Natural disasters: the numbers

This is a talk about disasters and specifically about what we know about disasters in terms of the numbers. My name is Ilan Noy and I'm a chair in the Economics of Disasters and a professor of economics at Victoria University of Wellington in New Zealand. I have been working on this topic for probably the last 15 years. What I'm going to try to do today is to try and sort of distill some of the insights that I've found and what others have found about what we know about the basics; about disasters, mostly from an economic perspective, because I am an economist.

Before we delve into the economic aspects of disasters, first we will look at how frequently these events occur. We will look at that even before we define exactly what we mean by disasters, and we'll do that later on in the lecture. But in the first slide, you can see a graph of how often these events occur globally. Now you can see a few things in this graph. One is that you see that the frequency of these events is increasing, or at least it's been increasing in the last 30 or 40 years, which is what this graph represents. The other thing we will note is that what mostly seems to be increasing are the hydrological events. So, these are things like floods, mostly floods. Most of them are floods, but they also include landslides and avalanches that are triggered by rainfall events. There is some observed increase in the last maybe 10 years in storms, but it's not a very obvious increase.

So, another look at the frequency, and this comes from a different data set put together by another reinsurance company Swiss Re. There, if you look at the solid line, the solid line is what we now call natural disasters. We later decided that this is not a good name for them. But, what you actually observe in their data, first of all, you see the numbers are different. If in Munich, the data set said they had about 700 events in the last couple of years; in the Swiss Re data set, they had a lot fewer events, maybe 200. That's one thing to notice. So, that suggests that what Swiss Re calls as a natural disaster is very different than what Munich calls one. The other thing we notice is that the main increase in the frequency of these events occurred not recently, but rather between 1970 and 1995. That suggests that maybe what is going on is not necessarily a change in the frequency of these natural events, but that actually something else is going on. The most obvious hypothesis is that we know more about these events now because we have the Internet. The Internet was roughly introduced in the mid-1990s. So, it's very closely correlated with when we start to see these increasing events. So, our ability to identify these events and know that they occurred increased significantly between 1970 and 1995 and that's probably one reason why we observe an increasing frequency. Another possibility is that we just have more people and more people in more places and because of that, they are more exposed to these events and therefore, we get more reporting about them. So, one hypothesis was we observed an increased frequency of events because we have more people globally and our ability to identify those events or actually for those events to affect people and therefore, get registered in these databases has increased in these past years and especially we have seen a significant increase in population globally and maybe, more importantly, we have two trends that also affect how many people are exposed to these events. That is one trend is the basis that people are moving to urban areas. So, we see a large increase in the number of people living in urban areas in the last 40 years. A lot of times these urban areas, the places where people are going to in the urban areas, are now less safe or are riskier in terms of especially in terms of flood risk, because in every urban area, in every city, the first places that get settled are the safest places and only when a lot of people are joining in later do they start to settle in areas that are more flood-prone or more landslide-prone and so on. So, because we see a rapid increase in urbanization in the last 40 years, that's another explanation for why we observe more events. Then, another trend that we have observed in the last 40 years is people who are moving to the coasts and people are moving to the coasts in almost every country. This is true in the United States. This is true in Australia, is true in China, and this is true in many, many other countries where people are moving to the coasts and the coasts are more exposed to disasters. So, we have people are moving to urban areas, especially the risky parts of urban areas. They are moving to the coasts, and there are more people. Because of all of these reasons, we observe more events. Conspicuously absent from our explanation of why we have more events is climate change.