occupancy. As capacity utilization
rises. fewer jobs are supported per unit sales but at higher wages per employee. This
pattern indicates that hotels are more likely to pay existing workers for extra hours/efforts
instead of hiring additional employees. The incremental adjustment of wages and salaries

by positions (e.g., entry level versus managerial level) however is not clear.

Table 40. Economic Ratios for Tourist Hotels in Taiwan

 

 

Predicted Predicted Computed average salary

jobs per NT$ income per employee

million sales ratio to sales ratio (NT$ millions per year)

Occupancy rate (A) (B) (NH)
42% 0.88 0.37 0.42
52% 0.76 0.36 0.47
62% 0.66 0.35 0.53
72% 0.57 0.34 0.60
82% 0.49 0.33 0.67

 

 

 

38 “Returns to scale" is defined as the percentage increase in output by a one percent increase in all inputs.
representing different cost structures with respect to the volume of sales. This is different from the hotel
scale. which is related to the size ofindividual establishment. measured by number of rooms each hotels
has.

3., Total income Average income per job * total jobs

 

—— = Average income * thejobs to sales ratio
Total sales Total sales

91

 

 

 

 

 

 

1.00
Jobs to NTS million sales
0‘83 M‘ iiiiiii A '3 7 I Computed average salary per
\ ~ employee (NT$ million;I Year)
\ ~
0.70 ' ' '\ '~ -*
\ ~
\ .
\ .
.1 \ .
Income to sales ratio
0.25
35% 45% 55% 65% 75% 85% 95%

Occupancy rate

 

 

 

Figure 6. Job Ratio, Income Ratio and Average Salary per Employee by Occupancy Rates for

Tourist Hotels in Taiwan

The direction and magnitude ofbiases associated with fixed I-O ratios can be
approached from the perspective of returns to scale and cost structures. The coefficients
in the standard [-0 table represent the average production function. indicating the general
performance of businesses at a specific level of supply and demand in a year. For services
with constant returns to scale. the standard I-O model provides unbiased impact estimates
regardless of the level of capacity utilization (Table 41). For services with increasing
returns to scale (IRTS), the standard I-O model overestimates jobs and total income. and
underestimates profits when applying average coefficients to scenarios with capacity
utilization higher than the base-year point. The biases result from greater labor efficiency.
economies of scale. and substitutions between capital and labor inputs under higher
demand. For services following decreasing returns to scale. the biases will be in the

opposite direction.

Table 41. Sensitivity of Impact Estimation using Fixed [-0 Ratios by Returns to Scale

 

 

Increasing returns Constant returns Decreasing returns

Standard l-O model with fixed to scale to scale to scale

ratios (IRTS) (C RTS) (DRTS)
Scenarios with capacity utilization higher than the base year level

Jobs Overestimation Unbiased Underestimation

lncome Overestimation Unbiased Underestimation

Profits U nderestimation Unbiased Overestimation

 

Scenarios with capacity utilization lower than the base year level

Jobs Underestimation Unbiased Overestimation
lncome Underestimation Unbiased Overestimation
Profits Overestimation Unbiased Underestimation

 

Evaluation of the DTAP Policy

Economic Impacts

Achieving the goals of the “Doubling Tourists Arrivals‘ Plan“ would increase
Taiwan's total foreign receipts. The current lodging capacity constrains the growth of
international arrivals to 81% ofthe DTAP policy target of 4.8 million international
visitors by 2008 (scenario 1). Total visitor spending under this scenario is US$ 5.2
billion. With an expanded lodging capacity to 38.900 rooms and achieving 75% ofthe
DTAP target (scenario 2). visitor spending is projected to be US$49 billion. Achieving
100% ofthe visitation target in scenario 3 yields US$65 billion in visitor spending by
2008. The GDP contribution of inbound tourism is expected to grow from 0.76% in 2001
to 0.94% in scenario 1. 0.87% in scenario 2. or 1.16% in scenario 3 (direct et‘t‘ects3(l). The
percentage of GDP contribution will depend upon the Taiwan GDP annual growth and

future exchange rates.

 

'; . . _ . . . . . . . .
‘0 Gross Domestic Products of Taiwan was NT$ 9.45 trillion In 2001. Assuming GDP Increases with a 3%

annual growth rate from 2001. the forecasted GDP in 2008 would be NT$ 1 1.28 trillion.

93

Doubling international visitors from 2.6 million in 2001 to 4.8 million in 2008
does not guarantee the doubling ofemployment and personal income. Total economic
impacts for service sectors are determined simultaneously by final demand changes as
well as capacity utilization rates. The supply level influences employment, wages. and
business profits. The difference in impact estimates among the three scenarios between
fixed and varying ratios models is as much as 30% for hotel jobs and 7% for hotel wages
and salaries; 4% for total jobs and 2% for total income through out the entire economy.
As transportation. amusements and food services may experience similar changing
returns to scale. the reported differences in impact estimates using fixed and varying

ratios are conservative.

Implications

Government agencies should coordinate expansion of tourism supply with
increased demand so that a balance in capacity utilization can be maintained. Greatly
expanding the demand level will result in high capacity utilization rate. Most ofthe
increased sales will be absorbed as business profits, and the creation ofjobs and personal
income will not be in direct proportion to increased sales. Greatly expanding the supply
level on the other hand will result in lower capacity utilization. Under the proposed
“Tourist Hotel Development Plan”. hotel capacity expands more than the anticipated
increase in international visitors in the DTAP policy. The excess supply, while supporting
more jobs per million sales. will result in lower average income and business profits. and
may discourage future investment.

One approach to facilitate the balance of demand and supply growth is through

tax instruments. Tax breaks and financial incentives can be implemented in the early

94

stage of the policy to encourage private investment when an expansion of the supply level
is needed. Raising business tax when the market is matured. on the other hand. can help
to control a blooming tourism demand by diverting a proportion of business profits to
governments. The increased tax can be used to improve public welfares for the local
regions. and avoids the increased impacts solely benefit a handful of stakeholders.
Coordinating tourism development and economic growth across regions can also
facilitate the balance of demand and supply in tourism activities. Currently. most ofthe
economic impacts of international visitor spending accrue to northern Taiwan. This
region provides 85% of hotel room nights. In addition. Taipei. the capital area. provides
the majority of financial & business services. retailing and personal services for the
country. In other words. northern Taiwan captures most of the direct and indirect effects.
Balancing regional tourism development can reduce the risk of “putting all eggs
in one basket” and minimize the regional demand and supply fluctuations. Demand
stability can be enhanced by promoting diversified tourism attractions across the nation
so that the overall visitation volume (demand level) are more resistant to changes in
outside factors. such as economic crisis. earthquake or disease outbreaks in a single
region. Second. diverting demand to other regions simultaneously reduces the supply
pressure on northern Taiwan. and improves the utilization of tourism resources across the
nation. Infrastructure. transportation systems. and a diversity of tourism attractions are
therefore recommended to encourage visits to other regions. Especially nature-based
attractions, such as national parks. and culture tourism can be promoted to establish

distinct niches from the urban-based tourism in Taipei.

Last. a data collection instrument is recommended. C urrently. the Taiwan
govemment has an extensive system to keep track of inbound tourism in terms of
monthly visits. length of stay and average spending per visitor. This information provides
an accurate assessment of international visitor volume and spending in Taiwan.
Infomiation regarding supply levels. however. is relatively incomplete. Small-scale
service firms are large in numbers and their cost structures and manpower usage may
fluctuate. Systematically inventorying supply levels. in temis of the number and scale of

establishments. and collecting operational data for major service sectors is recommended.

Application of the Varying Ratios [-0 Model

The varying ratios model applies whenever the policy being evaluated causes
changes in capacity utilization from the base year point (the year 1-0 table is compiled).
The problem of using a fixed ratios model is particularly evident when evaluating special
events and festivals. These are generally held for short periods. creating peaks in demand
and high capacity utilization. In these instances. the fixed ratios model will overestimate
jobs and income effects and underestimate business profits. such as the example of the
British Open (Gelan. 2003). This is because increased sales in the short run are
accommodated by greater labor efficiency. Jobs and wage payments therefore are not
increased in direct proportion to additional sales. As documented by Hatch (1986) in the
study of the Grand Prix car racing event in Australia. profits and value added components
increased at higher utilization levels. Using the varying ratios model illustrates the nature
of biases and provides quantitative estimates of the errors associated with fixed ratios

models.

96

The varying ratios model is also appropriate when evaluating significant drops in
tourism due to weather. disease outbreaks. or terrorist activities. Using fixed ratios in
these cases will overestimate job and income losses because fimis cannot immediately
reduce labor inputs in proportion to reduced sales. The loss in business profits will be
greater than predicted using a fixed ratios model as costs per unit sale will be above the
baseline.

In evaluating long range policies. the rate of growth in supply must be considered
along with anticipated demand increases. If supply and demand expand at similar rates
from the base year. capacity utilization will remain relatively constant and standard I-O
models will yield unbiased impact estimates. If demand grows faster than supply.
utilization rates will increase and fixed ratio models will overestimate job and income
effects. If supply grows faster than demand. the bias with fixed ratio models will be in the

opposite direction.

Recommendations for Future Studies
The major finding ofthis study. the changing jobs and income by capacity
utilization. rests on the specified regression models. The current equations include four
independent variables: occupancy rates. hotel scales. consumer price index and regions.
The relatively low adjusted R-square values. especially for the income to sales ratio
(Table 22). indicate the potential to refine current regression models and to incorporate
other independent variables. Business operation structures. such as ownerships. human

resource management. and government tax policies. are some potential variables to be

97

 

considered. Refining current regression models helps to raise the credibility on its
findings and avoid possible specification errors.

Future research is also recommended to understand managerial responses to
fluctuation in demand for services. Impacts of demand changes ultimately depend on how
fimts adapt to these changes in the short- and long-term. Increased sales can be
accommodated by using existing workers versus hiring new employees. Also. the
addition of newjobs can be differentiated by management/ entry-level positions.
seasonal/ permanent jobs. or full-time/ part-time status. Labor practices influence the jobs
to sales ratios and subsequently determine the proportion oflabor costs in business
operation. Relationships between labor efficiency and wage compensation are also not
well documented. A rise in wages to compensate employees for increased workload can
be achieved through over-time pay. additional employee benefits or profit sharing.

The examination of varying ratios should be extended to other services. especially
food & beverage services. transportation and amusement sectors. These three sectors
have characteristics similar to accommodations in terms of product perishability and
pricing strategies. A clear definition and measure of capacity utilization for each industry
is required to quantify the relationships between capacity utilization and value added
components. Modifying the value added components for multiple sectors simultaneously
is also encouraged so that the joint effects on the rest of the economy can be modeled.

Value added is only part of the production function. To obtain comprehensive
insights between business operation and capacity utilization. examination of changes in
the Leontief technical coefficients that are caused by utilization rates is also

recommended. Technical input coefficients depend on output prices (such as changes in

98

room rate in response to a rising demand). as changes in output prices affect the relative
price ratio between intermediate inputs and final output. Technical coefficients also can
vary due to economies of scale in input factors. For example. cost for utilities or business
services per occupied hotel room is generally lower when hotels operate at higher
occupancy rates. Empirical data are therefore needed to establish relationships between 1)
price changes and capacity utilization. and 2) intermediate inputs and capacity utilization.
Given a rise in occupancy rates. the likely pattern is an increase in output price and
proportionally lower per unit cost for intermediate inputs. These patterns would lead to a
reduced type 1 sales multiplier and larger value added components. The joint influence on
the type 11 sales multipliers however is undetermined as it depends on the nature of each

industry.

Conclusion

The assumption of linearity is a frequent criticism of standard l-O analysis
(C rompton. 1995; West & Gamagc. 1997; West. 1995). It has been demonstrated that this
assumption leads to overestimation ofthe secondary effects. especially with respect to
income and employment. The proposed I-O model with jobs to sales ratios and income to
sales ratios varying as a function ofcapacity utilization admits changing returns to scale.
and provides better estimates ofthe value added components.

This study identifies the potential biases associated with the applications of
standard [-0 models to service industries as tourism generally experiences fluctuation in
demand and supply level. The most foreseeable challenge in expanding the current model

lies in data availability. as extensive observations for individual establishments under

99

different level ofcapacity utilization are required. Comprehensive data collection plans
and business cooperation are therefore needed. Such efforts will greatly improve existing

policy models.

100

 

APPENDICES

lOl

Appendix A. Allocation of Visitor Spending to [-0 Sectors

 

Spending category

l-O Sector

 

Spending inside the hotel

Food
Transportation
Entertainment

Others

Shopping

 

Lodging (90%)
Restaurant (10%)

Restaurant (100%)
Transportation (100%)
Entertainment (100%)

Theatrical. dance. music and other arts activities (50%)
Personal service (509-16)

Production sectorsa
Food & beverage manufacturing (40%)
Tobacco manufacturing (5%)
Apparel & clothing accessories manufacturing (10%)
Leather. fur & allied product manufacturing (5%)
Wood products (5%)
Paper and printing (5%)
Chemicals Petroleum & coal products manufacturing (10%)
Non-metallic mineral products manufacturing (3%)
Basic metal industries & products (3%)
Machinery & Electronic Products (5%)

 

a . . . . .
The percentages for allocating shopping expenses to ten production sectors was estnnated usmg the 2001

ITCTS.

Appendix B. Definitions of Taiwan [-0 Sectors

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Original Spending
New [-0 160 [—0 categories in the
Code New 39 [-0 Sectors Code Employment Data Category results table
I Agricultural products & l-lO Agriculture. animal husbandry Agriculture.
livestock forest. fisheries
2 Forest products I I Forestry & logging
3 Fisheries 12 Fishing
4 Minerals I3-I7 Mining & quanying Manufacturing
5 Process foods & beverages 18-32 Food & beverage manufacturing
6 Tobacco 33 Tobacco manufacturing
7 Textile mill products 34-39 Textiles mills
8 Wearing apparel and 40—42 Apparel. clothing accessories & other
accessories textile products manufacturing
9 Leather & leather products 43-45 Leather & fur & allied product
manufacturing
10 Wood & wood products 46-49 Wood & bamboo products
manufacturing
1 1 Paper & paper products 8; 50-53 Pulp. paper & paper products
rinted matter manufacturing
12 Chemical manufactures & 5.4-70 Chemical matter manufacturing
petroleum refining products Chemical products manufacturing
Rubber products manufacturing
Plastic products manufacturing&
Petroleum & coal products
manufacturing
13 Non-metallic Mineral 71-75 Non-metallic mineral products
Products Manufacturing manufacturing
14 Iron and Steel Products & 76-85 Basic metal industries & Fabricated
Miscellaneous Metals & metal products manufacturing
Metallic Products
15 Machinery & electronic 86-105 Electrical & Electronic Machinery
products Precision. optical. medical instrument
manufacturing. watches & clocks
Other industrial products
manufacturing
16 Transport equipment 106-109 Transport equipment manufacturing
17 Other products 1 10-1 13 Other industrial products
manufacturing
18 Electricity 1 14 Electric power supply Utilities
19 Gas 1 15 Gas supply
20 City water 1 16 Thermal energy supply
21 Construction 117-120 Construction Construction
22 Wholesale trade 121 Wholesale trade Wholesale trade
23 Retail trade 122 Retail trade Retail trade
24 International trade 123 Foreign trade hlntemational
rade

 

 

103

 

Appendix B. (cont'd).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Original iSpending
New [-0 160 1-0 categories in the
Code New 39 1-0 Sectors Code Emplovment Data Categosy results table
25 Food & beverage services 124 Eating & drinking places Food & beverage
services
26 Transportation 125-130 Transportation Transportation
27 Warehousing 13 | Warehousing & storage Business services
28 Postal services & telegram & 132-133 Communication
telephone
29 Financing & auxiliary 134 Finance & insurance
financinU
30 Securities & futures 135 Finance & insurance
31 Insurance carriers 136 Finance & insurance
32 Real estate & rental & leasing 137. 139 Finance & insurance
33 Hotel services I38 Accommodation service Hotel services
34 Business services 140-145 Business Services Business services
35 Public administration 146 Public administration
36 Sanitary. pollution controlling. I47-153 Sanitary. pollution controlling
education services services
37 Theatrical. dance. music and 154 Motion picture industries Theatrical.
other arts activities dance. music and
ther arts
ctivities
p8 Recreational & cultural 155 Entertainments Recreational &
services cultural services
9 Other personal services 156-I60 Personal services Other personal
services

 

 

 

Note. The original 160 sector [-0 table is regrouped into 39 sectors to facilitate computation and
interpretation.

104

 

Appendix C. Economic Ratios for Taiwan [-0 Sectors, 1999

 

 

lncome to Profits to Value added Jobs/million

sales ratio sales ratio to sales ratio NT$ sales
Agriculture & Other Nature Products
1. Agricultural Products & Livestock 0.303 0.100 0.456 1.809
2. Forest Products 0.675 0.053 0.814 1.684
3. Fisheries 0.283 0.241 0.593 0.713
4. Minerals 0.400 0.113 0.609 0.138
Manufacturing
5. Process Foods 8; Beverages 0.108 0.038 0.245 0.264
6. Tobacco 0.1 16 -0.010 0.731 0.038
7. Textile Mill Products 0.149 0.039 0.241 0.269
8. Wearing Apparel and Accessories 0.243 0.028 0.300 0.886
9. Leather & Leather Products 0.155 0.031 0.215 0.653
10. Wood & Wood Products 0.271 0.063 0.391 1.153
1 1. Paper & Paper Products 8; Printed Matter 0.239 0.052 0.360 0.427
12. Chemical Manufactures & Petroleum 0.125 0.052 0.275 0.212

Refining

l3. Non-metallic Mineral Products 0.202 0.074 0.369 0.318
14. Iron. Steel Products & Miscellaneous Metals 0.143 0.058 0.246 0.384
15. Machinery & Electronic Products 0.136 0.066 0.256 0.264
16. Transport Equipment 0.141 0.050 0.308 0.258
17. Other manufacturing products 0.218 0.060 0.315 0.318
Utilities
18. Electricity 0.121 0.156 0.508 0.072
19. Gas 0.100 0.022 0.145 0.155
20. City Water 0.339 0.086 0.594 0.233
Construction
21. Construction 0.261 0.058 0.342 0.813
Services
22. Wholesale Trade 0.496 0.131 0.687 0.480
23. Retail Trade 0.485 0.192 0.738 1.383
24. lntemational Trade 0.339 0.179 0.599 0.433
25. Food & Beverage Services 0.527 0.123 0.707 1.841
26. Transportation 0.362 0.067 0.560 0.523
27. Warehousing 0.326 0.202 0.675 0.198
28. Postal Services & Telegram & Telephone 0.340 0.255 0.757 0.252
29. Financing & auxiliary financing 0.334 0.288 0.752 0.324
30. Securities & futures 0.377 0.169 0.654 0.324
31. Insurance carriers 0.41 1 0.248 0.726 0.324
32. Real estate & rental & leasing 0.028 0.692 0.819 0.041
33. Hotel Services 0.41 1 0.116 0.649 0.497
34. Business Services 0.371 0.143 0.562 0.522
35. Public administration 0.615 0.000 0.714 0.327
36. Sanitary. pollution controlling. education 0.707 0.067 0.818 0.911
37. Theatrical. dance. music and other arts 0.330 0.168 0.567 1.091
38. Recreational & Cultural Services 0.376 0.193 0.693 0.729
39. Other Personal Services 0.407 0.144 0.579 0.876

 

 

105

Appendix D. Multipliers for Taiwan [-0 Sectors, 1999

 

 

Type 1 sales Type 11 sales
multipliers multipliers
Agriculture & Other Nature Products
1. Agricultural Products & Livestock 1.969 4.024
2. Forest Products 1.246 4.235
3. Fisheries 1.412 2.909
4. Minerals 1.393 3.227
Manufacturing
5. Process Foods & Beverages 2.135 3.706
6. Tobacco 1.286 2.005
7. Textile Mill Products 2.156 3.567
8. Wearing Apparel and Accessories 2.135 3.962
9. Leather & Leather Products 1.560 2.635
10. Wood & Wood Products 1.535 3.118
1 1. Paper & Paper Products 8; Printed Matter 1,744 3,412
12. Chemical Manufactures & Petroleum Refining 1,674 2,679
13. Non-metallic Mineral Products 1.779 3.329
14. Iron. Steel Products & Miscellaneous Metals 7.028 3,454
15. Machinery & Electronic Products 1.633 2.700
16. Transport Equipment 1.908 3.215
17. Other manufacturing products 1.831 3.356
Utilities
18. Electricity 1.491 2.305
19. Gas 1.875 3.235
20. City Water 1.616 3.541
Construction
21. Construction 1,959 3.846
Services
22. Wholesale Trade 1.421 3.761
23. Retail Trade 1.363 3.531
24. lntemational Trade 1.519 3.394
25. Food & Beverage Services 1.423 3.768
26. Transportation 1.449 3.271
27. Warehousing 1.456 3.160
28. Postal Services 8; Telegram & Telephone 1304 2,913
29. Financing & auxiliary financing 1.319 2.968
30. Securities & futures 1.472 3.447
31. Insurance carriers 1.301 3.259
32. Real estate & rental & leasing 1,302 1.724
33. Hotel Services 1.507 3.558
34. Business Services 1.600 3.641
35. Public administration 1.316 3.975
36. Sanitary. pollution controlling. education 1,234 4.132
37. Theatrical. dance. music and other arts 1.564 3_449
38. Recreational & Cultural Services 1,427 3.261
39. Other Personal Services 1.573 3.680

 

 

106

Appendix E. Trade Margin, Transportation Margin and Regional
Purchase Coefficients for Manufacturing Sectors

 

Transportation

Regional purchase

 

Trade margin margin coefficient (RPC)
Process foods & beverages 20% 1% 82%
Tobacco 23% 0% 58%
Textile mill products 9% 1% 91%
Wearing apparel and accessories 23% 1% 80%
Leather & leather products 17% 2% 70%
Wood & wood products 15% 4% 67%
Paper products 19% 1% 81%
Chemical manufactures products 10% 1% 73%
Non-metallic mineral products % 4% 85%
Iron and steel products 10% 2% 7 %
Machinery & electronic products 8% 1% 67%
Transport equipment 14% 1% 83%
Other manufacturing products 14% 1% 49%

 

 

Note.

Trade margin: Margins of intemational trade. wholesale trade and retail trade.
Regional purchase coefficients: The proportion ofthe regional demand for a good or service that is fulfilled

by production in Taiwan.

107

Appendix F. Scatterplot of Linear and Natural Log for Jobs to Sales
Ratios versus Occupancy Rates

Linear (adjusted R-sguare value = 0.447)

30.0—

 

25.0-1

20.0—1

15.0—1

10.0—

Jobs to sales ratio

5.0—

 

0.0 -"

 

 

 

Occupancy rate

Loglinear (adjusted R-sguare value = 0.479)

 

109 jobs to sales ratio

 

0.5—

 

 

 

l l l |
0.2 0.4 0.6 0.8

Occupancy rate

108

Appendix G. Scatterplot of Personal Income to Sales Ratios versus

Personal income to sales ratio

Occupancy Rates

 

 

 

0.8—
6 o o o
0. —
0 0° 0
O O O a) O
o o o 000% o
o o OOOJ’ an o 0&9
0.4— .ooma .98 .4”? @908 o.
0 ° 08% 00 @119 000 o
9&3 o o (boo
o 0 Q o 1; 03%336
O Q) o O flooofioo
0.2_ o O O
o o
00 °
0
0.0—i
I j l I
0.2 (1.4 0.6 0.8

Occupancy rate

109

 

 

BIBLIOGRAPHY

Adams. P. D.. Dixon. P. B., McDonald. D.. Meagher. G. A.. & Parmenter. B. R. (1994).
Forecasts for the Australian economy using the MONASH model. International
Journal of Forecasting. 10(4). 557-571.

Allen. M. (1988). Strategic management ofconsumer services. Long Range Planning.
21(6), 20-25.

Archer. B. (1977). Tourism multipliers: The state ofartfardiff: University of Wales
Press.

Archer. B. (1978). Input-output analysis: Its strength. limitations and weaknesses. Annals
of'Tourism Research, 5(1). 185-186.

Archer. B. (1984). Economic impact: Misleading multiplier. Annals of Tourism Research,
11(3), 517-518.

Archer. B.. & Fletcher. J. (1988). The tourist multiplier. T eoros, 7(3). 69.

Archer. B.. & Fletcher. .1. (1996). The economic impact oftourism in the Seychelles.
Annals of Tourism Research. 23 (1), 32-47.

Arenberg. Y. (1991). Service capacity. demand fluctuations. and economic behavior.
A meriean Economist, 35(1). 52-55.

Armstrong. H.. & Taylor. J. (2000). Regional economies and policy (3rd ed.). Oxford.
UK: Blackwell Publishers.

Berrnan. S. D. (1994). The challenge of Cuban tourism. The Cornell Hotel and
Restaurant Administration Quarterly, 35(3). 10-15.

Berndt. E. R.. & Morrison. C. J. (1981). Capacity utilization measures: Underlying
economic theory and an alternative approach. The American Economic Review,
71(2). 48-52.

110

Blake. A. (2000). The economic effects of tourism in Spain. (Discussion Paper 2000/2):
Christel DeHaan Tourism and Travel Research Institute. Nottingham University.
UK.

Blake. A., & Gillham. J. (2001). A multi—regional CGE model of tourism in Spain. Paper
presented at the European Trade Study Group. Third Annual Conference,
Brussels. Belgium.

Blake. A., & Sinclair. T. (2002). Tourism crisis management: Responding to September
11 (Discussion Paper 2002/ 7): Christel DeHaan Tourism and Travel Research
Institute. Nottingham University. UK.

Borooah. V. K. (1999). The supply of hotel rooms in Queensland. Australia. Annals of
Tourism Research. 26(4). 985-1003.

Briassoulis. H. (1991). Methodological issues: Tourism input-output analysis. Annals of
Tourism Research, 18(3). 485-495.

Bryden. J. M. (1973). Tourism and Development: A Case Study of the Commonwealth
CaribbeanCambridge. England: University Press.

Caves. D. W.. & Christensen. L. R. (1988). The importance of economies of scale.
capacity utilization. and density in explaining interindustry differences in
productivity growth. Logistics and transportation review, 24(1), 3-32.

Cooper. A., & Wilson, A. (2002). Extending the relevance of TSA research for the UK:
General equilibrium and spillover analysis. Tourism Economics, 8(1). 7-3 7.

C opeland. B. R. (1991). Tourism welfare and de-industrialization in a small Open
economy. Economica, 58(232). 515-529.

Corcoran. K.. Allcock. A.. Frost. T.. & Johnson. L. (1999). Valuing tourism: Methods
and techniques. (BTR Occasional Paper. Number 28): Bureau of Tourism
Research. Canberra. Australia.

Council for Economic Planning and Development. (2002). Challenge 2008- National
Development Plan: Executive Yuan. Taiwan. R.O.C.

111

C rompton. J. L. (1995). Economic impact analysis of sports facilities and events: Eleven
sources ofmisapplication. Journal of Sport Management, 9. 14-35.

Directorate-General of Budget Accounting and Statistics. (2003). 1999 National Input-
Output Table. from http://www.dgbas.gov.tw/

Directorate-General of Budget Accounting and Statistics. (2003). 2001 Industry.
Commerce and Service Census: General report. Taipei. Taiwan: Executive Yuan.
Taiwan.

Directorate-General of Budget Accounting and Statistics. (2003). Manpower survey and
its supplementary surveys. from http://www.dgbas.gov.tw/census~n/four/e44.htm

Directorate—General of Budget Accounting and Statistics. (2004). Price statistics monthly
in Taiwan area ofthe Republic of China. Taipei. Taiwan: Executive Yuan.
Taiwan.

Dixon. P. B. (1994). Applied general equilibrium modelling: Achievement, failure and
potentialClayton. Australia: Centre of Policy Studies. Monash University.

Dwyer. L.. Forsyth. P.. Madden. J .. & Spurr. R. (2000). Economic impacts of inbound
tourism under different assumptions regarding the macroeconomy. Current Issues
in Tourism. 3(4). 325-363.

Dwyer. L.. Forsyth. P.. & Spurr. R. (2003). Inter-industry effects of tourism growth:
Implications for destination managers. Tourism Economics, 9(2). 117-132.

Dwyer. L.. F orsyth. P.. & Spurr. R. (2004). Evaluating tourism's economic effects: New
and old approaches. Tourism Management, 25(3). 307-317.

Eriksen. L.. & Ahmt. T. (1999). Measuring and modeling the regional impact oftourism
in Denmark. International Journal of Tourism Research, 1(5). 313-327.

Fletcher. J.. & Snee. H. (1989). Tourism multiplier effects. In S. F. Witt & L. Moutinho
(Eds). Tourism Marketing and Management Handbook (pp. 529-531).
Hertfordshire. UK: Prentice Hall lntemational Ltd. (UK).

Fletcher. J. E. (1989). Input-output analysis and tourism impact studies. Annals of
Tourism Research, 16(4). 514-529.

Frechtling. D. C. (1999). The tourism satellite account: Foundations. progress and issues.
Tourism Management. 20(1). 163-1 70.

Gelan. A. (2003). Local economic impacts: The British Open. Annals of Tourism
Research, 30(2). 406-425.

Ghosh. A. (1958). Input-Output approach in an allocation system. Economica, 25(97).
58-64.

Hatch. J. H. (1986). The impact of the Grand Prix on the restaurant industry. In J. P. A.
Burns. J. H. Hatch & T. J. Mules (Eds). The Adelaide Grand Prix - The impact of
a special event: The Centre for South Australian Economic Studies.

Hollenstein. H. (2001). Innovation modes in the Swiss service sector (WIFO working
papers. No. 156). Vienna: Austrian Institute of Economic Research.

Hudson. E. A., & Jorgenson. D. W. (1974). US. energy policy and economic growth.
1975-2000. The Bell Journal ofEconomics and Management Science, 5(2), 461-
514.

Hunn. C .. & Mangan. J. (1999). Estimating the economic impact of tourism at the local.
regional and state or territory level. including consideration of the multiplier
effect. In Valuing Tourism: Methods and Techniques (Occasional Paper No. 28).
Canberra: Bureau of Tourism Research. Australia.

Kimes. S. E. (1989). Yield management: A tool for capacity-considered service firms.
Journal of Operations Management, 8(4). 348-363.

Krakover. S. (2000). Partitioning seasonal employment in the hospitality industry.
Tourism il/Ianagement. 21(5). 461-471.

Krikelas. A. C. (1992). Why regions grow: A review of research on the economic base
model. Economic Review - Federal Reserve Bank of A tlanta. 77(4). 16-29.

113

Lapierre. J .. & Hayes. D. (1994). The Tourism Satellite Account (National income and
expenditure accounts. No. 13-001): Statistics C anada. Canada.

Leontief. W. W. (1936). Quantitative input and output relations in the economic systems
of the United States. The Review of Economic Statistics, 18(3). 105-125.

Lewis. R. C.. & Chambers. R. E. (1989). Marketing leadership in hospitality:
Foundations and practices. New York: Van Nostrand Reinhold.

Lin. B.-H.. & Lin. H.-H. (2000). A study of economies of scale and economies of scope

in Taiwan intemational tourist hotels. Asia Pacific Journal of Tourism Research.
5(2). 21-28.

Liu. J.. & Var. T. (1982). Differential multipliers for the accommodation sector. Tourism
management. 3(3). 177-187.

Lovelock. H. C. (1992). Seeking synergy in service operations: Seven things marketers
need to know about service operations. European illanagement .Iournal, 10(1).
22-29.

McDougall. R. A. (1995). Computable general equilibrium modeling: Introduction and
overview. Asia Pacific Economic Review, 1 (1). 88-91.

McGregor. P. G.. Swales. J. K.. & Yin. Y. P. (1996). A long-run interpretation of
regional Input-Output analysis. Journal of Regional Science, 36(3). 479-501.

Miller. R. E.. & Blair. P. D. (1985). Input-output analysisx‘ Foundations and extensions.
Englewood Cliffs. N.J.: Prentice-Hall.

Ministry of Economic Affairs. (2005). Economic indicators. from
http://2k3dmz2.moea.gov.tw/gnweb/english/e__main.aspx?Page=D

Moses. L. N. (1974). Outputs and prices in interindustry models. Paper ofthe Regional
Science Association, 32. 7-18.

Mullins. L. J. (1993). The hotel and the open systems model oforganisational analysis.
The Service Industries Journal, 13(1). 1-15.

114

Nelson. R. A. (1989). On the measurement of capacity utilization. The Journal of
Industrial Economics. 3 7(3). 273-286.

Ng. I. C. L.. Wirtz. J.. & Lee. K. S. (1999). The strategic role of unused service capacity.
International Journal ofService Industry Management, 10(2). 21 1-238.

O'Hagan. J.. & Mooney. D. (1983). Input-Output multipliers in a small open economy:
An application to tourism. The Economics and Social Review, 14(4). 273-280.

OANDA Corporation. (2004). FXHistory: Historical currency exchange rates.

Otto. D.. & Johnson. T. G. (1993). Microcomputer-based Input-Output modeling:
Applications to economic development. Boulder: Westview Press.

Pan. C .-M. Market structure and profitability in the international tourist hotel industry.
Tourism tl/Ianagement. In Press, Corrected Proof.

Peneder. M.. Kaniovski. S.. & Dachs. B. (2003). What follows tertiarisation? Structural
change and the role of knowledge-based services. The Service Industries Journal,
23(2). 47.

Porter. P. K. (1999). Mega-sports events as municipal investments: A critique of impact
analysis. In J. Fizel. E. Gustafson & L. Hadley (Eds). Sports Economics: Current
Research (pp. 61-74). Westport. Connecticut: Praeger Publishers.

Quevedo. J. (2002). The Tourism Satellite Account (TSA) from the regional point of
view: Reflections for debate. In Tourism Satellite Account: Implementation
project. Madrid. Spain: World Tourism Organization (WTO).

Rose. A. (1995). Input-output economics and computable general equilibrium models.
Structural change and economic dynamics, 6. 295-304.

Rose. A.. & Miernyk. W. (1989). Input-output analysis: The first fifty years. Economic
systems Research, 1(2). 229-271 .

Smeral. E. (2003). A structural view of tourism grow. Tourism Economics, 9(1). 77-93.

Smith. S. L. J. (1994). The tourism product. Annals of Tourism Research. 21(3). 582-595.

115

Song. H. (2002). Assessing the economic impacts of tourism: An overview of analytical
tools. from
http://w"ww.mediaexchange.info/2002/1II/downloads/DrHaiyanSong.pdf

Stynes. D. J.. & Sun. Y.—Y. (2003). Impacts ofvisitor spending on the local economy:
Sequoia and Kings Canyon National Parks, 2002.East Lansing. MI: Department
of Park. Recreation and Tourism Resources. Michigan State University.

Sugiyarto. G.. Blake. A.. & Sinclair. M. T. (2003). Tourism and globalization: Economic
impact in Indonesia. Annals of Tourism Research, 3 0(3). 683-701.

Taiwan Tourism Bureau. (1998-2004). Report on the R. 0. C domestic tourism survey.
Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (1999-2003). Monthly report on international tourist hotel
operating statistics. Taipei. Taiwan: Tourism Bureau. Ministry of Transportation
and Communications. R.O.C.

Taiwan Tourism Bureau. (1999-2003). The operating report ofinternational tourist
hotels in Taiwan. Taipei. Taiwan: Tourism Bureau. Ministry of Transportation
and Communications. R.O.C.

Taiwan Tourism Bureau. (2002). Domestic Travel by Domestic Residents Survey.
2001 .Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (2002). Taiwan Tourism Satellite Accounts. Taipei. Taiwan:
Taiwan Tourism Bureau. Ministry of Transportation and Communication.

Taiwan Tourism Bureau. (2003). Annual report on tourism 2002 Republic ofChina.
Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (2003). Monthly news letter- September. Taipei. Taiwan:
Tourism Bureau. Ministry of Transportation and Communications. R.O.C.

116

Taiwan Tourism Bureau. (2003). Regulation for the management of tourist hotel
enterprises. Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (2003). Standardfor buildings and fitcilities of tourist hotels.
Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (2003). Survey report on inbound traveler expenditures and
trends, 2002.Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Taiwan Tourism Bureau. (2003). Tourism hotels development plan. from
http://202.39.225.136lindexc.asp

Taiwan Tourism Bureau. (2004). Annual report on tourism 2003 Republic ofChina.
Taipei. Taiwan: Tourism Bureau. Ministry of Transportation and
Communications. R.O.C.

Upneja, A.. Shafer. E. I... Seo. W.. & Yoon. J. (2001). Economic benefits of sport fishing
and angler wildlife watching in Pennsylvania. Journal of Travel Research, 40(1).
68-78.

Wagner. J. E. (1997). Estimating the economic impacts of tourism. Annals of Tourism
Research. 24(3). 592-608.

Wang. T.-F.. & Wang. K. (1991 ). Ecom’m'zic impact analysis of tourism industry
development. Taichung. Taiwan: Economic Research Institute. ChungHsing
University.

Wanhill. S. R. C. (1988). Tourism multipliers under capacity constraint. The Service
Industries Journal. 8(2). 136-142.

West. G.. & Gamag .

e
Economics. 3 (

A. (1997). Differential multipliers for tourism in Victoria. Tourism
1). 57-68.

West. G.. & Gamagc. A. (2001). Macro effects of tourism in Victoria. Australia: A
nonlinear Input-Output approach. Journal of Travel Research. 40(1). 101-109.

117

West. G. R. (1995). Comparison of Input-Output. Input-Output Econometric and
Computable General Equilibrium Impact Models at the regional level. Economic
Systems Research. 7(2). 209-227.

Woollett. G.. Townsend. J.. & Watts. G. (2003). Development onGEM—T— A
Computable General Equilibrium Model of tourism. Queensland. Australia:
Office of Economic and Statistical Research. Queensland Treasury.

World Travel and Tourism Council. (2002). Chinese Taipei: The impact of travel &
tourism onjobs and the economy - 2002. plus special report on September 11th
impacts: World Travel and Tourism Council.

118

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1|111111111111111111111111111111111