Cost of Floods on Pakistan’s Economy

Table of Contents

Acknowledgements

ABSTRACT

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Appendix

REFERENCES

Acknowledgements

This thesis could not have been possible without the Will and Grace of Allah Subhanatala. I take this opportunity to express my gratitude towards Dr Hafiz A. Pasha, Dean of School of Social Sciences, my supervisor, under whose guidance and supervision I was able to produce this paper. It was a pleasure and an amazing experience to be taught by him. He has been an encouraging and inspirational teacher. There aren’t many people out there who could make a person stand up and want to achieve more in life than there is being offered. I like to also thank the MS Coordinator Muhammad Imran. His help was instrumental and mathematic skills well appreciated. I would especially thank my parents for always standing by me and tolerating my constant mood swings due to lack of sleep and long working hours. Thank you for making it possible for me to be able to produce this thesis.

List of Tables

Table 1.1 Natural Disasters Comparison

Table 1.2 PROVINCE-WISE DAMAGE ASSESSMENT

Table 1.3 Sector-wise Damage Assessment

Table 3.1 Damage Assessment-NDMA

Table 3.2: List of Variables in the Model

Table 4.1: Sectoral Growth Rates of GDP in Different Scenarios for 2010-11 and 2011-

Table 4. 2: GDP GROWTH RATES

List of Figures

Figure 3.1: Economic Impact of the Flood Shock

Figure 4.1: Comparison of Growth Rates with Floods and without Floods

ABSTRACT

The thesis reviews the available literature on measuring the direct and indirect impacts of natural disasters, both in the short run and in the long run. Based on this review, an attempt is made to quantify the economic impact of the recent floods in Pakistan. This will include not only the estimates emerging from the Damage Needs Assessment by the World Bank and ADB but also derivation of the second round multiplier effects on the GDP by use of the IPP Macroeconomic Model. In addition, an assessment will be made of the prospects for revival of growth.

Keywords: Cost of Floods, Economic impact, Short run and Long run effects, Revival of growth,

CHAPTER ONE

Chapter 1

Introduction

Pakistan is one of the most arid countries in the world with an average annual rainfall of less than 240 mm. It experiences heavy rainfall every year during the monsoon season from July to September. Every year in the summer season the melted ice from the glaciers caused the rivers to flood. However this year the unexpected substantial rainfall during the monsoon period resulted in heavy floods in the absence of adequate infrastructure (dams, barrages, reservoirs). The catastrophe started from the third week of July starting with a heavy rainfall first in the province of Balochistan; then with the second spell of rainfall hitting the province of KPK and trickling down to Punjab and ultimately hitting Sindh. With soaring prices, high level of poverty and the contractionary policy of the IMF program already burdening Pakistan this disaster had come at a time when it was least expected and least desired. The total estimated cost of damages and loss has been estimated to be US$10.06 billion; and the cost of reconstruction ranges from $6.8 billion to $8.9 billion, depending on the level of infrastructure desired.

What does the paper aim to Prove

Though there have been several different researches done on natural disasters around the world focusing on direct cost, this paper will also estimate the indirect effects on the other sectors of the economy; other than agriculture, using a simple structural simultaneous equation model. It is a supply side model and will be useful in defining the cost of different sectors of the economy and to the GDP.

The study consists of five chapters that are set out as follows. The first chapter introduces the topic of the thesis by summarizing how the floods proceeded and affected different regions. In addition it provides facts and figures of the damages and losses incurred in the floods. Chapter two reviews literature on the impact of natural disasters around the world. It summarizes different approaches to find out the economic cost of each disaster and why some countries are affected more than the others. Chapter 3 gives the specification of the model used to derive the effects on different sectors of the economy. Chapter 4 gives results of the simulations of the model. Chapter 5 summarizes the findings and the policy recommendations.

1.1 History of Natural Disasters in Pakistan

1935 Quetta Earthquake

In 1935, the city of Quetta in the province of Baluchistan saw the worst disaster ever seen in the history of Pakistan. 7.7 on the Richter scale the earthquake virtually leveled the city of Quetta. About 60,000 people were killed in one of the deadliest earthquakes of hit South Asia.

1945 Baluchistan earthquake

In 1945 Baluchistan once again faced a devastating earthquake. The earthquake measured 7.8 on the Richter scale with the epicenter about 98 km southwest of the town of Pasni, Balochistan. It not only damaged the property but also led to a 40 foot tsunami causing the death of over 4000 people.

1950 floods

The monsoon rains in 1950 caused a massive flood in Pakistan killing around 2900 people across the country. Even Lahore was not spared from the destruction when the River Ravi flooded and over 100,000 homes were destroyed; leaving around 900,000 people homeless.

1970 East Pakistan

In 1970 East Pakistan saw one of the deadliest tropical cyclone ever recorded. The Bholo Cyclone hit the territory causing up to 500,000 losses of lives primarily as a result of the storm surge that flooded the low-lying islands of Ganges Delta. This disaster is rated to be the worst natural disasters in modern times.

1974 Hunza Earthquake

6.2 on the Richter scale, this earthquake hit Kohistan and surrounding areas including parts of Swat, Hunza and Kashmir in Northern Punjab. About 5300 people were killed; 17000 injured and 97,000 affected.

2000 Drought

The 2000 drought in Baluchistan affected at least 1.2 million people. Over a 100 people died; mostly due to dehydration; according to the official figures. The livestock was greatly affected resulting in large losses. The drought lasted ten months.

2005 Kashmir Quake

In 2005, Kashmir and its surrounding areas were struck by a destructive earthquake measuring 7.6 on the Richter scale. It caused a great deal of damage, resulting in a loss of at least 73,000 live according to the official reports; more than 3.3 million were made homeless. The reconstruction of the infrastructure continues till today.

2007 Cyclone Yemyin

The Cyclone Yemyin hit the coastal areas of Pakistan in early July 2007. At least 730 people died, some 350,000 were displaced and more than 2 million livestock perished.

2010 Hunza Lake Disaster

A landslide in January 2010 in Attabad village in the northern areas of the country; killing 20 people; led to around 40 houses to slide in to the River Hunza. The debris from the landslide caused the river to dam, leading to a formation of a large lake. The lake threatened to flood downstream areas forcing some 20,000 people to leave their homes by June.

1.1 Natural Disasters Comparison

illustration not visible in this excerpt

Table 1.1

Comparison of Pakistan’s Floods with Major Global Natural Disasters

Source: Natural Disaster Management Authority (NDMA), 2nd October, 2010

The floods of 2010 were one of the worst natural disasters ever seen in history. Comparing the floods to the other natural disasters, it can be seen that this disaster caused a record damage to Pakistan. The death rate and the number of people injured may not be high when compared to other natural disasters in the table; but the geographical space and population affected are the largest in these recent floods. The area affected being mostly agricultural land and cropped area; played a devastating role. It not only cut short the food supply, it also affected both the service and manufacturing sectors and took away the source of livelihood of thousands of people. The population affected by the floods is so huge in both scale and destruction that it is more than all the above natural disasters put together.

1.3 How it all Began

Beginning of the monsoon season is the most looked forward time as it brings the much needed rain to the arid region of Pakistan. But in 2010 we saw the highest recorded rainfall in a decade; submerging vast areas and causing people to evacuate these areas. The heavy rainfall began and continued for days in the regions of Balochistan followed closely by a second spell of heavy monsoon rains over Khyber Pakhtunkhwa, which commenced in the last week of July 2010 and persisted up till first few days of August. Khyber-Pakhtunkhwa (KPK); flooding the cities of Peshawar, Nowshera, Swat, Charsada, etc. and cutting off these areas from the rest of the province and closing all routes towards the flood affected areas.

These rains generated unmatched flood flows in the major as well as secondary and tertiary rivers, including the nullahs in KPK, Punjab and then Sindh. The local rivers and nullahs in Baluchistan also saw extraordinary floods. The river Indus, at some of the control points in Punjab and Sindh, along with River Swat, Panjkora and Kabul experienced historic flood flows.

The heavy rainfall caused the river banks to burst, flooding the low regions and those nearer to the banks. Many houses, schools, roads were either severely damaged or destroyed leaving the people without food, shelter and medical attention. With half the province of KPK under water barely surviving the blow from the natural disaster, the rainfall then hit Punjab and affected the Indus River Basin.

The flow of the water from the heavy rainfall and the rivers pushed its way from KPK to Punjab as it made its way downstream to southern Punjab and Sindh. Though the Metrological (MET) Department and Natural Disaster Management Authority (NDMA) of Pakistan had forewarned the people to spare them from the demolition; the floods still managed to cause major damages to the people, land, infrastructure, etc.

Houses, schools, hospitals, electricity poles, transformers, cropped areas, standing crops, cattle, sheeps, animals; nothing was spared by the wrath of the flood water. Though millions of people lost their homes, shelter, and their income earning assets; the death toll wasn’t as high as compared to the other natural disasters seen in Pakistan.

People moved to safer areas in the hope of being spared from the destruction. But even reaching the safe areas the people were affected in one way or the other. The authorities were a slow in relief efforts for the flood victims. There were food and water shortages, and shelter wasn’t available to most of the victims. The people were in need of proper medical facilities. The army played a major role in trying to help the people reach safe grounds and support those who managed to get there on their own.

The private sector, civilians and a number of both national and international organizations have all put in every effort to generate funds for these people. Food items, clothing, shelter, etc. were collected in large quantity and taken to the flood affected areas. Till present day, relief efforts are still taking place. The government along with international agencies like the UN, World Bank and Asian Development Bank are working towards restoration of the affected areas. Estimate of the damage and costs of reconstruction have already been made and mentioned later in the study.

1.4 Damage Assessment Analysis

A total of 78 districts were hit, 2,092,600 hectares of cropped area was destroyed, 20,184,550 population was affected, 1,985 people died, 2,926 people were injured and 1,744,471 houses were damaged; according to the official report of the Natural Disaster Management Authority (NDMA) on 14th November.

The figures of the damage and reconstruction costs released on the 14th of November by the World Bank and ADB at the inauguration of the Pakistan Development Forum are listed in the table below. Most of the figures coincide with those presented by the officials in Pakistan with a few discrepancies here and there.

illustration not visible in this excerpt

Table 1.2 PROVINCE-WISE DAMAGE ASSESSMENT

Source: Pakistan Development Forum, Express Tribune, 15th November

Table 1.2 and 1.3 show the cost of the losses incurred due to floods and the reconstruction costs on both the provincial and sectoral level. The total cost of the losses and damage is taken out to be around $10,056 million and the total cost of reconstruction ranges from $6799 million to $8900 million depending upon the level of reconstruction. If the same infrastructure is built then the cost incurred will be $6800 million. If a better infrastructure is constructed then the cost will be $7418 million. For even higher standards of infrastructure the cost incurred will be $8915 million.

illustration not visible in this excerpt

Table 1.3 SECTOR-WISE DAMAGE ASSESSMENT

Source: Pakistan development Forum

CHAPTER TWO

Chapter 2

Literature Review

Hallegatte S. and Ghil M. (2008) with an endogenous business cycle model study the macroeconomic response towards natural disaster in which a recurring behavior takes place due to the flux of the investment-profit. The model concludes a greater response towards natural disasters during expansions than during recession because the exogenous shock amplifies the pre-existing disequilibria. In recessions, however, the existence of unused resources triggers the damping of stock. Higher output variability is also seen in response to the stochastic productivity shocks during expansions as compared to in recessions.

Long F. (1978) assembles the data available on consequences of natural disaster on Third World agriculture. These consequences provide an explanation of the lack of self capability of the low income countries; it also explains the occurrence of hunger and increasing poverty. He prefers a systemized data collection that can be used to study the effects of natural disasters. A proper mechanism can be planned through that data to protect agriculture in disaster prone countries. With a planning system the adverse effects of natural disasters can be mitigated.

Toya H. and Skidmore M. (2005) analyze the data of natural disasters impact over a period of time to study the extent to which the human and economic losses from natural disasters are reduced as economies develop. They undertake two sets of regressions for determining the relationship between the level of development and disaster impacts. One set of regressions analyze the impact of natural disasters on income whereas the second set of regressions focuses on economic damages. They establish that countries with higher income, higher educational attainment, and greater openness, more complete financial systems and smaller governments experience fewer losses.

Rodriguez-Oreggia E., Fuente A. and Torre R. (2008) examine the effects of natural disaster on social indicators such as HDI and different poverty levels at the municipal level in Mexico. To find the effects the analysis controls for a set of geographical and natural characteristics of location. Also control is set for precondition institutional, economic and demographic characteristics for heterogeneity. The paper uses an adjusted difference-in-difference regression on data for 2000 and 2005. The results demonstrate a significant fall in social indicators from natural disaster and especially floods and droughts, including HDI and the level of poverty. There is a major impact from natural disasters on reducing the HDI and also increasing poverty.

Bergholt D. and Lujala P. (2010) relied on econometric methods for the investigation of economic consequences of natural disasters and their linkage to armed civil conflict for the period 1980-2007; panel data and OLS regressions are used as the methodology. Geographical and hydro-meteorological disasters have a negative impact on income growth. But there is no significant evidence that supports the argument that negative effects of disasters increases the likelihood of armed civil conflicts.

Popp A. (2006) explores the relationship between natural disasters and long term economic growth. Several key macro-economic variables, most notably technology that can increase or decrease economic growth is affected by natural disasters. After the disaster recovery is very important and the institutions of the country determine how the economy makes progress.

Rasunussen N.T. (2004) provides an assessment of cross country natural disasters occurrence. The countries are compared along four dimensions: the number of events divided by land area; the number of events divided by population; the number of affected persons divided by total population and damages divided by GDP. The analysis demonstrates that the most disaster prone areas in the world are the small island states especially countries of Eastern Caribbean. Natural disasters are found to have a discernable macro-economic impact, including large affects on fiscal and external balances, pointing to an important role of stabilization measures.

Pelling M., Ozerdem A. and Barakot S. (2002) throw light on the impact of natural disasters on developing nations. The analysis also shows the types of losses the country faces and the effect on long run growth. In the paper three types of losses are mentioned; direct losses, indirect losses and secondary losses. The respective results depict that economic growth declines and the debt burden increases as a result of disasters. The post disaster period in the short run offers opportunities for acquiring foreign capital, remittances, foreign aid, and reinsurance payment; but it is only a short term solution and is insufficient to compensate for all the losses especially those due to systemic and secondary effects.

Vakis R. (2006) asserts that natural disasters have a huge impact on social and economic welfare. Policies to manage them need to be integrated and well grounded to the specifications of natural hazards as well as local capacities in terms of fiscal, administrative and economic capabilities. A well designed natural disaster management strategy crucially depends on carefully assessing and planning responses before, during and after the disaster occurs. This policy note discusses the complementary role that Social Protection can play in the formation of an effective strategy for natural disasters management.

Benson C. and Clay J. E. (2004) examine the short-term and long term economic and financial impacts of natural disasters, relying mostly on in-depth case studies of overall sensitivity to natural hazards in the small island economy of Dominica; public finance consequences of disasters in Bangladesh; and the economic consequences of climate variability in Malawi and Southern Africa. Policy implications are drawn, and, where appropriate, recommendations are made. Finally, directions for future research and cooperation are outlined.

CHAPTER THREE

Chapter 3

THEORETICAL FRAMEWORK AND EMPIRICAL RESULTS

Given that the current floods in Pakistan have had an impact on all the sectors of the economy. Handling these problems would require a model that could simultaneously capture the major relations among different sectors of the economy and thereby trace through the indirect and secondary effects on the economy.

The purpose of this chapter is to develop a formal economic framework. The model will be used to quantify the cost of floods to the Pakistan’s economy in terms of lost output. The model is particularly developed keeping in view the present economic scenario of the country and will be used to forecast short to medium run economic outlook of the country in the aftermath of the floods.

3.1 Structure of the Model

The model contains 4 equations: 2 behavioral equations and 2 identities. Including the exogenous variables, there are a total of 10 variables in the model database. These are enough to give a richer account of interaction between the three major sectors of the economy; the agriculture sector, manufacturing sector, and the services sector. Thus, through the model, what we are trying to see the impact of the flood on the GDP of the Pakistan.

3.2 Direct Impacts

Agriculture sector was most directly hit by the floods, which only damaged the standing crops but also damaged the agricultural infrastructure like irrigation channels and watercourses, farm to market roads, and the power distribution network in the rural areas. The water also drowned many livestock along with fodder storage facilities which resulted in shortages. This happened in all the four provinces. It is estimated that almost 2.245 million hectares of cropped area was affected. The livestock lost is around 1.4 million with 14 million at risk due to fodder shortages.

According to the PDMA of Punjab, 461 irrigation channels were damaged and 2899 kilometer of roads was affected badly by the floods. A summary of the damages has been given in the table 3.1.

illustration not visible in this excerpt

Table 3.1 DAMAGE ASSESSMENT

Source: NDMA (As of 2nd Oct, 2010)

Therefore, the most direct impact was on agriculture both crops and livestock. The medium term and long term affects are the indirect affects due to the damage of infrastructure which will limit the production capacity of the agricultural economy. Till the time the infrastructure is rebuilt, production losses will continue. According to the Provincial Governments the recovery could take two to three years.

Damage to shelter also affects the GDP. This is due to the loss in the rental values of property damaged by the floods. There are no precise calculations made by the National Income Accounts of Pakistan to capture the yearly fluctuations in the Ownership of Dwelling sector. Therefore, it is unlikely that this loss will be allowed for by FBS when the GDP for 2010-11 is estimated. It is also possible that rural small-scale and cottage industry has been affected by the floods here also the FBS is unable to accurately measure the value added of the in small-scale manufacturing, on an annual basis.

3.3 Indirect Impacts

We turn to the linkages of the agricultural sector with the rest of the economy. Understanding these relationships is necessary when the indirect effects of the floods on the economy are to be quantified. These linkages are identified below.

3.3.1 Agriculture to Manufacturing

Around 40% of the manufacturing sector of Pakistan is agro-based industry, both in terms of processing agriculture inputs and in providing inputs like fertilizer for agricultural production. Textiles, sugar, leather, and cigarettes are the major industries of Pakistan which rely on raw materials like cotton, sugarcane, hides and skins and tobacco. Production of all these inputs has been severely affected by the floods. Already, cotton (phutti) arrivals have slowed down and stand short by about 15 percent in relation to the level last year. Though a large quantity of cotton is going to be imported, the production of yarn and value added textiles will be adversely affected. The sugarcane crop is likely to have faced a damage of 15 to 20 percent. The sugar factories will face high prices for what is left of the sugarcane and achieve production levels which are significantly lower than last year.

To top it off, as the planting season starts for the Rabi crops, the demand for fertilizers is likely to be reduced if the land for cultivation is still under water or if the fertilizer and seeds packages being provided by the provincial governments do not reach the farmers. There may also be a delay in the return of the displaced population due to the slow progress of rehabilitation work in some areas.

The industries which have been impacted by the loss in agricultural production have a total weight of 50.63% in the Index of Industrial Production. Already the month of August 2010, there was a decline of 3 percent in the large scale manufacturing sector in relation to the level in the corresponding month of the previous year. Most of the agro-based industries have shown a negative growth and this will become more pronounced in the coming months.

A relatively less obvious impact is likely to be due to the lack of growth in the rural purchasing power due to lower marketed surplus of agricultural produce. With the phenomenal increase in wheat procurement price by the government in the year 2008-09, for example, rural incomes increased rapidly. This resulted in an increase in the demand of durable goods like tractors, cars, motorcycles, TVs, fans, freezers, etc. but in the year 2010-11 this is not likely to happen given the loss of production due to floods. Growth of industries catering to the rural market will probably be relatively low this year.

3.3.2 Agriculture to Services:

Agriculture is the primary sector and services the tertiary sector of the economy. There is, like with the manufacturing sector, a two way relationship between these sectors. The loss in the agriculture production due to the floods will have an adverse affect on a number of services as listed below.

- Wholesale and retail trade

According to the estimates of FBS trading margins on agricultural production in the wholesale and retail trade sector makes up about 75% of the value added. This includes commodities like milk, wheat, rice, cotton, vegetables, fruits and various manufactured goods. The growth in this sector, especially with the shortage of cotton, is probably going to be negatively affected in 2010-11 due to the floods.

- Transport and Communication

Direct affect of the floods on the infrastructure has already been explained. Other than that, the fall in the production levels of both the agriculture and manufacturing goods will imply less demand for transportation. Road traffic volumes, which contribute around 70 percent of the value added of the sector will be badly affected.

- Banking and insurance

Outstanding loans of farmers by the commercial banks and the specialized agricultural development bank are approximately Rs 200 billion. Some of these loans are at risk of turning into a bad debt because many farmers are not be in a position to pay back their loans after losing their assets and incomes due to the floods. The resulting decline in the profitability will reduce the value addition in the banking and insurance sector in 2010-11. The direct and indirect sectoral impacts of the floods are gives in figure 1.

illustration not visible in this excerpt

Figure 3.1: Economic Impact of the Flood Shock

3.4 Data Sources and Description

The data taken for this research is collected from the two standard sources of data in Pakistan. These include various issues of Economic Survey of Pakistan and Annual Report of the State Bank of Pakistan. The time period of the data is from 1980-81 to 2009-10. Note that all the variables taken are the growth rates and except for the real interest rate. In the table below each of the variable is described in detail.

illustration not visible in this excerpt

Table 3.2: List of Variables in the Model

3.4.1 Agricultural Sector

In the model, the agricultural sector is taken exogenously by the identity.

The growth in agricultural sector is taken as an exogenous variable because its growth rate is dependent upon a number of factors which cannot be predicted with certainty and that cannot be controlled. These factors make the growth rate of the agricultural sector uncertain too. The factors include weather conditions in the country, floods, and other natural calamities.

3.4.2 Manufacturing Sector

The growth rate in the manufacturing sector is behavioral determine by the equation:

GM =

As it can be seen from the equation, the growth rate in manufacturing sector, GM, is dependent upon the growth rate in the agricultural sector, GA, the overall growth rate of the economy, GY, the real rate of interest rate, R, and the growth rate in the exports of goods and services, GX. The agricultural sector is included as an independent variable because it provides raw materials to almost 40% of main agro-based industries in the manufacturing sector. Thus, the growth rate in the agriculture sector directly affects the growth rate in the manufacturing sector.

Other three variables in the equation measure the response on the demand. That is, if the GDP grows at a faster rate, it will increase the income of the people which will in turn increase the demand. More demand means, there will be more production taking place in the economy. Thus, when the GY is increases/decreases the growth in the manufacturing sector also increases/decreases.

Similarly, the faster the increase in the growth rate of exports, GX, the higher the growth rate of the manufacturing sector will be. Growth in export means that there is a high demand for the domestic goods in the international market, so there will an increase in the growth rate of the manufacturing sector. By trading externally the producers get an entrance in the foreign market, explore new avenues and sell their product at a higher price than in the domestic market.

The last variable in the equation is the real interest rate, R. It is a policy variable and has a negative impact on the manufacturing sector, from the investment side. As the rate of interest increases it makes costly for the producers to invest in expanding production capacity and ultimately reduces the growth of the manufacturing sector.

3.4.3 Services Sector

The services sector is the largest sector of the economy. It contributes around 55 percent to the GDP and is related directly and indirectly to both the manufacturing sector and the agricultural sector. In the model, it is behaviorally determined by the equation:

While

The growth rate in the services sector is dependent upon the sum of the growth rate in the agriculture sector and the growth rate in the manufacturing sector [ GA + GM ] and growth rate of the home remittances (GHR). When there is growth in the agriculture and manufacturing sector, there will be a need for certain services like transport, storage and communication; wholesale and retail trade; finance and insurance. Similarly, the growth in the GDP and home remittances will lead to an increase in the above mentioned services. The increase in the home remittances will lead to an increase in the banking and insurance sectors. Other variables which could impact on the growth of the services include the growth of the public expenditures, imports, etc.

3.4.4 Gross Domestic Product

The gross domestic product (GDP) is the sum of the value addition of all the sectors of the economy; it is, therefore, given by the following identity:

where,

It can be seen from the above identity, that the growth rate of the gross domestic product, GY, is the weighted sum of growth rates of the three main sector of the economy; agriculture, manufacturing, and services sector. The weights are measured as their percentage contribution to the overall economy of the Pakistan.

3.5 Linkages in the Model

The equations in the model are interlinked. The growth in the agriculture sector has an impact on the growth of the manufacturing and services sectors. Manufacturing sector in turn has an impact on the services sector. Similarly, the growth in the GDP has an impact especially on demand in the manufacturing sector.

[...]