Figures and data in The importance of choosing the data set for tax-benefit analysis

Abstract

Given the increased availability of survey income data, in this paper we analyse the pros and cons of alternative data sets for static tax-benefit microsimulation in Italy. We focus on all possible alternatives, namely using (a) SHIW or (b) IT-SILC data using a consistent net-to-gross microsimulation model, or (c) IT-SILC data using the gross incomes provided since 2007. Our results suggest that IT-SILC improves in the regional representativeness of the Italian population and does not perform worse than SHIW as for most demographic characteristics, SHIW provides more information regarding building and real estate incomes. Gross income variables simulated by using the net-to-gross module included in the TABEITA microsimulation model and calibrating for tax evasion provide a very precise fit with external statistics. Simulated IT-SILC gross income data fit external aggregate data even better than gross income data provided in IT-SILC, which tend to largely overestimate self-employment income. Finally, we suggest to match IT-SILC with SHIW to include in the former the information on building and real estate incomes that are contained only in the latter. This allows us to reach a very satisfactory validation of the final data set.

This article contains 14 tables.

Tables

Table 1. Comparison between 2008 SHIW and 2009 IT-SILC.

	SHIW	IT-SILC
Provider	Bank of Italy	Istat
Frequency of data collection	Biannual	Annual
Period of data collection	January-September 2009	September-October 2009
Income reference period	2008	2008
Coverage	Private households	Private households
Statistical units	Households and individuals	Households and individuals
Sample size (households)	7,977	20,492
Sample size (individuals)	19,907	51,196
Response rate	56.1%	85.5%
Geographical Representativeness	Three Macro regions	Regional
International comparability	No¹	Yes
Sample design	Two-stage (first stratified)	Two-stage (first stratified)
Data collection methodology	CAPI/PAPI	PAPI
Gross income variables	No	yes (starting from 2007)

Source: Our calculation and analysis of SHIW and IT-SILC.

Note: ¹International comparability is not among the main aims of the SHIW. It can however be obtained upon adequate harmonization of main variables, such as the “lissification” procedure (see LIS project).

Table 2. SHIW additional Information: Real-Estate Properties different from principal residence -Average yearly effective and imputed rent.

	Frequencies %	Rented properties Effective Rent (c) Euro	Unrented properties Imputed Rent (d) Euro	(d/c)-1*100 Diff. %
Total units of building and real estate	100	5,217	2,610	−49.97
Residential buildings	51.53	4,985	3,528	−29.22
Offices	1.46	7,914	5,686	−28.15
Storehouses	4.21	7,665	4,734	−38.23
Shops	4.11	7,195	5,743	−20.18
Workshops	0.52	4,860	11,892	144.68
Garages and basements	7.45	3,179	1,280	−59.74
Arable land	25.56	2,570	1,146	−55.39
Non-arable land	5.15	4,350	856	−80.33

Source: our elaboration on SHIW data.

Table 3. Population Statistics: frequencies and their standard errors.

	External source Frequencies (a)	Frequencies (b)	SHIW Standard error	Diff. % (b/a-1)*100	Frequencies (c)	IT-SILC Standard error	Diff. % (c/a-1)*100
Panel a: total population
Total	60,045,068	59,721,994	425,204	−0.54	59,726,481	187,436	−0.53
Males	29,152,423	29,039,856	298,942	−0.39	29,034,258	133,299	−0.41
Females	30,892,645	30,682,138	302,382	−0.68	30,692,223	131,737	−0.65
Panel b: Population by Geographical Partition
North West	15,917,376	15,832,738	243,504	−0.53	15,821,674	99,431	−0.60
North East	11,473,120	11,351,612	162,575	−1.06	11,387,593	63,830	−0.75
Center	11,798,328	11,708,006	192,699	−0.77	11,720,956	76,598	−0.66
South	20,856,244	20,829,638	237,817	−0.13	20,796,258	116,865	−0.29
Panel c: Population by Regions
Piemonte	4,432,571	4,040,054	69,762	−8.86	4,404,240	43,130	−0.64
Valle D’Aosta	127,065	213,437	9,954	67.97	128,551	1,773	1.17
Lombardia	9,742,676	9,877,494	209,934	1.38	9,684,939	64,349	−0.59
Trentino Alto Adige	1,018,657	926,708	42,773	−9.03	1,018,870	13,472	0.02
Veneto	4,885,548	4,985,755	122,846	2.05	4,849,150	34,935	−0.75
Friuli Venezia Giulia	1,230,936	1,268,878	30,745	3.08	1,221,728	15,022	−0.75
Liguria	1,615,064	1,701,753	44,885	5.37	1,603,944	26,705	−0.69
Emilia Romagna	4,337,979	4,170,271	87,184	−3.87	4,297,845	33,274	−0.93
Toscana	3,707,818	4,064,424	81,482	9.62	3,678,686	34,202	−0.79
Umbria	894,222	899,923	26,000	0.64	893,152	13,906	−0.12
Marche	1,569,578	1,632,901	35,183	4.03	1,559,290	18,397	−0.66
Lazio	5,626,710	5,110,758	146,875	−9.17	5,589,828	46,453	−0.66
Abruzzo	1,334,675	989,980	37,424	−25.83	1,322,333	22,113	−0.92
Molise	320,795	563,074	27,802	75.52	319,701	5,599	−0.34
Campania	5,812,962	5,121,838	125,890	−11.89	5,796,817	52,115	−0.28
Puglia	4,079,702	3,859,146	106,992	−5.41	4,074,872	41,321	−0.12
Basilicata	590,601	1,621,012	59,568	174.47	590,344	10,574	−0.04
Calabria	2,008,709	1,990,086	51,394	−0.93	2,008,038	30,027	−0.03
Sicilia	5,037,799	4,811,870	126,081	−4.48	5,024,861	62,789	−0.26
Sardegna	1,671,001	1,872,632	36,650	12.07	1,659,292	35,834	−0.70

Source: our elaboration on DemoIstat, values as 1st January 2009 (External Source) and SHIW and IT-SILC data.

Note: Following EUROMOD conventions, in IT-SILC individuals who were born after the income reference period have been dropped (see Ceriani et al., 2012).

Table 4. Population Statistics: frequencies and their standard errors.

Variable	External source Frequencies (a)	Frequencies (b)	SHIW Standard error	Diff. % (b/a-1)*100	Frequencies (c)	IT-SILC Standard error	Diff. % (c/a-1)*100
Age 0–19
Italy	11,408,746	10,669,166	179,935	−6.48	10,760,931	83,853	−5.68
North West	2,811,590	2,530,113	96,815	−10.01	2,673,483	43,519	−4.91
North East	2,069,019	2,164,051	72,273	4.59	1,958,606	28,342	−5.34
Center	2,106,040	1,737,163	75,409	−17.52	1,955,060	32,628	−7.17
South	4,422,097	4,237,839	109,069	−4.17	4,173,782	53,824	−5.62
Age 20–65
Italy	37,179,179	36,436,238	337,064	−2.00	36,948,154	151,911	−0.62
North West	9,872,235	9,598,375	194,968	−2.77	9,760,948	80,082	−1.13
North East	7,124,107	7,241,443	134,291	1.65	7,039,530	51,288	−1.19
Center	7,293,877	6,809,238	147,241	−6.64	7,254,419	62,513	−0.54
South	12,888,960	12,787,182	186,642	−0.79	12,893,257	95,458	0.03
Age > 65
Italy	11,457,143	12,616,590	185,011	10.12	12,017,396	70,712	4.89
North West	3,233,551	3,704,250	107,748	14.56	3,387,243	39,170	4.75
North East	2,279,994	1,946,118	53,937	−14.64	2,389,457	25,007	4.80
Center	2,398,411	3,161,605	98,630	31.82	2,511,477	29,813	4.71
South	3,545,187	3,804,617	98,284	7.32	3,729,219	40,583	5.19

Source: our elaboration on DemoIstat, values as 1st January 2009 (External Source) and SHIW and IT-SILC data.

Note: Following EUROMOD conventions, in IT-SILC individuals who were born after the income reference period have been dropped (see Ceriani et al., 2012).

Table 5. Population by highest level of education attained: frequencies and their standard errors.

Variable	External source Frequencies (a)	Frequencies (b)	SHIW Standard error	Diff. % (b/a-1)*100	Frequencies (c)	IT-SILC Standard error	Diff. % (c/a-1)*100
Elementary Education
Italy	12,379,000	11,980,676	174,686	−3.22	11,576,464	73,665	−6.48
North West	2,980,000	2,740,537	95,718	−8.04	2,854,287	37,062	−4.22
North East	2,323,000	1,748,543	51,924	−24.73	2,234,921	24,530	−3.79
Center	2,271,000	2,533,318	79,516	11.55	2,226,255	28,632	−1.97
South	4,805,000	4,958,278	110,357	3.19	4,261,001	48,345	−11.32
Lower Secondary
Italy	16,285,000	15,950,263	215,876	−2.06	15,690,681	99,741	−3.65
North West	4,393,000	4,307,785	130,857	−1.94	4,276,672	54,038	−2.65
North East	2,976,000	2,990,028	76,817	0.47	2,831,499	32,224	−4.86
Center	2,925,000	2,843,163	86,899	−2.80	2,786,859	38,068	−4.72
South	5,991,000	5,809,287	125,616	−3.03	5,795,651	63,650	−3.26
Upper Secondary
Italy	17,077,000	17,915,782	238,861	4.91	17,830,531	101,128	4.41
North West	4,751,000	5,009,707	135,033	5.45	4,886,090	53,287	2.84
North East	3,471,000	3,882,779	99,289	11.86	3,628,807	36,691	4.55
Center	3,666,000	3,529,855	107,904	−3.71	3,798,268	44,818	3.61
South	5,190,000	5,493,441	129,806	5.85	5,517,366	60,352	6.31
Tertiary
Italy	5,574,000	5,566,073	140,786	−0.14	5,780,476	60,781	3.70
North West	1,574,000	1,814,768	81,955	15.30	1,654,332	34,060	5.10
North East	1,053,000	983,817	57,247	−6.57	1,087,134	19,946	3.24
Center	1,297,000	1,421,667	80,259	9.61	1,305,711	25,910	0.67
South	1,650,000	1,345,821	52,594	−18.44	1,733,299	35,096	5.05

Source: our elaboration on Istat (2010d), SHIW and IT-SILC.

Note: Elementary education include the set of population who did not complete any education level or completed just the elementary school (grade 1 to 5). Lower secondary education is the group of people whose highest level of education is grade eight. Upper secondary education refers to people who received a diploma and Tertiary is the set of individuals who attain an undergraduate or graduate degree. Population is age 15 and more in the External Source and in SHIW, while it is age 16 and more in IT-SILC due to data availability.

Table 6. Population Statistics: frequencies and their standard errors.

Variable	External source Frequencies (a)	Frequencies (b)	SHIW Standard error	Diff. % (b/a-1)*100	Frequencies (c)	IT-SILC Standard error (c	Diff. % /a-1)*100
Employees
Italy	21,611,778	18,734,187	228,015	−13.31	22,485,222	118,991	4.04
North West	5,985,582	5,266,597	133,671	−12.01	6,325,447	64,905	5.68
North East	4,637,931	4,296,697	95,571	−7.36	4,273,878	43,170	−7.85
Center	3,944,874	3,752,849	104,983	−4.87	4,627,595	48,960	17.31
South	5,734,778	5,418,044	116,004	−5.52	7,258,302	70,304	26.57
Self-employed
Italy	6,117,343	5,438,250	155,809	−11.10	6,717,910	71,173	9.82
North West	2,067,939	1,712,138	98,638	−17.21	1,840,272	40,080	−11.01
North East	1,280,320	1,217,732	65,355	−4.89	1,308,766	24,974	2.22
Center	1,488,679	1,045,498	68,217	−29.77	1,393,713	26,769	−6.38
South	2,128,093	1,462,882	68,435	−31.26	2,175,159	43,040	2.21
Pensioners
Italy	15,323,148	14,844,649	198,842	−3.12	15,877,444	86,908	3.62
North West	4,439,474	4,671,715	122,667	5.23	4,531,924	47,265	2.08
North East	2,864,306	2,504,802	64,410	−12.55	2,957,492	29,571	3.25
Center	3,055,415	3,412,963	97,978	11.70	3,232,561	36,895	5.80
South	4,476,872	4,255,169	101,464	−4.95	5,155,467	51,590	15.16

Source: our elaboration on MEF (2013) (external source) and SHIW and IT-SILC data.

Note: We count among employees all those who receive employment income and similarly for self-employment and pensioners, consistently with the external sources. Notice that the sum of frequencies in the external sources by geographic partition does not sum up to the total. This is because there is income which cannot be assigned to any Italian region.

Table 7. Population by sector of activity: frequencies and their standard errors.

Variable	External source Frequencies (a)	Frequencies (b)	SHIW Standard error	Diff. % (b/a-1)*100	Frequencies (c)	IT-SILC Standard error (	Diff. % c/a-1)*100
Agriculture	1,287,100	878,337	44,619	−31.76	1,107,662	30,118	−13.94
Industry	6,988,500	7,486,766	150,641	7.13	7,145,046	70,310	2.24
Services	16,662,900	14,908,539	231,390	−10.53	15,034,130	102,976	−9.77

Source: our elaboration on Istat, Employment breakdown by industry (NACE Rev.2) – annual national data, 2009, wired at: http://dati.istat.it/?lang=en.

Table 8. Yearly Net Incomes: averages and their standard errors.

	SHIW		IT-SILC		Diff %
	euro (b)	Std. Err.	euro (c)	Std. Err.	(c/b-1)*100
Employment Income
Italy	15,950	2.17	15,887	2.13	−0.40
North West	17,633	4.63	17,277	4.24	−2.02
North East	15,724	3.82	16,212	4.32	3.11
Center	16,998	5.02	16,177	4.90	−4.83
South	13,759	3.66	14,115	3.60	2.59
Self-Employment Income
Italy	21,109	8.16	18,685	7.84	−11.48
North West	23,018	17.53	20,848	16.22	−9.43
North East	24,151	20.28	21,052	20.66	−12.83
Center	20,043	13.94	19,447	17.58	−2.98
South	17,105	9.78	14,527	9.00	−15.07
Pension Income
Italy	12,354	2.04	12,164	2.53	−1.54
North West	12,807	3.24	12,301	4.88	−3.95
North East	12,450	8.39	11,877	6.75	−4.60
Center	13,210	2.37	12,831	5.14	−2.87
South	10,881	2.11	11,708	2.95	7.61

Source: our elaboration on SHIW and IT-SILC data.

Table 9. Yearly Net Incomes: poverty index, inequality index and their standard errors.

	SHIW		IT-SILC		Diff %
	index (b)	Std. Err.	index (c)	Std. Err.	(c/b-1)*100
Poverty Headcount index	20.53	0.005	19.34	0.004	−5.82
Inequality Gini Index	35.00	0.006	31.60	0.003	−9.81

Source: our elaboration on SHIW and IT-SILC data.

Note: Poverty and Gini inequality computed on total household disposable income divided by the square root of household size

Table 10. Comparison between External Sources and TABEITA (SHIWT and IT-SILCT) gross incomes.

Variable	External Source		No evasion				Evasion
			SHIW_T		IT-SILC_T		SHIW_T		IT-SILC_T
	Frequencies	Amounts euro	Freq Diff%	Amount Diff%	Freq Diff%	Amount Diff%	Freq Diff%	Amount Diff%	Freq Diff%	Amount Diff%
Total taxable income	41,466,397	782,593,452	3.54	−4.47	−0.72	9.64	3.54	−8.69	−0.72	3.23
Employment income	21,611,778	418,740,720	−13.31	−14.61	4.04	5.94	−13.31	−14.67	4.04	5.72
Self-employment income	6,117,343	109,565,087	−11.10	30.30	9.82	44.63	−11.10	0.96	9.82	0.34
Pension income	15,323,148	213,594,560	−3.12	−1.75	3.62	6.53	−3.12	−1.96	3.62	6.28
Total deductions	12,687,840	21,721,425	−44.31	30.25	−22.60	43.59	−44.55	5.74	−22.05	7.10
Net taxable income	40,249,514	753,556,569	5.19	−5.81	2.16	8.43	5.17	−9.48	2.16	2.82
Total tax credit	39,423,594	62,917,813	−6.00	1.04	2.10	0.67	−5.67	2.55	2.49	2.40
Net personal income tax Regional	31,087,681	146,157,039	6.32	−11.54	7.57	9.20	5.39	−18.17	5.59	−0.85
additional income tax	30,652,846	8,633,217	7.83	−9.20	9.10	7.87	6.88	−13.46	7.09	1.13
Cadastral income main residence	29,776,305	10,551,000	−18.70	0.00	−21.19	0.00	−18.70	0.00	−21.19	0.00
Cadastral income other buildings	17,513,880	7,723,079	−14.19	2.69	−86.97	−55.97	−14.19	2.69	−86.97	−52.71

Source: our elaboration on MEF (2013) (External Source) and SHIW and IT-SILC data.

Note: Freq Diff % = frequencies SHIWT (IT-SILCT) over external source frequencies; Amount Diff % = amount in euro SHIWT (IT-SILCT) over external source amounts. The net personal income tax is the due tax, net of all tax credits. The net taxable income is the tax base, i.e. gross taxable income net of tax allowances.

Table 11. Comparisons between External Sources and TABEITA (IT-SILCT) and IT-SILC gross incomes.

Variable	External source		IT-SILC_T		IT-SILC
	Frequencies	Amounts million euro	Frequencies	Amounts million euro	Frequencies	Amounts million euro
Total taxable income	41,466,397	782,593	40,993,736	795,174	41,048,992	848,645
Employment income	na	na	21,485,120	428,534	21,485,120	431,602
Self-employment income	na	na	7,498,613	129,802	7,498,613	171,175
Pension income	na	na	16,541,269	240,099	16,677,270	245,868

Source: our elaboration on MEF (2013) (External Source) and IT-SILC data.

Note: “na” stands for not available. In fact, IT-SILC definition of gross employment, self-employment and pension income is not consistent with what provided in MEF aggregate tables used for validation of TABEITA (Table 10). As there exist no external sources to validate IT-SILC aggregated variables except for total taxable income we aggregated gross variables simulated using TABEITA (IT-SILCT) consistently with IT-SILC definition of gross variables and indirectly validate IT-SILC data. IT-SILC gross incomes are released as aggregate variables subject to administrative exact matching that we are not able to totally replicate in TABEITA. This explains the small differences between IT-SILC_T and IT-SILC frequencies for pension income.

Table 12. Comparisons between External Sources and TABEITA with calibration (IT –

{SILC}_{T}^{C}

) of cadastral incomes for building different from principal residence.

Variable	External Source		IT – ${SILC}_{T}^{C}$
	Frequencies	Amounts euro	Evasion Freq Diff %	Amount Diff %	No evasion Freq Diff %	Amount Diff %
Total taxable income	41,466,397	782,593,452	0.59	10.39	0.59	4.04
Employment income	21,611,778	418,740,720	4.04	6.10	4.04	5.87
Self-employment income	6,117,343	109,565,087	9.82	44.76	9.82	0.53
Pension income	15,323,148	213,594,560	3.62	6.62	3.62	6.39
Total deductions	12,687,840	21,721,425	−22.72	43.75	−22.07	7.20
Net taxable income	40,249,514	753,556,569	3.35	9.19	3.36	3.65
Total tax credit	39,423,594	62,917,813	2.12	0.38	2.36	2.05
Net personal income tax	31,087,681	146,157,039	9.27	10.55	7.39	0.59
Regional additional income tax	30,652,846	8,633,217	10.82	8.82	8.91	2.16
Cadastral income main residence	29,776,305	10,551,000	−21.19	0.00	−21.19	0.00
Cadastral income other buildings	17,513,880	7,723,079	−56.83	−6.65	−56.83	0.26

Source: our elaboration on MEF (2013) (External Source) and IT-SILC data.

Note: Freq Diff % = frequencies IT – SILCf over external source frequencies; Amount Diff % = amount in euro IT – SILCf over external source amounts.

Table 13. TABEITA (SHIW_T and IT-SILC_T) gross incomes. Averages and their standard errors.

Variable	No evasion				Evasion
	SHIW_T		IT-SILC_T		SHIW_T		IT-SILC_T
	Average euro	Std. Err.	Average euro	Std. Err.	Average euro	Std. Err.	Average euro	Std. Err.
Total taxable income	17,413.61	176.07	20,842.63	113.96	16,644.55	162.32	19,624.06	99.61
Employment income	19,086.86	184.06	19,728.63	119.78	19,072.34	183.68	19,687.35	119.43
Self-employment income	26,251.73	831.85	23,588.21	400.25	20,341.09	648.51	16,365.29	280.38
Pension income	14,136.62	205.80	14,331.29	93.27	14,106.22	206.52	14,297.66	92.81
Total deductions	4,003.99	82.94	3,176.11	40.75	3,264.82	69.93	2,352.12	30.27
Net taxable income	16,764.81	170.18	19,870.83	106.74	16,113.46	158.51	18,843.63	95.18
Total tax credit	1,715.53	7.12	1,573.64	3.97	1,735.05	7.16	1,594.42	3.95
Net personal income tax	3,911.55	79.73	4,772.70	47.73	3,650.65	73.93	4,414.69	41.75
Regional additional income tax	237.15	2.80	278.49	1.69	228.04	2.61	265.96	1.51
Cadastral income main residence	435.84	3.68	449.60	2.82	435.84	3.68	449.60	2.82
Cadastral income other buildings	527.71	20.24	1,490.07	39.08	527.71	20.24	1,600.44	41.98

Source: our elaboration on SHIW and IT-SILC data.

Note: Std. Err. refers to the standard error of the average.

Table 14. TABEITA with calibration (IT – SIL^C_τ) of cadastral incomes for building different from principal residence. Averages and their standard errors.

Variable	IT- ${SILC}_{T}^{C}$
	No evasion		Evasion
	Average Euro	Std. Err.	Average Euro	Std. Err.
Total taxable income	20,712.56	113.76	19,520.17	99.66
Employment income	19,758.79	119.96	19,716.68	119.53
Self-employment income	23,609.04	400.31	16,395.09	280.91
Pension income	14,343.90	93.15	14,312.55	92.81
Total deductions	3,184.75	40.78	2,355.07	30.36
Net taxable income	19,780.29	106.7	18,774.24	95.33
Total tax credit	1,568.60	3.95	1,591.00	3.96
Net personal income tax	4,756.20	47.40	4,404.03	41.51
Regional additional income tax	276.55	1.68	264.19	1.51
Cadastral income main residence	449.60	2.82	449.60	2.82
Cadastral income other buildings	953.51	13.88	1,024.14	14.91

Source: our elaboration on SHIW and IT-SILC data.

Note: Std. Err. refers to the standard error of the average.