“2% per degree Celsius . . . the magic number for how worker productivity responds to warm/hot temperatures”

Solomon Hsiang shares some bad news:

Persistently reduced labor productivity may be one of the largest economic impacts of anthropogenic climate change. . . .

Two percent per degree Celsius . . . That’s the magic number for how worker productivity responds to warm/hot temperatures.

In my 2010 PNAS paper, I [Hsiang] found that labor-intensive sectors of national economies decreased output by roughly 2.4% per degree C and argued that this looked suspiously like it came from reductions in worker output. Using a totally different method and dataset, Matt Neidell and Josh Graff Zivin found that labor supply in micro data fell by 1.8% per degree C. Both responses kicked in at around 26C.

Chris Sheehan just sent me this NYT article on air conditioning, where they mention this neat natural experiment:

[I]n the past year, [Japan] became an unwitting laboratory to study even more extreme air-conditioning abstinence, and the results have not been encouraging. After the Fukushima earthquake and tsunami knocked out a big chunk of the country’s nuclear power, the Japanese government mandated vastly reduced energy consumption. To that end, lights have been dimmed and air-conditioners turned down or off, so that offices comply with the government-prescribed indoor summer temperature of 82.4 degrees (28 Celsius); some offices have tried as high as 86.

Unfortunately, studies by Shin-ichi Tanabe, a professor of architecture at Waseda University in Tokyo who has long been interested in “thermal comfort,” found that while workers tolerated dimmer light just fine, every degree rise in temperature above 25 Celsius (77 degrees Fahrenheit) resulted in a 2 percent drop in productivity. Over the course of the day that meant they accomplished 30 minutes less work, he said.

Personally, I like the hot weather (as long as I’m not in a crowded bus or train), but then again I enjoy the occasional spot of low productivity.

Hsiang discusses Neidell and Zivin’s paper in detail here, including the killer graphs shown above.

Matt’s graph uses micro-data from the American Time Use Survey combined with interpolated daily weather station data while mine [Hsiang’s] uses total national production from UN national accounts combined with degree-day reconstructions from NCEP reanalysis, so they are completely different data sets utilizing completely different methods, but the results look extremely similar! On hot days, output in non-agricultural sectors drops and workers work less.

Furthermore, not only do the shape of the response-functions look similar, but the magnitudes of the responses are similar. . . . Matt and Josh found time worked fell by about 1.8% per 1 C in daily mean temperature [12.5%/((39.2 C-25.3 C)/2)] while I found that national output fell by about 2.9% per 1 C in daily mean temperature [10/(30.5-27)]. . . .

Hsiang concludes:

Reductions in worker output have never been included in economic models of future warming (see here and here) despite the fact that experiments fifty years ago showed that temperature has a strong impact on worker output (see here and here). In my dissertation I did some back-of-the-envelope estimates using the above numbers and found that productivity impacts alone might reduce per capita output by ~9% in 2080-2099 (in the absence of strong adaptation). This cost excedes the combined cost of all other projected economic losses combined.

10 thoughts on ““2% per degree Celsius . . . the magic number for how worker productivity responds to warm/hot temperatures”

  1. Very interesting results! I wonder if there are regional variations in this ‘magic number.’ Do workers in tropical climates, who may be used to high heat, respond in the same way as workers in more temperate climates? This could mitigate (or maybe exacerbate) losses in worker productivity – does it need to be folded in to economic models?

  2. Starting in the 1910s, Ellsworth Huntingon did a lot of work on worker productivity and climate. I don’t remember the specific temperature he arrived at as optimum, but I think I recall him saying that a lot of storms is good for mental activity.

  3. Has anyone looked at a possible increase in worker productivity during milder winters for line-men and others who work outside?

  4. “Over the course of the day that meant they accomplished 30 minutes less work…”

    That sounds like 6%, not 2% (unless Japanese workers work 24 hours a day…!)

  5. Pingback: More studies on the economic effects of climate change « Statistical Modeling, Causal Inference, and Social Science

  6. Scott,

    the paper by Neidell and Graff Zivin did look separately at the effect on workers who labored indoor vs outdoor. The effect is [unsurprisingly] strongest for those outdoors. The figure at the top is the response for this group of “high risk” workers.

    Francois,

    (1) Yes, the Dell et al. paper is closely related and we’ve spent time talking with that group trying to figure out how all these results are related. The central point of my paper was to actually show that structures in the basic response that they identify are consistent with a labor productivity mechanism.

    (2) We are looking into that right now in two other projects (unpublished). But in the PNAS paper, I see that the effect is driven almost entirely by the hot season (see the Table here) with not offsetting effect in the cooler seasons, although the Caribbean Basin doesn’t ever get that cold. In lab experiments, people do perform worse in cold temperature for many tasks, but it seems to be primarily related to lost dexterity (at moderately cold temps). However, in more natural settings, it seems like cold temps (eg. in the USA) don’t have such a strong effect. This is probably because people can keep themselves warm fairly easily (eg. putting on a sweater or going inside). Keeping oneself cool tends to be substantially harder and more expensive, although it is possible if people have access to AC. However, I am still concerned about this impact WRT climate change since in many medium and low income countries, like many of the ones in my PNAS sample, AC is not really a practical option for most people.

    Xi’an,

    I don’t think the 30 min correspond to the exact change in temperature that was cited, although I can’t be sure. I emailed Tanabe for the paper back when I saw this, but he said it was in Japanese. So I really don’t know the details of the study.

  7. Call me skeptical. Please explain how post-tsunami productivity can be taken as a laboratory for studying the air conditioning effect on productivity. There are other factors at play here like, I don’t know…a tsunami that certainly affected productivity. Also, what about the fact that many people take vacation during hotter times of the year reducing the total person hours? One week out of 52 for vacation during the summer amounts to something pretty close to the 2% number.

    • Yes, the tsnunami affected productivity, but did it affect productivity on days of different temperature in a way that could be confounded with a direct temperature effect?

      Anyway, I think the NYT article may be conflating two studies. The 2%/degree productivity loss appears to come from this call center experiment.

      The tsunami study might be this one, which appeared to look at self-reported summer productivity loss following the tsunami, and did not correlate this with temperature. In that case, your first criticism would be valid: productivity dropped after the tsunami, but why? (Your second criticism doesn’t apply directly because they’re looking at summer-only, not the whole year, although it’s possible that vacation-taking patterns changed after the tsunami.) But one could imagine in principle studying this more systematically, by measuring productivity as a function of temperature during the summer preceding and following the tsunami.

Comments are closed.