Temperature and crime

It is well known that crime rate are higher in the summer than the winter. During the winter it is simply harder for criminals to get around and their victims are also less likely to be outside. That doesn't mean if you make the summers warmer, that the crime rate will go up. I would like to have seen the crime data being used be monthly or even daily data, and it would have been nice to see that other factors that impact crime be controlled for. The reason for data over the course of the year is that it shouldn't be just the average temperature, but also the how high and low the temperatures go and how long they stay at those temperatures. There might be a lot more crimes at 70 degrees then if it is half the time at 30 and half the time at 110. It is also hard to think of many economists that would still use national time series data to evaluate changes in crime rates.

Using U.S. government data on average yearly temperatures and the number of violent crimes between 1950 and 2008, the researchers estimate that if the annual average temperature in the U.S. increases by 8°F (4.4°C), the yearly murder and assault rate will increase by 34 per 100,000 people -- or 100,000 more per year in a population of 305 million. . . .

Now this part is clearly wrong. Higher temperatures will increase the area of the world that can grow food and increase the growing season.

In addition to the "heat hypothesis," they report that rising global temperatures also increases known risk factors for the development of aggression in violence-prone individuals -- such as increasing poverty, growing up amid scarce resources, malnutrition and food insecurity. They contend that one of the most catastrophic effects of climate change will be food availability, producing more violence-prone individuals in the process. . . .