The Marginal Tax Wedge of Labor in Sweden from 1861
to 2009
Gunnar Du Rietz
Dan Johansson
Mikael Stenkula
April, 2010
Abstract: Although it is mainly the marginal tax rates that will influence the behavior of
economic agents, long-time series covering marginal tax rates are rarely available. This paper
presents annual Swedish time series data on marginal tax rates and marginal tax wedges of
labor, including top tax rates, for three representative income groups covering the time-period
1861 to 2009. Sweden stands out as an interesting example of a European country with a
sharp increase of the tax revenues as a percentage of GDP, and can be used as an illustration
of the European evolution though in a more radical way. The marginal tax wedges were low
(about 10 per cent) at the beginning of the period to the turn of the 19th century and were
roughly the same for the three income groups. The temporary increases caused by the two
World Wars and the depression in the 1930s were all made permanent. Following World War
II, high inflation and new indirect taxes made the wedges increase further. The top marginal
tax wedge exceeded 90 per cent in parts of the 1970s and 1980s. After the introduction of a
marginal tax cap and two tax reforms, the top marginal tax wedge was reduced to about 75
percent in 2009. For an average blue-collar, white-collar and median corporate executives the
marginal tax wedge has been reduced from at most about 70, 78 and 90 percent in the 1970s
to 53, 73 and 75 percent in 2009.
JEL-codes: H21, H31, N44
Keywords: marginal tax rates, marginal tax wedges, tax reforms.
Gunnar Du Rietz
Research Institute of Industrial
Economics (IFN)
Phone:+46-8-756 90 78
Fax:
e-mail:gunnar@durietz.com
Dan Johansson
The Ratio Institute
P.O. Box 3203
SE-103 64 Stockholm
Phone: +46-8-441 59 03
Fax: +46-8-441 59 29
e-mail: dan.johansson@ratio.se
1
Mikael Stenkula
Research Institute of Industrial
Economics (IFN)
P.O. Box 55665
SE-102 15 Stockholm
Phone: +46-8-665 45 00
Fax: +46-8-665 45 99
e-mail: mikael.stenkula@ifn.se
1 Introduction
Taxation is one of the most common topics that engage the attention of economists.
Comparing the evolution of the tax system within Europe as well as comparing the European
evolution with other countries outside Europe over time is an interesting exercise. Table 1
shows the evolution of total tax revenues as a percentage of GDP for Sweden, EU-15, OECD
and USA from 1955 until 2005. As can be seen from the table the tax share was about the
same in 1955, just below 25 percent. Comparing the evolution of EU-15 with OECD and in
particular with US, one can conclude that taxes have increased faster within the EU since
then. How this may have affected the European economy is a disputed question. One reason
for this is lack of appropriate data.
There are several differences between the European countries. Looking at table 1, one can
see that Sweden has increased its tax share of GDP more than EU on average. In a European
context, it could be of interest to examine the countries with the most radical development in
more detail. As economic efficiency and productivity growth depends on a range of factors it
may be easier to trace potential negative or positive effects from a specific policy variable,
such as some form of tax measure, if there has been drastic changes in the variable. To
understand the European evolution it could, hence, be a good idea to analyze the Swedish
development where tax increases have been especially intense. The Swedish development can
be seen as an illustration of the political evolution in Western Europe during the 20th century,
though in a more radical way. Although, the “Swedish experiment” during this time period
includes specific Swedish phenomena, such as an active labor market policy, the Swedish
evolution may shed light also on problems and achievements of the Western European model.
Fast productivity growth during the long period 1870–1970 gave Sweden a higher
economic standard than most other countries. Sweden was able to combine a relatively rapid
economic growth with full employment and an even income distribution until about 1970,
when economic growth started to deteriorate and a sharp increase in unemployment occurred
in the early 1990s. This deterioration may partly have been an effect of the dramatically
expansion of government spending, taxes and regulation in the labor market in the 1960s and
the 1970, which strongly distorted economic incentives. The weak Swedish economic
performance induced policy retreats, like deregulation of domestic capital markets and
international capital movements in the 1980s, tax reforms at the mid-1980s and early 1990s,
and a shift to an anti-inflationary policy regime in the early 1990s (Lindbeck 1997). Many
2
other OECD-countries, including countries in the Western Europe, have been engaged in
similar reforms during the 1980s though not of the same extent (OECD 2006).
Time series analysis requires information for a long time period back in time if the
number of observations is supposed to be large enough for econometric analysis or to trace
long-term trends. Extracting information about this is time consuming and often cumbersome.
Furthermore, analyses should be based on marginal and not average tax rates.1 Instead of
looking at one form of taxation, like the labor income tax, a wider measure that is often used
is the so-called tax wedge of labor. The tax wedge includes the income taxes and social
security contributions/payroll taxes and sometimes also the consumption taxes. This measure
will better cover the total effect on individual decision-making in the labor market (Agell et al
1995, pp. 125–126 and Lindbeck 1997, pp. 53–59).
Lately, Barro and Redlick (2009) have added to the literature about taxes and their effects
by computing long-term marginal tax wedges of labor for the United States from 1912 to
2006. They also urge for long-term macroeconomic tax data for other suitable OECD
countries, though they fear that: “[i]t is unclear whether the necessary underlying information
exists to assemble analogous long-term time series for other countries.”(p. 33) Barro’s and
Redlick’s pessimistic view may be true for many countries within Europe. However, with
special interest in Sweden, we know that Sweden has well documented information about its
tax system far back in time.
The purpose of this paper is to describe the dramatic long term evolution of the marginal
tax wedge of labor for Sweden and hence, complement Barro’s and Redlick’s time series for
the US with a European example. Sweden is of general interest from a European point of
view as it mimics the European evolution in many aspects and is the country that
experimented with largest increase in tax rates since the mid-60s. Sweden has had the highest,
tax-to-GDP ratio in the world with the exception for Denmark some occasional years
(Rodrigues 1981, table 2.1; Riksskatteverket 2002, table 14.3). The Swedish development
often receives attention within other European countries as well as outside Europe.
Furthermore, if one wants to derive and analyze really long-time series covering World War I
and World War II, it is a problem that most OECD and European countries experienced
massive destruction of their economies during the wars, severely affecting the outcome
1
Most cross-country and time-series analyses do not use marginal tax measures but often some form of average
tax rate in itself or as a proxy for the marginal tax rate. As concluded by, e.g., Engen and Skinner (1996, p. 636)
“[…] a major shortcoming with nearly all cross-country and time-series studies is the difficulty of measuring the
marginal tax burden appropriately. The average tax rate does not reflect the marginal tax burdens hypothesized
to affect economic decisions.” Notable exceptions include, e.g., Koester and Kormendi (1989) and Padovano and
Galli (2002) who analyze the effects of both average and marginal taxes.
3
pattern. The economic performance will to a large extent be affected by these events.
According to Barro and Redlick, Canada or Latin American countries could be suitable
countries to analyze to avoid these problems. However, as a neutral country during both world
wars, Sweden avoided massive destruction making long time data analysis from this
European country appropriate. Finally, we focus on Sweden because of our particular
knowledge about Swedish conditions and its data availability. The Swedish data can be
complemented and compared with other European country data in the future, derived from
researchers with specific knowledge about the tax system in other European countries.
The analysis begins in 1861 when Sweden introduced a new tax system. Hence, our study
covers a longer time period than Barro and Redlick (2009). The decades around the 1850s are
historically important for the economy, representing a break with previous periods. The
Swedish economy was extensively deregulated, industrialization started and growth levels
took off. To our knowledge, tax data are not available before 1861 at a level that is detailed or
general enough. The latest data available are from 2009. Longer time series have previously
only been done by Du Rietz. The most recent update, covering the period 1952–2003, is
published in Johansson (2004, Table A1, pp. 93–94). No one has generated this kind of
dataset for Sweden before
Table 1. Total tax revenues as a percentage of GDP (%).
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Sweden
24,0
25,5
33,3
37,8
41,2
46,4
47,4
52,2
47,5
51,8
49,5
EU-15
25,8
25,9
27,6
29,7
32,1
34,9
37,5
38,1
39,0
40,6
39,7
OECD
23,9
24,5
25,5
27,5
29,4
30,9
32,6
33,7
34,7
36,0
35,7
USA
23,6
26,5
24,7
27,0
25,6
26,4
25,6
27,3
27,9
29,9
27,5
Source: OECD stat extracts
2. The marginal tax wedge on labor income
Taxes imposed in an economy drive a wedge between the price of labor, paid by firms, and
the net return of labor, received by the employee. This difference is formally denoted the tax
wedge on labor income. This tax wedge influences the incentive to supply and demand labor
as well as the wage formation process.
Although the definition of the tax wedge seems clear-cut at first sight, there exists no
consensus regarding what taxes to include when the wedge is calculated in practice. The main
dispute concerns the treatment of consumption taxes (de Haan et al 2004). Consumption taxes
reduce the real value of wages and consumption opportunities. The tax wedge should refer to
4
how much taxes influence the difference between the gross labor cost and net consumption
wage, i.e., what one can buy for the wage.
Including the consumption taxes, the tax wedge, tw, can formally be defined as
tw
1
(1 t1 )(1 t 2 )
,
(1 t 3 )
(1)
where t1 is the income tax, t2 consumption taxes (expressed as a share of the total price) and t3
is the payroll taxes and the social security contributions (SSCs) which are added on the wage
and paid by the employer. Part of the SSCs may often correspond to a distinct social security
benefit with a direct link between fee and benefit (e.g., sickness pay and pension benefits).
Strictly speaking, only the tax component of the SSCs should be included in the tax wedge.
Further, some of the SSCs is paid by the employee and should be added as a third tax in the
numerator.
An augmented definition of the tax wedge, which will be used in this paper, is
tw
1
(1 t1 )(1 t 2 )(1 t 4 ) b1 b2
,
(1 t 3 )
(2)
where, t4 is SSCs paid by the employee and b1 and b2 is the benefit share of the SSCs paid by
the employee and employer which should not be included in the tax wedge.
In this paper, we will calculate the marginal tax wedge. The marginal tax wedge is
defined as in equation (2), but the tax rates will refer to marginal effects.2 The marginal tax
wedge measures the difference between the total labor cost paid by the employer and the net
consumption wage received by an employee, as a result of a marginal increase in the labor
income. The wedge is expressed as a percentage of the change in labor compensation
including the SSCs.
As the marginal tax rates often change with income, we will compute the marginal tax
wedge of labor at three different income levels referring to three large professional employee
groups covering the essential part of the labor market: an average blue-collar worker, an
average white-collar worker and a median corporate executive. The income distribution
among corporate executives is skewed, with few persons having relatively high incomes.
Using the median income instead of the average may hence provide a better description of the
corporate executive population. The annual labor income levels for these groups are taken
from Edvinsson (2005) before 1952 and from the income statistics of the Swedish Employers’
2
See Appendix B for a further specification.
5
Confederation from 1952 and onwards. Parallel to these measures, we will also compute the
top marginal tax wedge.3
The concept of marginal tax wedge that influences the economy is simple in theory;
however, in reality it is pure drudgery and quite complicated to calculate this kind of detailed
long time-series. Estimates from statutory tax rates are difficult to use as the tax system is
highly complex and often includes a whole range of exemptions, allowances and exceptions
that changes over time and which often may be difficult to grasp.
For part of the period investigated, the marginal tax wedges may to some degree depend
on marital status, child allowances, taxpayer wealth, deductions of interest payments
(especially on mortgages), and other deductible costs necessary for acquisition of income. To
simplify analysis, the marginal tax rate and marginal tax wedge have been calculated
assuming that the taxpayers are single and have no children, no wealth and no net interest
costs or deductible journey costs.4
To avoid an extensive and time-consuming section containing a comprehensive
description of all details and problems involved, this is excluded in this version. A description
of the general structure can be seen in Du Rietz et al. (2010).5
3. The development of components of the marginal tax wedge
This section will briefly present the development of the state and municipality income tax
system, consumption taxes and the SSCs. The presentation of the state income tax system is
more extensive as it includes several major changes. Figure 1 gives an overview of all the
specific taxes over the time period examined, including temporary taxes.
3
The average blue-collar worker is defined as a male, full-time working industrial worker with no overtime pay.
The average white-collar worker is defined as a male, full-time working salaried employee with no overtime pay
(formally position 3–5 in the white-collar statistics from the Swedish Employers’ Confederation). The median
corporate executive is defined as a male, full-time working financial manager in a large firm (formally position
9002 in the statistics). Edvinsson (2005) only includes data about the blue-collar worker. The income levels for
the other two groups are estimated by assuming that the wage relationship between the average blue-collar
worker and the other two groups were the same as in 1952. As the wage dispersion has decreased during the 20 th
century, this assumption will probably underestimate the true income level of the white-collar worker and, in
particular, the median corporate executive. For median corporate executives with extremely high incomes, the
top tax wedge applies.
4
Calculations taking interest costs and other deductible costs into consideration have been made by Du Rietz
(1994) between 1952 and 1993 with updated figures to 2003 in Johansson (2004). Comparing the marginal tax
wedge from that study with our results, one can conclude that the differences are often only minor. The largest
difference is about 5 percentage points (for the median corporate executive between 1977 and 1982).
5
In Du Rietz (2010) the reader can also find an extensive data material which the evolution depicted in this
article is based upon.
6
1860
1870
1880
1890
Municipality income tax 1861–
1900
1910
Defense surtax 1918
Defense tax 1915–1919
Social security contributions paid by employee
1913–1974, 1993–2005
State income tax 1903–
Municipality progressive income tax 1921–1937
1920
1930
7
1940
1950
1960
State appropriation tax 1861–1911 (1920)
State equalization tax 1928–1937
Extra state income tax 1931–1937
Defense tax 1938–1947
Sales tax 1941–1946, 1960–1968
Social security contributions paid by employer 1955–
Value added tax 1969–
Figure 1. Summary of taxes, 1861–2009
1970
1980
1990
2000
3.1 Central government taxation, the state income tax6
In Sweden, income taxes are paid both to the municipalities (local government) and to the
state (central government) throughout the studied period. This section presents the state
income tax system whereas the next section describes the municipality income tax system.
The municipality tax was deductible from the state taxable income base before 1971
The presentation below is divided into eight subsections following the major state tax
reforms during the period examined. Figure 2, shows the marginal tax rates paid to the state
and the municipalities. The total marginal tax effect, i.e. the sum of the state and municipality
tax rates considering the deductibility of the municipality taxes, are depicted in Appendix.
Figure 2. State and municipality marginal tax rates, 1861–2009
70,0
60,0
50,0
Percent
40,0
30,0
20,0
10,0
20
08
20
01
19
94
19
87
19
80
19
73
19
66
19
59
19
52
19
45
19
38
19
31
19
24
19
17
19
10
19
03
18
96
18
89
18
82
18
75
18
68
18
61
0,0
Municipality tax rate
State marginal tax rate, average blue-collar worker
State marginal tax rate, average white-collar worker
State marginal tax rate, median executive
Note: Between 1920 and 1937 there was an additional municipality progressive tax rate implying a 0.5
percentage point increase of the municipality tax rate for the median corporate executive. To avoid cluttering,
this effect is not shown in the figure. The state marginal tax rate refers to the statutory marginal tax rates on
taxable income, i.e., it does not take into account that municipality taxes are deductible. The total effect can be
seen in appendix A.
Sources: Gårestad (1987), Rodriguez (1980), Söderberg (1996), Skattebetalarnas förening (1966–2009) and own
calculations.
6
If not otherwise stated, this section is based on Genberg (1942), Gårestad (1987) and Söderberg (1996). We
will use the term marginal tax rate referring to the state marginal tax rates in this section.
8
The tax system 1861–1902
During the 19th century, Sweden had a state tax system based on so-called appropriations,
which was a rather heterogeneous tax system based on the economic and social order present
at that time. In 1861, this system was reformed and simplified, reducing the income tax
groups from eight to two (appropriation on real estate income and appropriation on labor or
capital income) and generalizing and broadening the concept of income applied. Parallel to
these taxes, there were also some basic taxes, such as armament fees and personal protection
fees, which can be characterized as lump-sum taxes. These taxes were mainly phased out
during the 1890s.
The tax level on labor and capital income according to the appropriation system was
normally set to one percent. Occasionally, additional appropriation taxes were levied if the
ordinary appropriation taxes yielded insufficient tax revenues. The tax level could then be
increased to two percent of your income.
The tax system 1903–1919
In 1903, a new progressive form of income tax system was introduced. The new tax system
was accepted without major conflicts, partly because the proposed progressivity was very low
and partly because a strong public opinion supported a new income tax that could be used to
rearm the military. The main objective for the reform was to increase funding of public
expenditures.
Although the tax system was progressive, the progressivity was rather moderate. The
marginal tax rates varied from one to five percent. Taxpayers had to start paying the lowest
tax rate, 1 percent, for incomes above SEK 1,000, corresponding to the income of an average
white-collar worker. Hence, most blue-collar workers did not pay any new income taxes. The
highest marginal tax rate had to be paid for incomes above SEK 80,000, corresponding to the
average wage of 100 blue-collar workers in 1903. There was also an average tax cap that
limited total state tax to at most 4 percent of the taxable income. The old appropriation system
was still used parallel to the new system. In 1911, the tax income brackets were slightly
revised and the old appropriation system was now abolished as a state income tax.
As a result of World War I, additional temporary progressive “defense taxes” were
introduced to finance needed military expenditures. The tax rates could be relatively high (up
to 13.5 percent on the margin) but affected only people with very high incomes.7
7
During 1915–1917, the defense tax was based on the income in 1913 and did not affect the marginal tax rate.
9
The tax system 1920–1937
After World War I, a new state tax system, that was supposed to replace the ordinary tax
system and the temporary defense taxes, was implemented. This tax system was more flexible
and more stable than the earlier systems. Technically, the structure of the tax system, i.e. the
tax income brackets and the imposed progressivity, was fixed, but the specific tax rates were
flexible and were decided on an annual basis by the politicians. The politicians could easily
change the state tax rates, in accordance with supposed financial needs. Another invention
within this tax system was the introduction of general state (and municipality) income tax
allowances. The allowances were rather generous and could be as large as one third of an
average blue-collar worker’s wage.
The system was rather progressive, with marginal tax rates running from about 4.5–5.5
percent to 22–28 percent.8 As before, there was also a tax cap, which now restricted the
average tax to about 17.5–21.5 of taxable income. In practice, however, the first tax income
bracket was very wide (the upper limit corresponds to almost ten times the taxable income of
an average blue-collar worker in 1920) including the majority of all taxpayers.9
The tax system 1938–1947
Just before World War II, the ordinary tax system was sharpened at the same time as the state
equalization tax and extra income tax were abolished. That is, the temporary tax increase
caused by these taxes was in this way made permanent within the ordinary income tax system.
The average tax cap was also removed from the tax system.
Technically, the system consisted of one flexible tax rate (the bottom tax / bottenskatt),
which was decided on an annual basis by the politicians, and one fixed tax rate (the surtax /
tilläggsskatt), i.e. it was partly built up in the same way as the tax system it replaced. All in
all, these changes resulted in an increase of the progressivity of the tax system.
Although the equalization tax and extra income tax were abolished to simplify the tax
system, another supposedly temporary tax, the defense tax (värnskatten), was introduced in
1938. This was a highly progressive income tax which had to be paid by most taxpayers. It
was also sharpened in 1941. Due to military tensions in the world, there was not much debate
or critics of the 1938 tax reform and it was almost an unanimous political decision.
8
As the tax rates were flexible it is not possible to give a fixed tax rate. The tax rates refer to the tax rates used in
practice during this time.
9
Parallel to this new tax system, two additional temporary state income taxes were introduced during the period
(se figure 1), which are included in our calculations. In practice, most people did not pay these extra taxes as
their income were too low.
10
The tax system 1948–1970
After World War II, the tax system was changed once again. The progressive defense tax was
abolished at the same time as the tax level and progressivity in the ordinary income tax
system was increased. The highest state marginal tax rate was now 70 percent but was only
paid by taxpayers with extremely high income corresponding to taxable income of more than
50 average blue-collar workers. This rate was almost twice as high as in the ordinary tax
system that was replaced, but almost the same if the temporary defense tax is included. The
higher tax level that had been a result of World War II was in this way made permanent for
many taxpayers.
This tax reform can be seen as the foundation of the Swedish system with a high (and
progressive) tax schedule and a high level of public expenditures. The tax system had a
distinct redistributional objective besides the financial needs that the tax revenues were
supposed to meet. The fiscal policy debate was, as a result, unusually intense in parliament
before this new tax system was passed.
The income tax schedule was slightly adjusted several times during the 1950s and 1960s
(1952, 1953, 1957, 1962 and 1966). In nominal terms, these adjustments were minor tax
reductions, but all insufficient to prevent tax increases in real terms caused by price and wage
inflation shifting taxpayers up into tax income brackets with higher marginal tax rates.
The tax system 1971–1982
In 1971, a new tax system was introduced. At least two problems had evolved from the
existing tax system. Firstly, due to the increase of municipality tax rates, the state taxable
income was heavily reduced as the municipality tax was deductible. This diminished state
revenues at the same time as it favored high-income earners with high marginal tax rates.
Secondly, a tax system with a high tax progression and joint taxation of families made it
unfavorable for a second income-earner (mostly the wife) to work outside the household.
With the new tax system, the deductibility of municipality taxes was abolished and the
statutory marginal tax rates were decreased somewhat to take this into account. The
progressivity of the system was also further strengthened. For most taxpayers, the total
marginal effect of both state and municipality tax rates was only slightly increased or
decreased due to this change. Compulsory individual taxation of spouses was also introduced.
The on-going and increasing inflation in combination with a nominal progressive tax
system made it necessary to adjust tax rates on a regular basis, as in the 1950s and 1960s, to
keep the real tax level constant and to avoid high wage increases and an accelerating
11
inflationary process. However, the changes made often only reduced tax rates for low and
middle income taxpayers, which increased the progressivity of the tax system further.10 To
“finance” these nominal tax cuts, the SSCs were increased between 1973 and 1977.11
In 1978, the tax income brackets were tied to a consumer price index. In this way,
inflation no longer pushed up taxpayers into income brackets with higher marginal tax rates.
However, the total marginal tax rate could still increase due to increased municipality tax
rates and real wage increases. In 1980, an explicit marginal tax cap was introduced to avoid
excessive marginal tax rates. Initially, the tax cap restricted the total marginal tax rate to at
most 80 and 85 percent in the two highest tax income brackets.
The tax system 1983–1989
The marginal tax rate increases in Sweden came to an end when the marginal tax cap was
introduced in 1980. However, marginal tax rates were still very high. With high marginal tax
rates and favorable deduction possibilities taxpayers had strong incentives to avoid taxes by
incurring deductible costs and debts. As interest payments on housing, e.g., were fully
deductible, the real cost of housing could be substantially reduced. Politicians wanted to
reduce distortionary behaviors caused by the tax system. In 1981, the politicians decided to
reduce marginal tax rates gradually.12
Between 1983 and 1985, the marginal tax rates were decreased between five and 15
percentage points at the same time as deduction possibilities were reduced. It became
considerable more expensive for taxpayers with high marginal tax rates to borrow money and
pay mortgage interests. The tax reform 1983–1985 can be characterized as a tax-switchover
from labor income tax to SSCs, consumption taxes and capital taxation.13
Parallel to these changes, the marginal tax cap in the highest tax income bracket was
decreased to 84 percent in 1983, 82 percent in 1984 and 80 percent in 1985. The marginal tax
rates were also slightly reduced between 1987 and 1989 and the number of tax income
brackets was sharply reduced. The marginal tax cap was no longer necessary and was
abolished in 1987.
10
Jakobsson and Normann (1972), Lodin (1975) and Söderberg (1996). The inflation was on average 9.2 percent
during the 1970s, compared to 4.1 percent during the 1960s.
11
This policy, to finance decreases of income taxes by increasing the SSCs, has been called “Haga policy” after
the negotiations that were conducted in the Haga mansion between the government, the opposition parties and
the organizations of the labor market during the 1970s.
12
This tax reform is known as the tax reform of the “wonderful night” (den underbara nattens skattereform).
13
Although marginal tax rates were decreased, the marginal tax wedges of labor were not much lowered as SSCs
and consumption taxes were raised (see section 4).
12
The tax system 1990–2009
At the end of the 1980s, leading politicians and parties of the labor market urged for a major
tax reform.14 Major tax reforms had also been implemented in many other western countries.
As a result, a major tax reform was implemented in two steps between 1990 and 1991, the socalled “tax reform of the century” (århundradets skattereform). The tax reform reduced
marginal tax rates substantially and further diminished the tax effects of deducting interest
costs. The reform, which was supposed to be revenue-neutral, was financed by an extended
VAT, an increased taxation of employee benefits and partly increased capital income
taxation.15
The tax schedule only consisted of one state income tax rate, 20 percent, and was payable
only by high-income earners. Most taxpayers only paid labor income tax to the municipality.
Due to the crises and depression in the 1990s, the tax rate was first increased to 25 percent
and then split up into two new tax income brackets with tax rates of 20 and 25 percent. In
2007–2010, an earned income tax credit (skattereduktion för arbetsinkomster, the so-called
jobbskatteavdrag) was introduced in four steps. Due to its construction it slightly reduced
marginal tax rates for most low- and middle-income earners by three to six percentage points.
3.2 Local government taxation, the municipality income tax
Parallel to the central government taxation system there was a local system, including a
municipality income tax. In 1862, a new municipality tax system was introduced which to a
greater extent consisted of a percentage of the taxpayer’s annual taxable income. As
mentioned above, the municipality tax was deductible and reduced state taxable income and,
hence, the required tax payments to the central government.
During the 19th century, the marginal municipality tax rate was low and gradually
increasing from about two to five percent. In 1930, the tax rate had increased to about 10
percent and it fluctuated around this level until the end of World War II. With the state tax
reform in 1920, income tax allowances were introduced for the municipality income tax. An
extra municipality progressive tax was also introduced parallel to the ordinary income tax.
The top marginal tax rate was eight percent, but it had an average tax cap of six percent. The
high tax rates were only applicable on very high incomes; people with low or average
incomes paid very low or no tax at all.
14
In 1988, e.g., the Minister of Finance, Kjell-Olof Feldt, as well as the leader of the Swedish Trade Union
Federation (Landsorganisationen, LO), Stig Malm, meant that the Swedish tax system had become “rotten and
perverse” (Feldt 1991).
15
See Agell et al (1995, 1998) for a detailed examination of the tax reform.
13
In 1928, there was a major municipality tax reform which mostly affected the technical
and legal part of the tax system. The municipality progressive tax was, however, rearranged
and part of it was transformed to an additional state income tax.16 The remaining tax was
abolished 1938, just before World War II.
After World War II, the ordinary municipality tax rate started to increase sharply: in 1950,
the tax rate was still ten percent, but in 1960 it was 15 and in 1970, 20 percent, i.e., it doubled
in 20 years. The reasons were mainly increased responsibilities for different activities decided
by politicians (often at the national level) which required increased expenditures in
combination with the increased urbanization and its associated costs which were financed by
the municipalities. As the tax was deductible, the effect of the sharply increasing taxes was
reduced. The municipality income tax allowance was also heavily increased in 1958, which
also reduced the effect from increasing tax rates. The tax continued, to increase during the
1970s, approaching almost 30 percent in 1980 and these tax increases now had full effect as
the deduction of the municipality tax had been removed in 1971. After 1980, the tax rate has
only slowly increased or been unchanged. The municipality tax has roughly tripled after
World War II.
Figure 3. Consumption tax rates, 1861–2009
30,0
25,0
Percent
20,0
15,0
10,0
5,0
20
08
20
01
19
94
19
87
19
80
19
73
19
66
19
59
19
52
19
45
19
38
19
31
19
24
19
17
19
10
19
03
18
96
18
89
18
82
18
75
18
68
18
61
0,0
Note: There is a change in the way VAT was calculated in 1996 which may overestimate the true consumption
tax rate somewhat.
Sources: Gårestad (1987), Rodriguez (1980), SAF (1991, 1994, 2000), Statistisk Årsbok (various years) and
Konjunkturinstitutet (2009).
16
See fn 9.
14
3.3 Consumption taxation
The marginal consumption tax has been computed as the sum of value added taxes, sales
taxes, all selective purchase taxes and excise duties (including energy and environmental
taxes) less part of the food and interest-cost subsidies, as a percentage of total private
consumption.17 This should be an acceptable proxy of the marginal consumption tax.
The development of consumption taxes as a ratio of private consumption is depicted in
Figure 3. These taxes constituted a big part of total tax revenues during the 19th century. The
consumption taxes at this time included import duties and commodity excises on goods such
as liquor and sugar. Due to increased custom duty yields, our estimated consumption tax rate
increased slowly during the 19th century from slightly more than three to about five percent of
total consumption. The tax rate stayed at this level until the beginning of the 1930s with the
exception of 1913–1919, when it sharply dropped. The increase of consumption taxes was
much smaller than the increase of private consumption in current prices, which more than
doubled during this time period. This was partly due to consumption taxes being more unit
taxes (a constant amount per unit of output) than ad valorem taxes (a constant fraction of
price).
During the 1930s, consumption taxes were raised sharply on liquor, tobacco and
automobiles and under World War II, a temporary sales tax on five percent was also
introduced for the first time as a way to finance increasing military expenditures. At the end
of the war the consumption tax rate had increased to almost twelve percent, but it decreased to
about ten percent in 1947 when the sales tax was abolished.
During the end of the 1940s and 1950s, many commodity excises were increased. The
sales tax was reintroduced at a level of four percent in 1960. Even if the new sales tax started
at a low level, it opened up for later substantial tax increases. In the second half of the 1960s,
the sales tax was ten percent and the estimated consumption tax rate was just below 20
percent.
In 1969, the sales tax was replaced with the value added tax of 10 percent. The value
added tax avoided the “chain taxation” of the sales tax and was considered attractive because
its tax base was large and the tax rate was possible to increase much more than the tax rate on
regular income. The consumption tax rate had roughly doubled between 1950 and 1970. In
1971, the tax rate was increased to 15 percent and it continued to slowly increase during the
17
We only want to reduce the sum of the taxes with the part of the subsidies that reduce the price level. We
estimate that roughly half of the interest- and food subsidies and one third of the other firm subsidies affect the
price level.
15
1970s and 1980s. The estimated consumption tax rate has been fluctuating around 20 per cent
during the 1970s and 1980s.
After the tax reform in 1990–1991, the value added tax was increased to 25 percent and
the base was broadened with tax-emption for only some services such as dental care. After the
reform the VAT has been differentiated with a decreased tax rate on, e.g., food and hotels
(from 25 to 12 percent), and passenger transports and books (from 25 to 6 percent). Some
goods and services such as newspapers, cinemas and concerts which were earlier taxexempted are now, on the other hand, taxed at a rate of 6 percent. There has also been an
increase in energy-, environmental and tobacco taxes. All in all, the estimated tax rate has
fluctuated around 20 to 25 per cent during the 1990s and 2000s, but with a small increase
during the last years.18
Figure 4. The fiscal part of the SSCs paid by employee, 1861–2009
12
percent
9
6
3
18
61
18
68
18
75
18
82
18
89
18
96
19
03
19
10
19
17
19
24
19
31
19
38
19
45
19
52
19
59
19
66
19
73
19
80
19
87
19
94
20
01
20
08
0
Marginal fiscal part of SSCs, average white-collar worker
Marginal fiscal part of SSCs, average blue-collar worker
Note: The fiscal part of the SSCs paid by employees is zero for the median corporate executive.
Sources: SAF (1991, 1994, 2000), Skattebetalarnas förening (1966–2009) and own calculations.
18
Unfortunately, there was a shift in calculation of the VAT in 1996 in the data from Statistics Sweden. Part of
the small spike in 1996 is due to this change. The estimated rate may, hence, be somewhat exaggerated during
the end of the 1990s and 2000s. This will, however, not change the main pattern or the conclusions.
16
Figure 5. The marginal fiscal part of the SSC paid by employer, 1861–2009
45
40
35
percent
30
25
20
15
10
5
18
61
18
68
18
75
18
82
18
89
18
96
19
03
19
10
19
17
19
24
19
31
19
38
19
45
19
52
19
59
19
66
19
73
19
80
19
87
19
94
20
01
20
08
0
Marginal fiscal part of SSCs, median corporate executive
Marginal fiscal part of SSCs, average white-collar worker
Marginal fiscal part of SSCs, average blue-collar worker
Sources: SAF (1991, 1994, 2000), Skattebetalarnas
Konjunkturinstitutet (2009) and own calculations.
förening
(1966–2009),
Table 2. SSCs paid by employee and employer
SSCs paid by employer
Estimated
SSCs paid by employee
fiscal part (%)
Health insurance
0–100
National basic pension
1955–
1913–1973
Industrial injuries
0
Health insurance
insurance
1955–1974
1960–
National supplementary 0–100
Universal SSCs
pension
1993–2005
1960–
Unspecified payroll tax
100
1969–
National basic pension
100
1974–
Parental insurance
100
1998–
Survivors’ pension
100
1999–
Söderberg
(1996),
Estimated
fiscal part (%)
100
50
100
Note: The universal SSCs in 1993–2005, were called universal pension and universal health insurance
contribution in 1996–1997 and universal pension contribution after 1997. A couple of other minor SSCs were
introduced as well, which are included in our calculations. See Nordling (1989, pp. 69–90) for a more detailed
description. There are also a few negotiated contributions (not prescribed by law). These are benefit-connected
and are hence not included in our marginal tax wedges.
17
3.4 Social Security Contributions (SSCs)
The SSCs consist of many contributions several of which have been introduced and abolished
during the studied period. An overview and the estimated fiscal part can be seen in table 2. As
the contributions differed substantially between incomes before 1982, it is difficult to give a
general picture of the SSCs.
In 1913, the first SSC paid by employees were introduced, the national basic pension
contribution (folkpensionsavgift). Until 1935, the contribution was rather small and specified
as a fixed amount within certain income brackets. After 1935, it was 1 percent of the taxable
income (up to a cap). The rate increased slowly to 5 percent in 1973, when it was transformed
to a SSC paid by employers.
In 1955, a compulsory health insurance (sjukförsäkringsavgift) was introduced, which was
partly financed by a second SSC paid by employees and the first SSC paid by employer. The
health insurance contribution paid by employee was also rather small and specified as a fixed
amount within certain income brackets. The SSC paid by employer was 1.14 percent of the
wage. Both the contribution paid by employee and employer had an upper income cap, above
which no contribution had to be paid and the marginal effect was zero.
In 1960, two new SSCs paid by employer were introduced, the national supplementary
pension contribution (ATP-avgift) at a rate of 3 percent (up to an income cap) and an
industrial injury insurance contribution (arbetsskadeavgift) at a rate of 0.4 percent. The rate
paid to the contributions was increased during the 1960s and in 1969 an unspecified payroll
tax (allmän arbetsgivaravgift) was introduced as well, at a rate of initially 1 percent but
increasing to 4 percent in 1973.
Due to the Haga-policy, discussed above, the SSCs paid by the employers continued to
increase during the 1970s. In 1974, the only contribution paid by the employee (the national
basic pension contribution) was converted to a contribution paid by the employer. The caps in
the SSCs paid by the employer were removed in two steps in 1976 and 1982, which mainly
affected workers with high incomes. In 1982, when all caps had been removed, the rate of the
SSCs had increased to 33 percent and was the same for all workers independent of income. In
the 1990s, the SSCs paid by the employer started to decrease slowly.
In 1993, SSCs paid by employees were reintroduced. It increased sharply during the 1990s
from 0.95 percent to 7.95 percent (up to an income cap). In 2000, the rate started to decline
and in 2006 the contributions were completely removed. These contributions can be seen as
pure fiscal and a way to increase the tax revenues during and after the depression at the
beginning of the 1990s.
18
Figure 6. The marginal tax wedge on labor income, 1861–2009
100
90
80
70
percent
60
50
40
30
20
10
01
08
20
94
20
87
80
Median corporate executive
19
19
66
59
73
19
19
19
52
Average white-collar worker
19
38
45
19
19
31
19
24
19
17
Average blue-collar worker
19
03
96
89
82
75
68
10
19
19
19
18
18
18
18
18
18
61
0
Top
Note: information about the defense surtax in 1918 is not included in the calculations of the top marginal tax
wedge.
Source: Gårestad (1987), Rodriguez (1980), Söderberg (1996), Skattebetalarnas förening (1966–2009) and own
calculations.
Figure 7. Top marginal tax wedge and top marginal tax rate, 1861-2009
100
90
80
70
percent
60
50
40
30
20
10
0
1861 1871 1881 1891 1901 1911 1921 1931 1941 1951 1961 1971 1981 1991 2001
Top marginal tax wedge
Top marginal tax rate
Top state marginal tax rate
Top state marginal tax rate*
Note: Top marginal tax rate refers to the combined effect from the state and municipality income tax rates.
Top state marginal tax rate includes the effect from the deductible municipality taxes until 1971.
Top state marginal tax rate* shows the marginal effects excluding the deductible municipality taxes before 1971.
Information about the defense surtax in 1918 is not included in the calculations.
Source: Gårestad (1987), Söderberg (1996), Skattebetalarnas förening (1966–2009) and own calculations.
19
4. Results – the marginal tax wedge of labor
This section presents the development of the marginal tax wedge of labor, i.e. the combined
marginal effect of all taxes described above. As described, we will present the marginal effect
for three income levels. We also present the top marginal tax rate and top marginal tax wedge
that the tax system induced over time. Figure 6 and 7 depict the marginal tax wedge for the
three groups and the top marginal tax wedge and marginal tax rate between 1861 and 2009.
4.1 The marginal tax wedge for an average blue- and white-collar worker and for a median
corporate executive
As can be seen from figure 6, the marginal tax wedges were about six percent in 1861 for all
three income groups. At the turn of the century they had increased to slightly more than ten
percent. The main reason for this was increasing municipality and consumption taxes. Still,
the marginal tax wedges were low compared to later levels.
Until the tax reform in 1920, the marginal tax wedges continued to be stable or increase
slowly for the three income groups. Although the state income tax schedule was progressive,
the marginal tax wedges were about the same. Exceptions include the end of World War I,
when the median corporate executive had to pay a defense tax. The defense tax 1915–1917
had no marginal effects.
During the beginning of the 1920s, the marginal tax wedges start to increase due to the
new state tax system, and due to increasing municipality and consumption taxes. At the end of
the 1920s, the wedges were about 17 percent for the three income levels. Still, there were no
big differences in the wedge between the income groups. During the depression, the
introduction of temporary taxes and the increase in the ordinary tax rates led to an increase of
the marginal tax wedges. The marginal tax wedges did not decline after the depression and the
wedges were about 22 percent in 1937 for the three income groups.
The defense taxes during World War II together with the tax reform of 1938 made the
marginal tax wedges increase further. At this time, the wedge started to diverge between the
three income groups. Hence, the system became progressive in practice. The marginal tax
wedge was slightly above 30 percent for an average blue- and white-collar worker and 40
percent for a median corporate executive at the end of the war. The state marginal tax rate
after World War II had almost tripled for the average blue- and white-collar worker and
quintupled for the median corporate executive compared to the situation ten years before.
Besides higher formal tax rates, the progressive nominal tax schedule in combination with
20
high inflation and high wage increases automatically increased marginal tax rates at the
beginning and during World War II.
The increase of the wedges was made permanent after World War II when the defense
taxes were abolished and a new tax system was introduced. The marginal tax wedge had
roughly been doubled in 20 years. After World War II, the marginal tax wedge started to
increase sharply again. In 1960, the marginal tax wedge was almost 50 percent for the average
blue-collar worker, slightly above 50 percent for the average white-collar worker and 60
percent for the median corporate executive. The driving force behind this sharp increase was,
again, the price and wage inflation in combination with the highly progressive tax schedule
introduced in 1948, which pushed up taxpayers into higher income brackets with higher
marginal tax rates. The phenomenon, that inflation pushes taxpayers’ incomes into higher
income brackets, is often called “bracket creep” and is a well established effect within the tax
literature (e.g, McKee 1986). This inflation-driven tax increase mechanism implied that
politicians did not have to pass new tax laws to increase tax rates and taxes paid by income
earners.
In the 1960s, this development continued, but now the increases in the marginal tax
wedges were also a result of increasing SSCs and increasing consumption taxes. In 1970, the
marginal tax wedge was about 60, 63 and 70 percent for the three income groups analyzed.
The marginal tax wedge had roughly been doubled in 20 years again.
In 1970, a new tax reform was implemented and the so-called Haga-policy in the 1970s,
described above, was an attempt to dampen marginal tax rate increases. However, even if the
statutory state marginal tax rates were reduced, especially for low and medium income levels,
the municipality income tax rates, the consumption tax rate and, in particular, the SSCs
continued to increase. The municipality tax was, further, no longer deductible. Parallel to this
development, inflation accelerated during the 1970s, increasing bracket creeping. As a result,
marginal tax wedges continued to increase, especially for the median corporate executive. At
the end of the 1970s, the wedges were about 70, 78 and 90 percent for the three income levels
analyzed. The marginal tax wedges had been tripled in 40 years.
In 1980, the marginal tax cap reduced the marginal income tax and the marginal tax
wedge for the median corporate executive. The 1983–85 tax reform reduced the marginal tax
wedge for all three income groups. In 1989, just before the major tax reform, the wedge had
decreased by about five percentage points.
The 1990–1991 tax reform decreased the marginal tax wedges by about 10–15 percentage
points. After the reform, the marginal tax wedge was about 55, 65 and 70 percent for the three
21
income groups. During the 1990s and 2000s, the marginal tax wedge increased somewhat for
the average white-collar worker and for the median corporate executive, but decreased
somewhat for the average blue-collar worker. This was mainly caused by an increase in the
state income tax and changes in the fiscal part of the SSCs. The earned income tax credit,
introduced in 2007, has also slightly reduced the marginal tax wedge. At the end of the period
examined, the marginal tax wedge was 53, 73 and 75 percent, respectively.
4.2. Top marginal tax wedge and top marginal tax rate
Besides looking at the marginal tax wedge at three income levels, it is of interest to see how
the top marginal tax wedge and the top marginal tax rate have evolved over time.
To reduce extreme tax rates, tax caps have occasionally been introduced, as described
earlier. Average tax caps were in place between 1903 and 1937 regarding the state income
tax, and between 1921 and 1937 regarding the municipality progressive tax. These tax caps
reduced the marginal tax rates on very high incomes, but the top marginal tax rate was still the
same. An explicit marginal tax rate cap was in place between 1980 and 1987 for the total
marginal tax rate, i.e., including both the state and the municipality taxes. This cap directly
reduced the top marginal tax rate.
In figure 6, the top marginal tax wedge is depicted together with the marginal tax wedge
for the three income levels examined. As can be seen, the top marginal tax wedge has often
been substantially higher than the marginal tax wedge for the median corporate executive.
The figures start to deviate at the beginning of the 20th century with the new tax system. Still,
the top marginal tax wedge and the top income tax rate was moderate at that time compared to
later levels. The marginal tax wedge paid by the median corporate executive deviated sharply
from the top marginal tax wedge between the wars. At the end of the 1930s, the top marginal
tax wedge was about 50 percent while the marginal tax wedge associated with the median
corporate executive was less than half that value.
After World War II, high inflation and bracket creep pushed the median corporate
executive towards the top marginal tax rate. At the end of the 1970s, the formal top marginal
tax wedge and the actual marginal tax wedge of the median corporate executive coincided at
about 90 percent, and they continued roughly to coincide throughout the rest of the examined
period.
22
In figure 7, the top marginal tax wedge and top marginal tax rate (state and total) are
shown.19 As already noticed, the top marginal tax wedge and the top marginal tax rate were
low during the 19th century and the beginning of the 20th century. During World War I, the top
wedge started to rise sharply. The tax reform after the war in combination with the
introduction of a local progressive tax, implied that the top marginal tax wedge increased
from about 10 to 40 percent in 20 years. About half of the effect can be attributed to the state
marginal tax rate.
During the 1920s, the top marginal tax wedge and top marginal tax rate slightly decreased
when the economy was booming. During the 1930s and the depression, new taxes were
imposed and the ordinary tax rates were increased. As a result, the top marginal tax rate and
the top marginal tax wedge increased again; the top marginal tax wedge was about 50 percent.
The top marginal tax wedge did not decrease after the depression and when World War II
started, the top marginal tax wedge increased from about 50 to 75 percent, mainly due to
supposedly temporary tax increases to strengthen the military capacity. This level was,
however, more or less maintained after the World War and in the coming decades. The top
marginal tax wedge increased slowly due to increasing municipality taxes and due to
increasing indirect taxation. The top state marginal tax rate was, however, slightly reduced in
nominal terms in 1953. The increasing and deductible municipality tax also diminished the
state marginal tax rate slightly during the 1950s and 1960s. In the 1970s, the top marginal tax
wedge increased more sharply again due to increased income taxes and increased SSCs,
implying that the top marginal tax wedge was about 90 percent at the end of the 1970s.
The top marginal tax rate and the top marginal tax wedge were slightly reduced due to the
marginal tax cap and due to the tax reform in the first half of the 1980s. However, it was not
until the major tax reform at the beginning of the 1990s that the top marginal wedge
decreased substantially to about 70 percent and the top marginal tax rate was reduced to
slightly above 50 percent. After the reform, the top marginal tax wedge has slightly increased
to about 75 percent.
The evolution clearly shows how temporary tax increases, from the World Wars and the
depressions (in the 1930s and partly the 1990s) are made permanent after the crises. The top
marginal tax rate and top marginal tax wedge increased stepwise until the beginning of the
19
Unfortunately, we have not been able to find information about every temporary tax during World War I. In
1918 there was a defense surtax which is not included in our calculation. The maximal tax rate in 1918 was,
hence, somewhat higher (but probably at the same level as in 1919).
23
1980s, and then they decreased.20 The early development gives support to an idea originally
put forward by Peacock and Wiseman (1961), i.e., in crises the acceptable burden of taxation
increases and the acceptance of the higher tax level remains after the crises, giving rise to a
stepwise increasing function of the tax rates (cf. Rodriguez 1980). The sharply decreased
marginal tax rates after the tax reform at the beginning of the 1990s represents a break from
this pattern.
5. Conclusions
This study has derived homogenous series of the marginal tax wedge and the marginal tax
rate on labor income in Sweden between 1861 and 2009. The marginal tax wedge and the
marginal tax rate are computed for three income levels corresponding to an average bluecollar worker, an average white-collar worker and a median corporate executive. We have
also computed the top marginal tax wedge and the top marginal tax rate on labor for the same
period. These datasets are unique as no one has compiled this kind of dataset for Sweden
before. We are aware of no such long time series for any other country. The Swedish
development can be used to shed light on the general Western European evolution with an
increasing level of taxation, though Sweden stands out as one of the European countries that
has experimented with highest increase of tax rates. This study can be complemented with
other European country studies in the future.
The analysis shows that the marginal tax wedge was rather low and about the same for
both an average blue-collar worker, an average white-collar worker and a median corporate
executive until the 1930s. It increased slowly during the 19th century to about 12 percent, and
then sharply increased due to World War I, the new tax system after World War I and the
depression in the 1930s. Before World War II, it exceeded 20 percent. During World War II,
the marginal tax wedge increased further. It also started to deviate between the average blueand white-collar worker, and the median corporate executive.
After World War II, a highly progressive state income tax schedule was introduced, but
most people did not have an income high enough to reach the highest income brackets with
the highest marginal tax rates. Due to high inflation and bracket creep, and due to increased
focus on indirect taxes, like VAT, and the introduction of social security contributions paid by
employers, marginal tax wedges continued to rise sharply until the end of the 1970s. At
20
Though the development after the World War II may be more accurately depicted as a slow increase during
the 1950s and the 1960s and a higher increase during the 1970s.
24
the beginning of the 1980s, an explicit marginal income tax cap and a minor tax reform
that slightly decreased the marginal tax rates were introduced. Before these reforms, the
marginal tax wedge was about 90 percent for the median corporate executive, close to 80
percent for the average white-collar worker and about 70 percent for the average blue-collar
worker. The marginal tax wedge had tripled in 40 years. After a major tax reform in the
1990s, the marginal tax wedge decreased significantly and at the end of the 2000s, the
marginal tax wedge is about 53, 73 and 75 percent for the three income levels analyzed.
The top marginal tax rate and the marginal tax wedge for the median corporate executive
were about the same during the 19th century and the beginning of the 20th century. The top
marginal tax wedge started to diverge and increase faster between the world wars. Already
during the depression in the 1930s it was about 50 percent. After World War II, the top
marginal tax wedge and the marginal tax wedge for the median corporate executive started to
converge due to inflation and bracket creep and reached 90 percent at the end of the 1970s.
From the beginning of the 1980s, the top marginal tax wedge and the marginal tax wedge for
the median corporate executive was again the same.
All temporary increases during crises, like the world wars and depressions, have been
maintained after the crises. A highly progressive tax system, in combination with high
inflation has also automatically increased the marginal tax wedge. An increased reliance on
indirect taxation, which is less transparent, has evolved over time when direct taxation met
increased political resistance among the public and a broadening of the tax base was deemed
necessary. There is a clear increasing trend in the marginal tax wedge until the end of the
1970s and a clear decreasing trend after that, though with the exception of a minor increase
due to the depression in the 1990s.
25
Appendix A
Figure A1. Marginal tax rates, 1861–2009
100,0
90,0
80,0
70,0
percent
60,0
50,0
40,0
30,0
20,0
10,0
20
08
20
01
19
94
19
87
19
80
19
73
19
66
19
59
19
52
19
45
19
38
19
31
19
24
19
17
19
10
19
03
18
96
18
89
18
82
18
75
18
68
18
61
0,0
Average blue-collar worker
Average white-collar worker
Median executive
Note: The marginal tax rate is the sum of the state and municipality marginal tax rates, considering that the
municipality income taxes were deductible from the state income tax base before 1971.
Source: Gårestad (1987), Rodriguez (1980), Söderberg (1996), Skattebetalarna förening (1966–2009) and own
calculations.
26
Appendix B
The marginal tax wedge of labor is defined as:
t wc
1
(1 t1c )(1 t 2c )(1 t 4c ) b1c b2c
,
(1 t 3c )
where t’1 is the marginal (income) tax, t’2 marginal consumption taxes, t’4 marginal SSCs paid
by the employee, t’3 marginal SSCs paid by the employer and b’1 and b’2 is the marginal
benefit share of the SSCs paid by the employee and employer which should not be included in
the marginal tax wedge.
The income tax, t1, is dependent upon income level, y, and tax allowances, a, and tax
credits, c, where a and c also may vary with income. The relationship can be written as:
t1
s s ( y a s ( y)) s m ( y am ( y)) c( y)
where s(.) refers to the tax function, i.e. how much tax the taxpayer has to pay given his
taxable income (y - a), a(.) refers to the tax allowance function and c(.) refers to the tax credit
function. Subscripts, s and m, refers to state and municipality. y-as and y-am is denoted the
state and municipality taxable income. Hence, the marginal effect, t’1, will be:21
t1c
dt1
dy
s cs * (1 a cs ) s mc * (1 a mc ) c c
where s’ and a’ refer to the derivate of the s and a functions. When the allowance is fixed and
not changing with income, the term a’i is zero, i.e. (1-a’i) is one and can be skipped.
21
To avoid cluttering, the arguments of the functions are skipped.
27
References
Agell, Jonas, Englund, Peter and Södersten, Jan (1995). ”Svensk skattepolitik i teori och
praktik. 1991 års skattereform” in Ekonomiska Rådets rapport till kommittén för
utvärdering av skattereformen, bilaga 1, SOU 1995: 104.
Agell, Jonas, Englund, Peter and Södersten, Jan (1998). Incentives and Redistribution in the
Welfare State. The Swedish Tax Reform, London: Macmillan.
Barro, Robert J. and Charles J. Redlick (2009). “Macroeconomic Effects from Government
Purchases and Taxes.” NBER Working Papers 15369.
de Haan, Jacob, Sturm, Jan-Egbert and Volkerink, Björn. (2004). “How to Measure the Tax
Burden on Labor at the Macro-Level?” in Peter Birch , Sørensen (ed.), Measuring the
Tax Burden on Capital and Labour. Cambridge: MIT Press.
Du Rietz, Gunnar (1994). Välfärdsstatens finansiering [The Financing of the Welfare State],
Stockholm: City University Press.
Du Rietz, Gunnar, Johansson, Dan and Stenkula, Mikael (2010). ”The Marginal Tax Wedge
of Labor in Sweden from 1861 to 2009.” Ratio Working Paper, forthcoming.
Edvinsson, Rodney (2005). Growth, Accumulation, Crises. With New Macroeconomic Data
for Sweden. Ph.D. thesis in Economic History, Department of Economic History,
Stockholm University, Stockholm: Almquist & Wiksell International.
Engen, Eric and Skinner, Jonathan (1996). ”Taxation and Economic Growth”, National Tax
Journal, vol. XLIX (4), p. 617–642.
Feldt, Kjell-Olof (1991). Alla dessa dagar: i regeringen 1982–1990. Stockholm: Norstedt.
Genberg, Torsten (1942). ”Skatteutvecklingen sedan sekelskiftet”, Stockholm:
Skattebetalarnas Förening.
Gårestad, Peter (1987). Industrialisering och beskattning i Sverige 1861–1914
[Industrialization and Taxation in Sweden 1861–1914], Ph.D. thesis in Economic
History, Department of Economic History, Uppsala University, Uppsala studies in
Economic History 28.
Jakobsson, Ulf and Normann, Göran (1972). Inkomstbeskattningen i den ekonomiska
politiken, En kvantitativ analys av systemet för personlig inkomstbeskattning 1952–
1971, Stockholm and Uppsala: Industriens Utredningsinstitut and Almqvist &
Wiksell.
Johansson, Dan (2004). ”Skatternas utveckling, omfattning och fördelning” [”The
Development of the Taxes, their Extension and Distribution”], in Johansson, Dan,
Karlson, Nils and Johnsson, Richard (eds.), Skatter och värdighet [Taxes and Dignity],
Stockholm: Ratio.
Koester, Reihard and Kormendi, Roger (1989). “Taxation, Aggregate Activity and Economic
Growth: Cross Country Evidence on Some Supply Side Hypotheses”, Economic
Inquiry, vol. 27, nr 3, pp. 367–387.
Konjunkturinstitutet [National Institute of Economic Research](2009). Analysunderlag,
december 2009.
http://www.konj.se/download/18.5f70f696125498bca9b800012708/AU_dec2009hem.
pdf.
Lindbeck, Assar (1997). The Swedish Experiment. Stockholm: SNS.
Löwnertz, Susanne (2003). ”De svenska skatternas historia” [”The History of the Swedish
Taxes”], Stockholm: Riksskatteverket.
Lodin, Sven-Olof (1975). ”1970-talets skatteomläggningar - varför så många?” [”Tax reforms
in the 1970s – why so many?”], Ekonomisk Debatt, vol. 8. pp. 448-467.
28
McKee, Michael., Visser, Jacob, and Saunders, Peter (1986). “Marginal Tax Rates on the Use
of Labor and Capital in OECD Countries”, OECD Economic Studies, no. 7, pp. 45–
101.
Nordling, Danne (1989). Skattetryck och skattepolitik. En introduktion för skattebetalare,
Stockholm: Skattebetalarnas förening.
OECD (2006). Fundamental Reform of Personal Income tax.Paris: OECD.
Padovano, Fabio and Galli, Emma (2002). “Comparing the Growth Effects of Marginal vs.
Average Tax Rates and Progressivity”, European Journal of Political Economy, vol.
18, no. 3, pp 529–544.
Peacock, Allan and Wiseman, Jack (1961). The Growth of Public Expenditure in the United
Kingdom”, Princeton: Princeton University Press.
Rodriguez, Enrique (1980). Offentlig inkomstexpansion: En analys bakom de offentliga
inkomsternas utveckling i Sverige under 1900-talet, Ph.D. thesis in Economic History,
Department of Economic History, Uppsala University, Lund: CWK Gleerup.
Riksskatteverket (2002) Skattestatistisk årsbok 2002. Stockholm: Riksskatteverket.
SAF (1976–2000). ”Fakta om Sveriges ekonomi”, rapportserie. [Swedish Employer’s
Association (1976–2000). ”Facts about the Swedish Economy”, Report series.]
SFS 1902, förordning nr 84. [Swedish Statutes 1902, regulation, no. 84]
SFS 1910, förordning nr 115. [Swedish Statutes 1910, regulation, no. 115]
Skattebetalarnas Förening (1966–2009). ”Fakta för skattebetalare”, rapportserie. [Taxpayers’
Association (1966-2009). ”Facts for Taxpayers”, Report series]
Statens offentliga utredningar [Swedish Government Official Reports series] 2001: 24. Ur
fattigdomsfällan [Out of the Poverty Trap], Stockholm: Fritzes.
Statistisk Årsbok (1920), tabell 182. [Statistical Yearbook of Sweden (1920), Table 182].
Statistisk Årsbok (1925), tabell 180. [Statistical Yearbook of Sweden (1925), Table 180].
Statistisk Årsbok (1928), tabell 205. [Statistical Yearbook of Sweden (1928), Table 205].
Statistisk Årsbok (1940), tabell 214. [Statistical Yearbook of Sweden (1940), Table 214].
Statistisk Årsbok (1952–2010) [Statistical Yearbook of Sweden (1952–2010)].
Svenskt Näringsliv (2002–2007). ”Fakta om Sveriges ekonomi”, rapportserie. [Confederation
of Swedish Enterprise (2002-2007). ”Facts about the Swedish Economy”, Report
series.]
Söderberg, Hans (1996). Inkomstskattens utveckling under 1900-talet. En vägvisare för
skatteberäkningar 1921–1996, Stockholm: Skattebetalarnas Förening.
29