Auxiliary Technical Report (Only Available Online)

Auxiliary technical report (only available online)

Auxiliary technical report FOR

“Valuing Flood Risk Reductions”

A.  Reynaud and M-H Nguyen

(Only available online)

Note for the reader

We present here an English translation of the two main sections of the questionnaire used for valuing a flood risk reduction in the choice experiment (the original Vietnamese survey is available from authors upon request). Section C is the risk and probability section. Section D provides the script used in the choice experiment with an example of choice sets.

Design and implementation of the choice experiment

The design of the questionnaire followed a highly iterative process. During the design of the survey (June 2011 - December 2011), experienced stated choice researchers, natural resource economists and water management experts (both Vietnamese and French) reviewed different versions of the questionnaire. Various meetings involving research team members, water experts and local representatives were organized in Vietnam (Nghe An Province). An initial meeting was held in October 2011 with the representative of the Vietnamese Ministry of Agriculture and Rural Development (MARD) in charge of flood protection in the Nghe An Province. The questionnaire was significantly modified after this meeting. A second version was presented in November 2011 to a household representative in the city of Vinh (Cua Lo district), to a farmer's representative in the Cua Lo district and to a politician in the Quy Hop district. These meetings turned out to be useful in checking the respondents' understanding of the survey and resulted in several adjustments to the formulation of explanations and questions. The questionnaire was also significantly reshaped after those meetings. After incorporating these comments the pilot version of the questionnaire was tested in December 2011 using face-to-face interviews. Six trained and carefully supervised interviewers interviewed 30 randomly selected households. A preliminary econometric model was then estimated using the pilot data and some modifications were made to the questionnaire. In particular the number of choice sets was increased from six to eight since it appeared that such a number was not incompatible with cognitive capacities of respondents.

The final survey took place from April 4th to June 10th 2012, a period of time during which no floods and no natural disasters were recorded in the Nghe An Province.

To minimize cultural biases, all interviews have been conducted in Vietnamese by faculty members and/or students from the Vanxuan University of Technology at Cua Lo (Nghe An Province). At the beginning of each household interview, it what mentioned that our survey had been officially approved by the head of the village People's Committee. As a result, less than 10% of households contacted refused to participate to the survey. We interviewed (face-to-face interviews) 448 households observed in 28 villages/communes from 14 districts in the Nghe An province.

SECTION c:

RISK AND PROBABILITY TRAINING

[For interviewer only] This section aims at training households to manipulate mortality rates and probabilities that will be used in the next sections.

[Interviewer] In sections E and F of this questionnaire, you will have to manipulate some complex concepts of mortality rate and probability of occurrence. In this section, I am going to give you a brief presentation of those concepts. First let me show you the different sources of mortality in Vietnam in 2008.

In 2008, 576635 people have died in Vietnam. Approximately 40% of those deaths can be imputed to cardiovascular diseases. The second source of mortality is malignant neoplasms (cancers) representing 14% of deaths in Vietnam.

People dying due to floods are included in the unintentional injuries category which represents 8% of deaths (in orange color).

On the right side, we have disaggregated the unintentional injuries category into subcategories. The most important subcategory is road traffic accidents representing 3.66% of total deaths (21,000 persons). As you see, floods represent less of 0.08% of deaths in Vietnam (474 persons in 2008).

[Interviewer] Let me talk now about mortality rates in Vietnam. The mortality rate is a measure of the number of deaths (in general, or due to a specific cause) in some population, scaled to the size of that population, per unit time. Mortality rate is typically expressed in units of deaths per 10,000 individuals per year. In the following picture I have represented the annual mortality rate for Vietnam in 2008 which is 67 per 10000 persons. There are 10000 squares representing 10000 persons. The 67 Red squares represent the 67 deaths. The surface inside the red circle represents the 67 deaths.

Annual mortality rate for Vietnam in 2008, all causes (67/10 000)

Imagine now that the Vietnam by improving for instance the healthcare system is allows to divide almost by two the mortality rate to reach 30 / 10000. On the following picture, I represent this reduced mortality rate.

Reduced annual mortality rate (30/10 000)

As you may see, the number of red squares representing deaths has been almost divided by 2. The surface inside the red circle has also been divided by 2.

Let me now show you the mortality rate for the category unintentional injuries. In 2008, 45506 persons died because of an unintentional injury. The associated mortality rate is 5 per 10000 persons.

Annual mortality rate for Vietnam in 2008, unintentional injuries (5/10 000)

Let me now show you the mortality rate for the subcategory floods. 474 persons died in Vietnam because of a flood in 2008. The associated mortality rate is 0.05 per 10000 persons. This this number is very small we will express it in deaths per million of persons: 5 per million persons.

Annual mortality rate for Vietnam in 2008, floods (0.05/10 000)

/
.

Of course, the mortality rate for floods is an average for Vietnam. You may live in an area prone to floods and your own mortality rate due to floods may be higher. On contrary, you may live in an area with a limited impact of floods. In such a case your own mortality rate due to floods may be lower.

SECTION d:

choice experiment for flood policies

[Interviewer] We would now like to know what you think about possible flood management strategies in your place. Let me first start by giving you some general information about floods in Vietnam and in particular in the Nghean Province.

Natural disasters (in particular floods) in Vietnam are a direct impediment to economic development, sustainable development and poverty reduction; and a huge obstacle to the process of striving for the development goals of Vietnam. Vietnam has more than 80% of its population living at risk of direct impacts of natural disasters. In the last 10 years (1997-2006), natural disasters have killed nearly 5,000 people, more than four million hectares of rice fields have been destroyed, and total losses are estimated at 50,000 billion VND.

The Nghean Province is particularly affected by floods. In 2005, floods killed 28 people in the Nghean province and inundated 2,496 houses (98 houses were completely destroyed). The same year 32,000 hectares of rice have been damaged and 1,736 hectares of aquaculture ponds by floods. Total damage due to flood in 2005 in the Nghean Province have been estimated at 372.5 billion (3.6% of the province GDP).

To reduce the cost of floods, public authorities can implement various programs such as building dikes, building reservoirs, modifying land use, improve population awareness about flood risks, etc. The objectives of these policies are usually to reduce the number of death due to floods, to reduce the number of houses destroyed by flood and to limit the agricultural losses due to floods.

In this section, we will try to better understand your preferences for different types of programs that can be used for reducing the risk of floods.

As you will see, each program is described by 5 attributes that I am going now to describe to you. Let me know explain to you what is the exact meaning of each program’s attribute.

Attribute 1: “Chance per year that your house will be flooded and substantially damaged by a flood occurrence”

/ This means the chance per year for the next 10 years that your property (house, equipment, etc.) will be substantially affected by floods as a result of rains exceeding those that would be expected normally.
Be careful:
• We are talking about substantial damage, not small damages that occur each year in case of normal flood conditions
Consider:
• Without any change in flood programs, the average risk for the next 10 years in the Nghean province is 2%.
• Even if you do not live near flood prone areas, floods can affect your, recreation opportunities, family and friends, and income levels in your community.

Attribute 2: “Chance per year that paddy and agricultural land will be flooded and substantially damaged by a flood occurrence”.

/ This means the chance per year for the next 10 years that paddy and agricultural land (including land for fish) will be flooded and substantially affected by a flood as a result of rains exceeding those that would be expected normally.
Be careful:
• We are talking about substantial damage, not small damages that occur each year in case of normal flood conditions
Consider:
• Without any change in flood programs, the average risk for the next 10 years in the Nghean province is 20%.

Attribute 3: “Chance per year that you or one member of your household may die in case of a flood (fatality rate due to floods)”.

/ This means the chance per year for the next 10 years that you or one member of your household may die.
Consider:
• Without any change in flood programs, on average 30 persons will die each year due to floods in the Nghean province. The average annual risk of death is 12 per million.

Attribute 4: “Level at which flood management policies are set and enforced”

/ This means at which political or administrative level should the flood management policy be implemented. Flood management policies include building an operating dikes and reservoirs, education campaigns, management of warning systems, emergency plans, etc.
Consider:
• Various level of institutions are involved in flood policies
• These institutions can be the State, the Province, the District or the Commune.

Attribute 5: “Flood management fee”.

/ How much you will pay (in VND per person with age between 18 and 60) each year during the next 10 years to fund flood policies
Consider:
• Please keep in mind your available income and all the other things you have to spend your money on.
• The current flood management fee paid by farmers is 5,500 VND representing the value for 1 kg of rice. The current flood management fee paid by non-farmers is 11,000 VND representing the value for 2 kg of rice.

Be careful, the flood management fee is a per capita fee, which varies if the household head is a farmer or not. Let me give you an example:

Imagine the flood management fee is:

-  3kg of paddy for a farmer (16,500 VND)

-  6kg of paddy otherwise (33,000 VND)

A farmer with 5 persons between 18 and 60 years old will have to pay each year: 16,500* 5= 82,250 VND for flood control.

A household which is not a farmer with 4 persons between 18 and 60 years old will have to pay each year: 33,000* 4= 132,000 VND.

We would now like to know what you think about possible flood management programs in the Nghean Province. You will be presented with 6 choice sets, each of which contains three alternative flood management programs:

-  Status quo: most likely situation in the Nghean Province for the next 10 years without implementing any additional program.

-  Program A: possible flood management program that could be implemented

-  Program B: possible flood management program that could be implemented

For each choice set, carefully consider the implications of each flood management program by looking at the attribute values listed in the associated table. Each program can have both positive and negative outcomes for your household and will last for the next 10 years.

Please complete ALL choice sets. For each choice set, please select the flood management program that you most prefer. Take all the time you need for answering. What is really important for me is to get your own opinion. When deciding the program you would select, please remember to keep in mind your available income and all other things you have to spend your money on.

We will now start with an example of the choice set you will have to complete.

Instructions for Completing Choice Sets

[Interviewer] This page provides an example choice set to help you understand the choices you will be asked to make on the following 8 pages.

On this page, three different flood management programs will be presented to you: program A, program B and Status Quo. Each of these programs has different advantages and disadvantages associated with them, reflecting different asset protection priorities and alternative levels of emphasis on flood suppression.

Please select the one management strategy that you most prefer.

Example only

/ Status Quo: Best guess of the impact of flood in your area for the next 10 years. The outcomes in this column reflects how each attribute is affected in a typical year. For example, in the status quo, your house is damaged seriously 2 years over 100. There are 20% of chance that paddy and agricultural land are significantly damaged each year and the mortality rate due to flood is 12 per million. This policy is set at the National level and the cost corresponds to the current level of the flood management fee. You will always have the option to select the status quo in each choice set.
Program A: The program A column represents a different program than the Status Quo. As you can see, the outcomes are different from the status quo column. For example, with program A, there are now 10% of chance that paddy and agricultural land are significantly damaged each year. Program A costs you 50% more than the status quo. Attributes of program A will be different in each choice set.
Program B: The Program B column represents another management strategy that is different from both the Status Quo and program A. With program B your house is damaged seriously 2 years over 100. . Program B costs you 1.5 times more than the cost of the status quo. Program B will also be different in each choice set.

1.  [Interviewer] Now, you start the real test. Choice set 1.