National Institute Economic Review · 2012-09-20 · between Danish activation policies and flexicurity on the one hand, and economic and social performance on the other, are lacking

http://ner.sagepub.com

National Institute Economic Review

DOI: 10.1177/0027950107086172 2007; 202; 99 National Institute Economic Review

Jon Kvist and Lisbeth Pedersen Danish Labour Market Activation Policies

http://ner.sagepub.com/cgi/content/abstract/202/1/99 The online version of this article can be found at:

Published by:

http://www.sagepublications.com

On behalf of:

National Institute of Economic and Social Research

can be found at:National Institute Economic Review Additional services and information for

http://ner.sagepub.com/cgi/alerts Email Alerts:

http://ner.sagepub.com/subscriptions Subscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

© 2007 National Institute of Economic and Social Research. All rights reserved. Not for commercial use or unauthorized distribution. at Socialforsknings Institut on November 2, 2007 http://ner.sagepub.comDownloaded from

http://www.niesr.ac.uk

http://ner.sagepub.com/cgi/alerts

http://ner.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav


� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ! " # # $ � % & ' � � ( ) * � � + � $ � , " ' � - � � . � � � & � / � � " � 0 " $ & � + � � � � � & � / � ! � � � � � � � " � � $ � � � � � � " � , � & � ' + � " #� � � � 0 � � & � / � � � , $ � ( , " � � � " & ' � , ' � � - 0 $ " ( - � � � � � � � " , " " + � � - 0 $ " ( - � � � 1 2 � � � � � ) � � � " , 3 � & � / � + ) � � � ! � � ! � " ' 0 + � � �� + ! � " # � & � / � � " � 0 " $ & ( � � / � ' � � � � ! " � � � � � - � � . � , $ � � � � � + # " � - � � " � " / � � � � � 0 � + � # # � � � � ( � � � + 1 4 � " - � � � � � � $ & ' � , � !" # ' � � - 0 $ " ( - � � � � - " � ! � � � � + ' � � � ' � � - 0 $ " ( � � � � � " ' ! � + � � � � � � � � & � / � � " � " # # � � + ) � � � � - � " � � ( + � $ + " � " � & � � � + �$ � , " ' � + ' 0 0 $ ( � - " � ! � " � 5 � & " � " - & � & � / � ! � " ' 0 + ' + � ! � � / � ' � $ $ ( � � $ " � � � 0 � " ! � � - - � + 1 * � � + � � & � / � � " � 0 " $ & � + � � ' +� - , � � & � � " � " � $ ( � & � / � $ � , " ' � - � � . � � 0 " $ & � + , ' � � $ + " + " & � $ � � � � � � ! � � � " � 0 " $ & � + 1 * � + 0 � � 6 � � + 0 � � � � 0 " 0 ' $ � � � ( � � �, � $ � # � � � � 0 " + � / � � # # � & � + " # � & � / � � " � ) $ � � $ � + � & � ' � $ $ ( . � " 6 � � , " ' � � + " / � � � $ $ - 0 � & � " � � � � * � � + � � & " � " - ( 17 � ( 6 " � � + 8 9 & � / � � " � 0 " $ & � + : # $ � % & ' � � ( : � - 0 $ " ( - � � � � # # � & � + : * � � - � � .; < = > $ � + + # & � � " � + 8 ; ? @ : ; ? A

B C D E F G H I J K L M H N J O I H P Q R I N S R T O S C O U J H P V R K R H S U L W R X Y H J P Z [ \ ] K T J W ^ _ W

` a b c d e f g b h d aLabour market activation policies (referred to henceforth asactivation) form a crucial part of Danish economic policy.From initially combating unemployment in the mid-1990s,the chief aim of activation policies in the 2000s has nowshifted to that of increasing labour supply. Denmark todayhas not only the lowest unemployment rate in the EuropeanUnion but also the highest employment rate, outperformingthe United States (see table 1). No wonder therefore thatthis employment performance makes Denmark’s activationpolicies the focus of other European governments,especially those of continental Europe.

Many believe that new activation policies from 1994onwards have contributed to a marked fall inunemployment in Denmark. The combination of these

activation policies, lax employment protectionlegislation and encompassing social security has createdwhat is seen as a golden ‘triangle’, often described as aparticular flexicurity model. This model focuses not onemployment protection, i.e. making it difficult for anemployee to be fired, but on social and job protection,i.e. combining high unemployment benefits with activelabour market policies. Throughout the European Unionthere is now a movement towards this ‘Danishdirection’, as witnessed, for example, in the recentEuropean Commission (2007) Communication, ‘Moreand better jobs through flexibility and security’.

This ability to meet both economic and social objectiveshas aroused international interest. In relation to otheri j k l m n o p q m r s l t u r m q v j q w m r s l t u r m q v x j v m y z q { m q r j x | } ~ m x r j q u } v � m p q z v m w � z q � w t r j q w v � m p q z v m w � v j v m y }n � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ¡ � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ¢ � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� £ � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ¤ � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ¥ ¦ § ¨ © ª « ¬ § ¨ ¦ ® ¯ ® ° ± ² ² ³ ¯ ´ µ



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �parts of Europe, Denmark and the other Nordiccountries have a higher degree of equality between menand women, a lower share of people at risk of poverty,and less inequality between rich and poor.Economically, international rankings by the WorldEconomic Forum (2006) and others place the Nordiccountries in the top ten for competitiveness.

Because of stronger economic performance and moreinnovative activation policies, Denmark has in recentyears overtaken Sweden – where trade union economistsGösta Rehn and Rudolph Meidner had developed activelabour market policies from the 1940 to 1960s – as amodel for others. Thus Europe is now looking toDenmark, not Sweden, for solutions to problems ofunemployment and lack of labour supply.

In reality, however, empirical evidence on the linkbetween Danish activation policies and flexicurity onthe one hand, and economic and social performance onthe other, are lacking.

This paper sheds light on the design and development ofDanish activation policies as well as their effects onindividuals and the economy, to the extent that thesehave been investigated.� � � � � � � � ! � ! � " # � $ $ % & ! & � ! # 'Denmark has a population of 5.4 million people, ofwhom 3.7 million are of working age, i.e. 15–66 years,see Box 1. The labour force consists of 2.9 millionpersons, 2.8 million of whom are in employment and105,000 of whom are unemployed. Nearly all of the800,000 persons between 15 and 67 years out of thelabour force rely on social security.

In a country with a high degree of economic openness,activation has become a necessary structural tool toestablish stable macroeconomic development. This canbe illustrated by the present employment situation in

Denmark. At 3.3 per cent (July 2007), the unemploymentrate is so low that it defies prediction by many labourmarket economists. Due to labour shortages, many firmsfind it difficult to meet the demand for products andservices, and collective agreements in 2007 have beenmarked by demands for significant wage increases. Inthis situation activation becomes an important means ofensuring that as many as possible work, that those out ofjobs are available for work, and that the relativelygenerous social protection system retains its legitimacy.

Activation policies enjoy political and popular supportin Denmark. In Parliament there is a tradition for broadpolitical compromise over reform of social and labourmarket policies. Social partners have a large role in theplanning and implementation of activation policies.Professional case officers working with these policiesbelieve in its merits. Three out of four participants aresatisfied with their offer of activation, and thepopulation at large supports these labour marketpolicies.

The rationale behind activation is twofold. First, awelfare model like that of Denmark needs a very highlevel of employment to be able to offer and financepublic benefits to the extent and quality necessary tomeet the demands of the population. Second,employment has proven again and again to be the safestway for the individual to avoid social exclusion. Fromthis perspective, activation can also be seen as a socialpolicy initiative. One of the goals of the Danish welfaremodel is to enable everyone in Denmark to be an activecitizen, preferably by participating in the labour market.

Activation plays a crucial dual role in policing moralhazard, so that everyone who can work actually does so,and in upgrading the skills of the unemployed to meetlabour demand at market wages. The alternative – ofincreasing economic incentives to work by loweringbenefits, or increasing employment opportunities byreducing wages – is generally seen as undesirable.

Activation policies are associated with considerableexpense. Denmark spends more as a proportion of GDPon labour market policies – active as well as passive –than any other country in the European Union; see table2. More than 100,000 persons (measured as full-timeequivalents) participate in activation schemes everyyear. Because of its high costs in both economic andhuman terms, activation must work to be legitimate. Astable 2 shows, Danish employment rates for both menand women outrank those of other EU countries. In fact,

( ) * + , - . / 0 1 2 ) 3 4 5 1 4 6 . 7 8 7 1 7 9 8 7 9 : 8 ; < . 4 8 ) = 8 1 > . ?+ @ A B B / . 1 4 8 ; C . = 5 1 4 6 D E E BF G H I J K L M G N O P Q R S P R R RT K U G I V W G V X Y Z P [ \ S P R R R] ^ H J G _ Y ` Z P a R [ P R R Rb N Y ^ H J G _ Y ` \ R S P R R Rc I L G W J K U G I V W G V X Y Q [ R P R R Rd G I V X Y e d L K L M f L M X f g Y N ^ K V h i Z R R Q j k © 2007 National Institute of Economic and Social Research. All rights reserved. Not for commercial use or unauthorized distribution.

at Socialforsknings Institut on November 2, 2007 http://ner.sagepub.comDownloaded from


� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

as described below, surprisingly little is known about thecontribution of activation and other labour marketpolicies to the performance of the labour market.� � � � � � � � � ! � � " � # " $ � � � % � � � � & ' " � ( % '� " � % ) To grasp the role that activation plays in Denmark, onemust appreciate that it forms an integral part of a largersystem made up of social protection, labour marketpolicies and collective agreements (Kvist, 2002a). Insome respects, Denmark shares many of the traits of theother Nordic countries, although there are alsoimportant differences. With regard to their similarities,Denmark belongs to a group of countries with a Nordicwelfare model characterised by:

• Comprehensiveness. The state has a big socialresponsibility vis-à-vis the market and civil society.

• Universalism. The whole population enjoys a widerange of social rights in the form of social securityand social services.

• Individualism. Social rights are to a large extentindividualised; thus benefits are provided andcalculated in relation to the situation of the individual,without regard to family circumstances (with socialassistance and family allowances being the mainexceptions).

• Goal of high employment. Policies aim to contributeto full employment and the reduction ofunemployment, especially long-term unemployment.

• Goal of equality. Policies often aim to contribute toincreased equality between groups based on gender,age, class, family situation, ethnicity, regions etc.

• High quality and generosity. Services are of high

quality provided by persons with relevant professionalbackgrounds in the social, health and educationalsectors.

• Decentralised organisation. Most services are providedby municipalities.

In relation to the other Nordic countries, the Danishmodel stands out in four respects. First, the Danishwelfare model is to a larger extent financed throughtaxes rather than social security contributions.Contribution financing is limited to unemploymentinsurance, voluntary early retirement pay andsupplementary labour market schemes. Secondly,benefits are less generous for middle and high incomegroups than in similar Finnish, Norwegian and Swedishschemes. In fact, Danish social security and socialassistance are particularly generous for low incomegroups, often at the same or a higher level than in theother Nordic countries (Kvist, 2002b). In other words, itis most generous for the groups that are most likely toface periods of unemployment, sickness or workincapacity. Where the other Nordic countries secure acertain income compensation in social security tomiddle income groups, this has been left to negotiationbetween the social partners in Denmark. Extensivelabour market schemes, which pay out, in particular,pensions for old age, invalidity and survivors, are thethird way in which Denmark differs from the otherNordic countries.

Finally, since the middle of the 1990s, Denmark has ledthe Nordic countries in employment policy innovation(Kvist, 2003). Most recently, Denmark has become well-known for its flexicurity model, combining limitedprotection of jobs with an active labour market policyand a social protection system with easily accessible,generous benefits. Two of the elements, flexibleemployment protection legislation and generous socialprotection, have been part of the Danish welfare model

* + , - . / 0 1 2 3 . 4 5 6 7 8 9 . : 4 - + , : 8 9 ; + 9 < . 7 3 : - 6 = 6 . > + 4 5 . ; 3 - : ? ; . 4 7 9 + 7 . >@ A B C D E F G H F I J D K A L F M I C N B M A L O N L P F K H M C D E D F Q R S M T U V @ R W X X Y Z O H C M [ O F I K L N K F R W X X \] E K D ^ F O F N Q A L F Q @ N Q Q D ^ F O F N Q A L F Q _ M K N C ` F I a M O F IZ b c Y X d Y e Y c d e c f c d g Y h i f d Y Y h d YV F I O N L P c d e f f W d Y c e f d g e i h c d W i f d eU F L O N I [ X d \ c \ W d f e \ W d g \ W i W d h \ W d Wb I D K F J j D I k J M O X d c c \ X d c h i X d f X f i i d f \ Y d hl m n o p q r s n o m t u v u w x � � y z { |} m u q r ~ p u � � q � q v t n o q t � � p � n � q u o v � � � � � � � m z o m u v u � m � � � m z t � v o � � � � q � � � m � � q � u � � p q � u � � q t � � � u q � o v u � m � m � u � q � � t v z � q � � � � o q p u � m z p o q v u � m � v � � t u v o u �n � � � p q � u � � q t | � v t t � � q � q v t n o q t � � p � n � q m n u � m � � � m o � z q � q � � u t � � m u v z � � n � q � � � m � � q � u � � t n o v � p q � v � � q v o � � o q u � o q � q � u |



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �for decades, even in periods of very high unemployment.The activation element, however, dates back to the 1994labour market reform.

Lax employment protection means that Danish firms areable to adjust quickly to changing demand, making theeconomy flexible and robust to economic shocks. Tosecure the support of trade unions, and to avoidunintended distributional effects, the social protectionsystem offers benefits that, by international standards,are generous with regard to the length of benefit periodsand the levels of benefits for low-income groups.

Potentially, however, generous benefits create twoproblems. First, high benefits mean that firms must offerhigher wages to attract workers. Those unemployed withlow qualifications may find it difficult to deliver what isneeded to get the lowest wage. Secondly, high benefits maydistort incentives to work. These problems are addressed inDenmark by the following means. First, education, jobtraining and other activation policies aim to upgrade theskills of the unemployed so that their employment chancesare improved. Second, monitoring, guidance, availabilitycriteria and tests, as well as sanctions for non-compliance,all aim to avoid misuse of benefits and ensure that theunemployed are available for work.

In sum, activation is at the heart of the Danish welfaremodel. It gives the unemployed qualifications andpolices moral hazard, which in turn reduces the costs ofpotential free-riding, increases the number of tax payers,and secures broad public support for an extensivewelfare model.� � � � � � � � ! " � # $ % & ' $ ( � & $ ( #The concept of activation originates with the poor lawsof the 18th century, where receipt of poor relief wascombined with compulsory work. But the type ofactivation we know today began at the end of the 1970s.Five years of rising unemployment had shown that highunemployment had become a permanent phenomenonthat was leading still more people to rely on long-termpublic support and which could not be addressed byconventional demand-side policies. The 1978 WorkOffer scheme is the first modern activation offer, havingthe dual objective of helping the unemployed back towork through qualification offers, and giving the long-term unemployed a chance to become re-entitled tounemployment insurance through participation inactivation. At this time, activation was in other wordsnot only employment policy, but also social policy.

Nothing much happened to employment policy duringthe Conservative-led coalition governments from 1981to 1993. The main focus of these governments was toimprove the balance of payments and the major deficitsof general government. Although unemployment washigh in the 1980s it took second place to the economicimbalances that were the main priority of economicpolicy.

When the Social Democrats took office late in 1993,their declared chief objective was to curbunemployment. The underfinanced Tax-Benefit schemeand the Labour Market reforms introduced in 1994 kick-started the economy and revolutionised the relationshipbetween the welfare system and the labour market witha switch from a passive to an active approach. Thisswitch occurred almost simultaneously in several OECDcountries, but was perhaps spearheaded by the UnitedKingdom, the Netherlands and Denmark. Civil servantsfrom these countries have exchanged knowledge andtheir colleagues in other countries have also looked atthe design and experiences of activation in the threecountries as promoted by the OECD Job Strategy and theEuropean Employment Strategy (see also Nedergaard,2006).

Four features of Denmark’s labour market policies in the1990s stand out in comparison with the policies of thetwo previous decades. First, it became impossible to‘recycle benefits’. Before the labour market reform of1994, the unemployed could receive unemploymentinsurance for a maximum of 1½ years, after which theywere entitled to an offer of active labour market policyfor six months. Participating in this offer in turn re-entitled the person to another period on unemploymentinsurance, thus making a virtuous ‘benefit carousel’; seefigure 1. Now only ordinary employment countstowards qualification for benefits.

Second, the benefit period was gradually reduced.Initially, the maximum benefit period was set at sevenyears in 1994, a very long period by internationalstandards that must be seen in the context of the previousbenefit carousel. The maximum benefit period wasreduced to six years in 1996, five years in 1998 and hasbeen four years since 1999; see figure 1. After havingexhausted the four-year benefit period, the unemployedare no longer eligible for unemployment insurance andmust instead claim social assistance that is subject to ameans-test that includes the income of the partner, if any,of the claimant. Four years is also a long period byinternational standards. But because of favourable



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ! " # � $ % � � & % � � � ' ( ) * + � $ # � + � , - . ' / , � + % � + $ � � * + & � * + 0 * & % � 1 * % � ' + - � � � ' 0 $2 3 4 5 6 3 7 8 9 3 : 6 8 7 8 9 3 : 6 8 7 8 9 3 : 6 8 7 8 9 3 : 67 ; ; < = > 3 ? @ A 5 9 ? 3 > B 9 3 : 6 = > 3 ? @ A 5 9 ? 3 > B 9 3 : 6 = > 3 ? @ A 5 9 ? 3 > B 9 3 : 6 = > 3 ? @ A 5 9 ? 3 > BC > D = 6 : > E 3 5 4 4 3 6 C > D = 6 : > E 3 5 4 4 3 6 C > D = 6 : > E 3 5 4 4 3 6 C > D = 6 : > E 3F 4 B 3 6 < 9 3 : 6 D G 9 3 : 6 D7 ; ; < @ : D D C H 3 : E B C H 3F 4 B 3 6 G 9 3 : 6 D G 9 3 : 6 D7 ; ; I @ : D D C H 3 : E B C H 3F 4 B 3 6 J 9 3 : 6 D G 9 3 : 6 D7 ; ; K @ : D D C H 3 : E B C H 3F 4 B 3 6 7 9 3 : 6 G 9 3 : 6 D7 ; ; ; @ : D D C H 3 : E B C H 3" * L . � M ! N � & � + % . * L ' � � , * � O � % � � ( ' � , $P Q Q R S T U V W X Y T X Z [ \ X [ ] V X Y ^_ ` a b c d e f g h g i j c j b g k g i j j a l i g k m b a n k g i j c i d l f o i p g j e f a l q e o p j c r o j c a i st o u c k l k l i g k m b a n k g i j c i d l f o i p g ` g i g v c j m g f c a w b c k c j g w j a x n g o f d sy j f c p j g f z a f { o r o c b o ` c b c j n w g k o i w d z c j e a v v g f a v d l c j o ` b g z a f { o v j g f | } k a i j e d s~ i w c r c w l o b o p j c a i m b o i d sP Q Q � � [ X � � � [ � � [ � Z V ] S T U V W X Y T X Z [ \ X [ ] V X Y ^ � � W � � [ \ Q � �y e a f j g i c i q a v j e g k o u c k l k l i g k m b a n k g i j c i d l f o i p g ` g i g v c j m g f c a w j a � n g o f d sy j f c p j g f z a f { o r o c b o ` c b c j n p f c j g f c o s_ w � l d j k g i j a v b g o r g d p e g k g d s� c q e j o i w a ` b c q o j c a i j a v l b b h j c k g o p j c r o j c a i c i j e g z e a b g o p j c r g m g f c a w sP Q Q � S T U V W X Y T X Z [ \ X [ ] V X Y ^ ^ � � W � � [ \ Q � �� f o w l o b d e a f j g i c i q a v k o u c k l k l i g k m b a n k g i j c i d l f o i p g ` g i g v c j m g f c a w j a � n g o f d s� c q e j o i w a ` b c q o j c a i j a v l b b h j c k g o p j c r o j c a i o v j g f j z a n g o f d s� a l i q m g f d a i d l i w g f } � n g o f d a v o q g f c q e j o i w a ` b c q o j c a i j a v l b b h j c k g o p j c r o j c a i o v j g f � k a i j e d s� a f { f g p a f w j a ` g p a k g g b c q c ` b g v a f l i g k m b a n k g i j c i d l f o i p g w a l ` b g w v f a k } � j a � } z g g { d a v z a f { � ` a j e z c j e c i � n g o f d sP Q Q � S T U V W X Y T X Z [ \ X [ ] V X Y ^ ^ ^ � � W � � [ \ Q � �y e a f j g i c i q a v k o u c k l k l i g k m b a n k g i j c i d l f o i p g ` g i g v c j m g f c a w j a � n g o f d sy j f c p j g f z a f { o r o c b o ` c b c j n w g k o i w d z c j e a v v g f a v d l c j o ` b g z a f { o v j g f � k a i j e d s� o f b c g f f c q e j h o i w h a ` b c q o j c a i j a o p j c r o j c a i sP Q Q � S T � V � � � \ � � [ � V � � T � � V � � � �_ p j c r o j c a i g u j g i w g w j a d a p c o b o d d c d j o i p g p b o c k o i j d s� � S T U V W X Y T X Z [ \ X [ ] V X Y � ¡ V X [ � � � V X Z �_ ` a b c j c a i a v j e g w g k o i w v a f � � ¢ o p j c r o j c a i c i j e g o p j c r g m g f c a w st c i c k l k w g k o i w a i o p j c r o j c a i c d g r g f n � k a i j e d s~ i j f a w l p j c a i a v c i j g i d c r g p a i j f o p j d p e g k g d o j o k c i c k l k a v g r g f n � k a i j e d s� R � V � � T � T � � � � \ T � � [ X [ ] V X Y � � � [ � � � T � � [ ] V X [ � [ X � U V � � �� c q e j o i w a ` b c q o j c a i j a o p j c r o j c a i v a f d a p c o b o d d c d j o i p g p b o c k o i j d z c j e a j e g f m f a ` b g k d j e o i l i g k m b a n k g i j z e a e o r g f g p g c r g w` g i g v c j d v a f k a f g j e o i | } k a i j e d s� � £ [ � ] T X [ T � X [ [ Y [ � \ � [ � W � T \ � V � ¤ � Y Y � � X T \ � V � T � � � � \ [ � X T \ � V � ¤ � T U V W X Y T X Z [ \ T � � X [ \ � X [ Y [ � \ �� o f b c g f f c q e j o i w a ` b c q o j c a i j a o p j c r o j c a i � c s g s o v j g f x k a i j e d s~ i j g i d c r g o p j c r o j c a i o v j g f } ¥ n g o f d a i l i g k m b a n k g i j c i d l f o i p g s¦ l c p { � a ` p a l i d g b b c i q o i w o r o c b o ` c b c j n j g d j d g r g f n � k a i j e d s� b c k c i o j c a i a v d m g p c o b f l b g d v a f g b w g f b n l i g k m b a n g w c i f g q o f w j a b g i q j e a v ` g i g v c j m g f c a w o i w m o f j c p c m o j c a i c i o p j c r o j c a i § ` l j ` g i g v c jm g f c a w v a f m g f d a i d o q g w � ¨ © n g o f d c i p f g o d g w v f a k } ¥ n g o f d j a � n g o f d ª s



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �economic development and intense activation offers, notleast in the phase immediately prior to exhaustingbenefit rights, very few insured unemployed actually movefrom unemployment insurance to social assistance.

Thirdly, the concept of ‘right and obligation’ wasintroduced. The period of unemployment was dividedinto a passive and an active period. In the initial passiveperiod the person claimed benefits without compulsoryactivation, hence the label ‘passive’. Having exhaustedthis period, a person would enter the active period, wherereceipt of unemployment insurance benefits becameconditional upon participating in activation offers.

Fourth, the introduction of individual action plansmarked the end of standard offers lasting six months andgiven at the end of benefit periods. Individual actionplans highlighted the need for longer and more coherentperiods of activation.

Different channels of activation served to reach theoverall goal of higher employment. The measures wereintended to change and raise the qualifications of theunemployed so that they matched employer demands. Atthe same time, the compulsory nature of activationprovided a test of the availability for work of theunemployed and a means of motivating them to intensifytheir job search. The availability for work criteria wasalso intended to prevent misuse of unemploymentinsurance and legitimate generous benefits. Activationaimed to enhance the productivity, flexibility andmobility of the unemployed.

Throughout the 1990s employment rose and there wasever more focus on the flexibility and availability of theunemployed in order to avoid labour force shortages.Annual adjustments to policy therefore brought forwardthe activation offer in the benefit spell and increaseddemands on claimants’ geographical and occupationalmobility.

In 1998 the Law on Active Social Policy extendedactivation to claimants of social assistance and therebyalso to municipalities in charge of social assistance; seetable 3. Indeed social policy in general gradually becamemore and more employment-oriented, as illustrated inparticular by the transfer of social assistance from theMinistry of Social Affairs to the Ministry of Employment(previously the Ministry of Labour) in 2001.In 2001 a new liberal-conservative coalition governmentcame into office, which revised the laws on activationthe following year. Principles of rights, obligation and

monitoring of availability remained, but the focuschanged towards more guidance and job counsellingrather than activation per se. The purpose was to findand use the shortest and most effective way fromunemployment to a job.

In 2006 the measures were directed towards those groupsof unemployed with the biggest employment problems.One of the significant initiatives was to ‘turn the casepile’ for the group of social assistance claimants whohad not received an offer of activation within the lasttwelve months. The focus on this group is a logicalconsequence of the very low rate of unemploymentwhich leaves people with significant employmentproblems as the main group without jobs.� � � � � ! � " # � � � $ " � � � % " � � # & ' !Danish unemployment insurance follows the Ghentmodel (Carroll, 1999), where insurance is voluntary andorganised in funds affiliated to trade unions. Insuredunemployed receive unemployment insurance benefits ifthey fulfil certain criteria, most notably that of havingworked 52 weeks within the last three years, havingbeen a member of an insurance fund for one year, andbeing able, willing and capable of working. Non-insured unemployed normally receive social assistancefrom the municipality. Social assistance is paid to peoplewho have been subject to a social contingency – likeunemployment, sickness or divorce – that involves theperson being unable to support herself and her family.Because of the employment criteria associated withunemployment insurance, there are typically many moreunemployed with better employment chances in thegroup of insured unemployed than in the group of non-insured.

Activation has historically been offered by the labourmarket exchange system for insured unemployed and bymunicipalities for non-insured. The two systems havebeen merged and from 2007 incorporated in 91 JobCentres located in the municipalities.

The target group for activation has gradually widened.Young claimants of social assistance were the first to besubjected to activation in 1978. During the 1990s, thetarget group was expanded to other age groups and thegroup of uninsured unemployed. Gradually, there havealso been moves to include more marginal groups, suchas persons on sickness benefits and disability pensions.Most recently, activation has become de facto targetedat certain groups of ethnic minorities who are long-term



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �claimants of social assistance. Activation of ethnicminorities is perceived as salient both to integrate morepersons from ethnic minorities in the workforce and toincrease the labour force.

Activation efforts have, since their beginning, beentargeted towards both insured and non-insuredunemployed. There has also been a series of labourmarket initiatives for groups that are more marginallyplaced in relation to the labour market. It was theintention not only to get jobs to those already close to thelabour market, but also to those labelled ‘unemployedwith problems besides unemployment’; for example,employment policy initiatives have been promotedunder headings like ‘Need for all’ (Brug for alle), ‘Morein work’ (Flere i arbejde) and most recently ‘A newchance for everybody’ (Ny chance til alle).

Employment policy in Denmark covers broad groupswith different types of labour market problems. Targetgroups for activation include:

• insured unemployed on unemployment insurancebenefits,

• non-insured unemployed on social assistance,

• claimants of social assistance with problems besidesunemployment such as ill health, social and personalproblems,

• sickness benefit claimants,

• those who have totally or partly lost the capacity towork but who are expected to regain it at a later stage,

• those with a permanent reduction in capacity to work,

• disability pensioners.

The organisation of activation has also changed. Recentreforms mean that today there is, in principle, no differencein offers to claimants of unemployment insurance andsocial assistance. After the 2002 Labour market reform,offers have been provided under the headings of guidanceand qualifications, in-work traineeships and wagesubsidies. In practice, offers will differ between those whoare close to the labour market and those who are furtheraway. Instruments of activation include:

• private and public job training – salaried, but with apublic wage subsidy to the employer,

• education in the form of introductory courses, shorterlabour market education, and shorter further education(18 months),

• individual job training – non-salaried, e.g. in the formof specially organised projects and in-worktraineeships.

Social assistance claimants typically get different offersfrom unemployment insurance claimants; see table 4.The most frequently used activation instruments forsocial assistance claimants are specially organisedprojects and in-work traineeships, as well as offers onguidance and clarification. Private and public jobtraining coupled with wage subsidies, as well asordinary education, is rarely used by social assistanceclaimants, but quite often by the insured unemployed.

Persons with reduced work capacity may be offeredrehabilitation or jobs with permanent wage subsidies bythe state. The latter type of jobs – flex jobs, shelteredjobs and service jobs – are taken up by nearly 50,000persons. Over the past few years the number of people inflex jobs has risen markedly. In part this can beexplained by the fact that, since the 2003 DisabilityPension Reform, people can no longer apply for adisability pension (førtidspension), but first have toexhaust their potential work capacity. The intention ofthe 2003 reform was to create a labour market where asmany people as possible were in ordinary employmentand where a disability pension was awarded only to� � � � � � ! " # $ % � � � � & ' ( ) � ( * � # + # � # $ + # $ " + , - & ' ( # . / ' � ( # $ � � & -0 1 1 2 , 3 ' ) � � ( & - - ' � � 4 # $ ) � 5 � ( + & 3 + 6 7 8 9 : ; < = > ? @ A B C > D A B9 7 ; E F B > G ; < A 8 8 @ 8 D A 7 ? ;H I J K L M N O P Q R ST U V W K X Y Z [ \ U K L V ] V Y K J V I X ^ J I J K L _ M P O Q M= ` > : D a 9 @ < A 7 ? ; A 7 <? B A : @ b @ ? A D @ > 7 M P R N S O P O N c d P S N Q= F ; ? @ A B F : > e ; ? D 8 M N P M N M M N P M N M= F ; ? @ A B ; < 9 ? A D @ > 7 f O S P Q _ f S P R _ _g : < @ 7 A : G ; < 9 ? A D @ > 7 M M P d S M Q N f M S P _ _ Nh U i i I j J Z W Z k i L I l k Z X J J I J K L Q c P _ M MC : A @ 7 ; ; 8 ` @ F 8 d R N O P Q O M c P O S Mm A a ; 8 9 n 8 @ < @ ; 8 f P _ S N S P _ M f M M P d _ fo B ; p e > n 8 O S P c O _= ` ; B D ; : ; < e > n 8 c P Q R N= ; : q @ ? ; e > n 8 f M Qr < 9 B D A F F : ; 7 D @ ? ; 8 ` @ F 8 Q P N M Ms t u v w x y s z { z | } z | w } � x ~ � { v � � � � � � � �



� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �those who could not have a flex job. Although thenumber of persons in flex jobs has risen, and theeconomic climate has been favourable, there are nofewer persons on disability pensions. In 2006, there wasa total of 237,000 persons on disability pensions and inflex jobs, which was 34,000 more than estimated whenthe reform was launched.

There are just as many people in activation programmesas there are people unemployed. However, if we excludejobs for persons with reduced work capacity, there are55,000 persons or nearly 2 per cent of the labour forceparticipating in activation.� � � � � � ! " # � $ % � & ' � " � � ( ' � ) ! � � ' * ! � ' + " ' ", � " - ! & .Theoretically, activation may have a series of differenteffects. These effects may be split into two, depending onwhether effects are caused by the reactions ofunemployed facing activation, or by reactions from theemployed and persons outside the labour market. InDenmark it is common to distinguish between threetypes of effects for the unemployed in the target group assuggested by Calmfors (1994):

• Motivation effects: participation in activation as acondition of benefit entitlement makes the unemployedintensify their job search which increases theprobability of getting work.

• Lock-in-effects: participation in activation makes theunemployed search less intensely for work because ittakes time to participate in activation or because theunemployed prefer to finish the activation offer beforeapplying for a job.

• Qualification effects: the labour market qualificationsof the unemployed are improved, thus making it easierto find a job matching their qualifications./ 0 1 2 3 4 1 2 0 5 4 6 7 8 8 7 9 1 : 4 5 ; 6 0 9 < = 2 5 = 7 8 8 7 9 1 :

The motivational effects are strong for the insuredunemployed. In a seminal study, Geerdsen (2002; 2006)analysed the ‘threat’ effect of activation by comparingthe exit from unemployment for groups facing activationwith groups not facing activation.

Figure 2 illustrates the results. The zero point marks thestart of activation and the point ‘–12’ marks the 12thmonth after start of activation. The figure shows a

marked increase in the exit rate from unemploymentinsurance in the start of the activation period. Theincrease occurs after the start of the activation period,and not before, as perhaps expected, because initiallyactivation consists of writing up an action plan, findingand selecting a suitable offer of activation in a firm or inthe educational system. The ‘threat’ of activation was inother words not very serious at the start of the activationperiod, involving relatively little disutility.

Eight months inside the activation period the curve falls.At this point in time the voluntarily unemployed who donot wish to participate in activation have left the systemand the search intensity has fallen for the remainingpersons participating in activation. The shape of thecurve thus both illustrates the motivational and the lock-in effects. The lower search activity of participants inactivation is confirmed in other Danish studies. Forexample, Madsen et al. (2007) find that participation inactivation is one of the reasons for not wanting a joband not actively looking for a job. An older studyshows that lock-in effects were largest for activationtaking the form of education (Bolvig, Jensen andRosholm, 2002).

Activation has also been shown to have motivationeffects on social assistance claimants. The effects aresmall overall (Graversen, 2004), but stronger for personsunder 30 years of age, the target group for so-called‘immediate activation’ (Beskæftigelsesministeriet, 2006).

> ? @ A B C D E F G C H B I J K J ? L ? M N I O L C K P ? Q @ M G C A Q C R H L I N R C Q MJ C Q C O ? M S N S M C R

T � U V W X Y Z X X V [ \ X ] ^ � � � � _ �-60

-40

-20

0

20

40

60

80

100

120

12 10 8 6 4 2 0 -2 -4 -6 -8 -10 -12

Remaining passive benefit months

Perc

enta

ge p

oin

ts



JON KVIST AND LISBETH PEDERSEN DANISH LABOUR MARKET ACTIVATION POLICIES 107� � � � � � � � � � � � � � � � �Qualification effects are hard to measure. Whether theactivated persons become employed due to newqualifications or increased job search is difficult to assess.

Changes in skill and search costs can hardly bemeasured directly. Instead, changes in the transitionrates from unemployment into employment may bestudied. For at least two reasons it is not easy to get acorrect estimate of the employment effect of activation.One reason is that it is very difficult to measure theisolated effect of activation due to the lack of a goodcontrol group. A part of the group that becomesemployed after activation would have found a jobanyway. Another reason is that we do not have verygood data on the exact extent of individual employment.To remedy these data problems, the effect of activationin recent Danish studies is estimated by the change, ifany, in numbers receiving social security (i.e.unemployment benefits, social assistance benefits,sickness benefits, rehabilitation benefits, and parentalleave benefits). In a Danish context this is a fairly correctway to measure the extent of employment, since onlyvery few adults rely on financial support from theirfamilies. The degree to which a person is living on socialsecurity benefits within a year is measured on a scalefrom 0 to 100, where 0 indicates full employment and100 indicates full public support.

The employment effect of activation has been analysedfor different types of activation measures and separatelyfor non-insured and insured unemployed in activation(Graversen and Weise, 2001; Bach, 2002).

Figures 3a and 3b illustrate the estimated employmenteffect of different activation measures for, respectively,insured unemployed and social assistance claimants aswell as the shares of unemployed in different measures.Some measures have the same effects for the two groupsof unemployed. Private job training has incomparablythe largest direct employment effect. Individuals over 25years of age that participate in this type of activationwill, on average, reduce their dependence on socialsecurity by 20 percentage points, which is equivalent toabout two months a year. However, only one in ten inactivation participates in private job training and theeffects of other measures are significantly smaller.

For social assistance claimants the most frequentlyapplied measure is the use of employment projects.Participation in this type of programme reduces

dependence on social security by only 3 percentagepoints, or about one-fifth of the effect of private jobtraining. Individual job training in public workplaces isanother measure frequently used by social assistanceclaimants. This type of measure reduces theparticipants’ dependence on social security by 6percentage points, equivalent to three weeks per year.

Figure 3b. Effects of activation on social assistance

claimants

-30

-20

-10

0

10

20

30

40

Job t

rain

ing

- pri

vate

Job t

rain

ing

- public

Indiv

idual

job t

rain

ing

- public

Em

plo

yment

pro

ject

Per

cent

Proportion of persons in the scheme

Per cent change in number of benefit claimants

Figure 3a. Effects of activation on insured unemployed

-30

-20

-10

0

10

20

30

40

50

Job

trai

nin

g -

pri

vate

Job

trai

nin

g -

public

Indiv

idual

job t

rain

ing

- public

Ord

inar

y

educa

tion

Speci

al

pla

nned

educa

tion

Per

cent

Proportion of persons in the scheme

Per cent change in number of benefit claimants



108 NATIONAL INSTITUTE ECONOMIC REVIEW No. 202 OCTOBER 2007

In a study of rehabilitation, Filges et al. (2002) foundthat the effects of this measure are very small and maybe even smaller than the effect of activation. This isquite surprising, since rehabilitation is offered to personswho are expected to be able to regain working ability.On the other hand, a significant part of the group wouldprobably have left the labour market if they had notbeen offered rehabilitation.

In Denmark it has generally been believed thateducation secures employment. Therefore, it has causedsurprise and concern that education in the form ofactivation shows small effects for the unemployed. Onereason may be that it is difficult to measure the effect ofeducation, which may only materialise over the longterm, with very different effects for different targetgroups. Another reason may be that education offers arenot properly designed to match the individual and thedemand of the labour market.

Not only does the content of activation haveimplications for its effects, but so also does the timing.To give the greatest effect, activation should be usedearly in the benefit spell for men, but later for women(Bolvig, Jensen and Rosholm, 2002). Similarly, earlyactivation has positive qualification effects in somecases (Graversen, Damgaard and Rosdahl, 2007).Finally, the sequencing of activation offers is important.Where more than one activation offer is made,education should preferably be offered before jobtraining (Graversen 2004).� � � � � � � � � � � � � �Although division into three different types ofemployment effects is widely used in European andAmerican evaluations, it is common in Denmark also toinclude so-called welfare effects. Because manyactivation programmes in Denmark have a broaderobjective than merely to improve employment for thetarget group, there is also focus on non-employmenteffects. In particular, for weaker groups in the labourmarket the goal of activation is also to improve thegeneral quality of life for the persons concerned. In fact,activation could have positive effects for such groupswithout it being measured by employment effects. Toinclude effects beyond that of employment alone it iscommon in Denmark to include a fourth set of effects:

• Welfare effects: the effects of activation onparticipants’ well-being.

Participants may have both positive and negative utilityof activation. The importance of activation forparticipant’s life quality can be related both to expectedemployment effects or to the offer itself. No Danishstudies have had the explicit purpose of measuringwelfare effects. However, there are a series of studies onparticipants’ attitudes. Most participants believe thatactivation leads to a better daily life (70 per cent) andimproved self-esteem (58 per cent) (Bach, 2002). Abouthalf believe that activation improves their labourmarket qualifications. Generally participants have apositive view on activation, but one in four have anegative view, either because they do not thinkactivation provides new qualifications or because theythink that there is no purpose to activation.

To balance the goal of improved quality of life againstthe goal of employment is hard. At the moment theemployment goal ranks very high on the politicalagenda. But a total assessment of activation is likely toinclude both employment effects and welfare effects. � � � � � � � � � � � � � � � � �The effects of activation on the individual depend notonly on the type, timing and sequencing of the activationoffers but also on the causes of unemployment and thepreferences of the unemployed for work, activation andleisure. To interpret the different effects of activation ondifferent groups of unemployed, Pedersen andSøndergaard (2004) have suggested a fourfold typologyof the unemployed; see table 5.

Estimating the number of persons in each type ofunemployment is difficult because of problems in revealingthe preferences of individuals towards work, activation andleisure, which make it difficult to provide precise estimates.However, discretionary accounts can be based on theunemployment statistics and two recent studies on themotivation and availability for work of the unemployed.

The potential target group for activation is about 220,000persons. By 2006 there were 135,000 registeredunemployed and activated. About 80 per cent of this groupwere insured unemployed relying on unemploymentinsurance, whereas the remaining 20 per cent were notinsured, but claiming social assistance. These are personswho are expected to be able to work. Another group of85,000 social assistance claimants are not ready for thelabour market, including 25,000 in activation. Thesepersons have been found by the authorities not to be ableto work immediately.



JON KVIST AND LISBETH PEDERSEN DANISH LABOUR MARKET ACTIVATION POLICIES 109

Among the registered unemployed and non-insured socialassistance claimants, 80 per cent express a desire to workand 20 per cent do not (Madsen, Filges, Hohnes, Jensenand Petersen, 2007). Among the social assistance claimantsthat are not ready for work, 60 per cent say that they wantto work (Bach and Petersen, 2007). In crude terms, thismeans that the potential target group for activation of220,000 consists of about 49 per cent who are involuntarilyunemployed, 12 per cent who are voluntarily unemployed,23 per cent who lack the necessary qualifications, and 15per cent who are facing the double constraint of not beingable or willing to work.

Of course such figures must be treated with a great deal ofcaution. In the group of involuntarily unemployed therewill be those wanting a job only under certain conditions.On the other hand, in the group of voluntarily unemployed,there will be persons waiting to get into the labour marketand thus not against being in employment.

The number of persons in the group that qualifies foractivation is greater than the number of participants inthese programmes. Whereas persons above 30 years ofage are obliged to participate in activation programmesafter one year of unemployment, those below 30 areobliged to participate somewhat earlier. What is more,not everybody who is entitled and obliged to participatein activation actually receives an offer.� � � � � � � � � � � � � � � � � � � � � � �Activation can also affect the employed and personsoutside the labour market:• Employment participation may change. Persons

attracted to an offer of activation may register asunemployed and job seekers. Persons deterred by theperspective of activation may stop participating inthe labour market.

• Substitution effects. The unemployed person whoobtains employment may substitute for another whois therefore pushed into unemployment.

• Wage effects. Activation may affect the level of wagesbecause labour supply goes up and thus also competitionfor jobs. In addition, by providing some groups withbetter qualifications and employment opportunities,activation may affect the wage distribution.

• Tax effects. Activation may have a positive or negativeimpact upon the level of taxes, depending on whetheractivation in its entirety has a good or bad impactupon society as a whole.

Generally, indirect effects are hard to measure. Thereare no Danish empirical analyses of activation andaggregate employment. The substitution effect for firmswith a stable production level has been estimated at 40per cent for persons hired on wage subsidies (Hussainand Rasmussen, 2007).

Activation may have positive or negative effects on taxlevels. If the net employment effect of activation issufficiently high, this may cause a drop in the tax level,because productivity has increased while, at the sametime, the fall in expenses for social security benefits hasexceeded expenditure on activation. But activation neednot always give a positive cost-benefit result. Activationin the private sector has been found to have positivebudget effects, whereas most other activationinstruments are close to cost neutral (Christensen, 2002).Most recently, the Economic Council of Wise Men (DetØkonomiske Råd, DØR) suggested that the net annualeconomic costs of activation ran at 3 billion DKK (£275million) (DØR, 2007). This masks significant differencesacross activation instruments. Education was expensive,with a net cost of 5 billion DKK, whereas job training

Table 5. Typology of unemployed

Wants to work

No Yes

Can work No Double constrained: Do not want to work Qualifications do not match labour demand (23%).and cannot work with current type and levelof qualifications (15%).

Yes Voluntarily unemployed in the sense that they Mostly temporarily involuntary unemployed activelydo not want to work at current levels of wages looking for work (49%).and benefits, but who do have qualificationsmatching labour demand (12%).

Source: Pedersen and Søndergaard (2004), Bach and Petersen (2007), and Madsen, Filges, Hohnen, Jensen and Petersen (2007).




was the only instrument with a positive return due topositive employment effects. The positive effects ofeducation were in general greater for those withoutqualifications than for those with a formal educationand with job training having positive employmenteffects. The motivational effects of activation were high,leading to a positive return of 2 billion DKK.� � � � � � � � � � � � � � � � � � � � � � � � � � � Job rotation policies and economic incentives have beenperceived as the two main alternatives to activation inDenmark. Job rotation was widely used in the early1990s, but today is only available for older workers. Incontrast, make-work-pay strategies were not used inthe early 1990s, but have gradually become moredominant.

Job rotation has not proved to be a success in increasingemployment. The 1979 voluntary early retirementscheme (efterløn) aimed at allowing workers aged 60 orolder to retire so that the (younger) unemployed couldtake their place in the labour market. It did not work.Many people have, over the years, used voluntary earlyretirement to make an early exit from the labour market.The rotation principle in the leave schemes foreducational, parental or sabbatical purposes codified inthe 1994 Labour market reform has not worked either.Its immediate popularity helped reduce the official rateof unemployment nearly overnight, but contributed tobottlenecks in the improving labour market and wasgradually phased out in the 1990s (Pedersen, 1996).Finally, the so-called transitional allowance(overgangsydelse) for the long-term unemployed, aged55 to 59 years of age, was abolished in 1996. The idea ofredistributing employment and unemployment throughjob rotation is today dead.

But the voluntary early retirement scheme is still inplace. With around 180,000 claimants since 1998, thereare more people on this scheme than on eitherunemployment insurance (varying from 117,000 to145,000) or social assistance (87,000 to 102,000persons). Several attempts have been made to reduce itsuse. The demand for membership of an unemploymentinsurance fund was increased in 1980, i.e. the year afterthe scheme was introduced, from five years within tenyears to ten years within fifteen years. Membershiprequirements were further increased to 20 years within25 years in 1992, when the so-called 63-year rule madeit more economically attractive to postpone retirementto the age of 63 rather than at 60 years of age. The 1999

reform further increased the membership requirement to25 years within the last 30 years, introduced separatecontributions for the scheme (previously part of generalunemployment insurance) and economic incentives topostpone retirement. Membership requirements wereincreased to 30 years in 2006 as part of the WelfareAgreement that will also gradually raise the age ofeligibility from 60 to 62 years in the period from 2019 to2022, and introduce regulation of the age of eligibilitywith demographic factors from 2025.

Improving economic incentives by cutting benefits hasgenerally not been used. But there are exceptions to thisrule, just as the tendency to use economic incentives inemployment has increased under the liberal-conservative government that came into office in 2001.Under the previous government, a broad coalition ofpolitical parties in Parliament agreed on the ‘youthpackage’ in 1996 which, inter alia, meant that personsunder 25 years of age without education had theirbenefit halved after six months and were obliged toparticipate in education or job training if they were notto lose their benefits. Secondly, a ceiling on the totalamount of social assistance for couples on socialassistance was introduced in 2005. Thirdly, peopleentering the country would normally be entitled to socialassistance if they could not find a job, but as of 2002 theso-called Start Help replaced social assistance for peoplein the first seven years of their stay in Denmark, de factohalving the previous level of social assistance. Fourthly,from 2008, couples who both claim social assistanceneed to have worked 300 hours or more over the last twoyears, non-compliance resulting in the withdrawal ofsocial assistance for one person. Finally, the earnedincome tax credit of 2.5 per cent of gross incomeintroduced in 2004 with a ceiling of 7,500 DKK (2007)will next year be raised to 4.25 per cent with a ceiling of13,100 DKK. � � � � � � � � � � � � � � Danish activation policies have undergone massivechanges over the past fifteen years. Most notably, thetarget groups have been widened from young socialassistance claimants to other age groups and to otherbenefit schemes and policy areas. With changing targetgroups, the role of activation has also changed fromcombating unemployment to increasing employmentand, most recently, to integrating those with disabilitiesand those from ethnic minority groups with lowemployment rates.



JON KVIST AND LISBETH PEDERSEN DANISH LABOUR MARKET ACTIVATION POLICIES 111

Danish experience confirms that activation has manytypes of effects: for unemployed individuals, whopotentially or actually face an activation offer orrequirement, and indirectly on the labour market itself.

Studies of Danish activation policies find significantmotivation effects. Both the voluntarily andinvoluntarily unemployed who are already looking forjobs receive an extra incentive to increase their searchintensity. The group of voluntarily and involuntarilyunemployed with the necessary work qualificationsmakes up a significant share of all unemployed.Motivation effects are therefore one of the mostimportant effects of activation.

Qualification effects of educational activation areparticularly important for those whose employmentprospects are limited by the lack of appropriate skills.Educational activation, however, has negative, if any,effects for persons who already have medium or highqualifications. Overall, the qualification effects ofactivation so far appear to be modest and only positivelysignificant for job training. Recent studies show thattargeting, content, procedure and timing are essentialfor securing an efficient upgrading of qualifications forthe large group of unemployed lacking qualifications.Although recent studies show that early interventions aregood, considerable lock-in effects may be a barrier forfurther expansion.

Rapidly falling unemployment has meant that the weakestsocial assistance claimants now constitute a large share ofthe group without labour market qualifications. Typicallythey are long-term claimants of social assistance withsocial and personal problems, a condition making itdifficult for them to find appropriate activation offers; thosewho participate in activation schemes often go from onescheme to the next. Although many social assistanceclaimants say that activation has had a positive effect ontheir quality of life, a considerable part of this group doesnot get any closer to the labour market. This result poseschallenges for finding ways to bring this group graduallycloser to the labour market, just as it becomes important forresearchers to go beyond immediate employment effects tomeasure the effects that reflect a progressive developmenttowards employment.

Research challenges also consist of measuring theindirect labour market effects of activation and itscombined effects with the lax employment protectionand encompassing social protection. The Danishemployment miracle is often accredited to its flexicurity

model, especially the role played by activation since1993. But there is little research evidence to back up thisclaim. So far research points mainly to the directmotivation effects whereby activation has brought theunemployed into work. Activation may also have had apositive indirect effects, by reducing the numbers ofthose becoming unemployed and dampening wageincreases due to the threat of activation. But suchindirect effects have not yet been proved.

Other countries may well learn from Danish experience:job sharing has proved to be a dead end. Insiders haveused stepping stones into the labour market to get out.Educational activation is not the panacea for allpopulation groups. And labour market integration of thevery weak non-employed demands a special focus absentin general activation offers. However, policymakerscannot pick and choose particular activation policies àla carte and expect to get positive ‘Danish’ resultswithout paying attention to their own national labourmarket structure and organisation, the economic cycle,and other policies, especially – in the jargon offlexicurity – those of employment protection and socialprotection.� � � � � � � � � �Bach, H.B. (2002), � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � , Copenhagen, Danish National Institute of Social

Research.Bach, H.B. and Petersen, K.N. (2007), � � � � � � � � � � � � � � � ��

, Copenhagen, Danish National Institute for SocialResearch.

Bolvig, I., Jensen, P. and Rosholm, M. (2002), ‘The employmenteffect of active social policy in Denmark’, IZA DiscussionPaper 737, Bonn: IZA.

Beskæftigelsesministeriet (2006), � � � � � � � � � � � �� , Copenhagen, Ministry of Employment.Calmfors, L. (1994), ‘Active labour market policy and

unemployment – a framework for the analysis of crucial designfeatures’, ! " # ! $ � � � � � $ % � � � � � , No. 22.

Carroll, E. (1999), ! � � � � � � $ � � � % � � $ � � � � � � � % � $ � � & � � � � � $ �& � � � � � � � � , Stockholm, Swedish Institute for Social Research.Christensen, T.Q. (2002), " � � ' � � � � $ ' � � ( � � � � � � � � � � � �� $ � � � � � � � � , Copenhagen, Danish National Institute of Social

Research.Det Økonomiske Råd (2007), # � � � ) � � � � � � � � � � � � � � *

,Copenhagen, Economic Council of Wise Men (DØR).

European Commission (2007), + � , � � � $ � � � � � � � � � $ � � � � � � �� - � $ � � � ( . � � � � � � � � � � � � � � � � � � � � � � � - � � � � � ( � � � � $ � � � y, COM

(2007) 359 final.Eurostat (2007a), Structural indicators. Retrieved from

www.ec.europa.eu/eurostat 14 September 2007.—(2007b), Public expenditures on labour market policies and

employment rates retrieved from www.ec.europa.eu/eurostat




14 September 2007.Filges, T., Harsløf, I. and Nord-Larsen, M. (2002), � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � , Copenhagen, Danish National

Institute of Social Research.Geerdsen, L.P. (2002), ‘Does labour market training motivate job

search?’, Working paper 23:2003, Copenhagen, DanishNational Institute of Social Research.

—(2006), ‘Is there a threat effect of labour market programmes?A study of AMLP in the Danish UI system’, � � � � � � � � � � � � � � � ,116, 513, pp. 738–50.

Graversen, B.K. (2004), ‘Employment effects of active labourmarket programmes: do the programmes help welfare benefitrecipients to find jobs?’, PhD thesis, Aarhus, Aarhus University.

Graversen, B.K., Damgaard, B. and Rosdahl, A. (2007), � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� , Copenhagen, DanishNational Centre for Social Research.

Graversen, B.K. and Weise, H. (2001), ‘Effekter afaktiveringsindsatsen over for kontanthjælpsmodtagere’,Arbejdspapir 2:2001, Copenhagen, Danish National Instituteof Social Research.

Hussain, M.A. and Rasmussen, M. (2007), ‘Do wage subsidiesreduce ordinary employment? A firm level panel data analysis’,Working Paper, Copenhagen, Danish National Centre forSocial Research.

Kvist, J. (2002a), ‘Activating welfare states. How social policiescan promote employment’, in Clasen, J. (ed.), � � � � � � � � � � � �� , Bristol, Policy Press, pp. 197–210.

—(2002b), ‘Changing rights and obligations in unemployment

insurance’, in Sigg, R. and Behrendt, C. (eds), � � � � � � � � � � � � � � � �� ! � � � � � " � � � � � , International Social Security Series No. 8.

New Brunswick, Transaction Publishers, pp. 227–45.—(2003), ‘Scandinavian activation strategies in the 1990s: recasting

social citizenship and the Scandinavian Welfare Model’, � � � � �� # � � � � � � � � � � � � � , 4, pp. 223–-48.Madsen, M., Filges, T., Hohnen, P., Jensen, S. and Petersen, K.N.

(2007), " � � � � � � � � � � � � � � � � � � $ � � Copenhagen, DanishNational Centre for Social Research.

Madsen, P.K. and Pedersen, L. (eds) (2003), % � � � � � � � � � � � � � � � $ � � � � � � � � � & � � � � � � � � � , Copenhagen, Danish National

Institute of Social Research.Nedergaard, P. (2006), ‘Which countries learn from which? A

comparative analysis of the direction of mutual learningprocesses within the open method of coordinationcommittees of the European Union and among the Nordiccountries’,

' � � & � � � � � � � � � � ' � � � � � � , � � � � � � � � � � � ( � � � � �) � � � � � � � � � � � � � � � � � � � # � � � � � � � � � � , 41, 4, pp. 422–42.Pedersen, L. (1996), * � � � � � � � � � � � � � � � � � � � � � � � � , Copenhagen,

Danish National Institute of Social Research.Pedersen, L. and Søndergaard, J. (2004), ‘The role of workfare in

the Scandinavian Model of Social Security: soft incentives, skillupgrading or quality of l i fe improvement for thedisadvantaged?’, in Saunders, P. (ed.), � � � � � � � � � � � � � � � � � � � � � � ,Aldershot, Ashgate.

Statistics Denmark (2007), Statistikbanken, retrieved data fromwww.dst.dk 14 September 2007.

World Economic Forum (2006), ! � � � � � ' � � & � � � � � � � � � � � � � & � � � + , , - + , , . , Lausanne.



Documents

National Institute Economic Review · 2012-09-20 · between Danish activation policies and flexicurity on the one hand, and economic and social performance on the other, are lacking