The size of labor market crime in Norway

In this report, we have estimated the size of labor market crime in Norway between 1973 and 2015. Labor market crime is characterized by profit-driven crime in the labor market. The phenomenon is complex and include various offenses. Actions related to labor market crime include violations of the pay and working conditions, tax evasion and social security fraud.

Both employers and employees may have incentives to commit such actions, but the society will always lose. Labor market crime results in loss of income for law-abiding businesses, deterioration of individual industries, and unfair competitive conditions for law-abiding businesses.

There is an increasing awareness that labor market crime represents a major social problem. Both former and present governments have compiled strategies, and implemented several measures to combat such crime. Nevertheless, there is a lack of information on of the scope. Better knowledge about and understanding of the scope of this form of crime is necessary to implement appropriate measures.

Two estimation methods are combined to calculate the development of labor market crime

Criminal acts are hidden, and hence are difficult to identify. It is therefore necessary to use special estimation methods, to estimate the scope. In general, these methods can be divided into direct approaches (when one has access to crime data from the control agencies) and indirect approaches (when one does not have such data). However, there will always be considerable uncertainty attached to this type of analysis.

In this project, we have been granted access to rarely available data for revealed tax evasion. Therefore, we can utilize both direct and indirect approaches. The direct approach gives us an estimate of the level of labor marked crime in Norway as a share of GDP, while the indirect gives us the development over time.

In our direct approach, we utilize control data for Norwegian businesses to calculate the value of labor market crime related to taxation and fraud of social security. The size of tax evasion is calculated from predicted probabilities for labor marked crime for all businesses in Norway. Combined with social security fraud figures from NAV, it gives us an estimate of the level of labor market crime, measured as a share of GDP.

Then we have used an indirect approach called MIMIC (Multiple Indicators Multiple Causes) to estimate changes in the size of labor market crime over a longer period (1973 to 2015). In this model, we use cause and indicator variables, which are assumed to be highly correlated with labor market crime, to say something about the changes of labor market crime over time.

The size is estimated based on tax evasion and hidden created value

We use two different measures of tax-related labor marked crime. The first one calculates the amount of tax evasion. The second one calculates the size of created value associated with illegal activities that are hidden from the outside world. Calculations of tax evasions have been carried out previously, but not related to labor market crime. The hidden created value, as far as we know, has never been estimated. We estimate hidden created value using data for unreported income and social security contributions that were revealed by controls. We believe that this estimate captures more of the activities related to tax evasion. This is because created value captures both black labor and tax evasion, as well as the fact these businesses often also commit other offenses (for example breach of the Working Environment Act). Therefore, we find that the measure based on created value provides significantly higher estimates, than when based on tax evasion only.

To estimate the scope of social security fraud, we have utilized aggregate statistics from NAV and existing literature. To create a level of labor marked crime as a share of GDP, we add the estimate of the average social security fraud to the results of tax-related labor market crime.

Tax evasion amounted to around NOK 12 billion in 2015, while hidden created value amounted to NOK 28 billion

The figure below illustrates our main estimates for the scope of labor market crime in Norway. The different paths represent the estimates based on tax evasion and hidden created value, both as part of mainland GDP. Estimates for 2010 to 2014 are based on microdata and have a greater yearly volatility relative to the earlier years, which are based on predictions from the indirect approach.

For 2015, we estimate that the share of labor market crime measured with hidden created value, was 1.2 percent of mainland GDP, or about NOK 28 billion. This includes both hidden income and social security contributions, in addition to an estimate of relevant social security fraud. If we consider labor market crime as tax evasion and social security fraud, the proportion is 0.5 percent, or about NOK 12 billion.

The figure exhibit a declining trend in the 1970s and 1980s. This reflects a period of a steadily increasing proportion of wage earners, which in our estimates contribute to less labor market crime. The figure show that labor market crime increased in 1988, and remained high for several years. This was during and right after the Norwegian banking crisis, and the collapse of the housing market. Many enterprises and individuals experienced financial difficulties and unemployment rose. The relatively steep decline that begins in 1993 coincides with a period of strong growth in the Norwegian economy, unemployment slowed and the proportion of wage earners increased more markedly. Labor market crime seems to fall when economic growth is strong and vice versa. Higher economic growth is correlated with more opportunities in the legal market.

The growth in labor market crime in the 2000s coincides with a period of strong labor immigration. High demand for labor, and far higher wage levels than in Eastern Europe, made Norway a very attractive destination for job seekers from the east. The supply of relatively low-skilled workers has probably allowed more labor market crime, as labor immigrants are easier to exploit, than Norwegian workers.

Towards the end of the 2000s the growth in labor market crime slows down. This can indicate that the implemented measures to reduce labor market crime and social dumping has had an effect. This includes measures such as the generalization of collective agreements (allmenngjøring av tariffavtaler), regulations on pay and working conditions in public procurement processes, infringement fees, information and duty obligations, introduction of HSE cards and approval schemes, equal treatment rules for hired labor and enhanced collaboration and information sharing between government agencies. In addition, the reduction in oil prices in 2014 led to a slowdown of the Norwegian economy and emigration of many migrant workers. Our estimates of the hidden created value and tax evacuations in 2015 are on the same level as the average for the period from 2009 to 2015. This indicates that the size of labor market crime is no longer increasing.

Robustness and uncertainty

The estimates presented in the figure above are based on median values or predicted probability of labor market crime and tax evasion. We have also calculated the scope based on average values. However, these are characterized by extreme observations, causing the average values to be several times higher than the median values. The main reason is that the distribution of businesses turns out to be highly skewed, both in terms of evasion and predicted probability. This means that most businesses have a low probability of committing labor market crime and evade small amounts, while a few have high probability and evade large amounts. These extreme observations pull the average values up. Thus, there is reason to believe that estimates for evasion and created value based on average values overestimate the size of labor market crime, and that estimates based on median values re closer to reality. Estimates based on average values an therefore be understood as an upper bound.

Nevertheless, using average values s important if one is to compare our labor market crime estimates with estimates for other types of crime, because the average is most commonly used in the literature. Our estimates based on median values provides a more robust estimate, that is, it is less affected by major disclosures that occur a few times, which raise average values. The relatively large difference between the median and average estimates indicates high uncertainty.

We have conducted different specifications of the models, and the results are robust in terms of model selection. However, none of the estimates captures the value of breach of labor and wage conditions, which harm workers and activity in the law-abiding part of business. This indicates that the estimates should be interpreted with caution and seen in conjunction with other surveys, including those who do not analyze the size as a percentage of GDP.

Recommendation

A key issue related to both an assessment of scope, development over time and evaluation of implemented measures aimed at combating labor market crime is the lack of high quality data sources. Access to statistics is necessary to be able to follow the development and size of labor market crime in total and across different parts of the economy. It is therefore very important that the agencies currently working on this issue continue the work of collecting high quality statistics related to labor market crime.

Although our estimate is uncertain, the need to advance and strengthen the measures against labor market crime is significant. In particular, interaction and information sharing between agencies is essential. Labor market crime is a societal issue that affects the responsibilities of several governmental agencies and underlying units. A continued joint effort could contribute to increase the overall impact of the agencies' different instruments.

Scenarioes for future skills demand in Norway

Scenarioes for future skills demand in Norway

This report summarizes four different scenarios with projections of skills demand in Norway. As a part of the project we organized a workshop, in close collaboration with the Ministry of Education and Research, where relevant uncertainties were discussed.

Based on the discussions at the workshop, we decided to focus on the following two uncertainties; Uncertainty related to the future of the “Norwegian worklife model”, and uncertainties related to the level of ambition and commitment of an international climate change agreement.

The costs of inadequately educating asylum seekers and refugees

The costs of inadequately educating asylum seekers and refugees

Economic Analysis Norway (Samfunnsøkonomisk analyse AS) has calculated the socio-economic costs of children and young people who come to Norway as asylum seekers or refugees receiving inadequate primary and secondary education.

On average, quantified factors amount to about NOK 3.8 million in 2015 prices per child. In addition, there are non-monetised effects on the individual’s quality of life, on crime rates and democracy.

Moreover, the analysis looks at conditions in schools that can improve the pupils’ chances of going on to achieve success in education, the labour market and society in general. The analysis also sets out proposals for measures that schools can implement to facilitate the future success of pupils.

Evaluation of «SkatteFUNN» - A feasibility study

Evaluation of «SkatteFUNN» - A feasibility study

This report proposes a comprehensive methodology for evaluating the Norwegian tax deduction scheme; SkatteFUNN. The SkatteFUNN research and development (R&D) tax incentive scheme was introduced in 2002, and is a governmental program designed to stimulate R&D in the business sector. The rationale behind initiating an R&D tax incentive scheme is the overall issue with companies not investing at a socially optimal amount in R&D, because of positive external effects are not fully internalised by the decision makers.  SkatteFUNN should, first and foremost, stimulate R&D investments in the business sector (first order effect), and ideally also lead to innovations (second order effect) and to a more knowledge based economy (third order effect).