Effects of Dividend Taxes on Tax Planning, Investments and Employment
Since the turn of the millennium there has been a dramatic increase in dividend payouts from closely held corporations (CHC:s) in Sweden. During the same time period, the tax treatment of dividends from CHC:s has become significantly more generous. In this research project we will use de-identified micro data, linking firms to their active owners. We first describe how corporate owners act to minimize taxes on distributed profits given the so-called “3:12-rules”, which are in place to prevent income shifting from wages to dividends. We will then analyze the effects of dividend tax rules on real outcomes such as investments, wages and employment. Moreover, we wish to test the hypothesis that expectations of higher taxes, inducing a temporary acceleration in dividend payouts, also impacts on investments and real firm behavior. We plan to combine quasi-experimental and descriptive empirical methods with insights from modern public finance. From an international perspective our data are excellent: the data cover the entire universe of individuals and firms, and they contain detailed and relevant information, e.g. identifiers for corporate groups. Finally, since CHC owners tend to locate at the top of the income distribution, we also believe that our project will deepen our understanding of income inequality.
Final report
Sweden has a dual income tax system, which means that labor income is taxed progressively, while capital income is taxed at a lower proportional rate. In order to prevent income shifting from highly taxed labor income to low taxed capital income, rules are needed to regulate how much income active owners of closely-held companies can tax as capital. The purpose of this project was to describe and understand tax planning under the Swedish so-called 3:12 rules and, more generally, the effects of tax planning on other economic activities. In line with the original project description, a special focus has been placed on studying the role of expectations in corporate behavior.
The project was led by Håkan Selin. In studying the role of expectations in tax planning, Selin has collaborated with Johan Holmberg, who is a researcher at the Umeå School of Business, Economics and Statistics. The research has been data intensive and the work has involved research assistants. We have mainly used data from the so-called FRIDA database provided by Statistics Sweden (SCB). FRIDA contains income tax returns for all Swedish companies. It is possible to link closely held companies (unlisted companies with a high concentration of ownership, where partners work in the company) with the individual owners' declarations for a longer period of time. Since 2012, it has also been possible to unlisted companies that are not closely held to their individual owners. We have also compiled data on groups of companies, with the aim of treating a group as a single company in the analysis.
The main part of the project is to study the role of expectations (Holmberg and Selin, 2025a). An increase in the dividend tax on shares in Swedish closely-held companies from 20 to 25 percent was scheduled to take effect on January 1, 2018. However, due to political turmoil in the summer of 2017, the tax increase was not implemented. Using a difference-in-difference approach, we compare closely held companies that would have been subject to the planned tax increase with other non-listed companies whose dividend tax was not affected by the withdrawn proposal. We track these groups of companies from 2012 to 2019. We draw the following conclusions:
- Dividends in closely held companies were at an unusually high level in 2016 and 2017 and then fell sharply in 2018, especially in cash-rich firms. The sharp drop in dividends between 2017 and 2018 is a result of tax planning; there are no other plausible explanations.
- We find no effects on real economic activity in firms, such as investment and wage costs. Cash declined among closely held companies during the period of high dividend distributions, but recovered quickly.
- During the period we study, dividends are on an increasing trend in closely held companies, and there are also many companies that switch to being closely held companies during the period. We describe and discuss the role of the growing dividend allowance in this development.
There has been a lot of research on how firms respond to changes in dividend taxes, but it is very unusual to evaluate a reform that never happened. We describe in detail, both qualitatively and quantitatively, the political process that created the expectation of a tax increase and why the reform was canceled. An interesting aspect of our study is that we examine unlisted firms, which are characterized by a higher degree of ownership concentration, and the partners often work in the firm. In contrast to large public companies, there is a close alignment of interests between owners and managers, which can facilitate tax planning. There is a complete lack of prior knowledge on how such firms react to canceled tax reforms. Furthermore, our estimates are made over a long period of time and take advantage of the fact that there is a natural control group (other non-listed firms) that did not expect a tax increase. This provides a credible picture of firms' reactions to expectations of higher taxes.
In addition to the study on the canceled tax increase, we have made substantial descriptive work on the Swedish tax rules in a Nordic context. Selin (2025) has provided an overview of the splitting rules in Sweden, Finland and Norway, i.e. the rules for how corporate owners may classify labor and capital income. The article includes a discussion of the conflicting objectives that arise when designing such tax rules. An important result of this comparison is that the Swedish and Finnish tax systems give firms very different incentives: In Sweden, the so-called dividend allowance (the income that can be taxed as capital income) depends, in simplified terms, on the size of the wage sum in the company. In Finland, the dividend allowance instead depends on the net assets of the company. The work in this part of the project has generated new research questions. Selin has initiated a collaboration with a Finnish research group, where the aim is to compare company outcomes and reforms in Finland and Sweden.
The project's research has been presented at several research seminars and conferences. Holmberg and Selin (2025a), for example, was presented at "The 2025 Annual Congress of the International Institute of Public Finance (IIPF)" in Prague. Selin (2025) was presented at the "Workshop on Taxation of Profits and Dividends" in Bergen, Norway, 2024.
We have also interacted with the wider community. We have published a popular science report based on Holmberg and Selin (2025a) in Swedish (Holmberg and Selin, 2025b). There we describe the results for a broader public without technicalities. We have been invited by the Swedish Tax Agency and the Nordic Tax Research Council (NSFR) to present the results of the project in the spring of 2025.
The project has generated one publication in an open access scientific journal (Selin, 2025), one working paper that we intend to publish in a strong scientific journal under open access license (Holmberg and Selin, 2025a) and one popular science report in Swedish (Holmberg and Selin, 2025b). The papers in IFAU's working paper series are reviewed by two expert reviewers and an editor before publication. It should also be mentioned that both Holmberg and Selin (2025a,b) are publicly available to everyone via the IFAU website.
Personal website: https://sites.google.com/site/researchselin/home
Website at the Institute for Evaluation of Labor Market and Education Policy (IFAU): https://www.ifau.se/Om-IFAU/Medarbetare/Forskare-Utredare/Hakan-Selin/
Publications:
Article in peer-reviewed journal
Selin, H. (2025) “Taxing Dividends in a Dual Income Tax System. The Nordic Experience with the Income Splitting Rules” Nordic Tax Journal, published online (open access).
Working paper
Holmberg, J. and Selin, H. (2025a) “Firm-level Responses to a Canceled Dividend Tax Increase” IFAU Working Paper 2025:3.
Popular science report
Holmberg, J. and Selin, H. (2025b) “Skattehöjningen som aldrig blev av. En studie av hur förväntningar om utdelningsskatter påverkar företag” IFAU-rapport 2025:8.
The project was led by Håkan Selin. In studying the role of expectations in tax planning, Selin has collaborated with Johan Holmberg, who is a researcher at the Umeå School of Business, Economics and Statistics. The research has been data intensive and the work has involved research assistants. We have mainly used data from the so-called FRIDA database provided by Statistics Sweden (SCB). FRIDA contains income tax returns for all Swedish companies. It is possible to link closely held companies (unlisted companies with a high concentration of ownership, where partners work in the company) with the individual owners' declarations for a longer period of time. Since 2012, it has also been possible to unlisted companies that are not closely held to their individual owners. We have also compiled data on groups of companies, with the aim of treating a group as a single company in the analysis.
The main part of the project is to study the role of expectations (Holmberg and Selin, 2025a). An increase in the dividend tax on shares in Swedish closely-held companies from 20 to 25 percent was scheduled to take effect on January 1, 2018. However, due to political turmoil in the summer of 2017, the tax increase was not implemented. Using a difference-in-difference approach, we compare closely held companies that would have been subject to the planned tax increase with other non-listed companies whose dividend tax was not affected by the withdrawn proposal. We track these groups of companies from 2012 to 2019. We draw the following conclusions:
- Dividends in closely held companies were at an unusually high level in 2016 and 2017 and then fell sharply in 2018, especially in cash-rich firms. The sharp drop in dividends between 2017 and 2018 is a result of tax planning; there are no other plausible explanations.
- We find no effects on real economic activity in firms, such as investment and wage costs. Cash declined among closely held companies during the period of high dividend distributions, but recovered quickly.
- During the period we study, dividends are on an increasing trend in closely held companies, and there are also many companies that switch to being closely held companies during the period. We describe and discuss the role of the growing dividend allowance in this development.
There has been a lot of research on how firms respond to changes in dividend taxes, but it is very unusual to evaluate a reform that never happened. We describe in detail, both qualitatively and quantitatively, the political process that created the expectation of a tax increase and why the reform was canceled. An interesting aspect of our study is that we examine unlisted firms, which are characterized by a higher degree of ownership concentration, and the partners often work in the firm. In contrast to large public companies, there is a close alignment of interests between owners and managers, which can facilitate tax planning. There is a complete lack of prior knowledge on how such firms react to canceled tax reforms. Furthermore, our estimates are made over a long period of time and take advantage of the fact that there is a natural control group (other non-listed firms) that did not expect a tax increase. This provides a credible picture of firms' reactions to expectations of higher taxes.
In addition to the study on the canceled tax increase, we have made substantial descriptive work on the Swedish tax rules in a Nordic context. Selin (2025) has provided an overview of the splitting rules in Sweden, Finland and Norway, i.e. the rules for how corporate owners may classify labor and capital income. The article includes a discussion of the conflicting objectives that arise when designing such tax rules. An important result of this comparison is that the Swedish and Finnish tax systems give firms very different incentives: In Sweden, the so-called dividend allowance (the income that can be taxed as capital income) depends, in simplified terms, on the size of the wage sum in the company. In Finland, the dividend allowance instead depends on the net assets of the company. The work in this part of the project has generated new research questions. Selin has initiated a collaboration with a Finnish research group, where the aim is to compare company outcomes and reforms in Finland and Sweden.
The project's research has been presented at several research seminars and conferences. Holmberg and Selin (2025a), for example, was presented at "The 2025 Annual Congress of the International Institute of Public Finance (IIPF)" in Prague. Selin (2025) was presented at the "Workshop on Taxation of Profits and Dividends" in Bergen, Norway, 2024.
We have also interacted with the wider community. We have published a popular science report based on Holmberg and Selin (2025a) in Swedish (Holmberg and Selin, 2025b). There we describe the results for a broader public without technicalities. We have been invited by the Swedish Tax Agency and the Nordic Tax Research Council (NSFR) to present the results of the project in the spring of 2025.
The project has generated one publication in an open access scientific journal (Selin, 2025), one working paper that we intend to publish in a strong scientific journal under open access license (Holmberg and Selin, 2025a) and one popular science report in Swedish (Holmberg and Selin, 2025b). The papers in IFAU's working paper series are reviewed by two expert reviewers and an editor before publication. It should also be mentioned that both Holmberg and Selin (2025a,b) are publicly available to everyone via the IFAU website.
Personal website: https://sites.google.com/site/researchselin/home
Website at the Institute for Evaluation of Labor Market and Education Policy (IFAU): https://www.ifau.se/Om-IFAU/Medarbetare/Forskare-Utredare/Hakan-Selin/
Publications:
Article in peer-reviewed journal
Selin, H. (2025) “Taxing Dividends in a Dual Income Tax System. The Nordic Experience with the Income Splitting Rules” Nordic Tax Journal, published online (open access).
Working paper
Holmberg, J. and Selin, H. (2025a) “Firm-level Responses to a Canceled Dividend Tax Increase” IFAU Working Paper 2025:3.
Popular science report
Holmberg, J. and Selin, H. (2025b) “Skattehöjningen som aldrig blev av. En studie av hur förväntningar om utdelningsskatter påverkar företag” IFAU-rapport 2025:8.