Abstract
The purpose of this paper is to investigate whether civilian disability pensions have been used as an antipoverty measure in Italy. We applied a two-step analysis to Italian provincial data for the years 2003–2005. We implemented a classic panel analysis, followed by a two-step GMM (Generalised Method of Moments) analysis in which we introduced the spatial variable. The analysis shows that the number of civilian disability pensions is not related to disabling disease, but it is related to the unemployment rate in some areas and to the rate of poverty everywhere. These results robustly hold when the spatially lagged dependent variable is introduced among the regressors. The spatial variable allows us to take into account the local dimension and the possible social, historical, and cultural links among provinces that go beyond administrative boundaries. In discussing the results, we stress that the figures reflect the number of civilian disability pensions granted, not those requested. Moreover, the national legislation on the attribution of civilian disability pensions is administered locally; therefore, its application may reflect degrees of discretionary interpretation. We propose that there is room to interpret the use of civilian disability pensions as an antipoverty policy instrument in areas characterised by economic difficulties. However, we suggest that civilian disability pensions are particularly unsuited to play the role of an assistance policy instrument; once granted, they are seldom withdrawn despite possible changes in the financial situation of the recipient.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
This choice excludes from our data income support transfers to the poor aged over 65 with insufficient incomes, which are included under the category of civilian disability pensions.
- 3.
That is, estimates that do not take into account spatial dependence.
- 4.
Prevalence measures the proportion of “events” in a given population at a given time. In other words, prevalence measures the proportion of individuals in a given population who suffer from a specific illness. In our case, the concept of disability replaces that of illness.
- 5.
We only show maps for 2005 as those of previous years do not add any relevant information due to spatial persistence. The number in parenthesis next to each range indicates the number of provinces for which the prevalence rate falls within that range.
- 6.
We define as discharge rate the percentage of individuals discharged from a hospital after a period of hospitalisation. The discharge rate refers to the last contact with the institution where the individual has been hospitalised.
- 7.
Previous empirical works on aggregate data have never considered variables related to the state of health as precise as ours. For instance, Nolan and Fitzroy (2003) use the number of visits to the hospital and the mortality rate as health indicators; Autor and Duggan (2003) also use the mortality rate, as do many other authors; and Stapleton et al. (1998) use a measurement of the incidence of AIDS. Note that the mortality rate is criticised in the literature when used as a regressor to explain the incidence of disability (McVicar and Anyadike-Danes 2007).
- 8.
On the contrary, we talk of a uniform spatial structure.
- 9.
The Moran I Index is similar to the correlation coefficient: it varies between zero and one, −1 and +1. When I equals zero, there is no spatial autocorrelation; when I is close to −1 or to +1 there is high negative or positive spatial autocorrelation, respectively.
The I index has the following main characteristics when compared with the coefficient of correlation:
-
1.
It takes one, and not two, variable into account.
-
2.
It incorporates the weights (wij) that index the relative areas. In our case, these are organised in a matrix of inverse distances expressed in km. This choice of weights allows us to formalise the hypothesis that the relationships among individual areas tend to decrease in strength as distances between these areas increase.
-
3.
It is appropriate to think of it as "the correlation between neighbouring values on a variable” (O'Sullivan and Unwin 2003).
-
1.
- 10.
The statistics to measure the degree of spatial autocorrelation at local level (LISA) allow us to identify the contribution of each province to the overall autocorrelation (Moran I Index); therefore, we can investigate the variation in spatial autocorrelation across the whole area. By focusing on each individual province, these techniques can be used to identify spatial clusters.
- 11.
If work disability is linked to socio-economic conditions (e.g., poverty rate) and if health disability is only partially linked to work disability, then health disability is, as a consequence, related to socio-economic conditions (the poverty rate) with a circular link.
- 12.
By estimating a model ignoring spatial clustering, an inefficient estimate of the parameters is obtained. The standard errors are underestimated, and a type I error is made. Luckily, local or global measures of spatial autocorrelation are estimated to decide whether the data show spatial dependence, and here, the Moran I index indicates a clear process of positive spatial autocorrelation to be considered in further estimates.
- 13.
The literature suggests using spatial lags of the regressors as instruments (Anselin 1988).
- 14.
Notes on the tests in Table 8.4: the Shea partial R2 appears to be rather high, while the F-test of joint nullity of the instruments rejects the null hypothesis; this allows us to conclude that the instruments are valid and relevant. The rank tests, Kleibergen-Paap rk LM statistic and Kleibergen-Paap rk Wald statistic, reject the null hypothesis and allow us to conclude that the model is identified. The Anderson-Rubin Wald (F), Anderson-Rubin Wald (Chi-sq) and Stock-Wright LM S statistic (Chi-sq) tests reject the null hypothesis and reassure us of the robustness of the instruments that we used. In addition, we observe that the Kleibergan-Paap rk Wald F-statistic is much bigger than the value of the Stock-Yogo, shown four rows later; thus, we conclude that the instruments are strong. As a general conclusion, the estimated model is identified and contains valid, relevant and strong instruments. Finally, to test whether the instruments are uncorrelated with the residuals, we calculate the Hansen (J-test) of overidentification and orthogonality, as well as the C statistic of exogeneity of the instruments. As all of the tests do not reject the null hypothesis, the instruments can be considered exogenous (Baum et al. 2002).
- 15.
The calculation of the Variance Inflation Factor (VIF) allows us to control for multicollinearity and to exclude it, as for each variable, VIF <10; the highest VIF, found for the variable “genital urinary apparatus”, is 5.21, which is well below the threshold. The average VIF is 2.99. The tolerance associated with each variable, 1/VIF, is >0.1, which allows us to exclude multicollinearity (a tolerance value <0.1 is comparable with a VIF equal to 10). Finally, the conditional number, which is generally used to assess the global instability of the regression coefficients, suggests stability, as it is 7.7270 where 10 is the threshold value.
References
Anselin L (1988) Spatial econometrics: methods and models. Kluwer Academic Publishers, Boston
Anselin L, Bera A (1998) Spatial dependence in linear regression models with an introduction to spatial econometrics. In: Ullah A, Giles DEA (eds) Handbook of applied economic statistics. Marcel Dekker, New York
Arbia G (1988) Spatial data configuration in statistical analysis of regional economics and related problems, advanced statistical theory and applied econometrics. Kluwer Academic Publisher, Boston
Arbia G (2005) Spatial econometrics, statistical foundations and applications to regional convergence. Springer Verlag, Ney York
Arbia G, Basile R, Mirella S (2002) Regional convergence in Italy 1951–1999: a spatial econometric perspective. ISAE Working Papers 29, ISAE–Institute for Studies and Economic Analyses, Rome, Italy
Autor DH, Duggan MG (2003) The rise in the disability rolls and the decline in unemployment. Q J Econ 118:157–206
Baldacci E, De Santis G (2003) Disability pensions in Italy: the lawand the numbers. In: Prinz C (ed) European disability pension policies. European Centre, Vienna, pp 225–252
Baldacci E, Milan G (1998) Gli effetti di redistribuzione territoriale della spesa pensionistica di invalidità. In: Rossi N (ed) Il lavoro e la sovranità sociale, 1996–1997. Il Mulino, Bologna
Baltagi BH (2005) Econometric analysis of panel data. Wiley, Ney York
Baum CF, Schaffer ME, Stillman S (2002) Instrumental variables and GMM: estimation and testing. Boston College Economics Working Paper 545
Beltrametti L (1996) Le pensioni di invalidità: effetti redistributivi tra le regioni italiane (1951–1993) in Politica Economica n. 3, pp 391–403
Cardoso FH, Faletto E (1979) Dependency and development in Latin America. University of California Press, Berkeley, CA
Castellino O (1976) Il labirinto delle pensioni. Il Mulino, Bologna
Cracolici MF, Cuffaro M, Nijkamp P (2007) Geographical distribution of unemployment: an analysis of provincial differences in Italy. Growth Change 38(4):649–670 Gatton College of Business and Economics, University of Kentucky
Currie J, Madrian BC (1999) Health, health insurance and the labor market. In: Ashenfelter RO, Card D (eds) Handbook of labor economics, vol 3C. Elsevier, Amsterdam, pp 3309–3416
Disney R, Webb S (1991) Why are there so many long-term sick in Britain? Econ J 101:252–262
Huckfeldt R (1986) Politics in context: assimilation and conflict in urban neighborhoods. Agathon Press, New York
McVicar D (2006) Why do disability benefit rolls vary between regions? A review of the evidence from the USA and the UK, regional studies. Taylor Francis J 40(5):519–533 July
McVicar D, Anyadike-Danes M (2007) Panel estimates of the determinants of British regional male incapacity benefits rolls 1998–2006. Applied Economics (fort)
Ministero del Lavoro e delle Politiche Sociali (2006) Rapporto di Monitoraggio sulle politiche sociali, I trasferimenti monetari per invalidità, pensioni sociali, integrazioni al minimo, pensioni di guerra e ai superstiti, Direzione generale per la gestione del Fondo Nazionale per le politiche sociali e monitoraggio della spesa sociale
Ministero del Lavoro e delle Politiche Sociali (2008) Rapporto di Monitoraggio sulle politiche sociali, I trasferimenti monetari per invalidità, pensioni sociali, integrazioni al minimo, pensioni di guerra e ai superstiti, Direzione generale per la gestione del Fondo Nazionale per le politiche sociali e monitoraggio della spesa sociale
Nolan M, Fitzroy F (2003) Inactivity, sickness and unemployment in Great Britain: early analysis at the level of local authorities. Mimeo, University of Hull
O’Sullivan D, Unwin DJ (2003) Geographic information analysis. Wiley, Hoboken, NJ
Ritchie J, Ward K, Duldig W (1993) GPs and IVB. Research Report No. 18, Department of Social Security/HMSO, London
Silva HB, Disney R, Martin SJ (2010) Disability, capacity for work and the business cycle: an international perspective. Econ Policy 25:483–536
Smith J (1998) Socioeconomic status and health. Am Econ Rev 88:192–196
Stapleton D, Coleman K, Dietrich K, Livermore G (1998) Empirical analysis of ID and SSI application and award growth. In: Rupp K, Stapleton D (eds) Growth in civilian disability pensions. W.E. Upjohn Institute for Employment Research, Kalamazoo, MI, pp 31–92
Vasquez JA (1995) Why do neighbours fight? Proximity, interaction, or territoriality. J Peace Res 32(3):277–293
Verbeek M (2009) Econometria. Zanichelli, Bologna
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Agovino, M., Parodi, G. (2012). Civilian Disability Pensions as an Antipoverty Policy Instrument? A Spatial Analysis of Italian Provinces, 2003–2005. In: Parodi, G., Sciulli, D. (eds) Social Exclusion. AIEL Series in Labour Economics. Physica, Heidelberg. https://doi.org/10.1007/978-3-7908-2772-9_8
Download citation
DOI: https://doi.org/10.1007/978-3-7908-2772-9_8
Published:
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-7908-2771-2
Online ISBN: 978-3-7908-2772-9
eBook Packages: Business and EconomicsEconomics and Finance (R0)