This updated paper on agricultural insurance is interesting and well-written. There are 6 BRM programs (Business Risk Management programs) within the GF2 framework (Growing Forward 2). The bulk of the paper deals with the first BRM program, production insurance. (Production insurance nicely illustrates general actuarial pricing concepts. Self-sustainability is a critical part of the pricing of production insurance; stochastic methods are used.)
- 1 Pop Quiz
- 2 BattlePlan
- 3 In Plain English!
- 4 BattleCodes
- 5 POP QUIZ ANSWERS
What is F-PIP-RIGOR?
Based on past exams, the main things you need to know (in rough order of importance) are:
- concepts around self-sustainability (the specific question varies)
- BRM programs (Business Risk Management programs) of GF2 (Growing Forward 2)
- agricultural insurance: types of plans & calculating indemnity payments
- historical adjustments to probable yield
- stabilizing methods applied to probable yield data
Important: Questions from 2015.Spring and prior are from the old agriculture reading. Some of the old problems may still be relevant, but don't spend too much time on them.
- 1 Agricultural insurance is the same thing as production insurance in this paper.
In Plain English!
This is an updated (and much improved) reading on Canadian agricultural insurance. It first appeared in 2015.Fall, and the questions on that exam were pretty easy. Unfortunately, the difficulty of the questions has been trending upwards ever since. (You now have to study the reading thoroughly to know the answers) This is probably the most memorization-intensive reading on the syllabus.
Sections 1,2: Intro & Glossary
Question: what is this reading about
- - it is about something called GF2 (Growing Forward 2)
Question: what is GF2
- - a comprehensive federal-provincial-territorial framework for Canada's agricultural sector
GF2 has 6 BRM programs (Business Risk Management programs):
BRM # BRM program Description funding 1 AgriIns (Agricultural Insurance) 1 protects against...production loss producer-provincial-federal partnership 2 AgriStability (Agricultural Stability) protects against...margin decline producer-provincial-federal partnership 3 AgriInv (Agricultural Investment) an investment fund...for small losses producer-provincial-federal partnership 4 AgriRecov (Agricultural Recovery) protects against...disaster provincial-federal partnership 5 AdvPmtsProg (Advance Payments Program) provides...low-interest loans for cash flow management federal funding 6 WLPIP (Western Livestock Production Insurance Program) protects against...fluctuation in livestock prices producer-provincial-federal partnership
- 1 This program of agricultural insurance, also called production insurance, is the most important of the 6 BRM programs discussed in this paper.
The rest of Sections 1 & 2 is a glossary. See the BattleCards. (Many of the glossary terms are obvious and are not included in the BattleCards.)
Section 3: Overview of Agricultural Risk Management in Canada
This section feels like an expanded version of Section 1. It is extremely detailed, and I'm not sure it's worth reading (judgment call on my part) - the other sections have the potential for many good exam questions.
Section 4: Production Insurance Programs
This is a long and detailed section, so I've broken it into 3 subsections. (This is where most of the good exam questions come from.) First, some basic facts about the Agri-Insurance program:
- all provinces have an Agri-Insurance program
- they are administered by Crown corporations OR branches of the provincial agricultural departments
- they are non-profit, experience-based (to encourage "good" behaviour), and the province may access provincial or federal reinsurance if its surplus is depleted
Actuarial Certification and regulation are covered in the mini BattleQuiz.
Section 4a (Types of Plans)
There are so many testable facts here that it's hard to know what to focus on. Let's start with the most basic facts, things that have been asked on past exams:
Question: what are the different types of Agri-Insurance plans?
- yield-based: can be individual or collective
- non-yield-based: examples are weather derivative, acre-based, mortality for livestock
Question: what is a yield-based plan
- a plan where the indemnity payment is based on the actual yield versus the insured yield
- Example: A corn farmer buys production insurance based on a yield of 10 tons / hectare.
- if actual yield = 11 tons / hectare → no coverage is triggered
- if actual yield = 8 tons / hectare → coverage would be triggered, and the insurer would compensate the farmer for low production according to the contract terms.
- Example: A corn farmer buys production insurance based on a yield of 10 tons / hectare.
- Pretty easy! (Section 4b covers the formulas for this.)
Question: how do non-yield-based plans work
- For this type of production insurance, coverage triggers are NOT based on yield.
- Example: A weather derivative plan is triggered when a pre-determined meteorological threshold is breached REGARDLESS of actual yield. (This might be rainfall that exceeded a specific amount.)
This section of the paper also introduces pricing of production insurance.
- Obviously, if an insurer is going to offer a yield-based or non-yield-based production insurance plan, they have to have a valid pricing methodology
- Interestingly, yield-based and non-yield-based plans use similar methodologies EXCEPT the non-yield-based plans have some EXTRA considerations. (Section 4b)
Question: identify the 1st step of pricing a yield-based production insurance plan
- An obvious first step is for the insurer to calculate the average production yield for the farmer over the appropriate experience period. (This is like looking at someone's driving record if they were applying for auto insurance.) This average production yield is called the probable yield.
- Pop Quiz: If the probable yield = 10 tons / hectare, which policy would be more expensive:
- Policy A: The farmer buys insurance for 15 tons / hectare. The contract provides for compensation of $1,000 x shortfall.
- Policy B: The farmer buys insurance for 12 tons / hectare. The contract provides for compensation of $1,000 x shortfall.
- Obviously, Policy A would be more expensive because the shortfall (hence payout) is always greater for Policy A.
Let's finish with 2 important topics, both of which were asked on (2017.Spring #8bc) related to data adjustment. Instead of calculating probable yield as a simple average of historical production yield, sometimes certain adjustments are appropriate:
Adjustment #1: briefly describe the concept of adjusting historical probable yield data for estimating current yield
- adjustments to historical production yields to reflect CURRENT production capabilities.
- Example: If there has been an increase in efficiency in farming practices, then production yield for older years may need to be 'leveled up' (This is similar to on-leveling premiums to account for rate changes.)
- Note that in the exam problem, you're asked to describe historical yield adjustments whereas in the BattleCards, the question asks for triggers to historical adjustments. It's the same thing.
Adjustment #2: briefly describe the concept of stabilizing methods used in historical yield methodologies
- various adjustments to lessen the effect of outliers (Ex: use a long-term averages of 15-25 years versus a shorter-term average)
The specific triggers & methods for these adjustments are covered in the mini BattleQuiz.
Section 4b (Formulas)
- Let's introduce some notation and formulas then apply them to (2016.Spring #8c)
Symbol Meaning PG Production Guarantee AP Actual Production A Area that is insured (or insured area) P Probable yield per unit of area C Coverage level % Indem$ Indemnity (in dollars) paid by insurer to insured
- Basic Formulas
PG = A x P x C Indem$ = MAX ( 0, PG - AP ) x (insured unit price)
- For (2016.Spring #8c), we're given these 5 pieces of information:
- A = 100 acres
- C = 70%
- P = 584 kg/acre
- insured unit price = $0.41 / kg
- AP = 25,000 kg
- Simple substitution gives:
- PG = A x P x C = 40,880
- Indem$ = MAX ( 0, PG - AP ) x (insured unit price) = MAX ( 0, 40,880 - 25,000 ) x $0.41 = 6,511
- This is an easy problem. You should definitely get this one right.
- Now, there are other formulas in this section you should probably know, although I think they are less likely to be asked.
Symbol Meaning L$ Liability (in dollars) that insurer assumes IndemRt Indemnity Rate in dollars per unit insured Indem$ Indemnity (in dollars) paid by insurer to insured PremRt Premium Rate in dollars per unit insured Prem$ Premium (in dollars) paid by insured to insurer
- Pop Quiz: What is the difference between Indem$ and Prem$?
- Indem$ is the dollar value paid by the insurer to settle a claim, and Prem$ is the dollar value received by the insurer for purchase of a policy.
- Obviously, Prem$ should be greater than Indem$.
- So, starting with expected Indem$, you would load in a bunch of stuff like expenses to arrive at the Prem$ you have to charge.
- Ok, here are a few more formulas:
L$ (yield-based plans) = PG x (insured price) L$ (non-yield-based plans) = (# of insured units) x (insured price) Indem$ = IndemRt x L$ Prem$ = PremRt x L$
- NOTE: The above table highlights a difference in pricing methodology for yield-based versus non-yield-based plans.
- A non-yield-based plan can't have a PG (Production Guarantee) because production, or yield, is not how those plans work.
- This is all a lot to take in. Let's summarize a step-by-step method for calculating the final premium:
- Calculate Indem$: Use the formula from the beginning of this section: MAX ( 0, PG - AP ) x (insured unit price)
- Calculate L$: Use the above formula for L$ for either a yield-based or non-yield-based plan
- Calculate IndemRt: Rearrange the formula for Indem$ given in the above table to get IndemRt: Indem$ / L$
- Calculate PremRt: Start with what you just calculated for IndemRt then load in:
- uncertainty margin
- balance-back factor
- individual discount/surcharge
- reinsurance load
- self-sustainability load
- Calculate Prem$: Use the formula in the above table: PremRt x L$
- Here's one final memory trick before I let you loose on the BattleCards!
- The trick I use to remember these load factors is: IRt(UB+/-RS)
- IRt is an abbreviation for Indemnity Rate
- U = Uncertainty load
- B = Balance-back factor
- +/- means discount/surcharge
- R = ReInsurance load
- S = Self-Sustainability load
- If this trick doesn't work for you, invent your own!!. This feels to me like a potential exam question.
- The trick I use to remember these load factors is: IRt(UB+/-RS)
Be sure to check out user j5nguyen's alternate memory trick for this ==> SIR BUDs!!!
This mini BattleQuiz mainly covers the formulas discussed above and other related facts.
Section 4c (Self-Sustainability & Govt Role)
Agricultural insurance involves 4 major players: producers, insurers, provincial govt, and federal govt. If a program isn't sustainable, then these players will have to contribute more money and I'm pretty sure they won't want to do that!
Concept: self-sustainability is a BFD!!!
Let's examine this idea of self-sustainability.
Question 1: how is self-sustainability defined?
- Conceptual defn: RECOVER from severe loss scenarios WITHIN a reasonable time
- Statistical defn: FOR ALL (base, adverse) scenarios with INITIAL DEFICIT = 6th yr, 95th percentile: MUST RECOVER DEFICIT IN (15yrs on avg AND 25yrs with 80% probability)
Question 2: how can we design a self-sustainable program
- The basis for a self-sustainability test is a 25-year stochastic simulation of financial position. Without getting into the details of how this simulation is designed, just note that the simulation should take into account various adverse scenarios. The simplest type of adverse scenario is adverse claims experience. (See BattleCards for all 6 examples.)
Question 3: What is the actuary's role regarding self-sustainability
- The actuary should design or confirm the methodology used for calculating the self-sustainability load.
Question 4: how does the self-sustainability test (from Question 1 above) compare to DCAT?
- Similarity: both consider (base, adverse) scenarios
- Difference: Agri self-sustainability uses a fully stochastic simulation over a longer time horizon
The last topic in this section concerns the roles of various "players" in the delivery of agricultural insurance.
Question: what is the role of government and other players in agricultural insurance
- The government, both provincial and federal, consider agriculture to be integral to the economy of Canada. They support the traditional system of private insurance between an insurer and producer/insured. Private insurers may provide coverage for perils (like fire) not covered by normal production insurance. Private insurers may also offer reinsurance. (The details are straightforward and are contained in the BattleCards.)
- Note: These facts are a pain to memorize but you only have to look at past exams to see the ridiculous level of detail they expect you to know. You can go even deeper into this topic in Level 3: Custom Battles, if you dare. But first things first...
This is a fun little section where I made up some Bloom's Taxonomy types of questions. Thinking about them will be mind-expanding! This is because they tie together ideas from multiple papers, and force you to see things from a broader perspective. There are 4 questions altogether, and the first one is:
Compare the different triggers for: (1) Actuarial Certification (2) Historical Adjustments to Probable Yield (3) Risk Transfer Test
Note that (1) and (2) are from this paper, while the third is from CIA.ReIns, a paper you probably haven't read yet. Anyway, the answer I came up with is:
- Trigger for Actuarial Certification:
- significant changes in program design or methods
- new crops
- Trigger for Historical Adjustments to Probable Yield:
- a change in farming or management practices
- a change in insurance program design
- a change in data source or data collection technique
- maturity of perennials (yield would vary over their life cycle)
- quality variation of crop from year-to-year (due to insured perils or other cause)
- Trigger for Risk Transfer Test:
- inception of contract
- when a change to the existing contract significantly alters expected future cash flows
Check out the mini BattleQuiz below for the rest of them...
- 6 BRMs (Business Risk Management Programs)
- BRM #1 - Production Insurance (or agricultural insurance):
- 2 types of plans: yield-based versus non-yield-based + examples (weather derivative,...)
- 5 adjustments to historical yields
- 6 stabilizing methods to apply to data sets
- IndemRt load factors: IRt(UB+/-RS)
- Formulas: L$ (for yield-based & non-yield-based plans), IndemRt, PremRt
- See BattleCards for further details.
- Why is the government involved in agricultural insurance? (Why don't we just leave it entirely to the free market?)
- Compare & contrast DCAT with self-sustainability.
- Hint: A good first step is to briefly describe separately what DCAT and self-sustainability are. (That part is just memorization.) Then use your imagination to come with ways that they're similar and ways that they're different. (This is the Bloom's Taxonomy part of the question. The graders of your exam are testing whether you just memorized a bunch of stuff, or if you really understand how different parts of the syllabus are related.)
- probable yield (NO missing data) = avg of production yields over experience period
- probable yield (with missing data): see BattleCards (Calculational problems template for this coming soon)
- PG = A x P x C
- Indem$ = MAX ( 0, PG - AP ) x (insured unit price)
- Prem$ using the 5-step procedure from Section 4b (I really doubt you'd be asked to perform the full calculation here, but it's a good idea to know the steps.)
- Expect roughly 2.0 - 2.5 pts from this reading on the exam
POP QUIZ ANSWERS
F-PIP-RIGOR is my abbreviation for the 9 P&C risk categories from Appendix B in the CIA.DCAT reading:
F - P I P - R I G O R Frequency-Severity - Policy Liabilities Inflation Premiums - Reinsurance & CounterParty Investment Government Off-B/S Related Company
Each of these categories can be used to construct various adverse scenarios. This fact from the DCAT reading is related to something called self-sustainability from Section 4c in this agriculture reading. A self-sustainability test is similar to DCAT testing in that both must consider adverse scenarios. Now, back to our regularly scheduled programming!