Skip to main content

Advertisement

SpringerLink
Log in

Timely Indicators for Inequality and Poverty Using the Italian Labour Force Survey

  • Original Research
  • Published:
Social Indicators Research Aims and scope Submit manuscript

Abstract

Income distribution measures obtained from standard household income surveys are usually published with a significant delay, of about 2 years. In this paper we propose a methodology for obtaining timelier indicators using the Italian Labour Force Survey (ILFS), a database which collects detailed information not only on individuals’ labour market status, but also on their households and wages. We develop a framework to estimate household labour income and we use it to construct distributional indicators that are available 1 year and half before the standard measures. After discussing the limitations of our approach, we show that the ILFS-based indicators closely track those calculated on standard household income surveys. The suggested indicators are not meant to substitute standard income measures but can offer up-to-date information, with the aim of better monitoring the distributional impact of changing macroeconomic conditions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Source: EU-SILC and SHIW, sample weights are used. Equivalised disposable income is equal to family disposable income divided by the OECD-modified equivalence scale. As for the EU-SILC, we consider only those years in which the SHIW is available

Fig. 2

Source: SHIW, sample weights are used. Equivalised disposable income is equal to family disposable income divided by the OECD-modified equivalence scale; equivalised labour income is analogously defined with respect to family labour income. We consider both all families, without any restriction, and those families in which there are not retirees and whose Reference Person is between 15 and 64 years old

Fig. 3

Source: SHIW, sample weights are used. Equivalised disposable income is equal to family disposable income divided by the OECD-modified equivalence scale; equivalised labour income is analogously defined with respect to family labour income. Indices are normalized with respect to the 2008 value. In the left panel we consider all families, without any sample restrictions. In the right panel we focus on those families in which there are not retirees and whose Reference Person is between 15 and 64 years old

Fig. 4

Source: SHIW, sample weights are used. Equivalised labour income is equal to family labour income divided by the OECD-modified equivalence scale. “month” refers to monthly equivalised labour income, obtained as the ratio between yearly labour income and the number of months worked, then aggregated at family level; “year” refers to yearly equivalised labour income; “delta month” is the difference between the Gini index on monthly equivalised labour income computed in a given year (i.e. 2010) and that in the two-year before (i.e. 2008); “delta year” is analogously defined. We consider households with no retirees and in which the reference person is 15–64 years

Fig. 5

Source: ILFS and SHIW, sample weights are used. The ILFS wage is the net regular salary earned one month before (no 13th or 14th month’s salary), excluding those extra-payments that are not commonly included in the monthly pay. The salary is reported in bin of 10 euros for amounts between 250 and 3000 euros; for lower and higher levels, there are two categories, respectively: 250- euros and 3000 + euros. The SHIW monthly wage is obtained as the ratio between annual earnings and months worked in the reference year (for employees only). Extremes values for the SHIW variable are winsorized at level 1 and 99% levels for each year

Fig. 6

Source: ILFS and SHIW, sample weights are used. The ILFS self-employed income has been imputed as described in Sect. 3, by estimating a Mincerian equation of hourly wage on observable individual and family characteristics. The SHIW self-employed income is obtained as the ratio between annual income and months worked in the reference year (for self-employed only). Extremes values are winsorized at level 1 and 99% levels for each year

Fig. 7

Source: ILFS and SHIW, sample weights are used. The ILFS monthly equivalised labour income is obtained by aggregating labour incomes at household level, divided by the OECD-modified equivalence scale to take into account economies of scale within the household. The SHIW monthly equivalised labour income is analogously defined. Extremes values are winsorized at level 1 and 99% levels for each year

Fig. 8

Source: ILFS and SHIW, sample weights are used. Gini index computed on equivalised labour incomes. We consider only those years in which both ILFS and SHIW are available. “delta SHIW”(right axis) is the difference between the Gini index computed in a given year (i.e. 2010) and that in the two-year before (i.e. 2008); “delta ILFS” is analogously defined. “delta ILFS” in 2010 is the difference between the Gini index in 2010 and that in 2009 since the value for 2008 is not available. Gini index computed in the ILFS in 2008 refers to 2009 incomes. We consider households with no retirees and in which the reference person is 15–64 years

Fig. 9

Source: ILFS and SHIW, sample weights are used. Labour Income Poverty Rate (LIPR): share of individuals with equivalised monthly labour income lower than the 60% of the national median value. We consider only those years in which both ILFA and SHIW are available. “delta SHIW” (right axis) is the difference between the LIPR computed in a given year (i.e. 2010) and that in the two-year before (i.e. 2008); “delta ILFS” is analogously defined. “delta ILFS” in 2010 is the difference between the corresponding LIPR in 2010 and that in 2009 since the value for 2008 is not available. LIPR computed in the ILFS in 2008 refers to 2009 incomes. We consider households with no retirees and in which the reference person is 15–64 years

Fig. 10

Source: ILFS and SHIW, sample weights are used. Gini index computed on equivalised labour incomes. Monthly wage in the ILFS is corrected for censoring: we assume that employees’ monthly wages censored at 250 euros are distributed according to a uniform distribution [0;250] and assign them the mean value, i.e. 125 rather than 250. For values censored at 3000 euros, we assume that monthly wages are distributed according to a Pareto distribution. “delta SHIW” (right axis) is the difference between the Gini index computed in a given year (i.e. 2010) and that in the two-year before (i.e. 2008); “delta ILFS” and “delta ILFS corrected” are analogously defined. “delta ILFS” and “delta ILFS corrected” in 2010 are the difference between the respective Gini index in 2010 and that in 2009 since the value for 2008 is not available. Gini index computed in the ILFS in 2008 refers to 2009 incomes. We consider households with no retirees and in which the reference person is 15–64 years

Fig. 11

Source: ILFS and SHIW, sample weights are used. Gini index computed on equivalised monthly labour incomes. Monthly wage in the SHIW is censored censored from below (at 250) and from above (at 3000), analogously as in the ILFS. “delta SHIW” (right axis) is the difference between the Gini index computed in a given year (i.e. 2010) and that in the two-year before (i.e. 2008); “delta ILFS” and “delta SHIW censored” are analogously defined. “delta ILFS” in 2010 is the difference between the respective Gini index in 2010 and that in 2009 since the value for 2008 is not available. Gini index computed in the ILFS in 2008 refers to 2009 incomes. We consider households with no retirees and in which the reference person is 15–64 years

Fig. 12

Source: ILFS and SHIW, sample weights are used. Eurostat provides indicators based on EU-SILC and SILC/EUROMOD. As for ILFS and SHIW indicators, we consider households with no retirees and in which the reference person is 15–64 years. Information on wages in the ILFS are available only from 2009; we consider in 2008 the value observed in 2009. The SHIW survey is run every two years and it is available with one year of delay with respect to the survey period. EU-SILC and SILC/EUROMOD refer to income data of 1 year before the survey year

Fig. 13

Source: ILFS, sample weights are used. We consider households with no retirees and in which the reference person is 15–64 years. The “labour income effect” is the change in the Gini index driven by the change in the distribution of equivalised labour income across people having it. The “family-employment effect” expresses how much inequality is associated to changes in the share of individuals living in families with at least one employed member. The decomposition of the change over time in the Gini index on equivalised labour income (including zero values) is based on the formula: ∆G ≅ e∆G_e + (G_e-1)∆e (Atkinson and Brandolini (2006); see also "Appendix A")

Fig. 14

Source: ILFS and HBS, sample weights are used. Adults 18–59 and children 0–17 are the share of people living in jobless households, distinguishing by age. Labour Income Poverty Rate (ILFS) is the share of individuals with equivalised monthly labour income lower than the 60% of the national median value computed in the ILFS. Incidence of relative poverty (HBS) is the share of individuals whose consumption expenditure is lower than the relative poverty line, measured in the HBS. Incidence of absolute poverty rate (HBS) is the share of individuals whose consumption expenditure is lower than the absolute poverty line, measured in the HBS

Similar content being viewed by others

Notes

  1. Italy followed a similar pattern: while overall income inequality did not change much, individuals along the entire income distribution became poorer (Brandolini 2014).

  2. Nowcasting is the prediction of the present, the very near future and the very recent past. Nowcasting techniques rely on timely information in order to nowcast key economic variables, such as e.g. GDP, that are typically collected at low frequency and published with long delays (BańBura et al. 2013).

  3. The survey is harmonized across European countries but run by national institutes of statistics, thus the date of the release differs across countries.

  4. Moreover, the trends in Gini index on equivalised labour income are similar to those observed for the Gini index on equivalised disposable income.

  5. Works that used the SHIW for the analysis of income inequality are Brandolini et al. (2002) and Jappelli and Pistaferri (2010) among the others; for a comparison across the different datasets see Galbraith et al. (2016).

  6. Equivalised disposable income is equal to family income divided by an equivalence scale (that can be equal to the square root of the household size or the OECD-modified equivalence scale); this normalization allows taking into account the existence of economies of scale within the household.

  7. Poverty indicators similarly to the ones provided by Istat are not available at international level.

  8. The share of individuals at risk of poverty or social exclusion is the leading indicator within the Europe 2020 framework and used for comparisons among European countries. See "Appendix A" for definitions.

  9. The absolute poverty rate increased from 2.1% in 2007 to 6.1 in 2018 among households whose head of the family is employed, if the reference person is unemployed, the incidence of absolute poverty rose from 7.0 to 27.6 over the same considered period. This evidence suggests that being employed seems to no longer ensure against the risk of poverty. However, some caveats are important. First, headcount employment provides only a partial picture of labour market developments (Brandolini and Viviano 2015): being simply the proportion of working-age people who have been working for at least 1 h in the reference week (ILO definition), it does not consider working times and contract duration, as well as important determinants of earnings. Second, employment is measured at individual level, while welfare at family level: labour supply interactions within the household affects family welfare and are not detected by the simple employment rate.

  10. Levels of the Gini index on equivalised labour and disposable income are largely similar when focusing on “younger” families, those where there are not retirees and the Reference Person is 15–64 years old. These families are less likely to rely on pension income. These represent 60% of Italian families, involving around 70% of the population. Levels of the Gini index differ when looking at all families; the Gini index on equivalised labour income is higher that the Gini index on equivalised disposable income since families relying on other incomes than labour income are counted as having zero income. However, trends are largely similar (Fig. 3).

  11. Labour income is referred to all the population, thus it takes into account also those who are not employed (labour income is equal to zero).

  12. Monthly labour income in the SHIW is obtained by the ratio between yearly labour income and months worked over the year.

  13. The SHIW refers to the yearly economic conditions of households and the reported employment condition is the prevalent employment status over the year. The ILFS reports current wage, which is the last regular monthly wage at the time of the interview.

  14. However, in years 2008–2016, the shares of employees who reported monthly wages below 250 and above 3000 are, respectively, on average 1.2 and 1.8% by year (from the SHIW; monthly wage is obtained as the ratio between annual earnings and months worked in 1 year).

  15. This is obtained as the ratio between monthly wage and weekly working hours times the average number of working weeks in one month, 4.3.

  16. We preferred this correction with respect to one based on the variance of self-employed income estimated in the SHIW; however, results look similar and are available upon request.

  17. These developments would reflect the fact that self-employed income is more cyclical than wages and that in Italy self-employment is on a decreasing trend (Bovini and Viviano 2018); thus, it is possible that some negative not-observed selection into self-employment would play a role in explaining this evidence.

  18. For those years in which SHIW is not available (odd years) we take the average gap among the closest available years.

  19. The OECD-modified equivalence scale is a factor computed on the basis of the size of the family and on the age of its members. It represents the needs of the family allowing for the presence of economies of scale in consumption. The OECD-modified scale assigns a value of 1 to the household head, 0.5 to each additional adult member and 0.3 to each child.

  20. For comparability, we normalize the indicators such that the value in 2008 is set equal to 1.

  21. Their decomposition look at the Gini index on individual earnings. The first component is the share of non-employed individuals; the second component is the dispersion of labour income among employed individuals. In this paper, we discuss the analogue of this decomposition looking at the Gini index on equivalised labour income.

  22. It is useful to remind that the analysis focuses only on those households who are more likely to rely on labour income—those with no retirees and whose reference person is 15–64 years old.

  23. This definition of jobless household is slightly different from the one used in Sect. 5.1; however, observed patterns are broadly the same.

  24. For a detailed discussion about jobless households see Mocetti et al. (2011).

References

  • Atkinson, T., & Brandolini, A. (2006). From earnings dispersion to income inequality. In F. Farina & E. Savaglio (Eds.), Inequality and economic integration (pp. 35–62)., Routledge Siena studies in political economy London: Routledge.

    Google Scholar 

  • BańBura, M., Giannone, D., Modugno, M., & Reichlin, L. (2013). Now-casting and the real-time data flow. In Handbook of economic forecasting (Vol. 2, pp. 195–237). Elsevier.

  • Böheim, R., & Jenkins, S. P. (2006). A comparison of current and annual measures of income in the British household panel survey. Journal of Official Statistics,22(4), 733.

    Google Scholar 

  • Bovini, G., and Viviano, E. (2018). The Italian “employment-rich” recovery: a closer look, Bank of Italy’s Occasional papers no. 461.

  • Brandolini, A. (2000), ‘The problematic measurement of income from self-employment’: a comment.In Seminar on household income statistics organized by Eurostat, Luxembourg, 13–14 Dec 1999.

  • Brandolini, A., (2014), Il Grande Freddo. I bilanci delle famiglie italiane dopo la Grande Recessione, In C. Fusaro e A. Kreppel (a cura di), Politica in Italia. I fatti dell’anno e le interpretazioni. Edizione 2014, Bologna, Il Mulino.

  • Brandolini, A., Cipollone, P., Sestito, P. (2002). Earnings dispersion, low pay and household poverty in Italy, 1977–1998 (pp. 225-264). In Cohen, D., Piketty, T. and G. Saint-Paul (ed.) The Economics of Rising Inequalities. Oxford university press, Oxford

  • Brandolini, A., Gambacorta R., Rosolia, A. (2018). Inequality amid income stagnation: Italy over the last quarter of a century, Bank of Italy’s Occasional Papers no. 442.

  • Brandolini, A., & Viviano, E. (2015). Behind and beyond the (head count) employment rate. Journal of the Royal Statistical Society: Series A (Statistics in Society),179(3), 657–681.

    Article  Google Scholar 

  • Ciani, E., R. Torrini (2019). The geography of Italian income inequality: recent trends and the role of employment, Bank of Italy Occasional Paper no. 492.

  • European Parliament (2016). Poverty in the European Union: The crisis and its aftermath. EPRS European Parliamentary Research Service Author: Marie Lecerf Members’ Research Service March 2016 – PE 579.099.

  • Fontaine, M., Fourcot, J. (2015). Nowcasting of poverty rate by microsimulation. Documents de Travail (INSEE), No F1506.

  • Galbraith, J. K., Choi, J., Halbach, B., Malinowska, A., & Zhang, W. (2016). A Comparison of major world inequality data sets: LIS, OECD, EU-SILC, WDI, and EHII. In L. Cappellari, S. W. Polachek, & K. Tatsiramos (Eds.), Income inequality around the world (research in labor economics, volume 44) (pp. 1–48). Bradford: Emerald Group Publishing Limited.

    Google Scholar 

  • Gasior, K., Rastrigina, O. (2017). Nowcasting: timely indicators for monitoring risk of poverty in 2014–2016 (No. em7/17). EUROMOD at the Institute for Social and Economic Research.

  • Istat (2014). I Nuovi Conti Nazionali in SEC 2010, Nota Informativa.

  • Jappelli, T., & Pistaferri, L. (2010). Does consumption inequality track income inequality in Italy? Review of Economic Dynamics,13(1), 133–153.

    Article  Google Scholar 

  • Jenkins, S. P. (2009). Distributionally-sensitive inequality indices and the GB2 income distribution. Review of Income and Wealth,55(2), 392–398.

    Article  Google Scholar 

  • Jenkins, S. P., Brandolini, A., Micklewright, J., & Nolan, B. (Eds.). (2012). The great recession and the distribution of household income. Oxford: Oxford University Press.

    Google Scholar 

  • Mocetti, S., Olivieri E., E. Viviano (2011). Italian households and labour market: structural characteristics and effects of the crisis, Stato e mercato, Società editrice il Mulino, Issue 2, pp 223–243.

  • Navicke, J., Rastrigina, O., & Sutherland, H. (2014). Nowcasting indicators of poverty risk in the European Union: a microsimulation approach. Social Indicators Research,119(1), 101–119.

    Article  Google Scholar 

  • Raitano, M. (2016). Income inequality in Europe since the crisis. Intereconomics,51(2), 67–72.

    Article  Google Scholar 

  • Sestito P. (2016). Audizione preliminare sulla delega recante norme relative al contrasto della povertà, al riordino delle prestazioni e al sistema degli interventi e dei servizi sociali (collegato alla legge di stabilità 2016), Camera dei deputati, Roma, 4 aprile 2016.

  • Shorrocks, A. (1982). Inequality decomposition by factor components. Econometrica,50(1), 193–211.

    Article  Google Scholar 

  • Stoyanova, S., Tonkin, R. (2016). Nowcasting household income in the UK: financial year ending 2015.

  • Vacas-Soriano, C., Fernández-Macías, E. (2018). Income inequality in the great recession from an EU-wide perspective. In CESifo Forum (Vol. 19, No. 2, pp. 9–18). München: ifo Institut–Leibniz-Institut für Wirtschaftsforschung an der Universität München.

Download references

Author information

Authors and Affiliations

  1. DG Economics, Statistics and Research, Bank of Italy, Rome, Italy

    Francesca Carta

  2. Dondena, Milan, Italy

    Francesca Carta

Authors
  1. Francesca Carta
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Francesca Carta.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A: Definitions

1.1 Indicators

Absolute poverty (provided by Istat based on the HBS): A household is in absolute poverty if its consumption expenditure is lower or equal to the monetary value of a basket of goods and services considered as essential to avoid severe forms of social exclusion. The monetary value of the basket of absolute poverty is reviewed every year in the light of trend in prices; it differs across household’s composition, age structure, macro-area and place of residence.

Gini index (provided by Istat based on the EU-SILC and by the Bank of Italy based on the SHIW): It measures the extent to which the distribution of income among individuals or households within an economy deviates from a perfectly equal distribution. The Gini index measures the area between the Lorenz curve and the hypothetical line of absolute equality, expressed as a percentage of the maximum area under the line. A Gini index of zero represents perfect equality and 100, perfect inequality.

Income quintile share ratio (also called the “S80/S20 ratio” based on EU-SILC): It is calculated as the ratio of the total income received by the 20% of the population with the highest income (= 1st or top quintile) to that income received by the 20% of the population with the lowest (= 5th or bottom quintile).

Decomposition of the Gini index by subgroups is not an easy task. However, when the income distribution of the groups do not overlap and are distinct, the decomposition by group is intuitive (the rank correlation of the income distribution is zero; P. Lambert e J. Aronson, Inequality Decomposition Analysis and the Gini Coefficient Revisited, Economic Journal, 103, issue 420, 1993). This condition is met when we split the population between those who live in households without labour income and individuals in families with positive labour earnings.

Share of adults (18–59 y.o.) or children (0–17 y.o.) living in jobless households (provided by Eurostat based on the European LFS): A household is jobless if no working age adult is employed (see "Appendix A" for a more detailed definition). The reference population is made of those families in which there is at least one working age adult; for example, families composed by only retirees are not considered. A working age adult has to meet the following conditions: (1) she is between 18 and 59 years old; (2) she is not a full-time student with less than 25 years old living with parents. Eurostat provides two statistics: the proportion of children (0–17 y.o.) and that of individuals (18–59 y.o.) living in jobless households.

Relative poverty (provided by Istat based on the HBS): A household is in relative poverty if its consumption expenditure is lower or equal a poverty line. The poverty line is set such that it defines as poor a household of two components with a consumption expenditure level lower or equal to the mean per-capita consumption expenditure. To define the relative poverty line for different household sizes an equivalence scale is used (Carbonaro equivalence scale) to take into account different needs and economies/diseconomies of scale that can be achieved in bigger/smaller households.

Risk of poverty or social exclusion (provided by Eurostat based on the EU-SILC): People at risk of poverty or social exclusion were in at least one of the following situations:

  • at risk of poverty after social transfers (income poverty), or

  • severely materially deprived; or

  • living in households with very low work intensity.

Income poverty Equivalised disposable income (after social transfer) below the at-risk-of-poverty threshold, which is set at 60% of the national median equivalised disposable income after social transfers.

Severely materially deprivation: Inability to pay for at least four of the following items (1) to pay their rent, mortgage or utility bills; (2) to keep their home adequately warm; (3) to face unexpected expenses; (4) to eat meat or proteins regularly; (5) to go on holiday; (6) a television set; (7) a washing machine; (8) a car; (9) a telephone.

Household with very low work intensity Household where working age (18–59 y.o.) individuals worked less than 20% of their total potential during the previous 12 months. The work intensity of a household is the ratio of the total number of months that all working-age household members have worked during the income reference year and the total number of months the same household members theoretically could have worked in the same period. Students in the age group between 18 and 24 years are not considered at working age. Households composed only of children, of students aged less than 25 and/or people aged 60 or more are completely excluded from the indicator calculation.

Appendix B: Data Definitions

Monthly wage (definition from the questionnaire, variable RETRIC): Net salary earned last month with the exception of other monthly payments (13th or 14th month’s salary, etc.) and those extra-payments not regularly included in the monthly pay (productivity premium, extraordinary overtime, overdue, etc.).

Italian wording: Retribuzione netta del mese scorso escluse altre mensilità (tredicesima, quattordicesima, ecc.) e voci accessorie non percepite regolarmente tutti i mesi (premi di produttività annuali, arretrati, indennità per missioni, straordinari non abituali, ecc.)

figure a

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Carta, F. Timely Indicators for Inequality and Poverty Using the Italian Labour Force Survey. Soc Indic Res 149, 41–65 (2020). https://doi.org/10.1007/s11205-019-02238-1

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11205-019-02238-1

Keywords

JEL Classification

Search

Navigation