Extent of undeclared work in the European Union
Colin Williams2023, European Labour Authority
visibility
…
description
48 pages
link
1 file
This report is an update of the 2017 European Platform tackling undeclared work report, An evaluation of the scale of undeclared work in the European Union and its structural determinants: Estimates using the Labour Input Method (Williams et al., 2017), that provided estimates of undeclared work in the private sector for 2013. Employing the same methodology, this study provides estimates on the prevalence of undeclared work in the European Union (EU) in 2019. The finding is that in 2019, 11.1 % of total labour input in the private sector in the EU is undeclared (11.6 % in 2013), and undeclared work accounts for 14.8 % of gross value added (GVA) in the private sector (16.4 % in 2013). However, these are unweighted averages, not considering the relative size of the labour force in each Member State. The weighted averages are that 9.7 % of total labour input in the private sector in the EU is undeclared (10.2 % in 2013), and undeclared work constitutes 14.6 % of GVA in the private sector (14.9 % in 2013). Therefore, between 2013 and 2019, there has been a decline in undeclared work at the EU level and a decline in 19 of the 26 Member States (estimates could not be produced for Malta in 2013), the exceptions being Bulgaria, France, Germany, Italy, Lithuania, the Netherlands and Romania. This displays the progress made in tackling undeclared work both in the EU as a whole and in most Member States.
Sign up for access to the world's latest research.
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Related papers
An evaluation of the scale of undeclared work in the European Union and its structural determinants: estimates using the Labour Input Method
On average, 11.6% of total labour input in the private sector in the EU is undeclared, and undeclared work constitutes on average 16.4% of gross value added (GVA) (the difference due to undeclared labour being concentrated in sectors with higher labour productivity). These, however, are unweighted averages, and do not take into account the relative size of the labour force in each Member State. The weighted averages, therefore, are that 9.3% of total labour input in the private sector in the EU is undeclared, and undeclared work constitutes 14.3% of GVA in the private sector. The reason for the weighted average being lower than the unweighted average is due to the influence of larger countries such as Germany, France and the UK, which have larger labour forces and relatively lower levels of undeclared work. There are significant differences in the size of undeclared work between Member States: For undeclared work as a proportion of total labour input – Poland, Romania and Lithuania have the highest levels of undeclared work, and countries exceeding the EU average are largely new EU Member States (NMS). Only the Czech Republic from the NMS has a smaller than EU average undeclared economy, with the lowest share of undeclared work in terms of labour input being found in the UK, Germany and the Netherlands (less than 3% of total labour input being undeclared). For undeclared work as a proportion of GVA – the distribution is similar to above, with undeclared work as a proportion of GVA being highest in Poland, Romania and Lithuania (all with undeclared work being over 25% of total GVA created in the private sector), and those with undeclared economies above the EU average again being mostly new Member States (Hungary, Latvia, Estonia, Bulgaria, Cyprus, Croatia and Czech Republic) with three older EU members: Greece, Spain and Italy. Only Slovakia and Slovenia from the NMS countries have undeclared economies slightly below the EU average. There are significant differences in the prevalence of undeclared work amongst different groups of the employed: The proportion of self-employment which is undeclared by Member State ranges from 77.6% of all self-employment in Latvia, 73.2% in Romania and 70.6% in Cyprus, to just 6.5% in Belgium, 5.3% in Bulgaria, 3.4% in Italy and 2.5% in Poland; The proportion of work conducted in the context of an employment relationship which is undeclared ranges from 25.3% Poland, 19.3% in Bulgaria and 18.5% in Lithuania, to just 2.1% in Portugal, 1.5% in Germany, 1.4% in United Kingdom and 1.1% in the Netherlands; and The proportion of all family work which is undeclared ranges from 89.8% in Latvia, 69.4% in Estonia and 68.2% in Romania, to 4.2% in Austria, 2.9% in Italy and 2.1% in Sweden. Examining the structure of the undeclared labour market in the EU, 61.8% of all undeclared work is work conducted in the context of an employment relationship, 37.3% is self-employment and 0.3% is family work. This masks considerable national variations – countries where the majority of undeclared work is conducted through self-employment include Cyprus, Netherlands, Portugal, Denmark and Germany; while those where over 90% of all undeclared work is conducted in the context of an employment relationship include Poland, Bulgaria and Italy. These different structures have significant implications for tackling undeclared work. Policy initiatives to help business start-up on a legitimate basis, such as smoothing the transition from unemployment to self-employment, will be useful in countries where most undeclared work is conducted as self-employment (e.g. Denmark). Conversely, in countries where most undeclared work is conducted in the context of an employment relationship (e.g. Poland), policy initiatives to address unregistered or under-declared waged employment, such as the use of notification letters to employers to change behaviour, will be most relevant. It is not only the development of tailored policy measures, however, that Member States need to pursue to tackle undeclared work. Evaluating whether cross-national variations in the size of undeclared work are associated with variations in various structural conditions, there is found to be: A strong significant relationship with GDP per capita in purchasing power standards (the greater the level of GDP in PPS, the lower the prevalence of undeclared work); and the quality of government (based on the European Quality of Government Index, where the higher the quality of government the lower the prevalence of undeclared work); A moderate significant relationship with: the Transparency International Corruption Perceptions index (which ranks countries according to perceived public sector corruption); a trust in authorities index based on World Economic Forum indicators; the impact of social transfers on poverty reduction; public expenditure on labour market interventions to protect vulnerable groups; the migration rate; the Gini coefficient and income inequality. Overall, the higher the perceived level of corruption, the lower the level of trust in public authorities, and the higher the perceived level of inequality, the higher the level of undeclared work; and A weak but significant relationship with the long-term unemployment rate; and with the very long-term unemployment rate – in both cases the higher the respective rates, the higher the level of undeclared work. No significant relationship was identified between undeclared work and the job vacancy rate or the implicit tax rate on labour – in the case of the latter, this refutes the common assumption that undeclared work is directly related to taxation levels. In conclusion, lower levels of undeclared work are found in Member States where there are higher levels of GDP per capita, more modernised systems of government, higher levels of trust in authorities and lower levels of corruption, where social transfers are effective at reducing poverty, there are higher levels of public expenditure on labour market interventions to protect vulnerable groups, and where there is greater equality, lower levels of long-term unemployment, and net in-migration rather than out-migration.
Explaining the Variations in the Magnitude of Undeclared Work Across the 28 European Union Member States
Economic Alternatives, 2020
This paper evaluates the competing theories that seek to explain the greater magnitude of undeclared work in some nations than others. These theories variously explain the higher levels of undeclared work in some nations to be determined by either economic under-development and the lack of modern governance (“modernisation” theory), higher taxes and state over-interference (“neo-liberal” theory) or inadequate government intervention in work and welfare (“political economy” theory). Reporting data on the magnitude of undeclared work in the 28 member states of the European Union using the labour input method, the finding is that undeclared work is higher in EU member states with lower levels of GDP per capita, less modernised systems of government, higher levels of corruption, social transfers are less effective at reducing poverty, and there are lower levels of public expenditure on labour market interventions to protect vulnerable groups. The theoretical and policy implications are then discussed.
Determinants of Undeclared Work in the EU Member States
Procedia Economics and Finance, 2016
In the context of European economic recovery, undeclared work has become a true challenge for the labour market policies, mainly because it affects tax revenue, social security and labour standards. Without better understanding of the main determinants of undeclared work, little can be done to prevent it. Having these in mind and based on the Eurobarometer survey, the paper aims to find the main factors that trigger undeclared work. Our findings suggest that after the economic crisis, the trust in national authorities has dropped and generated more undeclared payments, especially for the case of small and medium firms.
Tackling undeclared work in Europe: lessons from a 27-nation survey
2009
The aim of this article is to unravel the heterogeneous nature of undeclared work across the European Union and to evaluate the consequences for tackling such work. Until now, most studies of undeclared work have sought to measure the variations in its magnitude. Far fewer have evaluated the nature of undeclared work and whether it is configured in different ways across nations.
Evaluating the variations in undeclared work in the European Union
Drawing upon the results of 26,659 face-to-face interviews in the 27 member states of the European Union (EU-27), the aim of this paper is to evaluate its varying magnitude and whether such work is mostly wage-based (of an envelope wage or undeclared variety) or own-account work (of a market-oriented or solidarity-oriented variety). The finding is that in 'welfare capitalist' work and welfare regimes, which are more equal (measured by the gini coefficient) pursue more extensive labour market interventions, have higher levels of social protection and greater state redistribution, undeclared economies are smaller and are composed mostly of solidarity-oriented own-account work. Meanwhile, in 'neo-liberal' regimes which are more unequal, with less labour market intervention, social protection and redistribution via social transfers, undeclared economies are larger and are composed largely of envelope wage arrangements. The paper concludes by discussing the theoretical and policy implications.
Explaining cross-national variations in the prevalence and character of undeclared work in the European Union
The aim of this article is to evaluate the competing theories that variously explain the greater prevalence of undeclared employment in some countries either as: a legacy of under-development ; a result of the voluntary exit from declared employment due to the high taxes, state corruption and burdensome regulations and controls, or a product of a lack of state intervention in work and welfare which leads to the exclusion of workers from the declared economy and state welfare provision. Analyzing the cross-national variations in the prevalence of, and reasons for, undeclared employment across the European Union using evidence from a 2007 Eurobarometer survey, the finding is that undeclared employment is less prevalent and more of the voluntary variety in wealthier , less corrupt and more equal societies possessing higher levels of social protection and redistribution via social transfers. The theoretical and policy implications are then discussed.
Face Recognition System using OpenCV
IRJET, 2023
The face recognition system using OpenCV is a system which recognizes know faces and store it with time stamp. Then recognize the unknown faces and store it in another dataset with time stamp. The versatile face recognition system utilizing OpenCV, designed to streamline attendance tracking in online classes, meetings, and residential security applications. Leveraging advanced image processing techniques, the system identifies and verifies participants' faces in real-time. Its adaptable nature allows easy customization of recognized individuals, facilitating efficient management of attendance lists. This technology proves beneficial for remote learning scenarios, virtual meetings, and enhancing security protocols in residential settings. By seamlessly integrating with existing platforms, the system offers a user-friendly and efficient solution to automate attendance management, thereby improving overall convenience and security in various contexts.
BAO CAO THỰC TẬP
Tên tiếng Anh: ĐAI TIN BUSSINESS AND DEVELOPMENT REAL ESTATE JOINT STOCK COMPANY Ngày thành lập: 11/10/2015 Trụ sở chính: Lô P9-10, Đường D9M4, Khu Phố Thương Mại RICH TOWN, Phường Hòa Phú,TP Thủ Dầu Một, Bình Dương. Vốn điều lệ: 20.000.000.000 đồng Điện thoại: 02838.99.99.89 2.Thực trang Tổ chức bộ máy 2.1. Sơ đồ bộ máy tổ chức: Hình 1.1.1: Sơ đồ bộ máy tổ chức Nguồn: Tińh đêń ngaỳ 15/12/2017 Công ty Cp đầu tư và phát triển Bất Động Sản Đại Tín co: Tổng số công chức viên chức và cań bộ hơ ̣p đồng là 108 người. Trong đó nam: 66 người (chiêḿ 61.1 %); nữ 32 người( chiêḿ 38.9%) -Trình độ được đaò ta ̣o và cań bô ̣ giữ chức vu ̣ lañh đa ̣o + Cań bô ̣ có triǹh đô ̣ đại ho ̣c: 69 (chiêḿ 63.9%) BAN GIÁM ĐỐC PHÒNG KINH DOANH PHÒNG PHÁP LÝ PHÒNG TỔ CHỨC PHÒNG TÀI CHÍNH -Trưởng Phòng Kinh Doanh -Phó phòng Kinh Doanh -Nhân viên kinh doanh -Chuyên viên pháp lý -Chuyên viên Marketing -Chuyên viên tuyển dụng -Chuyên viên tiền lương -Kế toán trưởng -Kế toán hành chính GVHD-Th.s Từ Vân Anh Báo cáo thực tập SVTT-Trần Phương Nguyên + Phòng tổ chức: có chức năng tham mưu tổ chức sắp xếp bộ máy quản lý, CBCNV lao động trong toàn công ty. Tổ chức tuyển dụng, sắp xếp bố trí công việc cho nhân viên, tổ chức luân chuyển, sa thải, nâng chức vụ... Quản lý thời gian cán bộ CNVLD trực tiếp và gián tiếp công ty
Extent of
undeclared work in
the European Union
February 2023
2021
LEGAL NOTICE
© European Labour Authority, 2023
Reproduction is authorised provided the source is acknowledged.
For any use or reproduction of photos or other material that is not under the copyright of the European Labour
Authority, permission must be sought directly from the copyright holders.
Neither the European Labour Authority nor any person acting on behalf of the European Labour Authority is
responsible for the use which might be made of the following information.
The present document has been produced by Josip Franic (Institute of Public Finance, Croatia), Professor Ioana
Alexandra Horodnic (Alexandru Ioan Cuza University of Iasi, Romania) and Colin C Williams (University of Sheffield,
UK) and it has been financed by the European Labour Authority. The document has been prepared for the European
Labour Authority; however, it reflects the views of the authors only. The information contained in this report does
not reflect the views or the official position of the European Labour Authority or the European Platform tackling
undeclared work.
2
Contents
1.0 Executive summary .......................................................................................................1
2.0 Introduction ....................................................................................................................4
3.0 Methodology used for estimating the scale of undeclared work: the Labour
Input Method ..................................................................................................................5
3.1 Operational definition of undeclared work ............................................................................................. 5
3.2 The Labour Input Method ...................................................................................................................... 6
3.3 Data and variables used ........................................................................................................................ 8
3.3.1 Overview of the datasets and key definitions .............................................................................. 8
3.3.2 Comparable data on covered industries.................................................................................... 10
3.4 Analytical approach ............................................................................................................................. 10
3.4.1 Labour input from LFS ............................................................................................................... 10
3.4.2 Reconciliation of LFS with SBS on labour input ........................................................................ 12
3.5 Deficiencies of data sources and suggestions for improvements ....................................................... 13
4.0
Estimates of undeclared work in the EU27 using the Labour Input Method...............14
4.1 The size of undeclared work in the EU ................................................................................................ 15
4.2 Undeclared work by type of employment relationship ......................................................................... 21
4.3 Structure of the undeclared labour market .......................................................................................... 23
5.0 Correlations between undeclared work and structural conditions .........................24
5.1 Structural determinants of undeclared work in the EU ........................................................................ 24
5.2 Government ineffectiveness and public sector corruption ................................................................... 27
5.3 Level of development and state intervention in work and welfare ....................................................... 29
5.4 Quality and power of formal institutions ............................................................................................... 30
5.5 Instability and uncertainty of formal institutions ................................................................................... 33
5.6 Lack of adherence to the formal rules and trust in peers .................................................................... 34
References ............................................................................................................................37
Annex 1: Data sources of the structural conditions and results .....................................38
1.0 Executive summary
This report is an update of the 2017 European Platform tackling undeclared work report, An evaluation of
the scale of undeclared work in the European Union and its structural determinants: Estimates using the
Labour Input Method (Williams et al., 2017), that provided estimates of undeclared work in the private
sector for 2013. Employing the same methodology, this study provides estimates on the prevalence of
undeclared work in the European Union (EU) in 2019.
This report provides only estimates, and highly conservative ones, of the real size of the undeclared
economy which need to be treated with caution and not as indisputable facts. Caution is required because
the Labour Input Method (LIM) primarily focuses on detecting undeclared work attributable to the labour
input. However, the multifaceted and complex realm of undeclared work embraces many other
transactions that do not necessarily include undeclared labour relations such as unregistered employment
and envelope wages (e.g., businesses not declaring all their transactions in reported turnover). In general,
economic output comes from both labour and capital, with only the former being fully captured by LIM.
This method is hence more effective in estimating undeclared work in labour-intensive sectors where
capital inputs play a less important role. Contrariwise, many undeclared transactions in capital-intensive
sectors most likely remain undetected by LIM. There are also data limitations, with for example agriculture,
forestry and fishing excluded from the scope of this study, and whose inclusion (if the data was available)
would doubtless lead to higher estimates of undeclared work due to the likely higher prevalence of
undeclared work in this sector.
The finding is that in 2019, 11.1 % of total labour input in the private sector in the EU is undeclared
(11.6 % in 2013), and undeclared work accounts for 14.8 % of gross value added (GVA) in the private
sector (16.4 % in 2013). However, these are unweighted averages, not considering the relative size of
the labour force in each Member State. The weighted averages are that 9.7 % of total labour input in
the private sector in the EU is undeclared (10.2 % in 2013), and undeclared work constitutes 14.6
% of GVA in the private sector (14.9 % in 2013).
Therefore, between 2013 and 2019, there has been a decline in undeclared work at the EU level and
a decline in 19 of the 26 Member States (estimates could not be produced for Malta in 2013), the
exceptions being Bulgaria, France, Germany, Italy, Lithuania, the Netherlands and Romania. This displays
the progress made in tackling undeclared work both in the EU as a whole and in most Member States.
Nevertheless, there remain substantial differences in the level of undeclared work across the EU:
For undeclared work as a proportion of total labour input in the private sector, undeclared work
is most pervasive in Romania (21.7 %), Lithuania (20.8 %), and Bulgaria (19.3 %). The lowest levels
are found in Germany (3.9 %), the Netherlands (4.8 %), and Austria (5.1 %). On the whole, countries
above the EU average are mostly Member States that joined the EU in or after 2004.
For undeclared work as a proportion of GVA in the private sector, the distribution is similar, with
undeclared work as a proportion of GVA being again highest in Romania (27.1 %), Lithuania (26.0 %),
and Bulgaria (23.8 %). The lowest levels are found in Austria (5.3 %), Luxembourg (7.0 %), and
Sweden (7.5 %).
There are also significant differences in the pervasiveness of undeclared work in various types of the
employment relationship:
1
The proportion of self-employment that is undeclared ranges from 3.6 % in Belgium to 65.2 % in
Cyprus.
The share of waged employment that is undeclared ranges from just 0.5 % in the Netherlands to 20.2
% in Bulgaria.
The proportion of family work that is undeclared is highest in Luxembourg (68.9 %), Malta (63.0 %),
Cyprus (62.2 %), and Romania (60.1 %) and lowest in Sweden (2.1 %), Italy (2.4 %), Czechia (2.6 %),
and Slovakia (3.3 %).
Analysing the structure of the undeclared labour market in the EU, 62.9 % (61.8 % in 2013) of all
undeclared work in the EU is waged employment, 36.3 % (37.3 % in 2013) is self-employment and
0.8 % (0.9 % in 2013) is family work. Nonetheless, there again exist substantial differences between the
Member States. Countries in which the majority of undeclared work is self-employment include the
Netherlands (90 %), Cyprus (82.2 %), Ireland (71.4 %), and Finland (68.1 %). Meanwhile, in Poland,
Belgium, Bulgaria, and Italy over 90 % of all undeclared work is waged employment.
These differences in the structure of the undeclared workforce have important implications for tackling
this practice. Countries where undeclared work is predominantly self-employment may find it more
beneficial to prioritise policy initiatives supporting enterprise to start-up legitimately, such as smoothing
the transition from unemployment to self-employment, and making it easier and beneficial for enterprises
to move towards full legitimacy. On the contrary, countries where undeclared work is mostly waged
employment may find it more beneficial to focus on policy measures making unregistered and underdeclared employment more expensive and/or less appealing.
Evaluating how these 2019 cross-national variations in the magnitude of undeclared work are associated
with differences in various structural conditions, there is found to be1:
A strong significant negative relationship (i.e., the higher the value of the indicator, the lower the
prevalence of undeclared work) with the Social Progress Index, Democracy Index, government
effectiveness, GDP per capita in current prices (euro per capita), Corruption Perceptions Index (high
values of this indicator indicated that the public sector is perceived as very clean), Human Development
Index, rule of law, control of corruption, regulatory quality, voice and accountability, social capital,
judicial independence and reliability of police services and a strong significant positive relationship
(.e., the higher the value of the indicator, the higher the prevalence of undeclared work) with severe
material deprivation rate.
A moderate negative significant relationship with Research & Development expenditure, the
perceived share of undeclared work (higher value of this indicator indicates a high share of citizens
estimates a low share of undeclared work in their society), labour market policy expenditure, trust in
parliament, workers’ rights, the impact of social transfers on poverty reduction, acceptability of
undeclared work by a firm for another firm (higher value of the indicator indicates a high share of
citizens finding this form of undeclared work unacceptable) and trust in government and a strong
positive significant relationship with labour productivity and people at risk of poverty/social
exclusion.
1
The strength of the correlation/ relationship is assessed based on the Spearman`s rho coefficient (the descriptors
apply to both positive and negative relationships): 0.81 to 1 very strong relationship, 0.61 to 0.80 – strong relationship,
0.41 to 0.60 – moderate relationship, 0.21 to 0.40 – weak relationship, 0.01 to 0.20 – very weak or negligible
relationship. The significance of the relationship is given by the p value and it is significant at: *** p<0.01, ** p<0.05,
* p<0.1.
2
A weak negative but significant relationship with the acceptability of undeclared work by firms for
private households (higher value of the indicator indicates a high share of citizens finding this form of
undeclared work unacceptable), political stability, trust in labour inspectorate, the burden of
government regulation, trust in tax and social security authorities, acceptability of someone partially
or completely concealing income and tax compliance, attitudes towards firms hiring workers on an
undeclared basis (higher value of the indicator indicates a high share of citizens finding this form of
undeclared work unacceptable), and a weak positive but significant relationship with inequality of
income distribution and Gini coefficient distribution.
A very weak or insignificant negative relationship with business flexibility index, tax revenue as %
of GDP, social contributions as % of revenue, personally knowing people engaged in undeclared work,
and attitudes towards undeclared work by individuals for private households and a very weak or
insignificant positive relationship with the expense of government.
In conclusion, lower levels of undeclared work are found in Member States with:
i.
Higher government effectiveness and lower perceived levels of corruption.
ii.
Higher levels of development (whether measured in terms of GDP per capita, the Human
Development Index or Social Progress Index) and greater levels of state intervention in work
and welfare (to enhance workers` rights and labour productivity, investments in research and
development and implementing measures for reducing poverty and inequalities).
iii.
Higher quality more powerful formal institutions (higher reliability of policy services, higher
judicial independence, stronger rule of law, stronger voice and accountability, positive perceptions
towards the regulatory quality and higher trust in state institutions).
iv.
Lower levels of instability and uncertainty in formal institutions (better transparency in
policymaking and reduced perception of political instability).
v.
Greater symmetry between the norms, values and beliefs of citizens, workers, employers
and businesses and the formal rules (termed “vertical trust” and measured by the level of tax
compliance, and the acceptability of undeclared work) and a higher trust in peers to adhere to
the formal rules (termed “horizontal trust” and measured by personally knowing people engaged
in undeclared work and their estimates of the share of undeclared work).
3
2.0 Introduction
This report is an update of the 2017 European Platform tackling undeclared work report, An evaluation of
the scale of undeclared work in the European Union and its structural determinants: Estimates using the
Labour Input Method (Williams et al., 2017), that provided 2013 estimates of undeclared work in the private
sector. Using the same methodology, the aim is to update this report by providing figures on the
prevalence of undeclared work in 2019 for all EU Member States.
Given that undeclared work is hidden from and unreported to the authorities, estimating its magnitude is
a challenging task. As the European Commission (2007: 4) explains:
“
“Undeclared work can be measured both directly and indirectly. Indirect methods are
based on the comparison of macroeconomic aggregates (such as national accounts,
electricity consumption, cash transactions). Indirect (especially monetary) methods often
over-estimate the level of undeclared work and say little about its socio-economic
characteristics. Direct methods, on the contrary, are based on statistical surveys and
have advantages in terms of comparability and detail, but tend to under-report the extent
of undeclared work.”
There is a consensus among both practitioners and academics that a direct approach should be used
more to explore the nature of undeclared work in terms of who engages in such work, why they do so,
which forms of undeclared work they undertake and their reasons for doing so (Eurofound, 2013; Williams
and Schneider, 2016; Williams and Horodnic, 2021). When estimating the magnitude of undeclared work,
the consensus is supportive of the use of (indirect) methods that search for discrepancies in comparable
sets of secondary macroeconomic data (i.e., data constructed and/or collected for other purposes).
In line with this consensus, and replicating the 2017 report, this study again employs the Labour Input
Method (LIM) to estimate the scale of undeclared work in the EU. Developed and first applied by the Italian
National Institute of Statistics, this indirect measurement technique assesses the gap between the selfreported supply of labour (e.g., through representative national surveys) and the official employment
figures reported by businesses (declarations to the tax administration, national statistical offices, labour
or social security authorities, etc.). The misalignment between the two sources is ascribed to undeclared
work (ISTAT, 2016). However, there are a range of steps that need to be taken to ensure the
exhaustiveness and comparability of the datasets used for this purpose.
A detailed description of this methodology is provided in the next section. Section 4 then reports the LIM
estimates of undeclared work in all EU Member States for 2019, including a comparison with the 2013
results, along with an analysis of the variations in the level of undeclared work across different types of
employment relationship and how the structure of the undeclared workforce varies between Member
States. Section 5 then evaluates the relationship between these cross-national variations in the level of
undeclared work and various economic and social structural disparities between countries.
4
3.0 Methodology used for estimating
the scale of undeclared work: the
Labour Input Method
This section provides an overview of the assumptions and procedures underpinning the Labour Input
Method. After discussing the operational definition of undeclared work used in this study, a detailed
description is provided of the method, data sources and variables used in the analysis. The section ends
with an overview of the key limitations of this estimation approach and suggestions for possible
improvements.
Key conclusions
The Labour Input Method measures the difference workers’ reports of labour supplied and
employers’ reports of labour used.
To analyse this, the Labour Force Survey and Structural Business Statistics are used.
The report presents the estimates for 2019, the latest year for which both sources of data were
available.
Estimates refer to all private sector activities, except for those in the financial sector, agriculture,
forestry and fishing, which are excluded due to the lack of data.
The scarcity of information provided by Structural Business Statistics is the key obstacle to
producing more reliable estimates.
The increasing prevalence of atypical forms of employment undermines the credibility of data
provided in the Labour Force Survey.
3.1
Operational definition of undeclared work
Undeclared work is defined as “any paid activities that are lawful as regards their nature, but are not
declared to the public authorities, taking into account the differences in the regulatory systems of the
Member States” (European Commission, 2007: 1). Member States have adopted a variety of different
definitions focusing upon non-compliance with either labour, tax and/or social security legislation or
regulations.2
If there are additional forms of non-compliance, it is not undeclared work. If the goods and services
provided are unlawful (e.g., trafficking of drugs, firearms, persons, or money laundering forbidden by law),
it is part of the wider criminal economy i.e., the shadow economy (often defined as including both the
undeclared economy and the criminal economy), and if there is no monetary payment, it is part of the
unpaid sphere. As such, illegal economic activities are excluded from the definition of undeclared work.
However, sometimes what is lawful in one country is illegal in others. For instance, in some countries,
prostitution is legal (e.g., Germany, Greece and Hungary) but not in others, and in some countries (e.g.,
the Netherlands) some drugs are legal but not in others. Here, therefore, the definition of undeclared work
2
For the definitions used in Member States, see http://ec.europa.eu/social/main.jsp?catId=1322&langId=en.
5
excludes these transactions that are legal in some countries but not others. As such, undeclared work
excludes all activities that are unlawful as regards their nature, but also some lawful activities in some
Member States, but not others, for which data is not collected in survey databases (i.e., EU- LFS and
SBS).
In consequence, undeclared work is composed of the following four broad types:3
Unregistered employment: an employment relationship which is not registered with the authorities
when it should be registered. Such employees often do not have written contracts or terms of
employment and their remuneration is most probably undeclared in nature.
Under-declared employment: when formal employers pursue the illegal practice of reducing their tax
and social security payments, and therefore labour costs, by under-declaring the remuneration of
employees. This occurs when employers pay their formal employees two salaries: an official declared
salary and an additional undeclared (“envelope”) wage which is hidden from the authorities for tax and
social security purposes. Alternatively, an employer can under-declare the number of hours an
employee works, such as to evade paying the minimum wage.
Envelope wages: often used in the context of under-declared employment, an envelope wage is
a cash-in-hand wage paid by a formal employer to a formal employee in addition to their official
declared salary, to reduce their tax and social security payments and therefore labour costs. It
arises from an agreement between the employer and employee, and additional conditions may be
attached to its payment, which are not in the formal written contract or terms of employment.
Undeclared self-employment: paid activity conducted by the self-employed where income is not
declared for the purpose of evading either tax and/or social insurance contributions owed. The selfemployed may not declare either some or all their income.
Unregistered family work: labour input by individuals who are not directly paid but do contribute to
the for-profit family business.
3.2
The Labour Input Method
As mentioned, the Labour Input Method searches for discrepancies between two contrasting sources of
information on labour usage: the self-reported labour supply by workers and the reported use of labour by
employers (ISTAT, 2016). Although any datasets providing exhaustive figures on these two sides of the
same coin would generally serve the purpose, the European Labour Force Survey (LFS) and the
Structural Business Statistics (SBS) are the only reasonable options when the goal is to deliver
comparable EU-wide figures. LFS is a representative large-scale survey of households, which is
conducted on a regular basis by national statistical offices of all Member States (see Eurostat, 2021).
Primarily focused on citizens’ participation in the labour force, this survey also provides rich information
on other characteristics of the interviewees that can ease the estimation procedure (e.g., country of work,
type of work engagement, sector in which activities are pursued, the average number of hours completed
during the observed period, the existence of additional jobs, etc.). SBS, on the other hand, represents a
compilation of figures that various administrative and statistical bodies collect from enterprises during a
year (Eurostat, 2022a).
3
https://www.ela.europa.eu/sites/default/files/2021-09/Glossary%20v6-final_0.pdf
6
In essence, the Labour Input Method treats LFS as a trustworthy source of information on labour input,
while SBS is believed to be deficient due to deliberate misreporting. In other words, it is presumed that
individuals interviewed as part of LFS have no reason not to report all paid work, given that no question
about the legitimacy of their activities is asked in the Survey. More importantly, they are assured not only
that the Survey serves strictly statistical purposes, but also that their identity would remain concealed if
they accept participation. In contrast, it is highly unlikely that companies employing workers on an
undeclared basis would provide any information about them to the relevant authorities. What is more, selfemployed individuals operating fully undeclared are not even encompassed by regular data-gathering
arrangements on the part of government bodies.
The estimation procedure, which is illustrated in Figure 1, can be hence summarised as follows4:
Project the figures on labour input from LFS to the total population and for the period of interest
(commonly the whole year).
Adjust and supplement the obtained results with external figures (e.g., on immigration, cross-border
work, and population changes) to ensure exhaustiveness.
Take relevant data on labour demand from SBS and make any adjustments if needed.
Harmonise the two sets of figures by converting them to the same units (hours worked or work efforts
expressed in terms of full-time equivalent employment).
Compare the resulting numbers to assess the discrepancies.
Put the resulting estimates of undeclared work in the context of economic output.
Figure 1. Overview of the Labour Input Method
Estimate total labour
input from the
supply-side
Estimate labour
inputs from the
demand-side
• Treat data obtained
from labour suppliers
as a ground truth
given that survey
interviewees generally
do not have reason to
conceal their
employment. Adjust
the dataset using
external figures (e.g.
census data, data on
illegal immigration,
work permits issued,
etc.).
• Estimate demand for
labour from the official
reports by enterprises
to the relevant
authorities. Adjust the
resulting figures using
external data if
necessary.
Standardise the total
labour input estimates
Compare the two
estimates
• Express the supplyside and demand-side
projections in the
same units to enable
comparison.
• Compare the
resulting estimates. A
surplus of labour input
reported by workers
over that reported by
companies is ascribed
to undeclared work.
The last step above requires the assumption that undeclared work does not differ from declared work in
terms of productivity because there is no data on the comparative productivity of declared and undeclared
work. To increase the credibility of the resulting estimates, it is thus important to take into account the
nature of work and the sector in which an activity takes place. Accordingly, in this study the procedure
expressed above is completed separately for each level of economic activity and each type of employment
(namely self-employment, waged employment and family work). For both sources of data, a table akin to
4
See also OECD (2002).
7
the one illustrated below (see Table 1) was produced. The figures were then compared on a cell-by-cell
basis and the resulting discrepancies were summed to obtain economy-wide estimates.
Table 1. Sample matrix for LIM method, an illustration
Sector
Self-employment
Waged employment
Family work
Total
Sector 1
Sector 2
Sector 3
Sector 4
⁞
Source: Authors’ illustration
In the rest of this section, an overview is provided of the variables used and the steps taken to reconcile
these two data sources and, consequently, assess the extent of undeclared work across the EU.
3.3
Data and variables used
3.3.1
Overview of the datasets and key definitions
For the purpose of this study, national LFS datasets were obtained directly from Eurostat upon request.
SBS data, on the other hand, are publicly available on the Eurostat website (Eurostat, 2022a). LFS
datasets provide a rich set of employment-related information essential for this task. The central variables
in this respect are the type of employment (self-employed, dependent employee, and unpaid family
worker), the number of jobs a person holds, the number of hours worked as part of each job, and the
sectors in which these jobs are performed. Besides various socioeconomic characteristics of the
respondents (such as gender, age, marital status, household size), which are essential for simulating
missing values, a range of other auxiliary variables is used to increase the exhaustiveness of the resulting
figures. For instance, there is a variable denoting the exact country in which an individual works. It allows
the identification and proper classification of cross-border workers and other persons working out of the
country of residence. Since they essentially belong to the labour force of another country, these individuals
have to be moved from their domicile LFS to that of the country in which they work.
It is important to realize that LFS approaches the matter from the perspective of work completed, while
SBS does so in terms of jobs. The problem is that a single individual is counted several times in the
business statistics if they work for more than one employer. This means that the figures from the two
sources are not directly comparable. To reconcile the employment-based view of LFS with the job-based
view of SBS, the information extracted from each of them must be converted into equivalent units. The
most obvious candidates in this respect are the total number of hours worked and the number of jobs in
full-time equivalents. Only after the harmonisation is properly completed can one proceed with the
comparison of the resulting figures.
For this purpose, internationally agreed definitions and accompanying concepts established in the European
System of Accounts (ESA 2010) are used. Among other things, ESA 2010 provides a clear and concise
8
overview of existing employment types. According to this nomenclature, employment covers all individuals
involved in productive activities which fall within the production boundary of the national accounts (Eurostat,
2013). Persons in employment are further divided into self-employed individuals and dependent employees.
Employees are defined as “persons who, by agreement, work for a resident institutional unit and receive a
remuneration recorded as compensation of employees” (Eurostat, 2013, p. 307). This definition is in line
with the definition of paid employment by the International Labour Organisation (ILO, 1982). On the other
hand, self-employed are “persons who are the sole owners, or joint owners, of the unincorporated
enterprises in which they work, excluding those unincorporated enterprises that are classified as quasicorporations” (Eurostat, 2013, p. 307). In cases when an individual is engaged in waged employment, but
also holds a job that falls within the latter scope, she/he will be classified as self-employed if, and only if, the
main part of their income comes from this self-employment. The third important group of individuals, which
is sometimes seen as a sub-category of the self-employed (ILO, 1982), are unpaid family workers
(Eurostat, 2013, p. 308). These are persons either engaged in a family business or working on a family farm,
who are not directly paid for their contribution. Instead, they share a household with the owner of the
particular business/farm and receive remuneration in the form of in-kind payments or fringe benefits.
When it comes to labour input, ESA 2010 defines total hours worked as the “aggregate number of hours
actually worked as an employee or self-employed person during the accounting period, when their output
is within the production boundary” (Eurostat, 2013, p. 310). Since the definition of employees also includes
persons with a formal attachment who are temporarily not at work, as well as part-time workers, total
hours worked provide more realistic information about labour productivity than the mere number of
persons employed.
Another important concept is employment in full-time equivalents (FTE). This indicator represents a
standardized view of the number of existing jobs, given that it takes into account the variability in the
efforts invested. Specifically, FTE in a certain sector/type of employment is obtained by dividing the sum
of all hours worked (i.e., including all workers) in this sector/type of employment by the average annual
number of hours completed per individual worker. Although total hours worked are commonly the most
accurate measure of labour input, FTE represents not only a good but sometimes an even desirable
alternative. Besides being useful for checking the robustness of the estimates based on the total hours
worked, in some situations this proxy actually gives more tangible insight into the workers’ effort. The
concept of FTE is particularly useful for sectors that are characterised by a stronger reliance on atypical
forms of employment (on-call work, platform work, employee sharing, etc).
Returning to the estimation strategy, it was to a large extent determined by the (limited) availability of data
provided as part of SBS. The SBS figures are broken down by size classes and the level of industry
(NACE Rev.2 classification). However, while separate figures are provided for the jobs held by dependent
employees (including FTE), the remaining two categories of interest are jointly presented. Moreover, data
on hours worked are only given for a small number of sectors (namely mining and quarrying;
manufacturing; construction; electricity, gas, steam and air conditioning supply; and water supply,
sewerage, waste management and remediation activities). It is thus clear that the information provided in
SBS is not sufficient to produce projections for either of the two target concepts. On the whole, the scarcity
of data available in SBS has remained the main obstacle to producing more robust estimates of
undeclared work using the LIM approach. More details about the limitations of SBS and strategies applied
to address them are provided in the next section.
Before moving to the industry coverage, it is important to mention that the resultant estimates of
undeclared labour input were converted into comparable monetary figures using value added at factor
9
cost. This variable, which is provided by SBS, denotes “the gross income from operating activities after
adjusting for operating subsidies and indirect taxes” (Eurostat, 2022b).
3.3.2
Comparable data on covered industries
LFS provides data for a representative sample of the country's population, which makes it possible to
project employment figures to the whole economy. On the other hand, SBS covers solely the “business
economy”, but not in its entirety. Specifically, besides education and health, which are mainly (but not
universally) non-market activities, also excluded are agriculture, forestry and fishing (NACE Rev.2 Section
A). In addition, data for financial services (section K) are not directly comparable with those for the
remaining parts of the business economy. This is not only because of the specific nature of financial
entities (insurance companies, pension funds and credit institutions) but also due to the limited availability
of most types of standard business statistics for this sector.
Consequently, SBS provides comparable figures for activities falling into the following sections of the
economy: Mining and quarrying (NACE Rev.2 Section B); Manufacturing (C); Electricity, gas, steam and
air conditioning supply (D); Water supply, sewerage, waste management and remediation activities (E);
Construction (F); Wholesale and retail trade, repair of motor vehicles and motorcycles (G); Transportation
and storage (H); Accommodation and food service activities (I); Information and communication (J); Real
estate activities (L); Professional, scientific and technical activities (M); and Administrative and support
service activities (N). Accordingly, the estimates of undeclared work presented in this study will be
restricted to these sectors only5. Since the previous wave of the study covered the same scope of activities
(see Williams et al., 2017), this will make it possible to directly compare the resulting estimates with those
for 2013.
Unfortunately, even this rather limited set of information on the demand side is not free of missing values.
The issue is particularly emphasised in small economies given that data for certain sectors could not be
made publicly available for confidentiality reasons. This has to do with a rather small number of units
operating in these sectors, which implies that provided numbers could be easily matched with individual
companies. Due attention to the deficiencies of the datasets used in this study is given in the following
section.
3.4
Analytical approach
This section explains the steps taken to produce the estimates of undeclared work in the EU. First, the
adjustments made to the LFS dataset and to enumerate the variables used to project the Survey data to
the population-level figures for each Member State are described. This is followed by a discussion of the
harmonisation procedure, which ensured direct comparability of the results from the two data sources.
The last part of this section reports the most important obstacles encountered during the estimation
process, alongside suggestions for improvements which would enable more reliable estimates.
3.4.1
Labour input from LFS
As mentioned, LIM treats the projection of LFS data to the population as an irrefutable truth and,
accordingly, compares it with “deficient” SBS figures to assess the scale of undeclared work. This rather
5 This means that LFS figures referring to these sectors only were taken into consideration during the data
reconciliation phase.
10
strong assumption implies that the quality of the resulting estimates of undeclared work in a certain country
is strongly affected by the quality of the accompanying LFS dataset. However, the problem is that LFS
cannot capture all labour suppliers. This deficiency is becoming more pronounced owing to the increased
dynamics of labour markets across Europe. The core of the problem lies in the fact that the sampling
frame for LFS includes only legitimate country residents. Consequently, the LFS of a particular country
lacks data about cross-border workers from other countries, non-resident seasonal workers, and
employed immigrants who legally entered the country during the year when the survey was conducted.
An additional challenge is posed by irregular immigrants (also not captured by LFS), whose contribution
to the labour input of certain Member States is far from marginal.
Part of the missing information about cross-border workers and recent legal immigrants could be restored
due to possessing datasets for all members of the European Economic Area (EEA) except Liechtenstein.
As explained previously, every interviewee who declared themselves as an active worker was asked in
which country this work was being performed. Individuals who worked in a country different from the
country where the interview was held were hence removed from the LFS of the resident country to the
LFS of the country of work. However, this only sorted out the issue for EEA citizens working in another
EEA country. There remained a significant portion of third-country nationals legitimately residing in the
EU without any permit whose labour input we were not able to effectively capture.6
When it comes to third-country nationals who require a valid work permit, their contribution was estimated
using the official data on work permits issued by the national authorities (Eurostat, 2022c). Since these
numbers refer to people rather than to jobs and hours worked, the exact labour efforts of each individual
had to be simulated. This was done under the assumption that the behaviour of these workers does not
deviate from that of foreign-national residents and cross-border EEA workers. The same strategy was
applied to incorporate irregular immigrants. However, the only comparable source of information in the
latter case is the Eurostat database on third-country nationals found to be illegally present on the territory
of the Member States (Eurostat, 2022d). Since detected individuals represent only a portion of irregular
immigrants, it is obvious that the adjusted LFS datasets are still far from being fully comprehensive.
Once the adjustments were completed, the next step was to project the data from these “refined” LFS
datasets to the whole population. Obviously, only survey participants working for pay were considered in
this process. Those were identified using the variable WSTATOR (labour status during the reference
week). Specifically, retained were only the individuals who either did any work for pay or profit during the
reference week or were not working during a reference week but had a job or business from which they
were absent (Eurostat, 2021). Their efforts were extrapolated with the help of the variables HWUSUAL
and HWACTUAL (Eurostat, 2021), which store the number of hours worked during the observed period.
The yearly equivalents were then projected to the whole population using the provided weighting factors.
To calculate the number of jobs in FTE, we controlled for the inherent differences not only between sectors
6 For instance, many citizens of Bosnia and Herzegovina do not require any special permit to work in Croatia. This is
an important, but certainly not the only reason why our estimates for Croatia are unusually low. Since joining the EU
in 2013, Croatia has witnessed strong emigration of its citizens to other EU Member States. To fill empty workplaces,
the authorities had lessened the immigration procedures for oversees citizens. As the work efforts of many of these
newcomers (and this especially applies to seasonal workers in tourism) were not included in LFS nor adequately
captured by other relevant statistics, the resultant estimates of undeclared work in Croatia are most likely severely
underestimated. Similar issues are pronounced in other countries. For instance, in 2019 citizens of six non-EEA
countries (Armenia, Belarus, Georgia, Moldova, Russia and Ukraine) were entitled to perform work in Poland without
a work permit under some conditions. Accordingly, the labour input of these individuals remained out of the scope of
our adjustment procedures.
11
and types of employment (as specified in Table 1), but also between different contract types (full-time
work vs. part-time work) and nature of the job (main employment vs. auxiliary/occasional work).
For each group of workers defined by these variables, the average number of hours completed during a
year were calculated. The total number of hours worked is then divided by these averages to get full-time
equivalents for each and every group. In the final step, the resulting figures are summed across contract
types and workload arrangements to get two matrices of totals (Table 1). One of these matrices contained
the total number of hours worked, while the other expressed the same information in terms of FTE jobs.
Besides being a good robustness-checking mechanism, the option with two different tables was, in fact,
the only feasible estimation strategy for a great many countries due to the scarcity of information provided
by SBS (as discussed previously).
3.4.2
Reconciliation of LFS with SBS on labour input
As explained, SBS figures alone are not sufficient to construct the corresponding matrices that could be
compared with LFS projections. The first problem, which affects both estimation approaches is the
absence of independent figures for the self-employed and family workers. Since for each covered sector
only a single number is available (referring to “unpaid persons employed”), it was first required to split
these values. Moreover, these numbers are expressed in absolute terms, which means that each of the
two newly generated values had to be further converted into FTE jobs and the total number of work hours.
This was achieved using split-ratios and average figures for work hours obtained from LFS. A matrix of
weights specifying the dispersion of the two groups of individuals within every single sector was first
developed from LFS data, taking into account the inherent differences between individuals depending on
the type of their contract (full-time vs part-time) and the nature of the job (main job vs
supplementary/secondary job). These weights were then used to obtain separate figures on the number
of jobs held by self-employed and family workers for every single combination of the three variables of
interest (i.e., sector, contract type and nature of the job). In the next step, the total number of hours worked
during a year was approximated for every subgroup of interest. This was achieved using the information
on the structure of the labour market and distributional characteristics of workers’ input (in terms of hours
completed) which was again assessed via a comprehensive analysis of the LFS data.
Although breaching the basic requirement that each of the two sides of the equation should provide
independent figures on the labour input, this was the only viable solution to the issue. This procedure is,
however, justifiable on the same grounds as the Labour Input Method is, given that both perceive the
information provided by LFS as an indisputable fact. Still, this only further emphasises the necessity to
have as accurate LFS data as possible for each Member State.
The same strategy was applied in the case of dependent employees, but only to produce hour-based
results. This is because SBS provide exact figures on total FTE jobs held by dependent employees for
sectors covered in this particular dataset. These figures enabled the production of quite robust estimates
of undeclared work conducted by persons in waged employment. The same cannot be said for the selfemployed and family workers, given the aforementioned assumptions their estimation procedure relies
on. Nevertheless, once comparable labour inputs were generated and subtracted, the resulting
differences were ascribed to undeclared work. These absolute values then made it possible to assess
which proportion of GVA came from undeclared work.
It should be stressed that the procedure described in this section is identical to the one applied in the
previous wave (see Williams et al., 2017). This means that results are fully comparable on a country level.
The only notable difference relates to the geographical scope of the analysis. Given that it left the EU in
12
the meanwhile, the United Kingdom was not covered in this round of the study. More importantly, this time
we were able to produce estimates for Malta, which was not the case in the previous wave.
Finally, the results were produced for 2019, the last year for which both sources of data were available at
the time this study took place. Nevertheless, it must be stated that 2019 would have been chosen even if
newer data had been available. The first and most important reason for this is the potential harmful
disruption of the data caused by the COVID-19 pandemic. It is likely that for the years 2020 and 2021,
discrepancies between the supply- and demand-side data may also be due to the pandemic. Workers
interviewed as part of LFS might for instance have reported that they were not in employment because
they were receiving short-term financial support under the employment retention schemes developed in
response to the pandemic, largely funded by a €100 billion “Support to mitigate Unemployment Risks in
an Emergency” (SURE) programme.7
Meanwhile, and because these workers were officially employed, SBS data is likely to have reported them
as still in employment. Therefore, the short-term financial support schemes most probably would cause
an additional discrepancy between the LFS and SBS data that is not due to undeclared work in 2020 and
2021. For the 2020 or 2021 data, then there would be no way of knowing whether the discrepancy between
the supply- and demand-side data was due to the misreporting by workers in LFS or whether it was entirely
due to undeclared work. In other words, if a portion was due to misreporting by workers of their
employment status, this could have led in 2020 and 2021 to an underestimate of the extent of undeclared
work compared with other non-pandemic years.
The above explanation illustrates just one of many important limitations of the Labour Input Method and
the datasets it relies on. In the next section, the shortcomings of the applied approach and possible
remedies are outlined.
3.5
Deficiencies of data sources and suggestions for
improvements
Despite a slight improvement in the quality of available data from the time this study was first carried out,
the estimation of the extent of undeclared work using LIM remains a complex and challenging task. First
of all, it must be stressed that this method provides highly conservative estimates of the real size of the
undeclared economy. This is because it is primarily focused on detecting undeclared work attributable to
the labour input. This multifaceted and complex realm, however, embraces many other transactions that
do not necessarily include undeclared labour relations such as unregistered employment and envelope
wages (e.g., businesses not declaring all their transactions in reported turnover). In general, economic
output comes from both labour and capital, with only the former being fully captured by LIM. This method
is hence more effective in assessing undeclared work in labour-intensive sectors where capital inputs play
a less important role. At the same time, many undeclared transactions in capital-intensive sectors most
likely remain undetected by LIM.
A further limitation of LIM can be found in its foundational assumption that data on the supply of labour is
undisputedly accurate. Undoubtedly, this is an overly optimistic hypothesis as it is highly plausible that a
certain portion of the survey participants decided to stay silent about their unregistered activities. Likewise,
some of them might have misunderstood the concept of work comprehended by the survey and hence
did not bother to mention their sporadic for-profit activities during the interview (e.g., via collaborative
European Commission 2020. COVID 19- Economic package – multiannual financial framework 2014-2020.
Brussels: European Commission.
7
13
platforms). Finally, there is also unintentional misreporting that is of vital importance for the LIM method.
This primarily relates to honest mistakes with respect to the sector in which survey participants work
and/or their type of employment. The negative impacts of such mistakes feed through the whole estimation
pipeline.
When it comes to the demand side, the bulk of deliberate underreporting of labour input on the part of
business owners unquestionably has to do with undeclared work. However, there are also some other
important sources of error in this respect. For instance, the authorities might accidentally skip certain units
while collecting business data. Honest mistakes in reporting by the companies also cannot be excluded.
The risk of inaccurate reporting or mismeasurement due to statistical factors has, however, been
significantly decreased in recent years following the endeavours of Eurostat and relevant national bodies
to ensure the exhaustiveness of business statistics. In addition, sophisticated techniques for data
gathering and processing are increasingly employed to resolve discrepancies across multiple data
sources.
Nonetheless, the scarcity of information available in SBS continues to represent the biggest obstacle to
producing highly reliable estimates of undeclared work in the EU. Due to the partial availability of data on
hours worked and jobs in full-time equivalents, a range of simulations had to be done to obtain harmonised
results. These simulations, however, required a number of assumptions to be made by the data analysts.
Generally speaking, the more assumptions an estimation technique relies on, the less robust the resulting
figures are. Consequently, although we did our best to develop as credible an estimation pipeline as
possible, the estimates of the magnitude of undeclared work presented in the subsequent sections should
not be treated as undisputable facts, but rather as educated guesstimates.
Given this, some important steps ought to be done in order to increase the reliability of LIM estimates in
future waves. The most important issue to address in this respect is the current paucity of data provided
by SBS. Besides a more exhaustive overview of the demand side in the private sector, it would be also
beneficial if there were comparable figures for the remaining parts of the economy. In this respect, it may
be worth considering the option of harmonising and unifying as much administrative and statistical data
held by the individual Member States as possible. Furthermore, the reliability of the resultant estimates
would be much higher if the Member States improved their statistics on the participation of both regular
immigrants from third countries and irregular immigrants in the labour market. This would enable finetuning of the figures on the labour supply obtained from LFS.
4.0 Estimates of undeclared work in the
EU27 using the Labour Input Method
This section reports the estimates of the extent of undeclared work in EU Member States in 2019 along with
how this changed since 2013. It also assesses how the occurrence of undeclared work varies across different
types of employment, and how the structure of the undeclared labour market differs across the EU27.
Key conclusions
The finding is that in 2019, on average, 11.1 % of total labour input in the private sector in the EU is
undeclared (11.6 % in 2013), and undeclared work is on average 14.8 % of GVA (16.4 % in 2013).
However, these are unweighted averages, not considering the relative size of the labour force in
each Member State. The weighted averages are that 9.7 % of total labour input in the private sector
14
in the EU is undeclared (10.2 % in 2013), and undeclared work constitutes 14.6 % of GVA in the
private sector (14.9 % in 2013).
Between 2013 and 2019, therefore, there has been a decline in the extent of undeclared work in the
EU. Indeed, analysing the changing size of undeclared work between 2013 and 2019 in 26 Member
States (estimates could not be produced for Malta in 2013), its magnitude has declined in 19 out of
these 26 countries (the exceptions being Romania, Bulgaria, Lithuania, Italy, France, Netherlands
and Germany). This displays the progress that has been made in tackling undeclared work both in
the EU as a whole and across most Member States.
Nevertheless, there remain substantial differences in the relative size of the undeclared economy
between the Member States.
Analysing undeclared work as a proportion of total labour input, Romania (21.7 %), Lithuania
(20.8 %), and Bulgaria (19.3 %) have the highest levels of undeclared work. The lowest shares
were found in Germany (3.9 %), the Netherlands (4.8 %), and Austria (5.1 %). In spite of a few
exceptions, countries exceeding the EU average are largely Member States in Central and
Eastern Europe.
The ordering of countries is similar when undeclared work is evaluated as a proportion of GVA.
The highest shares are evidenced in Romania (27.1 %), Lithuania (26.0 %), and Bulgaria
(23.8 %), while the best achievers are Austria (5.3 %), Luxembourg (7.0 %), and Sweden (7.5
%).
There are also significant differences in the prevalence of undeclared work in different types of
employment relationship.
The proportion of self-employment which is undeclared ranges from as low as 3.6 % in Belgium
to as high as 65.2 % in Cyprus.
The proportion of waged employment which is undeclared work is highest in Bulgaria (20.2 %)
and lowest in the Netherlands (0.5 %).
The proportion of family work that is undeclared work is highest in Luxembourg (68.9 %), Malta (63.0 %),
Cyprus (62.2 %), and Romania (60.1 %) and lowest in Sweden (2.1 %), Italy (2.4 %), Czechia (2.6 %),
and Slovakia (3.3 %).
Examining the structure of the undeclared labour market in the EU, 62.9 % (61.8 % in 2013) of all
undeclared work is waged employment, 36.3 % (37.3 % in 2013) is self-employment and 0.8 %
(0.9 % in 2013) is family work. However, there are significant national variations. Countries in which
the majority of undeclared work is conducted through self-employment include the Netherlands (90
%), Cyprus (82.2 %), Ireland (71.4 %), and Finland (68.1 %), while those where over 90 % of all
undeclared work is waged employment include Poland, Belgium, Bulgaria, and Italy.
4.1
The size of undeclared work in the EU
Figure 2 reveals that undeclared work represents 14.8 % of GVA in the private sector at the EU level.
However, undeclared work as a proportion of GVA in the private sector varies across the Member States,
from as low as 5.3 % of GVA in Austria to as high as 27.1 % in Romania. Despite a few notable exceptions,
it is post-socialist and Mediterranean countries that are recording the highest levels of undeclared work
as a proportion of GVA. For instance, undeclared work in the private sector accounts for 26.0 % of GVA
15
in Lithuania, 23.8 % in Bulgaria, 23.4 % in Malta and 20.4 % in Italy. On the other hand, it is lowest in
Luxembourg, where only 7.0 % of the official GVA is undeclared work, Sweden (7.5 %) and Germany
(8.6 %).
23.8
23.4
20.4
20.1
18.8
16.5
16.4
14.9
14.5
14.0
13.6
13.5
13.0
12.8
12.0
11.1
9.9
9.6
9.3
8.6
7.5
5.3
10
7.0
15
10.6
20
14.8
25
19.7
27.1
30
26.0
Figure 2. Undeclared work in the private sector as % of total GVA, LIM estimates for 2019
5
0
Note: Figure for EU represents an unweighted average
Source: Authors’ work
There has been a decline in the extent of undeclared work in the EU from 16.4 % to 14.8 % of GVA
between 2013 and 2019. Speaking in terms of weighted averages, which consider the relative size of the
labour force in each Member State, undeclared work in the EU has fallen from 14.9 % to 14.6 % of GVA
in the observed period. Indeed, and as Figure 3 displays, the magnitude of undeclared work has fallen
in 19 out of 26 countries (note that the earlier wave did not include Malta). These results display the
progress made in tackling undeclared work both in the EU as a whole and across most Member
States.
Romania, Bulgaria, Lithuania, Italy, France, Netherlands and Germany are the only countries in which the
portion of GVA which is undeclared work has increased between 2013 and 2019. Nonetheless, some
caution is required in interpreting this. As will be shown below, in some of them (namely Italy, Netherlands
and Germany), the share of the workforce engaged in undeclared work has decreased, which indicates
there has been a structural transformation inside the undeclared economy (i.e., undeclared work has
shifted towards more productive sectors and/or towards more productive types of work arrangement).
Figure 3 further reveals that Poland, Latvia, Czechia and Cyprus have been most successful in tackling
undeclared work between 2013 and 2019 when measured in terms of its reduction as a proportion of
GVA. Undeclared work in Poland accounted for 19.7 % of GVA in 2019, which was 7.6 percentage points
less than in 2013. In Latvia the share fell 7.4 percentage points (from 22.3 % to 14.9 %), while in Czechia
and Cyprus it fell 7.3 percentage points (from 16.9 % to 9.6 % and from 17.9 % to 10.6 %, respectively).
16
Figure 3. Undeclared work in the private sector as % of total GVA, 2013 and 2019
26.2
27.1
25.2
26.0
Romania
Lithuania
19.2
Bulgaria
23.8
Malta
23.4
17.2
Italy
20.4
Hungary
23.2
20.1
Poland
27.3
19.7
Greece
Spain
16.5
Estonia
18.8
17.9
21.3
16.4
Latvia
22.3
14.9
EU average
14.8
14.7
14.5
Slovakia
Belgium
14.0
11.0
France
16.4
15.4
13.6
Croatia
17.1
13.5
Slovenia
16.4
13.0
11.9
12.8
Netherlands
Portugal
15.5
12.0
13.0
11.1
Ireland
Cyprus
17.9
10.6
11.8
9.9
Finland
Czechia
16.9
9.6
Denmark
14.3
9.3
7.1
Germany
Sweden
7.5
Luxembourg
7.0
Austria
5.3
0
5
22.5
8.6
9.7
9.1
10.0
10
15
2013
Note: Figures for EU as a whole are unweighted averages
Source: Authors’ work
17
20
2019
25
30
Figure 4 graphically displays these changes between 2013 and 2019 in undeclared work as a proportion
of GVA. This shows that besides the ones just mentioned above (namely Poland, Latvia, Czechia and
Cyprus), many other Member States in Central and Eastern Europe that have had large undeclared
economies have managed to reduce its magnitude to some extent. Some notable examples are Estonia
(a -4.9 percentage points decrease), Croatia (-3.6), Slovenia (-3.4) and Hungary (-3.1).
Figure 4. Changes in the share of undeclared work in the private sector (2013-2019), in
percentage points
Source: Authors’ work
However, this is just one measure of the size of undeclared work and the changes over time. It can also
be measured as a percentage of total labour input in the private sector. In 2019, on average, 11.1 % of
total labour input in the private sector in the EU is undeclared (compared with 11.6 % in 2013).
Analysing weighted averages, the share has fallen from 10.2 % to 9.7 % during the observed period.
Therefore, there has been a decline of undeclared work as a proportion of total labour input in the
private sector in the EU between 2013 and 2019.
18
Figure 5 presents undeclared work as a percentage of total labour input, both for the private sector and
for the economy as a whole in each of the Member States 8. The highest proportions of undeclared work
as a percentage of total labour input in the private sector are found in Romania (21.7 %), Lithuania (20.8
%), and Bulgaria (19.3 %), whilst the lowest shares are found in Germany (3.9 %), the Netherlands (4.8
%) and Austria (5.1 %).
Although countries exceeding the EU average of 11.1 % are largely Member States in Central and Eastern
Europe, there are some exceptions. Explicitly, Czechia (7.0 %), Croatia (10.7 %) and Slovenia (10.8 %)
have below-average shares of undeclared work. At the same time, France (11.8 %), Italy (12.6 %) and
Greece (12.6 %) are the only Member States that joined the EU before 2004 in which undeclared work
accounts for shares higher than the EU average. However, this is true only when the analysis is restricted
to the private sector. The results are different when the whole economy is examined (e.g., France falls
below the average).
Figure 5. Undeclared work in the EU in terms of the labour input, LIM estimates for 2019
minimum % of total labour input in the economy
5
16.7
20.8
19.3
9.9
14.4
9.7
16.0
11.1
16.0
11.5
14.0
8.9
12.6
8.9
12.2
12.6
8.6
9.0
12.1
8.5
11.1
7.3
7.0
10.8
6.6
10.8
7.7
7.2
8.6
5.1
8.6
5.7
10
2.6 3.9
4.8
3.0
5.1
3.5
5.3
3.2
6.7
3.7
6.7
4.6
7.0
5.1
7.3
4.8
7.5
4.9
4.9 7.8
10.7
15
11.8
20
17.9
25
21.7
% of labour input in the corporate sector
0
Note: Figures for EU as a whole are unweighted averages. The word ‘minimum’ signals that the real
shares are probably somewhat larger, as explained in section 3.5.
Source: Authors’ work
Comparing the magnitude of undeclared work as a percentage of labour input in the private sector in 2013
and 2019, Figure 6 reveals that undeclared work has decreased as a percentage of labour input in
the private sector in 19 countries. Those where undeclared work as a percentage of labour input in the
private sector increased include France (+3.0 percentage points), Romania (+2.8), Bulgaria (+1.5),
Luxembourg (+1.3), Lithuania (+1.0), Portugal (+0.9), and Greece (+0.2). Although the results for Portugal
and Luxembourg seem to contradict the findings with respect to the shares of GVA, this is not the case.
Productivity differs significantly not only between sectors but also between different types of employment
8 The latter was calculated by dividing the estimated undeclared labour input by the total labour input across the
economy (i.e., including the public sector). Since undeclared work can appear in the public sector as well, these
figures essentially represent lower boundaries of the real extent of undeclared work in the economy.
19
arrangement. For instance, family workers commonly exhibit lower productivity than self-employed and
dependent employees. Any change in the structure of the undeclared workforce therefore manifests itself
in the results in terms of the size of undeclared work as a percentage of GVA, even if the numbers with
respect to the labour input stay unchanged.
Figure 6. Undeclared work in the private sector as % of the labour input, 2013 and 2019
18.9
Romania
21.7
19.8
Lithuania
Bulgaria
17.8
Malta
17.9
17.3
Hungary
16.0
Poland
19.3
20.8
16.0
Latvia
14.0
20.8
18.3
12.4
12.6
12.9
12.6
Greece
Italy
Estonia
12.2
Slovakia
12.1
14.8
13.4
8.8
France
11.8
11.6
11.1
11.9
10.8
EU average
Belgium
Slovenia
10.8
Croatia
10.7
Denmark
8.6
Cyprus
8.6
Finland
7.8
6.6
Portugal
Ireland
Spain
5.4
Sweden
5.3
Austria
5.1
5.2
4.8
4.4
3.9
Netherlands
Germany
0
5
9.6
13.8
9.3
8.6
8.8
6.7
Luxembourg
14.2
7.5
7.3
7.7
7.0
Czechia
13.2
6.7
7.7
8.7
10
15
2013
Note: Figures for EU as a whole are unweighted averages
Source: Authors’ work
20
2019
20
25
Figure 7 graphically displays the geography of these changes in undeclared work as a percentage of
labour input in the private sector between 2013 and 2019. The highest falls are found in Cyprus (a -5.2
percentage points decline in undeclared work as a percentage of labour input in the private sector), Poland
(-4.8), and Latvia (-4.3). Notable improvements are also noticeable in older Member States, with Austria
(-3.6), Sweden (-2.4) and Spain (-2.1) having the largest percentage point declines in undeclared work as
a proportion of labour input in the private sector.
Figure 7. Changes in the share of undeclared labour input in the private sector (2013-2019),
in percentage points
Source: Authors’ work
4.2
Undeclared work by type of employment relationship
There are also marked differences in the prevalence of undeclared work in different types of employment
relationship. Table 2 gives a detailed overview of the extent of undeclared work in different types of
employment relationship, namely self-employment, waged employment, and family work.
Starting with dependent employees, the highest share of waged employment in the private sector that
is undeclared as a percentage of total labour input is found in Bulgaria (20.2 % of waged employment is
undeclared measured as a share of total labour input), Lithuania (19.1 %), Romania (18.8 %), and Poland
(18.1 %). It is lowest in the Netherlands (where just 0.5 % of waged employment in the private sector is
21
undeclared measured as a share of total labour input), Cyprus (1.3 %), Germany (2.4 %), and Ireland
(2.4 %). Considering waged employment alone, the proportion that is undeclared is much higher in Central
and Eastern Europe. The exceptions are Greece, Italy and Belgium, which are the only Member States
that joined the EU before 2004 with incidence rates above 10 %.
Table 2. Undeclared work in the private sector in the EU, LIM estimates 2019
Source: Authors’ work
The differences between countries are even more acute when self-employment and family work are
considered. The proportion of self-employment that is undeclared (measured as a share of their total
labour input) ranges from only 3.6 % in Belgium to 65.2 % in Cyprus. Other countries with low levels of
undeclared work among the self-employed are Italy (4.1 %), Poland (4.6 %), Czechia (6.8 %), and
Slovenia (9.2 %). Undeclared work among the self-employed is higher in Romania (62.5 %), Latvia
(55.8 %), and Denmark (54.7 %).
Turning to family workers, the proportion declared is lowest in Luxembourg (with 68.9 % of their labour
input being undeclared), Malta (63.0 %), Cyprus (62.2 %), and Romania (60.1 %), whilst the proportion
declared is highest in Sweden (with just 2.1 % of their labour input being undeclared), Italy (2.4 %),
Czechia (2.6 %), and Slovakia (3.3 %).
22
4.3
Structure of the undeclared labour market
Figure 8 reveals the structure of the undeclared labour market and how this significantly varies across Member
States. In the EU, the proportion of undeclared labour input that is waged employment is 62.9 %. This is an
increase in comparison with 2013 when this share accounted for 61.8 % (Williams et al., 2017). Meanwhile, 36.3 %
of undeclared labour input is self-employment at the EU level, which is one percentage point less than in 2013,
and the proportion of undeclared labour input that is family work is 0.8 % in 2019 (0.9 % in 2013).
Figure 8. Undeclared work by type of employment, LIM estimates for 2019
0.2
95.5
4.4
0.8
93.3
6.0
0.1
92.9
7.0
0.3
90.9
8.8
0.1
89.6
0.2
82.5
17.3
0.1
82.2
17.7
0.1
81.9
18.0
0.3
80.9
18.8
0.0
80.5
19.5
0.4
78.3
21.3
0.0
77.6
0.3
67.4
3.6
32.3
10.3
0
Family workers
22.4
10
62.6
35.8
36.3
20
42.4
30
62.8
40
82.2
50
90.0
60
33.8
0.0
64.2
0.8
62.9
5.9
51.7
0.1
56.3
43.6
0.6
53.6
45.9
0.8
52.5
46.7
0.2
53.0
70
Dependent employees
46.8
1.3
46.2
52.4
0.2
45.9
54.0
1.2
36.0
0.1
31.7
68.1
0.7
27.9
80
71.4
90
13.5 4.3
100
9.1 0.8
Self-employed
Note: Figures for EU as a whole are unweighted averages
Source: Authors’ work
Nevertheless, the differences in the structure of the undeclared labour market across Member States are significant.
For instance, the bulk of undeclared work is conducted by self-employed individuals in the Netherlands (90 % of all
undeclared work), Cyprus (82.2 %), Ireland (71.4 %), and Finland (68.1 %). On the other hand, in Poland, Belgium,
Bulgaria, and Italy more than 90 % of undeclared work is undertaken through waged employment. Family work
represents a negligible slice of undeclared work in most countries. Luxembourg, Greece, and Cyprus are the only
Member States in which this share exceeds 2 %.
These differences in the structure of the undeclared workforce across the EU have significant implications for
tackling undeclared work. For instance, countries where undeclared work predominantly occurs through selfemployment may find it more beneficial to prioritize policy initiatives encouraging businesses to start-up legitimately
and making it easier and more beneficial for businesses to shift towards becoming fully legitimate. In contrast,
countries where waged employment is the main contributor to undeclared work will find it more beneficial to focus
upon policy measures that make unregistered and under-declared employment more expensive and/or less
appealing.
23
5.0 Correlations between undeclared work
and structural conditions
This section evaluates the relationship between cross-national variations in the extent of undeclared work and crossnational variations in economic and social structural conditions.
Key conclusions
Various structural conditions are strongly correlated with the growth or decline of undeclared work.
The finding is that significantly lower levels of undeclared work are found in Member States where there
exist:
Higher government effectiveness and lower perceived levels of corruption.
Higher levels of development (whether measured in terms of GDP per capita, the Human Development
Index or Social Progress Index) and greater levels of state intervention in work and welfare (to enhance
workers` rights and labour productivity, investments in research and development and implementing
measures for reducing poverty and inequalities).
Higher quality more powerful formal institutions (higher reliability of policy services, higher judicial
independence, stronger rule of law, stronger voice and accountability, positive perception towards the
regulatory quality and higher trust in the state institutions).
Lower levels of instability and uncertainty in formal institutions (better transparency in policymaking and
reduced perception of political instability).
Greater symmetry between the norms, values and beliefs of citizens, workers, employers and businesses
and the formal rules (sometimes termed “vertical trust” and measured by the level of tax compliance, and
the acceptability of undeclared work) and a higher trust in peers to adhere to the formal rules (sometimes
termed “horizontal trust” and measured by personally knowing people engaged in undeclared work and
their estimates of the share of undeclared work).
5.1
Structural determinants of undeclared work in the EU
In recent years, there has been growing recognition that there are various structural determinants of the scale of
undeclared work. In a 2020 Platform report that provides a comprehensive literature review of previous studies on
the structural conditions associated with the growth or decline of undeclared work in the EU,9 the structural economic
and social conditions found to significantly determine the extent of undeclared work were grouped as follows:
Formal institutional failings:
i. The lack of modernisation of government organisations (measured by government effectiveness) and
persistence of public sector corruption (measured by the Corruption Perception Index and control of
corruption), or what is termed formal institutional misallocations and inefficiencies.
9
Williams, C.C. and Horodnic, I. (2020) Horizon scanning: early warning signals of future trends in undeclared work, European
Commission, Brussels
24
ii. Lower levels of “development” (measured by GDP (current prices, euro per capita), Human Development
Index and Social Progress Index); lower levels of state intervention in work and welfare provision (measured
by the burden of government regulation, workers’ rights, business flexibility index, expense of government,
research & development expenditure, tax revenue, social contributions, impact of social transfers on
poverty reduction, labour market policy (LMP) expenditure) and higher levels of poverty and inequality
(measured by people at risk of poverty/social exclusion, severe material deprivation rate, inequality of
income distribution in terms of the income quintile share ratio, Gini coefficient and labour productivity), or
what is termed formal institutional voids and weaknesses.
iii. Higher quality more powerful formal institutions, measured by the reliability of police services, judicial
independence, rule of law, regulatory quality, voice and accountability, trust in government, trust in
parliament, trust in tax and social security authorities and trust in labour inspectorate.
iv. Formal institutional instability and uncertainty, measured by the Democracy Index and political stability.
Informal institutional failings, including the asymmetry between the norms, values and beliefs of citizens,
workers, employers and businesses and the formal rules (sometimes termed “vertical trust” and measured by
the level of tax compliance, and the acceptability of undeclared work) and a lack of trust in peers to adhere to
the formal rules (sometimes termed “horizontal trust” and measured by personally knowing people engaged in
undeclared work and their estimates of the share of undeclared work).
Evaluating whether the cross-national variations in the size of undeclared work (using the LIM estimate of the extent
of undeclared work as a percentage of total labour inputs in the private sector) and the cross-national variations in
the economic and social structural conditions listed above (see Table A1 in the Annex for the indicators used to
measure each), the findings are as follows ranked in order of the strength of the correlation 10 (see Table A2 in
the Annex for the results).
A “strong” negative significant relationship, ranked in order of the strength of the correlation, is identified between
the prevalence of undeclared work and:
Social Progress Index (rs=-.761***), as a measure of the lower level of development
Democracy Index (rs=-.751***), as a measure of formal institutional uncertainty and instability.
GDP per capita (rs=-.725***), as a measure of the lower level of development.
Corruption Perceptions Index (rs=-.710***), as a measure of the perceived higher level of public sector corruption.
Human Development Index (rs=-.706***), as a measure of the lower level of development.
Rule of law (rs=-.704***), as a measure of the powerlessness of formal institutions.
Government effectiveness (rs=-.699***), as a measure of the lower level of modernisation of government
institutions.
Control of corruption (rs=-.695***), as a measure of the higher level of public sector corruption.
Regulatory quality (rs=-.691***), as a measure of the lower level of modernisation of government institutions.
10
The strength of the correlation/ relationship is assessed based on the Spearman`s rho coefficient (the descriptors apply to
both, positive and negative relationships): 0.81 to 1 very strong relationship, 0.61 to 0.80 – strong relationship, 0.41 to 0.60 –
moderate relationship, 0.21 to 0.40 – weak relationship, 0.01 to 0.20 – very weak or negligible relationship. The significance of
the relationship is given by the p value and it is significant at: *** p<0.01, ** p<0.05, * p<0.1.
25
Voice and accountability (rs=-.690***), as a measure of the powerlessness of formal institutions.
social capital (rs=-.638***), as a measure of informal institutional failings.
Judicial independence (rs=-.636***), as a measure of the powerlessness of formal institutions.
Reliability of police services (rs=-.610***), as a measure of the powerlessness of formal institutions.
A “strong” positive significant relationship is identified between the prevalence of undeclared work and severe
material deprivation rate (rs=.701***), as a measure of the high level of poverty.
A “moderate” negative significant relationship between cross-national variations in undeclared work (using the
LIM estimates) and the following structural conditions is identified, ranked in order of the strength of the correlation:
Research & Development expenditure (rs=-.578***), as a measure of a lower level of state intervention in the
economy
Estimated share of undeclared work (rs=-.574***), as a measure of trust in one’s peers to act according to the
formal rules.
Labour productivity (rs=.550***), as a measure of a lower level of state intervention in the economy.
Trust in parliament (rs=-.502***), as a measure of trust in formal institutions.
Workers’ rights (rs=-.460**), as a measure of a lower level of state intervention in the economy.
Impact of social transfers on poverty reduction (rs=-.446**), as a measure of a lower level of state intervention in
welfare provision.
Acceptability towards undeclared work by firm for firm (rs=-.422**), as a measure of adherence to the formal
rules.
Trust in government (rs=-.420**), as a measure of trust in formal institutions.
A “moderate” positive significant relationship between cross-national variations in undeclared work (using the LIM
estimates) and the following structural conditions is identified, ranked in order of the strength of the correlation:
Labour productivity (rs=.550***), as a measure of a lower level of state intervention in the economy. People at
risk of poverty/social exclusion (rs=.516***), as a measure of higher levels of poverty.
There is a “weak” negative but significant relationship between cross-national variations in undeclared work (using
the LIM estimates) and the following structural conditions ranked in order of the strength of the correlation:
Acceptability of undeclared work by firm for private household (rs=-.396**), as a measure of adherence to the
formal rules.
Political stability (rs=-.392**), as a measure of formal institutional uncertainty and instability.
Trust in Labour inspectorate (rs=-.377*), as a measure of trust in government.
Burden of government regulation (rs=-.373*), as a measure of a lower level of state intervention in the economy.
Trust in Tax and Social Security authorities (rs=-.367*), as a measure of trust in government.
Acceptability towards someone partially of completely conceals their income (rs=-.346*), as a measure of
adherence to the formal rules.
26
Tax compliance (rs=-.326*), as a measure of adherence to the formal rules.
Acceptability of firms hiring workers on undeclared basis (rs=-.267), as a measure of adherence to the formal
rules.
There is a “weak” positive but significant relationship between cross-national variations in undeclared work (using
the LIM estimates) and the following structural conditions ranked in order of the strength of the correlation:
Inequality of income distribution (rs=.370*), as a measure of higher levels of inequality.
Gini coefficient distribution (rs=.364*), as a measure of higher levels of inequality.
There is a very weak negative or no significant relationship between cross-national variations in undeclared
work (using the LIM estimates) and the following structural conditions:
Business flexibility index (rs=-.178), as a measure of state intervention in the economy.
Tax revenue (rs=-.105), as a measure of state intervention in work and welfare.
Social contributions (rs=-.098), as a measure of state intervention in welfare provision.
Personally know people engaged in undeclared work (rs=-.096), as a measure of trust in one’s peers to act
according to the formal rules.
Acceptability of an individual doing undeclared work for a private household (rs=-.059), as a measure of
adherence to the formal rules.
There is a very weak positive or no significant relationship between cross-national variations in undeclared
work (using the LIM estimates) and the expense of government (rs=.094), as a measure of state intervention in
work and welfare.
To better understand these results, Figures 9 to 22 below graphically portray the association between the crossnational variations in undeclared work and the cross-national variation in those structural conditions.
5.2
Government ineffectiveness and public sector corruption
Figure 9 charts the cross-national variations in the size of the undeclared economy using the LIM estimates (% of
total labour inputs in the private sector) and cross-national variations in government effectiveness (which captures
perceptions of the quality of public services, the civil service and their independence from political pressure as well
as government commitments to policy formulation and implementation). This reveals a strong correlation between
cross-national variations in the perceived government effectiveness and cross-national variations in the prevalence
of undeclared work, measured by Spearman’s rank correlation coefficient (rs= -.699 ***). The greater the perceived
effectiveness of the government, the lower is the prevalence of undeclared work. Countries such as Romania,
Bulgaria and Greece with a perception of weaker government effectiveness have higher levels of undeclared work,
whilst countries where the government effectiveness is perceived as stronger (e.g., Netherlands, Denmark, Finland)
have lower levels of undeclared work.
27
Figure 9. Relationship between
undeclared work and government
effectiveness, 2019
Source: own calculations based on data from
World Bank, The Worldwide Governance
Indicators (WGI)
There is also a strong significant correlation between cross-national variations in the level of undeclared work and
cross-national variations in the perceived levels of corruption, measured using Transparency International’s
Corruption Perceptions Index (see Figure 10). A country’s score indicates the perceived level of public sector
corruption on a scale of 0 - 100, where 0 means a country is perceived as highly corrupt and 100 means it is
perceived as very clean. A country is ranked relative to other countries in the index. The higher is the perceived
level of public sector corruption, the higher is the level of undeclared work (rs=-.710***). Countries such as Bulgaria,
Romania and Hungary with perceptions of greater public sector corruption have higher levels of undeclared work,
whilst countries where the public sector is perceived as rather clean (e.g., Sweden, Finland, Denmark) have
relatively lower levels of undeclared work.
Figure
10.
Relationship
between undeclared work and
Corruption Perceptions Index
(CPI), 2019
Source: own calculations based on data
from Transparency International Corruption Perceptions Index (CPI)
28
5.3
Level of development and state intervention in work and
welfare
Starting with the level of development, Figure 11 displays the cross-national variations in the size of the undeclared
economy (using the LIM estimates of undeclared work as a percentage of total labour inputs in the private sector)
and cross-national variations in GDP per capita (current prices, euro per capita). This reveals a strong correlation
between cross-national variations in the level of GDP per capita (current prices, euro per capita) and cross-national
variations in the prevalence of undeclared work, measured by Spearman’s rank correlation coefficient (rs= -.725***).
The greater the level of GDP per capita (current prices, euro per capita), the lower is the prevalence of undeclared
work. Countries such as Bulgaria and Romania with relatively lower levels of GDP per capita have higher levels of
undeclared work, whilst countries with relatively higher levels of GDP per capita (e.g., Denmark, Ireland,
Luxembourg) have relatively lower levels of undeclared work.
The same strong relationship is found for other alternative measures of “development” such as the Social Progress
Index (rs=-.761***) and Human Development Index (rs=-.706***).
Figure 11. Relationship between
undeclared work and GDP (current
prices, euro per capita), 2019
Note: To avoid excessive influence, the GDP
of Luxembourg was capped at 80000 (from
100360) in the analyses presented here.
Source: own calculations based on data from
Eurostat database
Turning to the association between state intervention in work and welfare and the prevalence of undeclared work,
Figure 12 and Figure 13 display the results for two structural determinants, namely active labour market policy
expenditure and the severe material deprivation rate. As Figure 12 reveals, Member States in which there are higher
levels of state intervention in work and welfare have relatively lower levels of undeclared work. The finding is that
the higher the labour market policy expenditure (as a % of GDP) on labour market policy (LMP) interventions
covering the range of financial and practical supports offered by governments to people who are unemployed or
otherwise disadvantaged in the labour market, the lower is the level of undeclared work (rs=-.515***).
29
Figure 12. Relationship between
undeclared work and Labour
market policy (LMP) expenditure
(% of GDP), 2019
Source: own calculations based on data from
European Commission (Directorate-General
for Employment, Social Affairs and Inclusion)
There is also a close association between the prevalence of undeclared work and the severe material deprivation
rate (as per cent of total population). The finding is that the higher the per cent of the total population facing severe
material deprivation, the higher is the level of undeclared work (rs =.701***). Countries such as Bulgaria, Greece
and Romania with relatively high per cent of the total population facing severe material deprivation have higher
levels of undeclared work, whilst countries with relatively low percentage of people from total population facing
severe material deprivation (e.g., Finland, Sweden, Luxembourg) have relatively lower levels of undeclared work
(Figure 13).
Figure 13. Relationship between
undeclared work and severe
material deprivation rate (% of
total population), 2019
Source: own calculations based on data
from Eurostat database
5.4
Quality and power of formal institutions
In terms of the formal institution power in formulating laws and regulations and making people abide by them, many
structural conditions are associated with the prevalence of undeclared work. As Figure 14 and Figure 15 display,
both the rule of law and regulatory quality are associated with the level of undeclared work. The rule of law reflects
perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the
30
quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and
violence. The finding is that a strong correlation exists between cross- national variations in the perception of the
rule of law and cross-national variations in the prevalence of undeclared work, measured by Spearman’s rank
correlation coefficient (rs= -.704***). Member States such as Bulgaria and Greece where the rule of law is perceived
as weaker have higher levels of undeclared work than Member States where the rule of law is perceived as stronger,
such as Denmark, Austria and Finland.
Figure 14. Relationship between
undeclared work and Rule of law,
2019
Source: own calculations based on data from
World Bank, The Worldwide Governance
Indicators (WGI)
There is also a close association between the prevalence of undeclared work and regulatory quality, reflecting the
perceptions of the ability of government to formulate and implement sound policies and regulations that permit and
promote private sector development. The stronger the perceived governance performance in terms of regulatory
quality, the lower the level of undeclared work (rs= -.691***). Member States such as Romania, Greece and Bulgaria
where governance performance in terms of regulatory quality is perceived as weaker have higher levels of
undeclared work than Member States where the governance performance is perceived as stronger, such as
Sweden, Finland and the Netherlands (Figure 15).
31
Figure
15.
Relationship
between undeclared work
and Regulatory quality, 2019
Source: own calculations based on
data from World Bank, The Worldwide
Governance Indicators (WGI)
The prevalence of undeclared work is also strongly correlated with the voice and accountability indicator which
reflects perceptions of the extent to which citizens feel able to participate in selecting their government, their freedom
of expression, freedom of association, and the perceived existence of a free media (rs= -.690***). As Figure 16
displays, Member States such as Hungary, Bulgaria, Croatia and Romania where the voice and accountability
indicator is relatively lower have higher levels of undeclared work, whilst the opposite is the case for the Member
States such as Denmark, Sweden and Finland where voice and accountability indicator have relatively higher
values.
Figure 16. Relationship between
undeclared work and voice and
accountability, 2019
Source: own calculations based on data
from World Bank, The Worldwide
Governance Indicators (WGI)
To further compound the view that the power of the formal institutions is an important determinant of undeclared
work, and how perceptions of public sector corruption can lead to higher levels of undeclared work by discouraging
citizens from adhering the formal rules, the relationship between trust in authorities and level of undeclared work is
here analysed. As Figure 17 displays, the level of trust in parliament is significantly correlated with the prevalence
of undeclared work (moderate association, rs=-.502***). Member States such as Bulgaria, Croatia and Lithuania
32
where trust in parliament is lower have higher levels of undeclared work, whilst Member States such as the
Netherlands, Denmark and Sweden where trust in parliament is higher have lower levels of undeclared work.
Figure 17. Relationship between
undeclared work and trust in
Parliament (% tend to trust), 2019
Source: own calculations based on data from
Standard Eurobarometer 92 - Public opinion in
the European Union, Autumn 2019
Similar results are obtained when analysing the level of trust in the institutions charged with tackling undeclared
work, namely labour inspectorates and the levels of undeclared work (weak but significant association, rs=-.377*).
As Figure 18 displays, in countries where there is greater trust in labour inspectorates, the levels of undeclared
work are lower.
Figure
18.
Relationship
between undeclared work and
trust in labour inspectorate (%
tend to trust), 2019
Source: own calculations based on data
from Special Eurobarometer 498 Undeclared Work in the European
Union, 2019
5.5
Instability and uncertainty of formal institutions
Moving to the formal institutional instability and uncertainty, the Democracy Index is analysed. The Economist
Intelligence Unit`s Democracy Index is based on five categories (electoral process and pluralism, civil liberties, the
functioning of government, political participation, and political culture) ranging up to 10 which represents full
33
democracy. As Figure 19 displays, the finding is that a strong correlation exists between the Democracy Index value
and the prevalence of undeclared work (rs=-.751***). Member States such as Romania, Croatia and Poland where
the Democracy Index scores lower have higher levels of undeclared work, whilst Member States such as Ireland,
Finland and Sweden where the scores are higher have lower levels of undeclared work.
Figure 19. Relationship between
undeclared work and Democracy
Index, 2019
Source: own calculations based on data from
The Economist Intelligence Unit - Democracy
Index
5.6
Lack of adherence to the formal rules and trust in peers
It is also the case that the level of undeclared work is closely associated with structural conditions reflecting the
informal institutions (values, beliefs and norms) of the Member States` citizens. There is a strong association
between the level of undeclared work and the level of social capital, measuring the strength of personal and social
relationships, social norms, and civic participation in a country. The greater the level of social capital, the lower is
the level of undeclared work (rs=-.638***), as displayed in Figure 20.
Figure 20. Relationship between
undeclared work and social capital,
2019
Source: own calculations based on data from
Legatum Institute - The Legatum Prosperity
Index (Social Capital pillar)
34
There is also a close association between the level of undeclared work and the acceptability of different forms of
tax non-compliant behaviours. For example, as Figure 21 displays, the extent to which undeclared work undertaken
by firms for other firms is found acceptable by citizens is significantly correlated with the level of undeclared work
(rs=-.422**). The higher the proportion of citizens viewing such non-compliant behaviour as totally unacceptable,
the lower is the prevalence of undeclared work.
Figure 21. Relationship between
undeclared work and acceptability
of undeclared work (% total
‘unacceptable’) - Undeclared work
by firm for firm, 2019
Source: own calculations based on data from
Special Eurobarometer 498 - Undeclared Work
in the European Union, 2019
Finally, a significant association also exists between the level of undeclared work and citizens’ perceptions of the
share of undeclared work existing in their society (also known as the level of horizontal trust) (rs=-.574***), which is
a measure of their trust in their peers to be compliant. The higher the share of population in a country that estimates
that a low number of their peers engage in undeclared work (i.e., the estimated share of undeclared work in the
country as less or equal to 10 %), the lower is the prevalence of undeclared work. Figure 22 reveals that Member
States with higher proportions of the population who estimate that the share of undeclared work in their country is
no more than 10 %, have lower levels of undeclared work.
Figure 22. Relationship between
undeclared work and Estimated
share of undeclared work, 2019
Source: own calculations based on data from
Special Eurobarometer 498 - Undeclared Work
in the European Union, 2019
35
In sum, the structural economic and social conditions associated with lower levels of undeclared work have been
here highlighted. Undeclared work is lower in Member States with:
i.
Higher government effectiveness and lower perceived levels of corruption.
ii.
Higher levels of development (whether measured in terms of GDP per capita, the Human Development Index
or Social Progress Index) and greater levels of state intervention in work and welfare (to enhance workers`
rights and labour productivity, investments in research and development and implementing measures for
reducing poverty and inequalities).
iii.
Higher quality more powerful formal institutions (higher reliability of policy services, higher judicial
independence, stronger rule of law, stronger voice and accountability, positive perception towards the
regulatory quality and higher trust in the state institutions).
iv.
Lower levels of instability and uncertainty in formal institutions (better transparency in policymaking and
reduced perception of political instability).
v.
Greater symmetry between the norms, values and beliefs of citizens, workers, employers and businesses
and the formal rules (sometimes termed “vertical trust” and measured by the level of tax compliance, and
the acceptability of undeclared work) and a higher trust in peers to adhere to the formal rules (sometimes
termed “horizontal trust” and measured by personally knowing people engaged in undeclared work and their
estimates of the share of undeclared work).
36
References
Eurostat
(2013).
European
System
of
Accounts
ESA
2010.
Available
from:
https://ec.europa.eu/eurostat/documents/3859598/5925693/KS-02-13-269-EN.PDF.pdf/44cd9d01bc64-40e5-bd40-d17df0c69334?t=1414781932000
Eurostat (2021). EU Labour Force Survey Database User Guide (data up to reference year 2020). Available from:
https://ec.europa.eu/eurostat/documents/1978984/6037342/EULFS-Database-UserGuide.pdf
Eurostat
(2022a).
Structural
business
statistics
–
https://ec.europa.eu/eurostat/web/structural-business-statistics
Overview.
Available
from:
Eurostat (2022b). Structural business statistics (sbs). Reference Metadata in Euro SDMX Metadata Structure
(ESMS). Available from: https://ec.europa.eu/eurostat/cache/metadata/en/sbs_esms.htm
Eurostat (2022c). Residence permits – statistics on authorisations to reside and work. Available from:
https://ec.europa.eu/eurostat/statisticsexplained/index.php?title=Residence_permits_%E2%80%93_statistics_on_authorisations_to_reside
_and_work
Eurostat (2022d). Third-country nationals found to be illegally present - annual data. Available from:
https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=migr_eipre&lang=en
Eurofound (2013). Tackling Undeclared Work in 27 European Union Member States and Norway: Approaches and
Measures since 2008. Dublin: Eurofound.
European Commission (2007). Stepping Up the Fight Against Undeclared Work, COM(2007) 628. Brussels:
European Commission.
ILO (1982). Resolution concerning statistics of the economically active population, employment, unemployment and
underemployment, adopted by the Thirteenth International Conference of Labour Statisticians,
International Labour Organisation. Available from: https://www.ilo.org/wcmsp5/groups/public/--dgreports/---stat/documents/normativeinstrument/wcms_087481.pdf
ISTAT (2016). L’economia non osservata nei conti nazionali. Anni 2011-2014. Istituto Nazionale di Statistica.
https://www4.istat.it/it/files/2016/10/Economia-nonAvailable
from:
osservata_2014.pdf?title=Economia+non+osservata+nei+conti+nazionali+-+14%2Fott%2F2016++Testo+integrale+e+nota+metodologica.pdf
OECD (2002). Measuring the non-observed Economy. Organisation for Economic Co-Operation and Development.
Available from: https://www.oecd.org/std/na/1963116.pdf
Williams, C.C. and Schneider, F. (2016). Measuring the Global Shadow Economy: The Prevalence of Informal Work
and Labour. Edward Elgar Publishing Ltd.
Williams, C.C. and Horodnic, I. (2020) Horizon scanning: early warning signals of future trends in undeclared work,
European Platform tackling undeclared work, Brussels.
Williams, C.C., Bejakovic, P., Mikulic, D., Franic, J., Kedir, A. and Horodnic, I.A. (2017) An evaluation of the scale
of undeclared work in the European Union and its structural determinants: estimates using the Labour
Input Method, European Commission, Brussels.
37
Annex 1: Data sources of the structural conditions and results
Table A1. Macro Indicators used for bivariate correlation (description and source of data)
Indicator
Description
Source
A. FORMAL INSTITUTIONS
I. FORMAL INSTITUTIONAL RESOURCE MISALLOCATIONS AND INEFFICIENCIES
1. Level of modernisation of government
Government effectiveness
Government effectiveness captures perceptions of the quality of public
services, the quality of the civil service and the degree of its independence
from political pressures, the quality of policy formulation and implementation,
and the credibility of the government's commitment to such policies. Estimate
of governance ranges from approximately -2.5 (weak) to 2.5 (strong)
governance performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
2. Formal institutions acting in a corrupt manner
Corruption
Index (CPI)
Control
(index)
of
Perceptions
corruption
Transparency International’s Corruption Perceptions Index ranks countries
and territories based on how corrupt their public sector is perceived to be. A
country or territory’s score indicates the perceived level of public sector
corruption on a scale of 0 - 100, where 0 means that a country is perceived
as highly corrupt and 100 means it is perceived as very clean.
Transparency International - Corruption Perceptions Index (CPI),
2019. Available from: https://www.transparency.org/research/cpi
Reflects perceptions of the extent to which public power is exercised for
private gain, including both petty and grand forms of corruption, as well as
‘capture’ of the state by elites and private interests. Estimate of governance
ranges from approximately -2.5 (weak) to 2.5 (strong) governance
performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
II. FORMAL INSTITUTIONAL VOIDS AND WEAKNESSES
1. Modernisation Explanation - Level of ‘development’
GDP (current prices, euro
per capita)
Gross domestic product at market prices - current prices, euro per capita.
Eurostat - Main GDP aggregates per capita, 2019. Available from:
https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_
10_pc&lang=en
Human Development Index
(HDI)
The Human Development Index (HDI) is a summary measure of average
achievement in key dimensions of human development: a long and healthy
life, being knowledgeable and have a decent standard of living. The HDI is
the geometric mean of normalized indices for each of the three dimensions.
United Nations Development Programme, Human Development
Reports, 2009-2018. Available from: http://hdr.undp.org/en/data
38
HDI is ranked on a scale from 0 to 1.0, with 1.0 being the highest human
development.
The index doesn’t measure people’s happiness or life satisfaction, focusing
instead on actual life outcomes in areas from shelter and nutrition to rights
and education. This exclusive focus on measurable outcomes makes the
index a useful policy tool that tracks changes in society over time. The Social
Progress Index score is an average across scores for the three broad
dimensions: Basic Human Needs, Foundations of Wellbeing, and
Opportunity (maximum possible score: 100). Within each dimension, there
are four components that further divide the indicators into thematic
categories. This diverse selection of indicators allows for granular analysis of
the specific underpinnings of social progress in each country, while the broad
categories of the index framework helps to better understand global and
regional trends.
Social Progress Imperative - Social Progress Index, 2019. Available
from: https://www.socialprogress.org/
Response to the survey question ‘In your country, how burdensome is it for
companies to comply with public administration’s requirements (e.g., permits,
regulations, reporting)?’ (1 = extremely burdensome; 7 = not burdensome at
all).
World Economic Forum – Global Competitiveness Reports (The
Executive
Opinion
Survey),
2019.
Available
from:
https://www.weforum.org/reports
Workers’ rights
Score adapted from the ITUC Global Rights Index, which measures the level
of protection of internationally recognized core labour standards. The scale
of this indicator ranges from 0 (no protection) to 100 (high protection).
Dimensions of labour protection include civil rights, the right to bargain
collectively, the right to strike, the right to associate freely, and access to due
process rights. The indicator does not consider firing regulations.
World Economic Forum – Global Competitiveness Reports (The
Executive
Opinion
Survey),
2019.
Available
from:
https://www.weforum.org/reports
Business flexibility
Sub-index within Fraser Institute Economic Freedom Index, measuring the
degree of economic freedom present in one of the five major areas, namely
Regulation. When regulations restrict entry into markets and interfere with
the freedom to engage in voluntary exchange, they reduce economic
freedom. This major area of the index focuses on regulatory restraints that
limit the freedom of exchange in credit, labor, and product markets. Measure:
score out of 10.
Fraser Institute – Economic Freedom Index, 2019. Available from:
https://www.fraserinstitute.org/economicfreedom/dataset?geozone=world&year=2017&page=dataset&minyear=2&max-year=0&filter=0
Expense of government (%
of GDP)
This indicator measures the government size (proxy of the degree of
intervention). Expense is cash payments for operating activities of the
government in providing goods and services. It includes compensation of
employees (such as wages and salaries), interest and subsidies, grants,
social benefits, and other expenses such as rent and dividends.
World Bank - World Development Indicators, 2019. Available from:
https://data.worldbank.org/indicator/GC.XPN.TOTL.GD.ZS
Research & Development
expenditure (% of GDP)
Research and experimental development (R&D, all sectors) comprise
creative work undertaken on a systematic basis in order to increase the stock
of knowledge, including knowledge of man, culture and society, and the use
Eurostat - R&D expenditure, 2019. Available from:
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=rd_e_ge
rdtot&lang=en
Social Progress Index (SPI)
2. State intervention
Burden of
regulation
government
39
of this stock of knowledge to devise new applications. R&D expenditure are
shown as a percentage of GDP (R&D intensity).
Tax revenue (% of GDP)
Tax revenue refers to compulsory transfers to the central government for
public purposes. Certain compulsory transfers such as fines, penalties, and
most social security contributions are excluded. Refunds and corrections of
erroneously collected tax revenue are treated as negative revenue.
World Bank - World Development Indicators, 2019. Available from:
https://data.worldbank.org/indicator/GC.TAX.TOTL.GD.ZS
Social contributions (% of
revenue)
Social contributions include social security contributions by employees,
employers, and self-employed individuals, and other contributions whose
source cannot be determined. They also include actual or imputed
contributions to social insurance schemes operated by governments.
World Bank - World Development Indicators, 2019. Available from:
https://data.worldbank.org/indicator/GC.REV.SOCL.ZS
Impact of social transfers
on poverty reduction (%)
At risk of poverty rate before social transfers (pensions excluded from social
transfers) (cut-off point: 60 % of median equivalised income after social
transfers). Reduction in percentage of the risk of poverty rate, due to social
transfers (calculated comparing at-risk-of poverty rates before social
transfers with those after transfers; pensions are not considered as social
transfers in these calculations). The indicator is based on the EU-SILC
(statistics on income, social inclusion and living conditions).
Eurostat - Impact of social transfers (excluding pensions) on
poverty
reduction,
2019.
Available
from:
https://ec.europa.eu/eurostat/databrowser/view/tespm050/default/t
able?lang=en
Labour market policy (LMP)
expenditure (% of GDP)
Labour market policy (LMP) interventions cover the range of financial and
practical supports offered by governments to people who are unemployed or
otherwise disadvantaged in the labour market.
European Commission (Directorate-General for Employment,
Social Affairs and Inclusion) - LMP expenditure. Available from:
https://webgate.ec.europa.eu/empl/redisstat/databrowser/view/LM
P_EXPSUMM/default/table?category=lmp_expend
People at risk of poverty or
social exclusion (% total
population)
This indicator corresponds to the sum of persons who are: at risk of poverty
after social transfers, severely materially deprived or living in households with
very low work intensity. Persons are counted only once even if they are
affected by more than one of these phenomena.
Severe material deprivation
rate (% total population)
The material deprivation rate is an indicator in EU-SILC that expresses the
inability to afford some items considered by most people to be desirable or
even necessary to lead an adequate life. The indicator distinguishes between
individuals who cannot afford a certain good or service, and those who do
not have this good or service for another reason, e.g. because they do not
want or do not need it. Severe material deprivation rate is defined as the
enforced inability to pay for at least four of the deprivation items.
Inequality
of
income
distribution (income quintile
share ratio S80/S20)
The ratio of total income received by the 20 % of the population with the
highest income (top quintile) to that received by the 20 % of the population
with the lowest income (lowest quintile). Income must be understood as
equivalised disposable income.
Eurostat - Income quintile share ratio S80/S20, 2019. Available
from:
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_di11
&lang=en
Gini coefficient
The Gini coefficient (scale from 0 to 100) is defined as the relationship of
cumulative shares of the population arranged according to the level of
Eurostat - Gini coefficient of equivalised disposable income, 2019.
Available
from:
40
Eurostat - People at risk of poverty or social exclusion by age and
sex, 2019. Available from:
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_peps
01&lang=en
Eurostat - Severe material deprivation rate by age and sex, 2019.
Available
from:
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_mdd
d11&lang=en
Labour
productivity
(percentage change on
previous period)
equivalised disposable income, to the cumulative share of the equivalised
total disposable income received by them.
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_di12
&lang=en
Real labour productivity per person employed. The labour productivity =
GDP/ETO with GDP = Gross domestic product; ETO = Total employment, all
industries, in persons. Data are expressed as percentage change comparing
year Y with year Y-3.
Eurostat - Labour productivity and unit labour costs. 2019. Available
from:
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_1
0_lp_ulc&lang=en
III. FORMAL INSTITUTIONAL POWERLESSNESS
Reliability of police services
Response to the survey question ‘In your country, to what extent can police
services be relied upon to enforce law and order?’ (1 = not at all; 7 = to a
great extent).
World Economic Forum – Global Competitiveness Reports (The
Executive
Opinion
Survey),
2019.
Available
from:
https://www.weforum.org/reports
Judicial independence
Response to the survey question ‘In your country, how independent is the
judicial system from influences of the government, individuals, or
companies?’ (1 = not independent at all; 7 = entirely independent).
World Economic Forum – Global Competitiveness Reports (The
Executive
Opinion
Survey),
2019.
Available
from:
https://www.weforum.org/reports
Rule of law
Reflects perceptions of the extent to which agents have confidence in and
abide by the rules of society, and in particular the quality of contract
enforcement, property rights, the police, and the courts, as well as the
likelihood of crime and violence. Estimate of governance ranges from
approximately -2.5 (weak) to 2.5 (strong) governance performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
Regulatory quality
Reflects perceptions of the ability of the government to formulate and
implement sound policies and regulations that permit and promote private
sector development. Estimate of governance ranges from approximately -2.5
(weak) to 2.5 (strong) governance performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
Voice and accountability
(index)
Reflects perceptions of the extent to which a country’s citizens are able to
participate in selecting their government, as well as freedom of expression,
freedom of association, and a free media. Estimate of governance ranges
from approximately -2.5 (weak) to 2.5 (strong) governance performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
Trust in Government (%
tend to trust)
Standard Eurobarometer question: trust in the (nationality) Government (%
tend to trust).
Standard Eurobarometer surveys 92 - Public opinion in the
European
Union,
Autumn
2019.
Available
from:
https://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/Surv
ey/index#p=1&instruments=STANDARD
Trust in Parliament (% tend
to trust)
Standard Eurobarometer question: trust in the (nationality) Parliament (%
tend to trust).
Standard Eurobarometer survey 92 - Public opinion in the European
Union,
Autumn
2019.
Available
from:
https://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/Surv
ey/index#p=1&instruments=STANDARD
IV. FORMAL INSTITUTIONAL INSTABILITY AND UNCERTAINTY
Democracy Index
The EIU Democracy Index is based on five categories: electoral process and
pluralism; civil liberties; the functioning of government; political participation;
41
The Economist Intelligence Unit - Democracy Index, 2019.
Available from: https://www.eiu.com/topic/democracy-index
and political culture. Measure: Index score out of 10, 10 being best (full
democracy).
Political stability
Political stability and absence of violence/terrorism measures perceptions of
the likelihood of political instability and/or politically-motivated violence,
including terrorism. Estimate of governance ranges from approximately -2.5
(weak) to 2.5 (strong) governance performance.
World Bank - The Worldwide Governance Indicators (WGI), 2019.
Available from: https://info.worldbank.org/governance/wgi/
B. INFORMAL INSTITUTIONS
Social capital
Score on the Social Capital pillar of the Legatum Prosperity Index. The Social
Capital pillar measures the strength of personal and social relationships,
social norms, and civic participation in a country. The scale ranges from 0
(low) to 100 (high). The index is used by World Economic Forum in Global
Competitiveness Reports.
Legatum Institute - The Legatum Prosperity Index (Social Capital
pillar),
2019.
Available
from:
https://www.prosperity.com/about/resources
Tax compliance (index)
Sub-component within Fraser Institute Economic Freedom Index, measuring
the cost of tax compliance. Measure: score out of 10.
Fraser Institute – Economic Freedom Index, 2019. Available from:
https://www.fraserinstitute.org/economicfreedom/dataset?geozone=world&year=2017&page=dataset&minyear=2&max-year=0&filter=0
The acceptability of various behaviours across respondents (1 - ‘absolutely
unacceptable’ and 10 – ‘absolutely acceptable’).
Acceptability of undeclared
work
(%
total
‘unacceptable’)
Personally know people
engaged in undeclared
work (% ‘yes’)
Percent of total ‘unacceptable’ (1-4):
Firm hires worker on undeclared basis (A firm hires a private person and
all or a part of the salary paid to him or her is not officially declared)
Undeclared work by firm for firm (A firm is hired by another firm and it
does not declare its activity to tax or social security authorities)
Undeclared work by individual for private household (A private person is
hired by a private household and he or she does not declare the
payment received to tax or social security authorities although it
should be reported)
Undeclared work by firm for private household (A firm is hired by a
private household for work and does not declare the payment
received to tax or social security authorities)
Someone partially of completely conceals their income (A private person
or self-employed person evades taxes by not declaring all or part of
their income)
Response to the survey question ‘Do you personally know any people who
work without declaring all or part of their income to tax or social security
authorities?’ (% total ‘yes’).
42
Special Eurobarometer surveys 498 - Undeclared Work in the
European
Union,
2019.
Available
from:
https://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/Surv
ey/index#p=5&instruments=SPECIAL
Special Eurobarometer surveys 498 - Undeclared Work in the
European
Union,
2019.
Available
from:
https://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/Surv
ey/index#p=5&instruments=SPECIAL
Estimated
share
of
undeclared work: ≤ 10 %
(%)
Response to the survey question ‘Approximately, what percentage of the
population in (OUR COUNTRY) do you think work without declaring some or
all of their income to tax or social security authorities?’ (% ≤ 10 %).
Special Eurobarometer surveys 498 - Undeclared Work in the
European
Union,
2019.
Available
from:
https://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/Surv
ey/index#p=5&instruments=SPECIAL
Source: updated from Williams C.C. and Horodnic I.A. (2020). Horizon scanning: early warning signals of future trends in undeclared work. European Platform tackling undeclared
work.
43
Table A2. Bivariate correlation between extent of undeclared work and structural conditions in
the European Union: Spearman’s rank correlation coefficient
Undeclared work, % of labour input in
the corporate sector
Macro Indicators
rs
Sig.
A. FORMAL INSTITUTIONS
I. FORMAL INSTITUTIONAL RESOURCE MISALLOCATIONS AND
INEFFICIENCIES
1. Level of modernisation of government
-0.699
***
Corruption Perceptions Index (CPI) (0-100 (very clean))
-0.710
***
Control of corruption (-2.5 to 2.5 (strong performance))
-0.695
***
GDP (current prices, euro per capita)
-0.725
***
Human Development Index (HDI) (0-1 (highest development))
-0.706
***
Social Progress Index (SPI) (0-100 (high))
-0.761
***
Burden of government regulation (1-7 (best))
-0.373
*
Workers’ rights (0-100 (best))
-0.460
**
Business flexibility index (0-10 (high))
-0.178
Government effectiveness (-2.5 to 2.5 (strong performance))
2. Formal institutions acting in a corrupt manner
II. FORMAL INSTITUTIONAL VOIDS AND WEAKNESSES
1. Modernisation Explanation - Level of ‘development’
2. State intervention
0.094
Expense of government (% of GDP)
Research & Development expenditure (% of GDP)
-0.578
Tax revenue (% of GDP)
-0.105
***
Social contributions (% of revenue)
-0.098
Impact of social transfers on poverty reduction (%)
-0.446
**
Labour market policy (LMP) expenditure (% of GDP)
-0.515
***
People at risk of poverty/social exclusion (% total population)
0.516
***
Severe material deprivation rate (% of total population)
0.701
***
Inequality of income distribution (income quintile share ratio)
0.370
*
Gini coefficient (0-100 (perfect inequality))
0.364
*
Labour productivity (% change on previous period (Y-3))
0.550
***
Reliability of police services (1-7 (best))
-0.610
***
Judicial independence (1-7 (best))
-0.636
***
Rule of law (-2.5 to 2.5 (strong performance))
-0.704
***
Regulatory quality (-2.5 to 2.5 (strong performance))
-0.691
***
Voice and accountability (-2.5 to 2.5 (strong performance))
-0.690
***
Trust in Government (% tend to trust)
-0.420
**
Trust in Parliament (% tend to trust)
-0.502
***
Trust in Tax and Social Security authorities (% tend to trust)
-0.367
*
Trust in Labour inspectorate (% tend to trust)
-0.377
*
Democracy Index (0-10 (full democracy))
-0.751
***
Political stability (-2.5 to 2.5 (strong performance))
-0.392
**
III. FORMAL INSTITUTIONAL POWERLESSNESS
IV. FORMAL INSTITUTIONAL INSTABILITY AND UNCERTAINTY
44
Undeclared work, % of labour input in
the corporate sector
Macro Indicators
rs
Sig.
B. INFORMAL INSTITUTIONS
Social capital (0-100 (high))
-0.638
***
Tax compliance (0-10 (high))
-0.326
*
Acceptability of undeclared work (% total ‘unacceptable’)
Firm hires worker on undeclared basis
-0.267
Undeclared work by firm for firm
-0.422
Undeclared work by individual for private household
-0.059
Undeclared work by firm for private household
-0.396
**
Someone partially of completely conceals their income
-0.346
*
Personally know people engaged in undeclared work (% ‘yes’)
-0.096
Estimated share of undeclared work: less than 10% (%)
-0.574
Note: Significant at: *** p<0.01, ** p<0.05, * p<0.1
45
**
***
Related papers
El hábito no hace al monje. Materia y proceso en la arquitectura vernácula
Vestir la Arquitectura. XXII Congreso Nacional de Historia del Arte, 2019
Electronic voting: algorithmic and implementation issues
36th Annual Hawaii International Conference on System Sciences, 2003. Proceedings of the, 2003
Análisis crítico del aporte de Peter Drucker a la Ciencia Administrativa
Economía & Negocios
Religious coping and resilience among students of medical sciences
European Journal of Public Health, 2017
Manifestações orais associadas à infecção por COVID-19
Research, Society and Development
A Comparative Proteomic Study of Thermobifida Cellulosilytica TB100T Secretome Grown on Carboxymethylcellulose and Rice Straw
The Open Biotechnology Journal, 2020
Molecular and conventional identification of Malassezia spp. in Malassezia spp.-associated dermatoses
Journal of the Egyptian Womenʼs Dermatologic Society, 2015
Enhancing Wind Energy Harvesting Using Passive Turbulence Control Devices
Applied Sciences, 2019
Bacteremia in Childhood Life-Threatening Infections in Urban Gambia: EUCLIDS in West Africa
Open Forum Infectious Diseases, 2019
Stroke, spinal cord ischemia and upper limb ischemia in patients undergoing TEVAR with coverage of the left subclavian artery: a case series study
Egyptian Journal of Radiology and Nuclear Medicine, 2021