Spring Forward: Making Sense of New Research on Daylight-Saving Time
This Sunday, most people in the United States will once again set their clocks ahead to daylight-saving time. Yes, spring is about to arrive, even though signs of spring are not yet noticeable everywhere in the country. This means more daylight exposure — at least for those who follow conventional sleeping routines. But it also means the mini jetlag that comes with the clock change may disrupt your sleeping cycle.
With yet another mini jetlag approaching, it is an apt time to consider why we have daylight-saving time and whether there is a good reason to keep it, particularly given recent political momentum to scrap it. Throughout its history, daylight-saving time has produced both intended and unintended consequences. Based on recent research (by myself and others), it is unclear whether daylight-saving time continues to serve its intended purpose.
Energy conservation has long been the main justification for having daylight-saving time. During World War I, the Germans and their allies were the first to use daylight-saving time on a wide scale, in order to conserve energy resources needed for the war effort. Britain and its allies, including the United States, soon followed. After the war ended, many countries abandoned daylight-saving only to resume it when the need to conserve energy arose again during World War II. Unsurprisingly, countries, including the United States, turned to daylight-saving time once again to conserve energy during the energy crises of the 1970s.
Today, daylight-saving time is largely confined to the West (Canada, Europe, the United States, and parts of Australia) as can be seen in the figure below. Few countries in Africa, Asia, Latin America, and the Caribbean use it. Clearly, the closer a location is to the equator the less reasonable a time change would be, simply because sunrise and sunset times do not vary enough across the year.
Although daylight-saving time’s primary purpose has been to conserve energy, its use does not conserve all energy equally. Historically, the main goal of daylight-saving time has been to reduce use of electricity by substituting electric lighting with natural light as much as possible. But more efficient lighting alternatives[1] have made lighting a smaller share of total energy consumption, which means savings from reducing electric lighting will be less notable today and in the future.
In addition, work schedules have become more flexible for a larger share of the population. These flexible schedules give people the ability to start the day an hour earlier or later, potentially neutralizing time-change effects. Most of the population did not have this flexibility back in the 1970s when a larger share of the population worked in manufacturing or strict 9-to-5 jobs. Back then, behavior needed to change in response to daylight-saving time because of the rigidity of working hours. In the flexwork-age, however, daylight-saving time’s effect on behavior may be greatly diminished. (Of course, this argument does not apply to those with kids attending school, regardless of flexible work hours.) A recent working paper by Baylis et al. (2023) empirically confirmed that changing the clock does change the time of most activities, consistent with the intent of changing the clocks twice a year, but the effect is less than one for one, meaning people are not shifting their behavior a full hour with the time change.
Consistent with the observations above, the socio-economic literature provides little evidence that daylight-saving time actually saves energy. Recently, my research collaborators and I found that daylight-saving time in the Balearic Islands in Spain, a region with a large tourism industry, actually slightly increases the demand for electricity, but in most cases the effect is not significantly different from zero (Graf et al., 2023). This finding that daylight-saving time does not lead to expected energy conservation is in line with other studies on the topic (see e.g., Kellogg and Wolff, 2008, Kotchen & Grant, 2011; Verdejo et al., 2016). The consensus seems to be that the sign and magnitude of the effect depends very much on the region studied. Havranek et al. (2018) found in a meta-study that studies in locations closer to the equator tend to show smaller savings and that more recent studies tend to find smaller effects of energy savings. A potential explanation for this finding may be improved data availability and better methods to estimate the effect of daylight-saving time on energy consumption.
There are also studies suggesting that the transition period, i.e., the disruption to sleep patterns associated with the clock change, causes additional traffic accidents, reduces self-reported well-being, and produces other adverse effects (for references in that space, see e.g., Rivers, 2018 and Gibson and Shrader, 2018).
To summarize, there is no compelling scientific evidence that daylight-saving time conserves energy, but there is compelling recent evidence that the effect is smaller than previously thought and very much dependent on local circumstances.
The public also dislikes changing the clock twice a year — and this includes not only data nerds like me who frequently work with time-series data, but also most voters, as recently seen in California.[2] Politicians have taken note. In the United States, the Senate passed a bill in early 2022 to make daylight-saving time permanent, but the bill stalled in the House. The Senate reintroduced the bill earlier this year. Similarly the EU Transport and Tourism Committee voted to abolish daylight-saving time permanently in the European Union, but it also has yet to happen.
Moving forward, Policy Integrity’s recommended approach is to comprehensively weigh the costs and benefits of the daylight-saving time policy against each other. The cost of changing the system should be considered, as well as any behavioral changes that would occur in response. Furthermore, some states (Arizona and Hawaii) do not currently change their clocks twice a year, and others (e.g., Florida) have passed legislation that would follow suit if federal law allowed. But the latest research has shown that a standardized time (e.g., within a time zone or closely economically related jurisdictions) is a central device for coordinating (economic) activities and behavior (Baylis et al., 2023). So there may be more benefit if, for example, Arizona were in the same time zone year round with its economically linked neighbors rather than switching twice a year. Accordingly, the economics of coordination should also be given careful attention, regardless of whether we move to permanent daylight-saving time, standard time, or the current combination that we have.
By Christoph Graf.
[1] According to EIA, nearly half of U.S. households use light-emitting diode (LED) bulbs for all or most of their indoor lighting in 2020. In 2015 this estimate was only 4% (EIA). ENERGY STAR rated LED’s use at least 75% less energy, and last up to 25 times longer, than incandescent lighting (Energy.gov).
[2] 60% of voters in a 2018 election in California supported allowing the California State Legislature to establish permanent year-round DST if allowed by federal law (Ballotpedia).
