Young Workers’ Payroll Tax Cut in Sweden
Introduction
Saez, Schoefer, and Seim (2019) evaluate the impact of a 2007 payroll tax cut in Sweden on wages, employment, profits, and firm growth. Before the reduction in the payroll tax, Sweden had a flat payroll tax rate of 31.4 percent, paid by employers on the total payroll of the firm. In 2007, a newly elected government cut the payroll tax rate on young workers (defined as those between 19-25 years old) to 21.3 percent. In 2009, the government further cut the payroll rate on young workers to 15.5 percent and extended eligibility to the 19-26 age bracket. The stated goal of the policy was to curb the high youth unemployment rates observed at the time. Saez, Schoefer, and Seim (2019) use a population-wide administrative dataset linking employees to employers and a dataset containing firm-level accounting data to estimate the policy’s impact. To identify a causal impact, they exploit the fact that firms that employed a high share of young workers before the tax cut benefitted more from the policy change than firms that had a low share of young workers. They find that sales, profits, capital assets, wages and employment all grow faster after the reform in firms that were more exposed.
Paradisi (2021) uses the results in SSS to compute the MVPF of these payroll tax cuts to young workers in Sweden.
MVPF = 1.0
Net Cost
To compute net costs to the government, the Paradisi (2021) calibrates three cost components: i) mechanical cost of the tax cut, ii) fiscal externality due to increased employment and wages, which increase payroll and profit tax revenues, and iii) fiscal externality due to increased firm profits, which increases corporate tax revenue. To calibrate component i), Paradisi considers that a payroll tax cut mechanically reduces government revenue by an amount equal to the pre-period wage bill times the change in the tax rate. However, lower payroll taxes reduce firm costs, which mechanically increases firms’ profits and therefore government revenue through the profit tax. The total mechanical impact of a 1% payroll tax cut on young workers is then 1-26.2% times the average pre-period young workers’ wage bill (€31,460) times the reduction in the tax rate (1%), where 26.2% is the average corporate tax rate. This gives 0.78 x 31,460 x 0.01 = €232.
To calibrate component ii), the author multiplies the average firm-level change in the wage bill (due to increases in both employment and wages) caused by a 1% tax cut (€4.42) by the payroll and profit tax rates (40%), for a total of 0.4 x 4.42 = €1.77. To calibrate component iii), Paradisi (2021) multiplies the policy impact on average profits (€2.9) by the average corporate tax rate among affected firms in Sweden at the time of the reform (26.2%), for a total of 0.262 x 2.9 = €0.75. Net costs to the government are therefore 232-1.77-0.75=€229.
Willingness to Pay
To compute the WTP, Paradisi considers three terms: i) the WTP of workers employed in firms directly affected by the policy; ii) the WTP of newly employed workers for their increase in after-tax income, and iii) the WTP of firm owners and manager for the increase in their income due to increased firm profits. To estimate component i) Paradisi notes that incumbent employees benefit from the tax change because their net wages increase, and therefore their post-tax income increases. The WTP is given by the product of the average wage change due to a 1% payroll tax cut (€2.5) and the net of tax rate (92%), for a total of 0.92 x 2.5=€2.29. To calibrate component ii), the utility benefit of previously unemployed workers, Paradisi multiplies the average change in employment caused by the policy change (1.93) by the average net-of-tax wages of workers in affected firms (92%). Finally, to compute component iii) Paradisi assumes that producer surplus is measured by profits, which are proportional to the fall in labor costs caused by tax cut. The author then multiplies the average change in profits caused by the policy (€314.6) by the net-of-profit-tax rate (92%), arriving at a producer surplus of €232. The sum of the utility impacts on i) to incumbent workers, ii) outside workers, and iii) producers, gives the total WTP for the policy change 2.29+1.78+232 = €236.
MVPF Summary
Dividing WTP by the net cost to the government, Paradisi (2021) arrives at an MVPF of 236/229 = 1.03.
References
Saez, Emmanuel, Benjamin Schoefer, and David Seim (2019). Payroll Taxes, Firm Behavior, and Rent Sharing: Evidence from a Young Workers’ Tax Cut in Sweden. American Economic Review, 109(5), 1717-63.
https://www.aeaweb.org/articles?id=10.1257/aer.20171937
Paradisi, Matteo (2021). Firms and Policy Incidence. Working Paper.
https://uploads-ssl.webflow.com/5d8e3657fd776a7142924af1/60b107ce7b635e55ff194322_paradisi_2021_firms_policy_incidence.pdf
Policy Features
- Category
- Taxes
- Sub-Category
- Payroll Taxes
- Beneficiary Type(s)
- Adults, Children and Youth, Firms
- Average Age
- 19
- Country of Implementation
- Sweden
- Year of Implementation
- 2007
- Empirical Method
- Difference in Differences
- Research Type
- Secondary
- Peer Reviewed
- No
- MVPF Publication Link
- www.aeaweb.org/articles?id=10.1257/aer.20171937