Description
The Turnaround of the Swedish Economy Lessons from Business Sector Reforms
IFN Policy Paper No. 73, 2015
The Turnaround of the Swedish Economy:
Lessons from Business Sector Reforms
Fredrik Heyman, Pehr-Johan Norbäck and Lars
Persson
The Turnaround of the Swedish Economy:
Lessons from Business Sector Reforms
Fredrik Heyman
*
, Pehr-Johan Norba? ck,
*
and Lars Persson
*
November 2015
Abstract: How can a country overcome a long period of stagnant productivity growth in its business sector and
reach its growth potential? Sweden during the 1970–2010 period can serve as an example to help other countries
understand how to efficiently reform a business sector. In the 1990s, Sweden implemented a reform package that
ignited a successful reorganization of a business sector that had faltered for decades. To understand the economic
forces behind this process, we first survey the industrial restructuring literature and then examine the reform
package using Swedish matched plant-firm-worker data. The removal of barriers to growth for new and
productive firms and increased rewards for investment in human capital were crucial to the success of Sweden’s
reforms. Turning to cross-country evidence, we find that an increase in the World Bank’s Doing Business
indicator is associated with an increase in GDP per capita. These findings suggest that policymakers have much
to learn from country case studies and from analyses based on the World Bank’s Doing Business indices. Our
study of the Swedish experience can be a valuable case study for developing countries that are attempting to
promote growth by developing their business sectors and for countries that are facing economic problems in the
aftermath of the financial crisis.
JEL: D22, E23, J 21, J 23, K23, L11, L16, L51
Keywords: regulations, allocative efficiency, productivity, job dynamics, matched employer-employee data, industrial
structure and structural change
* Research Institute for Industrial Economics (IFN), P.O. Box 55665, SE-102 15 Stockholm, Sweden,
[email protected], [email protected], and [email protected].
Acknowledgements: We thank Vincenzo Denicolò, Nicola Fuchs-Schündeln, Magnus Henrekson, Assar Lindbeck, Per
Skedinger and Michelle Sovinsky for their helpful comments. We also thank Fredrik Andersson, Malin Olsson Tallås, Olga
Pugatsova and Hanna Thunström for their excellent research assistance. Funding: We are grateful to the Marianne and
Marcus Wallenberg Foundation for its financial support. Fredrik Heyman gratefully acknowledges financial support from
the Swedish Research Council for Health, Working Life and Welfare (FORTE) and from Torsten Söderbergs Stiftelse. Pehr-
J ohan Norbäck and Lars Persson gratefully acknowledge financial support from the Wallander and Tom Hedelius Research
Foundation and from Vinnova.
2
1. INTRODUCTION
Although the economic policy debate has long centred on the importance of macroeconomic policies, a growing awareness
acknowledges that the microeconomic functioning of markets is vital. This shift in focus can be illustrated in the following
quote from Kaushik Basu, Senior Vice President and Chief Economist of the World Bank, in the foreword to the Doing
Business 2015 report:
1
“The public discourse on economic policy is overwhelmingly focused on fiscal measures, monetary interventions, welfare
programs, and other such highly visible instruments of government action. Thus when an economy does poorly, a
disproportionate amount of our debate centers on whether or not it needs a fiscal stimulus, whether there should be liquidity
easing or tightening, whether its welfare programs have been too profligate or too paltry and so on. What gets much less
attention but is equally – and, in some situations, even more – important for an economy’s success or failure is the nuts and
bolts that hold the economy together and the plumbing that underlies the economy.”
A malfunctioning system of micro-economic regulation can hinder economic growth and make standard macro policies
less effective or even lead to deep macroeconomic crises. In this paper, we will argue that the Swedish experience from
1970?1980 is an illustrative example of how a dysfunctional system of micro-economic regulations can lead to a severe
macroeconomic crisis. In the 1980s and early 1990s, Sweden faced its most severe economic crisis in the post-war period:
Swedish companies lost in global competition, while the state became very highly leveraged. The 1991–1994 period was
characterized by a substantial decline in GDP and increasing unemployment. Our institutional and theoretical analysis
suggests that the substantial slowdown of the Swedish economy in the 1980s and the subsequent economic crisis in the early
1990s were to a large extent caused by increasingly interventionist business policies, reflecting a change in world view to
one in which economic incentives and private entrepreneurship not associated with large firms were regarded with suspicion
(Lindbeck, 1997).
However, in the 1990s, Sweden undertook a major structural reform package, which largely aimed to improve the
microeconomic functioning of markets, and ignited a successful industrial reorganization process in the business sector. The
Swedish reforms were implemented, and subsequent governments did not reverse them, which is a fundamental aspect of
their success. Ireland, Sweden and the US experienced the highest labour productivity growth in the OECD during the
1995–2011 period (OECD (2013)). Sweden’s productivity growth was primarily driven by factors that increased the
effectiveness of its business sector. This process took place while the welfare state was largely preserved. Although
inequality in Sweden, as in many other countries, has increased, it still exists at a low level compared to that in other
countries.
To understand the economic forces associated with the Swedish restructuring process, we first survey the industrial
restructuring literature in search of mechanisms that are important in explaining firm, employment, and productivity
dynamics. The overview indicates that productive, expanding firms are typically associated with active owners and up-to-
date and incentive-based management practises. Consequently, economically efficient decisions are made, and well-
functioning business cultures are developed, leading to a motivated workforce. The employment of skilled workers and the
early adoption of new technologies have also been shown to create competitive advantages in both local and global markets,
thereby spurring productivity growth. Start-ups and expansions are associated with high degrees of uncertainty, meaning
that many businesses fail; thus, the observed number of highly successful and expanding ventures is low, but they are still
1
See also two recent articles in the J ournal of Economic Perspectives (Besley, 2015; Thimann, 2015) stressing the importance of the microeconomic foundations.
3
an important factor in productivity and employment growth. Firms’ productivity and employment also crucially depend on
external factors, such as institutions and access to production factors
The picture emerging from our institutional analysis shows that Sweden developed good economic institutions in the late
1800s and experienced high levels of sustained growth in the 1870?1970 period. In the 1970s, however, considerably more
interventionist policies were developed, such that Sweden had one of the most regulated business sectors in the developed
world in the 1970s and 1980s.
2
Growth then slowed substantially, which led to an economic crisis in the early 1990s. This
experience paved the way for several reforms that intended to increase the efficiency of the Swedish economy, particularly
the Swedish business sector. Our main finding from this institutional and theoretical examination is that the removal of
barriers to entry and growth for new and productive firms and increased rewards for investments in human capital and for
efforts in the workplace were crucial to the success of these reforms. These actions, we argue, led to remarkable growth in
productivity and employment in the Swedish business sector in the two decades that followed.
We then describe empirical results found in an accompanying paper, Heyman, Norbäck and Persson (2015), and discuss
how their results provide indicative support for our institutional predictions. Heyman, Norbäck and Persson (2015) is based
on Swedish matched employer-employee data over the period 1996-2009 from Statistics Sweden (SCB), which allows for
analysing issues related to firm employment and productivity dynamics possible in greater detail than has been possible in
most other international studies.
3
First, there was an increase in allocative efficiency in Sweden, as measured as greater market shares for the more
productive firms in the economy, during the period 1996-2009. This suggests that the reforms mitigated the insider and
incumbency problems in the Swedish business sector and enabled more productive firms to better attract capital and
employees than they had previously. The relationship between firms’ productivity and wage increases was strengthened
over the period studied, which suggests that productive firms and productive employees became more rewarded in the
Swedish business sector. There was also an increase in jobs created in small firms, while most of the productivity gains
were created in large incumbent firms, suggesting that the reforms facilitated the division of labour between large
incumbents and small growing firms. Finally, foreign firms contributed significantly to productivity and employment
growth in the business sector during this period, which suggests that the liberalization of foreign direct investment (FDI)
was an important factor in driving the restructuring process.
Thus, a crucial element of the success of the Swedish business sector turnaround was the emergence of firms with
different ownership and organization (foreign firms and small firms). Therefore, the micro-economic reforms that made the
playing field between different ownership and organization forms more level were crucial for Sweden’s successful
transformation.
The reforms that Sweden undertook after the crisis included other measures as well. The government was forced to
operate under stricter budget rules. Monetary policy was changed so that price stability was targeted by an independent
central bank under a floating exchange rate. As interest rates fell and the krona depreciated in the mid-1990s, demand and
exports increased.
Identifying the most important factors – “micro reforms”, which restore incentives and the functioning of markets, or a
“macro” story of solid public finances and a flexible exchange rate leading to export-led growth – is of course complex. One
way of determining whether the microeconomic reforms stressed here were crucial in the Swedish economic turnaround
involves using cross-country data and proxies for the variables that capture different aspects of the reforms and then
correlating these variables with economic performance. Unfortunately, cross-country and time-series data covering a broad
range of policy indicators are rare. Still, the available data give us some basic information and can illustrate how reforms
2
See Hicks and Kenworthy (1998).
3
See Heyman, Norbäck and Persson (2015) for details of the empirical analysis.
4
relate to different outcomes – in our case, most notably GDP per capita. Using the Ease of Doing Business indicator
(measured as the distance to the top) and logged GDP per capita, we find that a one-point increase in the overall Doing
Business index is associated with approximately half a per cent increase in GDP per capita when country fixed-effects are
taken into account.
4
The above described results have several policy implications. Developing countries face the challenging task of
achieving their growth potential. Sweden’s experience during the 1970?2010 period can serve as an interesting example that
helps us understand how a business sector’s regulation and deregulation affect a country’s growth potential. In fact, the
GDP per capita level of Sweden in the 1960s corresponds with a large number of today’s developing countries, such as
Brazil, Bulgaria, Iran, Thailand, and South Africa. The Swedish case shows that imposing regulations without carefully
considering how these affect the incentives and efficiency of the business sector might be counterproductive. In particular,
our institutional and empirical analysis suggests that reforms that remove barriers to entry and growth for new and
productive firms and increase the return on capital and human capital investments, similar to those implemented in Sweden
during the 1990s, are likely to spur economic growth.
We also believe that our study is relevant for the ongoing academic and policy debate about the current crisis in Europe,
which tends to be dominated by macroeconomic and public sector considerations. Thimann (2015), for instance, notes that a
remarkable correspondence exists between impediments to growth and economic stresses in eurozone countries, such as
Greece.
5
Still, the top-level political discussion seems almost exclusively focused on macroeconomic issues, for instance,
how to handle the Greek government’s debt. The Swedish post-crisis experience suggests that the microeconomic
dimensions are crucial for turning an economy in crisis around. Therefore, to ensure recovery in these countries, the focus
on implementing long-run microeconomic reforms should increase.
2. PRODUCTIVITY AND EMPLOYMENT DYNAMICS IN SWEDISH INDUSTRY
Figure 1 compares the long-term development of GDP per capita in Sweden with that of US and a simple average of the EU
15 countries.
6
In 1970, Swedish GDP per capita exceeded the average GDP per capita of the EU 15 group but was lower
than that of the US. During the 1970–1990 period, Sweden performed worse than both the US and the EU 15 average. When
Sweden entered a severe crisis in the early 1990s, its GDP per capita fell below that of the EU 15 average.
In the years after the crisis in the 1990s, propelled by significant reforms to its economy, Sweden showed a much
stronger trend, and its GDP per capita grew faster than that of the EU 15 and kept up with US growth. At the end of the
1990s, Swedish GDP per capita again surpassed the EU 15 average. The gap in GDP per capita between Sweden and the EU
15 widened further after the turn of the millennium. In addition, Sweden clearly appears to have managed the recent crisis
better than the EU countries, as manifested by a further increase in the gap in GDP per capita as the financial crisis
unfolded. The latter observation suggests that Sweden’s performance, beginning in the mid-1990s, does not simply
represent the well-known “catch-up” phenomenon (growth increases simply because the economy takes off from a low level
of activity in which extensive idle production factors are available). Sweden’s apparent ability to sustain a relatively
4
As Besley (2015) writes in his overview article of the Doing Business project, case studies of countries complement analyses using the Doing Business
indices: “Policymakers in China or Brazil or Egypt have good reasons to be interested in how economies like Singapore or Sweden approach business
regulation without deciding blindly that they should copy these practices. The Doing Business rankings provide a way into this question – the basis
beginning a dialogue about policy reform. This discussion was not happening on a systematic basis before the Doing Business project came along.”
5
Using data fromthe World Economic Forum’s Global Competitiveness Index and the World Bank’s Doing Business dataset, Thimann (2015) finds that
several of the crisis countries have high and persistent barriers to firmformation and firmdynamics.
6
The EU 15 includes Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain,
Sweden and the United Kingdom, which were member countries of the European Union just prior to the accession of the East European countries on 1
May 2004. Several of these countries were not members of the EU (or EEC) at the beginning of the period considered. We use the term“EU 15” as a
convenient European benchmark.
5
superior performance over a very long period suggests that the Swedish economy underwent profound changes during the
1990s and that these changes laid the foundation for a sustained period of growth and a successful economic transformation.
Figure 1. Comparing long-term trends in GDP per capita for Sweden, the US, and the original 15 European Union
member countries, 1970–2013
Notes: GDP per capita is expressed in USD constant PPPs with base year 2005.
Source: OECD, National Accounts.
Let us now turn to the employment trend in Sweden during this period. Figure 2 compares the employment dynamics in
Sweden with those in the EU 15 countries and those in the US for the 1990–2011 period. Throughout the period,
employment as a share of the total labour force, the so-called participation rate, was very high in Sweden. After a substantial
post-crisis decline in the 1990s due to layoffs in the private and public sectors during the initial restructuring process, the
labour force participation rate steadily increased, and it was again substantially higher than those in the EU 15 and the US.
In particular, Sweden showed high growth in private sector employment as a share of its total labour force in the 1995–2011
period. Although substantially lower than the US, private sector employment surpassed the EU 15 levels after the financial
crises.
Figure 2. Total and private sector employment rates, 1990–2011
Notes: The EU 15 rate is based on our calculations. Private sector employment data are missing for Germany and Greece for the entire period. Total
employment data are missing for Germany in 1991 and for Greece until 1994. Private sector data for Austria are taken fromthe OECD Economic Outlook
No. 85, where historical data are only available until 2008. Employment rates are defined as employment as a share of the total labour force.
Source: OECD Economic Outlook No. 95, OECD Labour Market Statistics.
6
To understand the economic forces underlying the Swedish industrial and job restructuring processes, we will describe
the crucial economic reforms that Sweden implemented. However, first, we briefly survey the industrial restructuring
literature in search of the mechanisms through which these reforms could have had an impact on firm and job dynamics.
3. CONCEPTUAL FRAMEWORK: INDUSTRIAL RESTRUCTURING AND ECONOMIC REFORMS
To understand the restructuring processes that took place in Swedish industry in the 1990s and 2000s, we begin with an
overview of the basic economic mechanisms that have been important in explaining employment and productivity dynamics
in general.
7
Beginning from this general knowledge of the functioning of industrial restructuring processes, we analyse the
potential effects of the economic reforms undertaken in Sweden in the 1990s on employment and productivity dynamics.
This overview focuses on the economy’s microeconomic features, thus centring on what the World Bank refers to as the
“nuts and bolts” of the economy.
Figure 3 depicts a schematic picture of how firm and business sector employment and productivity can be understood.
The figure categorizes the “firm-specific factors” that firms can choose and influence. Firm-specific factors concern how
firms are organized or which business strategies are used. The overview suggests that expanding productive firms are
typically associated with active owners and up-to-date and incentive-based management practises. Thereby, economically
efficient decisions are made, and well-functioning business cultures are developed, thus leading to a motivated workforce.
Employing skilled workers and adapting new technologies early have also been shown to create competitive advantages in
local and global markets and to thereby spur productivity growth. Start-ups and expansions are often associated with high
degrees of uncertainty, meaning that many businesses will fail and, consequently, that the observed number of highly
successful and expanding ventures will be low.
The figure also includes “external factors”. Firms have no influence over these factors; however, these factors can
directly and indirectly affect firm productivity and employment through the limitations that they set or the incentives that
they provide regarding firm-specific choices. In our study of the Swedish business sector, we focus on external factors in a
large reform package implemented in Sweden in the early 1990s, which included reforms of the labour and product markets,
tax reforms and the removal of FDI barriers. We should also highlight that the information and communication technology
(ICT) revolution took place during this period. This external factor was very important to the industrial restructuring process
in most developed countries during this period. Finally, firm-specific and external factors affect firm performance in terms
of measurable productivity and employment dynamics. Different methods can be used to sum up these dynamics at the
aggregate business-sector level, as shown at the bottom of Figure 3.
7
See Acs and Audretsch (2005), Caves (1998), Santarelli and Vivarelli (2007), and Sutton (1997) for an overview of the literature on market structure and
firmdynamics. For specific articles, see, e.g., Audretsch (1991), Bartelsman et al. (2005), Ericson and Pakes (1995), Hjalmarsson (1974), Hopenhayn
(1992), Klepper (1996), J ovanovic (1982), and Luttmer (2007). In addition, see Nelson and Winter (1982) for an analysis of firmgrowth processes with
bounded rational decision makers. See Li and Rama (2015) for an analysis of firmdynamics, productivity growth, and job creation in developing countries.
7
Figure 3: The explanatory factors behind productivity and employment trends in the business sector
3.1. Firm-specific Factors
To compete in the marketplace, firms need to undertake efficient decisions in several dimensions. Short-term decisions
include optimal pricing and efficient marketing. Medium-term decisions involve aspects such as correct location of activity
and the hiring of productive staff. Finally, long-term decisions involve decisions on updated R&D and the optimal
organizational form.
8
We discuss some of the more important firm-specific factors below.
3.1.1. Business Strategy and Organization
Productivity and employment dynamics in firms essentially depend on the changes that firms – and their rivals – make to
their business strategies and organization. Firms that have good business ideas need to decide on how to grow – but what
should be produced internally, and what should be bought on the market when they expand? On the one hand, economies of
scale and scope imply that increasing firm size reduces costs and increases profits. On the other hand, larger firms face
problems related to free riding, a lack of control over activities, and lost motivation among staff, all of which, in turn, limit
the optimal firm size.
9
The optimal firm size also differs between individual firms and between industries and depends on factors such as
technology (ICT), market conditions (demand levels), and presiding institutions and laws (corporate tax system). Start-ups
8
See Besanko et al. (2003) for an overview.
9
The free-rider problemimplies that actors do not dare invest in development and hard work, as they then risk having competitors benefit fromthe returns
on these investments.
External factors:
• Labour market
• Corporate ownership market
• Product market
Firm-specific factors:
• Strategy and Organization
• Ownership and Management
• Capital and Human Capital
Firm characteristics:
• Productivity dynamics
• Employment dynamics
Private sector:
• Productivity dynamics
• Employment dynamics
8
and expansions are also associated with high degrees of uncertainty and problems of asymmetric information. Therefore,
many businesses will fail and, consequently, the observed number of highly successful and expanding ventures will be low.
Overall, these results indicate that firms with strong business ideas typically increase their productivity levels, but they
might not necessarily increase their employment levels due to labour saving or due to the outsourcing of non-core business
activities.
10
3.1.2. Ownership and Management
Expanding productive firms are typically associated with active owners and up-to-date and incentive-based management
practises. Thereby, economically efficient decisions are made at the right time. Moreover, active ownership typically creates
a well-functioning business culture, leading to a motivated work force.
11
Why then do inefficient firms not implement more efficient management? First, some firms are family owned with
management that is difficult to replace. In addition, firms possibly face problems of corporate control, whereby managers use
their superior information to shirk their responsibilities or to hide their incompetence. Different incentives and monitoring
systems have been developed to mitigate these problems (see, e.g., Tirole, 2006).
3.1.3. Capital and Human Capital
Some firms are able to expand and maintain high productivity because they educate and hire productive employees and
invest in high-quality capital. Efficient human resource management enables firms to acquire talent and further develop their
skills so that they can facilitate the generation of high profits and firm expansion.
12,13
The implementation of ICT is a
prominent example of how the adoption of new technology was able to spur firm growth and productivity.
14
The acquisitions
of small growing firms are another important explanation for growth in expanding firms’ employment and productivity.
15
Moreover, multinational enterprises (MNEs) use their firm-specific assets to undertake FDIs in different countries. FDIs are
typically achieved by either setting up new plants (greenfield investments) or acquiring existing domestic target firms (cross-
border M&As). These FDIs often have positive externalities on labour through higher wages and on local firms through
knowledge spillovers.
16
Let us now examine how the changes in external factors caused by the economic reform package undertaken in Sweden
in the 1990s affected firm performance.
10
The literature that addresses firmformation and size was founded by Coase (1937) and was further developed by Williamson (1979). Grossman and Hart
(1986) and Hart and Moore (1990) then developed formal frames of analysis to study these questions, focusing on how the division of ownership affected
the different stakeholders’ incentives to invest in a firm’s development. See Rajan and Zingales (2001) for an application for entrepreneurship and enterprise
development. See the “Part One: Firmboundaries” chapter in Besanko et al. (2003) for an overview.
11
Bloomand Van Reenen (2007) find that firms with higher-quality management are more productive, and they argue that the lower aggregate productivity
in the UK and France compared with that in the US can be explained by the lower level of competition in the UK and France, where, in the absence of
competitive pressure, inefficient firms are not forced out of the market. Bertrand and Schoar (2003) follow individuals who have been CEOs at different
companies and show that CEO quality has an effect on how profitable these firms are. Lazear (2000) and Bandiera et al. (2007 and 2009) also show the
connection between good leadership and high productivity.
12
See Gibbons and Roberts (2013) and Murphy and Topel (1990) for an overview.
13
Ilmakunnas et al. (2004) use Finnish data and show that productivity increases with the employee education level and age. At the same time, Fox and
Smeets (2011) show that considerable differences in productivity remain between firms when they control for education level, gender, work experience,
and employment duration. Thus, labour force quality explains some, though far fromall, differences in productivity between firms.
14
The findings in Van Ark, O’Mahony and Timmer (2008) suggest that the slower productivity development in the EU compared with that in the US can
partly be explained by lower investments in ICT.
15
Indeed, Maksimovic and Phillips (2001) show that a large share of the plants in the US manufacturing industry change owners each year (up to 7 per cent
in some years) and that the productivity levels of these plants increase.
16
See Barba Navaretti and Venables (2004)for an overview of the MNE literature, J avorcik (2015) on the literature on FDI and job creation and Saggi
(2002) on the literature on FDI and technological spillov ers.
9
3.2. External Factors: The expanded Regulations in the 1970s and 1980s and the Structural Economic Reforms of the
1990s
We first describe the institutional setting of the Swedish business sector and then turn to the crucial reforms that were
implemented in the 1990s. We rely on detailed descriptions of the Swedish business sector and the policy reforms that
affected firms in Sweden, as described by, e.g., Bergh (2014), Bergh and Erlingsson (2006), Calmfors (2012), Edquist and
Henrekson (2013), Henrekson and J akobsson (2005), J onung et al. (2008), Lindbeck (1997), and the references therein.
Let us begin with a brief description of the development of economic institutions in Sweden prior to the reforms. Staying
out of two world wars and engaging in international trade by exploiting its abundant natural resources while developing
efficient institutions, Sweden experienced a long period of sustained growth, the so-called “golden years” of 1870–1970. At
the end of this century-long period, Sweden was fourth in the OECD rankings of GNP per capita. In the decades after the
Second World War, a relatively rapid GDP growth rate was combined with full employment and a fairly egalitarian
distribution of income.
In the 1970s, government policies became increasingly interventionist under the influence of the more radical political
ideas that emerged during the decade. Tight labour market regulations were implemented in the early 1970s. So-called
“solidaristic wage policies” led to a compressed wage structure, and workers’ wages became detached from individual firm
productivity. Marginal tax rates gradually increased, ultimately culminating in a 1971 tax reform that made Sweden’s tax
rate very high in comparison with those of comparable countries.
In the business sector, the government, trade unions and bank-related business groups embodied an explicit tripartite
negotiating culture. A fairly small number of dominating owners or ownership groups of corporations acknowledged and
accepted that the government would use its political power to implement far-reaching welfare reforms as long as the labour
movement abstained from socializing the industrial sector. Moreover, the government attempted to influence aggregate
savings, the credit supply and investment through public sector savings, capital market regulations, taxes and subsidies,
which all affected the functioning of the business sector. As noted by Lindbeck (1997), this approach mirrors a world view
in which markets, economic incentives and private entrepreneurship not associated with large firms are regarded with
suspicion.
These interventionist policies reduced the efficiency of the economy and likely served as an important factor in
Sweden’s inferior performance compared with those of the EU 15 countries and the US during the 1970–1990 period in
terms of GDP per capita growth. Internal problems in the Swedish model and external shocks eventually led to deep
economic crises that included a significant decrease in output and soaring unemployment in the early 1990s. In response,
economic-efficiency and growth-induced institutional reforms were undertaken in the late 1980s and 1990s. In addition,
macroeconomic policy reforms were implemented in the 1990s to reduce the inflationary bias in the Swedish economy.
These reforms included the establishment of an independent central bank and a floating currency.
Most of the literature has focused on the importance of macroeconomic reforms. By contrast, this study focuses on the
reforms that were undertaken to improve resource allocation and the microeconomic functioning of the markets in response
to the underperforming Swedish economy.
17
Notable reforms included the decentralization of the wage negotiation system
and the liberalization of temporary work contracts, the deregulation of the product market, greater openness to inward FDI
and tax system reforms (Bergh, 2013, Calmfors, 2012, Edquist and Henrekson, 2013 and Lindbeck, 1997).
In the following sub-sections, we proceed with a detailed review of the Swedish reforms before attempting to present
empirical evidence showing that these reforms increased economic efficiency in the Swedish business sector. To provide
such evidence, we use Swedish micro data to study different facets of productivity and employment dynamics.
17
See Besley (2015) and Thimann (2015) for recent articles that focus on the importance of microeconomic foundations.
10
3.2.1. The Labour Market
Labour market regulations significantly affect firm employment and productivity development. On the one hand, labour
market institutions can increase hiring and productivity by reducing matching and search problems in the labour market. On
the other hand, rigid labour markets may offer too much protection to insiders, thereby hampering creative destruction
processes and, to a lesser extent, rewarding productive labour and firms. Various types of labour turnover costs give insiders
market power, which has implications for talent allocation, work incentives, and employment and unemployment patterns
(see Lindbeck and Snower, 2001, for an overview). In particular, such insider market power might distort incentives for firm
development, education, and efforts in the workplace. We refer to this problem as the insider-outsider problem of the labour
market.
Let us now use this background to discuss the impact of the crucial labour market reforms in Sweden in the 1990s, the
decentralization of the wage negotiation system and the liberalization of temporary work contracts on the efficiency of the
restructuring of the Swedish business sector.
3.2.1.1. The Decentralization of the Swedish Wage Negotiation System and the Liberalization of Temporary Work
Contracts
After the Second World War, wage bargaining was highly centralized in Sweden. In the 1950s and 1960s, economy-wide
wage increases were negotiated centrally between the Swedish Employers’ Confederation and the Trade Union
Confederation. In the 1960s and 1970s, wages were set according to solidaristic wage policies under the principle of “equal
pay for equal work”, whereby wages would be equalized between sectors for similar tasks and occupations. Ideally, the
system would have mimicked a competitive labour market, in which low productivity firms would be driven out of the
market, thus freeing labour to seek high-productivity firms that were able to support higher wages. In practice, however,
considerable wage compression occurred as ambitions moved from equity goals to promoting more outright equality (Davis
and Henrekson, 2000 and Lindbeck, 1997). A market mechanism through which high-productivity firms could attract labour
by paying higher wages did not exist; instead, active labour market policies were pursued, whereby resources were allocated
to help the unemployed gain new competences and to reduce frictions and search costs in the labour market. How well the
government was able fulfil this allocative task is debatable. With constraints on wage setting, the ability to incentivize
workers in firms was also hampered.
In 1974, a new employment protection law (LAS) was implemented. The law mandated that employees could not be
fired without reasonable cause, such as abuse or a lack of work opportunities. Insiders were also favoured with respect to
firing and hiring procedures through the so-called “last in, first out” rule, which further reduced workers’ incentives to
change jobs. Temporary contracts also became limited. Figure 4 shows that the introduction of LAS had a significant impact
on the so-called Allard index of the strictness of employment protections (Allard, 2005), which nearly doubled during this
period. This measure of the strictness of employment protections continued to increase until the beginning of the 1990s. The
average employment protection in the EU 15 countries also increased over this period, though not to the same extent as in
Sweden. The US labour market continued to maintain a low level of employment protection.
11
Figure 4. Employment protection legislation, 1950–2003
Notes: Scale 0–5, the higher the index is, the stricter the employment protection legislation is. The EU 15 values are based
on our calculations, and they exclude Luxembourg.
Source: Allard (2005).
The push for radical reforms produced a tenser climate in the relations between the unions and the Employers’
Confederation. Centralized bargaining for private-sector, blue-collar workers gradually broke down in the 1980s, and it was
replaced by uncoordinated industry-level bargaining. Intermediate industry-level bargaining is a form of collective
bargaining that should be more conducive to wage inflation.
18
In 1990, the Employers’ Confederation attempted to
introduce a more decentralized system. However, this attempt failed, and, instead, a fully centralized wage stabilization deal
was negotiated for 1991–1992. In 1994, state-owned firms joined the employers’ organization, which weakened the political
influence in wage setting (Nycander, 2008). In 1997, the so-called Industry Agreement was concluded. The agreement
included a system that continued industry-level bargaining but with strong informal coordination based on pattern
bargaining with the manufacturing sector to conclude initial wage agreements in a bargaining round. This system
established a norm for wage increases for others to follow. The reformed wage-bargaining system turned out to be
consistent with lower nominal wage increases than those in the past. Moreover, it allowed for greater individual wage
flexibility (Calmfors, 2012). Sweden thus progressed from a more coordinated wage negotiation system than those in other
EU countries in the 1980s to a moderately coordinated wage negotiation system in the 1990s.
In 1992, a major employment protection reform was implemented that permitted staffing agencies (Skedinger, 2010),
and the regulations concerning temporary work were relaxed. This development created what is referred to as the dual
Swedish labour market, with strong employment protections for regular workers and weak employment protections for
temporary workers. This reform was also evident in the Allard index of employment protection, which declined
significantly. Another measure of the strictness of employment protections from the OECD demonstrates that the strictness
of employment protections concerning temporary contracts was significantly reduced in Sweden – from a very high level in
1985 to a very low level in 2010. However, the strictness of the employment protections concerning regular contracts
remained at a relatively high level over the same period.
We conclude our description of the reforms of the labour market as follows:
18
Both highly coordinated wage bargaining and decentralized firm-level bargaining deliver higher wage moderation: highly coordinated bargaining does so because wage setters
are forced to make economy-wide considerations, and decentralized bargaining does so because wage setters have to consider competitive pressures. This hypothesis seems to be
borne out by the high wage increases in Sweden in the 1980s (Calmfors, 2012).
12
Conclusion 1. The incentive and insider-outsider problems in the Swedish labour market may have been mitigated by the
labour market reforms undertaken in the 1990s. These reforms may also have improved firms’ flexibility and thereby their
ability to adjust their workforce and invest in and reward human capital.
3.2.2. Product Market Regulation
The absence of artificial barriers to entry and expansion is crucial for employment and productivity growth. Incumbent
firms have incentives to exploit their market power to protect their market share by preventing rivals from expanding and
new firms from entering their markets. For instance, incumbent firms can practice different forms of predatory behaviours,
such as engaging in exclusive dealing contracts or input cartels, lobbying for special restrictions on entry, or making entry-
deterring acquisitions. Even if incumbents are ineffective, they may not be replaced by more productive entrepreneurs due
to excessive barriers to entry.
We refer to these product market problems as the problem of weak creative destruction. Well-functioning competition
policy and legislation can mitigate such entry-deterring and predatory problems (see Motta, 2004 and Tirole, 2006).
Moreover, a well-functioning competition policy must ensure that innovative firms are not deterred by rivals to expand and
are able to reap the benefits of temporary market power. Moreover, these innovative firms need to put competitive pressure
on firms that are lagging behind (Aghion et al., 2005, Norbäck and Persson, 2012, and Vives, 2008).
3.2.2.1. The Deregulation of Product Markets in Sweden
Throughout most of the twentieth century, many product markets in services in Sweden were public monopolies. Thus, new
firms had no or very few opportunities to enter these markets, and consumers’ influences were limited (SOU 2005:4).
Moreover, the competition law was rather lax for a long period. The first competition legislation was implemented in 1925,
which enabled authorities to investigate companies that could have monopolistic characteristics. In 1946, the legislation
became stricter and centred on the idea of monitoring competition restrictions in the business sector.
19
Nevertheless, from an
international perspective, the competition law was very lax. A new Competition Act was implemented in 1993 that was
based on three cornerstones: the prohibition of restrictive agreements, the prohibition of abuse of dominance, and the
prohibition of control of concentrations (mergers). This new competition law indicated that the competition policy had
become much stricter.
In the 1980s, discussions concerning how to reform the Swedish welfare state became increasingly intense. The centre-
right government that came to power in 1991 was seemingly intent on implementing an economic policy based on extensive
deregulation in response to the country’s economic crisis in the 1990s. However, the possibility of making such reforms,
e.g., the deregulation of the air traffic system, the electricity market, and the postal service, had already been thoroughly
investigated, and government policies had previously been outlined in government white papers prepared by the Social
Democratic government in the late 1980s and early 1990s. Moreover, in 1993 (SOU 1993:16), the so-called Lindbeck
Commission presented a number of proposals to improve the efficiency and functioning of markets in Sweden (Lindbeck et
al., 1994). Overall, the intensity of competition increased substantially in many Swedish product markets during the 1990s.
The OECD has long calculated an index of the “knock-on” cost that regulations in the service and utility industries
impose on manufacturing industries. This index is shown in Figure 5, where we see that regulations on utilities and services
19
This monitoring was enabled not only through investigations but also through the registration of cartels in a public record called the Cartel Register (CR), or Kartellregistret.
The idea behind the CR was to highlight the extent of anti-competitive agreements in the Swedish business community and, in doing so, to help prevent the potential adverse
effects of such agreements on competition.
13
imposed high additional costs on manufacturing in the 1970s and 1980s, but these costs decreased sharply as Sweden began
to deregulate in the late 1980s and early 1990s. Thus, beginning in the mid-1990s, the costs of regulation in the services and
utilities sectors were substantially lower in Sweden than the average of the EU 15, and they were even lower than such costs
in the US and the UK.
Figure 5. Regulation impacts, 1975–2007
Notes: Measurement of potential costs of anti-competitive regulation in intermediate input sectors. The EU 15 values are
based on our calculations, and they exclude Luxembourg.
Source: OECD Indicators of Regulation Impact.
Since the late 1990s, the OECD has also constructed a system of indicators to measure on-going developments in
product market regulation (PMR) across the OECD countries (Wölfl et al., 2009). For Sweden, the “barriers to
entrepreneurship” category has improved the most. Particularly between 1998 and 2008, considerable improvements were
made to licensing and permit systems and communications. Furthermore, simplifications of rules and procedures were
made; certain legal barriers were removed; antitrust exemptions were allowed; and barriers to competition in network
sectors and services were reduced.
The product market reforms substantially reduced the power of the iron triangle of the Swedish business sector: the
government, incumbent firms, and unions. A crucial feature of these product market reforms was that they not only made it
easier for new firms to enter industries but also made it more difficult for inefficient firms to remain in the product market.
We can summarize our description of Sweden’s product market reforms as follows:
Conclusion 2: The deregulation of the Swedish product markets and the strengthened competition policy may have
mitigated the weak creative destruction problem in the Swedish business sector. These reforms may have forced inefficient
firms out of the market, thus making room for more productive entrants, but they also may have caused incumbent firms to
reach their potential through more intensive development.
14
3.2.3. Corporate Taxation
The tax system affects both incentives for firms to invest and the firm formation process. Mirrlees et al. (2011) show that
the tax system’s treatment of the cost of capital can distort firms’ investment incentives. At zero inflation and for an asset
for which the true decline in value over its lifetime matches the tax depreciation schedule, the corporate tax rate does not
affect the required rate of return for corporate investments that are financed by debt. However, investments financed by
equity are affected. Shareholders require a positive rate of return to compensate for the income that they could have earned
by investing in an interest-bearing asset. However, this “opportunity cost” of equity financing is not deductible from taxable
profits, which means that debt financing is favoured over equity financing, thus discouraging corporate investments
financed by equity.
Several contributions in the corporate tax literature also focus on the progressiveness of the personal income tax
schedule as an obstacle to firm formation activity (e.g., Gentry and Hubbard, 2000). Keuschnigg and Nielsen (2004a, b)
focus on the effects of various tax policies when entrepreneurs face financial constraints and must enter into contracts with
venture capitalists under conditions of one-sided or two-sided moral hazards. Haufler et al. (2014) show that the tax system
might create distortions in the type of projects that entrepreneurial start-up firms undertake.
20
Thus, the corporate tax system runs the risk of reducing the incentives for investments and of distorting the market’s
efficiency by favouring certain types of corporate ownership over others, such as incumbents over start-ups. We refer to this
problem as the tax incentives and discrimination problem in the business sector. Let us now use this background to discuss
the implications of the Swedish reforms for the corporate ownership market to predict how these reforms might have
affected the performance of the Swedish business sector.
3.2.3.1 Corporate Taxation Reforms in Sweden
Owners of corporations in Sweden are taxed on two levels. They pay taxes indirectly through corporate taxes, and they pay
taxes on capital gains. Corporate taxes increased substantially in Sweden during the 1970s and the 1980s, leading to very
high corporate taxes from an international perspective. Due to the increased awareness of the negative effects of high
taxation on business activities and increased international competition, the tax was reduced from 52 to 30 per cent in the
1990–1991 tax reform package. The rate was then further reduced to 28 per cent in 1994.
The Swedish tax rate on capital gains for long-term holdings was zero until 1965. The tax changes implemented in 1976
sharply increased the top marginal tax rate to more than 30 per cent, and it peaked in 1979 at nearly 35 per cent. Thereafter,
it decreased to approximately 25 per cent prior to the 1990–1991 tax reforms (Stenkula et al., 2015). The 1990–1991 tax
reforms made all capital gains fully taxable irrespective of the holding period. However, capital gains were no longer taxed
jointly with labour income; instead, a separate capital income tax was implemented at a flat rate of 30 per cent.
21
Importantly, until 1991, the Swedish tax system favoured large firms and institutional ownership (e.g., pension funds
and insurance companies). The 1991 tax reform and subsequent minor reforms considerably levelled the playing field for
different combinations of owners and sources of financing (Davis and Henrekson, 1999 and Edquist and Henrekson, 2013).
20
The empirical literature on the productivity effects of corporate taxation has mostly suggested effects fromindirect channels, such as R&D and capital investment. The
relationship between tax policy (mostly R&D tax credits) and the volume or location of R&D across countries and the US is reviewed by Hall and van Reenen (2000). See
Auerbach (2002), Gordon and Hines (2002), Hasset and Hubbard (2002), and Hines (2005) for analyses on the relationship between taxation and investments. Evidence of the
effects of taxation on industry entry and exit rates is presented by Da Rin et al. (2011), Djankov et al. (2010), and Kneller and McGowan (2012). Carroll et al. (2000, 2001)
examine the effects of the US tax reforms in the 1980s on the investment and hiring decisions of small businesses and find significant effects. Arnold et al. (2011) demonstrate
that corporate taxation has a direct effect on firmproductivity growth by lowering the growth rate of firms that are in more profitable industries. Using data for 11 European
countries, Gemmell et al. (2013) find evidence that productivity growth in small firms is lower when the corporate tax rates are higher.
21
From1992–1993, this separate capital income tax rate was temporarily reduced to 25 per cent, and it was temporarily reduced to 12.5 per cent in 1994.
15
The reforms in the 1990s generated a tax system that was far more positive for individuals who wished to start, develop, and
act as controlling firm owners compared with the situation in the 1970s and 1980s.
However, only examining the (marginal) tax rates to judge a tax system’s effect on firm performance is insufficient.
Evaluating a tax system’s effect on corporate capital investment is a complicated task. Many aspects, such as the project
type and the financing form, need to be considered. Here, the devil is in the details. A generally accepted method of
evaluating a capital tax system is to calculate the marginal effective tax rate (METR) on capital based on the method
originally presented by King and Fullerton (1984).
Devereux et al. (2002) use a simplified version of this method to compare the marginal effective corporate taxes for a
number of EU countries, J apan and the US in 1982 and 2001. These comparisons are shown in Figure 6. The figure shows
that Sweden had the second highest rate in 1982 at approximately 53 per cent but the second lowest rate in 2001 at
approximately 22 per cent. Hence, corporate taxes were reduced from very high levels to low levels during the reforms of
the 1990s.
Figure 6. Effective marginal corporate tax rates
Notes: Replication of author’s Figure 5, pg. 462, “Calculations based on a hypothetical investment in plant and machinery
for one period, financed by equity or retained earnings (but not debt). Taxation at the shareholder level is not included. The
project is expected to break even, i.e., there is no economic rent. Other assumptions are that the real discount rate is 10 per
cent, the inflation rate 3.5 per cent, and the depreciation rate is 12.25 per cent”.
Source: Devereux et al. (2002)
Using the King and Fullerton (1984) approach, Stenkula et al. (2015) examine the METR on capital for a longer period for
Sweden. This study provides further evidence of Sweden’s high METR for external capital (not retained earnings or
institutional capital) in the 1980s (approximately 100 per cent) and its significant reduction in the 1991 tax reform. These
corporate and capital tax reforms created opportunities for firm development, particularly for the growth of new small firms
and firm formation. Thus, we come to the following conclusion:
Conclusion 3. Sweden’s corporate tax reforms in the 1990s may have mitigated the problem of outsider discrimination in
the market for corporate ownership. The reduced taxes on corporate external financing may have led to increased entry and
the growth of new, productive firms in the Swedish business sector.
16
3.2.4. Foreign Direct Investment (FDI)
Business regulation affects the actions that firms can take and the balance of power that exists between various firm
stakeholders. Politicians may benefit from protecting owners from competition by gaining political support or by sharing the
rents that result from such protection (Olson, 1965, Stigler, 1971, and Perotti and Volpin, 2007). Moreover, in more open
economies, lobbying for international protection might occur (Spencer and Brander, 1983, and Grossman and Helpman,
1994). Politicians might also have an incentive to favour domestic owners in the market for corporate control (Horn and
Persson, 2001, and Norbäck et al. 2014).
Thus, regulation might affect the efficiency of the corporate ownership market by favouring certain types of ownership
over others, such as domestic ownership over foreign ownership. We refer to the problem as the foreign discrimination
problem in the market for corporate ownership. Let us now use this background to discuss the implications of the Swedish
reforms for the corporate ownership market to predict how such reforms might have affected the performance of the
Swedish business sector.
3.2.4.1. The Liberalization of Foreign Direct Investment in Sweden
Foreign exchange controls were introduced in Sweden shortly after the onset of the Second World War. In practice, this
legislation excluded any substantial foreign ownership of Swedish industry. The purpose of this legislation was openly
protectionist, i.e., to ascertain that “Swedish firms remain controlled by Swedish interests” (SOU 1986:23, p. 143). As
expected, legal impediments ensured that foreign ownership remained low, with foreign ownership of listed stocks never
exceeding 8 per cent throughout the 1980s; in addition, less than 5 per cent of private sector employees worked in foreign-
owned companies (Henrekson and J akobsson, 2005).
Between 1989 and 1993, the government undertook measures that opened the market to foreign ownership. This change
could be considered the final deregulation of the Swedish capital market that began in the early 1980s, thus following a
global trend of credit market deregulation in response to the more globalized economy (see Henrekson and J akobsson, 2005,
for a description of the major steps in the international deregulation process).
From a mere 7 per cent in 1989, the share of foreign ownership skyrocketed to 40 per cent only ten years later
(Henrekson and J akobsson, 2005). This increase also led to significant growth in the share of employees working in foreign-
owned firms, which increased from approximately 5 per cent at the end of the 1980s to 23 per cent in 2011. This change in
inward FDI was quite dramatic, even from an international perspective. Figure 7 shows that the inward FDI stock, as a
percentage of Sweden’s GDP, was approximately 5 per cent in the early 1990s, which was approximately half the EU 15
average. After the 1990s, Sweden’s inward FDI stock became substantially higher than the EU 15 average.
22
22
The increase in foreign ownership was especially strong in the mid-1990s. Employment in foreign-owned firms almost tripled between 1995 and 2013, fromapproximately
240,000 in 1995 to 630,000 in 2013.
17
Figure 7. Inward FDI stock (per cent of GDP), 1990–2013
Notes: The EU 15 values are based on our calculations, and they exclude Luxembourg.
Source: World Investment Report.
The injection of foreign ownership likely improved productivity development in the Swedish business sector. Having a
larger pool of potential owners should increase the potential for synergies. Foreign ownership may increase productivity
through the better use of assets, but bidding competitions may also generate large asset returns for previous Swedish
owners, who can then use these proceeds to invest in new projects or industries (Norbäck and Persson (2007).
We conclude the description of Sweden’s liberalization process of inward FDI in the 1990s as follows:
Conclusion 4. The liberalization of inward FDI in Sweden may have substantially mitigated the problem of foreign
discrimination in the market for corporate control in the Swedish business sector. These reforms may have caused more
efficient foreign owners to acquire inefficient Swedish target firms, which should have improved these firms’ productivity.
Moreover, this development may have spurred the incentive to create start-ups for sale in the market for corporate control.
4. A SUMMARY OF MICRO-DATA EVIDENCE FROM SWEDISH MATCHED EMPLOYER-
EMLOYEE DATA
The previous section showed that, after a period of interventionist policies in the 1970s and 1980s, Sweden deregulated its
product and labour markets, reformed its corporate tax system and opened itself up to FDI. We have argued that these
reforms should have reduced incumbents’ and insiders’ advantages and benefitted the growth of new efficient firms. This
section presents a summary of our previous work on empirical evidence based on matched employer-employee data. Details
of the empirical analysis can be found in Heyman, Norbäck and Persson (2015).
Data
The empirical analysis requires that we follow firms and individuals over time, thus necessitating access to highly detailed
data. Therefore, the analysis in Heyman, Norbäck and Persson (2015) was based on detailed employer-employee data from
SCB, which cover the 1990–2009 period.
23
The data originate from several register-based datasets from SCB and cover all
23
See, e.g., Davidson et al. (2014) and Hakkala et al. (2014) for recent articles based on these data.
18
firms in the private sector. First, the financial statistics contain detailed firm-level information on all Swedish firms in the
private sector during the 1996–2009 period. Variables include value added, capital stock (book value), the number of
employees, total wages, ownership status, profits, sales, and industry affiliation. Second, the Regional Labour Market
Statistics (RAMS) include data on all plants for the 1990–2009 period. The RAMS add information on the composition of
the labour force with respect to educational level and demographics at the plant level, which we aggregate to the firm level.
Individual-, plant- and firm-level data can be linked together using unique tracking numbers. From an international
perspective, the data are rather unique in terms of both magnitude and the level of detail.
A potential problem in an analysis of employment dynamics is the difficulty of following firms over time. Using
organization numbers as a method of identifying continuing, entering, and exiting firms can be problematic because such
numbers can change for various reasons. To more reliably follow firms over time, we used additional data from SCB. These
data make identifying new firm entries and firm exits possible, which means that we can analyse employment changes in (i)
completely new units, (ii) continuing units and (iii) exiting units.
To measure productivity, we used labour productivity, defined as value added per employee. Value added per employee
is a commonly used measure of productivity and is easily comparable across countries. Value added is calculated as the
output value minus the costs of purchased goods and services, excluding wages and other personnel costs (calculated by
SCB according to the international definition).
24
4.1. Allocative Efficiency
In the previous section, we emphasized the structural reforms that began in the 1980s: the reforms in the product market, the
reforms affecting inward FDI, and the labour market reforms with more decentralized wage setting and less job security for
workers with temporary contracts. We also emphasized that the tax system discriminated against smaller firms with high
growth potential.
In this section, we summarize the evidence regarding the reforms’ effects on the efficiency of the economy in Heyman,
Norbäck, and Persson (2015). We begin by using a productivity decomposition proposed by Olley and Pakes (1996) to
analyse productivity and reallocation. The Olley and Pakes method divides aggregate productivity into two terms, thus
implying that the weighted productivity in the business sector can be written as the sum of the simple (unweighted) average
productivity over all firms and the covariance between their productivity and market share.
25
The second term has a natural
efficiency interpretation term, and it can be interpreted as the extent to which market share is allocated to high-productivity
firms. If the covariance between firms’ productivity and their share of labour is strictly positive, then more productive firms
will tend to attract larger shares of workers, which is what we would expect in a well-functioning market economy.
To examine whether such allocative efficiency has changed over time in Sweden, we computed the Olley and Pakes
covariance term at the two-digit industry level for each year during the 1996–2009 period. Results show an increasing
allocative efficiency in Sweden, which is consistent with the notion that the reforms should have improved the market’s
allocation of resources.
26
We observe much higher estimates of the allocative efficiency term in the final years of our
sample compared with the first years.
This calculation presents one drawback: we cannot compare developments in Sweden with those in other countries.
However, to analyse how structural policies affect resource allocation efficiency, Andrews and Cingano (2014) use firm-
level data from a commercial data source covering 21 OECD countries in 2005. Investigating the Olley and Pakes
24
Another measure of productivity is Total Factor Productivity (TFP). Studies that use both labour productivity and TFP typically find similar results, irrespective of
the measure used (see, for instance, Bartelsman and Doms (2000) and Syverson (2011) for discussions of different productivity concepts).
25
See, e.g., Foster el al. (2001) for details. One advantage of their cross-sectional decomposition method is that cross-sectional productivity differences are more persistent and
possibly less sensitive to measurement errors and temporary shocks. The Olley and Pakes approach also does not depend on how firmentries and exits are measured.
26
See Heyman, Norbäck and Persson (2015) for details.
19
covariance, they find that Sweden has the largest allocative efficiency. This result is consistent with the substantial changes
in Sweden that we have accounted for in the previous section and the increase in allocative efficiency. Interestingly,
Andrews and Cingano (2014) also examine the source of the variation in the allocative efficiency term. The authors report
that regulations related to employment protection, product market competition, and FDI are negatively related to
productivity through a worsening of allocative efficiency, which indicates a reduced ability to allocate resources to more
productive firms.
A high degree of allocative efficiency implies that highly productive firms are able to attract workers from less productive
firms. This mechanism was weakened under the solidaristic wage policy, as described in our institutional analysis. The
aggregate picture from our empirical analysis appears consistent with the view that the deregulation of the Swedish wage-
setting system implied that productive and expanding firms found hiring and rewarding productive employees easier.
4.2. Where does the increase in productivity come from?
Our institutional analysis suggested that labour market reforms and particularly product market reforms combined with tax
reforms may have reduced the barriers to entry and that these actions might have played an important role in the turnaround
of the Swedish business sector by improving the creative destruction process. To distinguish the effect of the entry of new
firms and the exits of incumbents from the expansion and contraction of existing firms, we have also applied a
decomposition method to analyse the drivers of overall productivity in greater detail (see Foster et al. (2001) for a
discussion of different decomposition methods). The decomposition allows us to distinguish aggregate productivity changes
at the intensive margin from those at the extensive margin. Again, see Heyman, Norbäck and Persson (2015) for details.
Using firm-level data allows us to disentangle overall productivity growth into different components. This can help us
determine if Swedish productivity growth originated from within-industry dynamics (firm-level productivity growth), a
reallocation of market shares between existing firms (incumbents), or the entry and exit of firms. Even if the productivity of
individual firms did not change, productivity could have changed substantially due to changes in the market shares of firms
with different productivity levels. These insights are difficult to obtain with more aggregate data.
Figure 8 summarize the results from our productivity decomposition for the entire 1996–2009 period. Here, we see that
more than half of the overall increase in productivity in the business sector originated from new firms. The new firms that
survived gradually became more efficient than the average firm and thus contributed positively to long-term productivity
growth. As seen in Figure 8, entering firms’ contribution to productivity growth exceeded that of incumbents (firms that
were active throughout the 1996–2009 period). Hence, over the period studied, the entry of new firms was clearly the main
driving factor behind the increase in productivity in the Swedish business sector.
27
This result is consistent with the lower
entry barriers in Sweden enhancing the creative destruction process (Conclusion 2). Increased entry also emerged due to
corporate tax reforms (which levelled the playing field between entrants and incumbents) by promoting new firm start-ups
and, as we will see below, opening up the economy to FDI (Conclusions 3 and 4).
27
Braunerhjelmand Carlsson (1993) show that the number of small firms in the Swedish business sector decreased substantially compared with that in other industrialized
countries during the 1970s and 1980s. Henrekson et al. (2012) show that, in 2008, the firmsize distribution in Sweden had again become more similar to other comparable EU
countries.
20
Figure 8: Labour productivity growth decomposition, 1996–2009
Figure 8 also shows similar results for firms in the manufacturing and service sectors. The overall change in productivity
between 1996 and 2009 appears to have been somewhat larger in the manufacturing sector than in the service sector. We
also observe a positive cross effect in the manufacturing sector. This result indicates that the established manufacturing
firms that expanded also increased their productivity (or that established manufacturing firms that reduced their productivity
also experienced decreasing employment shares). Interestingly, this cross effect is negative in the service sector. This result
is consistent with the fast-growing nature of the service sector, where many expanding firms experienced declining
productivity during their growth phase.
4.3. Liberalization of Foreign Direct Investment
One of the major reforms undertaken in Sweden was the lifting of restrictions on foreign ownership. This reform led to
remarkably strong employment growth in foreign-owned affiliates in Sweden between 1980 and 2013, when nearly one-
fourth of workers were employed by foreign-controlled firms. We have also argued that the increase in foreign ownership
represented a much-needed productivity boost in the business sector, as a much larger pool of potential owners became
available.
The impact of the foreign acquisitions of Swedish firms on productivity in the Swedish business sector is an empirical
question. Heyman, Norbäck and Persson (2015) also provide a detailed empirical analysis based on matched employer-
employee data. The results show that, on average, labour productivity increased by approximately three per cent when
ownership was transferred from Swedish ownership to foreign ownership. This effect was completely driven by local
Swedish firms without any foreign operations being acquired by foreign firms. When we examine the effect of the foreign
acquisition of Swedish multinationals, we find no statistically significant effect. This result seems to be consistent with the
theory described in Section 3, i.e., that synergies may be more easily generated when an MNE acquires a local firm than
when two MNEs merge.
We also repeated the analysis of foreign ownership and foreign acquisitions on other performance measures, such as the
average wage and employment. The results indicate the existence of a significant wage and employment premium when
considering both average differences and the average change after an acquisition. Thus, foreign ownership and acquisitions
also appear to have contributed to higher employment and wages, which is what we would expect if foreign firms provide
new knowledge, better management, and better products and production methods.
21
Aggregate effects
How important have foreign firms been for aggregate productivity growth? To provide an answer we extend the
decomposition method developed by Foster et al. (2001) to also distinguish between Swedish and foreign firms. Thus, we
compare the different components that contribute to productivity growth, but we then decompose each component into a
Swedish and a foreign part. The results are summarized in Figure 9. Foreign-owned firms clearly contributed more to
productivity growth than did domestic Swedish firms. In fact, both the within-firm increase in productivity and the
productivity increase from entry are almost twice as large for foreign-owned firms compared with Swedish-owned firms.
The reason for the smaller overall difference in productivity growth is that the cross effect is negative for foreign-owned
firms. However, as explained above, this result may be due to significant expansions of foreign firms, where the
productivity in the expansion phase is below average productivity as the firm is built up.
Figure 9: Decomposition of labour productivity growth for the entire economy separated by ownership status, 1996–
2009
5. CROSS-COUNTRY EVIDENCE
Of course, the reform programme that Sweden undertook after the crisis included other measures. Monetary policy was
changed so that price stability was targeted under a floating exchange rate. As interest rates fell and the Krona depreciated
in the mid-1990s, demand and exports increased. Causally sorting out the most important factors – the “micro reforms”
-0,05 0 0,05 0,1 0,15
Productivity
Growth
Within
Between
Cross
Entry
Exit
Swedish Foreign
22
stressed here, which restore incentives and the functioning of markets, or a “macro” story of a depreciating exchange rate
that leads to export-led growth – is difficult.
One way to show that the reforms stressed here were very important in the turnaround of the Swedish economy is to use
cross-country data and proxies for the variables that capture the different reform areas and then correlate them with
economic performance. Unfortunately, cross-country and time-series data covering a broad range of policy indicators are
rare. Some variables only cover a few countries, and others have limited information over time. Some indicators are only
available for recent years, after the major reforms were implemented. Still, the available data give us some basic
information and can illustrate how reforms relate to different outcomes – in our case, most notably GDP per capita.
Figures 10a?10e present basic scatter plots on the relationship between different indicators and log GDP per capita for
OECD countries. Details on the different variables are presented in a web appendix. In Figure 10a, we start by studying the
World Bank’s overall “Ease of Doing Business variable”, defined as the distance to frontier (the best practice country). This
variable is available for the 2009–2014 period and incorporates many elements that are crucial for a market to function
efficiently. It focuses on regulations that affect small and medium-sized businesses. The index measures how difficult – or
costly – starting a business is, for instance, how difficult obtaining permits for electricity or construction is. It also covers,
for example, how costly trading across borders is. Essentially, it gives information on an economy’s day-to-day functioning.
By measuring the stance of the legal and regulatory system, it enables researchers to explore how important the business
environment is for the functioning of the market economy (see Besley, 2015, for an overview of the World Bank’s Doing
Business project). Figure 10a shows a positive relationship between GDP per capita and the Ease of Doing Business
variable.
28
Because the reform package in Sweden aimed at improving market functioning, the evidence from the Doing
Business variable should indicate that reforms affecting micro-economic performance have macroeconomic effects. A
similar pattern is also present when using other overall indexes that capture economic freedom and regulatory practices, for
instance, variables from the Heritage Foundation and the Fraser Institute.
29
These measures include more macroeconomic
types of variables.
Turning to the variables that reflect specific reform areas, we observe similar relationships. As a measure of product
market competition, we use the OECD indicator of PMR, which is available for 1998, 2003, 2008, and 2013. The index
ranges from 0 to 6, where higher values indicate stricter regulations. Figure 10b shows a clear negative relationship between
the regulation index and GDP per capita across OECD countries. In Figure 10c, we observe the relationship between
openness to FDI, as captured by the FDI regulatory restrictiveness index from OECD.
30
The figure indicates a negative
relationship across OECD countries. However, limited variation in openness to FDIs exists among OECD countries. If we
instead study all the available countries, we observe a strong negative relationship between the regulation of inward FDIs
and GDP per capita. As an indicator of the impact of taxes, we use the paying taxes variable from the World Bank’s Doing
Business database. This variable, available for the 2005?2014 period, includes the taxes and mandatory contributions that a
medium-sized company must pay and measures the administrative burden of paying taxes. This indicator is defined as the
distance to frontier, where a higher score indicates a lower tax burden. Figure 10d shows a strong positive relationship
across OECD countries between the tax indicator and GDP per capita. Finally, Figure 10e provides some basic information
on the cross-country relationship between the strictness of employment protection and GDP per capita. We use the OECD
index on the strictness of employment contracts for temporary contracts.
31
From Figure 10e, we observe a weak negative
28
An economy’s distance to frontier is reflected on a scale from0 to 100, where 0 represents the lowest performance and 100 represents the frontier. For example, a score of 75
in a certain year means that an economy was 25 percentage points away fromthe frontier constructed fromthe best performances.
29
These scatter plots are available upon request fromthe authors.
30
This variable covers the seven different years during the 1997–2003 period and ranges from0 to 1, where a higher value indicates less openness.
31
This variable is available for the 1995–2013 period. The data range from0 to 6, where higher values indicate stricter regulations.
23
association between regulation on temporary employment contracts and GDP per capita. Using the World Bank’s rigidity of
employment index (not shown) produces a more negative relationship.
Figure 10a: Correlation between log GDP per
capita and the Ease of Doing Business, 2009–2013
Notes: Scale 0–100; the higher the value, the closer to the frontier,
i.e., the better the business climate. OECD countries only.
Source: Doing Business and WDI (World Bank)
Figure 10c: Correlation between log GDP per
capita and FDI regulatory restrictiveness, 1997–
2013
Notes: Scale 0–1; the higher the value, the stricter regulation of
foreign direct investments. Years available: 1997, 2003, 2006, and
2010–2013. OECD countries only.
Source: OECD and WDI (World Bank)
Figure 10e: Correlation between Logged GDP per
capita and Employment protection legislation of
temporary contracts, 1985–2013
Notes: Scale 0–6; the higher the value, the stricter the regulation of
temporary contracts. OECD countries only.
Source: OECD and WDI (World Bank)
Figure 10b: Correlation between log GDP per
capita and product market regulation, 1998–2013
Notes: Scale 0–6; the higher the value, the more regulated product
market. Years available: 1998, 2003, 2008, and 2013. OECD
countries only.
Source: OECD indicators of product market regulation and WDI
(World Bank)
Figure 10d: Correlation between Logged GDP per
capita and corporate tax burden, 1997–2013
Notes: Scale 0–100; the higher the value, the closer to the frontier,
i.e., the lower the tax burden. Corporate tax burden refers to taxes
and the administrative burden of paying taxes for a medium-sized
company. Years available: 1998, 2003, 2008, and 2013. OECD
countries only.
Source: Doing Business and WDI (World Bank)
Let us now turn to regression analysis. Table 1 presents basic country regressions on variables capturing the reform
areas. Unfortunately, as noted above, most of these variables cover a period after the reforms were implemented in
Sweden and in many other countries. There is also variation in the sample of countries included in the different
variables. With these caveats in mind, the table shows estimates for the variables that are depicted in Figures 10a?10e
above.
32
For each reform variable, we present estimates with and without country fixed effects. All regressions include
year fixed effects. The cross-sectional regressions without country fixed effects give us estimates on how differences
between countries relate to aggregate productivity. The country fixed effects estimates take into account unobserved
country heterogeneity and relate within-country variation in the reform variables to within-country variation in
productivity. One issue with the regressions of the country fixed effects is the limited within-country and over-time
variation in several of the reform variables, which makes obtaining precise estimates difficult.
Columns (1) and (2) show a positive relationship between the overall Ease of Doing Business indicator (measured
as the distance to the top) and log GDP per capita.
33
The estimate in column (1) implies that a one-point increase in the
overall doing business index is associated with a nearly 10 per cent increase in GDP per capita. The estimate drops to
around half a per cent when country fixed effects are taken into account. Quantitatively, the estimate in column (2)
implies that increasing the Ease of Doing Business indicator from the level of Italy in 2014 (68.5) to the corresponding
value for Sweden (80.6) is associated with a productivity increase of approximately five per cent. The corresponding
figures for Greece and Spain are six and three per cent, respectively. If we instead compare Sweden’s value with the
mean of all developing countries, its increase in GDP per capita would be equal to nine per cent.
34
Turning to an indicator of product market competition (the OECD NMR indicator), we see in column (3) that more
regulations are negatively related to GDP per capita.
35
For this variable, the estimate turns statistically insignificant
when adding country fixed effects. If we only analyse the cross-section estimate in column (3), a change in NMR from
the mean position across developing countries to the Swedish one in the final year of the sample (2013) is associated
with a productivity increase of approximately 35 per cent.
Columns (5)?(8) show the results for FDI restrictions and the paying taxes variable. Both of these variables have the
expected signs and are statistically significant in all columns. Fewer restrictions on FDI and a lower tax burden,
including the administrative burden of paying taxes, for medium-sized firms are negatively related to aggregate
productivity. Quantitatively, the estimate in column (6) implies an increase in GDP per capita of approximately two per
cent in Sweden during the 1997–2013 period (when data are available). This calculation is based on a strengthening of
the index for Sweden during this period from 0.079 to 0.059. A similar calculation for the taxation variable implies an
increase in aggregate productivity of approximately 0.7 per cent (based on column 8). In this case, data cover the 2005–
2014 period, and the index for Sweden increased from 80.9 to 83.3. If we also compare Sweden with Spain, Greece, and
the developing country mean for the final year 2014, shifts in the paying taxes variable from the position of these
countries to that of Sweden’s lower level are associated with gains in GDP per capita of approximately two, two, and
five per cent, respectively.
32
We do not use PMR because it is only available for 1998, 2003, 2008 and 2013, and we would like to include a longer period in our analysis. The results for PMR are
qualitatively similar and are available upon request.
33
See Besley (2015) for a summary on previous research using the doing business indicators. See also Djankov et al. (2006) and Eklund (2011) for examples of studies on
the relationship between GDP growth and the Doing Business indicators.
34
These country groupings are based on the UN’s regional coding; seehttp://unstats.un.org/unsd/methods/m49/m49regin.htm.
35
NMR stands for non-manufacturing sector regulations. This index fromthe OECD is available for a long period: 1975–2013. See the web appendix for details.
26
Table 1: Reform variables and productivity; results from firm cross-country regressions, 1996–2009
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Reform
Variable Ease of Doing Business
Product Market
Regulations FDI Index Paying Taxes
Strictness of Empl.
Protection: Temporary
Contracts
0.098*** 0.004*** -1.375*** -0.076 -4.583*** -1.013*** 0.041*** 0.003*** -0.113 0.046***
(0.005) (0.001) (0.129) (0.068) (1.116) (0.377) (0.005) (0.001) (0.074) (0.017)
Country FE No Yes No Yes No Yes No Yes No Yes
Year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations
871 871 143 143 348 348 1,556 1,556 801 801
R-squared
0.613 0.397 0.615 0.629 0.194 0.658 0.258 0.411 0.099 0.822
Year
2009–2013 2009–2013 1998–2013 1998–2013 1997–2013 1997–2013 2005–2013 2005–2013 1985–2013 1985–2013
Note: The dependent variable is log GDP per capita (in constant 2005 US dollars). See the web appendix for details on the reformvariables. Standard errors are adjusted
for clustering at the country level. ***, **, and * show significance at the 1 per cent, 5 per cent, and 10 per cent levels, respectively.
6. CONCLUSION
In this paper, we have argued that fundamental reforms in Sweden’s business sector in the 1990s can explain its
remarkable productivity and employment growth after it experienced a significant economic crisis in the early 1990s.
Why were these reforms so successful?
We have argued that they resolved fundamental market and political failures that affected the Swedish economy in
the 1970s and the 1980s. A fundamental market failure in the Swedish business sector during this period was that
incumbent firms and labour unions had gained too much power. This power imbalance enabled them to protect their
markets from competition, creating negative externalities for potential entrants, consumers, and labour market outsiders.
The dominance of incumbents and insiders in labour unions was substantially mitigated by the deregulation of the
labour and product markets.
A fundamental political failure regarding the business sector was that politicians favoured incumbent firms and
insider employees. The corporate tax system and FDI restrictions impeded ownership changes and business formation,
which, in turn, harmed entrepreneurs, labour, and consumers. Moreover, the political system underestimated the cost of
hampering economic incentives for the business sector when pursuing political goals, such as very low unemployment
and highly compressed wages. These political failures were greatly mitigated by the tax reforms, the opening of the
economy to FDI, and the decentralization of wage bargaining.
The process described above suggests that deregulating the business sector can explain part of Sweden’s turnaround.
However, one question remains: over the last decade, why has Sweden not only caught up to the OECD average but
also outperformed nearly all other countries in terms of productivity growth and employment growth in the business
sector?
One possible explanation is that Swedish firms have made more efficient investments in R&D and ICT (see Edquist
and Henrekson, 2015). Investments in intangibles including software, R&D, mineral exploration, copyright and
licensing costs, financial industry development, design, brand equity, vocational training, and organizational structure
are considered very important in Sweden. According to Edquist (2011), intangible investments constituted nearly 10 per
27
cent of GDP and accounted for nearly 30 per cent of labour productivity growth in Sweden’s business sector over the
1995–2006 period. Why then did this occur? First, Swedish society has been open to the adoption of new technologies
and trends. Second, young people were exposed to new ICT early, both in school and at home, which suggests that the
young generation drives the digitalization of the Swedish business sector.
The inflow of FDI to Sweden has also been extraordinary. FDI has created employment and increased productivity
in the target firms, but its indirect dynamic effects also appear to have been important. For example, the Wallenberg
group held controlling positions in companies, which accounted for 42 per cent of the Stockholm Stock Exchange’s
market cap in 1998. By November 2010, their control had declined to 17.1 per cent of the total market cap. Between
1999 and 2009, Investor AB nearly tripled the share of its portfolio, investing in new growth markets and scaling back
its more traditional investments, where it controlled a number of very large firms. Of these growth investments, 62 per
cent went to the Nordic region.
Thus, the deregulation of FDI flows into Sweden appears to have created synergies in the acquired target firms, and
it has generated financial capital for entrepreneurial firms in growth markets. Moreover, the possibility of selling
successful ventures to large foreign incumbents, such as Microsoft, may have been a driver of the vibrant Swedish start-
up markets in new services, computers, and computer and internet games. Skype and Mojang are prominent examples
of such Swedish tech start-ups, which were later sold for astronomical sums. Network effects in these businesses then
create synergies when large foreign incumbents obtain new products from smaller firms. Bidding competitions create
substantial gains for these sellers, who then invest in new projects. This internationalization of the industry appears to
have been particularly large and efficient in Sweden compared with internationalization in other countries.
The Swedish reforms were implemented, and subsequent governments did not reverse them, which is a fundamental
aspect of their success. Many strong groups lost power in the reforms, at least in the short term. One of the arguments
explaining the success of the reforms is the considerable power of bureaucrats and experts in formulating economic
policy in Sweden. Building on a history of trust and respect for knowledge, the political system, industry, and unions
have often been able to reach decisions through consensus on issues of great importance or through the efficiency of the
Swedish economy. The process has also been open to the influence of many different parties, which has generated
broad commitment to the reforms.
We believe that our study of the Swedish experience of industrial reorganization in the 1990s can be a valuable case
study for developing countries that are in search of efficient regulation of their business sectors – and for developed
countries facing economic problems in the aftermath of the financial crisis. Sweden’s experience can serve as an
important example of how an economy undergoing a deep crisis can respond and recover by undertaking economically
sound industrial reforms. Furthermore, by comparing insights from economic theory with an actual restructuring
process, we believe that we can provide valuable knowledge concerning the economic forces driving creative
destruction, which can potentially provide solid ground for policy discussions, particularly those relating to how
countries can improve their competitiveness and employment levels. Our study thus supports the view that addressing
micro-economic inefficiencies is important to provide a solid foundation for a country’s growth and prosperity.
28
REFERENCES
Acs, Z. J . and D. B. Audretsch (2005). ‘Entrepreneurship, innovation and technological change’, Foundations and
Trends in Entrepreneurship, 1(4), 149–195.
Aghion, P., N. Bloom, R. Blundell, R. Griffith and P. Howitt (2005). ‘Competition and innovation: an inverted-u
relationship’, The Quarterly Journal of Economics, 120(2), 701–728.
Allard, G. (2005). ‘Measuring job security over time: in search of a historical indicator for EPL (Employment
Protection Legislation)’, Working Paper WP05-17, Instituto de Empresa, Madrid.
Andrews, D. and F. Cingano (2014). ‘Public policy and resource allocation: evidence from firms in OECD countries’,
Economic Policy, 29(78), 253–296.
Arnold, J .M., B. Brys, C. Heady, A. J ohansson, C. Schwellnus, and L. Vartia (2011). ‘Tax policy for economic
recovery and growth’, Economic Journal, 121(550), F59–F80.
Audretsch, D. B. (1991). ‘New-firm survival and the technological regime’, The Review of Economics and Statistics,
73(3), 441–450.
Auerbach, A. J . (2002). ‘Taxation and corporate financial policy’, in A. J . Auerbach and M. Feldstein (eds.), Handbook
of public economics, 1st edition, 3(19), 1251–1292, Elsevier.
Bandiera, O. I. Barankay och I. Rasul (2007). ‘Incentives for managers and inequality among workers: evidence from a
firm-level experiment’, Quarterly Journal of Economics, 122(2), 729–773.
Bandiera, O. I. Barankay and I. Rasul (2009). ‘Social connections and incentives in the workplace: evidence from
personnel data’, Econometrica, 77(4), 1047–1094.
Barba Navaretti, G. and A.J . Venables (2004). Multinational Firms in the World Economy, Princeton: Princeton
University Press.
Bartelsman, E. J . and M. Doms (2000). ‘Understanding productivity: lessons from longitudinal microdata’, Journal of
Economic Literature, 38(3), 569–594.
Bartelsman, E., S. Scarpetta, and F. Schivardi (2005). ‘Comparative analysis of firm demographics and survival:
evidence from micro-level sources in OECD countries’, Industrial and Corporate Change, 14(3), 365–391.
Bergh, A. (2013). ‘What are the policy lessons from Sweden? On the rise, fall and revival of a capitalist welfare state’,
New Political Economy, Published online: 20 Dec 2013.
Bergh, A. (2014). Sweden and the revival of the capitalist welfare state, Edward Elgar, Northampton, MA.
Bergh, A. and G. Erlingsson (2006). ‘Resilience through restructuring: Swedish policy-making style and the consensus
on liberalisations 1980-2000’, Ratio Working Papers No. 110. Available at RePEc:http://ideas.repec.org/p/hhs/ratioi/0110.html.
Bertrand, M. and A. Schoar (2003), ‘Managing with style: the effect of managers on firm policies’, Quarterly Journal
of Economics, 118(4), 1169–1208.
Besanko, D., D. Dranove, M. Shanley, and S. Schaefer (2003). Economics of strategy, 3rd edition, J ohn Wiley & Sons,
Inc, NY.
Besley, T. (2015). "Law, Regulation, and the Business Climate: The Nature and Influence of the World Bank Doing
Business Project", Journal of Economic Perspectives, 29(3), 99-120.
Bloom, N. and J . Van Reenen (2007). ‘Measuring and explaining management practices across firms and countries’,
Quarterly J ournal of Economics, 122(4), 1351–1408.
29
Calmfors, L. (2012). ‘Sweden: from macroeconomic failure to macroeconomic success’, CESifo working paper: Fiscal
Policy, Macroeconomics and Growth, No. 3790.
Carroll, R., D. Holtz-Eakin, M. Rider and H. S. Rosen (2000). ‘Income taxes and entrepreneurs' use of labor’, Journal
of Labor Economics, 18(2), 324–351.
Carroll, R., D. Holtz-Eakin, M. Rider and H. S. Rosen (2001). ‘Personal income taxes and the growth of small firms, in
J . M. Poterba (ed.), Tax Policy and the Economy, 15, 121–148, MIT Press, Cambridge.
Caves, R. E. (1998). ‘Industrial organization and new findings on the turnover and mobility of firms’, Journal of
Economic Literature, 36(4), 1947–1982.
Coase, R. (1937). ‘The Nature of the Firm’, Economica, 4(16), 386–405.
Da Rin, M., M. Di Giacomo and A. Sembenelli (2011). ‘Entrepreneurship, firm entry, and the taxation of corporate
income: evidence from Europe’, Journal of Public Economics, 95(9), 1048–1066.
Davidson, C., F. Heyman, S. Matusz, F. Sjöholm and S. Chun Zhu (2014). ‘Globalization and imperfect labor market
sorting’, Journal of International Economics, 94(2), 177–194.
Davis, S. J . and M. Henrekson (1999). ‘Explaining National Differences in the Size and Industry Distribution of
Employment’, Small Business Economics, Springer, vol. 12(1), pages 59-83.
Davis, S. J . and M. Henrekson (2000). ‘Wage-setting institutions as industrial policy’, NBER Working Paper No. 7502.
Devereux, M. P., R. Griffith, and A. Klemm (2002). ‘Corporate income tax reforms and international tax competition’,
Economic Policy, 17(35), 449–495.
Djankov, S., C. McLiesh and R. M. Ramalho (2006). ‘Regulation and Growth’, Economic Letters,
doi:10.1016/j.econlet.2006.03.021.
Djankov, S., T. Ganser, C. McLiesh, R. M. Ramalho and A. Schleifer (2010). ‘The effect of corporate taxes on
investment and entrepreneurship’, American Economic Journal: Macroeconomics, 2(3), 31–64.
Edquist C. (2011). ‘Innovation policy design: identification of systematic problems’, CIRCLE Electronic Working
Papers No. 2011/6.
Edquist, H. and M. Henrekson (2013). ‘Product market reforms and incentives to innovate in Sweden’, in M. Choupres
and H. Edquist (eds.), What can we learn from economic reforms in Greece and Sweden?, European Liberal Forum,
Brussels.
Edquist, H. and M. Henrekson (2015). ‘Swedish Lessons: How Important are ICT and R&D to Economic Growth?,’
Working Paper Series 1073, Research Institute of Industrial Economics.
Eklund, J . (2011). ‘Regelbördans ekonomiska effekter’, Tillväxtanalys.
Ericson, R. and A. Pakes (1995). ‘Markov-perfect industry dynamics: a framework for empirical work’, The Review of
Economic Studies, 62(1), 53–82.
Faggio, G., K. G. Salvanes and J . Van Reenen (2010). ‘The evolution of inequality in productivity and wages: panel
data evidence’, Industrial and Corporate Change, 19(6), 1919–1951.
Foster, L., J . Haltiwanger and C. J . Krizan (2001). ‘Aggregate productivity growth. Lessons from microeconomic
evidence’, in C. R. Hulten, E. R. Dean, and M. J . Harper (eds.), New developments in productivity analysis,’
University of Chicago Press, Chicago.
Fox, J . T. and V. Smeets (2011). ‘Does input quality drive measured differences in firm productivity?’, NBER Working
Paper No. 16853.
30
Gemmel, N., R. Kneller, D. McGowan and I. Sanz (2013). ‘Corporate taxation and productivity catch-up: evidence
from European firms’, Working Papers No. 13001, Bangor Business School, Prifysgol Bangor University (Cymru /
Wales).
Gentry W. M. and G. Hubbard (2000). ‘Entrepreneurship and houshold saving’, NBER Working Papers No. 7894.
Gibbons, R. and J . Roberts (eds.) (2013). The handbook of organizational economics, Princeton University Press,
Princeton.
Gordon, R. H. and J . J r. Hines (2002). ‘International taxation’, in A. J . Auerbach and M. Feldstein (eds.), Handbook of
public economics, 1st edition, 4(28), 1935–1995, Elsevier.
Gibrat, R. (1931). Les Inégalités économiques, Librairie du Recueil Sirey, Paris, France.
Grossman, S. and O. Hart (1986). ‘The costs and the benefits of ownership: a theory of vertical and lateral integration’,
Journal of Political Economy, 94(4), 691–719.
Grossman, G. M. and E. Helpman (1994). ‘Protection for sale’, American Economic Review, 84(4), 833–850.
Hakkala, K., F. Heyman and F. Sjöholm (2014). ‘Multinational firms and job tasks’, European Economic Review, 66,
248–265.
Hall, B. and J . Van Reenen (2000). ‘How effective are fiscal incentives for R&D? A review of the evidence’, Research
Policy, 29(4), 449–469.
Hart, O. and J . Moore (1990). ‘Property rights and the nature of the firm’, The Journal of Political Economy, 98(6),
1119–1158.
Hassett, K. A. and R. G. Hubbard (2002). ‘Tax policy and business investment’, in: A. J . Auerbach and M. Feldstein
(eds.), Handbook of public economics, 1st edition, 3(20), 1293–1343, Elsevier.
Haufler, A., P.-J . Norbäck and L. Persson (2014). ‘Entrepreneurial innovations and taxation’, Journal of Public
Economics, 113(May), 13–31.
Henrekson, M. and U. J akobsson (2005). ‘The Swedish model of corporate ownership and control in transition’, in
Huizinga, H. and L. J onung (eds.), The internationalization of asset ownership in Europe, Cambridge University
Press, Cambridge.
Henrekson, M., J ohansson, D. and M. Stenkula (2012). "Den svenska företagsstrukturen – utvecklingen i de medelstora
företagen efter 1990-talskrisen" Ekonomisk Debatt, 40(2), 27–38.
Heyman, F. P.-J . Norbäck and L. Persson (2015). ‘ The Turnaround of Swedish Industry: Reforms, Firm Diversity and
J ob and Productivity Dynamics‘, IFN Working Paper No. 1079.
Hicks, A. and L. Kenworthy (1998). ‘Cooperation and Political Economic Performance in Affluent Democratic
Capitalism‘, American Journal of Sociology 103, 1631–1672.
Hines, J . J r. (2005). ‘Do tax havens flourish?’, NBER Working Papers No. 10936.
Hjalmarsson, L. (1974). ‘The size distribution of establishments and firms derived from an optimal process of capacity
expansion’, European Economic Review, 5(2), 123–140.
Hopenhayn, H. (1992). ‘Entry, exit, and firm dynamics in long run equilibrium’, Econometrica, 60(5), 1127–1150.
Horn, H. and L. Persson (2001). ‘The equilibrium ownership of an international oligopoly’, Journal of International
Economics, 53(2), 307–333.
Ilmakunnas, P., M. Maliranta and J . Vainiomäki (2004). ‘The roles of employer and employee Ccaracteristics for plant
productivity’, Journal of Productivity Analysis, 21(3), 249–276.
J avorcik, B. S. (2015). ‘Does FDI Bring Good J obs to Host Countries?’, World Bank Research Observer, 30(1), 74–94.
31
J onung, L., J . Kiander and P. Vartia (2008). ‘The great financial crisis in Finland and Sweden – the dynamics of boom,
bust and recovery, 1985–2000’, European Economy, Economic Papers 367, European Commission, Brussels.
J ovanovic, B. (1982). ‘Selection and the evolution of industry’, Econometrica, 50(3), 649–670.
Keuschnigg, C. and S. B. Nielsen (2004a). ‘Taxation and venture capital backed entrepreneurship’, International Tax
and Public Finance, 11(4), 369–390.
Keuschnigg, C. and S. B. Nielsen (2004b). ‘Start-ups, venture capitalists, and the capital gains tax’, Journal of Public
Economics, 88(5) 1011–1042.
King, M. and D. Fullerton (1984). The taxation of income from capital: a comparative study of the United States, the
United Kingdom, Sweden and West Germany, The Chicago University Press, Chicago.
Klepper, S. (1996). ‘Entry, exit, growth, and innovation over the product life cycle’, American Economic Review, 86(3),
562–583.
Kneller, R. and D. McGowan (2012). ‘Tax policy and firm entry and exit dynamics: evidence from OECD countries’,
Working Papers No. 12006, Bangor Business School, Prifysgol Bangor University (Cymru / Wales).
Lazear, E. P. (2000). ‘Performance pay and productivity’, American Economic Review, 90(5), 1346–1361.
Li, Y. and M. Rama (2015). ‘Firm Dynamics, Productivity Growth, and J ob Creation in Developing Countries: The
Role of Micro- and Small Enterprises’, World Bank Research Observer, 30(1), 3–38.
Lindbeck, A. (1997). The Swedish experiment, J ournal of Economic Literature, Vol. XXXV , 1273-1319.
Lindbeck, A., P. Molander, T. Persson, O. Sandmo, B. Swedenborg and N. Thygesen (1994), Turning Sweden Around.
MIT Press: Cambridge, MA.
Lindbeck, A. and D. J . Snower (2001). ‘Insiders versus Outsiders’, J ournal of Economic Perspective, Vol. 15, NO 1,
165-188.
Luttmer, E. G. J . (2007). ‘Selection, growth, and the size distribution of firms’, The Quarterly Journal of Economics,
122(3), 1103–1144.
Maksimovic V. and G. Phillips (2001). ‘The market for corporate assets: who engages in mergers and asset sales, and
are there efficiency gains?’, Journal of Finance, 56(6), 2019–2065.
Mirrlees, J ., S. Adam, T. Besley, R. Blundell, S. Bond, R. Chote, M. Gammie, P. J ohnson, G. Myles, and J . Poterba
(2011) Tax by design: the Mirrlees review, Oxford University Press for Institute for Fiscal Studies, Oxford.
Motta, M. (2004). Competition policy: theory and practice, Cambridge University Press, Cambridge.
Murphy, K. M. and R. H. Topel (1990). ‘Efficiency Wages Reconsidered: Theory and Evidence’, in Y. Weiss and G.
Fishelson (eds.), Advances in the theory and measurement of unemployment, Palgrave Macmillan, London.
Nelson, R. R. and S. G. Winter (1982). An evolutionary theory of economic change, Belknap Press, Cambridge, MA.
Norbäck, P.-J . and L. Persson (2007). ‘Investment liberalizations:wWhy a restrictive cross-border merger policy can be
counterproductive’, Journal of International Economics, 72(2), 366–380.
Norbäck, P.-J . and L. Persson (2012). ‘Entrepreneurial innovations, competition and competition policy’, European
Economic Review, 56(6), 488–506.
Norbäck, P.-J ., L. Persson and R. Douhan (2014). ‘Entrepreneurship policy and globalization’, Journal of Development
Economics, 110(September), 22–38.
Nycander, S. (2008). Makten över arbetsmarknaden: ett perspektiv på Sveriges 1900-tal, SNS Publishing, Stockholm.
OECD (2013). OECD compendium of productivity indicators 2013, OECD Publishing.
Olley, S. and A. Pakes (1996). ‘The dynamics of productivity in the telecommunications industry’, Econometrica,
64(6), 1263–1298.
32
Olson, M. (1965). The logic of collective action, Harvard University Press, Harvard.
Perotti, E. and P. Volpin (2007). ‘Investor protection and entry’, Tinbergen Institute Discussion Papers 07-006/2.
Rajan, R. G. and L. Zingales (2001). ‘The firm as a dedicated hierarchy: a theory of the origins and growth of firms’,
The Quarterly Journal of Economics, 116(3), 805–851.
Saggi, K. (2002). ‘Trade, Foreign Direct Investment, and International Technology Transfer: A Survey’, World Bank
Research Observer, 17(2), 191–235.
Santarelli, E. and M. Vivarelli (2007). ‘Entrepreneurship and the process of firms’ entry, survival and growth’, Industry
and Corporate Change, 16(3), 455–488.
Skedinger, P. (2010). ‘Employment protection legislation. Evolution, effects, winners and losers’, Edward Elgar,
Northampton, MA.
SOU 1986:23 (1986). Aktiers röstvärde, Report by the Voting Rights Committee, Liber, Stockholm.
SOU 1993:16 (1993). Nya villkor för ekonomi och politik, Report from the Economics Commison, Allmänna förlaget,
Stockholm.
SOU 2005:4 (2005). Liberalisering, regler och marknader, Ministry for Enterprise, Stockholm.
Spencer, B. J . and J . A. Brander (1983). ‘International R&D rivalry and industrial strategy’, Review of Economic
Studies, 50(4), 707–722.
Stenkula, M., D. J ohansson and G. Du Rietz (2015). ‘Capital income taxation of Swedish households, 1862 to 2010’,
Scandinavian Economic History Review, 63(2), 154–177.
Stigler, G. J . (1971). ‘The theory of economic regulation’, The Bell Journal of Economics and Management Science,
2(1), 3–21.
Sutton, J . (1997). ‘Gibrat’s legacy’, Journal of Economic Literature, 35(1), 40–59.
Syverson, C. (2011). ‘What determines productivity?’, Journal of Economic Literature 49(2), 326–365.
Thimann, C. (2015). "The Microeconomic Dimensions of the Eurozone Crisis and Why European Politics Cannot Solve
Them", Journal of Economic Perspectives, 29(3), 141-64.
Tirole, J . (2006). The theory of corporate finance, Princeton University Press, Princeton.
Van Ark, B., M. O’Mahony, and M. P. Timmer (2008). ‘The productivity gap between Europe and the United States:
trends and causes’, Journal of Economic Perspectives, 22(1), 25–44.
Vives, X. (2008). ‘Innovation and competitive pressure’, The Journal of Industrial Economics, 56(3), 419–469.
Williamson, O. E. (1979). ‘Transaction-cost economics: the governance of contractual relations’, Journal of Law and
Economics, 22(2), 233–261.
World Bank (2014). ‘Doing business 2015: going beyond efficiency’, A World Bank Group Flagship Report, World
Bank, Washington, DC.
Wölfl, A., I. Wanner, T. Kozluk, and G. Nicoletti (2009). ‘Ten years of product market reform in OECD countries:
insights from a revised PMR indicator’, OECD Economics Department Working Papers No. 695
doc_291818468.pdf
The Turnaround of the Swedish Economy Lessons from Business Sector Reforms
IFN Policy Paper No. 73, 2015
The Turnaround of the Swedish Economy:
Lessons from Business Sector Reforms
Fredrik Heyman, Pehr-Johan Norbäck and Lars
Persson
The Turnaround of the Swedish Economy:
Lessons from Business Sector Reforms
Fredrik Heyman
*
, Pehr-Johan Norba? ck,
*
and Lars Persson
*
November 2015
Abstract: How can a country overcome a long period of stagnant productivity growth in its business sector and
reach its growth potential? Sweden during the 1970–2010 period can serve as an example to help other countries
understand how to efficiently reform a business sector. In the 1990s, Sweden implemented a reform package that
ignited a successful reorganization of a business sector that had faltered for decades. To understand the economic
forces behind this process, we first survey the industrial restructuring literature and then examine the reform
package using Swedish matched plant-firm-worker data. The removal of barriers to growth for new and
productive firms and increased rewards for investment in human capital were crucial to the success of Sweden’s
reforms. Turning to cross-country evidence, we find that an increase in the World Bank’s Doing Business
indicator is associated with an increase in GDP per capita. These findings suggest that policymakers have much
to learn from country case studies and from analyses based on the World Bank’s Doing Business indices. Our
study of the Swedish experience can be a valuable case study for developing countries that are attempting to
promote growth by developing their business sectors and for countries that are facing economic problems in the
aftermath of the financial crisis.
JEL: D22, E23, J 21, J 23, K23, L11, L16, L51
Keywords: regulations, allocative efficiency, productivity, job dynamics, matched employer-employee data, industrial
structure and structural change
* Research Institute for Industrial Economics (IFN), P.O. Box 55665, SE-102 15 Stockholm, Sweden,
[email protected], [email protected], and [email protected].
Acknowledgements: We thank Vincenzo Denicolò, Nicola Fuchs-Schündeln, Magnus Henrekson, Assar Lindbeck, Per
Skedinger and Michelle Sovinsky for their helpful comments. We also thank Fredrik Andersson, Malin Olsson Tallås, Olga
Pugatsova and Hanna Thunström for their excellent research assistance. Funding: We are grateful to the Marianne and
Marcus Wallenberg Foundation for its financial support. Fredrik Heyman gratefully acknowledges financial support from
the Swedish Research Council for Health, Working Life and Welfare (FORTE) and from Torsten Söderbergs Stiftelse. Pehr-
J ohan Norbäck and Lars Persson gratefully acknowledge financial support from the Wallander and Tom Hedelius Research
Foundation and from Vinnova.
2
1. INTRODUCTION
Although the economic policy debate has long centred on the importance of macroeconomic policies, a growing awareness
acknowledges that the microeconomic functioning of markets is vital. This shift in focus can be illustrated in the following
quote from Kaushik Basu, Senior Vice President and Chief Economist of the World Bank, in the foreword to the Doing
Business 2015 report:
1
“The public discourse on economic policy is overwhelmingly focused on fiscal measures, monetary interventions, welfare
programs, and other such highly visible instruments of government action. Thus when an economy does poorly, a
disproportionate amount of our debate centers on whether or not it needs a fiscal stimulus, whether there should be liquidity
easing or tightening, whether its welfare programs have been too profligate or too paltry and so on. What gets much less
attention but is equally – and, in some situations, even more – important for an economy’s success or failure is the nuts and
bolts that hold the economy together and the plumbing that underlies the economy.”
A malfunctioning system of micro-economic regulation can hinder economic growth and make standard macro policies
less effective or even lead to deep macroeconomic crises. In this paper, we will argue that the Swedish experience from
1970?1980 is an illustrative example of how a dysfunctional system of micro-economic regulations can lead to a severe
macroeconomic crisis. In the 1980s and early 1990s, Sweden faced its most severe economic crisis in the post-war period:
Swedish companies lost in global competition, while the state became very highly leveraged. The 1991–1994 period was
characterized by a substantial decline in GDP and increasing unemployment. Our institutional and theoretical analysis
suggests that the substantial slowdown of the Swedish economy in the 1980s and the subsequent economic crisis in the early
1990s were to a large extent caused by increasingly interventionist business policies, reflecting a change in world view to
one in which economic incentives and private entrepreneurship not associated with large firms were regarded with suspicion
(Lindbeck, 1997).
However, in the 1990s, Sweden undertook a major structural reform package, which largely aimed to improve the
microeconomic functioning of markets, and ignited a successful industrial reorganization process in the business sector. The
Swedish reforms were implemented, and subsequent governments did not reverse them, which is a fundamental aspect of
their success. Ireland, Sweden and the US experienced the highest labour productivity growth in the OECD during the
1995–2011 period (OECD (2013)). Sweden’s productivity growth was primarily driven by factors that increased the
effectiveness of its business sector. This process took place while the welfare state was largely preserved. Although
inequality in Sweden, as in many other countries, has increased, it still exists at a low level compared to that in other
countries.
To understand the economic forces associated with the Swedish restructuring process, we first survey the industrial
restructuring literature in search of mechanisms that are important in explaining firm, employment, and productivity
dynamics. The overview indicates that productive, expanding firms are typically associated with active owners and up-to-
date and incentive-based management practises. Consequently, economically efficient decisions are made, and well-
functioning business cultures are developed, leading to a motivated workforce. The employment of skilled workers and the
early adoption of new technologies have also been shown to create competitive advantages in both local and global markets,
thereby spurring productivity growth. Start-ups and expansions are associated with high degrees of uncertainty, meaning
that many businesses fail; thus, the observed number of highly successful and expanding ventures is low, but they are still
1
See also two recent articles in the J ournal of Economic Perspectives (Besley, 2015; Thimann, 2015) stressing the importance of the microeconomic foundations.
3
an important factor in productivity and employment growth. Firms’ productivity and employment also crucially depend on
external factors, such as institutions and access to production factors
The picture emerging from our institutional analysis shows that Sweden developed good economic institutions in the late
1800s and experienced high levels of sustained growth in the 1870?1970 period. In the 1970s, however, considerably more
interventionist policies were developed, such that Sweden had one of the most regulated business sectors in the developed
world in the 1970s and 1980s.
2
Growth then slowed substantially, which led to an economic crisis in the early 1990s. This
experience paved the way for several reforms that intended to increase the efficiency of the Swedish economy, particularly
the Swedish business sector. Our main finding from this institutional and theoretical examination is that the removal of
barriers to entry and growth for new and productive firms and increased rewards for investments in human capital and for
efforts in the workplace were crucial to the success of these reforms. These actions, we argue, led to remarkable growth in
productivity and employment in the Swedish business sector in the two decades that followed.
We then describe empirical results found in an accompanying paper, Heyman, Norbäck and Persson (2015), and discuss
how their results provide indicative support for our institutional predictions. Heyman, Norbäck and Persson (2015) is based
on Swedish matched employer-employee data over the period 1996-2009 from Statistics Sweden (SCB), which allows for
analysing issues related to firm employment and productivity dynamics possible in greater detail than has been possible in
most other international studies.
3
First, there was an increase in allocative efficiency in Sweden, as measured as greater market shares for the more
productive firms in the economy, during the period 1996-2009. This suggests that the reforms mitigated the insider and
incumbency problems in the Swedish business sector and enabled more productive firms to better attract capital and
employees than they had previously. The relationship between firms’ productivity and wage increases was strengthened
over the period studied, which suggests that productive firms and productive employees became more rewarded in the
Swedish business sector. There was also an increase in jobs created in small firms, while most of the productivity gains
were created in large incumbent firms, suggesting that the reforms facilitated the division of labour between large
incumbents and small growing firms. Finally, foreign firms contributed significantly to productivity and employment
growth in the business sector during this period, which suggests that the liberalization of foreign direct investment (FDI)
was an important factor in driving the restructuring process.
Thus, a crucial element of the success of the Swedish business sector turnaround was the emergence of firms with
different ownership and organization (foreign firms and small firms). Therefore, the micro-economic reforms that made the
playing field between different ownership and organization forms more level were crucial for Sweden’s successful
transformation.
The reforms that Sweden undertook after the crisis included other measures as well. The government was forced to
operate under stricter budget rules. Monetary policy was changed so that price stability was targeted by an independent
central bank under a floating exchange rate. As interest rates fell and the krona depreciated in the mid-1990s, demand and
exports increased.
Identifying the most important factors – “micro reforms”, which restore incentives and the functioning of markets, or a
“macro” story of solid public finances and a flexible exchange rate leading to export-led growth – is of course complex. One
way of determining whether the microeconomic reforms stressed here were crucial in the Swedish economic turnaround
involves using cross-country data and proxies for the variables that capture different aspects of the reforms and then
correlating these variables with economic performance. Unfortunately, cross-country and time-series data covering a broad
range of policy indicators are rare. Still, the available data give us some basic information and can illustrate how reforms
2
See Hicks and Kenworthy (1998).
3
See Heyman, Norbäck and Persson (2015) for details of the empirical analysis.
4
relate to different outcomes – in our case, most notably GDP per capita. Using the Ease of Doing Business indicator
(measured as the distance to the top) and logged GDP per capita, we find that a one-point increase in the overall Doing
Business index is associated with approximately half a per cent increase in GDP per capita when country fixed-effects are
taken into account.
4
The above described results have several policy implications. Developing countries face the challenging task of
achieving their growth potential. Sweden’s experience during the 1970?2010 period can serve as an interesting example that
helps us understand how a business sector’s regulation and deregulation affect a country’s growth potential. In fact, the
GDP per capita level of Sweden in the 1960s corresponds with a large number of today’s developing countries, such as
Brazil, Bulgaria, Iran, Thailand, and South Africa. The Swedish case shows that imposing regulations without carefully
considering how these affect the incentives and efficiency of the business sector might be counterproductive. In particular,
our institutional and empirical analysis suggests that reforms that remove barriers to entry and growth for new and
productive firms and increase the return on capital and human capital investments, similar to those implemented in Sweden
during the 1990s, are likely to spur economic growth.
We also believe that our study is relevant for the ongoing academic and policy debate about the current crisis in Europe,
which tends to be dominated by macroeconomic and public sector considerations. Thimann (2015), for instance, notes that a
remarkable correspondence exists between impediments to growth and economic stresses in eurozone countries, such as
Greece.
5
Still, the top-level political discussion seems almost exclusively focused on macroeconomic issues, for instance,
how to handle the Greek government’s debt. The Swedish post-crisis experience suggests that the microeconomic
dimensions are crucial for turning an economy in crisis around. Therefore, to ensure recovery in these countries, the focus
on implementing long-run microeconomic reforms should increase.
2. PRODUCTIVITY AND EMPLOYMENT DYNAMICS IN SWEDISH INDUSTRY
Figure 1 compares the long-term development of GDP per capita in Sweden with that of US and a simple average of the EU
15 countries.
6
In 1970, Swedish GDP per capita exceeded the average GDP per capita of the EU 15 group but was lower
than that of the US. During the 1970–1990 period, Sweden performed worse than both the US and the EU 15 average. When
Sweden entered a severe crisis in the early 1990s, its GDP per capita fell below that of the EU 15 average.
In the years after the crisis in the 1990s, propelled by significant reforms to its economy, Sweden showed a much
stronger trend, and its GDP per capita grew faster than that of the EU 15 and kept up with US growth. At the end of the
1990s, Swedish GDP per capita again surpassed the EU 15 average. The gap in GDP per capita between Sweden and the EU
15 widened further after the turn of the millennium. In addition, Sweden clearly appears to have managed the recent crisis
better than the EU countries, as manifested by a further increase in the gap in GDP per capita as the financial crisis
unfolded. The latter observation suggests that Sweden’s performance, beginning in the mid-1990s, does not simply
represent the well-known “catch-up” phenomenon (growth increases simply because the economy takes off from a low level
of activity in which extensive idle production factors are available). Sweden’s apparent ability to sustain a relatively
4
As Besley (2015) writes in his overview article of the Doing Business project, case studies of countries complement analyses using the Doing Business
indices: “Policymakers in China or Brazil or Egypt have good reasons to be interested in how economies like Singapore or Sweden approach business
regulation without deciding blindly that they should copy these practices. The Doing Business rankings provide a way into this question – the basis
beginning a dialogue about policy reform. This discussion was not happening on a systematic basis before the Doing Business project came along.”
5
Using data fromthe World Economic Forum’s Global Competitiveness Index and the World Bank’s Doing Business dataset, Thimann (2015) finds that
several of the crisis countries have high and persistent barriers to firmformation and firmdynamics.
6
The EU 15 includes Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain,
Sweden and the United Kingdom, which were member countries of the European Union just prior to the accession of the East European countries on 1
May 2004. Several of these countries were not members of the EU (or EEC) at the beginning of the period considered. We use the term“EU 15” as a
convenient European benchmark.
5
superior performance over a very long period suggests that the Swedish economy underwent profound changes during the
1990s and that these changes laid the foundation for a sustained period of growth and a successful economic transformation.
Figure 1. Comparing long-term trends in GDP per capita for Sweden, the US, and the original 15 European Union
member countries, 1970–2013
Notes: GDP per capita is expressed in USD constant PPPs with base year 2005.
Source: OECD, National Accounts.
Let us now turn to the employment trend in Sweden during this period. Figure 2 compares the employment dynamics in
Sweden with those in the EU 15 countries and those in the US for the 1990–2011 period. Throughout the period,
employment as a share of the total labour force, the so-called participation rate, was very high in Sweden. After a substantial
post-crisis decline in the 1990s due to layoffs in the private and public sectors during the initial restructuring process, the
labour force participation rate steadily increased, and it was again substantially higher than those in the EU 15 and the US.
In particular, Sweden showed high growth in private sector employment as a share of its total labour force in the 1995–2011
period. Although substantially lower than the US, private sector employment surpassed the EU 15 levels after the financial
crises.
Figure 2. Total and private sector employment rates, 1990–2011
Notes: The EU 15 rate is based on our calculations. Private sector employment data are missing for Germany and Greece for the entire period. Total
employment data are missing for Germany in 1991 and for Greece until 1994. Private sector data for Austria are taken fromthe OECD Economic Outlook
No. 85, where historical data are only available until 2008. Employment rates are defined as employment as a share of the total labour force.
Source: OECD Economic Outlook No. 95, OECD Labour Market Statistics.
6
To understand the economic forces underlying the Swedish industrial and job restructuring processes, we will describe
the crucial economic reforms that Sweden implemented. However, first, we briefly survey the industrial restructuring
literature in search of the mechanisms through which these reforms could have had an impact on firm and job dynamics.
3. CONCEPTUAL FRAMEWORK: INDUSTRIAL RESTRUCTURING AND ECONOMIC REFORMS
To understand the restructuring processes that took place in Swedish industry in the 1990s and 2000s, we begin with an
overview of the basic economic mechanisms that have been important in explaining employment and productivity dynamics
in general.
7
Beginning from this general knowledge of the functioning of industrial restructuring processes, we analyse the
potential effects of the economic reforms undertaken in Sweden in the 1990s on employment and productivity dynamics.
This overview focuses on the economy’s microeconomic features, thus centring on what the World Bank refers to as the
“nuts and bolts” of the economy.
Figure 3 depicts a schematic picture of how firm and business sector employment and productivity can be understood.
The figure categorizes the “firm-specific factors” that firms can choose and influence. Firm-specific factors concern how
firms are organized or which business strategies are used. The overview suggests that expanding productive firms are
typically associated with active owners and up-to-date and incentive-based management practises. Thereby, economically
efficient decisions are made, and well-functioning business cultures are developed, thus leading to a motivated workforce.
Employing skilled workers and adapting new technologies early have also been shown to create competitive advantages in
local and global markets and to thereby spur productivity growth. Start-ups and expansions are often associated with high
degrees of uncertainty, meaning that many businesses will fail and, consequently, that the observed number of highly
successful and expanding ventures will be low.
The figure also includes “external factors”. Firms have no influence over these factors; however, these factors can
directly and indirectly affect firm productivity and employment through the limitations that they set or the incentives that
they provide regarding firm-specific choices. In our study of the Swedish business sector, we focus on external factors in a
large reform package implemented in Sweden in the early 1990s, which included reforms of the labour and product markets,
tax reforms and the removal of FDI barriers. We should also highlight that the information and communication technology
(ICT) revolution took place during this period. This external factor was very important to the industrial restructuring process
in most developed countries during this period. Finally, firm-specific and external factors affect firm performance in terms
of measurable productivity and employment dynamics. Different methods can be used to sum up these dynamics at the
aggregate business-sector level, as shown at the bottom of Figure 3.
7
See Acs and Audretsch (2005), Caves (1998), Santarelli and Vivarelli (2007), and Sutton (1997) for an overview of the literature on market structure and
firmdynamics. For specific articles, see, e.g., Audretsch (1991), Bartelsman et al. (2005), Ericson and Pakes (1995), Hjalmarsson (1974), Hopenhayn
(1992), Klepper (1996), J ovanovic (1982), and Luttmer (2007). In addition, see Nelson and Winter (1982) for an analysis of firmgrowth processes with
bounded rational decision makers. See Li and Rama (2015) for an analysis of firmdynamics, productivity growth, and job creation in developing countries.
7
Figure 3: The explanatory factors behind productivity and employment trends in the business sector
3.1. Firm-specific Factors
To compete in the marketplace, firms need to undertake efficient decisions in several dimensions. Short-term decisions
include optimal pricing and efficient marketing. Medium-term decisions involve aspects such as correct location of activity
and the hiring of productive staff. Finally, long-term decisions involve decisions on updated R&D and the optimal
organizational form.
8
We discuss some of the more important firm-specific factors below.
3.1.1. Business Strategy and Organization
Productivity and employment dynamics in firms essentially depend on the changes that firms – and their rivals – make to
their business strategies and organization. Firms that have good business ideas need to decide on how to grow – but what
should be produced internally, and what should be bought on the market when they expand? On the one hand, economies of
scale and scope imply that increasing firm size reduces costs and increases profits. On the other hand, larger firms face
problems related to free riding, a lack of control over activities, and lost motivation among staff, all of which, in turn, limit
the optimal firm size.
9
The optimal firm size also differs between individual firms and between industries and depends on factors such as
technology (ICT), market conditions (demand levels), and presiding institutions and laws (corporate tax system). Start-ups
8
See Besanko et al. (2003) for an overview.
9
The free-rider problemimplies that actors do not dare invest in development and hard work, as they then risk having competitors benefit fromthe returns
on these investments.
External factors:
• Labour market
• Corporate ownership market
• Product market
Firm-specific factors:
• Strategy and Organization
• Ownership and Management
• Capital and Human Capital
Firm characteristics:
• Productivity dynamics
• Employment dynamics
Private sector:
• Productivity dynamics
• Employment dynamics
8
and expansions are also associated with high degrees of uncertainty and problems of asymmetric information. Therefore,
many businesses will fail and, consequently, the observed number of highly successful and expanding ventures will be low.
Overall, these results indicate that firms with strong business ideas typically increase their productivity levels, but they
might not necessarily increase their employment levels due to labour saving or due to the outsourcing of non-core business
activities.
10
3.1.2. Ownership and Management
Expanding productive firms are typically associated with active owners and up-to-date and incentive-based management
practises. Thereby, economically efficient decisions are made at the right time. Moreover, active ownership typically creates
a well-functioning business culture, leading to a motivated work force.
11
Why then do inefficient firms not implement more efficient management? First, some firms are family owned with
management that is difficult to replace. In addition, firms possibly face problems of corporate control, whereby managers use
their superior information to shirk their responsibilities or to hide their incompetence. Different incentives and monitoring
systems have been developed to mitigate these problems (see, e.g., Tirole, 2006).
3.1.3. Capital and Human Capital
Some firms are able to expand and maintain high productivity because they educate and hire productive employees and
invest in high-quality capital. Efficient human resource management enables firms to acquire talent and further develop their
skills so that they can facilitate the generation of high profits and firm expansion.
12,13
The implementation of ICT is a
prominent example of how the adoption of new technology was able to spur firm growth and productivity.
14
The acquisitions
of small growing firms are another important explanation for growth in expanding firms’ employment and productivity.
15
Moreover, multinational enterprises (MNEs) use their firm-specific assets to undertake FDIs in different countries. FDIs are
typically achieved by either setting up new plants (greenfield investments) or acquiring existing domestic target firms (cross-
border M&As). These FDIs often have positive externalities on labour through higher wages and on local firms through
knowledge spillovers.
16
Let us now examine how the changes in external factors caused by the economic reform package undertaken in Sweden
in the 1990s affected firm performance.
10
The literature that addresses firmformation and size was founded by Coase (1937) and was further developed by Williamson (1979). Grossman and Hart
(1986) and Hart and Moore (1990) then developed formal frames of analysis to study these questions, focusing on how the division of ownership affected
the different stakeholders’ incentives to invest in a firm’s development. See Rajan and Zingales (2001) for an application for entrepreneurship and enterprise
development. See the “Part One: Firmboundaries” chapter in Besanko et al. (2003) for an overview.
11
Bloomand Van Reenen (2007) find that firms with higher-quality management are more productive, and they argue that the lower aggregate productivity
in the UK and France compared with that in the US can be explained by the lower level of competition in the UK and France, where, in the absence of
competitive pressure, inefficient firms are not forced out of the market. Bertrand and Schoar (2003) follow individuals who have been CEOs at different
companies and show that CEO quality has an effect on how profitable these firms are. Lazear (2000) and Bandiera et al. (2007 and 2009) also show the
connection between good leadership and high productivity.
12
See Gibbons and Roberts (2013) and Murphy and Topel (1990) for an overview.
13
Ilmakunnas et al. (2004) use Finnish data and show that productivity increases with the employee education level and age. At the same time, Fox and
Smeets (2011) show that considerable differences in productivity remain between firms when they control for education level, gender, work experience,
and employment duration. Thus, labour force quality explains some, though far fromall, differences in productivity between firms.
14
The findings in Van Ark, O’Mahony and Timmer (2008) suggest that the slower productivity development in the EU compared with that in the US can
partly be explained by lower investments in ICT.
15
Indeed, Maksimovic and Phillips (2001) show that a large share of the plants in the US manufacturing industry change owners each year (up to 7 per cent
in some years) and that the productivity levels of these plants increase.
16
See Barba Navaretti and Venables (2004)for an overview of the MNE literature, J avorcik (2015) on the literature on FDI and job creation and Saggi
(2002) on the literature on FDI and technological spillov ers.
9
3.2. External Factors: The expanded Regulations in the 1970s and 1980s and the Structural Economic Reforms of the
1990s
We first describe the institutional setting of the Swedish business sector and then turn to the crucial reforms that were
implemented in the 1990s. We rely on detailed descriptions of the Swedish business sector and the policy reforms that
affected firms in Sweden, as described by, e.g., Bergh (2014), Bergh and Erlingsson (2006), Calmfors (2012), Edquist and
Henrekson (2013), Henrekson and J akobsson (2005), J onung et al. (2008), Lindbeck (1997), and the references therein.
Let us begin with a brief description of the development of economic institutions in Sweden prior to the reforms. Staying
out of two world wars and engaging in international trade by exploiting its abundant natural resources while developing
efficient institutions, Sweden experienced a long period of sustained growth, the so-called “golden years” of 1870–1970. At
the end of this century-long period, Sweden was fourth in the OECD rankings of GNP per capita. In the decades after the
Second World War, a relatively rapid GDP growth rate was combined with full employment and a fairly egalitarian
distribution of income.
In the 1970s, government policies became increasingly interventionist under the influence of the more radical political
ideas that emerged during the decade. Tight labour market regulations were implemented in the early 1970s. So-called
“solidaristic wage policies” led to a compressed wage structure, and workers’ wages became detached from individual firm
productivity. Marginal tax rates gradually increased, ultimately culminating in a 1971 tax reform that made Sweden’s tax
rate very high in comparison with those of comparable countries.
In the business sector, the government, trade unions and bank-related business groups embodied an explicit tripartite
negotiating culture. A fairly small number of dominating owners or ownership groups of corporations acknowledged and
accepted that the government would use its political power to implement far-reaching welfare reforms as long as the labour
movement abstained from socializing the industrial sector. Moreover, the government attempted to influence aggregate
savings, the credit supply and investment through public sector savings, capital market regulations, taxes and subsidies,
which all affected the functioning of the business sector. As noted by Lindbeck (1997), this approach mirrors a world view
in which markets, economic incentives and private entrepreneurship not associated with large firms are regarded with
suspicion.
These interventionist policies reduced the efficiency of the economy and likely served as an important factor in
Sweden’s inferior performance compared with those of the EU 15 countries and the US during the 1970–1990 period in
terms of GDP per capita growth. Internal problems in the Swedish model and external shocks eventually led to deep
economic crises that included a significant decrease in output and soaring unemployment in the early 1990s. In response,
economic-efficiency and growth-induced institutional reforms were undertaken in the late 1980s and 1990s. In addition,
macroeconomic policy reforms were implemented in the 1990s to reduce the inflationary bias in the Swedish economy.
These reforms included the establishment of an independent central bank and a floating currency.
Most of the literature has focused on the importance of macroeconomic reforms. By contrast, this study focuses on the
reforms that were undertaken to improve resource allocation and the microeconomic functioning of the markets in response
to the underperforming Swedish economy.
17
Notable reforms included the decentralization of the wage negotiation system
and the liberalization of temporary work contracts, the deregulation of the product market, greater openness to inward FDI
and tax system reforms (Bergh, 2013, Calmfors, 2012, Edquist and Henrekson, 2013 and Lindbeck, 1997).
In the following sub-sections, we proceed with a detailed review of the Swedish reforms before attempting to present
empirical evidence showing that these reforms increased economic efficiency in the Swedish business sector. To provide
such evidence, we use Swedish micro data to study different facets of productivity and employment dynamics.
17
See Besley (2015) and Thimann (2015) for recent articles that focus on the importance of microeconomic foundations.
10
3.2.1. The Labour Market
Labour market regulations significantly affect firm employment and productivity development. On the one hand, labour
market institutions can increase hiring and productivity by reducing matching and search problems in the labour market. On
the other hand, rigid labour markets may offer too much protection to insiders, thereby hampering creative destruction
processes and, to a lesser extent, rewarding productive labour and firms. Various types of labour turnover costs give insiders
market power, which has implications for talent allocation, work incentives, and employment and unemployment patterns
(see Lindbeck and Snower, 2001, for an overview). In particular, such insider market power might distort incentives for firm
development, education, and efforts in the workplace. We refer to this problem as the insider-outsider problem of the labour
market.
Let us now use this background to discuss the impact of the crucial labour market reforms in Sweden in the 1990s, the
decentralization of the wage negotiation system and the liberalization of temporary work contracts on the efficiency of the
restructuring of the Swedish business sector.
3.2.1.1. The Decentralization of the Swedish Wage Negotiation System and the Liberalization of Temporary Work
Contracts
After the Second World War, wage bargaining was highly centralized in Sweden. In the 1950s and 1960s, economy-wide
wage increases were negotiated centrally between the Swedish Employers’ Confederation and the Trade Union
Confederation. In the 1960s and 1970s, wages were set according to solidaristic wage policies under the principle of “equal
pay for equal work”, whereby wages would be equalized between sectors for similar tasks and occupations. Ideally, the
system would have mimicked a competitive labour market, in which low productivity firms would be driven out of the
market, thus freeing labour to seek high-productivity firms that were able to support higher wages. In practice, however,
considerable wage compression occurred as ambitions moved from equity goals to promoting more outright equality (Davis
and Henrekson, 2000 and Lindbeck, 1997). A market mechanism through which high-productivity firms could attract labour
by paying higher wages did not exist; instead, active labour market policies were pursued, whereby resources were allocated
to help the unemployed gain new competences and to reduce frictions and search costs in the labour market. How well the
government was able fulfil this allocative task is debatable. With constraints on wage setting, the ability to incentivize
workers in firms was also hampered.
In 1974, a new employment protection law (LAS) was implemented. The law mandated that employees could not be
fired without reasonable cause, such as abuse or a lack of work opportunities. Insiders were also favoured with respect to
firing and hiring procedures through the so-called “last in, first out” rule, which further reduced workers’ incentives to
change jobs. Temporary contracts also became limited. Figure 4 shows that the introduction of LAS had a significant impact
on the so-called Allard index of the strictness of employment protections (Allard, 2005), which nearly doubled during this
period. This measure of the strictness of employment protections continued to increase until the beginning of the 1990s. The
average employment protection in the EU 15 countries also increased over this period, though not to the same extent as in
Sweden. The US labour market continued to maintain a low level of employment protection.
11
Figure 4. Employment protection legislation, 1950–2003
Notes: Scale 0–5, the higher the index is, the stricter the employment protection legislation is. The EU 15 values are based
on our calculations, and they exclude Luxembourg.
Source: Allard (2005).
The push for radical reforms produced a tenser climate in the relations between the unions and the Employers’
Confederation. Centralized bargaining for private-sector, blue-collar workers gradually broke down in the 1980s, and it was
replaced by uncoordinated industry-level bargaining. Intermediate industry-level bargaining is a form of collective
bargaining that should be more conducive to wage inflation.
18
In 1990, the Employers’ Confederation attempted to
introduce a more decentralized system. However, this attempt failed, and, instead, a fully centralized wage stabilization deal
was negotiated for 1991–1992. In 1994, state-owned firms joined the employers’ organization, which weakened the political
influence in wage setting (Nycander, 2008). In 1997, the so-called Industry Agreement was concluded. The agreement
included a system that continued industry-level bargaining but with strong informal coordination based on pattern
bargaining with the manufacturing sector to conclude initial wage agreements in a bargaining round. This system
established a norm for wage increases for others to follow. The reformed wage-bargaining system turned out to be
consistent with lower nominal wage increases than those in the past. Moreover, it allowed for greater individual wage
flexibility (Calmfors, 2012). Sweden thus progressed from a more coordinated wage negotiation system than those in other
EU countries in the 1980s to a moderately coordinated wage negotiation system in the 1990s.
In 1992, a major employment protection reform was implemented that permitted staffing agencies (Skedinger, 2010),
and the regulations concerning temporary work were relaxed. This development created what is referred to as the dual
Swedish labour market, with strong employment protections for regular workers and weak employment protections for
temporary workers. This reform was also evident in the Allard index of employment protection, which declined
significantly. Another measure of the strictness of employment protections from the OECD demonstrates that the strictness
of employment protections concerning temporary contracts was significantly reduced in Sweden – from a very high level in
1985 to a very low level in 2010. However, the strictness of the employment protections concerning regular contracts
remained at a relatively high level over the same period.
We conclude our description of the reforms of the labour market as follows:
18
Both highly coordinated wage bargaining and decentralized firm-level bargaining deliver higher wage moderation: highly coordinated bargaining does so because wage setters
are forced to make economy-wide considerations, and decentralized bargaining does so because wage setters have to consider competitive pressures. This hypothesis seems to be
borne out by the high wage increases in Sweden in the 1980s (Calmfors, 2012).
12
Conclusion 1. The incentive and insider-outsider problems in the Swedish labour market may have been mitigated by the
labour market reforms undertaken in the 1990s. These reforms may also have improved firms’ flexibility and thereby their
ability to adjust their workforce and invest in and reward human capital.
3.2.2. Product Market Regulation
The absence of artificial barriers to entry and expansion is crucial for employment and productivity growth. Incumbent
firms have incentives to exploit their market power to protect their market share by preventing rivals from expanding and
new firms from entering their markets. For instance, incumbent firms can practice different forms of predatory behaviours,
such as engaging in exclusive dealing contracts or input cartels, lobbying for special restrictions on entry, or making entry-
deterring acquisitions. Even if incumbents are ineffective, they may not be replaced by more productive entrepreneurs due
to excessive barriers to entry.
We refer to these product market problems as the problem of weak creative destruction. Well-functioning competition
policy and legislation can mitigate such entry-deterring and predatory problems (see Motta, 2004 and Tirole, 2006).
Moreover, a well-functioning competition policy must ensure that innovative firms are not deterred by rivals to expand and
are able to reap the benefits of temporary market power. Moreover, these innovative firms need to put competitive pressure
on firms that are lagging behind (Aghion et al., 2005, Norbäck and Persson, 2012, and Vives, 2008).
3.2.2.1. The Deregulation of Product Markets in Sweden
Throughout most of the twentieth century, many product markets in services in Sweden were public monopolies. Thus, new
firms had no or very few opportunities to enter these markets, and consumers’ influences were limited (SOU 2005:4).
Moreover, the competition law was rather lax for a long period. The first competition legislation was implemented in 1925,
which enabled authorities to investigate companies that could have monopolistic characteristics. In 1946, the legislation
became stricter and centred on the idea of monitoring competition restrictions in the business sector.
19
Nevertheless, from an
international perspective, the competition law was very lax. A new Competition Act was implemented in 1993 that was
based on three cornerstones: the prohibition of restrictive agreements, the prohibition of abuse of dominance, and the
prohibition of control of concentrations (mergers). This new competition law indicated that the competition policy had
become much stricter.
In the 1980s, discussions concerning how to reform the Swedish welfare state became increasingly intense. The centre-
right government that came to power in 1991 was seemingly intent on implementing an economic policy based on extensive
deregulation in response to the country’s economic crisis in the 1990s. However, the possibility of making such reforms,
e.g., the deregulation of the air traffic system, the electricity market, and the postal service, had already been thoroughly
investigated, and government policies had previously been outlined in government white papers prepared by the Social
Democratic government in the late 1980s and early 1990s. Moreover, in 1993 (SOU 1993:16), the so-called Lindbeck
Commission presented a number of proposals to improve the efficiency and functioning of markets in Sweden (Lindbeck et
al., 1994). Overall, the intensity of competition increased substantially in many Swedish product markets during the 1990s.
The OECD has long calculated an index of the “knock-on” cost that regulations in the service and utility industries
impose on manufacturing industries. This index is shown in Figure 5, where we see that regulations on utilities and services
19
This monitoring was enabled not only through investigations but also through the registration of cartels in a public record called the Cartel Register (CR), or Kartellregistret.
The idea behind the CR was to highlight the extent of anti-competitive agreements in the Swedish business community and, in doing so, to help prevent the potential adverse
effects of such agreements on competition.
13
imposed high additional costs on manufacturing in the 1970s and 1980s, but these costs decreased sharply as Sweden began
to deregulate in the late 1980s and early 1990s. Thus, beginning in the mid-1990s, the costs of regulation in the services and
utilities sectors were substantially lower in Sweden than the average of the EU 15, and they were even lower than such costs
in the US and the UK.
Figure 5. Regulation impacts, 1975–2007
Notes: Measurement of potential costs of anti-competitive regulation in intermediate input sectors. The EU 15 values are
based on our calculations, and they exclude Luxembourg.
Source: OECD Indicators of Regulation Impact.
Since the late 1990s, the OECD has also constructed a system of indicators to measure on-going developments in
product market regulation (PMR) across the OECD countries (Wölfl et al., 2009). For Sweden, the “barriers to
entrepreneurship” category has improved the most. Particularly between 1998 and 2008, considerable improvements were
made to licensing and permit systems and communications. Furthermore, simplifications of rules and procedures were
made; certain legal barriers were removed; antitrust exemptions were allowed; and barriers to competition in network
sectors and services were reduced.
The product market reforms substantially reduced the power of the iron triangle of the Swedish business sector: the
government, incumbent firms, and unions. A crucial feature of these product market reforms was that they not only made it
easier for new firms to enter industries but also made it more difficult for inefficient firms to remain in the product market.
We can summarize our description of Sweden’s product market reforms as follows:
Conclusion 2: The deregulation of the Swedish product markets and the strengthened competition policy may have
mitigated the weak creative destruction problem in the Swedish business sector. These reforms may have forced inefficient
firms out of the market, thus making room for more productive entrants, but they also may have caused incumbent firms to
reach their potential through more intensive development.
14
3.2.3. Corporate Taxation
The tax system affects both incentives for firms to invest and the firm formation process. Mirrlees et al. (2011) show that
the tax system’s treatment of the cost of capital can distort firms’ investment incentives. At zero inflation and for an asset
for which the true decline in value over its lifetime matches the tax depreciation schedule, the corporate tax rate does not
affect the required rate of return for corporate investments that are financed by debt. However, investments financed by
equity are affected. Shareholders require a positive rate of return to compensate for the income that they could have earned
by investing in an interest-bearing asset. However, this “opportunity cost” of equity financing is not deductible from taxable
profits, which means that debt financing is favoured over equity financing, thus discouraging corporate investments
financed by equity.
Several contributions in the corporate tax literature also focus on the progressiveness of the personal income tax
schedule as an obstacle to firm formation activity (e.g., Gentry and Hubbard, 2000). Keuschnigg and Nielsen (2004a, b)
focus on the effects of various tax policies when entrepreneurs face financial constraints and must enter into contracts with
venture capitalists under conditions of one-sided or two-sided moral hazards. Haufler et al. (2014) show that the tax system
might create distortions in the type of projects that entrepreneurial start-up firms undertake.
20
Thus, the corporate tax system runs the risk of reducing the incentives for investments and of distorting the market’s
efficiency by favouring certain types of corporate ownership over others, such as incumbents over start-ups. We refer to this
problem as the tax incentives and discrimination problem in the business sector. Let us now use this background to discuss
the implications of the Swedish reforms for the corporate ownership market to predict how these reforms might have
affected the performance of the Swedish business sector.
3.2.3.1 Corporate Taxation Reforms in Sweden
Owners of corporations in Sweden are taxed on two levels. They pay taxes indirectly through corporate taxes, and they pay
taxes on capital gains. Corporate taxes increased substantially in Sweden during the 1970s and the 1980s, leading to very
high corporate taxes from an international perspective. Due to the increased awareness of the negative effects of high
taxation on business activities and increased international competition, the tax was reduced from 52 to 30 per cent in the
1990–1991 tax reform package. The rate was then further reduced to 28 per cent in 1994.
The Swedish tax rate on capital gains for long-term holdings was zero until 1965. The tax changes implemented in 1976
sharply increased the top marginal tax rate to more than 30 per cent, and it peaked in 1979 at nearly 35 per cent. Thereafter,
it decreased to approximately 25 per cent prior to the 1990–1991 tax reforms (Stenkula et al., 2015). The 1990–1991 tax
reforms made all capital gains fully taxable irrespective of the holding period. However, capital gains were no longer taxed
jointly with labour income; instead, a separate capital income tax was implemented at a flat rate of 30 per cent.
21
Importantly, until 1991, the Swedish tax system favoured large firms and institutional ownership (e.g., pension funds
and insurance companies). The 1991 tax reform and subsequent minor reforms considerably levelled the playing field for
different combinations of owners and sources of financing (Davis and Henrekson, 1999 and Edquist and Henrekson, 2013).
20
The empirical literature on the productivity effects of corporate taxation has mostly suggested effects fromindirect channels, such as R&D and capital investment. The
relationship between tax policy (mostly R&D tax credits) and the volume or location of R&D across countries and the US is reviewed by Hall and van Reenen (2000). See
Auerbach (2002), Gordon and Hines (2002), Hasset and Hubbard (2002), and Hines (2005) for analyses on the relationship between taxation and investments. Evidence of the
effects of taxation on industry entry and exit rates is presented by Da Rin et al. (2011), Djankov et al. (2010), and Kneller and McGowan (2012). Carroll et al. (2000, 2001)
examine the effects of the US tax reforms in the 1980s on the investment and hiring decisions of small businesses and find significant effects. Arnold et al. (2011) demonstrate
that corporate taxation has a direct effect on firmproductivity growth by lowering the growth rate of firms that are in more profitable industries. Using data for 11 European
countries, Gemmell et al. (2013) find evidence that productivity growth in small firms is lower when the corporate tax rates are higher.
21
From1992–1993, this separate capital income tax rate was temporarily reduced to 25 per cent, and it was temporarily reduced to 12.5 per cent in 1994.
15
The reforms in the 1990s generated a tax system that was far more positive for individuals who wished to start, develop, and
act as controlling firm owners compared with the situation in the 1970s and 1980s.
However, only examining the (marginal) tax rates to judge a tax system’s effect on firm performance is insufficient.
Evaluating a tax system’s effect on corporate capital investment is a complicated task. Many aspects, such as the project
type and the financing form, need to be considered. Here, the devil is in the details. A generally accepted method of
evaluating a capital tax system is to calculate the marginal effective tax rate (METR) on capital based on the method
originally presented by King and Fullerton (1984).
Devereux et al. (2002) use a simplified version of this method to compare the marginal effective corporate taxes for a
number of EU countries, J apan and the US in 1982 and 2001. These comparisons are shown in Figure 6. The figure shows
that Sweden had the second highest rate in 1982 at approximately 53 per cent but the second lowest rate in 2001 at
approximately 22 per cent. Hence, corporate taxes were reduced from very high levels to low levels during the reforms of
the 1990s.
Figure 6. Effective marginal corporate tax rates
Notes: Replication of author’s Figure 5, pg. 462, “Calculations based on a hypothetical investment in plant and machinery
for one period, financed by equity or retained earnings (but not debt). Taxation at the shareholder level is not included. The
project is expected to break even, i.e., there is no economic rent. Other assumptions are that the real discount rate is 10 per
cent, the inflation rate 3.5 per cent, and the depreciation rate is 12.25 per cent”.
Source: Devereux et al. (2002)
Using the King and Fullerton (1984) approach, Stenkula et al. (2015) examine the METR on capital for a longer period for
Sweden. This study provides further evidence of Sweden’s high METR for external capital (not retained earnings or
institutional capital) in the 1980s (approximately 100 per cent) and its significant reduction in the 1991 tax reform. These
corporate and capital tax reforms created opportunities for firm development, particularly for the growth of new small firms
and firm formation. Thus, we come to the following conclusion:
Conclusion 3. Sweden’s corporate tax reforms in the 1990s may have mitigated the problem of outsider discrimination in
the market for corporate ownership. The reduced taxes on corporate external financing may have led to increased entry and
the growth of new, productive firms in the Swedish business sector.
16
3.2.4. Foreign Direct Investment (FDI)
Business regulation affects the actions that firms can take and the balance of power that exists between various firm
stakeholders. Politicians may benefit from protecting owners from competition by gaining political support or by sharing the
rents that result from such protection (Olson, 1965, Stigler, 1971, and Perotti and Volpin, 2007). Moreover, in more open
economies, lobbying for international protection might occur (Spencer and Brander, 1983, and Grossman and Helpman,
1994). Politicians might also have an incentive to favour domestic owners in the market for corporate control (Horn and
Persson, 2001, and Norbäck et al. 2014).
Thus, regulation might affect the efficiency of the corporate ownership market by favouring certain types of ownership
over others, such as domestic ownership over foreign ownership. We refer to the problem as the foreign discrimination
problem in the market for corporate ownership. Let us now use this background to discuss the implications of the Swedish
reforms for the corporate ownership market to predict how such reforms might have affected the performance of the
Swedish business sector.
3.2.4.1. The Liberalization of Foreign Direct Investment in Sweden
Foreign exchange controls were introduced in Sweden shortly after the onset of the Second World War. In practice, this
legislation excluded any substantial foreign ownership of Swedish industry. The purpose of this legislation was openly
protectionist, i.e., to ascertain that “Swedish firms remain controlled by Swedish interests” (SOU 1986:23, p. 143). As
expected, legal impediments ensured that foreign ownership remained low, with foreign ownership of listed stocks never
exceeding 8 per cent throughout the 1980s; in addition, less than 5 per cent of private sector employees worked in foreign-
owned companies (Henrekson and J akobsson, 2005).
Between 1989 and 1993, the government undertook measures that opened the market to foreign ownership. This change
could be considered the final deregulation of the Swedish capital market that began in the early 1980s, thus following a
global trend of credit market deregulation in response to the more globalized economy (see Henrekson and J akobsson, 2005,
for a description of the major steps in the international deregulation process).
From a mere 7 per cent in 1989, the share of foreign ownership skyrocketed to 40 per cent only ten years later
(Henrekson and J akobsson, 2005). This increase also led to significant growth in the share of employees working in foreign-
owned firms, which increased from approximately 5 per cent at the end of the 1980s to 23 per cent in 2011. This change in
inward FDI was quite dramatic, even from an international perspective. Figure 7 shows that the inward FDI stock, as a
percentage of Sweden’s GDP, was approximately 5 per cent in the early 1990s, which was approximately half the EU 15
average. After the 1990s, Sweden’s inward FDI stock became substantially higher than the EU 15 average.
22
22
The increase in foreign ownership was especially strong in the mid-1990s. Employment in foreign-owned firms almost tripled between 1995 and 2013, fromapproximately
240,000 in 1995 to 630,000 in 2013.
17
Figure 7. Inward FDI stock (per cent of GDP), 1990–2013
Notes: The EU 15 values are based on our calculations, and they exclude Luxembourg.
Source: World Investment Report.
The injection of foreign ownership likely improved productivity development in the Swedish business sector. Having a
larger pool of potential owners should increase the potential for synergies. Foreign ownership may increase productivity
through the better use of assets, but bidding competitions may also generate large asset returns for previous Swedish
owners, who can then use these proceeds to invest in new projects or industries (Norbäck and Persson (2007).
We conclude the description of Sweden’s liberalization process of inward FDI in the 1990s as follows:
Conclusion 4. The liberalization of inward FDI in Sweden may have substantially mitigated the problem of foreign
discrimination in the market for corporate control in the Swedish business sector. These reforms may have caused more
efficient foreign owners to acquire inefficient Swedish target firms, which should have improved these firms’ productivity.
Moreover, this development may have spurred the incentive to create start-ups for sale in the market for corporate control.
4. A SUMMARY OF MICRO-DATA EVIDENCE FROM SWEDISH MATCHED EMPLOYER-
EMLOYEE DATA
The previous section showed that, after a period of interventionist policies in the 1970s and 1980s, Sweden deregulated its
product and labour markets, reformed its corporate tax system and opened itself up to FDI. We have argued that these
reforms should have reduced incumbents’ and insiders’ advantages and benefitted the growth of new efficient firms. This
section presents a summary of our previous work on empirical evidence based on matched employer-employee data. Details
of the empirical analysis can be found in Heyman, Norbäck and Persson (2015).
Data
The empirical analysis requires that we follow firms and individuals over time, thus necessitating access to highly detailed
data. Therefore, the analysis in Heyman, Norbäck and Persson (2015) was based on detailed employer-employee data from
SCB, which cover the 1990–2009 period.
23
The data originate from several register-based datasets from SCB and cover all
23
See, e.g., Davidson et al. (2014) and Hakkala et al. (2014) for recent articles based on these data.
18
firms in the private sector. First, the financial statistics contain detailed firm-level information on all Swedish firms in the
private sector during the 1996–2009 period. Variables include value added, capital stock (book value), the number of
employees, total wages, ownership status, profits, sales, and industry affiliation. Second, the Regional Labour Market
Statistics (RAMS) include data on all plants for the 1990–2009 period. The RAMS add information on the composition of
the labour force with respect to educational level and demographics at the plant level, which we aggregate to the firm level.
Individual-, plant- and firm-level data can be linked together using unique tracking numbers. From an international
perspective, the data are rather unique in terms of both magnitude and the level of detail.
A potential problem in an analysis of employment dynamics is the difficulty of following firms over time. Using
organization numbers as a method of identifying continuing, entering, and exiting firms can be problematic because such
numbers can change for various reasons. To more reliably follow firms over time, we used additional data from SCB. These
data make identifying new firm entries and firm exits possible, which means that we can analyse employment changes in (i)
completely new units, (ii) continuing units and (iii) exiting units.
To measure productivity, we used labour productivity, defined as value added per employee. Value added per employee
is a commonly used measure of productivity and is easily comparable across countries. Value added is calculated as the
output value minus the costs of purchased goods and services, excluding wages and other personnel costs (calculated by
SCB according to the international definition).
24
4.1. Allocative Efficiency
In the previous section, we emphasized the structural reforms that began in the 1980s: the reforms in the product market, the
reforms affecting inward FDI, and the labour market reforms with more decentralized wage setting and less job security for
workers with temporary contracts. We also emphasized that the tax system discriminated against smaller firms with high
growth potential.
In this section, we summarize the evidence regarding the reforms’ effects on the efficiency of the economy in Heyman,
Norbäck, and Persson (2015). We begin by using a productivity decomposition proposed by Olley and Pakes (1996) to
analyse productivity and reallocation. The Olley and Pakes method divides aggregate productivity into two terms, thus
implying that the weighted productivity in the business sector can be written as the sum of the simple (unweighted) average
productivity over all firms and the covariance between their productivity and market share.
25
The second term has a natural
efficiency interpretation term, and it can be interpreted as the extent to which market share is allocated to high-productivity
firms. If the covariance between firms’ productivity and their share of labour is strictly positive, then more productive firms
will tend to attract larger shares of workers, which is what we would expect in a well-functioning market economy.
To examine whether such allocative efficiency has changed over time in Sweden, we computed the Olley and Pakes
covariance term at the two-digit industry level for each year during the 1996–2009 period. Results show an increasing
allocative efficiency in Sweden, which is consistent with the notion that the reforms should have improved the market’s
allocation of resources.
26
We observe much higher estimates of the allocative efficiency term in the final years of our
sample compared with the first years.
This calculation presents one drawback: we cannot compare developments in Sweden with those in other countries.
However, to analyse how structural policies affect resource allocation efficiency, Andrews and Cingano (2014) use firm-
level data from a commercial data source covering 21 OECD countries in 2005. Investigating the Olley and Pakes
24
Another measure of productivity is Total Factor Productivity (TFP). Studies that use both labour productivity and TFP typically find similar results, irrespective of
the measure used (see, for instance, Bartelsman and Doms (2000) and Syverson (2011) for discussions of different productivity concepts).
25
See, e.g., Foster el al. (2001) for details. One advantage of their cross-sectional decomposition method is that cross-sectional productivity differences are more persistent and
possibly less sensitive to measurement errors and temporary shocks. The Olley and Pakes approach also does not depend on how firmentries and exits are measured.
26
See Heyman, Norbäck and Persson (2015) for details.
19
covariance, they find that Sweden has the largest allocative efficiency. This result is consistent with the substantial changes
in Sweden that we have accounted for in the previous section and the increase in allocative efficiency. Interestingly,
Andrews and Cingano (2014) also examine the source of the variation in the allocative efficiency term. The authors report
that regulations related to employment protection, product market competition, and FDI are negatively related to
productivity through a worsening of allocative efficiency, which indicates a reduced ability to allocate resources to more
productive firms.
A high degree of allocative efficiency implies that highly productive firms are able to attract workers from less productive
firms. This mechanism was weakened under the solidaristic wage policy, as described in our institutional analysis. The
aggregate picture from our empirical analysis appears consistent with the view that the deregulation of the Swedish wage-
setting system implied that productive and expanding firms found hiring and rewarding productive employees easier.
4.2. Where does the increase in productivity come from?
Our institutional analysis suggested that labour market reforms and particularly product market reforms combined with tax
reforms may have reduced the barriers to entry and that these actions might have played an important role in the turnaround
of the Swedish business sector by improving the creative destruction process. To distinguish the effect of the entry of new
firms and the exits of incumbents from the expansion and contraction of existing firms, we have also applied a
decomposition method to analyse the drivers of overall productivity in greater detail (see Foster et al. (2001) for a
discussion of different decomposition methods). The decomposition allows us to distinguish aggregate productivity changes
at the intensive margin from those at the extensive margin. Again, see Heyman, Norbäck and Persson (2015) for details.
Using firm-level data allows us to disentangle overall productivity growth into different components. This can help us
determine if Swedish productivity growth originated from within-industry dynamics (firm-level productivity growth), a
reallocation of market shares between existing firms (incumbents), or the entry and exit of firms. Even if the productivity of
individual firms did not change, productivity could have changed substantially due to changes in the market shares of firms
with different productivity levels. These insights are difficult to obtain with more aggregate data.
Figure 8 summarize the results from our productivity decomposition for the entire 1996–2009 period. Here, we see that
more than half of the overall increase in productivity in the business sector originated from new firms. The new firms that
survived gradually became more efficient than the average firm and thus contributed positively to long-term productivity
growth. As seen in Figure 8, entering firms’ contribution to productivity growth exceeded that of incumbents (firms that
were active throughout the 1996–2009 period). Hence, over the period studied, the entry of new firms was clearly the main
driving factor behind the increase in productivity in the Swedish business sector.
27
This result is consistent with the lower
entry barriers in Sweden enhancing the creative destruction process (Conclusion 2). Increased entry also emerged due to
corporate tax reforms (which levelled the playing field between entrants and incumbents) by promoting new firm start-ups
and, as we will see below, opening up the economy to FDI (Conclusions 3 and 4).
27
Braunerhjelmand Carlsson (1993) show that the number of small firms in the Swedish business sector decreased substantially compared with that in other industrialized
countries during the 1970s and 1980s. Henrekson et al. (2012) show that, in 2008, the firmsize distribution in Sweden had again become more similar to other comparable EU
countries.
20
Figure 8: Labour productivity growth decomposition, 1996–2009
Figure 8 also shows similar results for firms in the manufacturing and service sectors. The overall change in productivity
between 1996 and 2009 appears to have been somewhat larger in the manufacturing sector than in the service sector. We
also observe a positive cross effect in the manufacturing sector. This result indicates that the established manufacturing
firms that expanded also increased their productivity (or that established manufacturing firms that reduced their productivity
also experienced decreasing employment shares). Interestingly, this cross effect is negative in the service sector. This result
is consistent with the fast-growing nature of the service sector, where many expanding firms experienced declining
productivity during their growth phase.
4.3. Liberalization of Foreign Direct Investment
One of the major reforms undertaken in Sweden was the lifting of restrictions on foreign ownership. This reform led to
remarkably strong employment growth in foreign-owned affiliates in Sweden between 1980 and 2013, when nearly one-
fourth of workers were employed by foreign-controlled firms. We have also argued that the increase in foreign ownership
represented a much-needed productivity boost in the business sector, as a much larger pool of potential owners became
available.
The impact of the foreign acquisitions of Swedish firms on productivity in the Swedish business sector is an empirical
question. Heyman, Norbäck and Persson (2015) also provide a detailed empirical analysis based on matched employer-
employee data. The results show that, on average, labour productivity increased by approximately three per cent when
ownership was transferred from Swedish ownership to foreign ownership. This effect was completely driven by local
Swedish firms without any foreign operations being acquired by foreign firms. When we examine the effect of the foreign
acquisition of Swedish multinationals, we find no statistically significant effect. This result seems to be consistent with the
theory described in Section 3, i.e., that synergies may be more easily generated when an MNE acquires a local firm than
when two MNEs merge.
We also repeated the analysis of foreign ownership and foreign acquisitions on other performance measures, such as the
average wage and employment. The results indicate the existence of a significant wage and employment premium when
considering both average differences and the average change after an acquisition. Thus, foreign ownership and acquisitions
also appear to have contributed to higher employment and wages, which is what we would expect if foreign firms provide
new knowledge, better management, and better products and production methods.
21
Aggregate effects
How important have foreign firms been for aggregate productivity growth? To provide an answer we extend the
decomposition method developed by Foster et al. (2001) to also distinguish between Swedish and foreign firms. Thus, we
compare the different components that contribute to productivity growth, but we then decompose each component into a
Swedish and a foreign part. The results are summarized in Figure 9. Foreign-owned firms clearly contributed more to
productivity growth than did domestic Swedish firms. In fact, both the within-firm increase in productivity and the
productivity increase from entry are almost twice as large for foreign-owned firms compared with Swedish-owned firms.
The reason for the smaller overall difference in productivity growth is that the cross effect is negative for foreign-owned
firms. However, as explained above, this result may be due to significant expansions of foreign firms, where the
productivity in the expansion phase is below average productivity as the firm is built up.
Figure 9: Decomposition of labour productivity growth for the entire economy separated by ownership status, 1996–
2009
5. CROSS-COUNTRY EVIDENCE
Of course, the reform programme that Sweden undertook after the crisis included other measures. Monetary policy was
changed so that price stability was targeted under a floating exchange rate. As interest rates fell and the Krona depreciated
in the mid-1990s, demand and exports increased. Causally sorting out the most important factors – the “micro reforms”
-0,05 0 0,05 0,1 0,15
Productivity
Growth
Within
Between
Cross
Entry
Exit
Swedish Foreign
22
stressed here, which restore incentives and the functioning of markets, or a “macro” story of a depreciating exchange rate
that leads to export-led growth – is difficult.
One way to show that the reforms stressed here were very important in the turnaround of the Swedish economy is to use
cross-country data and proxies for the variables that capture the different reform areas and then correlate them with
economic performance. Unfortunately, cross-country and time-series data covering a broad range of policy indicators are
rare. Some variables only cover a few countries, and others have limited information over time. Some indicators are only
available for recent years, after the major reforms were implemented. Still, the available data give us some basic
information and can illustrate how reforms relate to different outcomes – in our case, most notably GDP per capita.
Figures 10a?10e present basic scatter plots on the relationship between different indicators and log GDP per capita for
OECD countries. Details on the different variables are presented in a web appendix. In Figure 10a, we start by studying the
World Bank’s overall “Ease of Doing Business variable”, defined as the distance to frontier (the best practice country). This
variable is available for the 2009–2014 period and incorporates many elements that are crucial for a market to function
efficiently. It focuses on regulations that affect small and medium-sized businesses. The index measures how difficult – or
costly – starting a business is, for instance, how difficult obtaining permits for electricity or construction is. It also covers,
for example, how costly trading across borders is. Essentially, it gives information on an economy’s day-to-day functioning.
By measuring the stance of the legal and regulatory system, it enables researchers to explore how important the business
environment is for the functioning of the market economy (see Besley, 2015, for an overview of the World Bank’s Doing
Business project). Figure 10a shows a positive relationship between GDP per capita and the Ease of Doing Business
variable.
28
Because the reform package in Sweden aimed at improving market functioning, the evidence from the Doing
Business variable should indicate that reforms affecting micro-economic performance have macroeconomic effects. A
similar pattern is also present when using other overall indexes that capture economic freedom and regulatory practices, for
instance, variables from the Heritage Foundation and the Fraser Institute.
29
These measures include more macroeconomic
types of variables.
Turning to the variables that reflect specific reform areas, we observe similar relationships. As a measure of product
market competition, we use the OECD indicator of PMR, which is available for 1998, 2003, 2008, and 2013. The index
ranges from 0 to 6, where higher values indicate stricter regulations. Figure 10b shows a clear negative relationship between
the regulation index and GDP per capita across OECD countries. In Figure 10c, we observe the relationship between
openness to FDI, as captured by the FDI regulatory restrictiveness index from OECD.
30
The figure indicates a negative
relationship across OECD countries. However, limited variation in openness to FDIs exists among OECD countries. If we
instead study all the available countries, we observe a strong negative relationship between the regulation of inward FDIs
and GDP per capita. As an indicator of the impact of taxes, we use the paying taxes variable from the World Bank’s Doing
Business database. This variable, available for the 2005?2014 period, includes the taxes and mandatory contributions that a
medium-sized company must pay and measures the administrative burden of paying taxes. This indicator is defined as the
distance to frontier, where a higher score indicates a lower tax burden. Figure 10d shows a strong positive relationship
across OECD countries between the tax indicator and GDP per capita. Finally, Figure 10e provides some basic information
on the cross-country relationship between the strictness of employment protection and GDP per capita. We use the OECD
index on the strictness of employment contracts for temporary contracts.
31
From Figure 10e, we observe a weak negative
28
An economy’s distance to frontier is reflected on a scale from0 to 100, where 0 represents the lowest performance and 100 represents the frontier. For example, a score of 75
in a certain year means that an economy was 25 percentage points away fromthe frontier constructed fromthe best performances.
29
These scatter plots are available upon request fromthe authors.
30
This variable covers the seven different years during the 1997–2003 period and ranges from0 to 1, where a higher value indicates less openness.
31
This variable is available for the 1995–2013 period. The data range from0 to 6, where higher values indicate stricter regulations.
23
association between regulation on temporary employment contracts and GDP per capita. Using the World Bank’s rigidity of
employment index (not shown) produces a more negative relationship.
Figure 10a: Correlation between log GDP per
capita and the Ease of Doing Business, 2009–2013
Notes: Scale 0–100; the higher the value, the closer to the frontier,
i.e., the better the business climate. OECD countries only.
Source: Doing Business and WDI (World Bank)
Figure 10c: Correlation between log GDP per
capita and FDI regulatory restrictiveness, 1997–
2013
Notes: Scale 0–1; the higher the value, the stricter regulation of
foreign direct investments. Years available: 1997, 2003, 2006, and
2010–2013. OECD countries only.
Source: OECD and WDI (World Bank)
Figure 10e: Correlation between Logged GDP per
capita and Employment protection legislation of
temporary contracts, 1985–2013
Notes: Scale 0–6; the higher the value, the stricter the regulation of
temporary contracts. OECD countries only.
Source: OECD and WDI (World Bank)
Figure 10b: Correlation between log GDP per
capita and product market regulation, 1998–2013
Notes: Scale 0–6; the higher the value, the more regulated product
market. Years available: 1998, 2003, 2008, and 2013. OECD
countries only.
Source: OECD indicators of product market regulation and WDI
(World Bank)
Figure 10d: Correlation between Logged GDP per
capita and corporate tax burden, 1997–2013
Notes: Scale 0–100; the higher the value, the closer to the frontier,
i.e., the lower the tax burden. Corporate tax burden refers to taxes
and the administrative burden of paying taxes for a medium-sized
company. Years available: 1998, 2003, 2008, and 2013. OECD
countries only.
Source: Doing Business and WDI (World Bank)
Let us now turn to regression analysis. Table 1 presents basic country regressions on variables capturing the reform
areas. Unfortunately, as noted above, most of these variables cover a period after the reforms were implemented in
Sweden and in many other countries. There is also variation in the sample of countries included in the different
variables. With these caveats in mind, the table shows estimates for the variables that are depicted in Figures 10a?10e
above.
32
For each reform variable, we present estimates with and without country fixed effects. All regressions include
year fixed effects. The cross-sectional regressions without country fixed effects give us estimates on how differences
between countries relate to aggregate productivity. The country fixed effects estimates take into account unobserved
country heterogeneity and relate within-country variation in the reform variables to within-country variation in
productivity. One issue with the regressions of the country fixed effects is the limited within-country and over-time
variation in several of the reform variables, which makes obtaining precise estimates difficult.
Columns (1) and (2) show a positive relationship between the overall Ease of Doing Business indicator (measured
as the distance to the top) and log GDP per capita.
33
The estimate in column (1) implies that a one-point increase in the
overall doing business index is associated with a nearly 10 per cent increase in GDP per capita. The estimate drops to
around half a per cent when country fixed effects are taken into account. Quantitatively, the estimate in column (2)
implies that increasing the Ease of Doing Business indicator from the level of Italy in 2014 (68.5) to the corresponding
value for Sweden (80.6) is associated with a productivity increase of approximately five per cent. The corresponding
figures for Greece and Spain are six and three per cent, respectively. If we instead compare Sweden’s value with the
mean of all developing countries, its increase in GDP per capita would be equal to nine per cent.
34
Turning to an indicator of product market competition (the OECD NMR indicator), we see in column (3) that more
regulations are negatively related to GDP per capita.
35
For this variable, the estimate turns statistically insignificant
when adding country fixed effects. If we only analyse the cross-section estimate in column (3), a change in NMR from
the mean position across developing countries to the Swedish one in the final year of the sample (2013) is associated
with a productivity increase of approximately 35 per cent.
Columns (5)?(8) show the results for FDI restrictions and the paying taxes variable. Both of these variables have the
expected signs and are statistically significant in all columns. Fewer restrictions on FDI and a lower tax burden,
including the administrative burden of paying taxes, for medium-sized firms are negatively related to aggregate
productivity. Quantitatively, the estimate in column (6) implies an increase in GDP per capita of approximately two per
cent in Sweden during the 1997–2013 period (when data are available). This calculation is based on a strengthening of
the index for Sweden during this period from 0.079 to 0.059. A similar calculation for the taxation variable implies an
increase in aggregate productivity of approximately 0.7 per cent (based on column 8). In this case, data cover the 2005–
2014 period, and the index for Sweden increased from 80.9 to 83.3. If we also compare Sweden with Spain, Greece, and
the developing country mean for the final year 2014, shifts in the paying taxes variable from the position of these
countries to that of Sweden’s lower level are associated with gains in GDP per capita of approximately two, two, and
five per cent, respectively.
32
We do not use PMR because it is only available for 1998, 2003, 2008 and 2013, and we would like to include a longer period in our analysis. The results for PMR are
qualitatively similar and are available upon request.
33
See Besley (2015) for a summary on previous research using the doing business indicators. See also Djankov et al. (2006) and Eklund (2011) for examples of studies on
the relationship between GDP growth and the Doing Business indicators.
34
These country groupings are based on the UN’s regional coding; seehttp://unstats.un.org/unsd/methods/m49/m49regin.htm.
35
NMR stands for non-manufacturing sector regulations. This index fromthe OECD is available for a long period: 1975–2013. See the web appendix for details.
26
Table 1: Reform variables and productivity; results from firm cross-country regressions, 1996–2009
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Reform
Variable Ease of Doing Business
Product Market
Regulations FDI Index Paying Taxes
Strictness of Empl.
Protection: Temporary
Contracts
0.098*** 0.004*** -1.375*** -0.076 -4.583*** -1.013*** 0.041*** 0.003*** -0.113 0.046***
(0.005) (0.001) (0.129) (0.068) (1.116) (0.377) (0.005) (0.001) (0.074) (0.017)
Country FE No Yes No Yes No Yes No Yes No Yes
Year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Observations
871 871 143 143 348 348 1,556 1,556 801 801
R-squared
0.613 0.397 0.615 0.629 0.194 0.658 0.258 0.411 0.099 0.822
Year
2009–2013 2009–2013 1998–2013 1998–2013 1997–2013 1997–2013 2005–2013 2005–2013 1985–2013 1985–2013
Note: The dependent variable is log GDP per capita (in constant 2005 US dollars). See the web appendix for details on the reformvariables. Standard errors are adjusted
for clustering at the country level. ***, **, and * show significance at the 1 per cent, 5 per cent, and 10 per cent levels, respectively.
6. CONCLUSION
In this paper, we have argued that fundamental reforms in Sweden’s business sector in the 1990s can explain its
remarkable productivity and employment growth after it experienced a significant economic crisis in the early 1990s.
Why were these reforms so successful?
We have argued that they resolved fundamental market and political failures that affected the Swedish economy in
the 1970s and the 1980s. A fundamental market failure in the Swedish business sector during this period was that
incumbent firms and labour unions had gained too much power. This power imbalance enabled them to protect their
markets from competition, creating negative externalities for potential entrants, consumers, and labour market outsiders.
The dominance of incumbents and insiders in labour unions was substantially mitigated by the deregulation of the
labour and product markets.
A fundamental political failure regarding the business sector was that politicians favoured incumbent firms and
insider employees. The corporate tax system and FDI restrictions impeded ownership changes and business formation,
which, in turn, harmed entrepreneurs, labour, and consumers. Moreover, the political system underestimated the cost of
hampering economic incentives for the business sector when pursuing political goals, such as very low unemployment
and highly compressed wages. These political failures were greatly mitigated by the tax reforms, the opening of the
economy to FDI, and the decentralization of wage bargaining.
The process described above suggests that deregulating the business sector can explain part of Sweden’s turnaround.
However, one question remains: over the last decade, why has Sweden not only caught up to the OECD average but
also outperformed nearly all other countries in terms of productivity growth and employment growth in the business
sector?
One possible explanation is that Swedish firms have made more efficient investments in R&D and ICT (see Edquist
and Henrekson, 2015). Investments in intangibles including software, R&D, mineral exploration, copyright and
licensing costs, financial industry development, design, brand equity, vocational training, and organizational structure
are considered very important in Sweden. According to Edquist (2011), intangible investments constituted nearly 10 per
27
cent of GDP and accounted for nearly 30 per cent of labour productivity growth in Sweden’s business sector over the
1995–2006 period. Why then did this occur? First, Swedish society has been open to the adoption of new technologies
and trends. Second, young people were exposed to new ICT early, both in school and at home, which suggests that the
young generation drives the digitalization of the Swedish business sector.
The inflow of FDI to Sweden has also been extraordinary. FDI has created employment and increased productivity
in the target firms, but its indirect dynamic effects also appear to have been important. For example, the Wallenberg
group held controlling positions in companies, which accounted for 42 per cent of the Stockholm Stock Exchange’s
market cap in 1998. By November 2010, their control had declined to 17.1 per cent of the total market cap. Between
1999 and 2009, Investor AB nearly tripled the share of its portfolio, investing in new growth markets and scaling back
its more traditional investments, where it controlled a number of very large firms. Of these growth investments, 62 per
cent went to the Nordic region.
Thus, the deregulation of FDI flows into Sweden appears to have created synergies in the acquired target firms, and
it has generated financial capital for entrepreneurial firms in growth markets. Moreover, the possibility of selling
successful ventures to large foreign incumbents, such as Microsoft, may have been a driver of the vibrant Swedish start-
up markets in new services, computers, and computer and internet games. Skype and Mojang are prominent examples
of such Swedish tech start-ups, which were later sold for astronomical sums. Network effects in these businesses then
create synergies when large foreign incumbents obtain new products from smaller firms. Bidding competitions create
substantial gains for these sellers, who then invest in new projects. This internationalization of the industry appears to
have been particularly large and efficient in Sweden compared with internationalization in other countries.
The Swedish reforms were implemented, and subsequent governments did not reverse them, which is a fundamental
aspect of their success. Many strong groups lost power in the reforms, at least in the short term. One of the arguments
explaining the success of the reforms is the considerable power of bureaucrats and experts in formulating economic
policy in Sweden. Building on a history of trust and respect for knowledge, the political system, industry, and unions
have often been able to reach decisions through consensus on issues of great importance or through the efficiency of the
Swedish economy. The process has also been open to the influence of many different parties, which has generated
broad commitment to the reforms.
We believe that our study of the Swedish experience of industrial reorganization in the 1990s can be a valuable case
study for developing countries that are in search of efficient regulation of their business sectors – and for developed
countries facing economic problems in the aftermath of the financial crisis. Sweden’s experience can serve as an
important example of how an economy undergoing a deep crisis can respond and recover by undertaking economically
sound industrial reforms. Furthermore, by comparing insights from economic theory with an actual restructuring
process, we believe that we can provide valuable knowledge concerning the economic forces driving creative
destruction, which can potentially provide solid ground for policy discussions, particularly those relating to how
countries can improve their competitiveness and employment levels. Our study thus supports the view that addressing
micro-economic inefficiencies is important to provide a solid foundation for a country’s growth and prosperity.
28
REFERENCES
Acs, Z. J . and D. B. Audretsch (2005). ‘Entrepreneurship, innovation and technological change’, Foundations and
Trends in Entrepreneurship, 1(4), 149–195.
Aghion, P., N. Bloom, R. Blundell, R. Griffith and P. Howitt (2005). ‘Competition and innovation: an inverted-u
relationship’, The Quarterly Journal of Economics, 120(2), 701–728.
Allard, G. (2005). ‘Measuring job security over time: in search of a historical indicator for EPL (Employment
Protection Legislation)’, Working Paper WP05-17, Instituto de Empresa, Madrid.
Andrews, D. and F. Cingano (2014). ‘Public policy and resource allocation: evidence from firms in OECD countries’,
Economic Policy, 29(78), 253–296.
Arnold, J .M., B. Brys, C. Heady, A. J ohansson, C. Schwellnus, and L. Vartia (2011). ‘Tax policy for economic
recovery and growth’, Economic Journal, 121(550), F59–F80.
Audretsch, D. B. (1991). ‘New-firm survival and the technological regime’, The Review of Economics and Statistics,
73(3), 441–450.
Auerbach, A. J . (2002). ‘Taxation and corporate financial policy’, in A. J . Auerbach and M. Feldstein (eds.), Handbook
of public economics, 1st edition, 3(19), 1251–1292, Elsevier.
Bandiera, O. I. Barankay och I. Rasul (2007). ‘Incentives for managers and inequality among workers: evidence from a
firm-level experiment’, Quarterly Journal of Economics, 122(2), 729–773.
Bandiera, O. I. Barankay and I. Rasul (2009). ‘Social connections and incentives in the workplace: evidence from
personnel data’, Econometrica, 77(4), 1047–1094.
Barba Navaretti, G. and A.J . Venables (2004). Multinational Firms in the World Economy, Princeton: Princeton
University Press.
Bartelsman, E. J . and M. Doms (2000). ‘Understanding productivity: lessons from longitudinal microdata’, Journal of
Economic Literature, 38(3), 569–594.
Bartelsman, E., S. Scarpetta, and F. Schivardi (2005). ‘Comparative analysis of firm demographics and survival:
evidence from micro-level sources in OECD countries’, Industrial and Corporate Change, 14(3), 365–391.
Bergh, A. (2013). ‘What are the policy lessons from Sweden? On the rise, fall and revival of a capitalist welfare state’,
New Political Economy, Published online: 20 Dec 2013.
Bergh, A. (2014). Sweden and the revival of the capitalist welfare state, Edward Elgar, Northampton, MA.
Bergh, A. and G. Erlingsson (2006). ‘Resilience through restructuring: Swedish policy-making style and the consensus
on liberalisations 1980-2000’, Ratio Working Papers No. 110. Available at RePEc:http://ideas.repec.org/p/hhs/ratioi/0110.html.
Bertrand, M. and A. Schoar (2003), ‘Managing with style: the effect of managers on firm policies’, Quarterly Journal
of Economics, 118(4), 1169–1208.
Besanko, D., D. Dranove, M. Shanley, and S. Schaefer (2003). Economics of strategy, 3rd edition, J ohn Wiley & Sons,
Inc, NY.
Besley, T. (2015). "Law, Regulation, and the Business Climate: The Nature and Influence of the World Bank Doing
Business Project", Journal of Economic Perspectives, 29(3), 99-120.
Bloom, N. and J . Van Reenen (2007). ‘Measuring and explaining management practices across firms and countries’,
Quarterly J ournal of Economics, 122(4), 1351–1408.
29
Calmfors, L. (2012). ‘Sweden: from macroeconomic failure to macroeconomic success’, CESifo working paper: Fiscal
Policy, Macroeconomics and Growth, No. 3790.
Carroll, R., D. Holtz-Eakin, M. Rider and H. S. Rosen (2000). ‘Income taxes and entrepreneurs' use of labor’, Journal
of Labor Economics, 18(2), 324–351.
Carroll, R., D. Holtz-Eakin, M. Rider and H. S. Rosen (2001). ‘Personal income taxes and the growth of small firms, in
J . M. Poterba (ed.), Tax Policy and the Economy, 15, 121–148, MIT Press, Cambridge.
Caves, R. E. (1998). ‘Industrial organization and new findings on the turnover and mobility of firms’, Journal of
Economic Literature, 36(4), 1947–1982.
Coase, R. (1937). ‘The Nature of the Firm’, Economica, 4(16), 386–405.
Da Rin, M., M. Di Giacomo and A. Sembenelli (2011). ‘Entrepreneurship, firm entry, and the taxation of corporate
income: evidence from Europe’, Journal of Public Economics, 95(9), 1048–1066.
Davidson, C., F. Heyman, S. Matusz, F. Sjöholm and S. Chun Zhu (2014). ‘Globalization and imperfect labor market
sorting’, Journal of International Economics, 94(2), 177–194.
Davis, S. J . and M. Henrekson (1999). ‘Explaining National Differences in the Size and Industry Distribution of
Employment’, Small Business Economics, Springer, vol. 12(1), pages 59-83.
Davis, S. J . and M. Henrekson (2000). ‘Wage-setting institutions as industrial policy’, NBER Working Paper No. 7502.
Devereux, M. P., R. Griffith, and A. Klemm (2002). ‘Corporate income tax reforms and international tax competition’,
Economic Policy, 17(35), 449–495.
Djankov, S., C. McLiesh and R. M. Ramalho (2006). ‘Regulation and Growth’, Economic Letters,
doi:10.1016/j.econlet.2006.03.021.
Djankov, S., T. Ganser, C. McLiesh, R. M. Ramalho and A. Schleifer (2010). ‘The effect of corporate taxes on
investment and entrepreneurship’, American Economic Journal: Macroeconomics, 2(3), 31–64.
Edquist C. (2011). ‘Innovation policy design: identification of systematic problems’, CIRCLE Electronic Working
Papers No. 2011/6.
Edquist, H. and M. Henrekson (2013). ‘Product market reforms and incentives to innovate in Sweden’, in M. Choupres
and H. Edquist (eds.), What can we learn from economic reforms in Greece and Sweden?, European Liberal Forum,
Brussels.
Edquist, H. and M. Henrekson (2015). ‘Swedish Lessons: How Important are ICT and R&D to Economic Growth?,’
Working Paper Series 1073, Research Institute of Industrial Economics.
Eklund, J . (2011). ‘Regelbördans ekonomiska effekter’, Tillväxtanalys.
Ericson, R. and A. Pakes (1995). ‘Markov-perfect industry dynamics: a framework for empirical work’, The Review of
Economic Studies, 62(1), 53–82.
Faggio, G., K. G. Salvanes and J . Van Reenen (2010). ‘The evolution of inequality in productivity and wages: panel
data evidence’, Industrial and Corporate Change, 19(6), 1919–1951.
Foster, L., J . Haltiwanger and C. J . Krizan (2001). ‘Aggregate productivity growth. Lessons from microeconomic
evidence’, in C. R. Hulten, E. R. Dean, and M. J . Harper (eds.), New developments in productivity analysis,’
University of Chicago Press, Chicago.
Fox, J . T. and V. Smeets (2011). ‘Does input quality drive measured differences in firm productivity?’, NBER Working
Paper No. 16853.
30
Gemmel, N., R. Kneller, D. McGowan and I. Sanz (2013). ‘Corporate taxation and productivity catch-up: evidence
from European firms’, Working Papers No. 13001, Bangor Business School, Prifysgol Bangor University (Cymru /
Wales).
Gentry W. M. and G. Hubbard (2000). ‘Entrepreneurship and houshold saving’, NBER Working Papers No. 7894.
Gibbons, R. and J . Roberts (eds.) (2013). The handbook of organizational economics, Princeton University Press,
Princeton.
Gordon, R. H. and J . J r. Hines (2002). ‘International taxation’, in A. J . Auerbach and M. Feldstein (eds.), Handbook of
public economics, 1st edition, 4(28), 1935–1995, Elsevier.
Gibrat, R. (1931). Les Inégalités économiques, Librairie du Recueil Sirey, Paris, France.
Grossman, S. and O. Hart (1986). ‘The costs and the benefits of ownership: a theory of vertical and lateral integration’,
Journal of Political Economy, 94(4), 691–719.
Grossman, G. M. and E. Helpman (1994). ‘Protection for sale’, American Economic Review, 84(4), 833–850.
Hakkala, K., F. Heyman and F. Sjöholm (2014). ‘Multinational firms and job tasks’, European Economic Review, 66,
248–265.
Hall, B. and J . Van Reenen (2000). ‘How effective are fiscal incentives for R&D? A review of the evidence’, Research
Policy, 29(4), 449–469.
Hart, O. and J . Moore (1990). ‘Property rights and the nature of the firm’, The Journal of Political Economy, 98(6),
1119–1158.
Hassett, K. A. and R. G. Hubbard (2002). ‘Tax policy and business investment’, in: A. J . Auerbach and M. Feldstein
(eds.), Handbook of public economics, 1st edition, 3(20), 1293–1343, Elsevier.
Haufler, A., P.-J . Norbäck and L. Persson (2014). ‘Entrepreneurial innovations and taxation’, Journal of Public
Economics, 113(May), 13–31.
Henrekson, M. and U. J akobsson (2005). ‘The Swedish model of corporate ownership and control in transition’, in
Huizinga, H. and L. J onung (eds.), The internationalization of asset ownership in Europe, Cambridge University
Press, Cambridge.
Henrekson, M., J ohansson, D. and M. Stenkula (2012). "Den svenska företagsstrukturen – utvecklingen i de medelstora
företagen efter 1990-talskrisen" Ekonomisk Debatt, 40(2), 27–38.
Heyman, F. P.-J . Norbäck and L. Persson (2015). ‘ The Turnaround of Swedish Industry: Reforms, Firm Diversity and
J ob and Productivity Dynamics‘, IFN Working Paper No. 1079.
Hicks, A. and L. Kenworthy (1998). ‘Cooperation and Political Economic Performance in Affluent Democratic
Capitalism‘, American Journal of Sociology 103, 1631–1672.
Hines, J . J r. (2005). ‘Do tax havens flourish?’, NBER Working Papers No. 10936.
Hjalmarsson, L. (1974). ‘The size distribution of establishments and firms derived from an optimal process of capacity
expansion’, European Economic Review, 5(2), 123–140.
Hopenhayn, H. (1992). ‘Entry, exit, and firm dynamics in long run equilibrium’, Econometrica, 60(5), 1127–1150.
Horn, H. and L. Persson (2001). ‘The equilibrium ownership of an international oligopoly’, Journal of International
Economics, 53(2), 307–333.
Ilmakunnas, P., M. Maliranta and J . Vainiomäki (2004). ‘The roles of employer and employee Ccaracteristics for plant
productivity’, Journal of Productivity Analysis, 21(3), 249–276.
J avorcik, B. S. (2015). ‘Does FDI Bring Good J obs to Host Countries?’, World Bank Research Observer, 30(1), 74–94.
31
J onung, L., J . Kiander and P. Vartia (2008). ‘The great financial crisis in Finland and Sweden – the dynamics of boom,
bust and recovery, 1985–2000’, European Economy, Economic Papers 367, European Commission, Brussels.
J ovanovic, B. (1982). ‘Selection and the evolution of industry’, Econometrica, 50(3), 649–670.
Keuschnigg, C. and S. B. Nielsen (2004a). ‘Taxation and venture capital backed entrepreneurship’, International Tax
and Public Finance, 11(4), 369–390.
Keuschnigg, C. and S. B. Nielsen (2004b). ‘Start-ups, venture capitalists, and the capital gains tax’, Journal of Public
Economics, 88(5) 1011–1042.
King, M. and D. Fullerton (1984). The taxation of income from capital: a comparative study of the United States, the
United Kingdom, Sweden and West Germany, The Chicago University Press, Chicago.
Klepper, S. (1996). ‘Entry, exit, growth, and innovation over the product life cycle’, American Economic Review, 86(3),
562–583.
Kneller, R. and D. McGowan (2012). ‘Tax policy and firm entry and exit dynamics: evidence from OECD countries’,
Working Papers No. 12006, Bangor Business School, Prifysgol Bangor University (Cymru / Wales).
Lazear, E. P. (2000). ‘Performance pay and productivity’, American Economic Review, 90(5), 1346–1361.
Li, Y. and M. Rama (2015). ‘Firm Dynamics, Productivity Growth, and J ob Creation in Developing Countries: The
Role of Micro- and Small Enterprises’, World Bank Research Observer, 30(1), 3–38.
Lindbeck, A. (1997). The Swedish experiment, J ournal of Economic Literature, Vol. XXXV , 1273-1319.
Lindbeck, A., P. Molander, T. Persson, O. Sandmo, B. Swedenborg and N. Thygesen (1994), Turning Sweden Around.
MIT Press: Cambridge, MA.
Lindbeck, A. and D. J . Snower (2001). ‘Insiders versus Outsiders’, J ournal of Economic Perspective, Vol. 15, NO 1,
165-188.
Luttmer, E. G. J . (2007). ‘Selection, growth, and the size distribution of firms’, The Quarterly Journal of Economics,
122(3), 1103–1144.
Maksimovic V. and G. Phillips (2001). ‘The market for corporate assets: who engages in mergers and asset sales, and
are there efficiency gains?’, Journal of Finance, 56(6), 2019–2065.
Mirrlees, J ., S. Adam, T. Besley, R. Blundell, S. Bond, R. Chote, M. Gammie, P. J ohnson, G. Myles, and J . Poterba
(2011) Tax by design: the Mirrlees review, Oxford University Press for Institute for Fiscal Studies, Oxford.
Motta, M. (2004). Competition policy: theory and practice, Cambridge University Press, Cambridge.
Murphy, K. M. and R. H. Topel (1990). ‘Efficiency Wages Reconsidered: Theory and Evidence’, in Y. Weiss and G.
Fishelson (eds.), Advances in the theory and measurement of unemployment, Palgrave Macmillan, London.
Nelson, R. R. and S. G. Winter (1982). An evolutionary theory of economic change, Belknap Press, Cambridge, MA.
Norbäck, P.-J . and L. Persson (2007). ‘Investment liberalizations:wWhy a restrictive cross-border merger policy can be
counterproductive’, Journal of International Economics, 72(2), 366–380.
Norbäck, P.-J . and L. Persson (2012). ‘Entrepreneurial innovations, competition and competition policy’, European
Economic Review, 56(6), 488–506.
Norbäck, P.-J ., L. Persson and R. Douhan (2014). ‘Entrepreneurship policy and globalization’, Journal of Development
Economics, 110(September), 22–38.
Nycander, S. (2008). Makten över arbetsmarknaden: ett perspektiv på Sveriges 1900-tal, SNS Publishing, Stockholm.
OECD (2013). OECD compendium of productivity indicators 2013, OECD Publishing.
Olley, S. and A. Pakes (1996). ‘The dynamics of productivity in the telecommunications industry’, Econometrica,
64(6), 1263–1298.
32
Olson, M. (1965). The logic of collective action, Harvard University Press, Harvard.
Perotti, E. and P. Volpin (2007). ‘Investor protection and entry’, Tinbergen Institute Discussion Papers 07-006/2.
Rajan, R. G. and L. Zingales (2001). ‘The firm as a dedicated hierarchy: a theory of the origins and growth of firms’,
The Quarterly Journal of Economics, 116(3), 805–851.
Saggi, K. (2002). ‘Trade, Foreign Direct Investment, and International Technology Transfer: A Survey’, World Bank
Research Observer, 17(2), 191–235.
Santarelli, E. and M. Vivarelli (2007). ‘Entrepreneurship and the process of firms’ entry, survival and growth’, Industry
and Corporate Change, 16(3), 455–488.
Skedinger, P. (2010). ‘Employment protection legislation. Evolution, effects, winners and losers’, Edward Elgar,
Northampton, MA.
SOU 1986:23 (1986). Aktiers röstvärde, Report by the Voting Rights Committee, Liber, Stockholm.
SOU 1993:16 (1993). Nya villkor för ekonomi och politik, Report from the Economics Commison, Allmänna förlaget,
Stockholm.
SOU 2005:4 (2005). Liberalisering, regler och marknader, Ministry for Enterprise, Stockholm.
Spencer, B. J . and J . A. Brander (1983). ‘International R&D rivalry and industrial strategy’, Review of Economic
Studies, 50(4), 707–722.
Stenkula, M., D. J ohansson and G. Du Rietz (2015). ‘Capital income taxation of Swedish households, 1862 to 2010’,
Scandinavian Economic History Review, 63(2), 154–177.
Stigler, G. J . (1971). ‘The theory of economic regulation’, The Bell Journal of Economics and Management Science,
2(1), 3–21.
Sutton, J . (1997). ‘Gibrat’s legacy’, Journal of Economic Literature, 35(1), 40–59.
Syverson, C. (2011). ‘What determines productivity?’, Journal of Economic Literature 49(2), 326–365.
Thimann, C. (2015). "The Microeconomic Dimensions of the Eurozone Crisis and Why European Politics Cannot Solve
Them", Journal of Economic Perspectives, 29(3), 141-64.
Tirole, J . (2006). The theory of corporate finance, Princeton University Press, Princeton.
Van Ark, B., M. O’Mahony, and M. P. Timmer (2008). ‘The productivity gap between Europe and the United States:
trends and causes’, Journal of Economic Perspectives, 22(1), 25–44.
Vives, X. (2008). ‘Innovation and competitive pressure’, The Journal of Industrial Economics, 56(3), 419–469.
Williamson, O. E. (1979). ‘Transaction-cost economics: the governance of contractual relations’, Journal of Law and
Economics, 22(2), 233–261.
World Bank (2014). ‘Doing business 2015: going beyond efficiency’, A World Bank Group Flagship Report, World
Bank, Washington, DC.
Wölfl, A., I. Wanner, T. Kozluk, and G. Nicoletti (2009). ‘Ten years of product market reform in OECD countries:
insights from a revised PMR indicator’, OECD Economics Department Working Papers No. 695
doc_291818468.pdf