The Carbon Math of Hybrid Work: Commuting, Office Energy, the Case for a Third Option

Your company probably has an ESG page. It probably mentions Scope 3 emissions. And it almost certainly doesn’t account for the carbon cost of making 200 people drive to a half-empty building three days a week. 

The return-to-office debate has been argued on the grounds of culture, productivity, and real estate sunk costs. What it hasn’t been argued on — at least not rigorously — is emissions. That’s a problem, because the math here isn’t intuitive. Fewer commutes don’t automatically mean less carbon. Half-occupied offices can be worse per person than full ones. And working from home carries its own energy penalty that rarely shows up in the sustainability deck. 

This piece lays out what the research actually says, where the real savings are, and where the assumptions break down. If you’re a facilities lead, an HR director building a hybrid policy, or a sustainability officer trying to reconcile your company’s climate commitments with its workplace strategy, the numbers below should change how you think about all three. 

Key Takeaways 

• A typical hybrid worker’s carbon footprint is roughly 40% lower than a full-time office worker’s, but only at two or more remote days per week. One day barely registers. 

• Unused office space in the U.S. generates an estimated 22 million metric tons of CO2 annually. Half-empty buildings don’t use half the energy. 

• Fully remote work cuts emissions by up to 54%, but rebound effects (more errands, home heating, suburban migration) eat into the gains. 

• Coworking spaces produce an estimated 30% less carbon per employee than traditional low-density offices, largely because shared spaces run at higher occupancy. 

• The lowest-carbon configuration isn’t “everyone at home” or “everyone in the office.” It’s a right-sized portfolio: a smaller anchor office, remote days, and local flex workspaces filling the gaps. 

2.7 Tonnes per Desk: What Commuting Actually Costs 

Transportation accounts for 28% of total U.S. greenhouse gas emissions [1]. The majority comes from personal vehicles, and commuting is one of the largest single contributors. 

The average American round-trip commute runs about 37 miles, and roughly 69% of workers drive alone [2]. At approximately 400 grams of CO2 per mile, that’s 2.7 tonnes of carbon dioxide per person, per year, just getting to and from a desk. Scale that across a 500-person company where 80% drive solo, and you’re north of 1,540 tonnes annually from commuting alone — before a single light switch gets flipped. 

The COVID lockdowns proved the math directionally: global emissions dropped 7% in 2020, driven largely by the collapse in travel, then bounced back the moment people started driving again. A 2024 study published in Nature Cities [3] modeled what a sustained 10% increase in remote work could deliver: 192 million metric tons of CO2 cut from the U.S. transportation sector each year. 

But commuting is only half of the ledger. 

The Building Doesn’t Know You Left 

Commercial buildings consume about 17% of total U.S. energy [4]. Factor in indirect emissions from electricity generation, and the building sector accounts for roughly 30% of the country’s greenhouse gases [5]. 

The problem for hybrid work isn’t how much energy buildings use when full. It’s how much they use when empty. 

The Empire State Building offered a stark illustration during the pandemic: with the building nearly vacant, electricity consumption fell only 28% [6]. At UC Davis, 90% of occupants left campus, but electricity dropped by just 15% [7]. HVAC systems, base lighting, elevators, server rooms — these don’t scale down proportionally when people stop showing up. Nearly a quarter of a typical commercial building’s electricity is consumed when no one is inside at all [7]. 

That creates a perverse outcome for hybrid policies that reduce attendance without reducing space. U.S. office vacancy hit a record 19.7% by early 2025 [8]. An estimated 40% of all corporate office space is paid for but sits unused on any given day, generating roughly 22 million metric tons of CO2 and wasting about 32 billion kilowatt-hours of electricity annually [9]. 

A Cornell and Microsoft joint study found that reducing building attendance from 50% to 10% can actually double the carbon footprint per onsite worker [10]. The building’s fixed energy load gets divided among fewer people, and the per-capita math inverts. 

So if your hybrid policy cut attendance in half but kept the same square footage, you may have improved commuting emissions while making your building’s per-person footprint worse. The net result depends entirely on whether you right-sized the space. 

Two Days Changes the Equation. One Day Doesn’t. 

Fully remote workers showed a 54% lower carbon footprint compared to full-time onsite workers. That’s the ceiling. The floor is more interesting: hybrid workers who stayed home just one day per week reduced emissions by only about 2%, with the commuting savings almost entirely offset by increased non-commute travel, higher home energy use, and the tendency of hybrid workers to live farther from the office. 

The meaningful reductions started at two days. A two-to-four-day remote schedule cut carbon footprints by 11% to 29%, depending on specific patterns and whether the employer implemented seat-sharing — which alone could reduce emissions by 28% under full building attendance, because it lets companies shrink their physical footprint without reducing capacity. 

A separate activity-based study published in Applied Energy [11] quantified the daily trade-offs more granularly: switching to work-from-home eliminated about 4.23 kg of CO2 equivalent per person per day from commuting and 1.34 kg from reduced workplace energy, but residential emissions rose by 1.51 kg. Net daily savings came to roughly 4 kg per person — meaningful, but not the full commuting reduction that gets cited in most advocacy pieces. 

The consensus across multiple 2024 and 2025 analyses [10, 12, 13] converges on a roughly 40% reduction for hybrid workers on a two-to-three-day remote schedule. Below that threshold, the gains erode fast.

The Gains Aren’t as Clean as the Headlines 

Several rebound effects chip away at the theoretical savings, and most corporate sustainability reports don’t account for them. 

Non-commute travel climbs. When people skip the office, they don’t sit motionless at home. The Cornell-Microsoft study flagged non-commute travel as a significant and underappreciated offset. Remote workers run more midday errands, make more recreational trips, and generally drive more outside of the commuting window. For one-day-a-week hybrid workers, this rebound nearly zeroes out the commuting savings. 

Home energy isn’t free. The IEA documented residential electricity increases of 15% in the UK and 20–30% in parts of the U.S. during lockdowns. Heating or cooling a detached suburban house for one person is substantially less efficient, per occupant, than climate-controlling a shared office floor for fifty. If your remote employees live in older, poorly insulated housing stock, the home-energy penalty is even steeper. 

Hybrid workers tend to live farther out. This is the suburban migration effect. When you only commute two or three days a week, a longer drive feels tolerable, so people move to cheaper housing in outer suburbs or exurbs. The per-trip emissions go up even as the number of trips goes down, and this can also increase non-work travel distances over time as daily errands shift to car-dependent suburban patterns. 

Digital infrastructure has a cost. An hour-long HD video call generates between 150 and 1,000 grams of CO2, depending on the local electrical grid. Data centers consumed roughly 460 terawatt-hours of electricity in 2022 — about 1.5% of global consumption — and that figure is projected to more than double by 2030, according to the World Economic Forum [12]. For any individual worker, this is a rounding error. At enterprise scale across thousands of employees on video eight hours a day, it starts to register. 

Public transit takes a hit. A Nature Cities and MIT analysis estimated that a 10% shift to remote work could cost U.S. transit systems $3.7 billion per year in fare revenue — a 27% drop [3]. Less funding means less service lines, which means fewer people use transit, which means more cars. The irony: remote work policies designed in part to reduce emissions could undermine the low-carbon infrastructure that serves the workers who still need to commute. 

None of these factors erase the net benefit of hybrid work. But they do mean the real reduction is smaller than the simple “fewer commutes, less carbon” framing suggests. A 40% reduction is a more honest estimate than 54%, once the offsets are priced in. 

Shared Space, Shared Load 

The core carbon problem with hybrid offices is utilization. A building designed for 300 people that hosts 140 on an average Tuesday is running its HVAC, lighting, and base systems at roughly the same rate it would for 280. The energy cost per person is inflated because the total number of people dropped but the energy usage barely moved. 

Coworking spaces invert that dynamic. Because they serve multiple organizations under one roof, they consolidate demand. A single coworking location replaces fragments of underused space across ten or twenty companies, running at higher occupancy with lower per-person energy intensity. Industry data from DropDesk estimates coworking spaces produce about 30% less carbon per employee than traditional low-density corporate campuses.

The operational side reinforces the structural advantage. About 55% of coworking spaces now use smart sensors that adjust lighting and HVAC in real time based on occupancy, according to DropDesk’s 2026 industry survey. That sensor-driven responsiveness cuts energy waste by roughly 20% — a capability most conventional corporate offices haven’t retrofitted for, especially older Class B and C buildings.

By Q4 2025, U.S. coworking inventory reached approximately 8,900 locations spanning 159 million square feet, per CoworkingCafe. That’s up from about 5,600 locations in early 2023 — a growth curve that tracks almost exactly with hybrid work adoption. The infrastructure for a lower-carbon “third place” between home and HQ already exists at scale, and its growing quarter over quarter. 

The commute-distance argument strengthens the case further. If your employees are scattered across a metro area, a hub-and-spoke model that replaces a single downtown headquarters with a smaller anchor office plus coworking memberships near where people actually live can cut average commute distances dramatically. In the top 25 U.S. metros, CoworkingCafe lists thousands of locations across urban, suburban, and exurban zones.

Search CoworkingCafe for coworking spaces near your team’s zip codes. Filter by day pass availability if you’re testing the model before committing to memberships. 

A Practical Carbon Framework for Workplace Decisions 

Here’s how to apply the research to actual portfolio and policy decisions. The goal isn’t to be perfect, but to move away from a configuration that ignores emissions and toward one that accounts for the biggest variables. 

Step 1: Measure what you’re actually using. Pull badge-swipe or access-control data for the last 90 days, and calculate your average daily occupancy as a percentage of total capacity. If it’s below 60%, your building’s per-person carbon footprint is meaningfully inflated by the energy overhead of space nobody occupies. 

Step 2: Model the commuting baseline. Survey your workforce on commute distance and mode of transportation. The national average is about 41 miles round-trip by car, producing roughly 16 kg of CO2 per day for a solo driver. Your actual number will vary by metro, transit access, and remote-work eligibility. Even a rough estimate reveals how much of your Scope 3 footprint sits in employee commutes. 

Step 3: Set your hybrid floor at two remote days. One remote day per week delivers negligible carbon benefit after rebound effects, but two days is where the curve bends. Three days captures most of the commuting reduction without fully sacrificing in-person collaboration. If your policy is anchored at one mandated remote day as a perk, know that it’s doing almost nothing for your emissions profile. 

Step 4: Right-size the space or share it. Reducing attendance without reducing square footage just redistributes the carbon cost from commuting to building operations. Consider implementing seat sharing, sublease unused floors, and consolidating into fewer, higher-density locations. Every square foot you shed that was running HVAC for empty chairs is a direct emissions cut. 

Step 5: Add local flex nodes. For employees whose commute to headquarters exceeds 30 minutes by car, a coworking membership closer to home reduces per-trip emissions while preserving the structure and social interaction that full-time remote work sacrifices. Compare providers on CoworkingCafe — filter by location, amenities, and pricing to find spaces that match your team’s actual work patterns. 

FAQ 

Does working from home produce zero emissions? No. Residential energy use rises measurably when people work from home — by 15–30% depending on the region, per IEA data. Heating and cooling a single-occupant home during work hours is less efficient per person than a shared office. The net carbon savings of remote work are real but smaller than the commuting reduction alone would suggest. 

Is a one-day-a-week hybrid policy worth it for emissions? Barely. The Cornell-Microsoft study found only a 2% reduction for one-day-a-week hybrid workers, after accounting for rebound travel and home energy use. The inflection point is two remote days per week. Below that, the offsets eat the gains. 

How do coworking spaces reduce per-person emissions? Higher utilization. A coworking space serving multiple companies runs its HVAC and lighting for roughly the same energy cost as a single-tenant office at half it’s occupancy, but the per-person share is far lower. Sensor-driven climate controls, common in newer flex spaces, reduce waste further. The structural advantage is consolidation: shared space means less total space running at partial load. 

What about the carbon cost of video calls and cloud computing? It’s real but comparatively small at the individual level. An hour of HD video generates 150–1,000 grams of CO2 depending on the grid. At enterprise scale the numbers add up though. For any individual worker, commuting and building energy dwarf digital infrastructure emissions by orders of magnitude. 

Should our company factor carbon into RTO decisions? If your company has public climate commitments, employee commuting is almost certainly your largest unmanaged Scope 3 category. Ignoring it while mandating return-to-office creates a gap between your sustainability reporting and your workplace policy that investors, employees, and regulators will eventually notice. The fix isn’t necessarily “don’t RTO.” It’s “RTO into the right-sized space, with flex alternatives for the trips that don’t need to happen.” 

Sources

1. U.S. Environmental Protection Agency: Transportation Sector Emissions 
2. U.S. Census Bureau: 2024 American Community Survey: Commuting Data 
3. Nature Cities (2024): Impacts of remote work on vehicle miles traveled and transit ridership in the US
4. U.S. Energy Information Administration: How much energy is consumed in U.S. buildings?
5. Center for Climate and Energy Solutions: Decarbonizing U.S. Buildings 
6. Alliance to Save Energy: Saving Energy in Buildings When Nobody is in Them 
7. UC Davis Energy and Efficiency Institute: Energy Waste in Empty Buildings 
8. CommercialEdge (March 2025): National Office Vacancy at 19.7% 
9. Density.io: How Companies Can Take Climate Action with Occupancy Data 
10. Cornell University & Microsoft (PNAS, September 2023): Climate Mitigation Potentials of Teleworking 
11. Applied Energy (2023): Greenhouse Gas Emissions Under Work From Home vs. Office 
12. World Economic Forum (2025): The AI-energy nexus will determine AI’s impact. We must account for it better 
13. Grist (May 2024): The Carbon Cost of Return-to-Office Mandates