The books vs eBooks calculator above answers a question that has no single correct answer, and it is honest about why. An e-reader carries a large one-off manufacturing footprint and then costs almost nothing per book. A printed book carries a small footprint each time, forever. Somewhere the two lines cross, and where they cross depends entirely on four numbers that different studies choose very differently.
Arb Digital builds free calculators that put those numbers in front of you as editable inputs rather than burying them in the code. Every factor on this page is yours to set, every default is labelled as a placeholder rather than dressed up as a fact, and the boundary — what is counted and what is not — is stated on the page instead of implied.
What This Books vs eBooks Calculator Does
You supply the e-reader's manufacturing footprint, the footprint of one printed book, how many books you read a year, how long the device lasts, what share of your print books you buy new, and the energy and grid factor for charging. The calculator then finds the crossover: the number of books at which the cumulative footprint of reading digitally equals the cumulative footprint of reading in print.
It reports that crossover in books and in years, then shows the totals over the device's whole life on both sides so you can see whether the break-even is reached before the device is replaced. That last check is the one that decides the real-world answer, and it is the one most comparisons skip.
A negative or unreachable break-even is reported as such. If your print factor is small enough — because most of your books are borrowed, for instance — the e-reader may never pay back within any plausible device life, and the tool says so rather than printing a number.
The Boundary: What Is Counted and What Is Not
This model counts three things: the one-off manufacturing footprint of the device, the electricity used to charge it over its life, and the production footprint of each printed book that is avoided. That is a deliberately narrow boundary, and it is narrower than a full life-cycle assessment.
Not counted on the device side: shipping the e-reader to you, its packaging, its end-of-life disposal or recycling, the data centres and networks that store and deliver the files, and any replacement of the battery or the device itself before the lifetime you entered. Not counted on the print side: shipping books from printer to warehouse to shop to home, the retail floor space, returns and pulping of unsold stock, and end-of-life disposal. Not counted on either side: your reading lamp, and the fact that a printed book can be read by several people in succession while an e-book licence usually cannot be passed on at all.
That last point is not a rounding error. A library copy read forty times amortises its production footprint forty ways, which is why the "share bought new" input exists and why lowering it moves the answer so sharply. Any comparison that treats every printed book as read once by one person is quietly assuming the least favourable case for print.
The Formula / How It's Calculated
Per book, reading digitally costs the annual charging emissions divided by the number of books read that year: per-book use = (kWh per year × grid factor) ÷ books per year. Reading in print costs the book's production footprint scaled by the share bought new: effective print cost = print factor × new share.
The saving per book is the difference between them, and the break-even is the device's manufacturing footprint divided by that saving: break-even books = device footprint ÷ (effective print cost − per-book use). Dividing again by books per year converts it to years.
Work through the placeholder defaults. Charging is 1 kWh a year at a grid factor of 0.4, so annual use emissions are 0.4 kg, spread across 12 books, giving 0.0333 kg per book. Each print book avoided is worth 1.2 kg at a 100% new share. The saving per book is 1.2 − 0.0333 = 1.1667 kg. Dividing the 100 kg device footprint by that gives a break-even of 85.7 books, which at 12 books a year is 7.14 years.
That result is the whole point of the exercise: with these particular placeholder inputs the device does not break even within a five-year life. Over five years the e-reader totals 100 + (0.4 × 5) = 102 kg, while 60 print books total 72 kg. Change the device footprint, the print factor or the reading rate and the conclusion can flip entirely — which is exactly why none of them is hard-coded here.
Why Published Break-Even Figures Disagree So Much
Estimates in the literature range from a handful of books to around a hundred, and the spread is driven by assumptions rather than by disagreement about physics. A 2025 comparison by the German consumer testing organisation Stiftung Warentest, reported by heise online, found that reading on a device already owned — a tablet used for other things — reaches the payback point after only a few books, while a dedicated e-reader bought purely for reading takes considerably longer. Notably, that analysis excluded data-centre emissions on the digital side, which is itself a boundary choice.
Four assumptions do most of the work. The device footprint is the first: a small e-ink reader and a large tablet are not the same object, and allocating a shared tablet's footprint across all its uses rather than to reading alone changes the answer by an order of magnitude. The book footprint is the second: a 200-page mass-market paperback and a 600-page hardback on coated stock are very different products.
Reading rate is the third, and it cuts both ways: a heavy reader reaches break-even quickly but also replaces devices more often. Device lifetime is the fourth and the most decisive. If the average owner replaces a reader every two or three years, a break-even at eighty-five books is never reached, and the digital route is worse however good it looks on paper. The University of Michigan's Center for Sustainable Systems published one of the earliest comparative life-cycle assessments of printed scholarly books against e-book reading devices, and the pattern it identified — a device that must be used heavily and kept long enough to amortise its manufacture — has held up in the studies since.
Where the Print Factor Comes From
The footprint of a printed book is mostly paper, and paper factors are published. The US Environmental Protection Agency's Waste Reduction Model exists to provide greenhouse gas, energy and economic factors for materials management pathways including source reduction, recycling and landfilling, and its documentation is a defensible starting point for the paper component. It will not, on its own, give you a per-book figure — you still need the book's weight, the printing and binding energy, and the ink.
The practical approach is to work from mass. Establish the book's weight, apply a paper production factor from a published dataset, then add printing and binding. That gives a number you can defend and explain, which is worth far more than a rounder figure with no provenance. If you are estimating page counts and book length in the first place, the words per page calculator is a useful companion.
The Comparison Nobody Runs: Reading Less New
Both columns in this calculator assume every book is newly produced for you. The third option is the one that dominates either — borrowing. A library copy read repeatedly divides its production footprint by the number of readers, and second-hand books add nothing to production at all beyond the transport of moving them.
The "share bought new" input is how you model that, and it is worth experimenting with. Set it to 25% and the effective per-book print cost drops to a quarter, which multiplies the break-even by four. Under the placeholder defaults that pushes the crossover well past any realistic device lifetime. This is not an argument for or against either format — it is an observation that the variable with the most leverage is not the one the comparison is usually framed around.
How This Fits With the Other Footprint Tools
This page uses the same approach as the rest of the footprint set: named factors, editable inputs, and a boundary stated in plain language. The crypto carbon footprint calculator works the same way from published network figures you choose, the flight carbon footprint calculator from a published factor and distance, and the food carbon footprint calculator from editable life-cycle factors by food group. The plastic footprint calculator covers the single-use side of consumption. If your reading is going digital in a different direction, the audiobook time calculator handles listening time rather than emissions.
Arb Digital designs and builds free interactive calculators that cite their sources, state their limits, and earn links because they are genuinely useful. Browse what we have already published, or tell us what your audience keeps searching for.
Browse the Free Tools Hub Talk to Arb DigitalCommon Mistakes to Avoid
- Ignoring device replacement. A break-even at eighty books is meaningless if the device is replaced after thirty. Compare the break-even against the lifetime, not against a wish.
- Assuming every print book is read once. Library and second-hand copies spread production across readers, which is the single largest lever in the comparison.
- Using a tablet's whole footprint for reading. If the device does five other jobs, allocating all of its manufacture to reading overstates the digital side badly.
- Comparing across different boundaries. One study counting shipping and data centres and another counting neither will disagree by more than the effect being measured.
- Treating a placeholder as a source. The defaults here exist so the page loads with a working example. Replace them before quoting the result anywhere.
Related Free Tools From Arb Digital
Continue with the crypto carbon footprint calculator, the flight carbon footprint calculator, the food carbon footprint calculator and the plastic footprint calculator for the rest of the footprint set, the words per page calculator for estimating book length, and the audiobook time calculator for listening rather than reading. Everything else is on the free online tools hub.
Frequently Asked Questions
There is no single answer. Published estimates range from a handful of books to around a hundred, depending on the device's manufacturing footprint, the type of book being replaced, the reading rate and how long the device is kept. This calculator lets you set all four and see the crossover for your own assumptions.
No. An e-reader only pays back its manufacturing footprint if it is used heavily and kept long enough. A device replaced after two or three years by a light reader may never reach break-even, and borrowed or second-hand print books can beat both options.
Shipping and packaging on both sides, retail floor space, returns and pulping of unsold books, data centres and networks, end-of-life disposal, and the fact that a printed book can be read by several people while an e-book licence usually cannot be passed on.
Because grid emissions per kilowatt-hour differ enormously between countries and change every year. Hard-coding one would make the result wrong for most readers, so the tool takes the current figure from your own grid operator or national inventory.
Substantially. A library copy read many times divides its production footprint among all its readers. Lowering the share bought new reduces the effective per-book print cost proportionally, which lengthens the e-reader's payback by the same proportion.
They are placeholders chosen so the page loads with a working example, and they are labelled as such on the form. They are not sourced figures and should be replaced with numbers from a manufacturer disclosure or a published life-cycle study before the result is quoted.
Then the manufacturing footprint should be allocated across everything that device does, not charged entirely to reading. Enter a proportionate share in the device footprint box, and the break-even falls sharply.
This page is an estimating tool. Every emissions factor is an input you supply, the defaults are placeholders rather than sourced values, and the boundary is narrower than a full life-cycle assessment; results should not be quoted as a measurement of any specific product.
