A Century of Unquestioned Logic
Daylight Saving Time has rested on one premise since Germany first adopted it in 1916: push sunset later and people turn on lights later, so the nation burns less fuel. Congress codified that logic in the Uniform Time Act of 1966 and extended DST by four weeks through the Energy Policy Act of 2005. A 2008 Department of Energy report estimated the extension alone saved 0.5% of national electricity.
Nobody tested it rigorously. Why? Testing required something rare: a population that switches into DST while a comparable population keeps its clocks unchanged, all within the same state, same weather, same economy, so researchers could isolate the clock change from every other variable that moves electricity demand. You can't isolate DST's effect without a control group. Indiana provided one by accident.
Indiana's Accidental Experiment
Until 2006, Indiana was a patchwork. Fifteen counties near Chicago and Louisville observed DST; seventy-seven did not. When the legislature mandated statewide DST that April, millions of households that had never changed their clocks began springing forward and falling back overnight.
Economists Matthew Kotchen and Laura Grant seized the opportunity, obtaining more than seven million monthly billing records from Duke Energy Indiana covering residential customers in southern counties that had previously refused DST. Counties already observing DST served as a built-in control group, enabling a difference-in-differences analysis that could strip away seasonal patterns, weather fluctuations, and economic trends with unusual precision.
What Seven Million Bills Revealed
Residential electricity consumption rose. Not fell. Rose by an average of 1% in newly-adopting counties relative to controls. For Indiana's residential sector, that meant approximately $9 million in additional electricity bills every year.
Spring showed a mixed picture, with reduced evening lighting demand roughly offsetting other changes, but late summer and fall told a different story entirely, as electricity use surged 2–4% above what it would have been without DST. Shifting clocks pushes peak afternoon heat into post-work hours when families are home running air conditioners at full capacity, and darker autumn mornings force furnaces on earlier. Benjamin Franklin imagined the lighting savings in 1784. He could not have anticipated that residential air conditioning would bury them under a mountain of heating and cooling demand two centuries later.
Kotchen and Grant estimated the environmental costs too: increased generation meant increased emissions of sulfur dioxide, nitrogen oxides, and particulate matter, producing $1.7 to $5.5 million per year in pollution damages for Indiana alone. DST was not merely failing at its stated purpose; it was actively increasing both household costs and environmental harm.
Replication Across Borders
If this were an Indiana-only quirk, it would be a footnote. It isn't. Pedro Hancevic and Diego Margulis examined Argentina's repeated adoption and abandonment of DST between 2007 and 2009, publishing their results in El Trimestre Económico (doi:10.20430/ete.v85i339.311). Using a regression discontinuity design on hourly load data, they found DST increased residential consumption by 0.4–0.6%. Different hemisphere, different continent, same direction.
Not every study agrees. Research on Norway and Sweden by Arne Willén and colleagues found a 1.3% reduction in electricity use during DST months, suggesting latitude matters enormously: at 60°N, summer evenings already stretch past 10 p.m., so the extra hour barely registers on cooling demand and lighting savings dominate. A boundary condition appears to exist somewhere in the temperate zone, roughly between 35°N and 50°N, where air conditioning prevalence is high enough to overwhelm any benefit from delayed sunset.
An Original Calculation
Kotchen and Grant's $9 million figure covers 2.8 million Indiana residential accounts. Scale it nationally and the numbers become staggering: approximately 140 million U.S. residential electricity accounts exist, consuming an average of 10,500 kWh annually at a mean price of 16.6 cents per kWh according to the EIA's 2024 data, which means a 1% increase implies $4.5 billion in excess residential electricity costs per year; halve that to account for climate variation and regions where effects are smaller, and the figure still exceeds $2 billion. Congress has never published a cost-benefit analysis of DST that accounts for air conditioning.
Strongest Counterargument
A 2008 DOE study of the four-week DST extension found a 0.5% electricity savings based on extrapolation from national aggregate load data, and Hendrik Wolff and Ryan Kellogg found no significant effect in Australia in a 2009 working paper. Both challenge Kotchen and Grant's direction of effect rather than merely its magnitude, but the DOE study relied on aggregate grid data that cannot separate residential from commercial demand and could not control for weather or economic activity as precisely as seven million individual household meter readings in a clean natural experiment allow. Aggregate extrapolations and simulations are not the same as granular empirical measurement. Still, the disagreement across studies is genuine, and the magnitude of DST's energy penalty likely depends on latitude, climate, and local building stock.
What We Didn't Prove
Southern Indiana has hot, humid summers and cold winters, which maximizes the heating-and-cooling penalty. States with milder climates or lower air conditioning penetration may see smaller or no increases. Only residential electricity was measured; commercial and industrial patterns differ. Behavioral adaptation over longer time horizons remains unaddressed: households might eventually adjust thermostat schedules, window treatments, or insulation in ways that blunt the initial energy penalty. And electricity is one component of total energy consumption; natural gas for heating and gasoline for evening recreation are not captured in billing data.
The Bottom Line
Seven million monthly billing records, analyzed through one of the cleanest natural experiments in energy economics, showed that DST increased residential electricity consumption by 1–4%, cost Indiana households $9 million per year, and generated millions in pollution damages. Partial replication in Argentina confirmed the direction of the effect. DST's rationale belongs to the era of incandescent bulbs and hand-cranked windows; in a grid powered by central air conditioning, heating and cooling costs dwarf lighting savings.
What You Can Do
If your state is debating permanent DST versus permanent standard time, the energy evidence favors standard time for temperate climates. Cite Kotchen and Grant when you write to legislators. At the household level, program cooling to ramp down during shifted evening hours when DST pushes peak heat into your post-work schedule. Seal windows before fall, when darker mornings send furnaces running earlier. Track your electricity bills across the March and November clock changes; if you see a bump, the fix is insulation and thermostat scheduling, not waiting for Congress to resolve a debate it has ducked for over a century.