Insulation R-value chart

By Insulation Report Editorial. Last reviewed August 2026. How we research this.

Recommended attic R-value by climate zone

This chart is ENERGY STAR's published recommendation, from "Recommended Home Insulation R-Values", which states its guidance is based on Table R402.1.3 of the 2021 International Energy Conservation Code. ENERGY STAR groups some zones together because the values are identical; the chart expands them so every zone has its own row. The second column is for a bare or nearly bare attic. The third is the lower target ENERGY STAR gives an attic that already has 3 to 4 inches, which is the common real case: an older house that was insulated once and never topped up.

The geography column is orientation, not a lookup. Zones follow county lines, not state lines, and several states span three zones. The insulation depth checker resolves a ZIP to its zone using the Pacific Northwest National Laboratory county dataset for the 2021 IECC, which is the same dataset DOE points to.

Zone numbers also carry a moisture letter in the source dataset, A for moist, B for dry and C for marine, so a house near the Pacific coast sits in 4C while the mid-Atlantic sits in 4A. The letter matters for vapour control and material choice, not for the attic R target: ENERGY STAR's recommended levels depend on the number alone, which is why the chart above does not split them.

Two readings worth pausing on. First, R-49 or more is the recommendation for every zone except the southern tip of Florida and Hawaii, so "most of the country needs R-49 to R-60" is not an upsell, it is the published guidance. Second, the existing-insulation column is one full step lower, because the first inches of insulation do the most work. A house that already has some insulation faces a genuinely smaller job than a bare one, and a quote that prices both the same is not reading the attic.

R-value per inch by material

R per inch is carried here as a range, never a point value, because it varies with product, density and installed condition. A chart that gives one number per material is reporting a precision that does not exist. The last two columns convert each range into the inches needed to reach R-49 and R-60 from nothing, using the material's best and worst R per inch, so the pair brackets the honest answer.

Aged, settled values from published material data, carried as ranges in this site's reference data. For loose fill the inches columns are estimates rather than conversions, for the settling reason covered below. Spray foam is installed by a contractor to a specified thickness, so its inches columns describe the specification, not something a homeowner measures afterward.

Three notes the table cannot carry. Fibreglass batts only deliver their label when they are fitted: gaps and compression cost real R-value, and a batt jammed around a pipe underperforms the same batt cut to fit. Cellulose settles more than fibreglass, so installed depth should allow for it, which the bag label of the specific product accounts for. And a radiant barrier is deliberately absent from this table: it has no R per inch to list, for reasons covered further down. The material choices themselves, and where each one belongs, are the subject of the insulation types guide.

Recommendation, model code and enforced code are three different numbers

Nearly every R-value page collapses these into one claim, usually "code requires R-49" or "code requires R-60". There are three separate things, and for most readers only the first one applies.

The first two columns carry the same numbers, and that is not a coincidence: ENERGY STAR states its guidance is based on the model code table. What differs is force. The model code is what the International Code Council publishes; it binds nobody until a jurisdiction adopts it. The raised-heel allowance in the model code's Section R402.2.1 is worth knowing because contractors quote it both ways: Where the 2021 IECC calls for R-60 in the ceiling, installing R-49 over the whole ceiling satisfies it if the insulation carries over the wall top plate at full uncompressed height, which normally means a raised-heel truss.

The third column is where the common claim falls apart. State adoption runs from a voluntary code on a 1989 basis in Wyoming to the 2024 IECC in Rhode Island, and 8 states have no mandatory statewide residential energy code at all (basis: ACEEE State Policy Database, reviewed November 2024, a secondary source that goes stale between reviews). So a sentence like "code requires R-60 in your attic" is usually wrong twice: wrong about which edition applies, and wrong about whether any edition applies to a voluntary top-up in the first place. Energy codes apply to new construction and permitted work, not usually to voluntarily adding insulation to an attic you already have. Which edition applies where you live depends on your state and often your city, and your local building department is the only authority on it.

Why loose-fill R-value does not scale with depth

The materials chart above invites a simple multiplication: inches times R per inch. For batts and boards that works. For loose fill it drifts, because DOE documents that the settled density of loose-fill insulation rises as installed thickness rises. The deeper the pile, the more the lower material compresses under the weight above it, and each added inch buys slightly less than the inch before it.

One named product shows the effect in the manufacturer's own numbers. Owens Corning publishes coverage per bag for AttiCat blown-in fibreglass at several target R-values. Multiply each target R by its coverage and the product should be constant if R scaled cleanly with depth. It is not: it falls as the target rises.

Manufacturer coverage for the Owens Corning AttiCat 27.5 lb bag as published on the retail listing, read 2026-08-07. One product, used because its coverage table is public and checkable; the pattern is the general one DOE describes for loose fill.

What this means in practice. Reading depth off an attic floor and multiplying by R per inch gives an estimate, not a measurement, and the estimate flatters deep piles slightly. Going the other way, a calculator that divides a target R by a fixed R per inch understates the material needed at high targets. The bag label for the specific product is the authority in both directions, because the manufacturer has already folded settling into its coverage table. For reading the attic you have, the depth checker carries the ranges and the caveat built in.

The same settling is why an attic's history is not evidence of its present R-value. An invoice from 2003 saying R-38 was blown in describes the day it was installed. More than two decades of settling, a few trades walking through, and any moisture event since then all subtract, and none of them left paperwork. Measure the depth that is actually there and work from that, not from what the job was sold as.

What R-value does not capture

R-value measures one thing: resistance to conducted heat through the material, under test conditions. Four things that decide how an attic actually performs are outside the number entirely.

• Air leakage. Warm air moving through gaps around light fittings, plumbing stacks, the attic hatch and top plates carries heat past the insulation rather than through it. DOE's guidance on attics puts air sealing before insulating for this reason, and insulation laid over an unsealed ceiling plane underperforms its label, sometimes badly.
• Wind washing. At the eaves, air entering the soffit vents can wash through the edges of low-density insulation and strip its performance exactly where the depth is already thinnest. Baffles at the eaves exist to stop this, and their absence does not show up in any R-value.
• Moisture. Wet insulation conducts heat far better than dry, and loose-fill that has been soaked and dried usually stays degraded because the fibres mat. A roof leak can quietly delete the R-value of everything under it, and mould follows.
• Thermal bridging. Ceiling joists conduct heat better than the insulation between them, so a framed ceiling underperforms the label of the material in its bays until the insulation carries over the joist tops. This is one reason deep loose fill over the whole deck outperforms batts of nominally similar R fitted between joists.

A radiant barrier belongs in this list from the other direction: it addresses a heat flow R-value does not measure, and it has no R-value of its own. A radiant barrier reflects radiant heat rather than resisting conduction, so it has no useful R-value per inch and cannot be counted toward an R-value target. It is measured differently from bulk insulation and the two do not add together. Any product sheet quoting an "equivalent R" for foil is describing an assembly under favourable test conditions, not a property of the foil.

What your attic should target

Four steps, ending in a number.

1. Find your zone. Use the geography column in the first chart to orient yourself, or enter your ZIP in the depth checker for the county-resolved answer.
2. Measure what is there. Push a ruler through to the drywall in three or four places away from the hatch and use the typical reading. Ceiling joists are 5.5 to 9.25 inches deep depending on the framing (2x6 to 2x10 nominal), so visible joist tops mean you are under your joist depth, which is short in every zone.
3. Read your target. Bare deck or under about 3 inches: use the uninsulated column, R-60 in zones 4 to 8, R-49 in zones 2 and 3, R-30 in zone 1. Already at 3 to 4 inches or more: use the existing column, R-49, R-38 and R-25 for the same zones.
4. Convert the gap to inches with the materials chart, remembering the loose-fill caveat and the bag label.

Worked once: a zone 5 attic with 4 inches of settled blown fibreglass holds roughly R-9 to R-11. Its target is R-49, so the gap is about R-38 to R-40, which is roughly 10 to 12.5 inches of blown cellulose over the top. The old material stays put unless it is wet, mouldy, rodent contaminated or might be vermiculite; the remove-or-add-over guide runs that test properly.

One scope note: the targets on this page are attic targets. The model code sets separate, lower minimums for walls and floors because framing depth constrains them, and retrofitting a closed wall is a different job with different economics. This page owns the attic number, which is the one most homes are furthest from and the cheapest to fix.

What closing a gap like that costs, and how the price is built from wage data and material coverage rather than asserted, is on the insulation cost guide. Quotes from licensed local contractors are available there too, once you know the number you are buying.

You may not need more insulation

We do not sell insulation and we do not install it, which is why this section can exist. If your measurement lands within about R-10 of the recommendation, more depth is unlikely to be the best use of the money. Insulation has strongly diminishing returns: the first inches over a bare deck do most of the work, and the payback on the last few R-values is several times longer than on the first few. Going from R-11 to R-30 changes a heating bill; going from R-49 to R-60 is a rounding error on most of them.

The better spend at that point is almost always the ceiling plane itself. Air sealing the penetrations, the hatch and the top plates addresses the heat loss that R-value cannot see, usually costs less than a top-up, and makes whatever insulation is already there perform closer to its label. Hold off entirely if you are about to reroof, rewire or add ventilation, because work that disturbs the insulation wastes the money twice. And if the attic is spray foamed at the roofline as a conditioned space, floor depth is not the relevant measurement and these targets do not apply.

Related

• How much insulation do I have? The depth checker
• Insulation types, and which material belongs where
• What insulation costs, built from wage and coverage data
• Remove the old insulation or add over the top?