AbstractMost analyses of the potential for combined heat and power (CHP) in parallel with a central electricity generating industry value the electrical output by means of an assessment of the marginal benefit to the central supply of units and capacity provided by the local system. This approach has defects, especially when applied to a growing system with a choice of investments. This paper suggests an alternative approach without these defects. It concludes that CHP is extremely attractive if load factors on the local plant are high, as in industry. This conclusion is insensitive to future costs of fuel and plant. If there were enough sites of this nature then there would be no case for CHP from district heating loads. If, however, high‐load‐factor sites are not available then CHP from district heating could be a second best alternative. the paper investigates this hypothesis and concludes that this may be so, but that in this case the answer is very sensitive to future costs of fuel and pl