Deal first with the risks that score highest. Rate each risk from one to five for how likely it is and from one to five for what it would cost, multiply the two ratings, and sort the list by the result, the score. What you end up with is a sheet with one risk per line, its two ratings, its score and what you plan to do about it.
The method comes from a training course on farm risk written for the Platform for Agricultural Risk Management (PARM), a program hosted by the International Fund for Agricultural Development (IFAD), a United Nations agency. Its authors put the arithmetic in one line: A simple multiple is used to calculate the values.
Ask what could go wrong on a farm this year and the answer usually comes as a list with no order to it: a dry spell, a price drop, a combine that breaks down mid-harvest, a foreman who quits. Ask two people on the same farm which one is worst and you often get two different answers, because nobody wrote down, for each risk, how likely it is and what it would cost. The two ratings put exactly that in writing.
What are the two numbers you rate?
Likelihood and impact, each from one to five, against a scale written at the top of the sheet so that two people rating the same risk reach the same number. For likelihood, the PARM course has five levels, from remote to probable, each with a percentage chance that the event happens in the period you are planning for: remote translates to 0-10%; unlikely to >10-25%; possible to >25-50%; likely to >50-90%; and probable >90-100%. Remote is a one, probable a five.
Impact is what the risk would cost the farm if it happened, from negligible at one to catastrophic at five. In the course’s example, the lowest step is a loss of under 5% and the highest a loss of more than half. Its authors advise adapting those steps to each business, because where margins are thin a 15% loss can already be catastrophic. The ratings are judgment calls, and the authors say as much: the people doing the rating should question and debate the scales.
For two risks on almost every farm list, a price drop and a short crop, the impact rating can come from money rather than a guess. Estimate how much a likely fall in price, or in yield, would take off the season’s revenue, and pick the impact rating that matches that loss on your scale.
Price and yield can also fall in the same season. On separate lines each fall can look bearable, while the two together can wipe out the profit, because nobody adds up their losses. Give that case a line of its own and rate its impact from the combined loss. How far price, yield or both at once can fall before revenue stops covering costs is worked out step by step in another article on this site, on how bad a year your farm can take.
A request for the farm’s emissions figure, from a buyer or under a new carbon rule, belongs on the list as a risk from outside the farm. Its impact is usually rated by guesswork, because the farm’s emissions have rarely been measured. In a separate article on this site, on where the farm’s emissions come from, you work that figure out activity by activity; with it, you can estimate what meeting the request would cost and rate the impact from there.
How do the two numbers become one?
Multiply them. Two ratings from one to five give a score from 1 to 25, and sorting by the score is enough to put the list in order. What is left to decide is where one band ends and the next begins. On this site’s risk matrix calculator, the four bands are: 1 to 4 is low, 5 to 9 moderate, 10 to 15 high, and 16 to 25 critical.
Laid out as a grid, with the five likelihood ratings on one side and the five impact ratings on the other, the 25 cells make up the risk matrix. Only four cells reach critical, the ones where likelihood and impact are both four or five: the risks that are both likely and heavy.
Which risk comes first, on a real list?
The one with the highest score. In the PARM course, the authors work through an example farm with seven risks.
| Risk | Likelihood | Impact | Score |
|---|---|---|---|
| Drought | 4 | 5 | 20 |
| High input prices | 4 | 4 | 16 |
| Pest | 3 | 4 | 12 |
| Labor shortage | 4 | 3 | 12 |
| Low market prices | 3 | 3 | 9 |
| Disease | 2 | 1 | 2 |
| Flood | 1 | 1 | 1 |
Sorted by score, drought and input prices come out on top, which is the authors’ conclusion too: the highest priority risk for this farmer is drought, followed by high input prices. Anyone holding the sheet can redo each multiplication and check the order.
Where is the matrix easy to misread?
At a tie. The pest and the labor shortage both score 12, and they are not the same risk: the pest is a moderate chance (3) of a heavy loss (4), the labor shortage a high chance (4) of a middling loss (3).
From the score alone you cannot tell which of the two comes first. When two risks tie, look at their two ratings and settle the order with the people who gave them.
Does a rare disaster beat a frequent nuisance?
It can, though the two score the same: a rare disaster rated 2 and 5 and a frequent nuisance rated 5 and 2 both come to 10. Many people do not weigh them as equal.
In two online experiments with 2,699 British adults asked about imaginary flood risks, not farm risks, a Cambridge team led by Sutherland found an overall preference to reduce impact over likelihood. In a US government guide to risk assessment, written by NIST for computer systems rather than farms, a low-likelihood, very-high-impact event is rated moderate and a very-high-likelihood, low-impact one low. So when a rare risk would be catastrophic, compare it with the frequent risks of the same score, and if you judge it worse, move it up the list by hand and write down why.
What do you do with each band?
Decide what to do about each risk, and use the band to set the deadline. A critical risk gets an action now, with a person’s name and a date beside it. A high risk gets an action planned within the year. A moderate risk is watched and comes back at the next review of the sheet. A low risk is accepted and recorded, which is also a decision. Every action ends one of two ways: a fix, with a person responsible and a deadline, or a written decision that the farm will carry the risk. A sheet with no names beside the critical and high risks is still just a list.
Where to start
Set aside an afternoon with the people who know the operation, name the year’s worries out loud, and have someone write them down. Look for risks from outside the farm, such as prices, weather, regulation and buyers, and from inside it, such as people, machines and cash, which is the split the authors of the PARM course use.
The ratings go out of date long before the sheet wears out, which is why it carries a date. A drought rated four last year may deserve a five after two dry seasons, and whoever kept the old sheet can see the rating change and discuss why. With the risk matrix calculator, the multiplying and the sorting are done for you, and the ratings stay yours to give. Other articles on getting a farm through bad years are in the library’s sustainability section. Ranking risks is one part of farm management; the rest of it is on that page.