Nuclear Heat–Power Carbon Leverage

Ratio of direct gas-heat CO2 displacement to electricity CO2 displacement for equal reactor thermal energy

UK

3.93

against UK domestic supply mix

x (dimensionless ratio)

2025 · annual

Provisional source figure

Modelled backcast

Previous
3.77 2024
Change on the previous period
+4.4%
Base period
2005 = 100
Observations
63

The same period, against the other comparators

  • 1.59 against UK gas generation
  • 1.37 against UK non-renewable generation

US

1.45

against US non-CHP independent gas producers

x (dimensionless ratio)

2024 · annual

Modelled backcast

Previous
1.45 2023
Change on the previous period
-0.2%
Base period
2005 = 100
Observations
20

UK · domestic supply mix

Annual meanRange within the yearInterquartile range
22.533.54x (dimensionless ratio)2016201720182019202020212022202320242025

Series begins 2005 · 21 observations

Carbon leverage ratio against UK · domestic supply mix. Above 1.0 means the reference nuclear heat design displaces more CO2 as heat than the same thermal energy would as electricity.

What this figure cannot support

MODELLED REFERENCE-DESIGN BACKCAST. The reference design is held fixed while external electricity emissions change. It is not an observed history of commercial SMR operations and not a point-in-time backtest — current source vintages revise earlier periods.

How this index is calculated

Table view and CSV
YearMeanMinMaxP25P75n
20051.1961.1961.1961.1961.1961
20061.1341.1341.1341.1341.1341
20071.1571.1571.1571.1571.1571
20081.1711.1711.1711.1711.1711
20091.2881.2881.2881.2881.2881
20101.2661.2661.2661.2661.2661
20111.3231.3231.3231.3231.3231
20121.1991.1991.1991.1991.1991
20131.2871.2871.2871.2871.2871
20141.4391.4391.4391.4391.4391
20151.7141.7141.7141.7141.7141
20162.1372.1372.1372.1372.1371
20172.4012.4012.4012.4012.4011
20182.6212.6212.6212.6212.6211
20192.9042.9042.9042.9042.9041
20203.2473.2473.2473.2473.2471
20212.9012.9012.9012.9012.9011
20222.9812.9812.9812.9812.9811
20233.3073.3073.3073.3073.3071
20243.7693.7693.7693.7693.7691
20253.9343.9343.9343.9343.9341

Download the annual summary as CSV
Download every observation, with its lineage, as CSV

UK · gas generation

Annual meanRange within the yearInterquartile range
1.51.51.51.51.51.61.61.61.61.6x (dimensionless ratio)2016201720182019202020212022202320242025

Series begins 2005 · 21 observations

Carbon leverage ratio against UK · gas generation. Above 1.0 means the reference nuclear heat design displaces more CO2 as heat than the same thermal energy would as electricity.

What this figure cannot support

MODELLED REFERENCE-DESIGN BACKCAST. The reference design is held fixed while external electricity emissions change. It is not an observed history of commercial SMR operations and not a point-in-time backtest — current source vintages revise earlier periods.

How this index is calculated

Table view and CSV
YearMeanMinMaxP25P75n
20051.4381.4381.4381.4381.4381
20061.4261.4261.4261.4261.4261
20071.4691.4691.4691.4691.4691
20081.4731.4731.4731.4731.4731
20091.4621.4621.4621.4621.4621
20101.4801.4801.4801.4801.4801
20111.5191.5191.5191.5191.5191
20121.4961.4961.4961.4961.4961
20131.5131.5131.5131.5131.5131
20141.5001.5001.5001.5001.5001
20151.5151.5151.5151.5151.5151
20161.5331.5331.5331.5331.5331
20171.5061.5061.5061.5061.5061
20181.5301.5301.5301.5301.5301
20191.5431.5431.5431.5431.5431
20201.5461.5461.5461.5461.5461
20211.5361.5361.5361.5361.5361
20221.5211.5211.5211.5211.5211
20231.5311.5311.5311.5311.5311
20241.5381.5381.5381.5381.5381
20251.5851.5851.5851.5851.5851

Download the annual summary as CSV
Download every observation, with its lineage, as CSV

UK · non-renewable generation

Annual meanRange within the yearInterquartile range
1.21.21.31.31.41.4x (dimensionless ratio)2016201720182019202020212022202320242025

Series begins 2005 · 21 observations

Carbon leverage ratio against UK · non-renewable generation. Above 1.0 means the reference nuclear heat design displaces more CO2 as heat than the same thermal energy would as electricity.

What this figure cannot support

MODELLED REFERENCE-DESIGN BACKCAST. The reference design is held fixed while external electricity emissions change. It is not an observed history of commercial SMR operations and not a point-in-time backtest — current source vintages revise earlier periods.

How this index is calculated

Table view and CSV
YearMeanMinMaxP25P75n
20050.9050.9050.9050.9050.9051
20060.8700.8700.8700.8700.8701
20070.9230.9230.9230.9230.9231
20080.9530.9530.9530.9530.9531
20090.9740.9740.9740.9740.9741
20100.9820.9820.9820.9820.9821
20110.9570.9570.9570.9570.9571
20120.8380.8380.8380.8380.8381
20130.8490.8490.8490.8490.8491
20140.9000.9000.9000.9000.9001
20150.9450.9450.9450.9450.9451
20161.1771.1771.1771.1771.1771
20171.2151.2151.2151.2151.2151
20181.2611.2611.2611.2611.2611
20191.3631.3631.3631.3631.3631
20201.3621.3621.3621.3621.3621
20211.3301.3301.3301.3301.3301
20221.2961.2961.2961.2961.2961
20231.2911.2911.2911.2911.2911
20241.3001.3001.3001.3001.3001
20251.3741.3741.3741.3741.3741

Download the annual summary as CSV
Download every observation, with its lineage, as CSV

US · non-CHP IPP gas generation

Annual meanRange within the yearInterquartile range
1.41.41.41.41.41.41.51.5x (dimensionless ratio)2015201620172018201920202021202220232024

Series begins 2005 · 20 observations

Carbon leverage ratio against US · non-CHP IPP gas generation. Above 1.0 means the reference nuclear heat design displaces more CO2 as heat than the same thermal energy would as electricity.

What this figure cannot support

MODELLED REFERENCE-DESIGN BACKCAST. The reference design is held fixed while external electricity emissions change. It is not an observed history of commercial SMR operations and not a point-in-time backtest — current source vintages revise earlier periods.

How this index is calculated

Table view and CSV
YearMeanMinMaxP25P75n
20051.3571.3571.3571.3571.3571
20061.3651.3651.3651.3651.3651
20071.3721.3721.3721.3721.3721
20081.3791.3791.3791.3791.3791
20091.3981.3981.3981.3981.3981
20101.3981.3981.3981.3981.3981
20111.3941.3941.3941.3941.3941
20121.3991.3991.3991.3991.3991
20131.4051.4051.4051.4051.4051
20141.3971.3971.3971.3971.3971
20151.4051.4051.4051.4051.4051
20161.4001.4001.4001.4001.4001
20171.4121.4121.4121.4121.4121
20181.4281.4281.4281.4281.4281
20191.4421.4421.4421.4421.4421
20201.4481.4481.4481.4481.4481
20211.4541.4541.4541.4541.4541
20221.4541.4541.4541.4541.4541
20231.4511.4511.4511.4511.4511
20241.4471.4471.4471.4471.4471

Download the annual summary as CSV
Download every observation, with its lineage, as CSV

What was assumed

ParameterValueEvidenceBasis
nuclear heat delivery efficiency0.9assumedFrozen analyst reference design. Not a vendor performance guarantee, and not temperature-dependent in version 1.0.0.
nuclear electric efficiency0.33assumedFrozen analyst reference design.
boiler efficiency hhv0.85assumedReference gas boiler, higher heating value basis. The same value PIDI-E assumes, deliberately.
gas kg co2 per mmbtu52.91measuredEIA generic natural-gas coefficient, HHV, held constant for all markets and years. Applied to the UK as a scenario constant, NOT as a UK fuel assay.

Source attribution

Contains public sector information licensed under the Open Government Licence v3.0. Source: Department for Energy Security and Net Zero, DUKES 2026 table 5.14, accessed 12 September 2026. Gas coefficient: U.S. Energy Information Administration.

Source: U.S. Energy Information Administration, state electric power industry emissions and annual net generation (2024 vintage), and CO2 emissions coefficients, accessed 12 September 2026.

Definition

What it measures
Ratio of direct gas-heat CO2 displacement to electricity CO2 displacement, for equal reactor thermal energy
Headline metric
Leverage ratio (x); 2005=100 supporting series
Supporting metrics
Comparator sensitivities; component inputs; method version; historical reference series
Cadence
Annual US and UK; quarterly commentary retains latest annual observation
Panel
Nuclear Decarbonisation
Primary sources
eia-state-emissions;eia-state-generation;eia-gas-coefficient;desnz-dukes-514
Licence position
US edition: EIA public domain only. UK extension: free OGL attribution.
Release gate
Included dataset/code checks pass; source vintage, attribution, scenario assumptions and limits displayed

What this does not cover

US gas non-CHP independent-generator comparator (20 years) and UK domestic supply comparators (21 years). Not a marginal emissions, lifecycle abatement, site feasibility or observed SMR performance index. US edition uses only public domain; UK extension uses OGL.