INTRODUCTION

and related industries. Therefore, in order to investigate the direct and indirect impacts of the natural disaster on agricultures in Taiwan and China, the present study employ an supply-side Inter-Country Output Analysis proposed by Miller and Blair (2009). To this end, this paper uses data from cross-strait Inter- Country Input-Output table of 96 sectors in 2006 reported in Lin’s (2013) and the statistics of Agricultural natural disasters published by the Council of Agriculture Statistics in Taiwan during 2005 to 2011 and the database of agricultural productivity in China in 2005 to 2010.

LITERATURE REVIEW
Climate change may cast influence on environments in various ways. For instance, Ahmed et al. (2015) concluded that 33% of farmers in the Province of Punjab, Pakistan, were unwilling to pay for a planned climate change adaptation program, whereas 67% were willing to pay (WTP). The predominant reasons for this willingness included ‘having impacts on agricultural production’, ‘feeling responsible for my contribution to climate change’ and ‘concern for the risk posed by climate change’. Ahmed et al (2015) and McElhinney (2016) further found farmers who were more WTP for a climate change adaptation program were more highly-educated, had higher incomes and had greater concern for climate change. In short, climate change not only affects the natural environment but also affects social awareness about the environmental sustainability.

Furen et al. (2005) argued that climate change would affect growth in population and economic development. A decrease in the amount of rainfall would increase the risk of drought, which would in turn affect the productivity of related industries. The authors use Social Accounting Matrix (SAM) and simulate six situations to explore the economic impacts of water restrictions and conclude that the food and beverage sectors undergo most

Mainland China and Taiwan have followed different patterns of political and economic development since 1949, resulting in significant differences in economic activities, social environments and developmental trajectories between Mainland China and Taiwan. Nevertheless, both of them has suffered from great damage caused by natural disasters, which would seriously affect economic systems, traffic infrastructure, agricultural industry and so on. More specifically, Taiwan depends largely on imports from and exports to Asian countries. Dominated by sub-tropical and tropical climate, Taiwan is often affected by Typhoons during summer and autumn. Although typhoons bring abundant rainfall, they also trigger natural disasters such as floods and landslides, which threaten residents’ lives and cause losses in economic activities. For instance, typhoon Morakot caused death of more than 600 people and losses of 164 billion in agricultural productions and infrastructure in 2009.

In recent years, Mainland China also suffers from disasters such as typhoon and drought. In Mainland China, there is not much variance in the total annual rainfall across years, whereas the amounts of monthly rainfall vary greatly within a year. According to the statistics of the National Office of Flood Control and Drought Management, in 2009, there were 43% drought-stricken in wheat-field and 429 million people faced the problem of lacking drinking water.

To summarize, although Mainland China and Taiwan share different industrial structures, they both suffer from natural disasters and the losses caused by these disasters in agriculture

____________________
DOI: 10.2112/ SI96-001.1 received 13 May 2019; accepted in revision 8 July 2019.
*Corresponding author: [email protected]
©Coastal Education and Research Foundation, Inc. 2019

2 Lin and Chou _________________________________________________________________________________________________

Table 1. The losses in value-added by natural disasters in Taiwan. 2005 2006
Sectors GDP rank GDP effect effect

1. 001 Farming products 46,012 1 7,996

2. 002 Livestock products 602 6 53

3. 003 Forest products 778 5 37

4. 004 Fishery products 3,710 2 147

5. 005 Agricultural services 2,357 3 403

6. 006 Crude petroleum and natural gas 186 13 30
   extraction

008 Other non-metallic minerals 249 11 42 013 Animal feeds 112 16
026 Petrochemical raw materials and 171 14 24
petroleum refining products

28. 028 Basic chemical materials 15

29. 029 Chemical fertilizers 527 7 91

032 Pesticides and herbicides 140 15 24 038 Plastic products 109 17 18 066 Electricity supply 363 9 57

70. 070 Wholesale trade and retail trade 1,134 4 165

71. 071 Railroad vehicle transportation and 267 10 41
    land transportation

081 Telecommunication services 108 18 16
Finance, securities, futures
083 and other activities auxiliary to 429 8 66
financial service activities
085 Real estate services 108 19 16 087 Professional, scientific and 223 12 35
technologic services

94. 094 Public administration and social 95 20 15
    association services

95. 095 Repair, domestic and personal
    services
    Total 59,140 9,504

Unit: Million NT
losses in all situations. Okuyama (2007) discussed the advantages and disadvantages of Input-Output model, Computable General Equilibrium (CGE) model and Social Accounting Matrix (SAM). This paper concluded that the model estimation should be guided by economic theories to prevent overestimation or underestimation in empirical or simulated results.

Wu (2003) investigated the influence of the reconstruction budget for the 921 Earthquake on the industrial structure and the economic development of tourism in central Taiwan. After reviewing various related literatures, Wu (2003) found RAS method was the one with least measurable error to build Regional Input-Output model, and therefore it was found to explore the linkage contributions and the economic multiplier impacts.

Lin et al. (2010) combined the regional and supply side Input- Output analysis developed by Miller and Blair (2009) to evaluate the losses of agricultural productions caused by natural disasters in regional economy. They found the losses of agricultural productions from natural disasters substantially affected income and employment effect, as high as a loss accounting for 30% of Agriculture GDP.

2007 2008 2009
rank GDP rank GDP rank GDP effect effect effect
1 27,110 1 32,035 1 28,394 8 194 8 223 8 2,292 11 147 9 79 15 9,305 4 567 3 1,441 3 9,913 2 1,366 2 1,621 2 1,648
13 101 13 123 12 180
9 142 10 169 10 197
288 11 52 16

2010 2011
rank GDP rank GDP rank effect effect
1 20,630 1 8,027 1 4 311 5 19 15 3 112 10 19 14 2 1,633 2 538 2 5 1,056 3 406 3
15 83 13 31 11

14 83 14 106 13 211
19 51 19 61 19
5 308 5 365 5 357 15 81 15 96 14 107 16 62 16 74 16
7 195 7 238 7 345
3 565 4 705 4 1,372
10 139 11 170 9 311 17 56 17 69 17 121
6 225 6 275 6 495
18 54 18 67 18 131 12 119 12 145 11 233

13 76 14 28 12
15 19 8 236 6 91 5
20 62 15 24 13 49 17 19 16
9 162 8 61 7 6 511 4 181 4
10 118 9 44 8 18 48 19 18 17
7 191 7 70 6
17 48 18 17 18 12 100 12 37 10

20 51 20 61 20 135
111 19

15 20 9,943

32,347 39,038

57,817

14 111 11 43 9

16 41 20 26,279

Journal of Coastal Research, Special Issue No. 96, 2019

METHODS

According to Miller and Blair (2009), natural disasters affect the supply of economic activities and result in output impairment in the economic system. This study uses the Inter- Country Input-Output model to estimate the influence of agricultural products loss caused by climate change in cross- strait industry as follows.

X=Z+F (1) X, Z and F are the output vector, middle demand matrix and
the final demand. The multi-regional trade matrix is given below:

ZLLLZLK ij ij
Z=MOM
ZKL ij
ij
In equation (2), ZLL is the i-th commodity produced in the j-th
LK ij

LZ 

industry in the L-th region. Z
from the j-th industry and the inputs produced in region L.

KK  ij

indicates the outputs in region K

(2)

Natural Disasters Caused by Climate Change and Agricultural Products Loss in Cross-Strait Industry 3_________________________________________________________________________________________________

Table 2. The losses in value-added by natural disasters in China.
Sectors GDP rank
effect
01 Farming products 15,65 1 7
02 Livestock products 5,668 2 03 Forest products 1,281 4 04 Fishery products 2,417 3 05 Agricultural services 1,014 6 06 Crude petroleum and natural gas extraction 86 15 08 Other non-metallic minerals 104 13 11 Flour 82 17 13 Animal feeds 318 7 26 Petrochemical raw materials and petroleum refining products 91 14 29 Chemical fertilizers 194 9 32 Pesticides and herbicides
66 Electricity supply 188 10

70. 70 Wholesale trade and retail trade 1,109 5

71. 71 Railroad vehicle transportation and land transportation 168 11

81 Telecommunication services 63 19 83 Finance, securities, futures 297 8
and other activities auxiliary to financial service activities
85 Real estate services 86 16 87 Professional, scientific and technologic services 157 12

94. 94 Public administration and social association services 66 18

95. 95 Repair, domestic and personal services 53 20
    Total 30,04 7

Unit: Million NT
Assuming that the input coefficients are fixed, the relationship of input and output can be estimated and the coefficient of middle input in different regions can be expressed in equation (3):
zLL
ij ij

2007 2008 GDP ran GDP effect k effect
26,480 1 24,595
9,258 2 9,596 2,255 4 2,119 3,622 3 3,439 1,700 6 1,615 142 15 135 175 13 165 134 17 137 515 7 527 148 14 142 328 9 307 89 20
311 10 299 1,814 5 1,809 277 11 269 104 19 100
490 8 479
140 16 138 259 12 254 111 18 108

2009 2010

2005 2006
GDP ran

ran k
1
2 4 3 6 17 13 16 7 14 9
10 5 11 19
8
15 12 18

GDP rank

effect k
18,001 1
5,440 2 1,528 4 2,522 3 1,125 6 95 15 117 13 81 17 310 8 98 14 221 9 60 20 204 10 1,133 5 180 11 68 19
314 7
89 16 165 12 71 18

32,831

GDP ran
effect k effect

 L zLK   XL  XK

49,908

85 20 47,829

31,795 1
10,769 2 2,560 4 4,401 3 2,028 6 169 15 209 13 157 17 603 7 177 14 393 9 106 20 370 10 2,138 5 329 11 123 19
579 8
165 16 306 12 130 18

59,354

24,784 1
7,520 2 1,976 4 3,275 3 1,548 6 129 15 160 13 112 17 425 8 133 14 304 9 82 20 279 10 1,553 5 245 11 92 19
429 7
121 16 226 12 97 18

44,871

explain the new equilibrium outputs when the final demand changes:

(7) The input output table can make differences between domestic

X =(I−A)−1F

ALLLALK j j A=MOM=MOM

goods (

Zd

) and the imported goods (

Zm

), which can be expressed

 KL KK 
A L A  zKL 

 zKK 

as Z=ZD+ZM, F=FD+FM and X=ADX+FD. The Leontief’s
D
Matrix in domestic goods(I − A ) can be solved as:
(I−AD)X =FD (8) X and the equation that analyze the changes of final demand

ij
   XL  j

ij
L  XK

For similarity consideration, this study defines the total output matrix X and the final demand matrix F as the balance equation between supply and demand, which is written as:
X = AX + F (4) (I − A)X = F (5)
where (I − A) is a Leontief matrix and X can be solved when the matrix is non-singular:

can also be solved:

X = (I − A)−1 F

In equation (6), (I − A)−1 is a direct and indirect requirements matrix, also called as the inter-industry interdependence coefficients matrix or Leontief inverse matrix. Equation (6) can

Likewise, the total output effect under differ situations such as when the final demand changes, the domestic goods output effect and the value-added effect can be calculated in equation (12) to equation (14).

j (3)

(6)

V =v(I−AD)−1FD

(11)

Journal of Coastal Research, Special Issue No. 96, 2019

X =(I−AD)−1FD X =(I−AD)−1FD

(9) (10)

The output effect on Equation (10) can be transformed to a value-added effect:

4 Lin and Chou _________________________________________________________________________________________________

X =(I−A*)−1X X =(I−AD*)−1X V =v(I−AD*)−1X
ANALYSIS

(12) (13) (14)

added caused by natural disasters mainly occur in agriculture, forestry, fishery, wholesale trade and retail trade, animal feed and chemical fertilizers industries. These sectors account for 90% (Mainland China) and 87% (Taiwan) of total losses respectively. Second, the losses in agricultural products caused by natural disasters may result in losses in other industries through industrial linkage. The empirical results show that the agricultural loss in Mainland China (2.6) is smaller than the loss in Taiwan (2.9).

LITERATURE CITED
Ahmed, A.; Masud, M.; Al-Amin, A.; Yahaya, SRB.; Rahman, M., and Akhtar, R., 2015. Exploring factors influencing farmers’ willingness to pay (WTP) for a planned adaptation programme to address climatic issues in agricultural sectors. Environmental Science of Pollution Research, 22, 9494– 9504.

Albert, E. and Steenge, M., 2007. Thinking about imbalances in post-catastrophe economies: An input-output based proposition. Economic Systems Research, 19, 205-223.

Furen, J.; Hirokazu, T.; Yasuhisa, K., and Tomonori, M., 2005.

Economic loss estimation of water supply shortage based on questionnaire survey in industrial sectors. Japan: Disaster Prevention Research Group of National Research Institute for Earth Science and Disaster Prevention (NIED), 50p.

Lin, H.C., 2013. The Taiwan’s economic impact of cross-strait agricultural trade: Regional input-output analysis. Taiwanese Agricultural Economic Review, 19(1), 81-127.

Lin, H.C.; Kao, T.; Chou, L., and Chang, C., 2010. The economic impact from agricultural products loss caused by natural disasters and regional input-output analysis in Taiwan. Proceedings of the 1st Congress of East Asian Association of Environmental and Natural Resource Economics (Hokkaido University, Sapporo, Japan), pp. 10- 75.

McElhinney, J., 2016, Influencing the agricultural sector to embrace adaptation to climate change, for the sake of global food security. Environmental Science of Pollution Research, 23, 9245–9246.

Miller, R.E. and Blair, P., 2009. Input–Output Analysis: Foundations and Extensions. Cambridge: Cambridge University Press, 259p.

Okuyama, Y., 2007. Economic modeling for disaster impact analysis past present and future. Economic Systems Research, 19, 115-124.

Wu, Y.W., 2003. After the 921 earthquake reconstruction tourist attractions in central Taiwan's economic impact analysis. China: National Taiwan University, Master's thesis, 160p.

A supply-side Inter-Country Input-Output (ICIO) analysis was conducted to explore the impacts of natural disasters on the economic systems of Taiwan and Mainland China. The losses in value-added effects were estimated. Table 1 presents the value- added losses caused by natural disasters in Taiwan during 2005- 2010. The effects range from 9.9 billion to 59.1 billion NT. The largest reduction was the damage to agricultural products, accounting for 80.7% of the total losses. The second largest industry suffering from natural disasters is wholesale trade and retail trade industry, followed by chemical fertilizers.

As an industry with second largest losses caused by natural disasters, the wholesale and retail industry undertakes 0.07 units of the agricultural production. Given that the forward linkage is stronger in wholesale and retail, the economic effect on this industry accounts for 0.5% on average.

In addition, 0.10 units of agricultural production input were offered to chemical fertilizers industry, making it the third largest industry suffering from natural disasters. As the chemical fertilizers industry has a larger backward linkage with agriculture, the chemical fertilizers industry bears the production losses and the losses in value-added effects.

Table 2 presents the value-added losses by natural disasters in Mainland China from 2005 to 2010. The effects range from 30,047 to 59,354 billion RMB. Similarly, agriculture industry was the mostly affected, with a loss of 55.2% in 2010, followed by wholesale trade and retail trade, and animal feeds.

The input coefficient of agriculture is 1.14, which means the losses will bring damage through forward linkage. The losses of productions and the value-added in agriculture account for more than 65% in both cases.

To summarize, both in Taiwan and Mainland China, agriculture, wholesale trade and retail trade, animal feeds, and chemical fertilizers industry are the industries mostly affected by natural disasters. Comparing the multipliers in Taiwan (2.9) and the multiplier in Mainland China (2.6), it can be found that the losses in Taiwan is larger than Mainland China in the face of the same natural disaster.

CONCLUSION
This study used supply-side Inter-Country Input Output analysis to explore the impacts of natural disasters on the economic systems of Taiwan and Mainland China. Two main results emerge from the present study. First, the losses in value-