Crop Loss Damages

[HenryGeorgist]

How do crop losses work, what do they represent?

[HenryGeorgist]

@Tich1212 @Brennan1994
From the original post we had this link:

go-consequences/crops/crop.go

Lines 121 to 132 in 587076c

	func (c Crop) computeImpactedCase(e hazards.ArrivalandDurationEvent) float64 {
	// Determine crop damage percent based on damage dur curve and event dur
	dmgfactor := c.lossFunction.ComputeDamagePercent(e) / 100
	exposedProductionValue := c.productionFunction.GetExposedValue(e)
	totalProductionCost := c.productionFunction.productionCostLessHarvest
	percentProductionValue := exposedProductionValue / totalProductionCost
	totalMarketValue := c.GetTotalMarketValue()
	totalMarketValueLessHarvestCost := totalMarketValue - c.productionFunction.harvestCost
	loss := dmgfactor * percentProductionValue * totalMarketValueLessHarvestCost
	fmt.Println("loss = ", loss)
	return loss
	}

[HenryGeorgist]

I think @Tich1212 pointed out the important part, that how value is characterized is critical to the discussion.

We are using the rate of inputs added to the field as a proxy for the creation of value. When does the farmer make the profit? At the time he plants the seed? At the time he delivers it to market?

Within the context of this calculation I am suggesting my math pro-rates the value creation along the entirety of the production function at the rate of the value input to the field.

We could simplify this by removing the market value, and instead multiplying by the inputs only (not converted to a percentage). This would suggest we do not carry the farmer's potential profit as part of the benefit base. This would be an understatement of the loss, but possibly the cleanest way to proceed, and probably most consistent with policy.

We could simplify this by removing the percentage of production (and always pay full price less harvest to the farmer) which would be an overstatement of the loss because there will be cases where the farmer did not bear the cost of some inputs to production, but yet received the benefit of loss.

Instead, the math (as it stands in the op) effectively operates between the two cases, providing a percentage of the profit based on the amount of input contributed. If the loss occurs on the day before harvest, the farmer would receive profit commensurate with the work, if it occurs the day after planting (to reflect lost opportunity to change course), they would receive only a small portion of the profit (to reflect the opportunity to change to a different crop).

[Brennan1994]

Ok. I feel like I understand now. I'm just going to try to repeat what you're saying dumbed down a little more in my words.

There's three ways to handle this problem.
1.) Don't consider the value of the crops, just consider the cost of business. This would be multiplying the exposed value by the damage factor. It ignores the value of crops entirely. This is going to be an understatement of the loss. In this scenario we are assusming the value of the crops is 0 until they're harvested.

2.) Pay full value for the crops. This would be taking market value times damage factor, and calling it good. We're going to overcount cost though because the farmer didn't have to do the work of raising the crops that got damaged, but he's still gonna get credit for the value they would have had. In this scenario, we're assuming the value of the crops is full market value at planting.

3.) Your approach is to prorate the value of the crops using the percentProduction. Meaning, the value of crops increases from 0 to full market value at the same rate the work on them is being completed. If you've spent 50% of your budget raising those crops, we assume those crops have a present value 50% of their market value.

[Brennan1994]

In our metaphor of the the house, I mused that the losses when it flooded would be the value of the house in it's current state, plus the labor and materials to get it there, multiplied by the damage factor (how much of it I lost ex. 75%)

In the ag compute, I see now that we're getting at the present value of those crops by prorating them with the production function,, but we're not actually accounting for the labor anywhere, correct? Labor is only included in the calculation as a stepping stone to figuring out the present worth of those crops at the time they flooded. Shouldn't we be adding a term for that somewhere?

[HenryGeorgist]

We are assuming that profit covers labor essentially.

[Brennan1994]

Alright. we have to use the marketPrice as part of the equation because we want account for the value of the immature crops somehow. We don't want to look at crops as having 0 value until their picked, rather that their value is increasing from 0 to their market value over time parellel to the work being put into the field. Loss is going to be the value of the crops at the time of flooding + the labor put into the field x the damage factor. There's not a good way to seperate the profit from the value of crops at the time of flooding, so we drop the second factor (labor put into the filed) and say it's a wash.

Is that right?

[HenryGeorgist]

https://www.hec.usace.army.mil/confluence/agdocs/agriculture-damage-computation-proceedures-43820820.html

The link above is the current documentation on the computational procedures. There is always room for improvement.

[Brennan1994]

I will definitely contribute to that document once I've got a clear understanding