LIS is happy to announce the following data updates:

• Belgium (4 new datasets and all revised) – Further annualisation from BE18 to BE21 in the LIS Database
  Read more »
  Four new datasets from Belgium (BE18 to BE21) have been added to the LIS Database. The datasets are based on the respective waves 2019 to 2022 of the Survey on Income and Living Conditions (SILC) carried out by Statistics Belgium (StatBel).
  Earlier datasets of the Belgian data were revised for consistency with the later series; the revisions have no impact on the LIS Key Figures. Hourly wages (gross1) which had a substantial number of missing values were recovered by using the yearly wage. Furthermore, the variable pwgta is now provided in BE04 with the personal intergenerational cross-sectional weight to be used with the variables edmom_c and eddad_c that were asked only to a subsample of selected respondents, aged 26-66 years at the moment of the interview. The variable immigr_c is now provided in BE10 and filled with the country(ies) of birth of parents. The variables disabled and health_c are provided now for children as well in BE16.
• Brazil (6 new datasets and 1 revised) – Partial annualisation from BR17 to BR22 in the LIS Database
  Read more »
  LIS has released a first set of annual data for Brazil in the LIS Database. The six new datasets BR17 to BR22 are based on the National Continuous Household Sample Survey (PNADC) from the Brazilian Geographical and Statistical Institute (IBGE). Since income amounts were collected gross of taxes and contributions and taxes and social contributions were not collected nor simulated by the data provider, LIS added to the PNADC micro-simulated amounts computed by a team of experts from the University of Pernambuco (Prof. José Ricardo Bezerra Nogueira) and the Centro de Pesquisas Ageu Magalhães – CPqAM (Dr. Carlos Feitosa Luna). More information about this is available in Compare.It.
  The dataset BR16, the first available year from PNADC, has been re-harmonised with the slightly adjusted simulation techniques and newly available data version, taking into account adjusted weighting factors based on the newly available population census information.
• Israel (3 new datasets and 17 revised) – Addition of IL19, IL20 & IL21 to the LIS Database
  Read more »
  Three new datasets have been added to the LIS Database (IL19, IL20 & IL21 ). The datasets are based on the Household Expenditure Survey (HES) carried out by Central Bureau of Statistics and reworked by the National Insurance Institute of Israel.
  The previous datasets IL02 to IL18 have been slightly revised. Variable ptime1 (part-time employment (dummy), main job) is now available, approximated by using the overall working hours in all jobs, considering as part-time all persons who work 30 hours or less. The income section has been slightly adjusted, with a minor impact on the LIS Key Figures for some years.
• Italy (16 new datasets and all revised) – Addition of IT77 to IT84, IT02, IT06 & IT12 to the LIS Database;
  addition of IT98, IT02, IT06, IT12 & IT20 to the LWS Database
  Read more »
  LIS Database
  Various new data points from Italy have been added to the LIS Database. The series has been extended to include all of the datapoints currently available from the Survey of Household Income and Wealth (SHIW) carried out by the Bank of Italy, which implied the addition of annual data from IT77 to IT84, and the years IT02, IT06, and IT12. It should be noted that until IT12 the datasets refer to net income in hitotal (total current income). Conversely, from IT14 the availability of detailed variables for taxes and social contributions (as simulated by the Bank of Italy) allowed the addition of those amounts to all labour income and pension variables, so that the latter datasets are considered gross. Note that for the datasets IT04, IT108 and IT10 taxes and contributions were also simulated by the Bank of Italy, but only as total amounts on overall income; as a result, while income taxes (p/hxitax) and social contributions (p/hxscont) are provided, the income variables are all net of taxes and contributions so that hitotal is equal to dhi (as all other datasets until IT12).
  Note that the whole SHIW series was re-harmonised according to the latest variable list, in order to increase the coherence of the whole series. Thus, various consistency revisions have been carried out for the earlier available datasets IT86 to IT20, using the latest version of the historical database available at the Bank of Italy; this involved the filling of some variables which had been left empty in the older datasets (locsz_c, area_c, rural, farming, marital, weeks, net1), the consistency of the subset of the population for which the variables are filled (marital, ctrybrth, migrat_c, immigr, disabled, educ_c, lfs, weeks), the uniformisation of the categories across years (region_c, own, migrat_c, disabled, educ_c, lfs). In addition, a different recoding of the tertiary degrees has implied some changes in variable educlev.
  
  LWS Database
  Various new dataset from Italy have been added to the LWS Database. The series has been extended by the new data point IT20, and the remaining gap years in the LWS database IT02, IT06, and IT12 have been added. All datasets in the series are from the Survey of Household Income and Wealth (SHIW) carried out by the Bank of Italy. It should be noted that until IT12 included the datasets refer to net income in hitotal (total current income). See further remarks in the above `LIS Database’ section.
  The earlier available datasets in the series have been revised for consistency. Among other minor changes, the variable boef_c (country-specific expectations about household finances) has been added to IT10.

• Luxembourg (1 new dataset and 3 revised) – Addition of LU21 to the LWS Database
  Read more »
  LIS has added one more data point (LU21) to the Luxembourgish wealth data series in the LWS Database. The datasets are based on the Luxembourg Household Finance and Consumption Survey (HFCS), acquired from Banque Centrale du Luxembourg (BCL).
  LIS added to the earlier datasets in the series the number of businesses held by the household, which is available in the variable bus3_c in the datasets LU10, LU14 and LU18, and in bus2_c in LU21.
• Romania (3 new datasets and 12 revised) – Addition of RO14 , RO16 & RO20 to the LIS Database
  Read more »
  Three more datasets have been added to the recently announced annual Romanian data series. The three new datapoints RO14 , RO16 & RO20, extend the annual data series of RO06 to RO20. The data series is based on the Quality of Life Survey (ACAV) on which is based the Romanian Survey on Income and Living Conditions (SILC) and is provided by the Romanian National Institute of Statistics (INSSE). We are grateful to INSSE for making available information on the value of own consumption (LIS variable hi14) for these three datasets explicitly for LIS. Own consumption is part of the construction of LIS disposable income and it is an important source of in-kind income in Romania.
  New variables are provided for the series: in RO10 the variable pwgta (the additional person weight) which is the personal intergenerational cross-sectional weight to be used with variables edmom_c and eddad_c that were asked only to a subsample of selected respondents, aged 25-59 years at the moment of the interview; in RO19 hxloan (instalment for other loans) & public1 (public sector) are now additionally available.
• Switzerland (1 new dataset and 13 revised) -Addition of CH19 to the LIS Database
  Read more »
  One new dataset from Switzerland, CH19, has been added to the LIS Database. The dataset is based on Income and Living Conditions (SILC) data from the Swiss Federal Statistical Office.
  The datasets CH06 to CH18 have been revised for consistency in the variables ctrybrth, citizen, yrsresid, and ptime1. Variable hxmort (mortgage instalments) refers now to annual amounts (previously monthly amounts); this affects also the derived variable hhouscost (housing costs). Transfers from general social assistance have been now placed to LIS variable hi45 (general assistance), previously added directly at the higher aggregate hipublic (public transfer)
• United Kingdom (1 new dataset and 1 revised) – Addition of UK21 to the LIS Database
  Read more »
  One new data point for the United Kingdom (UK21) was added to the LIS Database. The dataset is based on the Family Resources Survey (FRS) from the Department for Work and Pensions (DWP) and the Office for National Statistics (ONS).
  A correction to the benefits section was applied to dataset UK20, the previously weeklyised amount for the ‘self-employment income support scheme (SEISS)’ (introduced in 2020 during the Coronavirus pandemic) has no longer been used; instead, the reported amounts are now treated as lump sum amounts. This change affects slightly the LIS Key Figures.

Design and Development

LIS and LISER invite you for a hybrid session of the (LIS)^2ER Friday seminar on In Search of a Paradox of Redistribution: An Analysis of Fiscal Redistribution in High-income Countries

In person room: MSH LIS-LISER corner 5th floor (Maison des Sciences Humaines).

Friday 03 November at 11h00 (UTC+01:00) Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna

The following presentation is scheduled:

Speaker: Dave Coady

Title: In Search of a Paradox of Redistribution: An Analysis of Fiscal Redistribution in High-income Countries

Abstract:
The last decade has seen a sharp increase in interest in the possible existence of a Paradox of Redistribution (PoR) whereby more narrow targeting of social transfers aimed at increasing their redistributive (poverty) impact has the perverse effect of increasing poverty over the medium term due to decreasing public support for such narrowly targeted spending. However, the empirical support for the existence of a PoR has been mixed. We revisit this issue using harmonized LIS household survey data covering recent decades by examining how the magnitude of fiscal redistribution (FR) from transfers varies across high-income countries and over time within these countries. Our analysis is embedded in the standard social welfare framework, which allows for a transparent and integrated evaluation of FR including by making explicit the value judgements necessarily inherent in such analyses. Our results support recent findings that FR has increased over the last four decades, although we do not find support for some recent results that FR decreased since 1995. While we find strong support for a PoR for social insurance transfers (dominated by pension transfers), we find little support in the context of social assistance transfers. We argue that, especially in the context of social assistance, more detailed country-specific studies of the political and economic dynamics in countries are needed to adequately determine the existence or otherwise of a PoR. Our high-level analysis can, however, help to identify possible candidates for such country case studies.

Participation
In person room: MSH LIS-LISER corner 5th floor (Maison des Sciences Humaines).

Please join us on Zoom meetings by following the link below
https://us02web.zoom.us/j/83022484736

Meeting ID: 830 2248 4736

Sebastian Will, (University of Freiburg, Germany)

The provision of affordable housing has been a matter of great concern in almost all middle- and high-income countries for at least the last two decades. For many households, the own home is the single biggest asset. Accordingly, homeownership plays a key role in shaping wealth inequalities. In this article, Sebastian Will first briefly discusses various housing regime indicators. Secondly, he calculates Gini coefficients in order to examine wealth inequalities with and without housing wealth and for subdivisions of renting and owning households.

Full article is available here.

Carmen Petrovici, (LIS)

Looking at the newly released long series of Romanian data in the LIS database Carmen Petrovici unravels what stands behind the remarkable increase in the gross wage income in the latest years. In 2018 Romania implemented a fiscal reform, which, among other measures, switched most of the social contributions paid by the employer over to the employees, and in order to absorb the new fiscal burden, the gross wages increased. However, this increase in gross wages is not followed by a similar increase in the net wages over the same period.

Full article is available here.

Gintare Mazeikaite, (LIS)

In June 2018, the Parliament of Lithuania adopted a six-reform package encompassing labour taxation, pensions, education, innovation, healthcare, and measures targeting the informal economy. One of the primary motivations for the tax reform was to increase the competitiveness of labour taxation. Gintare Mazeikaite describes some of the key elements of these reforms, and how they affect the disposable income of the household.

Full article is available here.

Josep Espasa Reig, (LIS)

The ’lissyrtools’ R package aims to make the use of R in LISSY simpler. It provides a set of commonly used functions that can easily reproduce LIS estimates such as those in DART, IKF and in the Compare.It dashboard. The package is currently in Beta (0.1) version. It is installed in LISSY and users can download it locally from the LIS GitHub repository.

Full article is available here.

LIS is happy to announce the following data updates:

• Georgia (2 new datasets) – Addition of GE20 & GE21 to the LIS Database
  Read more »
  Two new datasets from Georgia (GE20 and GE21) have been added to the LIS Database. The datasets are based on the respective waves of the Integrated Household Survey (IHS) carried out by the National Statistics Office of Georgia.
• Lithuania (2 new datasets and 10 revised) – Addition of LT19 & LT20 to the LIS Database
  Read more »
  LIS has added two more data points to the Lithuanian series in the LIS Database. The new datasets LT19 and LT20 are based on the Lithuanian Survey of Income and Living Conditions (SILC) carried out by Statistics Lithuania.
  Alongside with the harmonisation of the new datasets, a few consistency corrections were carried out. Country-specific codes in variable educ_c were modified for LT13 to LT18, with no impact on the standardised education variables. Imputed rent (hrenti) was corrected for the whole series. Improvements in the aggregation routines did cause minor updates in the variables relation and hhtype for the whole series.
• Mexico (2 new datasets and 6 revised) – Addition of MX20 & MX22 to the LIS Database
  Read more »
  Two new datasets from Mexico, MX20 and MX22, have been added to the LIS Database. The datasets are from the Household Income and Expenditure Survey (ENIGH – Nueva Serie) and are provided by the Mexican National Statistical Institute. A few consistency revisions have been carried out. For the datasets MX12 to MX18 variable hxloan (installment for other loans) has been corrected. Voluntary health contributions are now available in variable hxvcont from MX18 onwards.
• Norway (1 new dataset and 5 revised) – Addition of NO21 to the LIS and LWS Databases
  Read more »
  LIS has added one more data point (NO21) to the Norwegian data series to both the LIS and LWS Databases. The dataset is derived from the fully register-based Household Income and Wealth Statistics maintained by Statistics Norway (SSB). In addition, LIS received a new data version for NO20, which prompted two updates: 1) a modest change in the variable pipension, which affects all variables including these amounts, with a minor impact on the LIS Key Figures; and, for the LWS Database only, 2) a more accurate disaggregation of financial assets in its subcomponents hafc (deposit accounts and cash), hafi (financial investments, and its subcomponents) and hafo (other non-pension financial assets). The latter additional detail triggered a correction to the whole Norwegian LWS data series, as it clarified that the assets amounts that were previously included in the block of long-term saving (has) were mostly including financial assets, and they were thus moved to the block of financial assets (haf), hence implying an increase in disposable net worth (dnw). Finally, in NO16 minor revisions to the income blocks pensions and public social benefits were carried out.
• Romania (12 new datasets and 2 revised) – Partial annualisation from RO06 to RO19 in the LIS Database
  Read more »
  LIS is excited to announce the release of twelve new datasets from Romania, consisting in the (so far partial) annualization of RO06 to RO19 (all years except 2014 and 2016). The data series is based on the Quality of Life Survey (ACAV) on which is based the Romanian Survey on Income and Living Conditions (SILC) and is provided by the Romanian National Institute of Statistics (INSSE). Data for RO14, RO16, and RO20 will become available as soon as the information on own consumption will be received. This variable is part of the construction of LIS disposable income and it is an important source of in-kind income in Romania. In addition, the previous data points RO95 and RO97 have been revised for consistency, where possible, with no impact on the major income aggregates. Remaining consistency warnings are documented in Compare.It.
• Sweden (2 new datasets and 20 revised) – Addition of SE01 & SE21 to the LIS Database
  Read more »
  Two more datasets have been released for Sweden. Both datasets rely entirely on the Swedish income registers and are provided by Statistics Sweden: SE01 is based on the Household Income Survey (HINK/HEK); SE21 uses the Swedish Living Conditions Survey (ULF/SILC) sample and the Income and Tax Registers (IoT). Please note that as of this data release the previously available dataset SE00 has been also fully (re)harmonised with the latest information provided by Statistics Sweden. This revision causes a slight change in the LIS Key Figures. In addition, LIS has received new information on property taxes for the years SE08 to SE20 that imply substantial changes to variables hxptax (property taxes) and hhouscost (housing costs), but also a negligible change in p/hxitax (income taxes), and hence dhi and the LIS Key Figures, as the property taxes are removed from the total taxes paid.
• Taiwan (5 new datasets and 11 revised) – Addition of TW17 to TW21 to the LIS Database
  Read more »
  LIS has added five more data points to the Taiwanese data series in the LIS Database. The new datasets TW17 to TW21 are based on the Family Income Distribution and Expenditure Survey provided by the Taiwanese Directorate-General of Budget, Accounting and Statistics (DGBAS). The previous datasets TW81 to TW16 underwent some consistency revisions across various sections of data, among which the education, income and consumption blocks. Particularly, the consumption module has been (re)harmonised from scratch following the latest harmonisation decisions. LIS is happy to provide more details on the revisions in case needed.
• Uruguay (1 new dataset and 14 revised) – Addition of UY22 to the LIS Database
  Read more »
  One more dataset (UY22) has been added to the Uruguayan data series in the LIS Database. The dataset is based on the Continuous Household Survey (ECH) from the Uruguayan National Institute of Statistics (INE). LIS prioritised the release of this dataset, as it follows closely the structure of the UY19 data, whereas UY20 and UY21 essentially differed in the sampling and questionnaire routines due to the COVID-19 pandemic. Therefore, LIS postponed the work on those two datasets. The earlier datasets UY06 to UY19 have been revised for consistency, notably in-kind incomes were moved from hi521 (alimony and child support) and hi522 (remittances) to hi531 (in-kind transfers from private institutions) and hi532 (in-kind transfers from other households).

The US21 dataset, that was added to the LIS Database on the 14^th of June, has been corrected, as the Economic Impact Payment (or Coronavirus stimulus payment) has been added. This has a major impact on poverty and inequality indicators.

We strongly advise those who have carried out analyses on the US21 data to rerun their analyses.

Mauricio De Rosa, (Universidad de la República, Uruguay), Ignacio Flores, (City University of New York and Paris School of Economics), and Marc Morgan, (Geneva University)

Income inequality has regained attention in academia and politics, with rising trends observed globally over the past three decades. Latin America has been seen as an exception to this trend. This paper aims to reassess the prevailing narrative of declining inequality in Latin America by adopting an innovative approach. The authors build a comprehensive dataset that combines harmonized surveys, social security and tax data, and national accounts from ten Latin American countries. This approach allows them to reconcile micro and macro income data and address critical gaps, namely, in the coverage of top/capital incomes.

Full article is available here.