A leap forward or a ride on a carousel?

Ilse Julkunen

This article discusses and examines the potential of an active labour market policy in contributing to young people’s integration into paid work or education. It is based on a comparative study of a total of 7 800 young unemployed people in five countries: Finland, Denmark, Iceland, Norway and Sweden.1

There are many problems involved in exploring the potential of active labour market policies. First, the measurable outcomes are difficult to establish. Are we content with having wage labour as the critical hallmark of integration, or do we need more sophisticated measurements? Secondly, labour market schemes vary on local, national and cross-national levels when it comes to structures and approaches, and therefore we are not sure if we are comparing like with like. Thirdly, from a comparative perspective, the outcomes need to be contextualised in the different labour market conditions in the various Nordic countries.

 

The labour market context in the Nordic countries

It is evident that the current labour market situation strongly affects the chances of integration into working life. In 1995 – the first year of our study- Finland and Sweden showed an increase in the work force for the first time since the recession in the early 1990s (Employment Outlook, 1996). Similarly, Norway, Iceland, and even Denmark to a certain extent, had shown an increase, but earlier in the decade. This favourable labour market development continued in 1996, with the exception of Sweden, where unemployment rose and employment declined. Large differences in employment levels remained, however.2

Compared to Europe, unemployment is rather a new phenomenon in Scandinavia, but there are also differences between the Nordic countries. The unemployment situation in Denmark and Norway differs from that in the other Nordic countries on one essential point: these countries experienced an increased unemployment before the recession in the 1990s. The unemployment level in Denmark rose considerably after the second oil crisis at the beginning of the 1980s. This recession was delayed in Norway due to its oil boom, but even here there was a sharp increase in the unemployment level after 1985, and it has been rising particularly among young people since 1987 (Halvorsen & Marklund, 1993). The unemployment situation in the Nordic countries can be differentiated by using two dimensions: unemployment history and unemployment level. Unemployment history can be divided into short and long-term experience. Short-term experience refers to the increased unemployment level mainly after the recession of the 1990s, whereas long-term experience refers to an increase in the level since the second oil crisis. The current national unemployment levels can be categorised as high or low by comparing it with the EU average of 9 per cent. Combining these two dimensions produces following groups:

 

Figure 1: Unemployment dimensions in the Nordic countries.

 

	LOW unemployment level	HIGH unemployment level
SHORT unemployment history	Iceland Sweden	Finland
LONG unemployment history	Norway Denmark

 

Iceland has the shortest history of unemployment and the lowest unemployment level in the Nordic countries. The case of  Iceland is unique in that unemployment was practically non-existent before the 1990s. Finland experienced an unemployment peak in the 1970s, but this was levelled out due to economic ties with the Soviet Union. Unemployment increased dramatically after the recession in the 1990s, and such high unemployment is thus a relatively new phenomenon. Sweden was hit by the crisis somewhat after Finland. Thus, both Finland and Sweden can be categorized as having a short unemployment history, but although Finland currently has a much higher unemployment level than Sweden. Norway still has a relatively low level - even by international standards - and Denmark, generally speaking, has a slightly higher level than Norway. What also unites these two countries is that the number of long-term unemployed is higher (26–28 per cent) than in the countries with a shorter unemployment history, which points towards the difference between cyclical and structural unemployment. Still, Finland deviates from this pattern in that the number of long-term unemployed is even higher (37 per cent) and resembles that of the rest of Europe. 

 

Shifting trends in labour market policies

Nordic labour market policies seem at first sight quite convergent, since there has been a growing trend from passive income transfer to active measures, and also tendencies to strengthen both incentives and duties by tightening social security. Still, the countries have different emphasis. Active measures3 are being applied most vigorously in Sweden and Norway. Denmark has intensified its activating policies in the 1990s while developments in Finland have fluctuated from increasing activating policies in the 1980s, abandoning such obligations in 1993, and increasing them again in 1996 for young people under 20, and in 1997 for young people under 25 years. With the rise of unemployment in the 1990s Iceland has employed more activating policies. A comparison of  the costs of active versus passive labour market policy measures as percentages of GNP (Employment outlook, 1996) shows that Norway and Sweden can be categorised as active countries and Finland and Denmark as passive. There is no comparative cost information available for Iceland.

Even though unemployment benefit systems are largely based on insurance-based benefits, there has been a shift towards low level basic security and means testing for the young. These changes were largely brought about by the structural challenges of high unemployment and an increase in atypical employment. In all the Nordic countries, basic security is combined with income-related security, but there are differences in emphasis on these two dimensions.  Denmark has come quite close to a basic security model4 with unusually generous benefits and easy access to the system. It is also worth noting that Denmark is the only Nordic country which has not lowered the compensation level for unemployment. Sweden, Norway and Iceland still have an institutional social policy system, with combined income-related and basic security, while the Finnish institutional model for the young has changed to a combined basic  and means-tested system, as very few are eligible for income-related benefits.

Developments in the Nordic labour market policy for young people can be described by the ‘sticks and carrots’ metaphor and can be summarised as follows:

The criteria for insurance-based unemployment benefit have become stricter. Many young people have already ceased to qualify for universal benefits because of atypical employment (in all the Nordic countries except Denmark).

The means-testing criteria for flat-rate benefits and social assistance have been tightened (Sweden, Finland, Denmark).

The level of means-tested benefits has gone down (Finland, Sweden, Denmark). The benefits have not been index-linked for many years, meaning decreased spending power (Finland). This decreased level of benefits has mainly hit those living with their parents (Finland, Denmark).

Young people under 25 can no longer receive unemployment benefit without being activated. They have both the right and the obligation to accept a job offer or a training opportunity (Finland, Sweden, Norway, Denmark).

 

Why, then, should employment measures have any effect?

Employment measures aim generally at increasing labour market flexibility and maintaining and strengthening the individual’s working capacity with a view to improving his or her chances of finding employment. There is continuing debate on whether these activating programmes do, in fact, have these effects. The schemes are frequently criticized for having less ambitious functions, such as cleaning up the statistics or just storing the workforce (cf. Hyyppä, 1999). The debate, however, is seldom based on empirical facts – which is understandable given the fragmented body of knowledge in this field. An overview of previous studies on the effect of employment measures proves that the results are highly controversial and difficult to compare. Choices of control group, surveys, research design, methods and measurements vary. One might say that there is some unity in that it is difficult to prove beyond reasonable doubt the positive and negative effects of labour market policies. The main arguments put forward in Nordic studies on the effects of different labour market programmes are summarised below.

Qualifying measures, i.e. training (education), apprenticeships and job training in the private sector, as opposed to more traditional public activation, seem to produce more positive effects (Schröder, 1991; Try, 1994; Aho et al., 1996; Calmfors, 1996; Mikkonen, 1996).

 The duration of the activation programme also seems to affect the end result, since short as opposed to long educational courses have been found to produce less beneficial results (Torp, 1992; Madsen, 1994; see also Korpi, 1994).

If the measures are part of a long-term strategy or development plan, then the probability of success increases (Sehlstedt & Schröder, 1989).

 Positive results are produced not only through relevant education, but also because of  the fact that job training offers connections with potential employers (Rosdahl, 1996).

Experience in Sweden has shown that results depend highly on whether the initiative is part of a concrete and individual action plan (cf. Madsen, 1994; Schröder, 1994)

Even though the aims of labour market initiatives are integrative, there are still many side effects, and also the risk of marginalization. Research (Edin & Holmlund, 1991; Furlong, 1993; Hammer, 1996; Spies, 1996) has shown that the measures per se might form employment traps. Those who participate in programmes constantly circulate between initiatives and unemployment. The probability of finding employment has also been shown to diminish the more the person has participated in different programmes (Korpi, 1994). A contributing reason for this is that activation qualifies the unemployed for a renewed period of benefits. This kind of circulation is especially common in Sweden (Björkman & Harkman, 1995).

Evaluation of different schemes for unemployed youth has mainly been undertaken in a national context. We lack thus comparative knowledge of how schemes function in different settings. It is also important to evaluate schemes not only as labour market measures but as a social policy measure, which may help young people to cope with problems related to unemployment. Furthermore, increased educational motivation and return to education is also an important outcome.

 

Research methods and research design

This research is based on a Nordic survey on youth unemployment. The samples were drawn from national unemployment registers, and the respondents were young people between the ages of 18 and 25 who had been unemployed for a period of at least three months over the previous six months. The surveys were carried out in late 1995/early 1996, and were initially based on postal questionnaires, with additional telephone interviews carried out to minimise the bias caused by a skewed response rate. The postal questionnaires were completed six months after the sampling, at which time some of the young people had found jobs, entered schemes or returned to education, while others remained unemployed. The response rate in the survey reached 73 per cent in Finland, 60 per cent in Iceland, 56 per cent in Norway, 63 per cent in Sweden and 79 per cent in Denmark. The attrition analyses conducted for the countries involved showed that, all in all, the material is well balanced and that there is no need to correct skewness.

This article analyses the role of labour market schemes in relation to young people’s subjective experiences, and to the outcomes of employment or education, in both a short-term and a long-term perspective. The main research design (cf Carle, 1997; Carle & Julkunen, 1998) can be illustrated as follows:

 

Figure 2: Research design

 

 

 

 

     

 

 

 

 

 

 

 

 

At the starting point in spring 1995, all of the young people in our study were unemployed. Approximately a half a year later the situation had changed. In what way did the active measures influence the outcome in terms of the occupational status of the young at the time of the interview? Logistic regression was used to predict the occupational status. The study also analyses the short-term effects of the active measures from the responses to a question about the occupational status  three months after the activation

 

Something to do? Young people’s experiences of employment schemes

In our study, about half of the Swedish, Norwegian and Finnish subjects had participated in employment-enhancing measures at some point, but only a minority of the Danish and Icelandic subjects had done so. The numbers participating corresponded to the current employment policies of the countries. Sweden is famous for its active labour market policy, whereas in Norway the obligation to be active was introduced later, in 1994. Even though Finland can be characterized as a country with more passive labour market policies, a large number of young people had still been activated. This was due partly to the high level of unemployment, but also to increased investments in youth unemployment measures during the past few years, particularly since 1996 when young people under 20 were covered by activation measures. Denmark currently has a recognised activation policy, but this clearly has less effect on the insured young people who comprised the Danish group in the study. Moreover, the effect of the new activation policy cannot yet be seen since the reform came into force only in April 1996. Unemployment is a new problem in Iceland, and the number of young people participating in the measures was low.

The Norwegian and Swedish young people in our study had taken part in activating programmes more often than those in the other Nordic countries. One third of them had done so more than three times during their lives. When we studied the effect of the background factors, such as education, interrupted education and place of residence, on participation in labour market programmes, we did not find significant differences between those who had been activated in different programmes and those who had not. This indicates that activation is rather the norm, a natural part of the lives of young unemployed people in Scandinavia, just as unemployment has become a normal part of their labour market careers. In the transition period from youth to adulthood, new institutions between school and work have emerged and their importance seems to be growing. This ‘grey sector’, between school and working life, consists of employment schemes, apprenticeships and separate employment projects (cf. Nyyssölä, 1999). We did, however, uncover interesting differences between those who had repeatedly been activated and those who had participated only once in labour market programmes. The analysis shows that those who had been activated more frequently formed a more exposed group, and featured more young people with a background of only compulsory school, interrupted studies and longer unemployment experiences.

In general, the young people in all the Nordic countries reported positive experiences of labour market initiatives: they thought that they had learned new things, that they had been given something to do, that the job/education was interesting, and that the atmosphere at work was good. These positive experiences seemed to be related to the unemployment level, since the least positive experiences were found in Iceland and the most positive experiences were found in Finland. Still, the results were somewhat ambiguous. What the young people who were interviewed appreciated most was that the programmes gave them something to do, which implies that they have an important social function – to participate in schemes is a way of participating in a society in which work is the norm.  Or, as one young unemployed woman phrased it: ’Only to be able to say to people: I am working.’5 The labour market programmes also have a social function in that they create the opportunity to be socially active, and also to reflect on one’s own situation compared with others. This can strengthen one’s self-confidence, as Josef implies:

 

It was really good to join the computer course and see what kind of people are unemployed. That it is people who are not stupid. It can really be anybody and they are extremely productive people who are outgoing and not unstable, but people who are smart and nice.6

 

However, another distinctive feature was that the majority (about 80 per cent) disagreed most with the statement that the measures were too demanding. This tells us something about the quality of the employment measures, and that there is a discrepancy between the character of the employment/education opportunities and the resources and wishes of the young. The results are concordant with previous studies that have highlighted the need for developing the programmes and extending the competence of employment offices and officers (Virtanen, 1996). We could also raise the question as to whether this phenomenon is connected with volition: to what extent can young people themselves choose their employment place and to what extent is coercion used. Previous studies (e.g. Schröder, 1994) have emphasized the importance of the schemes being an integral part of an individual plan. An individual plan also implies an important psychological rationale: the importance of being able to choose one’s situation. This dimension has been acknowledged by Carstens (1998) in her study of activation in Denmark, where she implies that it strengthens self-value and identity development to believe that one has chosen one’s situation.

 

Employment and educational prospects

The analysis of employment prospects is divided into two parts. The first concerns employment three months after completing a programme, and the second is based on a logistic regression where the impact of factors affecting the probability of being employed was taken into account.

 

Table 1. Employment status three months after completing a programme by  country; percentages

 

Per cent	Finland	Iceland	Norway	Sweden	Denmark
in work	13	47	29	30	38
unemployed	69	35	46	54	30
in education	8	14	10	9	7
activated	2	5	6	4	6
other	2	0	5	3	14
participating right now	11	5	7	5	6
total	100	100	100	100	100

 

Three months after completing employment schemes, the majority of the subjects in Finland and Sweden, where the levels of unemployment were higher than in the other countries, were unemployed. The difficult employment situation in Finland was clearly visible: only 13 per cent of the subjects had found a job, 8 per cent had begun some form of education, and as many as 69 per cent were still unemployed. This shows that labour market schemes do not function well in times of high unemployment. The chances of finding a job were best in Iceland and Denmark, where respectively 47 and 38 per cent of the young had done so. Despite economic growth and a high demand for labour in Norway, surprisingly few had entered work there, which points to different selection mechanisms in the inflow and outflow of unemployment. At low unemployment levels there is stronger inflow selection, and those who become unemployed stay unemployed for a longer time. Structural unemployment affects the demand for labour. Thus, the qualifications of the young Norwegian people probably affected their chances of finding a job. Still, this does not seem to be the case in Denmark, where fewer people ended up in unemployment. An interesting feature, however, is that a much higher number of young people had found an alternative occupation in Denmark. This phenomenon seemed to apply mostly to women, who had either gone on maternity leave, stayed at home, or participated in specific activation projects.

When all the explanatory factors were taken into consideration (Table 2), it was found that the chance of finding a job was highly dependent on the unemployment history of the young person concerned. A shorter period of unemployment significantly improved the odds of becoming employed again. Logically enough, previous work experience also affected the probability of finding work. Those who had no experience stood a significantly worse chance of finding a job than those who had more working experience. The women in Denmark were less likely to find employment than the men. Other background factors did not seem to significantly affect the probability of finding employment, even though education can be said to have a positive effect: the higher the level of education, the more likely the person was to be employed within three months.

 

 

Table 2. Probability of becoming employed three months after participating in a labour market programme.

0=unemployed/participating in labour market programmes 1=employed

	Finland (n=697)		Iceland (n=281)			Norway (n=346)			Sweden (n=1 153)			Denmark (n=302)
	B	expB	B	expB	B		expB	B		expB	B		expB
Gender
Women/ reference: men	-0.1582	0.8537	-0.04	0.9564	-0.0216		0.9787	-0.0113		0.9888	-0.4506		0.6373***
Age group
Reference: 22-24 years old
18-19 years old	-0.2026	0.8166	-0.21	0.8133	-0.3169		0.7284	0.1150		1.1219
20-21 years old	0.2084	1.2317	-0.04	0.9647	-0.1609		1.1788	0.0892		1.0933	0.0229		1.0232
Education
Reference: university
No qualifications	-3.70	0.0245	0.2821	1.3259	-6.0234		0.0024	-3.9449		0.0194	-1.8007		0.1652
Compuls. school	0.1444	1.1553	-0.10	0.9010	0.1616		1.1754	0.4616		1.5866	0.3594		1.4325
Vocational train.	0.6316	1.8805	0.0997	1.1049	2.2639		9.6201	1.1513		3.1624	0.7244		2.0636
High school	0.9463	2.55762	-0.28	0.7554	1.3873		4.0039	1.1519		3.1641	0.6051		1.8314
Work experience
Reference: >2 years
No experience	-1.6016	0.2016***	-0.2576	0.7729	-0.3515		0.7036	-0.8357		0.4336***
Up to one years	0.3962	1.4861*	-0.29	0.7478	1.4929		4.4500***	-0.19		0.8305	-1.4187		0.2420**
1-2 years	0.1966	1.2172	0.3889	1.4753	-1.0266		0.3582**	0.2008		1.2223	0.7360		2.0876
Unemployment
Short/ reference: >2 years	0.6412	1.8987***	0.17	1.1840	0.6730		1.9602**	0.4058		1.5006***	0.3978		1.4886***
Residence
Reference: city
Community	0.3493	1.4181*	0.3103	1.3638*	0.1849		1.2031	0.1118		1.1183	0.1569		1.1698
Town	-0.2895	0.7486			0.6721		0.5106*	0.0141		1.0142	0.1089		1.1151
Labour market measures
Reference: employment measure
Labour market training	-2.3463	0.0957	1.8519	6.3716	2.0165		7.5119***	0.6684		1.9511	0.4371		1.5482
Job training	-2.7991	0.0609			2.0741		7.9571***	1.5919		4.9129	0.4371		1.5482
Other measure	-1.7783	0.1689	1.9023	6.7011	0.3152		1.3705	1.5282		4.6096	-0.3543		0.7016
Participation
Once/ reference: more	0.1882	1.2071	0.2900	1.3364	0.7706		2.1610	0.3163		1.3720	0.1148		1.1217
Constant	0.0182		-1.47		-1.6788			-0.68			-1.3531

*p<0.05; **p<0.01; ***p<0.001

 

 

Labour market schemes7 appeared to have a positive short-term effect in all the countries except Finland, with labour market training, job training and employment projects more likely to result in finding work than the employment measures. However, the results were only significant for Norway, where participation in labour market training and job practice clearly improved the chances of finding work. This may have been because the young people who had participated in these programmes had better preconditions. Our study cannot verify whether there is some selectivity, since we do not know for certain what the subjects’ previous qualifications were. Studies in Norway (Torp, 1992) have, however, proved that recruitment may be distorted, especially for labour market training. Another interesting feature is that repeated participation in labour market schemes seems to produce a slightly worse chance of finding work than one-off participation. This was true for all the countries, but especially for Norway.

We were also interested in finding out whether activation could be seen as a generally determining feature of future employment. We chose to limit this analysis (Table 3) to comparisons within the workforce, i.e. those who were studying, at home, or who were in the army at the time, were excluded. The explanatory variables were gender, age, education, place of residence, working experience, and duration of unemployment. We also considered non-participation in programmes.

 

Table 3. The impact of labor-market initiatives on the probability of being reemployed (occupation at the time of the interview), logistic regression 0=unemployed/participating in labour -market programmes 1=employed.

	Finland (n=1 040)			Iceland (n=645)		Norway (n=624)			Sweden (n=1 347)				Denmark (n=779)
	B		expB	B	expB	B	expB	B		expB		B		expB
Gender
Women / reference: men	0.0593		1.0611	-0.07	0.9288	-0.066	0.9358	-0.003		0.9975		-0.2174		0.8042**
Age group
Reference: 22-24 years old
18-19 years old	0.1631		1.1772	-0.05	0.9531	0.0528	1.0542	-0.098		0.9066
20-21 years old	-0.101		0.9040	0.0342	1.0348	-0.161	0.8514	0.0478		1.0490		-0.114		0.8923
Education
Reference: university
No qualifications	-3.70		0.0247	-0.29	0.7506	0.1779	1.1947	0.2744		1.3157		-0.4478		0.6390
Compuls. school	0.1865		1.2050	-0.09	0.9094	-0.689	0.5023**	-0.858		0.4240**		-0.447		0.6396*
Vocational train.	0.8544		2.3500	0.2715	1.3120	-0.28	0.7538	-0.14		0.8653		0.1435		1.1543
High school	0.9959		2.7073	0.6490	1.9137*	0.1163	1.1233	-0.06		0.9381		-0.134		0.9867
Work experience
Reference: >2 years
No experience	-1.06		0.3464***	-0.63	0.5325**	-1.021	0.3601***	-0.77		0.4631***
Up to one year	0.2490		1.2827*	0.1294	1.1382	-0.033	0.9680	0.2124		1.2367*		0.1024		1.1079
1-2 years	0.3482		1.4165*	0.2895	1.3357	0.4559	1.5776**	0.1687		1.1830		-0.6148		0.5407**
Unemployment
Short/ reference: >2 years	0.5614		1.7531***	0.4870	1.6274***	0.5348	1.7071***	0.5188		1.6800***		0.3897		1.4765***
Residence
Reference: city
Community	0.1873		1.2060	0.2507	1.2849*	0.2425	1.2744*	-0.81		0.9218		0.1351		1.1446
Town	-0.064		0.9385			-0.188	0.8289	-0.09		0.9119		0.1526		1.1649
Labour market measures
Reference: employment measure
Not participated	0.4089		1.5052	-0.14	0.8681	0.0971	1.1020	-0.01		0.9898		0.1002		1.1054
Labour market training	-0.258		0.7723	0.00	0.9977	-0.133	0.8751	-0.45		0.6400*		0.0788		1.0820
Job training	-0.354		0.7020			-0.09	0.9138	0.5401		1.7162***		-0.2546		0.7752
Other measure	-0.102		0.9035	0.2967	1.3455	0.3806	1.4632	-0.09		0.9164		-0.3294		0.7194
Participation
Once/ reference: more	0.0766		1.0796	0.3292	1.3898	0.315	1.3641	0.4553		1.5766**		0.1723		1.1881
Constant	-1.632			0.5950		-0.1668			-0.68		-0.408

*p<0.05; **p<0.01; ***p<0.001

 

 

The results of the analysis indicate that it was mainly the structural factors - unemployment and working experience - that most affected the probability of becoming re-employed. This was true for all the Nordic countries, but particularly accentuated in Iceland and Denmark, where the young people had significantly more work experience and stood a better chance of becoming integrated into the labour market. The level of education was also significant in all the countries except Finland, but despite this, level of education still seemed to be a positive factor in finding a job in Finland. These results are, thus, concordant with a study by Korpi (1994), who found that moving from unemployment to employment depends largely on the educational level of the person and the duration of the unemployment.

The effect of the different labour market programmes was not significant except in Sweden, where job training was of central importance. Those who had participated in job training thus stood a better chance of becoming employed. Those who had participated in labour market training, on the other hand, had a slightly worse chance of finding work. However, those who had participated in labour market training had a better chance of getting into education (see Table 4).

Table 4. The effect of labour market measures on the chances of education.             0=other 1= education.

 

	Finland (n=1 435)				Iceland (n=955)		Norway (n=914)			Sweden (n=1 804)		Denmark (n=1 011)
	B		expB	B		expB	B	expB	B		expB	B	expB
Gender
Women / reference: men	0.1582		1.2044**	-0.2759		0.7589**	-0.1824	0.8332	0.1659		1.1805*	0.0926	1.0970
Age group
Reference:22-24 years old
18-19 years old	-0.081		0.9222	0.4170		1.5173**	0.0908	1.0950	0.0419		1.0427
20-21 years old	0.2036		1.2258	-0.319		0.7269	0.2599	1.2968*	-0.0069		0.9931	0.1641	1.1783
Education
Reference university
No qualifications	0.2324		1.2616	-4.563		0.0104	-3.6421	0.0262	0.1744		1.1906	-0.7347	0.4797
Compuls. school	-0.4936		0.6104*	0.8510		2.3421	0.2322	1.2614	-0.3362		0.7145	0.2923	0.7466
Vocational train.	-0.52		0.5957*	1.0706		2.9171	0.2895	1.3358	0.5001		0.6065*	-0.5248	0.5917
High school	0.9726		2.6447***	2.0610		7.8536	1.2019	3.3265	0.4437		1.5585	1.1235	3.0755***
Work experience
Reference >2 years
No experience	0.1944		1.2146	0.717		2.048**	0.6098	1.8401***	0.2640		1.3021*
Up to one year	-0.0298		0.9706	0.1123		1.1188	0.1430	1.1537	0.0822		1.0857	0.2084	1.2317
1-2 years	0.2468		1.2799	-0.5153		0.5974**	-0.2174	0.8047	-0.0759		0.9269	-0.2418	0.7852
Unemployment
Short/ reference: >2 years	0.3426		1.4086***	0.3419		1.4077**	0.4149	1.5142***	0.2199		1.2459**	0.2585	1.2317*
Residence
Reference: city
Community	-0.3009		0.7401*	-0.3947		0.6739	-0.2985	0.7419*	-0.4987		0.6073***	-0.3139	0.7306*
Town	0.0128		1.0129				-0.1328	0.8756	0.1365		1.1462	0.3438	1.4103**
Labour market measures
Reference employment measure
Not participated	-0.1408		0.8687	-0.3947		0.6739	0.4285	1.5350	-0.1811		0.8344	-0.2975	0.7426
Labor-market training	-0.168		0.8463	0.5466		1.7273	0.3097	1.3630	0.3622		1.4365	0.1303	1.1391
Job training	0.4452		1.5608*				-0.4154	0.6601	0.2798		1.3229*	0.8227	2.2766
Other measure	0.1804		1.1977	-0.0520		0.9493	-0.3664	0.6932	0.2544		1.2897	-0.0764	0.9264
Participation
Once/ reference: more	0.2580		1.2943	0.430		1.5373	0.0130	1.0131	0.2250		1.2523	0.2984	1.3477
Constant	-1.2279				-2.946		-2.410			-1.4717		-1.8443

*p<0.05; **p<0.01; ***p<0.001

 

Repeated participation seemed to affect negatively the probability of finding work. The results were significant only in Sweden, however, where those who had been activated only once stood a better chance of becoming employed than those who had been activated repeatedly. The probability of finding employment seems, therefore, to diminish the more the person has participated in different programmes.

The results were somewhat different when the effect of labour market measures was related to the chances of further education. Women stood a better chance than men in this area in Finland and Sweden. The Icelandic women stood a worse chance, which was probably connected to the fact that a large number of them had children. Background factors such as level of education and place of residence were significant in all the Nordic countries, except Norway. The chances of being accepted for further education diminished in sparsely-populated areas; and the higher the level of education, the better the chance of being accepted for education. Work experience affected negatively further education possibilities - those who had no work experience stood a better chance of getting into educational programmes in Iceland, Norway and Sweden. Interestingly enough, participation in labour market schemes had a slight positive effect in all the Nordic countries except Norway. Job training, however, was of significance only in Finland and Sweden.

 

Addressing the elements of policy...

Unemployment and work experience as such were decisive in determining employment prospects.  This result is crucial in relation to how society itself explains and handles misfortunes. According to Therborn (1995), the culture of the policy-makers is deeply ambivalent about the value of (un)employment. To governments, unemployment is seen as failure, yet the success of economic policy is defined with complete disregard of employment. National economic institutions, ideologies and policies have thus lost their significance and the impact of international markets has increased. Although risks have also increased and become more global, paradoxically, they have also become more individualized. The individualization of modern society means, in some way or other, that situations – in this case unemployment – which would once have led to a call for political action are now interpreted as something that can be solved on an individual level through personal action. 

Does policy matter? Therborn (1991) has come to the general conclusion that institutional differences between different kinds of welfare states are the main explanation for differences in unemployment. In particular, it is an institutional commitment to full employment that matters and not so much anti-inflation or growth policies. Comparative research from the 90s (Esping-Andersen, 1996) has, stressed, however, that it is not only the aim of full employment, but social investments in employment, which seem to count. Were there differences according to the different labour market contexts? Did the countries with a more stressed activation policy have more positive employment prospects, and did labour market programmes make a difference?

Labour market programmes appeared to be influenced by the national context in that the schemes have different effects in the different countries. Generally, employment schemes seemed to have positive effects in active countries such as Sweden. In Norway, however, positive effects could be seen only in the short term and, contrary to expectations surprisingly many ended up unemployed. Norway, however, differs from Sweden in one essential way: it has a longer unemployment history and a higher number of long-term unemployed. Unemployment history, then seems to offer one explanation for the difference for Norway. However, this is not the case for Denmark, as fewer young people ended up unemployed. The young people in Denmark however, found alternative routes to a much higher degree, as they were either on maternity leave, stayed at home or participated in specific projects. One explanation for this phenomenon could be the more family-friendly policy, of which the leave-of-absence arrangements are a good example. Iceland and Finland seem to form the extremes of Scandinavia. Iceland has a short unemployment history, a low unemployment level and a favourable labour market, and young people there had found work to a much higher degree than those in the other countries. The opposite trend was found in Finland, the country with the highest unemployment level, and a rather passive policy, where there was higher selection in terms of who exits from unemployment, and very few entered the labour market.

 

... volition and competence

As far as the various programmes are concerned, the study revealed that job training had positive effects on the chances of getting a job. It seemed not only to offer relevant work experience, but also to provide connections with potential employers. These results are in accordance with Raffe (1987), who argues that the training content of the schemes is less important than the context in which the training is provided. Could this also explain why employment measures in the public sector appear to produce less good results?  Labour market training did not affect employment chances, but it appeared to affect positively the chances of a return to education.  It appears to enhance the motivation to learn in all the countries, regardless of whether the country has active or passive policies, or have low or high unemployment. One could then argue that activation works as a ‘qualitative push’, and that schemes may in fact strengthen the competences of young people. Still, we cannot say for certain that they will end up in work after a completed course of education.

Circulation in labour market schemes had negative effects in all countries, which points towards the unintended effects of activation policies. These may be related to employers’ recruitment practices, and the group concerned might be more vulnerable. The negative effect raises several questions. First, is labour market policy the right answer to this ‘problem’?  Other solutions might be more appropriate. An alternative could be a more family-friendly policy; support for families, day-care services (also for the unemployed), social policy, housing policy and social work. There are dangerous trends in the introduction of a market environment in the social sector. Because of the pressure for some kind of ‘success rate’ with regard to employment, there is a risk that the most vulnerable groups might be expelled not only from the labour market, but also from professional counselling processes.  However, as much as there is a risk in not looking for alternative tools, there is also the danger of a totalitarian tendency in that we are trying to integrate young unemployed people in one way or another. This leads us to the second question: should employment be stressed as the only ultimate end result? Generally, the aim of activation is to integrate unemployed people into the labour market, but many programmes also strive to enhance personal development – in different degrees. Still, the programmes are mostly evaluated only according to the mechanical criteria of employment; personal gain and personal development are seldom highlighted. The positive experiences of labour market programmes reported in this study indicate that they have values other than just improving employment. This is in line with previous results from the study revealing that those on training schemes have better mental health (Hammer 2000). An enhanced self-esteem, which was found particularly among the Finnish young people, strengthens personal competence and may as well be an important trigger to future labour market intergration. This implies that social aims should not be overlooked. But, as Carstens (1998) has questioned, should there also be a limit to how long people can circulate in activation if it does not lead to a higher quality of life? It is equally important to leave room for choice – including the choice not to be integrated – as it is to provide options or alternatives.

 

A final note

The success of labour market programmes seems to be embedded in the national context, but on different levels; on a national level (the labour market), on a practical level (policy and institutions) and on an individual level. We can learn from other countries, but in order to understand the function of these measures we need to thoroughly analyse them in accord with Therborn (1995), from two opposing sides: that of the policy-makers (employers) and that of the policy-takers (the unemployed). In this study, the viewpoints of the policy-takers have been analysed and the results indicate that activation can make a difference. However, it seems that it needs to be strong, as in the case of Sweden, or in the form of broader social investments as in the case of Denmark. Still, not all schemes are successful in reaching employment aims. The context appears to be important. But is the crucial question whether the context is the public or the private sector, or rather the ability to create jobs and opportunities where the unemployed can fulfil and develop their competences? Rothstein (1998) has pointed out that an active labour market policy should always be a quality product and subsidies should not be limited to specific jobs and sectors. Labour market programmes have then some potential, but they should not be implemented at any cost. Alternative tools and real choices must be stressed. Still, as Julkunen (1998) has argued, labour market policy alone is not enough without a receptive labour market. This could well be seen in the case of Finland. Ultimately it is the pressure and anticipations from the labour market that affect the outcome. It seems though that too little attention has been paid to the viewpoints of the employers or public officials on the issue of labour market programmes. And a crucial question remains: in what way is (and is not) labour market flexibility enhanced through labour market programmes?

 

Acknowledgements: I am grateful for the comments of the referees and the editorial board. This article has been made possible through the funding of the EU TSER programme and Svenska Kulturfonden.

 

Notes