Does Anybody Really Care What Time It Is?
Changing the clock alters energy use and economic activity only if it changes behavior. A new paper studies whether it does.
It’s March, and the magnolia trees say spring is arriving in California. Which means that the clocks in (most of) the country will “spring forward” by one hour. It’s happening this Saturday night. So, just as the bears emerge from hibernation, many time-shift haters are coming out of the woods arguing that we should live on daylight saving time (DST) year-round, while others say we should permanently be on standard time.
Originally DST was implemented in order to save energy, but the preponderance of research suggests that in most locations it doesn’t. Still, many argue for year-round DST in order to have more daylight in the evening hours, giving us more time for activities after work. Others argue for permanent standard time so it isn’t so dark in the mornings and to improve our health.
Similar cases are made in states that border a time zone boundary (or are bisected by one) and are debating which time zone their clocks should be set to. In fact, DST policies are, in essence, about how much of the year to spend following the clock of one time zone versus the adjacent one.
(Source)
Whether discussing the effect on energy consumption, retail sales, outdoor exercise, or hours of sleep, the policy debates about clock setting generally have one thing in common: they assume that moving the clock by an hour will move activities by that same amount. That is, next week, or shortly thereafter, they presume people will be getting up, eating meals, going to work, or taking a hike at the same clock time as usual even if it is an hour earlier in natural or “solar” time.
More than a decade ago, when two important papers found that DST doesn’t save energy, it raised the question of whether that is because decreased evening energy use was offset by morning increases, or because people respond to the clock shift by changing their schedules to stay in sync with solar time. Put differently, the numbering of hours on the clock is entirely arbitrary, so how much will it really change behavior if we change each number by one? Are social norms around clock times so powerful – 9-to-5 work days, eating lunch at 12 noon, going to bed at 10 PM – that changing clocks is all it takes to shift activity?
Analyzing this question around the shifts to and from DST is very difficult, because the weather and length of days are changing at the same time. But in a new paper that Patrick Baylis, Ed Rubin, and I are releasing today – “When we change the clock, does the clock change us?” – we use a different natural experiment to get traction on the question. People living at the same latitude in the same time zone, but east or west of one another, face permanently different relationships between clock time and solar time. Does that matter in their daily lives?
For instance, Grand Rapids, Michigan and Portsmouth, New Hampshire are both in the Eastern time zone and at almost exactly the same latitude, but the sun comes up an hour earlier in Portsmouth. The same distance further west at the same latitude is Valentine, Nebraska in the Central time zone. Are the schedules of Grand Rapidians more similar to Portsmouthers than to Valentinos, because the first two share a time zone?
To answer that question, we utilized three different data sources: information from Twitter on when people tweet in different locations, a 2000 census question that asked when the respondent leaves for work, and anonymized data on pedestrian traffic harvested from cell phones by SafeGraph. These are the same SafeGraph data that were used by reporters at the beginning of the pandemic to tell us how empty downtowns were and how much foot traffic at stores had plummeted. SafeGraph has made aggregated hourly data available to researchers for millions of businesses and other points of interest.
(Source)
If you want all the gory details of the analysis, you will have to download the paper, but here’s the Reader’s Digest version (a phrase that just lost half of our audience under 50). We analyze when individuals engage in these behaviors as a function of how far east or west they are in their time zone, controlling for latitude and the time of year. And we also control for how much people living in one location spend time in other time zones, a measure of “connectedness” with people who operate on different clock time.
What we find is that solar time does matter, but clock time matters more. Our estimates suggest that on average a person at the west end of a time zone tweets and goes to work about 20 minutes later (in clock time) than a person on the east end. That same one-hour difference in solar time on average causes people to visit the points of interest in the SafeGraph data about 10 minutes later. So, when the sun rises and sets an hour later, but with the same clock time, people are operating on a schedule that is 10-20 minutes later.
The Twitter data allow us to dig a bit deeper by utilizing keywords in the tweets. “Breakfast” gets tweeted nearly 30 minutes later on average at the west end of the time zone than at the east, but “lunch” and “dinner” only get tweeted about 5-10 minutes later.
We thought that the data might show a clear relationship between engaging in outdoor activities and more closely following solar time, but for the most part that wasn’t the case. There is a bit of evidence that people in rural counties follow solar time more than people in urban counties, but the data also suggest counties with more outdoor workers follow solar time less than counties with greater predominance of indoor workers.
The extra information in the foot traffic data also gives a mixed story. Visitors to medical offices, universities, and religious facilities on average follow clock time entirely, with no adjustment for solar time, while elementary and secondary schools shift by a bit more than 10 minutes.
(Source)
Visits to retail outlets and restaurants/bars are mostly comparable to our overall results, a 10-20 minute average response to solar time when it is an additional hour offset from clock time. Most surprising to us, parks, golf courses, and recreation centers also fall in that 10-20 minute range.
So, changing the clock does change the time of most activities, but less than one for one. That’s important for understanding how electricity demand is likely to shift with clock changes, particularly in comparison to production from solar plants (which don’t care about our clock). And it’s important for understanding how setting clocks affects commuting, work schedules, shopping, and other activities.
Our findings don’t argue for permanent daylight saving time or standard time, or for the switching we now do, but they do imply the behavioral differences are likely to be smaller than advocates of each policy suggest. Though one important caveat is that a biannual clock shift may have a different effect than a permanent one.
The core value of standardizing time is to coordinate activities across individuals, so it seems whichever system results in the most effective coordination with the lowest cost of implementation should be the frontrunner. I have written previously that I think the current system coordinates a shift in activity times as the seasons change that otherwise is likely to take place in a more chaotic fashion. Nonetheless, as we debate the value and drawbacks of each alternative, it’s important to recognize that shifting clocks is not the same as shifting behavior.
Suggested citation: Borenstein, Severin, “Does Anybody Really Care What Time It Is?”, Energy Institute Blog, UC Berkeley, March 6, 2023, https://energyathaas.wordpress.com/2023/03/06/does-anybody-really-care-what-time-it-is/
I’ve joined Mastodon and am now posting most days on energy topics @severinborenstein@econtwitter.net
Keep up with Energy Institute blog posts, research, and events on Twitter @energyathaas.
Categories
Severin Borenstein View All
Severin Borenstein is Professor of the Graduate School in the Economic Analysis and Policy Group at the Haas School of Business and Faculty Director of the Energy Institute at Haas. He received his A.B. from U.C. Berkeley and Ph.D. in Economics from M.I.T. His research focuses on the economics of renewable energy, economic policies for reducing greenhouse gases, and alternative models of retail electricity pricing. Borenstein is also a research associate of the National Bureau of Economic Research in Cambridge, MA. He served on the Board of Governors of the California Power Exchange from 1997 to 2003. During 1999-2000, he was a member of the California Attorney General's Gasoline Price Task Force. In 2012-13, he served on the Emissions Market Assessment Committee, which advised the California Air Resources Board on the operation of California’s Cap and Trade market for greenhouse gases. In 2014, he was appointed to the California Energy Commission’s Petroleum Market Advisory Committee, which he chaired from 2015 until the Committee was dissolved in 2017. From 2015-2020, he served on the Advisory Council of the Bay Area Air Quality Management District. Since 2019, he has been a member of the Governing Board of the California Independent System Operator.
Energy Institute Blog 
With utilities charging more for electricity between the hours of 4:00 pm and 9:00 pm, people with solar panels generate more electricity onto the grid during those hours when daylight savings time is observed. The one-hour shift can add 400 additional kilo watt hours per year on an 8,000-watt rooftop solar system being placed onto the grid during those critical hours when people are coming home, preparing meals and watching TV. With the utility charging 49 cents per kilo watt hour at that time from June1st until September 30th, that can add up to $200.00 more savings for owners of an 8,000-watt system and $300.00 more for owners of a 12,000-watt system. Even my winter output would be improved during those hours if daylight savings time was year around and could add another 150 kilo watt hours during the peak hours. Multiply that by the 1.4 million homes in California that have rooftop solar, and it really adds up.
Here’s an important benefit: the summer time crime rate falls with sunset shifted one hour later.
https://www.brookings.edu/blog/brookings-now/2015/10/29/fighting-crime-with-daylight-saving-time/
https://siepr.stanford.edu/publications/working-paper/under-cover-darkness-using-daylight-saving-time-measure-how-ambient
And summer time is when crime increases:
https://bjs.ojp.gov/library/publications/crime-seasonal
https://www.jstor.org/stable/43551937
Those benefits alone probably overwhelm the costs of any alternative. The current system works just fine because we are willing to forego outdoor activities in the winter in return for an earlier sunrise. The adjustment in the start and end of DST has fixed the primary flaw (including that DST used to start the weekend around my birthday!)
I hate ambiguity. It is the bane of us scientists. People always want certainty and science can only give them probabilities. You have given even less!
If the conclusion here is that there is no significant benefit to changing the clock, I would suggest that the benefit of not changing the clock is “not changing the clock”.
I’m retired, so I’m on solar time. Clock time is only important to me in the case of connecting with public transportation or with someone on clock time. Being a surfer, tide time is my most important clock.
I was going to write my representative to introduce a bill that would allow California to have no clock change, but my allotted time for this windmill has been taken up here.
Might it be that the change is driven by the desire to keep the same synching with TV programs that people watch in the morning and the evening? So, perhaps it is not so much the clock they look at but the viewing habits they have when eating breakfast and dinner?
Those proposing permanent daylight savings time must be young enough not to remember trying this in 1973 and how unpopular it was. It lasted for eight months. See: Google search terms “daylight savings time 1970s” for articles like these:
https://time.com/6157915/daylight-saving-time-history/
https://www.washingtonpost.com/history/2022/03/18/daylight-saving-seventies-history/
For those of us going to class or work that started at 8am we had to get up in the pitch dark and get to work as dawn was breaking. It was awful. Many schools adjusted their start times so students would not be walking to school in the dark. For whatever reason, humans like to get up at dawn or in the light. Earlier when clocks sprang forward in late April I would start to wake up an hour early in the morning by late March. Finally I would start to adjust my own watch one hour early in late March and work an hour earlier. Moving daylight savings time to earlier in March was wonderful for me personally.
Those proposing permanent daylight savings time should remember the lessons of history.
It would be a lot simple matter to just say that from x date to y date schools and businesses start an hour earlier. No messing with clocks. Businesses and other organizations, including schools, can all make their decisions. A problem with time changes, when not everyone makes the change, flight times appear to get an hour shorter or longe4!!
Your analysis needs to address two key economic issues – (1) what economic and social costs does the twice a year DST ritual impose on the population, and: (2) how do the costs compare to the benefits?