| Interview with a Hazard Expert
Download the MP3 here
Interviewer: Sarah Bucknall (SB) National Space Centre
Interviewee: Dr Paul Rockett (DR) Benfield Ltd.
SB: Hi, my name is Sarah Bucknall from the National Space Centre. I’m here with Dr Paul Rockett from Benfield Ltd. So Paul, please tell us about your job and a little bit about what you do.
DR: Well, first I ought to start off by just telling you a little bit about the kind of work that my company’s involved in because to most people it’s probably not familiar. Now if you imagine you live by a river, say, you won’t be surprised if your parents buy insurance against the possible flood risk. Now what we do, if you imagine a big storm event, there will be an awful lot of properties that will be affected, possibly a whole town. And for an insurance company that will have to provide the finance for this thing, sometimes they find that really big events are quite disastrous for them. So what they do is they actually go out and they buy insurance for themselves. And what the company that I work for does is actually place that insurance.
Now you might ask why is that interesting from a science perspective and actually what we need to do is understand the risk and understand the natural perils that are involved and to be able to quantify the risk that they’re facing. So our role would be to work out how often a particular insurance company might lose a particular sum of money and then we can place that risk in the markets. From the science perspective, what we need to do is actually understand those particular risks. And my job is involved very much in this kind of work.
SB: So how did you first get into this line of work?
DR: I started off looking at economics, financial economics in particular, and when I was coming towards the completion of my masters degree I started looking for jobs. I went for an interview with a reinsurance broker in London. Although I didn’t persuade them to give me a job, what I did persuade them to do was to fund a PhD in extreme value mathematics. But what quickly happened was that this turned into extreme mathematics applied to earthquake modelling and I got involved in earthquake risk assessment. And then once I finished my PhD I went and did a post doctorate in hurricane forecasting. So that’s generally the way I got into natural peril modelling.
SB: So have you always worked in the reinsurance industry?
DR: No. When I left Cambridge, when I did my PhD, I went to University College London and looked at hurricane forecasting. That was actually in a space science laboratory which was a really nice place full of people with white coats. So after being there for 3 years I decided that the lure of the city and the bigger pay cheques and doing the same kind of work in an industrial kind of setting. It does give you more pressure but ultimately it can be more rewarding.
SB: So what particular qualifications do you need for your line of work? Is a science background essential?
DR: Yes, a science background is essential. You don’t necessarily need a PhD but the kind of work that we do is fairly varied so the kind of people that take these kind of jobs are ones that have a background in the geographical sciences – geography or geographical information systems, which is the spatial pattern of things – or mathematical sciences, so it’s fairly general but you certainly need to have a degree and it helps if you have a masters degree as well.
SB: Does it matter what your masters is in or is it just a masters in general?
DR: A masters in a scientific kind of discipline. Obviously mine was in financial economics which you probably wouldn’t class as a science but anything that is statistical related or has that kind of reliance is also fine.
SB: So in your job, what kind of predictions are you making?
DR: Well what we’re looking at is the events. We’re trying to predict events, in a way, and the effect of events on loss – so how likely it is that a particular insurance company is going to lose a certain amount of money. Or we might be looking at a very large individual risk, for example the Prague Metro. And that would be how often does this Metro get flooded, what are the causes, what can they do about it, and things like that.
SB: So does that go right down to the weather predictions? Is that the part that you’re involved in?
DR: We try to take time out of it. We assume that events are fairly random and it’s the way that you treat that randomness is the way that we do it. But as part of the team that I work for, we are involved in projects that actually take MET Office forecasts of wind speed. Then what we do is we take their wind speed predictions, which would be up to 3 or 4 days in advance, and we map that onto the United Kingdom and from there we work out how much damage we can expect up to 3 or 4 days in advance. So that’s really quite a nice new tool that we’re building.
SB: Is that model ever used to assess risk to humans? Do you ever assess danger?
DR: We do but it’s always classed in terms of loss. A serious wind storm has financial implications and social implications. So we wouldn’t be involved in the social side of things, which we would leave to other agencies, but just by being able to forecast events ahead of time can also give the agencies a handle on these kinds of events.
SB: So to who do you communicate your information?
DR: Our information’s largely communicated to reinsurance managers and reinsurance executives who have to decide how much insurance their insurance company has to buy. Obviously they’re dealing with quite large amounts of money and risks that can be a very long way in the distance. We try and work out how often the kinds of losses you can receive say once every 250 years. So that would be an extremely large flood or extremely big windstorm or earthquake.
SB: What happens in a typical workday for you?
DR: The first thing I would typically do is check my email and check the news wires to see whether there has been any sort of natural disaster or natural catastrophe that is either imminent or has already occurred. It doesn’t have to be a seriously major event but as soon as we have an event to deal with the first thing I would do is to write a report about it and get other people involved. Assuming most days we don’t have a natural disaster, typically I work on quite large projects. The key project I’ve been working on recently has been flood risk in the Czech Republic. So what that will do is we’ll be looking at the geographical implications – the kind of flood extent that we can expect at different frequencies or different return periods of event, and what kind of properties are going to be affected. And so my day to day job is actually assessing these kinds of risks and actually trying to work out how they interact with things like flood defences.
I also do a lot of computer programming. We build these models which help us talk about the risk so we have to go right from the geographical side to the mathematical side to the financial side to work out the risk to the insurer. To put all that together you need to have computer programming skills to pull it all together. But we have various scientists that work specifically on the geographical side or the finance side and I like to link between them.
SB: So is all of your work based in Europe? Or is it global wide?
DR: I personally deal with Europe largely. Obviously there are natural perils throughout the world that are a big concern to my particular company and there are other people in my team that deal with South Africa and earthquakes that they get there. We’re also involved with the United States where they have hurricanes and also earthquakes. We’re pretty much involved in everything but I personally deal with Europe.
SB: So does that mean you get to travel to Europe or are you based in the office?
DR: It does, I’ve had lots of good trips through my time. In fact the whole of last year I spent actually working in the Czech Republic which was excellent. In my previous roles in hurricane forecasting I got as far as Trinidad and the United States, Singapore, Australia. So there are lots of opportunities to travel.
SB: What would you say are the best parts of your job?
DR: The best part for me is that I like the problem solving. Certain things are quite difficult to do geographically. So the implementation say of flood defences, you would think about how you would deal with that geographically but you can also think about it mathematically and trying to build these things in – they’re quite difficult problems and they can take many weeks sometimes to solve. But also I like the fact that I work on large projects, large natural perils projects. So I’m involved in the science in all the different elements of it and pulling it together with the computer programming as well gives a complete picture on the kind of things that are going on in the world.
SB: So you get a satisfaction from solving those problems and helping out?
DR: Absolutely, yes.
SB: What are the worst parts of your job?
DR: I must admit that’s quite difficult. I actually quite like most of my job. The only thing that is feasibly an issue is that we’re trying to do science in a tight deadline and sometimes that doesn’t work very well. Sometimes when you’ve got a week to do it, it may take 3 weeks. But when it all fits together it’s brilliant.
SB: Are there any major consequences of when you have to run over schedule?
DR: Sometimes we can’t. And then you have to find a fix that puts it together. It’ll be your own intuition together with the actual science that tries to pull it together. There can be big consequences like the Czech flood project which we ideally wanted to finish by renewal. There is a particular date for Europe where all the insurance companies buy their own insurance - this reinsurance. And that typically is the 1st of January. But for us, we have to do the natural hazard side of things by about the end of August/beginning of September so that all the financial side can be dealt with by other people. So if we miss that kind of deadline we can’t get results out in time for the markets so they won’t be able to come up with their best pricing based on the latest model in time for this 1st January reception date.
SB: What would be your advice to someone who would be interested in pursuing a similar career?
DR: This is quite interesting. I would guess that you would need to have a very strong interest in natural catastrophes/natural perils. And once you’ve got that kind of interest then you need to start thinking about the kind of subjects you would be doing at A-Level and the degree. But the great thing with what I do is that you can be fairly flexible within reason so I started off largely with economics but you could just as easily do this with geography or one of the other sciences. And in fact, because a lot of the stuff I do is mathematics, even a physics background or things like that would all be quite relevant. But you do need to start thinking about the sciences and going into that kind of field.
SB: Ok. Thank you very much.