The American Taxpayer Relief Act (ATRA) of 2012 repealed tax cuts for the top income earners in the United States and added a new top tax bracket with income thresholds of $400,000 for single filers, $425,000 for heads of households, and $450,000 for married joint filers. Filing units in the new top tax bracket experienced an increase of the marginal tax rate from 35% to 39.6% on ordinary income, and from 15% to 20% on dividends and realized long-term gains.
As of 2015, pass-through businesses accounted for approximately 80% of firms, 55% of business income, and more than half of employment across all major industries in the United States. The unique corporate structure of pass-through businesses, known as S-corporations, allows for business income to be taxed as personal income rather than as corporate income. Changes to the personal tax rate an of S-corporation business owner is effectively a change on business’s income tax rate. The population affected by the ATRA tax rate increases contained a disproportionate number of business owners, with 57% having some pass-through business income.
Risch (2024) exploits the implementation of ATRA to study the relationship between the top marginal tax rate faced by business owners and employee earnings. The paper uses a difference-in-differences model and data from the universe of tax records linking firms to owners and workers to study changes in employees’ incomes. The paper finds that $1 increase in business income tax liability following ATRA corresponded to a 11-15 cents per dollar of new tax liability passed onto worker earnings. The tax burden was primarily borne by business owners, who bore approximately 80% of the burden. The workers’ share of the burden was entirely borne by employees in the top 30% of the income distribution, with essentially none of the burden borne by the lower two thirds of earners in affected firms. While workers in exposed firms saw lower relative earnings growth through 2016, firms did not shrink or become less productive. In addition, there was not evidence of skill downgrading, nor was their evidence of changes in revenues, suggesting the firms did not passed the tax burden into consumers through higher prices.
Risch (2024) takes these estimates to estimate the MVPF of increasing the top personal income tax rate in a pass-through system, considering the taxation of business owners and non-business-owners. This MVPF estimate considers the welfare effects of raising the top personal income tax rate generally, assuming no spillovers to lower bracket taxpayers associated with the taxation of non-business owners and accounting for the estimated pass-through to worker earnings associated with the taxation of business income.
The paper separately estimates the MVPF of taxing private business income and the MVPF of taxing business owners through the personal tax income code.
MVPF = 1.4
Risch (2024) defines the cost of taxing private income as mechanical revenue increase + the revenue externality from behavioral response of business owners + revenue externality from transfer.
For this MVPF, the paper takes the net cost calculation used to estimate the MVPF of taxing business owners through the personal tax income code and weights it by the share of top-bracket taxpayers that are business owners (30%): (0.7)(1-0.2584) + (0.3)(1-0.2584-[(0.23)(0.058)])=0.7376.
Risch (2024) defines willingness to pay to be a function of the mechanical welfare effect + the direct welfare effect of the transfer. In this case, the transfer is defined as: the pass-through of each dollar of new total tax liability of business owners to worker earnings. In other words, it focuses on the pass-through of new business and non-business income tax faced by pass-through owners when the top marginal income tax rate increases. This pass-through rate is lower than that used in the MVPF for taxing business income because business owners do not appear to pass through their new non-business income liability.
The willingness to pay by non-owners is 1 and the willingness to pay for owners is defined as it was for the MVPF of taxing business owners through the personal tax income code. Given that 70% of the population are non-owners and 30% are owners, the weighted willingness to pay is then: (0.7*1) + (0.3)*(1+[(0.35-0.23)(0.058)]) = 1.0021.
The MVPF estimate is then 1.0021/0.7376 = 1.36
Risch, Max (2024). “Does Taxing Business Owners Affect Employees? Evidence from a Change in the Top Marginal Tax Rate.” The Quarterly Journal of Economics, 139(1): 637-692. DOI: https://doi.org/10.1093/qje/qjad040