Here in Ann Arbor, Michigan, this past March has been anything but spring-like. Even by our standards, we’ve had more than our fair share of cold, wind, and snow. And now winter seems to be extending itself into April…after a tease of spring with a warm and sunny Easter Sunday, this Monday brought us a high of 39F with scattered snow flurries. To which I say boo.
So, yes, warmer winter weather would be nice. And as it so happens, warmer winters are something we should expect to happen with climate change, so that’s good news. The bad news is that (as economists like to say) there is no free lunch, so climate change will bring us hotter summers too. And as anyone who lived through last summer in Ann Arbor will attest, three months of scorching weather can be pretty unpleasant too.
But does the discomfort associated with a hotter summer actually outweigh the increase in comfort that will come with a warmer winter here in Ann Arbor? And what about the rest of the U.S., some parts of which currently experience very little winter cold (Florida), while others experience very little summer heat (North Dakota)? In a new paper (Climate Amenities, Climate Change, and American Quality of Life), David Albouy, Walter Graf, Hendrik Wolff, and I try to answer these questions by assessing U.S. households’ willingness to pay to avoid hot and cold weather.
How do we come up with an estimate of households’ willingness to pay for climate, in dollars? We take advantage of the fact that local climate is a factor that potentially affects households’ choice of where to live. Compare Berkeley to Ann Arbor, for example. I’ll conjecture that most people think that the climate in Berkeley is superior to that in Ann Arbor–Berkeley has lots of days with warm afternoons and cool, comfortable evenings, while in Ann Arbor such days tend to occur only in the spring and fall. So why don’t people move from Ann Arbor to Berkeley as soon as they have a chance? Well, it costs a lot more to live in Berkeley than it does in Ann Arbor. For the cost of a studio in Berkeley, you can get a pretty nice two bedroom apartment in downtown Ann Arbor. So there’s a tradeoff here–if you live in Berkeley you get beautiful weather, but you have to pay a pile of money in rent (or mortgage payments if you own) to get decent housing, leaving you with less money to spend on other things you want. Thus, the cost of living difference between Berkeley and Ann Arbor reveals something about their residents’ willingness to pay for a comfortable climate.
In the paper, we extend this idea across the U.S. to get at households’ willingness to pay to avoid hot and cold weather. There are, of course, a number of important details that are important to get right along the way–for instance, differences in wages across cities matter too, as do differences in other amenities such as coastlines, mountains, and population density. Our main result is that, on average, households’ willingness to pay to avoid an excessively hot day is greater than the willingness to pay to avoid an excessively cold day. This result comes from the fact that in hot places like the South, the cost of living is quite low relative to wages, indicating that households are willing to pay very little to endure the South’s hot, humid summers. The result also makes (at least to us) a lot of intuitive sense. On a cold day, you can protect yourself outside by layering up with coats, hats, and mittens. On a hot day though, there is unfortunately a limit to how many clothes you can take off (most jurisdictions won’t even legally let you get down to zero), so you’re uncomfortable outside no matter what you do. Sure, you can go inside to your air conditioning, but if you’d rather be outside in the afternoon, there’s a cost to doing so.
What does all this mean for the potential effect of climate change on comfort? If heat is worse than cold on the margin, then the extra discomfort from hotter summers will outweigh the benefits of warmer winters. When we use “business as usual” climate projections for 2100, we find that unless technology or preferences change, the annual net loss of comfort from climate change will be worth between one to three percent of U.S. income. Those numbers might seem big or small to you depending on your perspective, but for reference they’re in the same range as forecasted losses in GDP from damages to market goods such as agricultural products that come from the Nordhaus DICE model and the Stern report.
Perhaps even more interesting than the average welfare loss is the distribution of this welfare loss across the U.S., which we depict in the map above (losses expressed as a percent of 2100 income). As you might expect, the South gets hit fairly hard, since their summers are already hot and their winters are fairly mild–there’s not a lot of scope for climate change to give southern residents benefits in the winter. But most of the North fares quite poorly as well, even in places like North Dakota that currently have harsh winters but mild summers (though they do already experience some pretty hot days). What’s going on there? This is where it’s important to note that not everybody has the same preferences for climate. Some people aren’t bothered that much by cold weather, while others can’t stand it. It turns out that people sort themselves according to these preferences, so that North Dakota is largely populated by people who don’t mind the cold so much, while Florida gets populated by people who strongly dislike the idea of winter. So if you live in North Dakota, the fact that climate change will result in a warmer winter won’t be worth that much to you, but the fact that your relatively mild summer will turn into an extremely hot one will be worth a lot. The take-away message here is that if we want to forecast who will gain and who will lose from climate change, taking regional preferences into account can matter a lot, and it may be that southerners won’t be the only ones to suffer if the latest climate projections are realized.