Covering cows

In Kenya's Marsabit District, up to one-third of the cattle die due to drought. But hundreds of herders now benefit from an innovative livestock insurance scheme, set up by the International Livestock Research Institute (ILRI) in Kenya. Payouts are triggered if satellite images predict that herders will lose more than 15 percent of their cattle, as the head of ILRI's livestock insurance scheme, Andrew Mude, explains.

Deutsche Welle: How many pastoralists does your service insure at the moment?

Right now, 599 persons have bought the insurance in January/February of 2011, and about 400 persons have purchased the insurance in August and September of this year.

How much does it cost to insure an average herd?

It depends on the size and composition of the herd. In Marsabit District, there are two clusters: the upper cluster and the lower cluster. For the upper cluster, it costs 5.5 percent of the value of livestock you're insuring to get insurance coverage for your area for one year. And in lower Marsabit, it's 3.25 percent. So, assuming that the value of a cow is 15,000 shillings, or about $150 (108 euros), to insure it for a year in the upper cluster would be $8.25 (6 euros), and in the lower area $4.88 (3.50 euros).

That's a considerable investment for the herders to get insurance - or is it something they can easily afford?

It's not a small investment in terms of cash. Marsabit is one of the poorer districts of Kenya, but they feel that this product is of value to them, and the population is beginning to conceptualize it as well. If someone can sell one cow to insure 20 cows, then it's fair for them.

A couple of days ago, you were present when some herders got their first insurance payouts. What was that like?

It was a very important milestone, for the pastoralists to actually see that those who bought contracts for this particular coverage period are actually getting money. It was an event. Not everyone was paid, but several farmers were paid, and there was testimony from some who were paid. They said, this has come at just the right time, and this is something that we can now trust, and maybe some of my other colleagues will pay more attention and be more interested. So, I do think it's a milestone for the project.

When you say it was the first payout, how much of the herds did the livestock keepers actually lose in the past year, and how much did they get paid out?

The model is an index-based insurance product. So it doesn't work according to individual losses. It works according to predicted losses, which we predict from satellite information on the forage availability or scarcity over the year in that particular area. During this time, predicted mortality for the different areas was between 18 percent and 33 percent. And the insurance has a deductible of 15 percent, so you get paid for any losses above 15 percent. The indemnity payments were an average of 3,000 shillings, which is about $30 (21.50 euros), and the max payout was 37,000 Kenyan shillings, which is about $370 (266 euros).

So they actually saw that the insurance benefits them.

Yes, I believe so. This was the general sentiment from talking to the people who received a payment. Of course, as with any product, you'll have some who are happier than others. Those in the area where the predicted mortality was 18 percent, they're only getting back 3 percent, which is the difference between the predicted mortality and the deductible of 15 percent. So, even though they experienced less loss, they were questioning why they were not getting the same payment as the herders in those areas where there were more losses.

And why does the insurance scheme work in that way, that insurance is not paid out when an animal dies, but based on satellite pictures determining how good or bad a grazing area is?

This is an innovation in insurance, which allows the risk management benefits of insurance to really come to bear to a very remote population, and for small insurance amounts. One reason it would be difficult for traditional insurance to serve this population, is that when you have losses, you need to verify them. This area is quite far from the main towns where you have insurance agents and so on, and for an agent to have to travel all the way to this area just to verify whether a particular farmer has lost two animals - valued at about $300 - is too cost-inefficient. This model allows for us to ascertain what the livestock condition in the area is, without actually having to go to that area.

So, you hope that in the future, via mobile phone technology, the banking - the payout and the sales - could be done completely via cell phone networks?

Yes, certainly. Most of the sales are already being done via telephone-based platforms and other points of sale, so once the capacities extend to the whole area, then I think it would be possible. This would really bring down the costs of implementation, and therefore also bring down the price of the product itself: the premiums.

What scope do you see for this insurance in the area, and also maybe in other countries?

We think there's scope for scaling up, but only for pastoral production systems ... But there are many areas across northern Kenya, in Ethiopia - we've started out a project in Uganda, and we're also thinking of moving this to areas in West Africa, where the pastoral system is a big part of livestock production. And there also are areas in Southeast Asia where such conditions hold true.

Over the next three or four years, our main aim is basically to expand the product, and continue to learn. And once we have learned lessons of where this product is best suited, we'll be able to write a process manual that can be used by various commercial entities, even other partners who are interested in implementing this program in different places.

Interview: Anke Rasper
Editor: Sonya Angelica Diehn