Marloes de Graaf-Zijl
Bas Scheer
Jonneke Bolhaar
The Future of Social Protection
What Works for Non-standard Workers?
Chapter 7. Netherlands: non-standard work and social protection
Abstract
Over the last decades, the share of non-standard work in The Netherlands has grown substantially and it is now among the highest of all OECD-countries. Approximately one in three workers work in a non-standard work arrangement (including own-account or temporary work, variable hours contracts and agency work). In the current Dutch institutional setting, firms can save social contributions, severance payments, re-integration obligations and tedious administrative procedures by hiring people as own account workers or, to a lesser extent, through one of the other non-standard work arrangements. By using temporary work arrangements, for example, employers can circumvent employment protection legislation – i.e. severance payments and time-consuming procedures to ask for permission to dismiss – and sickness related payments and re-integration obligations that end upon the end date of the contract. If policy makers wish to reduce the share of non-standard work arrangements, they should aim to reduce incentives to hire workers on non-standard contracts, by reducing differences in taxes and social security coverage between non-standard and standard work.
7.1. Introduction
In recent decades, the share of non-standard work in the Netherlands has grown substantially and it is now among the highest of all OECD countries. One of the potential reasons for this can be found in Dutch institutional settings that allow employers to circumvent various welfare state obligations attached to open-ended contracts by hiring people on various alternative forms of employment. The Netherlands is therefore an interesting case for studying the social protection of workers in these various work arrangements.
Part-time employment is not regarded as non-standard employment in the Dutch context. The share of part-time work is high, but mostly voluntary, and is treated by labour law in exactly the same manner as full-time employment. We therefore restrict this section to other flexible work arrangements, among which we distinguish between temporary contracts, agency work, own-account work and variable hours contracts (including own-account workers on variable hours contracts).
Figure 7.1. Forms of non-standard employment in the Netherlands
7.2. The incidence of various forms of non-standard employment
The number of non-standard jobs has increased steadily over the last 14 years, with only a brief slowdown during the great recession of 2008. However, the number of standard jobs did not increase between 2003 and 2009 and even declined after that period. . This decline came to a halt at the end of 2015, but there has still not been any significant growth in standard work (see Figure 7.2, Panel A). As a result, the share of standard work still declined (see Figure 7.2, Panel B). More recently (first quarter of 2018), with the increase of labour market tightness, the share of standard work has stabilised and even increased slightly. However, it is uncertain how this share will develop in the future, for example when the current economic boom comes to an end.
Figure 7.2. Employment has shifted from standard to non-standard employment relations
Note: non-standard is the sum of variable hour contracts (incl. on-call), temporary work, agency work and own-account work. Non-standard consists only of open-ended employment contracts. Self-employed with personnel are excluded from this graph.
Source: Statistics Netherlands.
Currently, approximately one-third of all workers are employed on a non-standard work arrangement. The share of workers employed in temporary work, agency work and on-call work has risen from 14% in 2003 to 22% in 2016 (23% in the third quarter of 2017, not in the figure), while the share of own-account workers has increased from 8% to 12% (Figure 7.2, Panel B). This increase came entirely at the expense of regular open-ended contracts, since the share of self-employed people with personnel hardly changed (Figure 7.2, panel A).
Among those in non-standard employment, own-account workers are the largest sub-group. Those on variable hours contracts (including on-call work), are currently the second largest sub-group. Their share nearly doubled, from roughly 6% in 2003 to 11% in 2016. The third largest sub-group is formed by those on temporary contracts (without variable hours), whose share increased from 6% in 2003 to 8% in 2016. The smallest sub-group is accounted for by agency work, which has been broadly stable at around 2.5% of total employment, increasing when economic activity expands and decreasing when it contracts (de Graaf-Zijl and Berkhout, 2007[1]).
We can subdivide temporary contracts (without variable hours) further into temporary contracts of one year or more, shorter than one year and temporary contracts that from the start are intended to be converted into open-ended contracts. Even though all temporary contracts can be converted into open-ended contracts, those in the latter category specifically envisage such an outcome, if the employee performs well and the financial situation of the firm allows it. 1 Even though there is no thorough legal basis for this conversion, around one-third of all temporary contracts are agreed on this basis. Their share of all temporary contracts varies over the business cycle, but they became more prevalent over the 2003‑16 period and 3.5% of all workers are currently on such contracts. Temporary contracts of one year or more recorded the largest increase (from 1.3% to 2.1%). The share of temporary contracts for shorter periods (less than one year) has remained broadly stable at around 2%.
7.3. Which groups work on a non-standard work arrangement?
Non-standard work is not equally distributed across the working population. Figure 7.3 presents descriptive statistics based on Dutch Labour Force Survey data. There are considerable differences in the share of non-standard work by gender, age, educational attainment and ethnicity. Differences are especially pronounced between age groups, but differences between educational levels and ethnic groups are substantial as well.
The differences between men and women are relatively small compared with the other differences. The overall shares of men and women in non-standard work are comparable, but the type of work differs. Temporary and on-call work are more prevalent among women, while agency work and own-account work are more prevalent among men.
Temporary work, agency work and on-call work are more prevalent among younger workers2 than among prime-age and older workers. The share of these types of non-standard workers decreases with age. However, for the 65-74 year-old post-retirement group, these shares increase again, especially for on-call work. This is most likely caused by a decrease in open-ended contracts, since many of these contracts end at the statutory retirement age (which until recently was 65). The share of own-account work increases with age: only 5% of 15-24 year-olds work as own-account workers, compared with 16% of 55-64 year-olds. After retirement age, this share increases to almost 50%.
Workers with a low level of education are more likely to work in agency work or on variable hours contracts than those with a higher level of education. There is a particularly marked difference for on-call work, which accounts for 20% of workers with a low level of education but only 5% of workers with a higher level of education. The share of temporary work, however, does not differ much across education levels. Own-account work is more common among highly educated workers (14%) than among those with a lower level of education (11%).
Regarding ethnicity, second generation non-western immigrants are particularly more likely to be temporary, agency or on-call workers and less likely to be own-account workers. This difference is at least partly due to the fact that this group consists of relatively young workers.
Figure 7.3. Nonstandard work most strongly associated with age
Note: Western countries include Europe, north-America and Oceania, Indonesa and Japan. Non-western countries include Africa, Central and South America, and Asia, excluding Indonesia, Japan, Morocco and Turkey (Morocco and Turkey are not shown).
Source: Own calculations based on Dutch LFS-Data.
The descriptive statistics presented above consider only one dimension at a time, and do not take possible correlations between gender, age, education and ethnicity into account. After controlling for correlations with other characteristics by means of regression analysis, age appears to be by far the most important determinant of the probability of working in non-standard employment (Bolhaar, Brouwers and Scheer, 2016[2]). The probability of working in all types of non-standard employment contracts decreases with age, whereas the share of own-account work increases with age. There are differences between men and women, between educational groups and between ethnic groups as well, but these differences are smaller than those between age groups. In some cases these differences appear to be almost fully driven by the variation in the age of the different groups.
Bolhaar, Brouwers, and Scheer (2016[2]) show that the gender of those in non-standard employment varies according to the unemployment rate. When unemployment is low, the difference between men and women for the probability of working on a flexible contract is largest. When economic circumstances improve, women remain in flexible employment, while men move to open-ended contracts. So far, we do not know what drives this pattern. It may be the result of better bargaining power or skills among men: possibly women are in a worse position to negotiate a better contract when the economic situation improves. It does not appear to be driven by preferences for flexible work arrangements. According to the Dutch Working Conditions Survey of 2015, working on a non-standard contract is a positive choice (i.e. they like arrangements that offer flexibility and have no need for security) for roughly 35% of men, compared with 29% of women. An alternative explanation is that women enter and leave the workforce more often than men, which means they have shorter tenure and are therefore less likely to move to open-ended contracts.
7.3.1. A snapshot of the social protection system in the Netherlands
This section presents the social protection of workers on open-ended contracts and other work arrangements in terms of employment protection, unemployment benefits, disability benefits, sickness pay, minimum wages, collective labour agreements, active labour market policies, pensions and the taxes they pay (including work-related deductions).
Dutch labour law regards employees as those who depend mostly on their employment contract for their income and are therefore in an unequal bargaining position with the employer. The employee is therefore protected against any unfair treatment by the employer, such as dismissal without notice, very long working hours, dangerous working conditions and so on. Employees are also insured against unemployment and disability risks by social security. And there are sectoral collective labour agreements that stipulate wage schedules, on-the-job-training rights and pension rights, and sometimes top-ups to sickness or unemployment benefits. There is a legal minimum wage in place that applies to all employees (not for own-account workers), and many sectoral collective labour agreements impose higher wage levels.
By law, employers in the Netherlands are obliged to treat all employees equally: they are not allowed to make a distinction based on the number of working hours in the contract or the permanency of the employment contract. Part-time and full-time workers therefore have the same rights by labour and social security law. This is also the case for workers on temporary contracts, with the exception of employment protection, and for variable hours contracts except on the matter of their working hours.
Since hiring firms are not regarded as the legal employer of agency workers, they fall under a different collective labour agreement. Therefore, their working conditions may differ from workers hired directly by the firms at whose workplace they are based.
Since own-account workers are not regarded as employees, they are not covered by labour law or collective sickness, disability and unemployment insurance systems. Their social protection therefore differs substantially from that of all other workers.
7.3.2. Social protection of workers on open-ended contracts
In labour law, employees on open-ended contracts are protected against involuntary termination of their contract. Without the consent of the employee (i.e. except in cases of mutual agreement), an employer can only terminate the employment contract if he can show fair grounds for dismissal and the employee cannot be reassigned to another position within the firm.3 These fair grounds need to be approved by the public employment office when the contract is ended for economic reasons or by a court if the termination is based on individual factors, before the termination can actually take place. Upon termination, the employer is obliged to make a transition payment to the employee (i.e. severance payment), if the worker was with the employer for two years or more. This severance payment is based on the years of service and gross salary. The employer is only exempt from these transition payments if the employee was fired on the grounds of gross misconduct.
All employees are protected against dangerous or unhealthy working conditions by the Working Conditions Act (Arbowet). The employer is responsible for the provision of proper working materials and is legally accountable for any harm that employees suffer on the job. There are rules for the minimum number of holidays, and holiday payments of 8% of the gross wage.
There is a statutory minimum wage that applies to all employees. The current minimum wage for adults aged 22 years or older is EUR 8.96 per hour worked. For younger people the minimum wage increases in steps from EUR 2.69 per hour for 15-year-olds to EUR 7.61 per hour for 21-year-olds. The minimum wage is adjusted on 1 July every year.
Collective labour agreements (CLAs) are concluded on a sectoral base between employers’ organisations and unions. Wage schedules, pension rights and training rights are all included in these arrangements, as well as the negotiated wage rise for the sector as a whole. Some CLAs also include top-ups of unemployment benefits or sick pay.
Social security law protects employees against income loss resulting from unemployment by the Unemployment Benefits Act (Werkloosheidswet) and afterwards by the universal means-tested social assistance and Act on Income Provisions for Older Unemployed (IOW or Wet Inkomensvoorziening Oudere Werklozen). The Unemployment Insurance Act (WW) provides insurance against involuntary job loss for employees. Eligible workers can receive unemployment benefits for a minimum of three months, while the maximum duration depends on the number of weeks the worker was employed before the job loss. For the first two months workers receive 75% of the monthly base wage, which is a function of their previously earned wage up to a certain ceiling, and 70% afterwards.
Employees are protected against income loss resulting from sickness by the Sickness Benefits Act (ZW, or ZiekteWet). Employers are obliged to keep sick employees on the payroll for two years. By law, the employer has to pay 70% of the wage during this period. In most collective bargaining agreements this is increased to 100% in the first year. Employers are required to attempt to reintegrate sick workers in their organisation or at another organisation. If employers cannot prove that they have made a reasonable attempt at reintegration, the period of continued wage payment can be extended to three years.
After these first two years, sickness becomes a fair ground for dismissal. Employees who have been on sick leave for more than two years are eligible for WIA disability benefits – WIA is the Work and Income (Capacity for Work) Act. The WIA is aimed at reintegration into the workforce, unless people are fully and permanently disabled. For older (partially) disabled unemployed people who are not eligible for the IOW there is another programme: Income Provisions for Older and Partially Disabled Unemployed Employees (IOAW).
Besides these employment-related social protection benefits, there is also a universal means-tested social assistance (Bijstand), which is not dependent on a person’s employment history. The means test consists of a wealth test and a household income test. The maximum income is 70% of the minimum wage for single person households and 100% for multi-person households.
7.3.3. Social protection of workers on temporary contracts
In the Netherlands, the same labour and social security laws, minimum wages and youth minimum wages apply to workers on temporary contracts as to workers on open-ended contracts. The same protection against unsafe or unhealthy working conditions and rules for holidays, vacation days and holiday allowances also apply. Temporary workers fall under the same collective agreement as their colleagues on open-ended contracts, but these agreements may stipulate different on-the-job-training rights or pension rights for the two categories. Unfortunately, there is no recent study on these specific exceptions in CLAs for temporary workers (De Beer and Verhulp, 2017[3]).
The fact that the coverage of many of the social security schemes depends on employment history means that temporary workers enjoy more limited benefit rights than workers on open-ended contracts. Like workers on open-ended contracts, temporary workers build up insurance against involuntary unemployment under the Unemployment Benefits Act. However, as discussed in the section above, the duration and amount of unemployment benefit depends on the prior wage and employment history. As a result, temporary workers are not only more likely to become unemployed (Van der Werff, Kroon and Heyma, 2016[4]; De Beer and Verhulp, 2017[3]), but also have shorter maximum benefit durations due to a shorter employment history (Van der Werff, Kroon and Heyma, 2016[4]).
Subject to certain restrictions the Public Employment Service (PES) compensates workers who move frequently between employment and unemployment. If an unemployed worker finds a job, but loses it again within 26 weeks, the unemployment benefit that the workers received before finding this job is reinstated if the new base wage4 is below 87.5% of the old base wage. This is done to compensate for the loss of benefit income due to accepting a lower-paying job. However, if the new job lasts longer than 26 weeks or pays more than 87.5% of the old job, reinstatement of benefit is not possible and the worker needs to start a new unemployment benefit application procedure. Only the employment history and wage of the new job counts towards this new unemployment benefit.
In essence, temporary workers also receive the same insurance against sickness and disability. However, the obligation to keep sick employees on the wage bill ends at the pre-specified end of the contract. From that moment, the PES pays the sickness benefit (ZW benefit or Ziektewet-uitkering) and is responsible for the reintegration of the sick employee into the labour market. Like workers on open-ended contracts the benefit is determined by the base wage: the default amount is 70% of the base wage.
The most important difference between open-ended and temporary contracts is that temporary contracts end at a prespecified date or event. This means that the employer is not obliged to provide a fair ground for dismissal at the end of the contract. By employing workers on temporary contracts, employers can thus avoid the need for a dismissal procedure and, in most cases, also the need to make a severance payment. The employee only receives a transition payment at the end of the contract if the duration of the contract (or consecutive contracts) exceeded two years. Most temporary contracts are shorter than that.
According to EU Directive 1999/70/EC, European countries are obliged to adopt regulations that restrict the use of temporary contracts. The Netherlands has chosen not to place restrictions on the reasons used to justify the use of temporary contracts, but rather on their maximum duration and the maximum number of consecutive temporary contracts with one employer. In 1999 the Dutch government introduced the Wet Flexibiliteit en Zekerheid (Flexwet) law, which set the maximum duration of consecutive temporary contracts at three years. A contract was considered consecutive if the period between the end of a contract and start of a new contract was less than three months. A temporary contract was automatically converted into an open-ended contract if the duration of consecutive contracts exceeded the maximum of three years. This regulation of consecutive fixed-term contracts was called the chain rule.
In 2015 the regulation of these “chains of temporary employment” became stricter with the introduction of the Work and Security Act (WWZ or Wet Werk en Zekerheid). Contracts are now considered consecutive if the period between the end of a contract and start of a new contract is less than six months, and the maximum duration of such consecutive contracts has been lowered to two years. In addition to the maximum duration of consecutive contracts, the WWZ stipulates that an employee can be on a maximum of three consecutive fixed-term contracts. Thus, a temporary contract is automatically converted into an open-ended contract if the total number of contracts in a chain exceeds three or the total duration of the chain exceeds two years. For example, if an employee renews a fixed-term contract and this takes the total duration of the contracts to more than two years, then the new contract is regarded as an open-ended contract by law. Note that this rule only applies to a chain of contracts. A single fixed-term contract for a period of more than two years is possible, but it cannot be extended with a second temporary contract. In that case, any extension must be an open-ended contract.
The effectiveness of these measures in reducing the share of temporary work is not evident. There is no thorough study yet on the effects of the WWZ. Causal inferences are difficult to find, because it is hard to disentangle the effect of various elements of the very broad act and changing economic circumstances. In addition, the new law introduced an incentive to hire people for shorter periods. Circumstantial evidence shows that a sequence of three temporary contracts lasting seven and eight months each (23 months in total) has become much more popular since the introduction of the WWZ law, because this so-called 7-8-8 scheme maximises the total length and number of consecutive temporary contracts.5
Since 1999 firms have been able to deviate from the standard chain rule in their CLAs for jobs that require more flexibility due to the nature of the work. The maximum number of consecutive fixed-term contracts can be extended to six (at most) and the maximum total duration of these contracts can be extended to (at most) four years. In addition, the required six-month break between contracts that is necessary to break the chain can be shortened to three months, but only when the work is of a seasonal nature. The most recent detailed study of these deviations dates from 2014, before the introduction of the stricter WWZ regulation. However, these older CLAs still show whether the chain rule was made stricter or less strict in the bargaining process. For 71% of workers covered by a CLA there was a deviation from the default chain rule that either applied to all employees or specific groups of employees covered by the CLA. A deviation for specific groups of employees was most prevalent: this applied to 60% of employees covered by a CLA (e.g. those aged 65 and over, seasonal workers or apprentices). For 17% of employees covered by a CLA the deviation applied to all employees.6 (De Beer and Verhulp, 2017[3]).
CLAs that included a deviation from the chain rule tended to decrease the maximum total duration of the chain, which effectively made the chain rule stricter. However, the rules for the maximum number of contracts and the minimum gap between chains were more often made less strict. So, overall it seems that unions were most interested in decreasing the total possible duration of the chain, but were less concerned about the maximum number of contracts and the minimum gap between chains.
For specific groups of employees (e.g. those aged 65 and over and apprentices) it is possible to agree to exclude specific groups from the chain rule completely. 18% of employees in such groups are not covered by the rule due to such an exemption (De Beer and Verhulp, 2017[3]). In these cases an indefinite number of consecutive temporary contracts can be offered.
Unfortunately, relatively little is known about deviations in CLAs after the introduction of the WWZ. In 2015 roughly 25% of employees covered by a newly created CLA were in agreements covered by a CLA that deviated from the WWZ chain rule. However, the exact nature of these deviations is currently unknown (De Beer and Verhulp, 2017[3]).
7.3.4. Social protection of agency workers
In the Netherlands, the same labour and social security laws apply to agency workers as to open-ended contracts, with a few exceptions. The same minimum wage and youth minimum wages apply as to workers on open-ended contracts or temporary contracts. Agency workers also build up rights for unemployment benefits and disability benefits in the same way as other workers, but like temporary workers their rights are usually shorter due to interrupted work histories (see below). Furthermore, agency workers have the same protection against unsafe or unhealthy working conditions and the same rules for holidays, vacation days and holiday allowances apply.
In short, employment conditions are the same for agency workers and other employees, as stipulated in the WAADI law on the provision of employees by intermediaries (Wet Allocatie Arbeidskrachten door Intermediairs). This also means that agency workers must be paid on an equal basis to other workers at the user company. However, this equal pay rule is difficult to apply in practice and Dutch unions complain that is not well respected (PPMi, 2017[5]). Finally, pension rights are not included in the WAADI law. Agency workers automatically enrol in a separate pension fund after 26 weeks under contract at the agency firm.
The rules concerning temporary work differ between standard employment agreements and agency agreements. By law, the chain rule for temporary contracts does not apply during the first 26 weeks of an agency agreement, as a result of which an indefinite number of temporary contracts can be put in place during this time, without any requirement for a trial period.7 Afterwards, a maximum of six consecutive temporary agency agreements is allowed (instead of three for standard employment contracts) with a maximum total duration of four years (instead of two for standard employment contracts). In addition, during the first 26 or 78 weeks, temporary agency contracts automatically end if the hiring firm ends the hiring. This is the case if the contract between the worker and the temporary employment agency contains a so-called “uitzendbeding” clause. Most contracts have this clause during the first 26 or 78 weeks, depending on the CLA.8 Dutch politicians and policy makers regard the temporary agency sector as an important means of getting unemployed workers back into the work force. The agency sector has therefore been granted these exceptions to the rules.
Only 13% of agency workers are employed for more than 26 weeks by the same temporary employment agency and 2% beyond 78 weeks (Vermeulen et al., 2016[6]).
Regarding unemployment benefits, disability benefits and social assistance, the same rules apply for agency agreements and standard employment agreements. However, due to interrupted work histories agency workers less often meet the eligibility criteria and if they are eligible, their entitlement to benefits is usually shorter. The difference in inflow to unemployment and unemployment benefits between standard and non-standard contracts will be discussed in Section 7.3.3.
Rights during periods of sickness are different for temporary agency workers than for other employees, though this depends on their contract with the temporary employment agency. If their contracts contains a “uitzendbeding” clause, which is usually the case during the first 26 or 78 weeks, the employment contract ends upon the start of a period of sickness. In that case the UWV employee insurance agency pays sickness pay of 70% of the base wage, and is responsible for the reintegration into the labour market. If the contract does not contain a “uitzendbeding” clause, which is usually the case if the contract with the employment agency exceeds 26 or 78 weeks (depending on the CLA), sickness regulations are the same as for regular employment. Recall thatonly 13% of agency workers are employed for more than 26 weeks by the same temporary employment agency and 2% beyond 78 weeks (Vermeulen et al., 2016[6]).
7.3.5. Social protection of on-call workers and those on other variable hours contracts
In the Netherlands, the same labour and social security laws apply to contracts with variable working hours as to other employment contracts. The same protection against unsafe or unhealthy working conditions and rules for holidays, vacation days and holiday allowances apply. The same hourly minimum wage and youth minimum wages also apply. In order to protect on-call workers against employers calling on them for only a very short amount of time, the employer is obliged by law to pay at least three hours’ wages every time he uses the worker for less than three hours of work.
On-call contracts and other contracts with variable hours can be either an employment contract or agency contract. Furthermore, this contract can be open ended or on a temporary basis. Basically, workers on variable hours contracts have the same social protection as other workers on the corresponding contracts with fixed working hours. So the employment protection of, for example, on-call workers on open-ended contracts is the same as the employment protection for other workers on open-ended contracts. And for on-call workers on temporary contracts the same chain rule applies as for temporary workers. On-call workers on an employment contract have the same pension rights as workers on a standard employment contract. The amount of pension rights is determined by the actual number of hours worked.
The employer is not free from obligations when deciding to call upon an on-call worker. If the employee is willing to work, and the employer has suitable work available, then the employer is obliged to offer the expected number of hours’ work to the on-call worker. If the employer does not adhere to this rule, the on-call worker can demand payment for the number of hours that could have reasonably been expected in a particular month. The on-call worker, on the other hand, is obliged to respond to the employer’s calls if he is available. The employer and employee are allowed to deviate from these rules, by agreeing at the start of the on-call contract that this rule does not apply, but only for the first 26 weeks.
Upon dismissal, the same rules apply as to other workers. A firm cannot just stop calling upon an on-call worker. The same grounds for dismissal are allowed and need to be proven in court or at the public employment office before dismissal can take place. Also, the same rules apply for severance payment, which is based on years of service and gross salary.9 Naturally, workers need to be aware of these rights for them to be effective. It is unclear whether all workers in these types of on-call arrangements are aware of these rights.
However, in certain cases the employer and on-call worker do not agree on the hours of work an on-call worker can reasonably expect. Or firms suddenly stop calling upon a worker, even though that is not allowed. In those cases the on-call worker may have to go to court to ascertain these rights (e.g. Rb. Amsterdam 3 February 2014, ECLI:NL:RBAMS:2014:522). It is unclear how many on-call workers are not aware these rules.
After having worked three months, the same rules for unemployment benefits and disability benefits apply to workers with variable hours contracts as to other employees. UWV calculates the rights for benefits based on the wage earned in the year before the start of the period of unemployment or sickness.10
Sick pay differs. Firms are obliged to pay at least 70% of the wage during the first two years of sickness (or until the end of the temporary contract), for the usual hours worked, or the hours for which the worker was already scheduled to work. However, if explicitly stipulated in the employment contract, firms may pay only for the actual hours worked (i.e. excluding sick days) during the first six months of on-call or other variable hours work. If a worker has worked for at least six months on a regular schedule, he has the right to receive sick pay of at least 70% of the wage based on this schedule.
7.3.6. Social protection of self-employed and own account workers
The social protection of self-employed and own-account workers differs substantially from the social protection of all employees mentioned above. Only protection against unsafe or unhealthy working conditions is the same as for employees.11 For all other social protection the situation for own-account workers is completely different due to the fact that the contract between a hiring firm and an own-account worker is, according to the law, not a labour contract. As a result, labour law does not apply. This implies that there is no employment protection, no employee insurance against unemployment, disability or sickness and no minimum wages. Own-account workers do not fall under any collective labour agreement either, so they have no training rights or obligatory pension savings. There are no rules for vacation days and holiday allowances.
Unlike employees, own-account workers are not automatically insured against unemployment, disability and sickness. They do not pay contributions to the public unemployment insurance or disability funds, and therefore have no rights to these types of benefits. Social assistance is therefore their safety net, unless they arrange disability insurance or pension savings by themselves.12
Additionally, there is a social assistance act specifically for the self-employed (Besluit bijstandverlening zelfstandigen, Bbz). Self-employed workers who experience a temporary income shortfall can get social assistance to supplement their income (up to the level of a regular social assistance benefit). It is also possible to receive a zero- interest loan instead of the social assistance. If the person’s wealth exceeds EUR 186 498, they are only eligible for a zero-interest loan. A number of criteria need to be met to be eligible for the Bbz. First, the loss of income must be temporary, with an expectation that the self-employed person will regain a sufficient level of income in the future. Moreover, the person must spend a minimum of 1 225 hours per year working at his own company and it is not possible to receive funding from a private party. The default maximum duration of this assistance is 12 months, but this can be extended for an additional 24 months if necessary.
Last, for elderly self-employed people with a low income, there is another separate benefits act (Wet Inkomensvoorziening oudere en gedeeltelijk arbeidsongeschikte gewezen zelfstandigen, IOAZ). This provides an income supplement to poor self-employed people who are older than 55 but not yet at statutory pension age (65)13.
Own-account workers are not insured against unemployment, disability and sickness through the social security system, but they can opt for private insurance. In practice 25% of own-account workers are insured against disability (Berkhout and Euwals, 2016[7]). Roughly 25% of own-account workers report that they have not made any arrangement for their pension (including savings, firm/house equity or pension funds). The most prevalent reasons are that they cannot afford to make pension contributions (53%), they have not gotten to it yet (27%) and that the pension age is still a long time away (20%). Another 10% report that they have a partner with adequate pension arrangements (Statistics Netherlands & TNO, 2017[8]).
In case of pregnancy, female own-account workers receive benefits equivalent to the minimum wage for 16 weeks. This duration is the same as for female employees, but the benefit level is lower.
Own-account workers pay much lower taxes and do not pay any social insurance contributions. The net income of an own-account worker is therefore much higher compared with an employee with the same gross income while the wedge of taxes and social security contributions is smaller (Figure 7.4). On average own-account workers pay 20% less tax from the same gross income (Bosch, De Graaf-Zijl and Van Vuuren, 2015[9]). Also they do not pay contributions for unemployment and disability insurance, and saving for an occupational pension is voluntary.
In certain situations a worker is not allowed to be classified as an own-account worker, because he matches the definition of an employee in Dutch labour law14 (see Section 7.4.2. for more information on the misclassification of own-account workers). However, employers have faced legal uncertainty regarding the classification of own-account workers. In order to clarify these rules a law was passed in 2016, called the Deregulation Labour Relations Assessment Law (Wet Deregulering Beoordeling Arbeidsrelatie, the DBA law). This required both the employer and the own-account worker to assess whether the worker is actually allowed to be classified as an own-account worker (rather than as an employee). If the parties are unsure about the classification, sector specific template agreements are available that define criteria for the correct classification of own-account work. If these criteria apply, the worker and employer can sign the agreement and both parties can be sure that the own-account worker has not been misclassified.
However, this law did not have the desired effect. Legal uncertainty increased even more, due to confusion about how strictly these new agreements would be monitored. Many employers and own-account workers reported that this made hiring own-account workers less attractive. Due to these protests and fears of job losses, the government decided to postpone monitoring the template agreements: no companies or own-account workers will be fined before 2020 (except in clear-cut cases of fraud or malicious intent). This is supposed to provide the necessary time for the government to redraft the laws concerning own-account workers and increase legal certainty.
Figure 7.4. The tax wedge is significantly higher for dependent employees
Note: The wedge is defined as the difference between labour cost and disposable income. Taxes and premiums for the unemployment benefits sickness/disability benefits are included in this wedge, whereas pension premiums are not. Self-employed earnings are the profits from self-employment as reported for tax purposes. The kinks in the marginal tax rates are caused by the phasing in and out of tax credits.
Source: Bosch, De Graaf-Zijl and Van Vuuren (2015[9]).
7.4. Incentives created by the system and implications for the financing of social protection
The different entitlements associated with different forms of employment (employment protection legislation, sickness benefit, unemployment and disability benefits, collective labour agreements and minimum wages) influence incentives for employers to hire people on temporary contracts, agency work, on-call work and own-account work as alternatives to open-ended contracts. In general, firms can save on social contributions, severance payments, reintegration obligations and burdensome administrative procedures by hiring people as own-account workers or on one of the flexible employment arrangements. The use of all these types of flexible work arrangements instead of regular open-ended contracts with fixed working hours is a much debated topic in the Netherlands. Dutch unions claim that employers use these types of contracts for the sole purpose of circumventing the social protection of open-ended contracts.
This section discusses the incentives the Dutch system of social protection, employment protection legislation and collective bargaining generates for employers and workers, looks at the evidence on whether these incentives actually influence firm and worker behaviour and at the incomes of non-standard workers and their reliance on tax-funded assistance compared with non-standard workers.
7.4.1. Incentives resulting from the system
By using temporary work arrangements, employers can circumvent employment protection legislation – i.e. severance payments and time-consuming procedures to obtain permission for dismissal – and avoid sickness-related payments and reintegration obligations which end on the termination date of the contract. So even though in principle temporary workers have the same social security rights and fall under the same labour law and collective labour agreements as workers on open-ended contracts, employers can avoid costly and onerous dismissal procedures and obligations to reintegrate sick workers if they hire people on a temporary basis. The WWZ law made the use of a chain of temporary contracts less attractive in 2015, with the aim of limiting the use of temporary employment contracts. However, its effectiveness is unclear (see Section 7.3.3).
However, establishing such a set-up with the purpose of avoiding the automatic transition to an open-ended contract can be unlawful, and in some cases, such structures are penalised in court. The temporary contract can then be converted to an open-ended contract, even if the employer adhered to the chain rule (e.g. Rb. Utrecht 12 December 2008, ECLI:NL:RBUTR:2008:BG9194). In other cases the contract is not converted to an open-ended contract, but the employee is compensated, for example, by being guaranteed an extension of the temporary contract (e.g. Rb. Groningen 14 March 2007, ECLI:NL:RBGRO:2007:BA1435). However, agreeing a pause in employment contracts to avoid having to offer an open-ended contract is not in itself a sufficient condition for the court to rule in the employee’s favour (e.g. HR 29 June 2007, ECLI:NL:HR:2007:BA2504). A decision to compensate the employee depends on the context of the case. For example, a ruling in favour of the employee is more likely if the employee has repeatedly worked on temporary contracts with the same employer for a long time.
By hiring temporary agency workers, employers can limit costs and risks, due to the fact that the hiring firm can end a hiring period without notice and is not responsible for reintegration obligations and sickness payments. In addition, in some sectors temporary agency workers are cheaper for employers than regular workers, because they fall under a different collective labour agreement with lower wage scales.
Moreover, employers can save on wage payments by hiring people on contracts with variable working hours because this means they do not have to pay for unproductive hours. Also, sickness payments can be avoided during the first six months of on-call work.
By hiring own-account workers, employers can avoid risks and costs. They are not subject to employment protection obligations, regulations in cases of sickness or collective labour agreements with their ensuing pension and training rights if they hire own-account workers. In addition, own-account workers pay considerably lower taxes due to tax reductions that are specifically aimed at self-employed workers and small companies, and firms may capture part of this benefit. Firms that hire own-account workers negotiate a fee with the worker. Whether lower taxes and social security contributions benefit the firm or the worker depends on their respective bargaining positions.
A report by senior civil servants from various Dutch ministries (2015[10]) indicated that own-account workers at the higher end of the wage distribution can use their bargaining power to gain benefit from the lower costs, while firms profit from the tax advantages at the lower end of the wage distribution by negotiating low fees. At the lower end of the wage distribution, they calculated that firms can save 30% of wage costs by hiring an own-account worker and capturing all the cost advantages, compared with hiring the same worker on an employment contract at the minimum wage level. On the other hand, at the higher end of the wage distribution (roughly two times the median wage) they calculated that own-account workers can earn a 43% higher net income than employees in cases where the hiring firm has the same wage costs, if the workers can capture all benefits for themselves.
Another topic under debate in the Netherlands is the position of workers employed by payrolling firms. Much like agency workers, workers on payrolling contracts work for a hiring firm, whilst under contract at another firm. However, the main difference between agency and payrolling firms is that agency firms actively search for suitable workers and match them with hiring firms. Payrolling firms do not perform such tasks; workers are recruited by the hiring firm and that firm then contacts a payrolling firm that formally employs the worker. Payroll contracts have a separate CLA, which means that firms that prefer the payrolling CLA over their sector-specific CLA have an incentive to hire workers on payrolling contracts. However, this does not allow employers to offer lower wages, because payroll workers must be paid on an equal basis to other workers doing similar jobs at the hiring firm; because payrolling is covered by the WAADI law (see Section 7.3.4). The incidence of payrolling is difficult to quantify, because payrolling workers are currently not included as a separate category in the labour market statistics.
7.4.2. Is there evidence that firms misclassify workers to save contributions?
In the Netherlands, the discussion on misclassification of workers is mostly focused on false self-employment. By hiring own-account workers instead of (temporary) employees, firms can save on social security contributions, which provides an incentive to opt for the former classification. Alternatively a worker can save on both taxes and social security contributions by performing a task as an own-account worker instead of on a labour contract. However, in certain situations a worker is not allowed to be classified as an own-account worker, because he matches the definition of an employee in Dutch labour law. If in such a case an employer nevertheless decides to hire someone as an own-account worker, this is called false self-employment. From a social protection viewpoint such structures are undesirable, since no social contributions are made for these own-account workers even though they are in fact employees. From an individual viewpoint this can be beneficial for both the employer and worker, since both the wage cost and net wage can increase due to the lower wedge.
Researching false self-employment is a difficult task, because it is not always clear whether an own-account worker is actually a misclassified employee15. As such, estimates of the size of the group of false self-employed generally have a large margin of error. In a 2013 study an attempt was made to estimate the share of false self-employment in four sectors: construction, health care, transport and management and organisation (including ICT) (SEOR, 2013[11]). By surveying own-account workers within these branches, the researchers estimated that roughly 3-15% of own-account workers should have been classified as employees.
According to the SEOR report, misclassified own-account workers do not necessarily earn less than actual own-account workers. The average gross hourly earnings of the misclassified group are higher, because they spend less time on unpaid tasks (administration, acquisitions etc.). However, employers do not pay a higher hourly rate to the misclassified group – overall their compensation is comparable. The transport and management branches are an exception, with misclassified own-account workers being paid more than their correctly classified counterparts and even more than the wage costs of an employee. This suggests that in these branches false self-employment is a benefit for the own-account worker (at the expense of the employer). In other branches misclassified own-account workers are paid comparable or lower amounts than correctly classified workers (employees and own-account workers), which suggests that either the employer or both the employer and the own-account worker benefit from the misclassification.
7.4.3. Incomes, risks and reliance on tax-funded assistance
It is hard to determine the causal relationship between working on non-standard work arrangements and income. The reason is that the people working in these arrangements differ substantially from workers on regular open-ended contracts. As shown above, workers in flexible work arrangements are in general younger, have a lower level of education and more often come from a non-native background. These characteristics generally correlate with human capital and work experience and workers on non-standard work arrangements are therefore expected to earn lower wages, have lower incomes and less wealth, for these reasons alone.
That being said, we do observe differences in standardised household income by contract type. People on open-ended contracts on average have the highest standardised household income, roughly EUR 33 000, followed by own-account workers (EUR 32 000, see Figure 7.5). People on non-standard employment contracts generally have a lower income, of which agency work has the lowest average (EUR 25 300). These differences may result from various factors, not only the wage/personal income as an own-account worker, but also partner income and the number of people in the household. Personal characteristics differ between the various groups as well (see Section 7.4).
Figure 7.5. Those working on standard contracts enjoy the highest incomes
Note: Standardised income is the disposable income corrected for differences in household size and composition. This adjustment is done using equivalence factors. These factors represent the economy of scale realised by sharing household costs. Using these factors all incomes within a household are adjusted such that they represent a separate one-person household. In this way the welfare level between households are made comparable. Because welfare is experienced on an individual level, a standardised income is assigned to each of the household members. The disposable income consists of gross income minus transfers of income such as alimony payments, premiums for social securities and taxes on income and wealth.
Source: Statistics Netherlands (2017[12]).
The inflow rate into unemployment of employees on non-standard contracts is over five times that of workers on standard contracts, and the unemployment benefit inflow rate is roughly three times higher. The probability of leaving the job market entirely is also much higher: five times that of workers on standard contracts (Table 7.1). Separate data show that temporary workers are also four times as likely to move into the unemployment benefit scheme as regular workers (Table 7.2). The fact that non-standard workers are more likely to enter unemployment than the unemployment benefits scheme implies that non-standard workers are less likely to be eligible for unemployment benefit than standard workers once they become unemployed. This is probably due to their interrupted employment histories, which makes them less likely to meet the eligibility criteria. It is therefore not surprising that workers on non-standard contracts are roughly nine times as likely to move from employment into social assistance (see Table 7.1).
Temporary and agency workers are also more likely to flow into the WIA disability benefit scheme compared with standard workers. For on-call workers, however, this is less likely (Table 7.2). This may be due to their younger age, but possibly also to the eligibility criteria being harder to meet for on-call workers. These statistics do not necessarily imply that on-call workers are less likely to suffer from disabilities. Nor does the higher rate of temporary and agency workers moving onto disability benefit automatically imply that these non-standard contracts increase the probability that a worker becomes unfit for work. Possibly jobs often performed on such non-standard contracts are riskier, but it may also be the case that people who work on non-standard contracts are more likely to develop health problems. Another explanation may be that reintegration is more difficult when the public employment office is responsible for this, rather than the employer. The latter is the case for sick temporary workers after the end of their contract.
Table 7.1. Non-standard workers are more likely to transition into unemployment-related situations
1 Defined as the proportion of workers who were employed in the previous year, but who have transitioned to one of these outcomes in the current year.
(a) Source: data from Statistics Netherlands’ Statline (link, downloaded on 13-11-2017). Own calculations: average yearly rate 2003‑17; quarterly rate converted to a yearly rate: Py = 1-(1-Pq)^4.
(b) Source: Van der Werff, Kroon and Heyma (2016[4]), using Statistics Netherland's microdata. Own calculation: average rate 2009‑14.
Table 7.2. Inflow rates into various benefit schemes are higher from non-standard contracts
(a) Source: UWV (2010[13]). Yearly rate in 2008.
(b) Source: De Beer & Verhulp (2017[3]), p.36 with own calculations. Average WIA inflow rate after 24‑36 months based on three measurement dates: October 2007, 2008 and 2009).
7.5. Conclusions
In recent decades, the share of non-standard work in the Netherlands has grown substantially and it is now among the highest of all OECD countries. Currently, approximately one-third of all workers are employed on a non-standard work arrangement. The share of workers employed in temporary work, agency work and on-call work has risen from 14% in 2003 to 23% in 2017, while the share of own-account workers has increased from 8% to 12%. The probability of working in a non-standard employment relationship is mostly correlated with age. The probability of working on all types of non-standard employment contracts decreases with age, whereas the share of own-account work increases with age. There are differences between men and women, between educational groups and between ethnic groups as well, but these differences are smaller than those between the age groups. In some cases these differences appear to be driven by the variation in the age of the different groups.
In principle, all employees have the same rights under labour and social security law. However, exceptions exist: workers on temporary contracts do not have the same employment protection rights as standard workers. If their contract lasts less than two years, they have no rights to severance payments and the employer does not need to provide justification for their dismissal. In addition, once the contract ends, temporary workers also have different rights to sick pay and reintegration assistance.
Workers on variable hours contracts do not have the same rights as regular employees regarding working hours, and during the first six months they often do not have the same rights to sick pay. In principle their rights to employment protection are the same as for regular workers, so that employers cannot just stop calling upon a worker or reduce their monthly working hours. But one may doubt whether all workers are aware of these rights and whether they in practice enforce these rights, since they are in a dependent position and going to court is time consuming and expensive.
Since temporary agency workers fall under a different collective labour agreement than workers hired directly by the firms at whose workplace they are based, and their conditions – such as wages, fringe benefits and pension rights – may differ.
Despite the legal equality of the employment contracts, the position of workers employed in these different relationships is not 100% equal. The inflow into social security benefits such as unemployment benefits, social assistance and disability benefits, is higher among non-standard workers. We cannot exclude the possibility that this is partly due to reversed causality, e.g. that non-standard workers are less healthy. But it is also apparent that the rules lead to different outcomes for different groups of workers. Temporary workers are likely to have shorter and less stable employment histories than workers on open-ended contracts. The fact that entitlement to many social security benefits depends on employment history means that temporary workers often do not accrue benefit entitlements to the same degree as standard workers. The benefits received are therefore usually lower and have a shorter maximum duration.
The most pronounced difference in social protection is between employees (with an employment contract), who are covered by labour law, and own-account workers, who are not. Since own-account workers are not regarded as dependent employees, they are not covered by labour law or collective employee insurance regimes for sickness, disability and unemployment. Their social protection therefore differs substantially from that of all other workers.
The discussion on misclassification of workers is mostly focused on false self-employment. By hiring own-account workers instead of (temporary) employees, firms can save on social security contributions and avoid minimum wage regulations, which provides an incentive to opt for the former classification. On the other hand, workers can save on both taxes and social security contributions by performing a task as own-account workers instead of under a labour contract. Both firms and workers can therefore have an incentive to misclassify an employment relationship as own-account work. A report by senior civil servants from various Dutch ministries (2015[10]) showed that own-account workers at the higher end of the wage distribution can use their bargaining power to benefit from the lower costs, while firms profit from the tax advantages at the lower end of the wage distribution by negotiating low fees. There are concerns that, especially at the lower end of the wage distribution, workers are forced into own-account work in situations of dependent employment, with the result that social security rights are withheld from this vulnerable group of workers.
Recently, the Dutch government has tried to curb the growth in temporary contracts, because they view further growth of all kinds of flexible work arrangements as undesirable. In 2015, the WWZ law tightened the rules concerning a “chain of temporary contracts”. The effectiveness of these measures in reducing the share of temporary work is not evident, however. There is no thorough study yet on the effects of the WWZ and a causal link between the introduction of the act and a lower share of temporary work is difficult to find. It does appear, however, that a sequence of three temporary contracts of seven and eight months each (23 months in total, which maximizes the total length and number of consecutive temporary contracts) has become much more popular since the introduction of the WWZ law. An attempt to combat false self-employment by clarifying the legal boundaries between own-account work and dependent employment (the so-called DBA law) did not have the desired effect, due to confusion about how strictly these new agreements would be monitored.
Despite these recent attempts to combat the growth of the various forms of flexible employment, and despite the economic upswing, its share is still increasing. This is probably due to the fact that the Dutch institutional setting provides firms with possibilities to save on social contributions, severance payments, reintegration obligations and laborious administrative procedures by hiring people as own-account workers or (to a lesser extent) on one of the flexible employment arrangements. By using temporary work arrangements, employers can circumvent employment protection legislation – i.e. severance payments and time-consuming procedures to obtain permission for dismissal – and sickness-related payments and reintegration obligations that end on the expiry of the contract. If policy makers wish to reduce the share of non-standard work arrangements, policy should be aimed at reducing incentives to hire workers on non-standard contracts and for workers to opt out of the social security system and pay lower taxes as own-account workers.
7.5.1. Policy options for reducing the share of non-standard contracts:
Combinations of several policy options offer ways of combating the further increase of non-standard work arrangements:
1. Stricter regulation of non-standard work, for example through:
Minimum hourly compensation for own-account workers (proposed in the recent Dutch coalition agreement).
Stricter rules concerning the use of temporary contracts (e.g. maximum duration and number of contracts, or the situations in which non-standard work arrangements are allowed).
Stricter monitoring of the current rules concerning non-standard contracts (proposed in the recent Dutch coalition agreement).
A disadvantage of these policy options is that already existent incentives to avoid the use of standard (i.e. permanent) contracts will remain in place. So stricter monitoring is necessary in any event. However, in practice effective monitoring can be difficult and costly (see for example Eurofund (2015[14])).
2. Reducing differences in taxes and social security coverage between non-standard and standard work, for example by:
Decreasing the difference in the tax wedge between own-account workers and standard workers, by reducing or abolishing the tax benefits of self-employment, or introducing an additional tax deduction for dependent employees.
Increasing the rights of non-standard workers e.g. by expanding the right to severance payment to workers with short employment histories (less than two years) and broadening social security laws to include all workers, not only dependent employees.
Reducing the social security rights and employment protection of workers on open-ended contracts and the duration of sick payment and reintegration obligations for employers of workers on open-ended contracts.
Netherlands: non-standard work and social protection
Netherlands: non-standard work and social protection
References
[7] Berkhout, E. and R. Euwals (2016), Zelfstandigen en arbeidsongeschiktheid, CPB.
[2] Bolhaar, J., A. Brouwers and B. Scheer (2016), De flexibele schil van de Nederlandse arbeidsmarkt: een analyse op basis van microdata, CPB.
[9] Bosch, N., M. De Graaf-Zijl and D. Van Vuuren (2015), Position paper t.b.v. ‘IBO Zelfstandigen zonder personeel’, CPB.
[3] De Beer, P. and E. Verhulp (2017), Dertig vragen en antwoorden over flexibel werk, Amsterdam Institute for Advanced labour Studies.
[1] de Graaf-Zijl, M. and E. Berkhout (2007), Temporary agency work and the business cycle, SEO Economic Research.
[14] Eurofound (2015), New forms of employment, Publications Office of the European Union.
[10] Ministerie van Financiën (2015), IBO Zelfstandigen zonder personeel, Ministerie van Financiën.
[5] PPMi (2017), Study to Support Impact Assessment on the Review of the Written Statement Directive, European Commission.
[11] SEOR (2013), ZZP tussen werknemer en ondernemer.
[12] Statistics Netherlands (2017), cbs opendata, https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83686NED/table?ts=1518193405374 (accessed on 28 July 2017).
[8] Statistics Netherlands & TNO (2017), Zelfstandigen Enquête Arbeid, Ministry of Social Affairs and Employment.
[13] UWV (2010), Groei van de flexibele arbeid en de gevolgen voor het beroep op WW, Kenniscentrum UWV.
[4] Van der Werff, S., L. Kroon and A. Heyma (2016), Beslag uitkeringen personen in flexibele schil 2006-2014, SEO.
[6] Vermeulen, H. et al. (2016), Uitzendmonitor, KBA.
Notes
← 1. This category consists of temporary workers who indicate that there is a prospect of becoming a standard worker. Since this is an indication based on the perception of the worker, the certainty of this prospect is unclear.
← 2. The combined share of temporary, agency and on-call workers among 15-24 year-olds increased considerably in the past decade, from 40% in 2003 to 65% in 2015. For 25-34 year-olds it rose from 12% to 25%. For all other age groups and for own-account work the prevalence of non-standard work was quite stable over 2003‑15.
← 3. Summary dismissal is possible in the case of gross misconduct by the employee.
← 4. The wage used as a basis for the amount of benefit payment.
← 5. An Internet search on the term “7+8+8 regel” gives 1 300 results, especially for HR consultancy firms.
← 6. These two shares do not add up tot he overall share of 71%, because some CLAs have a seperate deviation from the chain rule for both specific groups and all employees.
← 7. The temporary agency sector has two collective labour agreements. The biggest – the ABU CLA – does not extend the 26 weeks. The smaller one – the NBBU CLA – does extend the period to 78 weeks.
← 8. By law, this clause can apply only for the first 26 weeks of the agency agreement, but deviation by CLA is allowed. In the NBBU CLA this period has been extended to 78 weeks.
← 9. Since working hours vary in on-call contracts, the gross salary is based on the average monthly working hours for the calculation of the severance payment.
← 10. Idem.
← 11. This has been the case since 2012. Before then, only the rules concerning protection against the most severe dangers and creating dangerous situations to other people was in place for own account workers.
← 12. Own-account workers who started from a situation of dependent employment have the right to continue their insurance at their previous insurance agency. This is an opt-in situation, not an opt-out, as a result of which only 2% use this option (Bosch, De Graaf-Zijl and Van Vuuren, 2015[9]).
← 13. The following criteria must be met to be eligible for the IOAZ benefit: 1) Age between 55 and 65. 2) At least 10 years of working history as a self-employed person or three years self-employed after seven years as an employee. 3) At least 1 225 working hours per year as a self-employed person (this can be shared between partners of a household). 4) Income of no more than EUR 23 938 per year in the previous three years (including income of partners in a household). 5) Expected future income less than EUR 25 443 (as estimated by the municipality). In some cases partially disabled self-employed people are also eligible for the IOAZ, but this is determined on a case-by-case basis.
← 14. Criteria for misclassification of self-employment are: 1) the worker in question cannot freely choose who actually performs the work; 2) the employer has authority over the worker; 3) the worker receives a wage, which is defined as compensation for his work exceeding cost reimbursement. If these three criteria apply, the worker is not classified as self-employed, but as an employee.
← 15. The second criteria (the employer having authority over the worker) is the most diffuse, since the existence of authority is hard to objectively determine and context dependent. For example, the following situations suggest that the employer has authority over the worker: The employer decides how, when and where the job is performed; the employer decides on the amount of working hours/days per week; if the job requires team work, the employer can influence this co-operation and the worker gets the same instructions as any other employees at the firm.