In view of a compensation for the shortcomings of traditional carbon efficiency evaluation that tends to be based on absolute quantity indicators or a few single point evaluation indicators, relative quantity indicators are selected, with a regional carbon efficiency evaluation index system to be constructed from such multiple dimensions as economy, population, and welfare, taking into account regional development reality and fairness principles. Based on a combination of TOPSIS method with grey relational projection analysis, a dynamic comprehensive evaluation can be achieved of the carbon efficiency in 9 provinces (autonomous regions) of the Pan Pearl River Delta from 2011 to 2020 by utilizing time series average operator quadratic weighting. The results of static and dynamic evaluation research show that there are differences in the development level of carbon efficiency in the region. The overall level of carbon efficiency in Hainan Province is relatively high at 0.530, while the carbon efficiency in Guizhou Province is relatively low at 0.162. Hainan Province is 227% higher than Guizhou Province, with the carbon efficiency in Fujian Province on the rise, increasing by 52.3% from 0.327 in 2011 to 0.498 in 2020. It is therefore suggested that differentiated emission reduction strategies should be formulated, technology funding should be increased, social welfare should be enhanced, and talent introduction should be facilitated to achieve a low-carbon economic development..