Two Decades of Structural Reform in Denmark: A Review2f348b20-11b6-45ed-b7da-600d3aa8c31f/... ·...

Two Decades of Structural Reformin Denmark: A Review1

1 IntroductionIn an international perspective, Den-mark (and Scandinavia�s) economicperformance is notable for combin-ing a generous welfare state and highincome equality with a fairly stronglabour market performance and,more broadly, a well-functioningeconomy. At the same time, the to-tal tax pressure is among the highestin the world, close to 50% of GDP,and the number of transfer recipi-ents is high. In 2003, more than20% of Danes of working age (de-fined as 15—64 years) received somesort of transfer income, a figureboosted by more people being inearly retirement or on disabilitybenefits than there are unemployed.

The employment rate is thehighest among the EU-15, theU.S.A. and Japan (see chart 1a).

The unemployment rate stood atsome 5� percent in 2003, whichwas a good deal below that of mostEU-15 countries and lower than theU.S. (see chart 1b). In purchasingpower adjusted terms, GDP percapita is among the highest in theEU-15, (chart 1c). 2 Productivitygrowth per employee since 1995 ex-ceeds the EU-15 average, but hasfallen behind the strong performanceof the U.S.A. and some of Den-mark�s Scandinavian neighbours,namely Sweden and Finland whichboth have a strong(er) presence inIT-sectors (chart 1d).

Meanwhile, income inequality isamong the lowest in the OECD, andthe public expenditure ratio amongthe highest. A strong focus is main-tained on high public service stand-ards and avoiding poverty.

1 This paper is in large part inspired by Callesen (1997). The views expressed are those of the authors alone andnot necessarily those of the Danish Ministry of Finance.

��

��

��

�

�

�

�� !"�� ##��$

��

2 The gap with the U.S.A. largely reflects fewer annual hours worked per employee, and so does not imply a similargap in welfare. Fewer hours worked may reflect different preferences over leisure versus consumption as well asweaker work incentives at the margin from high marginal taxes.

Mads Kieler

Chief Advisor, Ministry of Finance, Denmark

Sflren Gaard

Ministry of Finance, Denmark

61�

vowi_tagung_2004 Seite 61 15.12.2004,08:32 schwarz rot

Public and private savings arehigh in an international comparison,providing both for a high level ofprivate gross fixed investment and asizeable current account surplus(i.e. total financial savings), (seetable 1).

The generally favourable employ-ment and savings performance shouldbe seen against the background of along period of structural reformsthat helped turn around the econ-omy from a dismal position in the1970s. The reforms have focused on

��

��

�

�

�

�

�

�� !"�� ##��$

� ��

��

��

�" �%%�

!

��

�

��

�

�

�� !"�� ##��$

� ��

�� !!"#� �

��"

&' �� ()��

!

�

�

�� !"�� ##��$

� ��

Mads Kieler

Sflren Gaard

62 �


strengthening private and public sav-ings and improving the functioningof labour and product markets.3 Inthis vein, the reforms have in someareas scaled back on easy access andrights to universal welfare services,while maintaining the basic functionsof the welfare state in terms of alle-viating exposure to economic risks,providing old-age security, alleviatingpoverty, and providing equal educa-tion opportunities.

In part, the key to Denmark�simproved performance is that policy-makers managed to move beyondthe traditional efficiency versusequity debate, with its strong ideo-logical overtones, to find ways ofcombining efficiency and equity. Inparticular, generous welfare supportsystems are maintained for thosethat qualify, while alleviating disin-centives e.g. by restricting access ortesting availability for work. The taxsystem has been reformed to reducemarginal tax rates despite a contin-ued (slight) increase in the total tax-to-GDP ratio since the mid-1980s —although marginal tax rates remainhigh.

A fairly wide political consensushas also formed behind the objectiveof running sustainable fiscal (andstructural) policies in view of the fu-ture financial pressures related tothe ageing of the population. Thegovernment�s so called 2010-strategyis focused on running public sur-pluses to reduce public debt, and in-creasing employment through struc-tural measures, to ensure the sus-tainability of public finances. Theadjustment burden still outstandingin terms of fiscal policy measures orstructural reform in order to pre-vent public debt from rising beyondbounds is, on available estimates,moderate compared to most EUcountries and the U.S.A., (see Fred-eriksen, 2003 and 2004 and Eco-nomic Policy Committee, 2003).

Significant policy challenges re-main to prepare for ageing, however,in particular with respect to tightexpenditure control, and continuedstructural reforms to raise sustain-able employment. A Welfare Com-mission has been established to ex-plore ways of reforming the Danishwelfare model with a view, notably,to boost labour supply and employ-ment.

3 In this paper, �structural reform� refers not only to measures that raise potential output and reduce rigidities, e.g.in labour markets, but also to changes in tax and pension systems that impact savings and investment incentives.

Table 1

Total, Real and Financial Savings , 2002

DK Euro area US SE UK

% of GDP

Total savings 22.9 21.3 14.0 21.2 14.5— of which private 19.6 21.0 14.0 17.2 14.6— of which public 3.3 0.3 0.0 4.1 —0.1

Total investment 20.8 20.2 18.6 17.1 16.3— of which private 19.1 17.6 15.2 14.1 14.9— of which public 1.8 2.6 3.4 3.0 1.4

Total financial savings 2.1 1.1 —4.6 4.1 —1.8— of which private 0.5 3.3 —1.2 3.1 —0.3— of which public 1.6 —2.3 —3.4 1.1 —1.5

Source: OECD, Economic Outlook 74 (2003) and Statistics Denmark.

Mads Kieler

Sflren Gaard

63�


2 Historical Policy Chal-lenges and the Timing ofMajor Policy Initiatives

Denmark entered the 1980s in anunfavourable macroeconomic posi-tion. As in the rest of the EU, un-employment and inflation had risenfrom very low levels in 1972 todouble-digit ranges in 1982, withDenmark experiencing a slightlyworse deterioration than the aver-age. Long-term interest rates hadsoared to 20% under the impact ofhigh foreign rates and a lack of cred-ibility of the exchange rate paritywithin the ERM in the aftermath ofrepeated devaluations of the kronein 1979—82.

Unlike other EU countries Den-mark had a seemingly chronic cur-rent account imbalance — with anuninterrupted 30-year run of defi-cits. The imbalance was rooted indisincentives to save in the privatesector and, from the mid-1970s,mounting public deficits. Under theimpact of rising unemployment andhigh interest rates, the governmentdeficit reached 9% of GDP in 1982.

Against this unfavourable back-ground, Denmark has had four ma-jor (stabilization) policy adjustmentssince the early 1980s, namely:— The introduction of the fixed

exchange rate policy in 1982backed by strong fiscal tightening

��$��%��

��

��*��

�

�

�

�

�� #�#�� "�� ##��"��$

�#� �#�� #� �#�� ## �##� �

�$%&��'

��

�(�)��$�

&�'��(��)��*+,��&��-��-��$

��

��+�%

�

�

��

�

�

�� #�#�� "�� ##��"��$

�#� �#�� #� �#�� ## �##� �

�)&*%�+�+)%,-�$�&�*%�$�$.��$ �%/+��*)%

Mads Kieler

Sflren Gaard

64 �


and the abolition of wage index-ation.

— The 1986 �potato diet� and 1987tax reform to redress overheatingpressures and external imbalan-ces by strengthening private sav-ings.

— The �kick start� of 1994 to pro-mote recovery and reduce unem-ployment.

— The �Whitsun package� of 1998to prevent overheating andstrengthen savings.

The timing of these packages wasrelated to major imbalances in theeconomy in the shape of significant

output gaps (chart 3a) and, on thefirst two occasions, substantial sav-ings-investment imbalances (charts 2band 2c).

The turnaround in macroeco-nomic performance was initiatedwith the introduction of the fixedexchange rate policy backed bystrong fiscal tightening from late1982 onwards. This episode hasbeen labelled an example of �expan-sionary fiscal tightening� (see Alesinaand Perotti, 1995, and others). Thefiscal tightening was more than off-set by an exceptional decline inlong-term rates from 20% to 10%

��%��.��$��%��

��

��+�%

�

�

0�

0�

0

�� #�#�� "�� ##��"��$

�#�� #� �#�� ## �##� � ��

�$%&��' �(�)��$�

��.��$��%��

��"

��+�%

�

�

0�

0�

0

0�

0�

�� #�#�� "�� ##��"��$

�#� �#�� #� �#�� ## �##�

�$%&��' �(�)��$�

��

Mads Kieler

Sflren Gaard

65�


within three years and a rebound inconfidence. Consequently, the Dan-ish economy boomed in 1983—86 ata substantially faster pace than therest of the EU, and the businesscycle deviated significantly from theEU average in the mid- to late-1980s (see chart 3b).

As private consumption and in-vestment soared, the current ac-count deficit widened to a record5% of GDP in 1986, and wage infla-tion accelerated. In response, the�potato diet� of 1986 consisted

largely of administrative credit mar-ket tightening, particularly in themortgage credit market (affectingtop-up loans, repayment profiles onnew loans, etc.). The tax reform of1987 lowered the tax value of inter-est deductions to strengthen savingsincentives (see below). The meas-ures had a strong contractionary ef-fect on private consumption and thehousing market. To help restorecompetitiveness, employers� socialsecurity contributions were sharplyreduced in 1988, fully financed by a

/��,��*��'�0�1��2�3��&��-

��!�

�

�

�

0�

0�

0!

0�

0�

�� #�#�� "�� ##��"��)��#,��#$

�#� �#�� #�� #� �#�� ##

�$%&��'

��##� �##� �## �##� �

�*&*%,�)/�&�1)��.��*+*2��*)%�*%*��*3$.

��$��

��!�

�

�

�

0�

0�

�� #�#�� "�� ##��"��)��#,��#$

�#� �#�� #� �#�� ## �##� �

�$%&��' �(�)��$�

Mads Kieler

Sflren Gaard

66 �


2.5 percentage point increase in (ef-fective) value-added tax (VAT) rates.

As the ensuing slump in domes-tic demand was prolonged by highinterest rates during the German re-unification boom, growth was mea-gre and unemployment climbed until1993. The former credit measureswere gradually lifted in 1993 and1994. Alongside a wave of mortgagerefinancing when interest rates fell,

this played an important role in�kick-starting� a strong recovery in1994. Helped also by fiscal stimulus,growth reached 6% in 1994. Thesubsequent period of falling unem-ployment was also the most activein terms of labour market reforms(see below).

When policy tightening becamenecessary in 1998 to prevent over-heating, the �Whitsun package�

,��+4��5�-��''��/��$

��

�

!

�

�

0�

0�

0!

0�

�� &�)��#,��#��#��) �� &�&-$.� � /) ��#�� #�� 0� � �� #�� #�� #��)�� #�� ' �� )��#��

'��*� #$

�#� �#�� #�� #� �#�� ## ��##� �##� �## �##� � ��

4(�5(��,�5�*. �+�$//$ � �6)�5�$3*)(.�7$��.

2��5��-+4��&��'�5�-��/��

��$

��

�

�

�

0�

0�

0!

0�

�� &�)��#,��#��#��) �� &�&-$.� � /) ��#�� #�� 0� � �� #�� #�� #��)�� #�� ' �� )��#��

'��*� #$

�#� �#�� #�� #� �#�� ## ��##� �##� �## �##� � ��

�*. �+�$//$ � 4(�5(��,�5

Mads Kieler

Sflren Gaard

67�


mainly took the form of further re-ductions in the tax value of interestdeductions and increased mandatorypension savings — rather than admin-istrative credit market measures,which entail distortions.

Further credit market liberaliza-tion — with the introduction of vari-able rate mortgages during the late1990s and mortgage loans free ofrepayment for up to 10 years (whichmay be rolled over) as of 2003 —are providing substantial support tohouse prices and private demandduring the emerging upswing in2004.

With changes to the tax value ofinterest rate deductions and creditmarket policies heavily influencingdomestic demand, demand manage-ment has not been restricted to (tra-ditional) fiscal policy, even thoughmonetary policy is bound by thefixed exchange rate. Fiscal policyhas, in contrast to many other EUcountries, played a notable counter-cyclical role (see chart 4a and 4b).

3 Policies to StrengthenNational Savings

Private savings incentivesThe chronic savings deficits up to themid-1980s appeared to be rooted inthree structural factors: the tax

treatment of interest payments, thetax-financed pay-as-you-go pensionsystem, and the tax treatment ofinvestments. Policies to deal withthese aspects were enacted from themid-1980s.

Until 1987, the tax value of in-terest deductions depended on themarginal income tax rate, which atthe time could be up to 73%.4 Thetax value was subsequently reducedthrough a series of tax reformsagreed in 1987, 1994 and 1998.The tax reform of 1998 reduced thetax value of interest expenditures toits current level of 33%, phased induring 1999—2001. These changeshave raised the real after-tax ratefaced by households despite a con-comitant decline in pre-tax real in-terest rates (see chart 5a).

Denmark has a tax-financed pay-as-you-go pension system with uni-versal coverage. However, the pen-sion support levels relative to aver-age income are fairly modest incomparison with many EU-coun-tries. The pay-as-you-go system iscomplemented by tax-favoured indi-vidual retirement savings accounts,collectively agreed labour marketpensions, and mandatory (legislated)labour market pensions. These arebased on actuarial principles with a

Table 2

Important Tax Rates Affecting Savings, 1980—2004

1980 1986 1993 1998 2002 2004

%

Maximum tax value of interest deductions 68.8 73.2 52.2 46.4 33.3 33.3Tax on return of private pensions 0.0 44.1 50.1 35.8 15.0 15.0Corporate tax rate 40.0 50.0 34.0 34.0 30.0 30.0

Source: Author�s calculations.

4 The very high income tax rates in Denmark should be seen in conjunction with the near-absence of payroll taxes(although part of the income tax since 1998 consists of an 8% so-called labour market contribution, it appliesto both wages and other personal income and is collected with the normal income tax). The total tax wedgebetween wage costs to the employer and the net wage of employees is not correspondingly higher in Denmark (seesection 7).

Mads Kieler

Sflren Gaard

68 �


clear link from individual contribu-tions to future pension rights. Thewidely perceived need to top upmoderate state pensions with per-sonal pension savings underpins in-centives to work and save.

In the late 1980s, private pen-sion savings covered about one thirdof the labour force, mainly white-collar workers. Subsequently, a grad-ual build-up of supplementary pen-sion schemes has taken place for al-most the entire labour market. Thedecisions to build up labour marketpensions were taken by the em-ployer and labour organizations inthe context of collective bargainingagreements. Individual retirementaccounts are also widespread.

Notwithstanding the introductionof a tax on pension returns (see be-low), as well as altered rules for taxdeductions of pension contributionsand taxation of pension pay-outs,the tax treatment of pension savingsremains preferential to other savings.

There are no signs that the reducedtax subsidy has restricted the build-up of private pensions savings (seechart 5b).

The build-up of collectivelyagreed and mandatory labour marketpensions presumably contributes tohigher total private savings, thoughthey may partly substitute for otherindividual savings. Particularly house-holds with short planning horizonsand households at a point in theirlife-time income cycle, where theydo not wish to save for pensions,may save more than they would oth-erwise have done.5 Without thebuild-up of new schemes, net pen-sion payouts might have ensued bynow, affected by the very high taxfree investment yields up to theearly 1980s for previous cohorts ofpension savers.

Formerly very generous depreci-ation rules for investment expendi-ture reduced the efficiency of invest-ment decisions, and distorted the

5 This applies to households that cannot easily offset mandatory pension savings by running down liquid asset hold-ings or take reasonably-priced loans, including many first-time house owners and many who live for rent. Asliquidity constraints ease over time for such households, their savings may decline relative to a counterfactual withless mandatory saving. But new generations of constrained households may offset this decline.

��%'��+,�6�&��-��-�5��3��-��-

��

�

�

�

0�

�� #�#�� "�� ##��"��)��#,��#$

�#� �#��

�)&*%�+�+)%,-�$�&�*%�$�$.��$

�#�� #�� #� �## �##� �##� ��

�$�+��*%�$�$.��$��/�$��8�$�+�*%�$�$.��$

Mads Kieler

Sflren Gaard

69�


relative price between the produc-tion factors capital and labour.Depreciation rates have been re-duced, and the tax base for companytaxation broadened, which has al-lowed a reduction in company taxrates (see table 2).

Strengthening public savingsAs the rise in unemployment due toadverse economic shocks during the1970s and early 1980s interactedwith still unreformed Danish institu-tions, the high unemployment rateincreasingly turned out to be of astructural nature. High interest ratesalso made for a rapidly rising debtservice burden, and fiscal policy had

to be tightened to halt the build-upof debt.

Much of the movement in theactual budget balance over the lasttwo decades reflects changes in the(estimated) structural budget balance(see chart 6a). Discretionary fiscaltightening was the main reason forthe sharp improvement in the struc-tural balance in the mid-1980s. Es-pecially since 1995, an estimated re-duction in structural unemploymenthas provided a major contribution tothe improvement in the structuralbudget balance (see chart 6b).

Tighter control of real publicconsumption growth was institutedfrom 1982 to 1992 (see chart 6c).

%��2��.�� !7 #� �

��

��+�%

!

0!

0

0#

�� #�#�� "��)��#,��#$

�#�

�(",$��+�% $ ��( �(��+��+�% $

��#�� #�� #� �#�� ## �##� �##� �## �##� �

��-��3��-��-��-

��

��+�%

�

�

!

�

�

�� #�#�� "�� ##��"��)��#,��#$

�#� �#�� #� �#�� ## �##� �

�)��+�$85$%"*�(�$�$��)��+�5�7&$%�. �)��+�5�7&$%�.

Mads Kieler

Sflren Gaard

70 �


Some relaxation was permitted inthe 1993—2001 period, but spendingrestraint has been reinforced in2002—04 with an average real con-sumption growth of less than 1%per year. Public transfers (and publicsector wages) are largely indexed toprivate sector wages.

The total tax-to-GDP ratio in-creased significantly during the1980s and, to a lesser extent, during

the 1990s. Still around 50%, the taxratio has eased down in recentyears, helped by the current govern-ment�s institution of a tax freezefrom 2001. The tax freeze fixes alltax rates whether set in percentageterms or in nominal amounts (e.g.excise duties). For local and regionalgovernments, the tax freeze appliesto the average tax rate across the na-tion, with flexibility allowed for tax

��3��'�)��2��&��

��2��.��-�� !!"

��

��+�%

!

�

�

0�

�� #�#�� "��)��#,��#$

�##�

�)%��*�(�*)%�/�)&�.��( �(��+�(%$&5+)7&$%� �(&(+��$"� ��%,$�*%�.��( �(��+��+�% $

�!�## �##� �##� �### � ��

��3��/��-��-

��

��+�%

�

�

�� #�#�� "��)��#,��#$

�85$%"*�(�$��(�"$% ��8��(�"$%�$3$%($��(�"$%

�#� �#�� #�� ##� �## ��#�� #� �## �##� �##� ��

��3��-��

��"

�

�

�

0�

�� #�#�� "��)��#,��#$

�#� �#� �##�#�� #�� ##� �

��

��

��

��

Mads Kieler

Sflren Gaard

71�


changes in individual municipalitiesor counties, as long as they are off-set by opposite movements else-where.

An important contribution to fis-cal consolidation was the introduc-tion of a tax on investment incomein the private (and semi-public) pen-sion funds and life insurance compa-nies as of 1983. Originally in theform of a real interest rate tax, thetax regime has gone through a seriesof changes and today consists of a15% flat rate on all types of invest-ment returns (see table 2).

4 Key Features of theDanish Labour Market

The Danish labour market is charac-terized by relatively generous unem-ployment benefits combined with

highly active labour market policies,relatively stringent availability-for-work requirements and liberal firingrules. These institutions, togetherwith generally strong basic schoolingand fairly high educational attain-ment levels, contribute to a highlyadaptable labour force, with a jobturnover among the highest in theworld. Today, government involve-ment in wage setting is minimal,with no legislated minimum wage,and wage negotiations are predomi-nantly decentralized.

Unemployment benefitsUnemployment benefits are gener-ous for low-income individuals cov-ering 90% of previous income — thehighest rate in the OECD. Benefitsare capped at a maximum, however,

��--��-�%��*��5��4��-� !!!

��

�

�

�

�� "�1 � �#��2�� #�3� �4$.� � +��##�� #�� ) ��#�� #�� * � �� #$�5��6��"�) ��)�� #��) �-�*��1 � �$�/)

��7��#�) �#��) ��7�� #�� 3864�� ###�� 38&4�* � �#$/) ��#�� 9�� #�() ��) �� #�� $

� �� 9 �� 4

0�6��'��&��-��%�.��'��

��

: ��#

�

�

�� "�1 � �#��2�� #�3� �4$.� � +��##�� #�� ) ��#�� #�� * � �� #$�5��6��"�) ��)�� #��) �-�*��1 � �$�/)

��7��#�) �#��) ��7�� #�� 3864�� ###�� 38&4�* � �#$/) ��#�� 9�� #�() ��) �� #�� $

� �� 4 �� 9��

Mads Kieler

Sflren Gaard

72 �


which corresponds to roughly 60%of average wage income for the em-ployed. Hence, only 10% to 15% ofthe unemployed receive benefits atthe 90% compensation rate. Benefitsare less generous for high-incomeindividuals than in some EU-15countries. Benefits are not reducedover the course of the unemploy-

ment spell and the maximum dura-tion is relatively long. Total compen-sation is the highest in the OECDwhen measured over a five-year pe-riod (see chart 7a and 7b).6

The potential adverse incentiveeffects from generous benefits may,however, to some extent be counter-balanced by relatively tight �availabil-

,�*��--��'�%��3��+'��+8��$��9��-�� :

��

6��

�

!

�

�

�� "�;�## ��/)��*��3� �4$.� � /) �� 7��'��*��0��0(�� <�� #�#��#�� #��' �� ()�) �� ) �=�*��-��

2��+�� ) ��%��0�8�� 7� �#$�/) �� 7��' �#�� <�� #�� *��($

� ��

��* �%$..�*%��$��3�*+��*+*�7��(+$.

� :��##!

��%)0��6��-�-��#

��-��)�3��5�� !!!

��

�

��

�

�

�

�

�� "�;�## ��/)��*��3� �4$.� � /) �� 7��'��*��0��0(�� <�� #�#��#�� #��' �� ()�) �� ) �=�*��-��

2��+�� ) ��%��0�8�� 7� �#$�/) �� 7��' �#�� <�� #�� *��($

�85$%"*�(�$.�)%��;�:��*) �� ! �

�� 4

��

�9

��

��

��

��

��

��

��

��

��

�

��

��

��

6 The OECD summary measure of gross replacement rates over a five-year period does not take into account that insome countries the unemployed may be entitled to a new benefit period after participating in an ALMP program.Hence, the maximum period of obtainable benefits may be much longer than indicated by the OECD measure.Also, in Denmark, periods on activation benefits are counted in the duration of unemployment but may be consid-ered a separate activity in other countries. This impedes international comparisons using the OECD summarymeasure or statutory maximum durations.

Mads Kieler

Sflren Gaard

73�


ity-for-work� requirements for theunemployed (see chart 8a).

Active labour market policies(ALMP)Denmark spends the second highestshare of GDP in the EU on ALMPwhen measured relative to the levelof unemployment (see chart 8b).The extensive focus on ALMP hasseveral objectives. First, ALMP is —if suitably implemented — skill en-hancing either through formal edu-cation or on-the-job training. Sec-ond, ALMP may alleviate bottle-necks in a tight labour market.Third, and potentially most impor-tant, ALMP may be an effective wayof testing availability-for-work dueto the requirement of participationin active measures already after sixor twelve months of unemployment,depending on individual circumstan-ces. ALMP is considered at least as

important in this respect as formalavailability-for-work requirements.

Flexible institutions andan adaptable work forceThe level of employment protectionlegislation (EPL) is low (see chart 9a).Turnover in the labour market ishigh, with Danish workers changingjobs more often than in other coun-tries for which data are available(see chart 9b). Across countries,low EPL appears to be correlatedwith low structural unemployment(see chart 9c), Portugal being amuch-noted exception. Productmarkets seem reasonably competitive(although considerable scope for im-provement remains) and productmarket regulation is liberal com-pared to continental Europe (seechart 9d). Basically very few at-tempts are made at halting structuraladjustment in the private sector

��)�*�-��

��#�

6�� 7

�

�

�

�� 3�>>�"��>>94"��.�� $�3�>>>4"�2��1��3� !4"�1� �� $�3� !4"��'#��;��(�� 3�>>>4$

��

;�3�,��

��#�

�

�

�

�� 3�>>�"��>>94"��.�� $�3�>>>4"�2��1��3� !4"�1� �� $�3� !4"��'#��;��(�� 3�>>>4$.� � /) ��?�*��' ��' ��) ��# ��$

� ��

Mads Kieler

Sflren Gaard

74 �


through government intervention orsubsidies. State subsidies for privatebusinesses are low and there is littleor no direct state ownership of e.g.manufacturing entities or financialinstitutions.

At least in the Danish context, astrong social safety net and generousunemployment benefits may be anecessary prerequisite for maintain-ing broad political support for theseflexible institutions.

Labour supply-enhancing welfarespendingSome welfare spending acts to raiselabour supply, notably subsidizedchildcare and care for the elderly,i.e. functions that were formerlytaken care of at home or in the fam-

ily. Most local governments guaran-tee availability of childcare for pre-schoolers, with parental co-paymentlimited to a maximum of 33% ofcosts. For older children, after-school care is widely available atsubsidized rates, and elderly care iswell developed, including care andassistance in elderly persons� ownhomes.

Supply and demand for labour acrossskill levels and wage-settingThe Danish education system ap-pears to have met the trend-like in-crease in demand for skills, perhapsto a larger extent than in othercountries. Hence, shifts in demandfrom low to high skilled labour ow-ing to technological progress and re-

2��

��)�*�-��+ "��

��#

�

�

�

�

�

�� 3�>>�"��>>94"�.�� $�3�>>>4"��2��1��3� !4"�1� �� $�3� !4"��'#��;��(�� 3�>>>4$

�)6� �! ��

�&5+)7&$%��5�)�$ �*)%:�*%"$8 <*,�

��

��

�

��

��

�

��

��

��

��

��

��0��$��*�� !!7

��#"

6�� 7

�:

�:�

�:

:�

:

�� 3�>>�"��>>94"�.�� $�3�>>>4"�2��1��3� !4"�1� �� $�3� !4"��'#��;��(�� 3�>>>4$

� � �� 9 �� 4 ��

Mads Kieler

Sflren Gaard

75�


inforced by globalization seem tohave been met by similar shifts onthe labour supply side (see Fosgerauet al., 2001). Relative unemploy-ment and non-employment rates byskill have remained broadly constant,or even narrowed, and so have wagedifferentials across skill levels (seechart 11a and 11b).

Wage setting has become increas-ingly decentralized during the 1990s.Collective bargaining now mainlycovers areas such as pensions, ma-ternity leave etc. Among workerscovered by collective bargaining, theshare of workers for which wagesare fixed in collective (sectoral)agreements has fallen from 34% in

1989 to 15% by 2000. The vast ma-jority of wages are set at companyand individual level.

6 Successes and Failuresin the 1990s

The labour market policies of the1990s comprised successes but alsounsustainable elements, which havesubsequently been curtailed. Thekey success was a marked reductionin structural unemployment duringthe 1990s (see chart 12a).7 How-ever, labour force participation con-tracted in the first half of the decadeand has only picked up slightly sincethen (see chart 12b).

��%��*�8��

��

�

�

�

�

!

�� "�=�*��5�� ' �$

��

,��

��

��(��

�

�

�� "�=�*��5�� ' �$

��

�#� �#�� #�� ##� �## ��#�� #� �## �##� �##� ��

��-�� $%&��'

�*.�)%� �*�$�*�

7 The structural unemployment rate is estimated in an unobserved components model using the Kalman filter. Simi-lar results are found by the OECD and the EU Commission, although the drop in both actual and structural un-employment is less pronounced when the EU/ILO harmonized measure of unemployment is used instead of thenational measure shown in chart 12a.

Mads Kieler

Sflren Gaard

76 �


6.1 Measures to Reduce StructuralUnemployment

In the first half of the 1980s, policy-makers were preoccupied with whatseemed an excessively high averagewage level in the economy, a diagno-sis founded on the combination ofpersistent current account deficitsand high unemployment. Repeatedattempts at income policies, eitherthrough direct government involve-ment or pressure on the labour mar-ket parties to settle on �responsible�wage agreements, turned out lesssuccessful than hoped-for — althoughthe abolition of price indexation ofwages as of 1982 was an importantand useful step, particularly in thecontext of disinflation.

The experience of acceleratingwage inflation during the upswing inthe mid-1980s — despite official gov-ernment policies for continued mod-erate wage developments — shiftedthe focus of policymakers towardsstructural labour market policy. Inline with the findings of the theoret-ical literature, the government�s am-bition in the 1980s was to lower re-placement rates in the unemploy-ment benefit system and alter theirstructure, but it proved impossibleto rally sufficiently broad politicalsupport behind such measures.

As of 1992—94, the agendashifted to reforms of active labourmarket policies and other parame-ters in the unemployment insurancebenefits (UIB) system than the gen-eral replacement rates.

��-�3��2$��

��''��'��,��

��

6�� 7

��

�

�

�� #�#�� "��"��)��#,��#$

<*,�-.'*++$"

�#� �#�� #�� ##� �## ��#�� #� �## �##� �##�

�)6-.'*++$" �'*++$"

��8�*�-�3��2$��

��

&' �� (�� @��

�

��

��

�

�

�� #�#�� "��"��)��#,��#$

<*,�-.'*++$"

�#� �#�� #�� ##� �## ��#�� #� �## �##� �##�

�)6-.'*++$" �'*++$"

Mads Kieler

Sflren Gaard

77�


Active labour market policies(ALMP)Prior to 1994, the maximum benefitduration was 2� years, but the un-employed could qualify for a newbenefit period by participating inactive labour market programs.Effectively, the maximum durationof benefits was approximately eightyears. Availability-for-work require-ments were relatively lax. By 1994,substantial reforms were initiated,and these were extended in the sub-sequent years.

Firstly, participation in active pro-grams would no longer qualify theunemployed for a renewed benefitperiod. This was a crucial elementin the reforms, shifting the focus ofthe ALMP system more firmly togetting people into work rather thankeeping the unemployed in the bene-fit system. The formal maximumbenefit period was set at seven years.Subsequently, the maximum period

was reduced to five years (in 1998)and currently to four years (since2000).

Second, participation in activeprograms became obligatory in apolicy known as �right-and-duty� ofactivation. Sanctions where put inplace in case of refusal or for peoplethat dropped out. Initially, the right-and-duty to participate became ef-fective after three years of unem-ployment, but this was movedforward gradually and by the end of2002 it was effective after one yearof unemployment. The activationmeasures may be private or publicjob placement, education or retrain-ing schemes, or other (quantitativelyless important) activities. Over theperiod, various changes have beenmade to the exact content and scopeof the active measures.

Third, among reform initiativestargeted at specific groups, unem-ployed aged 25 years or less without

%��2��

��

��*��

��

�

�

�� #�#�� "��"��)��#,��#$

��( �(��+�(%$&5+)7&$%�

�#� �#�� #�� ##� �## ��#�� #� �## �##� �##� ��

�%$&5+)7&$%�

)�3��5��

��

��(��

�

��

�

��

�� #�#�� "��"��)��#,��#$

�#� �#� �##�#�� #�� ##� �

Mads Kieler

Sflren Gaard

78 �


a qualifying education are no longerallowed more than 26 weeks onpassive benefits (since 1996). There-after, they must either work orenter education at a benefit levelcorresponding to roughly half ofthe UIB.

There was also an increased fo-cus on alleviating bottlenecks on lo-cal labour markets, and a host ofother changes of an administrativenature.

Tightened eligibility criteriaand formal availability-for-workrequirementsWork requirements for qualifying forUIB were extended from 26 weeksto 52 weeks of work in the preced-ing three years. When the benefitperiod was shortened in 1998, thework requirement was lowered to26 weeks for people who exhausttheir benefit entitlement due to con-cerns that an unacceptable numberof unemployed would exhaust theirentitlement. An attempt in 2002 torestore the work requirement at theprevious level was unsuccessful.

Formal availability-for-work ruleshave been tightened in several stepsand are, by now, relatively tightcompared to most EU countries (seechart 8a above).

Evaluating the impact of reformThe sharp fall in actual unemploy-ment from 1993 to 2002 withoutresurging inflation provides primafacie evidence that the structural un-employment rate has fallen since theearly 1990s. This conclusion is con-firmed by estimates of the structuralunemployment rate in unobservedcomponents models estimated by theKalman filter employed by, e.g., theOECD, the EU, and the Ministry ofFinance (see chart 12a).

To help quantify the contributionof individual policies to the reduc-tion in structural unemployment, apanel-data analysis of 19 OECDcountries is used, in a first step, toattribute variation in unemploymentrates to observable differences in la-bour market institutions, productmarket regulation, taxation etc. (seebox 1). Applying these coefficientsto the change in Danish institutionssuggests that the reduction in un-employment chiefly came from the

��-��'��*��%�� !!:# !!!

��!�

% �� #

0�

0�

0!

�� &�)��#,��#$

��;� �3�*+��*+*�7��$=(*�$&$%�. �+*,*�*+*�7��$=(*�$&$%�. �$.*"(�+��)..��$5+� $&$%��$

Mads Kieler

Sflren Gaard

79�


increased focus on ALMP. Tightereligibility and availability require-ments as well as reductions in thefive-year gross replacement ratethrough shortening of the benefitperiod also contributed substantially(see chart 13a).

Meanwhile, active labour marketpolicies entail sizable (direct) budg-etary costs amounting to roughly1� percent of GDP, with lesser de-viations from year to year. More-over, the lowering of unemploymentthrough ALMP schemes is by someconsidered to be partly �cosmetic�.

Despite the bringing forward of thepoint of activation, the total numberof persons in active schemes hasbeen relatively constant at around23/4 percent of the labour force (seechart 13b) (although this implies anincrease in numbers relative to thelower number of unemployed).

Private and public job trainingmay also potentially �displace� regu-lar workers, despite formal require-ments that attempt to prevent dis-placement. The size of displacementis not well-known (though studies inSweden indicate that it may be sig-

��-��-��%)0��2��-

��!�

��*��

��

�

�

�� &�)��#,��#$

�## �##� �##� �## �##� � ��

�%$&5+)7&$%��5+(.��;��5��* *5�%�.�%$&5+)7&$%�

Explaining Cross-Country Variation in Unemployment Rates

Panel data regressions for 19 OECD-countries are used to attribute variation in actual unemployment rates todifferences in labour market institutions, product market regulation and taxation. The results show that highunemployment insurance benefits (UIB), long duration of UIB, passive labour market policies, and limitedtesting of availability-for-work are all factors which can raise structural unemployment (see table 3).

Table 3

Impact of institutions on unemployment, 1963/83—1999

Coefficient Standard deviation

Higher UIB compensation 0.054 (0.023)Longer duration of UIB1 0.006 (0.005)Tighter availability for-work requirements —0.020 (0.010)Higher expenditure on ALMP2 —0.001 (0.000)Tighter eligibility (prior work) requirements —0.037 (0.020)Stronger job protection 0.004 (0.002)More centralized wage bargaining —0.013 (0.004)Higher degree of unionization 0.189 (0.010)Total labour taxes1 0.047 (0.015)

Source: Gaard (2004).Note: The coefficients were obtained from several separate regressions. In addition to the listed variables, deviations from HP-trend of log real GDP,country dummies, and, in some specifications, time dummies enter the regressions.1 Not significant in the 1983—99 sample, presumably owing to difficulties in constructing the indicator.

The coefficient is based on a sample covering 1963—99.2 Measured in units of GDP per unemployed in relation to the labour force.

Box 1

Mads Kieler

Sflren Gaard

80 �


nificant). Private job-training is lesspredominant in Denmark, however,than in countries where displace-ment effects have been more activelydiscussed, perhaps because wagesubsidies are relatively moderate.

The positive impact of active la-bour market policies may in part beattributed to a motivational effecton job search among the unem-ployed. In this vein, a considerablebody of evidence suggests that the�right-and-duty of activation� enhan-ces job search activity in the run-upto the time when activation becomesmandatory (see chart 14a). The bring-ing forward of the point of activa-tion from 36 to 24 months between1996 and 1998 led to a significantincrease in search at 24 months.

Experience shows that job place-ment in private firms has a highersuccess rate at getting the unem-ployed lastingly into employment

than training measures (see chart14b); — although this may in partreflect a selection bias in that privatejob placement may be offered tothose unemployed that are most job-ready.

6.2 Labour Supply Management andMismanagement in the 1990s

Many European countries, includingDenmark, have attempted to reducehigh unemployment — even whenmainly of a cyclical nature — by re-ducing labour supply, e.g. throughearly retirement schemes. Suchmeasures are, however, short-sightedin that the labour resources are nolonger available, once the cyclicalupswing materializes and businessstarts demanding labour again. Inthis way, cyclical downturns in em-ployment tend to become structural,with adverse medium-term conse-quences for public finances.

��'��

��

&�'�� #��#�3��4

�

�

�� + ��# ��3� �4$

�� ! !

�%$&5+)7&$%��.5$++:�&)%��.

��##� �##

&��'�2��'+2�''��2�6�0��-��'��%��

��

�

�

!

�

�

�� + ��# ��3� �4$

��*3��$�1)��5+� $&$%� �"( ��*)% �)��+�(�+* �1)��5+� $&$%� 4��$�� *3��*)%

Mads Kieler

Sflren Gaard

81�


Unemployment in Denmark rosesharply during the cyclical downturnfrom 1987 to 1993, and from 1992to 1994 a pre-early retirementscheme was gradually introduced forlong-term unemployed 50—59 yearolds (this scheme was in addition tothe already existing early retirementscheme for 60—66 year olds, insti-tuted during the recession in 1979).In addition, leave schemes were in-troduced that gave workers the rightto take leave of absence for educa-tional purposes (at 100% of the UIBrate), childcare (at 80% of UIB),and sabbatical (80%). Particularly asof 1994, the uptake on theseschemes reduced the labour forcesignificantly (see chart 15a).

The schemes have subsequentlybeen tightened and to a large extent

abolished. Hence, the permanent ad-verse effects on labour supply havebeen much smaller than if cuts inworking hours or retirement ageshad been introduced.

The pre-early retirement schemeprovides an interesting and caution-ary tale as it acted mainly as a sub-stitute for employment, not for long-term unemployment, as intended.Despite the large uptake on thescheme from 1994 to 1996, long-term unemployment among the50—59 year olds fell only briefly,and it subsequently developed largelyin parallel with long-term un-employment among 40—49 year olds(see chart 15b). It appears that thenatural flow into long-term un-employment (and out of long-termunemployment in the absence of the

��-��-��)��5��+,��9��-

��

�" �� #��#

�

�

�

�� #�#�� "��)��#,��#$

�##�

��*+"� ��$

�##� �## �##� �##� �### � �� !

�"( ��*)%�+�5(�5).$. ��* �+

��+��'��)��*+,��

�" �� #��#

�

�

�� #�#�� )��#,��#$��8� �� )��)#$

�##

�(&�$��)/�5$�.)%.�)%�5�$-$��+7��$�*�$&$%�

�##� �##� �## �##� � ��

�)%,-�$�&�(%$&5+)7&$%�:��,$"��0�#�)%,-�$�&�(%$&5+)7&$%�:��,$"��0�#

Mads Kieler

Sflren Gaard

82 �


scheme) had been underestimated.In the run-up to 1996, when accessto the scheme was possible, employ-ment among 50—59 year olds fell,while it rose for other age groups.After 1996, when access to thescheme was closed, employmentamong 50—59 year olds developedstronger than total employment (seechart 16a).

Participation in the temporaryleave schemes has dwindled to al-most nothing as compensation levelswere reduced to first 70% and then60% of UIB for childcare and sab-batical. Access to educational leaveand sabbatical has been closed.Lately, the extension of maternityleave periods to one year has causeda sharp decline in the number of

persons taking leave for childcarepurposes.

Other welfare schemes have alsobeen reformed, but participation isstill high. Reforms of the disabilityschemes have cut the annual inflowand reduced the total number ofrecipients to below 1995 levels (seetable 4).

The early retirement scheme stillremains in force. Despite a reformin 1998, which is estimated to re-duce the number of recipients by15,000 persons (almost 10%) overtime, the total number of people inearly retirement amounts to 6% ofthe labour force in 2003.

Labour supply has also been re-duced through reductions in collec-tively agreed weekly working hours

��%��*�" #"!�4��/��-

��

�

�

��

��

�

�� #�#�� "��)��#,��#"��(�$

�##

�&5+)7&$%��$:��,$"��0�# �&5+)7&$%��$:�)��$�.

�##� �##� �##! �##� �##� �## �##� �##� �### �

(��3��'�,��-'��-�%*�� "#<<

��

��(��

�

��

�

�

�� #�#�� )��#,��#"��(�$

�#�

�* '��%"�&��$�%*�7�+$�3$

�#�� #� �#�� ## �##� �

�*.��*+*�7�. �$&$.�%$&5+)7&$%��. �$&$. �$�*�$&$%��. �$&$.

Mads Kieler

Sflren Gaard

83�


(from 40 to 37 hours, phased induring 1987—90) and an extensionof annual leave from five to sixweeks of vacation through 2000 to2004 (see table 4).

Despite the cut back on leaveschemes introduced in the mid-1990s, the total number of transferrecipients of working age is stillhigh, at more than 20% in 2003(see chart 16b).

7 Alleviating theDisincentive Effectsof High Taxes

Total tax revenues are high, but thedistortion compared to other coun-tries is probably smaller than mightappear at first look. First of all, thetax ratio is exaggerated by the factthat transfer incomes in Denmark(as well as some other countries,including Sweden) are taxed, whilein most other countries they arepaid �net-of-tax�. This feature aloneboosts the measured tax ratio in

Denmark by 4% to 5% of GDP rel-ative to most OECD countries.8 Sec-ondly, exemptions and loopholeshave been virtually eliminated fromthe tax system, and this has madepossible a reduction in marginal taxrates. Third, means-testing of in-come transfers or public services topersons of working age is not verypredominant in Denmark. In manyother countries, means-testing e.g.for parental payments to child careimplies that composite marginalrates can be higher than would ap-pear from the tax system alone.

The marginal effective tax rate,including employers� social contribu-tions, is close to the average of EUcountries for incomes at 100% ofthe average wage, but among thehighest for incomes at 150% of theaverage wage (see chart 17a and17b).

High taxes have not preventedhigh participation rates, in part be-cause the negative substitution effect

Table 4

Labour Supply (Mis)Management

Introduced Eligibility Participation Reform

Pre-early retirement 1992—94 Age 50—59 years,long-term unemp-loyed UIB member

46,000 in 199611,500 in 2003

Access closedin 1996

Temporary leave 1992—94 Children andeducation

77,000 in 19955,000 in 2003

Gradually tighte-ned 1995—2002

Disability Before 1979 Disabilityor in needof permanentsocial support

270,000 in 1995259,000 in 2003

1998 and 2000:Annual inflow

reduced strongly

Official working hours 1987—90

2000—04

From 40 to 37hours a week.

From 5 to 6 weeksof annual vacation

Early retirement 1979 Age 60—66(from 2004

60—64)1

UIB member

115,000 in 1995179,000 in 2003

(6% of totallabour force)

1998 (Impactassessment

15,000)

Source: Author�s calculations.1 The official retirement age was reduced from 67 to 65.

8 Correction for other technical factors, especially the inclusion of indirect taxes in the GDP measure in the denom-inator, would pull in the opposite direction.

Mads Kieler

Sflren Gaard

84 �


on work effort from high marginaltaxes is off-set by a positive incomeeffect, requiring people to workmore to attain a given level of in-come net of tax. Econometric evi-dence of labour supply elasticities ismixed and estimated elasticities are

subject to considerable margins ofuncertainty. Tax progression may havecontributed to lower average workinghours, through more vacation daysand fewer weekly hours, while con-tributing to the tendency for mostfamilies to rely on two incomes.

0��*��''��,�6��-�� =��'�%��*��8�*�

��

�

�

�

!

�

�

�� ##��$.� � /(�0 �� ()�(��)�� $

��

0��*��''��,�6��-�� " =��'�%��*��8�*�

��

�

�

�

!

�

�

�� ##��$.� � /(�0 �� ()�(��)�� $

��

Table 5

Marginal Tax Rates, 1980—2004

1980 1986 1993 1998 2003 2004

%

Marginal income tax, average employed x 57.7 59.3 54.3 53.8 52.2Marginal income tax, highest bracket 68.8 73.2 68.7 62.0 62.9 62.9Marginal income tax, first bracket 43.6 48.0 50.6 45.2 43.7 42.9

Source: Ministry of Finance, Denmark.

Mads Kieler

Sflren Gaard

85�


For the average employed per-son, the marginal tax rate has beenreduced from close to 60% in 1993to 52% in 2004 (see table 5). Lowermarginal tax rates have been fi-nanced e.g. by lower mortgage in-terest deductions and higher indirectand green taxes.

The income distribution remainsamong the most even in the OECD,and (on standard measures) consid-erably more even than in the averageof the EU (see table 6).

8 The Danish Experience

Structural reform is feasible withoutweakening the basic functions of thewelfare state. Reforms of labour andproduct markets generally require agreater willingness to change andadapt. But flexible institutions donot necessarily imply high incomeinequality, at least not if skills arenot too unevenly distributed and theeducation system meets the in-creased demand for skills. Welfarespending has been reformed withoutimportant adverse distributional con-sequences. Also, it has been possibleto reduce distortions within the taxsystem, e.g. lowering marginal taxrates, without reducing the effectiveprogression of the system or overalltax revenue. The tax wedge on la-bour remains high, though, and isabove the EU average, at least forhigher incomes.

Welfare expenditures may be a pre-requisite for public support for flexible

institutions. Liberal firing rules, rela-tively light product market regula-tion and a hands-off approach tostate subsidies or state involvementin market-based production mightnot have been politically feasible ifnot accompanied by a strong socialsafety net for those affected bystructural change and unemploy-ment. The emphasis is not on pre-venting job losses associated withstructural change but on cushioningthe income loss and providing op-portunity as well as incentives fordisplaced workers to transition tonew jobs.

Deep recessions feed bad policies —it is important not to do irreversibledamage. During recessions, externalcompetition and technological prog-ress are often blamed for the eco-nomic woes. Pressure may arise forincreased government intervention,subsidies, working hours reductions,early retirement schemes, and some-times protectionist measures. Suchmeasures can do medium-termdamage to output and public finan-ces, and are not easily reversible.Reduced labour supply should beavoided or, at the very least, re-stricted to measures of a temporarynature that can be undone whenlabour demand picks up.

For countries with a fixed exchangerate or participants in a monetaryunion, there may be some scope for de-mand management through administra-tive credit measures or borrowing in-

Table 6

Indicators of Income Distribution

DK1 EU2

Income quintile ratio (s80/s20) 3.2 4.4Gini-coefficient 23 28Share of persons living in low-income households 4 9

Source: Authors� calculations.1 National calculations based on incomes from 2002.2 Joint Inclusion Report, 2004. Based on incomes from 2000.

Mads Kieler

Sflren Gaard

86 �


centives, particularly those that affectthe housing market. Opportune tim-ing of changes in credit market reg-ulations and borrowing incentives —including mortgage interest deduct-ability — have had strong and gener-ally stabilizing demand effects inDenmark. Hence, demand manage-ment is not restricted to (tradi-tional) fiscal policy even thoughmonetary policy is bound by thefixed exchange rate.

Significant policy challenges re-main to prepare for the ageing of thepopulation. The Danish government�s2010-strategy aims at ensuring thesustainability of current policies, i.e.ensuring that current rules govern-ing social transfers (indexation, re-tirement age, early retirement etc.)and public service standards can besustained without imposing a needfor future generations to raise taxesto prevent an unsustainable debtpath. The strategy entails runningfiscal surpluses of 1�—2� percent ofGDP on average from now to 2010and implies that the relatively largecurrent generations of working agecontribute to the financing of theirfuture state pensions, elderly careetc. The strategy presupposes realpublic consumption growth of 0.5%per year from 2005 to 2010 andfurther structural reforms that canraise sustainable employment byroughly 2% by 2010 (see Ministryof Finance, 2003). Further reformsmay be considered in the context ofthe Danish Welfare Commission�sreform proposals due by 2005, e.g.with a view to make the welfare sys-tem more robust to possible changesin life expectancy (relative to the in-crease assumed in current baselineprojections). Future generations maywish to make different priorities no-tably with respect to taxes, public

services, or working hours. Withthe 2010-plan already in place, thereis reasonably ample time to discussand consider possible reforms. §

References

Alesina, A. and R. Perotti. 1995. FiscalAdjustment: Fiscal Expansions andAdjustments in OECD Countries. Eco-nomic Policy. October.

Botero, J., S. Djankov, R. L. Porta, F.Lopez de Silanes and A. Shleifer.2003. The Regulation of Labor. NBER.Working Paper 9756.

Callesen, P. 1997. The Policy Experienceof Structural Reforms in Denmark. Paperfor the Conference: The Effects and Pol-icy Implications of Structural Adjustmentsin Small Open Economies. Amsterdam.October 23 and 24. Available from theDanish Ministry of Finance.

Davis, S. and J. Haltiwanger. 1999.Chapter in Handbook in Macroeconom-ics. Elsevier Science. Amsterdam.

Economic Policy Committee. 2003.The Impact of Ageing Populations onPublic Finances: Overview of AnalysisCarried out at the EU Level and Pro-posals for a Future Work Programme.http://www.europa.eu.int/comm/economy_finance/epc/documents/2003/pensionmaster_en.pdf

Fosgerau, M., S. E. H. Jensen and A.Sflrensen. 2001. Relative DemandShifts for Educated Labour. CEBR Discus-sion Paper 2000—11. Copenhagen.

Frederiksen, N. K. 2003. Fiscal Sustain-ability in OECD Countries. December2002. Updated March and November2003. Mimeo. Danish Ministry of Finance.

Frederiksen, N. K. 2004. Fiscal Policy inOECD Countries, 2000—2005. March.Mimeo. Forthcoming in: Danish Ministryof Finance. Working Paper Series.

Gaard, S. 2004. Assessing the Impact ofLabour Market Reforms in the 1990s —A Panel Data Study of 19 OECD Coun-

Mads Kieler

Sflren Gaard

87�


tries. Ministry of Finance Working Paper.Forthcoming.

Geerdsen, L. P. 2002. Does LabourMarket Training Motivate Job Search? AStudy of Incentive Effects of CompulsoryALMP in the Danish UI System. SFIWorking Paper 23.

Hasselpflug, S. and A. K. B. Thorball.2004. Availability Criteria in 25 OECDCountries. Ministry of Finance WorkingPaper. Forthcoming.

Ministry of Finance. 2003. ConvergenceProgramme Update. http://www.euro-pa.eu.int/comm/economy_finance/about/activities/sgp/scplist_en.htm

Nicoletti, G., S. Scarpetta and O.Boylaud. 1999. Summary Indicators ofProduct Market Regulation with anExtension to Employment ProtectionLegislation. OECD Economics Depart-ment Working Paper 226.

Scarpetta, S. 1996. Assessing the Roleof Labour Market Policies and Institu-tional Settings on Unemployment: ACross-Country Study. OECD EconomicStudies 26. 43—98.

World Bank. 2003. Doing Business Data-base. http://rru.worldbank.org/DoingBusiness/default.aspx

Mads Kieler

Sflren Gaard

88 �


Two Decades of Structural Reform in Denmark: A Review2f348b20-11b6-45ed-b7da-600d3aa8c31f/... ·...

Documents

Transcript of Two Decades of Structural Reform in Denmark: A Review2f348b20-11b6-45ed-b7da-600d3aa8c31f/... ·...