+ A Markov system analysis application on labour market - PowerPoint PPT Presentation

+ A Markov system analysis application on labour market dynamics: The case of Greece This project has received funding from the European Union’s Horizon 2020 research and innovation programme Maria Symeonaki under grant agreement No 649395 Glykeria Stamatopoulou Panteion University of Social and Political Sciences IWPLMS, 22-24 June, Athens, Greece

+ The structure of the presentation 2 n Motivation – Objectives n Data n Methodology n Results – Conclusions n Future work IWPLMS, 22-24 June, Athens, Greece 2

+ Motivation – Objectives 3 n To examine: n the flow data and transition rates between labour market states on a micro-level, i.e. the labour market outcomes of the individuals n Using: n quantitative analysis and Markov system theory for Greece * *Source: Hellenic Statistical Authority IWPLMS, 22-24 June, Athens, Greece 3

+ Data 4 4 EU LFS n The EU LFS is a large household sample survey providing results on labour participation of people aged 15 and over as well as on persons outside the labour force n All definitions apply to persons aged 15 years and over living in private households n Persons carrying out obligatory military or community service are not included in the target group of the survey, as is also the case for persons in institutions/collective households IWPLMS, 22-24 June, Athens, Greece

+ Data 5 EU LFS n Information about the labour market transitions in Greece from 2006 to 2013. n To study and compare the distribution of transition probabilities from employment, unemployment and inactiveness and vice versa n The current labour status (MAINSTAT) and the situation one year before the survey (WSTAT1Y) are the variables that will be used in the present analysis IWPLMS, 22-24 June, Athens, Greece

Country Year: 2013 Country Year: 2013 6 + Belgium 8.4 Luxembourg 5.9 by country (%), 2013 Unemployment rates (total population) 15-74 Bulgaria 13.0 Hungary 10.2 Czech Republic 7.0 Malta 6.4 Denmark 7.0 Netherlands 7.3 Germany 5.2 Austria 4.9 Estonia 8.6 Poland 10.3 Ireland 13.1 Portugal 16.4 Greece 27.5 Romania 7.1 Spain 26.1 Slovenia 10.1 France 10.3 Slovakia 14.2 Croatia 17.3 Finland 8.2 Italy 12.2 Sweden 8.0 Cyprus 15.9 United Kingdom 7.6 Latvia 11.9 Iceland 5.4 Lithuania 11.8 Norway 3.5 Source: Eurostat, Labour Force Survey (lfsa_urgan), http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=lfsa_urgan&lang=en

Country Year: 2013 Country Year: 2013 7 + Belgium 23.7 Luxembourg 16.9 country (%), 2013 Youth unemployment rates, 15-24 by Bulgaria 28.4 Hungary 26.6 Czech Republic 18.9 Malta 13.0 Denmark 13.0 Netherlands 13.2 Germany 7.8 Austria 9.2 Estonia 18.7 Poland 27.3 Ireland 26.8 Portugal 38.1 Greece 58.3 Romania 23.7 Spain 55.5 Slovenia 21.6 France 24.8 Slovakia 33.7 Croatia 50.0 Finland 19.9 Italy 40.0 Sweden 23.6 Cyprus 38.9 United Kingdom 20.7 Latvia 23.2 Iceland 10.7 Lithuania 21.9 Norway 9.1 Source: Eurostat, Labour Force Survey (lfsa_urgan), http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=lfsa_urgan&lang=en

10.0 20.0 30.0 40.0 50.0 60.0 0.0 + Unemployment rates (2013) Belgium Bulgaria Czech Republic Denmark Germany Estonia Ireland Greece IWPLMS, 22-24 June, Athens, Greece Spain France Croatia Italy Cyprus Latvia Lithuania Luxembourg Hungary Malta Netherlands Austria Poland Portugal Romania Slovenia Slovakia Finland Sweden United Kingdom Iceland Norway Youth UR Total UR 8

9 IWPLMS, 22-24 June, Athens, Greece

10 IWPLMS, 22-24 June, Athens, Greece

+ Basic parameters of a non homogeneous Markov system { } S = 1,2,..., k p 11 ( ) t p ( ) t p o 1 ( ) t + { } t = 0 1 , k 1 ∞ P ( t ) 1 { } t = 0 ∞ p 12 ( ) t p 13 ( ) t p o ( t ) p 31 ( ) t p 21 ( ) t { } t = 0 ∞ p o 2 ( ) t p o 3 ( ) t p k + 1 ( t ) p 32 ( ) t 2 3 ∞ { } Q ( ) t t = 0 p 23 ( ) t p 22 ( ) t p 33 ( ) t T t ( ) [ ] p ( ) t p ( ) t = N ( ) t N t N ( ), ( ), t … , N ( ) t + + 2 , k 1 3 , k 1 1 2 k k = 3 + = + Δ N ( t 1 ) N ( ) ( ) t Q t T t ( ) p ( ) t o Transition diagram of a Markov system with Δ T t = − − 1 ( ) T t ( ) T t ( ) three states IWPLMS, 22-24 June, Athens, Greece

+ States in LFS 12 Carries out a job or profession, including unpaid work 1. for a family business or holding, including an apprenticeship or paid traineeship, etc Unemployed 2. Pupil, student, further training, unpaid work experience 3. In retirement or early retirement or has given up 4. business Permanently disabled 5. In compulsory military service 6. Fulfilling domestic tasks 7. Other inactive person 8. IWPLMS, 22-24 June, Athens, Greece

+ States in NHMS and possible 13 transitions 1: Employment (1) 2: Unemployment (2) 1 → 1 2 → 1 1 → 2 2 → 2 1 → 3 2 → 3 3: Inactivity (5, 6,7,8) 3 → 1 3 → 2 3 → 3 IWPLMS, 22-24 June, Athens, Greece

+ Basic parameters 2006 - 2013 14 ⎡ ⎤ p 11 ( t ) p 12 ( t ) p 13 ( t ) ⎢ ⎥ P ( t ) = ⎢ ⎥ p 21 ( t ) p 22 ( t ) p 23 ( t ) ⎢ ⎥ p 31 ( t ) p 32 ( t ) p 33 ( t ) ⎢ ⎥ ⎣ ⎦ ⎡ ⎤ p o ( t ) = p o 1 ( t ), p o 2 ( t ), p o 3 ( t ) ⎣ ⎦ ( ) ⋅ ′ ⎡ ⎤ p k + 1 ( t ) = I − P ( t ) 1 = p k + 1 , 1 ( t ), p k + 2 , 2 ( t ), p k + 1 , 3 ( t ) ⎣ ⎦ ⎡ ⎤ q 11 ( t ) q 12 ( t ) q 13 ( t ) ⎢ ⎥ ⎢ ⎥ Q ( t ) = q ij ( t ) = p ij ( t ) + p i , k + 1 ( t ) p oj ( t ) q 21 ( t ) q 22 ( t ) q 23 ( t ) ⎢ ⎥ q 31 ( t ) q 32 ( t ) q 33 ( t ) ⎢ ⎥ ⎣ ⎦

+ Mobility indices k k M B = 1 ∑ ∑ q ij i − j n Bartholomew’s index of k mobility i = 1 j = 1 ⎛ ⎞ 1 M ( PS ) = ⎟ ( n − tr ( Q )) ⎜ n Shorrocks’ index of mobility ⎝ ⎠ n − 1 ⎛ ⎞ tr ( Q ) IM = ⎜ ⎟ n Immobility index ⎝ ⎠ n ( ) M T = 1 − tr Q n Prais – Bibby mobility index k IWPLMS, 22-24 June, Athens, Greece

+ Results 2006, 2007 16 Greece 2006 Greece 2007 N = 257,228 N = 262,884 ⎡ ⎤ ⎡ ⎤ p o (2006) = p o (2007) = 0.395 0.372 0.233 0.367 0.357 0.276 ⎣ ⎦ ⎣ ⎦ ⎡ ⎤ ⎡ ⎤ 0.955 0.029 0.016 0.957 0.027 0.016 ⎢ ⎥ ⎢ ⎥ Q (2006) = Q (2007) = 0.237 0.722 0.041 0.234 0.726 0.040 ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ 0.026 0.023 0.951 0.024 0.020 0.956 ⎣ ⎦ ⎣ ⎦ IWPLMS, 22-24 June, Athens, Greece

+ Results 2008, 2009 17 Greece 2008 Greece 2009 N = 258,428 N = 255,701 ⎡ ⎤ ⎡ ⎤ p o (2008) = p o (2009) = 0.343 0.352 0.305 0.309 0.418 0.273 ⎣ ⎦ ⎣ ⎦ ⎡ ⎤ ⎡ ⎤ 0.944 0.040 0.016 0.957 0.026 0.017 ⎢ ⎥ ⎢ ⎥ Q (2009) = Q (2008) = 0.208 0.748 0.044 0.248 0.709 0.043 ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ 0.024 0.029 0.947 ⎢ ⎥ ⎣ ⎦ 0.023 0.021 0.956 ⎣ ⎦ IWPLMS, 22-24 June, Athens, Greece

+ Results 2010, 2011 18 Greece 2010 Greece 2011 N = 262,172 N = 236,551 ⎡ ⎤ ⎡ ⎤ p o (2010) = p o (2011) = 0.195 0.537 0.268 ⎣ 0.257 0.431 0.312 ⎦ ⎣ ⎦ ⎡ ⎤ ⎡ ⎤ 0.917 0.067 0.016 0.934 0.050 0.016 ⎢ ⎥ ⎢ ⎥ Q (2011) = Q (2010) = 0.117 0.848 0.035 ⎢ ⎥ 0.171 0.787 0.042 ⎢ ⎥ ⎢ ⎥ 0.015 0.034 0.951 ⎣ ⎦ ⎢ ⎥ 0.022 0.034 0.944 ⎣ ⎦ IWPLMS, 22-24 June, Athens, Greece

+ Results 2012, 2013 19 Greece 2012 Greece 2013 N = 212,307 N = 215,244 ⎡ ⎤ ⎡ ⎤ p o (2012) = p o (2013) = ⎣ 0.154 0.555 0.291 ⎦ 0.172 0.552 0.276 ⎣ ⎦ ⎡ ⎤ ⎡ ⎤ 0.903 0.078 0.019 0.904 0.076 0.020 ⎢ ⎥ ⎢ ⎥ Q (2012) = Q (2013) = 0.092 0.876 0.032 0.100 0.872 0.028 ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ 0.012 0.045 0.943 0.013 0.041 0.946 ⎣ ⎦ ⎣ ⎦ IWPLMS, 22-24 June, Athens, Greece

+ Mobility and Immobility Indices 20 Greece M(PS) IM M(B) M(T) 2006 0.186 0.876 0.138 0.124 2007 0.180 0.879 0.133 0.120 2008 0.189 0.874 0.139 0.126 2009 0.181 0.879 0.134 0.120 2010 0.168 0.888 0.124 0.111 2011 0.142 0.905 0.105 0.095 2012 0.139 0.907 0.103 0.093 2013 0.098 0.907 0.104 0.092 IWPLMS, 22-24 June, Athens, Greece

+ Mobility indices 2006 - 2013 21 0.2 0.15 M(PS) 0.1 M(B) M(T) 0.05 0 2006 2007 2008 2009 2010 2011 2012 2013 IWPLMS, 22-24 June, Athens, Greece

+ Immobility index 2006 - 2013 22 0.92 0.9 0.88 IM 0.86 0.84 2006 2007 2008 2009 2010 2011 2012 2013 IWPLMS, 22-24 June, Athens, Greece

+ Conclusions 23 Transition probabilities Input probabilities E è E U è U E è U è E è U è IA 2006 0.955 0.722 0.029 2006 0.395 0.372 0.233 2007 0.957 0.726 0.027 2007 0.367 0.357 0.276 2008 0.957 0.709 0.026 2008 0.343 0.352 0.305 2009 0.944 0.748 0.040 2009 0.309 0.418 0.273 2010 0.934 0.787 0.050 2010 0.257 0.431 0.312 2011 0.917 0.848 0.067 2011 0.195 0.537 0.268 2012 0.903 0.876 0.078 2012 0.154 0.555 0.291 2013 0.904 0.872 0.076 2013 0.172 0.552 0.276 IWPLMS, 22-24 June, Athens, Greece

+ Future work 24 n First, we will study the asymptotic behaviour of the system n The same methodology will be applied for all European countries (starting with the Southern European countries) n We will also study gender and age differences IWPLMS, 22-24 June, Athens, Greece