
The Financial Times published the following chart last night. It shows the cumulative number of COVID-19 cases around the world, across the number of days since the 100th case in that particular country. The message here is that most western countries appear to be on a similar trajectory. (The grey dotted line represents a 33% daily increase.) Whereas in Asia, and in particular Hong Kong and Singapore, they have seemingly managed to slow the spread.

Now, there are a number of possible explanations for the outliers; everything from stricter quarantine rules to more rigorous testing. There's also an argument that Hong Kong and Singapore were better prepared as a result of the SARS outbreak in 2002. (More on these explanations, here.) But the other factor at play seems to be climate.
A recent study (by Jingyuan Wang, Ke Tang, Kai Feng, and Weifeng Lv) has concluded that, like the flu, the transmission of COVID-19 appears to be significantly impacted by both air temperature and relative humidity. In their research, they looked at the reproductive number (R), or the severity of infectiousness, for all Chinese cities with more than 40 cases between January 21 to 23, 2020. (Large-scale government interventions began on January 24, 2020 and would have therefore skewed the numbers.)

What they found was that for every one degree Celsius increase in temperature and every one degree Celsius increase in relative humidity, the reproductive numbers drop by 0.0383 and 0.0224, respectively. Air temperature, in other words, has more of a positive impact on containing spread than relative humidity -- which feels right. That is also apparent when you look at the above charts. Take note of Korea, Iran, and Italy near the top left corner of the temperature chart.
If you'd like to download a full copy of the research paper, click here.
In 2015, Marshall Burke, Sol Hsiang, and Ted Miguel published a paper in Nature that looked at the relationship between temperature (climate) and economic output. They examined the historical impact of temperature changes (1960-2010) on 166 countries and then used this data to try and predict the potential future impacts of climate change on GDP per capita.
What they discovered is that temperature has a non-linear impact on economic production. Put differently, there’s an optimal annual average temperature. And it turns out to be 13 degrees celsius. If a country sits below this average number, then warming increases productivity. But if a country sits above this number, then warming has a negative impact on productivity. And the impact gets worse (stronger negative correlation) at higher temperatures.
Some of you are probably wondering whether the correlations they found should be interpreted as causation. For what it’s worth, the study tries to correct for non-temperature related economic changes (such as a recession or policy changes) and it also looks at how individual countries perform against themselves during temperature fluctuations. So the control and treatment groups are arguably pretty tight.
All of this suggests that there are a number of countries that stand to benefit from climate change (at least from this perspective). They are the ones that are cold today.
For more on the study, click here.

What is happening in California right now is both sad and scary. I woke up to these photos in the New York Times. So I spent the morning reading up on wildfires and what causes them.
I am sure many of you are thinking: Is climate change doing this?
It’s important to note that California is designed to burn. The characteristics that make the state a highly desirable place to live – mild winters and hot and dry summers – also make it a highly flammable place.
Below is a map from Popular Science showing how much of California has burned over the last 5 years. Most of California’s hottest fires (13 of the top 20) have occurred since 2000.

But these naturally occurring fires are actually important for the regeneration of its forests. So one argument is that the current policy of “total fire suppression” is actually partially responsible for this increase in severe wildfires.
Low-intensity forests aren’t burning like they used to and it is creating more densely packed forests for even larger wildfires. In other words, through our actions we could be exchanging smaller and more frequent fires for bigger and more severe ones.
The other concern is development.
California had a population of about 24 million people in 1980. Today it is closer to 40 million. And much of that growth has occurred outside of existing urban areas (too hard to build). This has meant more people living in suburban and rural areas – adjacent to wildlands.
Ed Glaeser has repeatedly argued that the best way to respect nature is to actually stay as far away from it as possible. Humans are a destructive species. Living in a dense city is good for the planet.
Now let’s address the climate change issue.
Daniel Swain, who is a climate scientist and author of the Weather West blog, published this popular thread on Twitter about a day ago. At the time of writing this post, it had been retweeted and liked over 10,000 times.
From his perspective as a climate scientist, this is not all about climate change. That’s the wrong question to ask. It’s more complicated and nuanced than that. But climate does indeed act as a “threat multiplier” to the other factors we’ve already discussed.
Fire season typically begins and ends with rain. When the fall rainy season starts, the fire season ends. One of the biggest risk factors is a dry fall, particularly after a hot and dry summer (or after multiple hot and dry summers, as has been the case in California).
Because according to Swain, fall means the start of “offshore wind” season, which can help to stoke these wildfires.
Unfortunately, one of the projected outcomes of climate change for California is more precipitation concentrated in the “winter” months, at the expense of precipitation in the shoulder seasons (spring and fall). So that means a longer fire season.
Swain believes that if Northern California had received close to its typical amount of rain this fall, it is almost certain that this tragedy would have been avoided.
That, however, makes me wonder about the relative importance of all the factors we have discussed today.
For Daniel Swain’s full thread, click here.