With a growing concern of greenhouse gas (GHG) emissions due to climate change, many activities and efforts onthe greenhouse gas reduction have been implemented in solid waste sectors. Since recycling is the major managementoption for solid waste in Korea, it is important to estimate the reduction of the greenhouse gas emission during recyclingprocesses. In this study, two common methodologies, Prognos method of EU and waste reduction model (WARM) methodof USA, have been critically reviewed and compared to estimate the reduction for recycling of waste paper in terms ofsystem boundary, recycling processes, and emission factors. As a common point of two methodologies, the reductionfactors for the paper recycling have been developed by subtracting the recycled product emissions from the virgin productemissions to get the greenhouse gas savings. While the recycling losses and transportation are considered in twomethodology development, there are a number of differences between the methodologies in system boundary,transportation distance and forest carbon sequestration. As a result, it caused the difference in final greenhouse gasreduction factor of paper recycling. The reduction factor was −820kgCO2eq/ton in Prognos method, while −3,891kgCO2eq/ton was found in the WARM method. When both methods were applied to recycling of waste paper in Korea,the greenhouse gas reductions by the Prognos method and the WARM method were found to be 3,485.2tCO2eq/day and2,248.8tCO2eq/day, respectively. When the carbon sequestration by forest is considered in the WARM method, thereduction rate was estimated to be 16,538.3tCO2eq/day. The main reasons for such difference can be attributed to systemboundary and forest carbon sequestration. Especially, forest carbon sequestration can be an important factor in Korea thatusually manufactures papers from imported pulp from abroad. This study implies that the applications and results of bothmethods to estimate greenhouse gas reduction by waste recycling should carefully reviewed and acknowledged beforeuse due to the different assumptions and results that are anticipated.