1. WIIET Exercise 1: Updating an Index Using Farmer Information

1.1. Topic: Updating an Index Using Farmer Information

1.1.1. Objective

The goal of this exercise is to demonstrate how to adapt an initial drought index based on discussions with farmers.

1.1.2. Introduction

Imagine that you are bringing a preexisting index to a new community. Discussions with the farmers will help you to determine if the index is useful for this new community and if it should be adjusted.

This exercise uses hypothetical indices and interviews for a hypothetical community. However, it is based on rainfall data, as well as modified versions of the indices and farmer discussions that have previously occurred in the R4 project, in Ethiopia.

Weather Index Insurance Educational Tool (WIIET):
For this exercise we will be using the Weather Index Insurance Educational Tool (WIIET), pronounced “wheat”. This tool can be accessed at: http://iri.columbia.edu/wiiet. It is intended to teach the index design process for index-based insurance by allowing the user to create an index by working through a series of modules. In this way, the user is able to walk through the index design process step by step: from the collection of crop and climate information to selecting the index parameters; from tuning those index parameters to eventually evaluating the index”s performance. The end result is to have an index that successfully models and hedges against the climate risk that farmers face in real-world conditions and that offers protection at a price that is not prohibitively expensive.

In the following exercises we will primarily focus on creating and pricing indices for drought index insurance. In the Create Contract module, you can create an index and apply it to a historical or simulated rainfall dataset to see what payouts would have occurred. The Pricing module allows you to calculate an insurance premium that is based on the payouts calculated in the Create Contract module. If you need further clarification at any point during this exercise, please refer to the User Guide, which can be found by clicking on the respective icon at the top right of any WIIET page.

Software note:
We recommend the Firefox browser for these WIIET modules because it has been the most tested. The tool is designed to work with other browsers, however you may encounter some occasional “bugs” in the system. We are currently working to fix these kinks and have included work-arounds in this material. For example, depending on your specific browser, you may need to reload the page or return to the main menu to ensure saved datasets are visible for future activities.

1.1.3. Exercise Overview

Background Information:
Tasks 1 to 4, below, walk you through the following scenario: You are bringing a preexisting drought index to a new community to see if it is appropriate for them. Discussions with the farmers will help you to determine if the index is useful and if it should be adjusted. Using the computer web-tool, WIIET, you follow up on that discussion and develop a new drought index based on their responses, if necessary. After you have a new index, you will return to the farmers and discuss the strengths and weaknesses of your new index and if they want to move forward with the insurance process.

During the following exercises you will be working on creating an accurate drought index for a crop known as Teff. This crop is native to Ethiopia and similar to wheat. Teff, in this area, is planted in the middle of the rainy season, when there has always been enough rain (according to the rainfall data and the farmers’ memories). This crop is primarily vulnerable to lack of rainfall at the end of the season. Thus, the Teff index only focuses on the end of the rainfall season.

This exercise is focused on information provided by the farmers because it is the most important part of the process. Future exercises will use scientific analysis tools to further validate the indices. These include crop water stress calculations, and statistical modeling of rainfall. It is possible to use advanced modeling to validate and improve these indexes. The indices that you will create in these exercises are designed to reflect the information contained in the sophisticated models in a way that can also leverage information from farmers and experts. They are also structured to be as easy to understand as possible.

Initial Index Information:
The information for the preexisting index is below. Task 1 will walk you through how to enter this information into WIIET. This will serve as a starting point for your new index:
  • Contract start date: 11 August
  • Length: 70 days
  • Cap: 25
  • Index Trigger: 82
  • Index Exit: 60
  • Payout Years (Since 1995): 1995, 1997, 2000, 2004, 2008

1.2. Task 1: Bringing a preexisting index to a nearby community

Note: Now we are going to put this existing information into the computer:

  1. Log on to WIIET. Go to: http://iri.columbia.edu/wiiet and enter your assigned user name and password (or create a new account, if one has not been provided)
  2. Go to Create Contract module on the left hand side of the page
  3. Click on Example Satellite Precipitation in the Step 1 window on the left, to select that rainfall dataset
  4. In the Step 2 window it asks, “When would you like the contract to begin?” Click on Contract Start Dekad and set the contract start dekad as: 11-Aug. (If you are using Internet Explorer 8, you may have difficulty with this step. A temporary work around for this is to click on Contract Start Window and then select a Start of Aug 11 and an End of Sept 1. The Contract Start Requirements should be set to 0 mm. We are in the process of fixing this bug.)
  5. Length of Contract Period: select 7 dekads
  6. At the bottom of the Step 2 box there is a matrix referring to phases covered. Click on the remove phase button until you have only Phase 1. Next, make sure each of the circles in that phase are blue by clicking on them.
  7. Under Step 3 you can set the Dekadal Cap to 25. The use of a cap is further described in Module 1.
  8. Set the Contract Failed Start Liability to 100. This is not applicable for the index you are using. For some more complicated indices this is the payout if sowing conditions are not met.
  9. On the bottom most table, set the Trigger to 82.
  10. Set the Exit to 60.
  11. Set the liability for the phase to 100. This is the maximum payout for each individual phase. Assume this means that the maximum payout is 100 dollars. Because we have chosen 100, this means that the payouts reported by the software can be thought of either in terms of dollars or in terms of percentage.
  12. Set the Maximum liability to 100. This is the total amount of money insured (100 percent) across all phases. Because there is only one phase in our simple contract, this simply is the maximum liability of the individual phase. For contracts with multiple phases, this parameter becomes more important.
  13. Now you are ready to Run Simulation: Click on the bottom right side of the page
  14. Once you get results (a Payout Calculation graph and table), click on ‘save parameters’, in blue on the left hand side of the screen. Use the name: “original”, and description “original index”. Then click save. Then click close.

Interpreting the Results: The graph and table you see on your screen are the results of applying this preexisting drought index in the new community. The graph shows what years would have paid out historically and the size of the payout you would have received if you had gotten the rainfall for that year. Because we have set the maximum payout to $100, you can think of this either as dollars or percentage of maximum payout. Because WIIET is designed to work with many different kinds of indices, some of the data presented will not be meaningful for the simple index we are working with for this exercise.

The resulting table has 5 columns: “Harvest Year” is the year that the data reflects, according to the year in which the crop would have been harvested. The “Sowing Dekad” is the number of the dekad, or approximate ten-day period when the contract began. There are 360 dekads in a year. For example, dekad 23 means that the index started in the 23rd dekad of the year, or the 11th of August. For some more complex indices, this date changes from year to year, depending on the rainfall that year. The “Rainfall for Phases”- total adjusted rainfall for the index window after the dekadal cap is applied for each phase. The “Payout for Phases” is the payout that would occur for each phase of the contract. The “Total Payout” - total payout accounting for every phase (in this example there is only one phase, so this is the same as the “Payout for Phase”).

Questions:

  1. What years would have paid out? ______________________________________________________________
  2. How many payouts would there have been? _____________________________________________________________
  3. Are there any years with full payouts? If Yes, how many? ___________________________________________________________________________

1.3. Task 2: Adjusting the drought index based on farmer input

Context: Now, imagine you have just returned from your trip to the new community. Below are the notes you took from your discussion with the new farmers.

  • Dates; start: 21 August
  • Length of desired insurance: 40 days
  • Bad years from farmers: 2000, 2004, 2006, 2007, 2009
  • Very bad years from farmers: 2000, 2004, 2009
  • Crops desired to be insured: Teff

The steps outlined here will help you take their feedback into account:

  1. Compare the information that the farmers in the new community gave you to the results that the WIIET tool generated. Note that the difficult years described by the farmers do not completely agree with the existing contract that you designed.
  2. Next you will try to make a more appropriate index to fit the farmers” needs. The below steps will help you make these adjustments. You will need to adjust the start date and length of insurance to better match the farmer’s concerns:
  • To navigate back so you can adjust the information you previously entered, click on “view parameters” at the top of the screen.
  • Change the contract to reflect the start and length the farmers reported. This means reentering these two numbers, using the new information given above. Remember to translate the length given into dekads (10 day periods).
  • Then, click on “Run Simulation” at the bottom.

Questions:

  1. How many payouts would this community have experienced in the last 15 years if this new index had been used? ___________________________________________
  2. Is this a realistic index? _________________________________________________
  3. What else can be adjusted to create a better index that also addresses the farmers” concerns? _________________________________________________

1.4. Task 3: Making a More Appropriate Index

Context: In this section you will adjust the trigger and exit until you have 5 payouts, one of which is approximately a full payout. This will be a complicated process and you will probably need several tries.

The goal is to try to have as many payout years agree with the farmers as possible. Try to not have too many very small payouts or very large payouts. However, it is impossible to do both of these things perfectly.

Hint: You may always switch between the “view parameters” screen and the “view results” screen by clicking on the gray bars at the top or bottom of the page. However, each time that you enter in new parameters you must hit the “run simulation” button in order to calculate your new index and historical payouts.

Now it is time for you to adjust and play with the exit and trigger values. If you need some additional guidance, the “Hint” at the very end of this Task may be useful. Once you have some results to be proposed, click on ‘save parameters’. Use the name: “draft”, and description “draft index”.

Questions:

  1. What trigger value have you chosen? _______________________________________________________________
  2. What exit value have you chosen? _______________________________________________________________

Hint: Look at the “rain by phase” column of the output to decide what the trigger and exit should be. If the trigger is higher than the rain by phase for a particular year, that year will have a payout. If you want five payouts, you can find a trigger that has five years in which the rain by phase is less than that number. A full payout occurs when the exit is above the rain by phase. To have one year with a full payout, you can set the exit close to the lowest rain by phase value.

1.5. Task 4: Farmer discussion follow-up notes

Context: Now that you have your improved adjusted index (from Task 3), you have returned to the community to present it to the farmers. During this discussion they tell you that they want an index that would have paid more money in years like 2009. It is important to explain to the farmers that this would probably lead to an index that costs much more because all payouts would be much larger. The farmers should also be reminded that the future will be different from the past. The farmers understand your concerns; however, they are confident that the expense is worthwhile if payments in years like 2009 could be increased.

In response, you must work with WIIET again to increase payout sizes, without impacting the number of payouts. Return to adjust your parameters once again. Adjusting one single parameter can fulfill the farmers’ request. Adjust and play with the parameter until you have an index that will satisfy the farmers.

Questions:

  1. What parameter will you adjust to fulfill the farmers’ request? ________________
  2. Will this number be increased or decreased? ______________________________
  3. Will this result in a more expensive index? Why or why not? ___________________ _____________________________________________________________________