This Replication folder generates Figures 1, 2, 7, 8, 9, and 10 (i.e. the primary results) from "Accounting for Factorless Income" by Loukas Karabarbounis and Brent Neiman, NBER Macroeconomics Annual, 2018. Please cite accordingly.

Start by running "Macroannual_DataGenerator.do", which will import, clean, and merge data from four primary sources: (i) Kara McDaniel's Tax Tables, (ii) US NIPA Fixed Asset Tables, (iii) US NIPA Annual GDP Tables, and (iv) the FRB St. Louis database FRED. 

This .do file will produce a panel dataset called KN_MA_Input_Data, which is then saved in a subfolder called "Output_Files". 

Note that the NIPA series have been downloaded as flatfiles (stored in "Input_Files") and so require the codebooks TablesRegister.txt and SeriesRegister.txt to locate and retrieve specific series. 

Following the paper, we decompose the economy into three major sectors: Private Non-residential, Residential, and Government. 

The script imports and cleans variables related to value added, labor, and capital/investment for all three sectors. The capital/investment variables are decomposed into different fixed asset types, which we aggregate into ICT, non-ICT, and housing capital. We calculate ICT and non-ICT capital price growth as the (average, corresponding to a Tornqvist index) current-dollar share weighted price growth of more detailed fixed asset types within each category.

Next, generate the figures by running "Macroannual_Analysis.do", which takes KN_MA_Input_Data as an input and outputs the figures as .eps and .pdf files to the folder "Output_Files".