Abstract
Pensions for reduced earnings capacity (REC) have yearly made up about 20% of all pensions granted by the German Pension Insurance in the past 25 years. The dataset described in this article is a time series of cross-sectional observations for the retirement cohorts 2001–2020. It enables researchers to have a detailed insight into the population of REC-pensioners and allows observing trends and effects of reforms over time. The article provides general information on the data, first descriptive analyses, examples on possible matching with other data of the FDZ-RV and information on how to access the data.
1 Introduction
The vast majority of the German population is statutorily insured against the case of long-term illness with the German Pension insurance. Aside from old-age pensions, pensions for reduced earnings capacity (REC) make up most of the pensions paid by the insurance. Every year about 180,000 persons are granted a pension for reduced earnings capacity and statistics show that the share of REC-pensions of all pensions has been stable at about 20% over the last 25 years (www.statistik-rente.de).
Thus, the Research Data Centre of the Pension Insurance (FDZ-RV) possesses high quality data on this part of the population which is often underrepresented in surveys or other administrative data. In addition, since health issues are the relevant factor of receiving a REC-pension, the FDZ-RV can offer important health information which is usually difficult to obtain through surveys.
The new Scientific Use File “Pension for Reduced Earnings Capacity and Diagnoses 2001–2020” (DOI: 10.57701/SUF.RTZN.2001-2020.EM.1-0) enables a detailed insight into the population of reduced earnings capacity-pensioners and allows to observe trends and effects of pension reforms over time. In addition, the option to combine the data with other data products of the FDZ-RV broadens the fields of research opportunities.
This article provides general information on the data about recipients of REC-pensions in Germany, a description of the background of the data as well as first descriptive analyses, examples on possible matching with other data of the FDZ-RV and information on how to access the data.
2 Pensions for Reduced Earnings Capacity in Germany
Introduced already in 1889, the REC-pension is still an important element of the German social security system. It aims at securing individuals’ material welfare if they are no longer able to work for example due to chronical illnesses, but are too young to receive a regular old-age-pension (Brussig et al. 2020).
After an extensive reform in 2001, several conditions determine whether a person is entitled to receive a pension for reduced earnings capacity. Mika (2017: 34) sums up the current regulations as follows: “The main condition is that the person is no longer able to work for six or more hours a day due to chronic illness. However, a full pension is only credited if the earnings capacity is limited to less than 3 h daily. Working ability above 3 h and less than 6 h daily entitles to a partial pension, which in some cases is upgraded to a full pension if unfavorable labor market conditions predominate. A physician has to confirm a negative outlook for the chronic disease(s) in the next future.”[1]
Further reforms in 2014 and 2019 have aimed at improving the situation of REC-pensioners. This has led to different outcomes in the pension calculations between different retirement-cohorts.
3 Research Data
3.1 Data Source
In order to grant the pension, the German Pension Insurance follows a strictly regulated procedure, which produces highly reliable data for researchers interested in issues of retirement. Therefore, the FDZ-RV regularly offers cross-sectional and longitudinal datasets – scientific use files – which are samples drawn from the pool of individuals who are insured in the Federal Pension Insurance and receive a pension.
For the years 2003–2014, the FDZ-RV yearly published a scientific use file which included the newly awarded REC-pensions in the respective year. Besides demographic variables, the datasets included relevant indicators concerning the awarded pension as well as details on the diagnosis having led to the awarding of the pension. The dataset we present in this article is a time series including updated versions of the existing scientific use files for the years 2003–2014 as well as information on the years 2001–2002 and 2015–2020. Thus, these harmonized data offer an opportunity to analyze the demographic and political process of pension reforms regarding the pension for reduced earnings capacity.
3.2 Sample
The SUF.RTZN.2001-2020.EM.1-0 is a subsample of the pool of pensions awarded in the years 2001–2020. For every year in the dataset, the total of pensions awarded has been filtered for those individuals receiving a pension for reduced earnings capacity. Among these, only individuals receiving a pension for the first time in the respective year were included. Pension revaluations were excluded from the population as well as pensions awarded for the coming year(s). Out of the remaining cases a 20% random sample has been drawn leading to an average number of 34,000 cases per year. In sum, the time series dataset includes more than 680,000 cases.
3.3 Variables
Generally, the dataset can be divided into the following parts:
Data-related and demographic variables (age, sex, marital status, citizenship, geographical variables, occupation, and education)
Pension type, age, and start
Parameter for care and health insurance and special circumstances
Variables on rehabilitation and insurance including the diagnoses leading to the status of reduced earnings capacity
Parameters for pension calculation, including incomes of the past three years before pension as well as the amount of the pension payment
As geographical variables, the dataset includes the federal state (Bundesland) as well as a variable to differentiate between eastern and western German federal states (including east/west differentiation for Berlin). New in this dataset are four additional geographical variables: 96 spatial planning regions (Raumordnungsregionen), 50 core labor market regions (Kernarbeitsmarktregionen), and two variables distinguishing rural from urban areas.
Also new in this dataset are three vocational classifications that allow for the analysis of the last reported occupation before the award of the REC-pension. These variables are based on the German classification of occupations (KldB) 2010 and allow for horizontal differentiation (5 ‘occupational sectors’ and 14 ‘occupational segments’) as well as vertical differentiation between 4 skill levels of the occupation. Additionally, a 3-level education variable allows for analyzing education over time and different classification standards.
Furthermore, the dataset includes detailed information on the diagnoses which were the reason for awarding a REC-pension. These diagnoses are based on the ICD-10-GM classification and have been categorized due to anonymization efforts. However, the variables include more than 140 categories enabling a deep insight into different medical fields.
3.4 Data Quality
Given the administrative nature of the data, its quality is high compared to survey data. Especially for pensions the information in the insurance account is thoroughly verified and completed. However, this only applies for parameters relevant to the calculation of the pension. Data users are asked to carefully read the documentation to the dataset to be aware of the problems that might occur when analyzing some demographic parameters.
The diagnoses indicated in the dataset stem from pension awards and therefore have undergone consecutive verification steps by medical doctors. This means the information is highly reliable and the diagnosed diseases are highly relevant for the persons’ lives in the sense that they restrict life in a way that disables the person from working on a regular basis.
4 Descriptive Statistics
As mentioned in Section 3.2, the time series dataset includes an average number of 34,000 cases per year and more than 680,000 cases in sum.
Tables 1 and 2 show that REC-pensions are split almost evenly between men and women. However, there is a tendency of shifting towards a larger share of women receiving a REC-pension (59.6% men and 40.4% women in 2001 compared to 47.6% men and 52.4% women in 2020).
Overview of important socio-demographic variables (continued on page 6).
| 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Male | 23,819 (59.6%) | 20,323 (58.1%) | 19,854 (57.9%) | 18,924 (56.7%) | 17,937 (55.5%) | 17,241 (55.5%) | 17,419 (55.2%) | 17,286 (53.8%) | 18,127 (53.1%) | 19,012 (52.8%) |
| Female | 16,129 (40.4%) | 14,635 (41.9%) | 14,441 (42.1%) | 14,442 (43.3%) | 14,381 (44.5%) | 13,809 (44.5%) | 14,152 (44.8%) | 14,859 (46.2%) | 16,023 (46.9%) | 17,012 (47.2%) | |
| Age | Mean age (SD) | 52.2 (8.25) | 51.7 (8.31) | 51.2 (8.38) | 50.9 (8.36) | 50.9 (8.32) | 50.9 (8.25) | 51.0 (8.26) | 51.1 (8.20) | 51.2 (8.30) | 51.4 (8.33) |
| Region | West | 30,883 (77.3%) | 26,729 (76.5%) | 26,168 (76.3%) | 25,455 (76.3%) | 24,605 (76.1%) | 23,742 (76.5%) | 23,970 (75.9%) | 24,649 (76.7%) | 26,080 (76.4%) | 27,855 (77.3%) |
| East | 7955 (19.9%) | 7263 (20.8%) | 7275 (21.2%) | 7076 (21.2%) | 6847 (21.2%) | 6475 (20.9%) | 6783 (21.5%) | 6725 (20.9%) | 7335 (21.5%) | 7485 (20.8%) | |
| Other countries | 1078 (2.7%) | 930 (2.7%) | 822 (2.4%) | 810 (2.4%) | 846 (2.6%) | 799 (2.6%) | 765 (2.4%) | 761 (2.4%) | 727 (2.1%) | 677 (1.9%) | |
| Education indexa | Low (valid %) | 2679 (27.8%) | 3891 (23.8%) | 4312 (22.8%) | 4544 (23.5%) | 4421 (23.1%) | 4209 (23.0%) | 4374 (22.9%) | 4542 (23.7%) | 4527 (22.6%) | 4889 (23.2%) |
| Mid (valid %) | 6745 (70.1%) | 11,960 (73.1%) | 13,983 (73.9%) | 14,135 (73.2%) | 14,036 (73.3%) | 13,416 (73.4%) | 14,019 (73.4%) | 13,956 (72.8%) | 14,774 (73.9%) | 15,392 (73.0%) | |
| High (valid %) | 203 (2.1%) | 516 (3.2%) | 626 (3.3%) | 624 (3.2%) | 679 (3.5%) | 663 (3.6%) | 701 (3.7%) | 677 (3.5%) | 700 (3.5%) | 804 (3.8%) | |
| Unknown/NA (total %) | 30,321 (75.9%) | 18,591 (53.2%) | 15,374 (44.8%) | 14,063 (42.1%) | 13,182 (40.8%) | 12,762 (41.1%) | 12,477 (39.5%) | 12,970 (40.3%) | 14,149 (41.4%) | 14,939 (41.5%) | |
| Occupation task level | Unskilled tasks (valid %) | 1802 (21.5%) | 2780 (19.3%) | 3388 (20.0%) | 3572 (20.3%) | 3763 (21.2%) | 3654 (21.2%) | 4031 (22.4%) | 4300 (23.6%) | 4678 (24.0%) | 5024 (24.1%) |
| Skilled tasks (valid %) | 6019 (71.7%) | 10,438 (72.5%) | 12,164 (71.8%) | 12,566 (71.5%) | 12,581 (70.8%) | 12,185 (70.6%) | 12,574 (69.9%) | 12,626 (69.3%) | 13,458 (69.1%) | 14,334 (68.8%) | |
| Complex tasks (valid %) | 384 (4.6%) | 810 (5.6%) | 949 (5.6%) | 1000 (5.7%) | 967 (5.4%) | 988 (5.7%) | 954 (5.3%) | 914 (5.0%) | 923 (4.7%) | 1006 (4.8%) | |
| Highly complex tasks (valid %) | 187 (2.2%) | 365 (2.5%) | 438 (2.6%) | 431 (2.5%) | 449 (2.5%) | 444 (2.6%) | 425 (2.4%) | 376 (2.1%) | 419 (2.2%) | 474 (2.3%) | |
| Unknown/NA (total %) | 31,556 (79.0%) | 20,565 (58.8%) | 17,356 (50.6%) | 15,797 (47.3%) | 14,558 (45.1%) | 13,779 (44.4%) | 13,587 (43.0%) | 13,929 (43.4%) | 14,672 (43.0%) | 15,186 (42.2%) | |
| Retirement diagnosisb | Somatic disease | 29,692 (74.3%) | 25,055 (71.7%) | 24,184 (70.5%) | 22,750 (68.2%) | 21,658 (67.0%) | 20,730 (66.8%) | 20,821 (65.9%) | 20,548 (63.9%) | 21,217 (62.1%) | 22,000 (61.1%) |
| Psychological disease | 10,256 (25.7%) | 9903 (28.3%) | 10,111 (29.5%) | 10,616 (31.8%) | 10,660 (33.0%) | 10,320 (33.2%) | 10,750 (34.1%) | 11,597 (36.1%) | 12,933 (37.9%) | 14,024 (38.9%) | |
| Total | 39,948 | 34,958 | 34,295 | 33,366 | 32,318 | 31,050 | 31,571 | 32,145 | 34,150 | 36,024 |
-
Source: FDZ-RV: DOI: 10.57701/SUF.RTZN.2001-2020.EM.1-0, own calculations. a“Low”: School-leaving qualification unknown, without school-leaving qualification, secondary/grammar school without vocational training; “Mid”: Secondary/gramma/high school diploma with vocational training, high school diploma without vocational training, school-leaving certificate unknown, but vocational training diploma; “High”: Master tradesman/technician, Bachelor, diploma/Magister/Master/state examination, doctorate. b “Psychological Disease” includes ICD-Codes F00-F99, “Somatic Disease” includes all other non-missing ICD-Codes.
Overview of important socio-demographic variables – continued.
| 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Male | 18,604 (52.2%) | 18,228 (51.6%) | 17,753 (50.8%) | 17,079 (50.6%) | 17,323 (50.3%) | 16,984 (49.4%) | 16,111 (49.1%) | 16,004 (48.1%) | 15,605 (48.7%) | 16,593 (47.6%) |
| Female | 17,007 (47.8%) | 17,086 (48.4%) | 17,160 (49.2%) | 16,680 (49.4%) | 17,142 (49.7%) | 17,427 (50.6%) | 16,694 (50.9%) | 17,236 (51.9%) | 16,410 (51.3%) | 18,238 (52.4%) | |
| Age | Mean age (SD) | 51.5 (8.41) | 51.8 (8.46) | 52.1 (8.42) | 52.3 (8.42) | 52.7 (8.26) | 52.8 (8.45) | 53.1 (8.37) | 53.4 (8.40) | 53.9 (8.45) | 54.5 (8.37) |
| Region | West | 27,770 (78.0%) | 27,621 (78.2%) | 27,252 (78.1%) | 26,404 (78.2%) | 27,165 (78.8%) | 27,097 (78.7%) | 25,824 (78.7%) | 26,132 (78.6%) | 25,239 (78.8%) | 27,691 (79.5%) |
| East | 7273 (20.4%) | 7160 (20.3%) | 7164 (20.5%) | 6808 (20.2%) | 6828 (19.8%) | 6844 (19.9%) | 6612 (20.2%) | 6795 (20.4%) | 6459 (20.2%) | 6839 (19.6%) | |
| Other countries | 562 (1.6%) | 525 (1.5%) | 490 (1.4%) | 543 (1.6%) | 468 (1.4%) | 462 (1.3%) | 369 (1.1%) | 313 (0.9%) | 317 (1.0%) | 301 (0.9%) | |
| Education indexa | Low (valid %) | 4789 (23.2%) | 4300 (21.2%) | 3846 (20.6%) | 3647 (20.3%) | 3846 (20.1%) | 3879 (20.2%) | 3769 (20.3%) | 3648 (18.9%) | 3372 (17.8%) | 3497 (16.5%) |
| Mid (valid %) | 15,117 (73.1%) | 14,998 (74.1%) | 13,909 (74.3%) | 13,443 (74.7%) | 14,231 (74.3%) | 14,217 (74.2%) | 13,788 (74.1%) | 14,412 (74.9%) | 14,406 (75.9%) | 16,266 (76.5%) | |
| High (valid %) | 763 (3.7%) | 939 (4.6%) | 957 (5.1%) | 897 (5.0%) | 1066 (5.6%) | 1072 (5.6%) | 1041 (5.6%) | 1193 (6.2%) | 1204 (6.3%) | 1492 (7.0%) | |
| Unknown/NA (total %) | 14,942 (42.0%) | 15,077 (42.7%) | 16,201 (46.4%) | 15,772 (46.7%) | 15,322 (44.5%) | 15,243 (44.3%) | 14,207 (43.3%) | 13,987 (42.1%) | 13,033 (40.7%) | 13,576 (39.0%) | |
| Occupation task level | Unskilled tasks (valid %) | 5105 (24.3%) | 5652 (24.8%) | 6735 (26.7%) | 7562 (29.1%) | 8274 (29.3%) | 8733 (30.3%) | 8104 (29.3%) | 8847 (30.4%) | 8276 (29.2%) | 8694 (27.5%) |
| Skilled tasks (valid %) | 14,389 (68.6%) | 15,141 (66.3%) | 15,854 (62.9%) | 15,678 (60.3%) | 16,609 (58.9%) | 16,771 (58.1%) | 16,164 (58.5%) | 16,753 (57.7%) | 16,444 (58.0%) | 18,537 (58.7%) | |
| Complex tasks (valid %) | 1020 (4.9%) | 1325 (5.8%) | 1700 (6.7%) | 1719 (6.6%) | 2056 (7.3%) | 2067 (7.2%) | 2126 (7.7%) | 2074 (7.1%) | 2207 (7.8%) | 2610 (8.3%) | |
| Highly complex tasks (valid %) | 472 (2.2%) | 704 (3.1%) | 930 (3.7%) | 1050 (4.0%) | 1255 (4.5%) | 1270 (4.4%) | 1259 (4.6%) | 1382 (4.8%) | 1442 (5.1%) | 1732 (5.5%) | |
| Unknown/NA (total %) | 14,625 (41.0%) | 12,492 (35.4%) | 9694 (27.7%) | 7750 (23.0%) | 6271 (18.2%) | 5570 (16.2%) | 5152 (15.7%) | 4184 (12.6%) | 3646 (11.4%) | 3258 (9.3%) | |
| Retirement diagnosisb | Somatic disease | 20,897 (58.7%) | 20,343 (57.6%) | 19,986 (57.2%) | 19,123 (56.6%) | 19,584 (56.8%) | 19,520 (56.7%) | 18,541 (56.5%) | 18,960 (57.0%) | 18,507 (57.8%) | 20,204 (58.0%) |
| Psychological disease | 14,714 (41.3%) | 14,971 (42.4%) | 14,927 (42.8%) | 14,636 (43.4%) | 14,881 (43.2%) | 14,891 (43.3%) | 14,264 (43.5%) | 14,280 (43.0%) | 13,508 (42.2%) | 14,627 (42.0%) | |
| Total | 35,611 | 35,314 | 34,913 | 33,759 | 34,465 | 34,411 | 32,805 | 33,240 | 32,015 | 34,831 |
-
Source: FDZ-RV: DOI: 10.57701/SUF.RTZN.2001-2020.EM.1-0, own calculations. a“Low”: School-leaving qualification unknown, without school-leaving qualification, secondary/grammar school without vocational training; “Mid”: Secondary/gramma/high school diploma with vocational training, high school diploma without vocational training, school-leaving certificate unknown, but vocational training diploma; “High”: Master tradesman/technician, Bachelor, diploma/Magister/Master/state examination, doctorate. b “Psychological Disease” includes ICD-Codes F00-F99, “Somatic Disease” includes all other non-missing ICD-Codes.
A second tendency, which can be observed, is the increasing mean age of pensioners. This tendency leads to a grown skewness of the age distribution. In 2020, individuals were on average 2.3 years older than in 2001 when being awarded a REC-pension for the first time (on-site version of RTZN data).
As far as regional distribution (east–west) and education are concerned, there is no tendency to be seen. On average 80% of REC-pensions are awarded in West-Germany representing the overall share in the general population.
Regarding education one can state that REC-pensions are mostly awarded to individuals with secondary-level education (low- and mid-level education). However, Tables 1 and 2 show that in this group the share of individuals without apprenticeship (low-level education) has declined whereas the share of those with mid-level education has risen between 2001 and 2020. In addition, the share of REC-pensioners with high education levels (tertiary education) has more than tripled (2.1% in 2001 compared to 7.0% in 2020). In line with this tendency, the share of pensioners whose last reported occupation included mainly unskilled and skilled tasks has declined whereas the share of those with complex and highly complex tasks has about doubled.
A first look at the retirement diagnoses shows that the share of somatic diseases as reason for awarding a REC-pension has been declining between 2001 and 2020 (−16% points) whereas the share of psychological diseases as reason for retirement has risen to 42% of awarded REC-pensions in 2020.
5 Application Fields, Selected Publications, and Links with Other Data
With a time span of 20 years, the dataset allows for analyses of several topics over time and across retirement law reforms that have been enacted in the covered period. The descriptive analyses in Section 4 show first examples of such topics which could be further investigated. In addition, this chapter shows examples of analyses that have been carried out with the existing scientific use files “Pension for Reduced Earnings Capacity and Diagnoses” of the years 2003–2014. The new dataset enables replications with more recent data as well as broadening analyses of these studies.
5.1 Geographical Analyses
In their two publications Hagen and Himmelreicher (2014a, 2014b) aim at understanding whether the socio-economic and -demographic characteristics differ between REC-pensioners with psychological and somatic diagnoses. They also focus on their regional distribution across Germany. They use the 96 spatial planning regions and calculate differentiated risks to fall into REC-pension for psychological and somatic diseases. Subsequently, they match the data with economic data for each spatial planning region and analyze whether the calculated risks are for example higher in economically deprived regions.
5.2 Match with Other FDZ-RV Data
Generally, matching the FDZ-RV-data across dataset-type and years is allowed if agreed upon in the data usage contract. So far, for scientific use files, this is only possible via statistical matching. Since 2011, the data which can be used on site as a guest researcher include an identifier that allows to match data over time and dataset-types. The FDZ-RV is currently working on a concept to offer this possibility for scientific use files, too.
In the face of the growing need of care staff in Germany, Harling et al. (2010)Harling et al. 2010 compare care staff with other occupation groups concerning rehabilitative benefits, awarded REC-pensions and the diagnoses leading to the awarding of a REC-pension. For their analyses they combine the data on pensions awarded with statistics on rehabilitation outcomes (SUF.RSD).
In order to analyze the link between occupation and the risk of falling into REC-pension due to disorders affecting the back and spine, Weske et al. (2008)Weske et al. 2008 match longitudinal career data (SUF.VVL) with data on REC-pensions using the 20 variables which are identical in both datasets.
Söhn and Mika (2015, 2019 use an on-site version of the longitudinal data of VVL in combination with the data on REC-pensions in order to explain the often very low pension levels in the case of reduced earnings capacity.
Müller et al. (2013)Müller et al. 2013 compare the data on REC-pensions with a sample of all individuals insured at the German Pension Insurance (SUF.AKVS) in order to analyze social inequalities in the access to REC-pensions.
6 Data Access and Documentation
The data can be used for research purposes, is provided for the software-packages Stata and SPSS and can be applied for free of charge via the FDZ-RV data-application form (“Nutzungsantrag”) at www.fdz-rv.de. On this website one can also find the codebook to the scientific use file. Both, dataset and codebook are available in German and English. For further information, interested researchers may contact the author of the article.
References
Brussig, M., Drescher, S. E., and Kalina, T. (2020). Aktivierende Erwerbsminderungsrente? Zur Rückkehr auf den Arbeitsmarkt nach Erwerbsminderung. Berliner J. Soziol. 29: 237–271, https://doi.org/10.1007/s11609-020-00400-y.Search in Google Scholar
Hagen, C. and Himmelreicher, R.K. (2014a). Erwerbsminderungsrente in Deutschland – Ein unterschätztes Risiko(?). Vierteljahrshefte Wirtschaftsforsch. 83: 115–138. https://doi.org/10.3790/vjh.83.2.115.Search in Google Scholar
Hagen, C. and Himmelreicher, R.K. (2014b). Starke Zunahme von Erwerbsminderungsrenten wegen psychischer Erkrankungen bei westdeutschen Frauen: analysen zu individuellen, sozialen und regionalen Unterschieden beim Zugang in Erwerbsminderungsrente in Deutschland. Informationsdienst Soziale Indikatoren 51: 6–11.Search in Google Scholar
Harling, M., Schablon, A., and Nienhaus, A. (2010). Abgeschlossene medizinische Rehabilitationen und Erwerbsminderungsrenten bei Pflegepersonal im Vergleich zu anderen Berufsgruppen. DRV-Schriften Band 55: 72–85.10.1055/s-0031-1283475Search in Google Scholar
Mika, T. (2017). Different employment pathways into the pension for reduced earnings capacity. Sozialer Fortschr. 66: 31–47.https://doi.org/10.3790/sfo.66.1.31.Search in Google Scholar
Müller, R., Hagen, C., and Himmelreicher, R. K. (2013). Risiken für eine Erwerbsminderungsrente - Bremen im Ländervergleich: Eine Analyse des Rentenzugangs in Erwerbsminderungsrente auf Basis von Daten der Deutschen Rentenversicherung. Arbeitnehmerkammer Bremen, Bremen.Search in Google Scholar
Söhn, J. and Mika, T. (2015). Die erwerbsbiografische Vorgeschichte der Frühverrentung wegen Erwerbsminderung. Z. Sozialreform 61: 461–492.10.1515/zsr-2015-0406Search in Google Scholar
Söhn, J. and Mika, T. (2019). Veränderung der Versicherungsbiografien im Rentenzugang 2003 bis 2014 im Vergleich. Dtsch. Rentenversicher. 74: 259–288.Search in Google Scholar
Weske, M., Thiede, M., and Mueller, U. (2008). Zusammenhänge zwischen beruflicher Tätigkeit und dem Risiko vorzeitiger Berentung wegen Dorsopathie-Stochastische Verknüpfung von Querschnitts- und Längsschnittdaten. DRV-Schriften Band 55: 109–120.Search in Google Scholar
Article Note
This article is part of the special issue “A New Decade of Research on the Economics of Climate Change: Towards an Integrated View on a Sustainable Use of the Biosphere” published in the Journal of Economics and Statistics. Access to further articles of this special issue can be obtained at www.degruyter.com/jbnst.
© 2022 the author(s), published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
