3.3.3. Households versus Tax Units

In this paper, the terms ‘household’ or ‘tax unit’ are used similarly, one as being the substitute for the other. However, there is a difference between ‘tax unit’ and ‘household’. The tax-unit series was constructed by Piketty and Saez (2004, p. 4) according to the following definition. “A tax unit is defined as a married couple living together (with dependents) or a single adult (with dependents), as in the current tax law.” Due to the limited availability of data at the state level, the tax-unit series could not be identically constructed at the state level, hence, the resort to the household population as a proxy variable for tax units. The household series comes from the Bureau of the Census who defines the term ‘household’ as follows:

‘A household includes all the persons who occupy a housing unit. A housing unit is a house, an apartment, a mobile home, a group of rooms, or a single room that is occupied (…). The occupants may be a single family, one person living alone, two or more families living together, or any other group of related or unrelated persons who share living arrangements.’

Therefore, to believe that households approximate well tax units is a delicate question because more than one tax unit may be included in one household (e.g. spouses filling returns separately, students co-renting the same housing, etc.) In other words, the number of tax units is usually greater than the number of households. Whether income per tax unit or income per household is chosen, the latter exceeds income per tax unit (by 30 percent on average, according to Piketty and Saez, 2004, p. 30).

This is the reason why the number of households had to be adjusted upward so that the national aggregate of the time-series I collected fits Piketty and Saez’s series of tax units, leaving the initial state proportions intact. In other words, I calculated the percent share that state i represents in the U.S. number of households using BEA state data. Then, the same percentages (state / U.S.) were applied to Piketty and Saez’s denominators to get a new series by state. This is the most significant adjustment made for the N* variable.