Israel Finance Ministry, International Division

Excerpts from the
Annual State Revenue Report for 1997

PERSONAL INCOME TAX IN ISRAEL

	GENERAL
	ANNUAL UPDATE OF THE TAX MODEL
	MAIN FINDINGS OF THE STATE REVENUE ADMINISTRATION TAX MODEL
	STUDY OF INCOME DISTRIBUTION ON THE BASIS OF HISTORICAL DATA

Main Findings of the Tax Model

This chapter discusses income-tax liability from labor and property income and the incidence of direct taxation in Israel-as recorded in the State Revenue Administration tax model, updated to January 1998. The findings point to an average tax burden (income tax and National Insurance contributions, including health tax) of 27 percent. Nearly half (48.1 percent) of taxpayers do not reach the tax threshold, and the 5.2 percent of taxpayers who reside in the highest tax bracket generated 53.5 percent of income tax collected (at the regular rates). The average monthly gross income, adjusted to a twelve-month labor-year, was NIS7,684. The self-employed earned 2.6 times more than employees on average.

1. GENERAL

The State Revenue Administration tax model is based on the Income Tax Administration databases-which include the annual tax returns of the self-employed, employers' reports on their employees, and payments to National Insurance. The tax model datafiles contain information on taxpayers' income, their basic demographic characteristics, and various auxiliary data that, taken together, make it possible to calculate individual taxpayers' tax liability.

The tax model is based on annual income. In the past, data on average monthly income were released without any adjustment to actual employment period during the year. This adjustment pertains to wage-earning taxpayers who entered or left the labor market during the tax year, e.g., those who newly joined the market and those who left it for good (through retirement, for example) or temporarily (by becoming unemployed, by being seasonal workers, or by being quitting their jobs to care for children, etc.). Evidently, when taxpayers are employed for only part of the year, their monthly income (calculated from the annual income) is higher when adjusted to labor-months than when unadjusted.

Part-time employment is commonplace among low-paid employees located in the low deciles of annual income distribution; the opposite-employment in more than one post, with more than twelve labor-months per year-is typical of persons in the high deciles of the income distribution. (See Table VII-1.) The immediate result of the adjustment is a narrowing of gaps in the distribution of adjusted monthly income, an adjustment, however, that has nothing to do with inequality in income distribution as measured in annual terms.

As stated, the adjustment for labor-months was performed only for wage and salary income; monthly income from a business, property, and capital was left unadjusted. Because the adjustment also disregards part-time positions, it falls within the range bounded by unadjusted annual income and income per labor-hour, for which the tax model has no data. It is proper to note here that the differences among population groups in income per labor-hour have additional origins: differences in schooling, scniority, occupation, and industry, to mention only a few, as well as discrimination.

Table VII-1
Distribution of Average Monthly Taxable Income-Adjusted and Unadjusted
(NIS, January 1998 Prices)

	All taxpayers	Employees only
Decile¹	Unadjusted monthly income	Adjusted monthly income	Unadjusted monthly income	Adjusted monthly income	Average labor-months per year
1	502	1,396	475	1,303	4.3
2	1,068	2,023	967	1,814	6.1
3	1,752	2,674	1,580	2,353	7.7
4	2,476	3,291	2,250	2,872	9.1
5	3,269	3,965	2,965	3,416	10.2
6	4,192	4,800	3,791	4,104	11.0
7	5,384	5,927	4,839	5,029	11.5
8	7,067	7,651	6,309	6,452	11.7
9	10,082	10,631	8,835	8,840	12.0
10	24,727	25,530	18,118	17,940	12.1
Average	6,053	7,684	5,013	6,190	9.6
Gini index	0.5244	0.4582	0.4909	0.4186

Source:State Revenue Administration tax model for 1995.

Note to the table:

1. Annual taxable income deciles for all taxpayers and for employees separately.

2. ANNUAL UPDATE OF THE TAX MODEL

The State Revenue Administration tax model is updated each year in two senses: database and tax function. The most recent adjustment was performed in two phases:

First, the 1994 taxpayer database, used for the 1996 tax model (updated to January 1997), was replaced with the 1995 database, used for the 1997 tax model (updated to January 1998). Second, the tax function (income tax and National Insurance contributions, including health tax) that was correct as of January 1997 was replaced with that for January 1998. Since this adjustment resembles the process invoked the previous year, the findings of the two years' tax models are comparable. The changes in income distribution between 1996 and 1997 are a consequence of several factors that affect the database: differences in the level of total economic activity and activity in various industries, and differences in growth rates of the self-employed and wage-earning sectors, to name only two. Legislative changes between January 1997 and January 1998 also prompted changes in income distribution. In personal income tax, the effect of legislative changes in 1997 was estimated at approximately NIS1 billion, foremost because tax brackets and credit points were not adjusted. These changes raised the share of personal income tax in GDP by half a percentage point but are not reflected in the comparison between January 1998 and January 1997 because the tax brackets and credit points were adjusted in the beginning of 1998.

Table VII-2
Tax-Model Population by Vocation, Sex, and Marital Status¹ in 1995

(Thousands)

	Marital status
	Taxpayers, total	Single	Married	Divorced	Widowed	Unknown
Total	2,321	600	1,507	84	108	22
Men	1,352	360	925	25	32	10
Women	969	240	582	59	76	12
Employees, total²	2,027	575	1,269	70	100	13
Men	1,088	340	701	16	29	2
Women	939	235	568	54	71	11
Self-employed, total	294	25	238	14	8	8
Men	264	20	223	9	3	8
Women	30	5	14	5	5	1

Source: State Revenue Administration tax model for 1995.

Notes to the table:

1. The marital status of the self-employed is reported as shown in 1995 tax returns; that of employees is as shown in Population Registry records as of the end of 1995. The "Unknown" category includes taxpayers who are separated from their spouses.
2. Directors of companies that they control are counted among the self-employed.

In addition to the data released annually-such as average monthly income and direct tax burden by deciles of the whole population and of employees, the self-employed, men, and women (Tables VII-3-VII-8)- this year's data include the total composition of taxable income of employees and the self-employed (Table VII-9). In the past, the State Revenue Administration published such statistics randomly, most recently in 1983 for employees and in 1976 for the self-employed.

On the occasion of Israel's fiftieth anniversary, the distribution of income and in-kind transfers in pre-Israel Palestine and after statehood was proclaimed is reviewed. The available raw data are not copious, especially for the early period, and are not comparable because of differences in the definition of income and its incidence. To reconcile the historical data and create a common denominator, a processing method based on Monte-Carlo numerical simulations was developed. The method and the main findings are described in Part 4. The incidence of in-kind transfers in the early 1960s and the 1990s, reckoned on the basis of the simulation method, is presented in Part 5.

3. MAIN FINDINGS OF THE STATE REVENUE ADMINISTRATION TAX MODEL

1. In January 1998, 48.1 percent of taxpayers-52 percent of employees and 21.8 percent of the self-employed-were under the tax threshold.

2. Some 5.2 percent of taxpayers were in the highest bracket. Their taxable income (at regular rates) was 27.6 percent of all income and the taxes on their income generated 53.5 percent of income tax receipts, similar to the previous year's level.

3. The average income tax burden in January 1998 was 19.9 percent and the average National Insurance burden was 7.1 percent. Thus, the total direct-tax burden was 27.0 percent, as it was in January 1997.

4. The Gini index for gross income was 0.524 as against 0.506 a year previous. The index declined to 0.458 after income tax was remitted (0.439 in 1996) and to 0.448 after remittance of income tax and National Insurance contributions (0.428 in 1996). Thus, inequality in gross-income distribution rose in 1997 relative to 1996.

5. On the average, 87.4 percent of taxpayers were employees in January 1998-1 percentage point higher than in the beginning of 1997-but their proportion fell steadily in the upper deciles of income distribution. Only 31.6 percent of taxpayers in the uppermost centile were employees. The average gross income of self-employed taxpayers was 2.6 percent higher than that of employees; this ratio fell to 2.1 after gross income was adjusted to the number of labor-months. The disparity between employees and the self-employed widens in the high deciles: in the uppermost decile and the uppermost centile, the self-employed out-earned employees by factors of 3.3 and 5.2, respectively. The Gini index for gross income of employees was 0.491 and that of self-employed 0.542. Directors of corporations are included among the self-employed.

6. The average income-tax burden was 16.8 percent for employees and 28.0 percent for the self-employed. The average burden in National Insurance contributions was 7.0 percent and 7.2 percent, respectively. The Gini index for net income was 0.421 and 0.468, respectively.

7. The gross income of men was 1.9 times higher than that of women. The differential widened as men and women climbed the income scale, men out-earning women by factors of 2.3 in the uppermost decile and 3.1 in the uppermost centile. On average, women employees work fewer months during the year and fewer hours per week (30.5 as against 42.2). After the adjustment for labor-months is made, the gross income gap narrows to 1.7. According to the 1995 Income Survey, urban male employees had 1.2 times the gross wage per labor-hour of their women counterparts. The gross income of men is less equitable than that of women, as reflected in Gini indices of 0.528 and 0.462, respectively.

8. Men had average burdens of 23.1 percent in income tax and 7.2 percent in National Insurance contributions; the corresponding figures for women were 11.6 percent and 6.9 percent. Several factors help explain the disparity in income-tax burdens: Women receive larger personal tax credits than men, tend more than men to hold part-time jobs, and are paid less per hour. Thus, their average total income is lower and is taxed at lower marginal and average rates.

9. Almost all the taxable income of employees originates in wages and pension (Table VII-9), whereas much income of the self-employed comes from businesses and vocations. A large portion of the wage and pension income of the self-employed comes from the earnings of corporate directors. The share of income that is taxed at special rates, and the share of capital gains, are substantial only in the uppermost decile of the self-employed and stand at 40 percent in the uppermost centile.

Table VII-3
Distribution of Taxpayers, Income, and Tax Liability, by Tax Brackets:¹ January 1998

(Percent)

Marginal tax rate	Percent of taxpayers in each tax bracket	Percent of population under tax threshold	Percent of income in total income	Percent of tax liability in total tax revenue
Taxpayers, total
10	26.4	26.4	4.1	0.0
20	22.6	18.0	10.3	0.3
30	35.1	3.7	35.2	18.8
45	10.7	-	22.8	27.4
50	5.2	-	27.6	53.5
Employees, total	100.0	48.1	100.0	100.0
10	29.1	29.1	5.2	0.0
20	23.7	19.0	12.7	0.4
30	34.1	3.9	39.8	24.8
45	9.4	-	23.6	33.8
50	3.6	-	18.7	41.0
Self-employed, total	100.0	52.0	100.0	100.0
10	8.2	8.2	0.8	0.0
20	15.0	11.2	3.5	0.1
30	41.7	2.4	21.9	8.9
45	19.0	-	20.5	17.0
50	16.1	-	53.3	74.0
Total	100.0	21.8	100.0	100.0

Source: State Revenue Administration tax model for 1995.

Note to the table:

1. Income liable to tax at regular tax rates; tax liability at regular rates only.

Table VII-4
Average Monthly Income, Direct-Tax Burden, Average Marginal Tax Rate,
and Ratio of Employees to Self-Employed, by Income Deciles,¹ January 1998

(NIS)

Decile	Average gross monthly income per taxpayer²	Gross income distribution (pct.)	Direct tax burden as percent of income			Net monthly income per taxpayer	Net income distribution (pct.)	Self-employed in decile (pct.)
			Total	Income tax	National Insurance³
1	502	0.8	4.8	0.0	4.8	478	1.1	2.4
2	1,068	1.8	4.9	0.0	4.9	1,015	2.3	4.1
3	1,752	2.9	5.1	0.0	5.1	1,662	3.7	6.2
4	2,476	4.1	5.9	0.3	5.6	2,333	5.3	7.5
5	3,269	5.4	7.5	1.6	5.9	3,025	6.8	8.8
6	4,192	6.9	11.7	4.9	6.8	3,704	8.4	11.3
7	5,384	8.9	16.3	8.9	7.4	4,507	10.2	14.6
8	7,076	11.7	21.3	13.3	8.0	5,566	12.6	17.3
9	10,082	16.7	27.8	19.2	8.6	7,275	16.5	22.1
10	24,727	40.9	40.8	34.0	6.8	14,631	33.1	34.6
Uppermost centile	77,319	12.8	43.0	40.2	2.8	44,065	10.0	68.4
Average	6,053		27.0	19.9	7.1	4,418		12.6
Gini index		0.524					0.448

Average marginal income tax rate, weighted by number of taxpayers: 25.1 percent
Average marginal income tax rate, weighted by taxpayers' taxable income (regular tax rates): 37.4 percent

Source: State Revenue Administration tax model for 1995.

Notes to the table:

1. Deciles of annual income liable to income tax for the population at large; corporate directors are counted among the self-employed.
2. Taxable income is not adjusted for full-time positions or twelve months' salary. The adjustment is described in Part 1.

3. Including health tax. For employees, only employee payments are included.

Table VII-5
Employees: Average Monthly Income, Direct Tax Burden,
and Average Marginal Tax Rate, January 1998

(NIS)

Decile	Average gross	Direct-tax burden as percent of income			Average net
	monthly income¹	Total	Income tax	National Insurance²	monthly income
1	475	4.6	0.0	4.6	453
2	967	4.9	0.0	4.9	916
3	1,580	5.0	0.0	5.0	1,490
4	2,250	5.6	0.3	5.3	2,123
5	2,965	6.3	1.0	5.4	2,778
6	3,791	9.6	3.5	6.1	3,427
7	4,839	14.2	7.3	6.9	4,153
8	6,309	19.0	11.6	7.4	5,108
9	8,835	25.1	17.0	8.1	6,620
10	18,118	38.7	31.1	7.6	11,109
Uppermost centile	41,420	44.5	39.9	4.6	22,898
Average	5,013	23.8	16.8	7.0	3,818
Gini index	0.491				0.421

Average marginal income tax rate, weighted by number of taxpayers: 23.9 percent.
Average marginal income tax rate, weighted by taxpayers' taxable income (regular rates): 35.1 percent.

Source: State Revenue Administration tax model for 1995.

Notes to the table:

1. Taxable income is not adjusted for full-time positions or twelve months' salary. The adjustment is described in Part 1.

2. Including health tax. For employees, only employee payments are included.

Table VII-6
Self-Employed: Average Monthly Income, Direct Tax Burden,
and Average Marginal Tax Rate, January 1998

(NIS)

Decile	Average gross	Direct-tax burden as percent of income			Average net
	monthly income²	Total	Income tax	National Insurance¹	monthly income
1	1,334	7.1	0.3	6.8	1,240
2	2,771	8.1	0.7	7.4	2,545
3	3,995	12.2	3.7	8.5	3,506
4	5,165	16.8	7.7	9.1	4,299
5	6,455	20.4	11.0	9.4	5,135
6	8,060	24.0	14.2	9.8	6,122
7	10,128	27.9	17.6	10.3	7,307
8	13,386	33.9	23.4	10.5	8,852
9	19,737	40.6	30.1	10.5	11,718
10	59,840	43.1	39.1	4.0	34,061
Uppermost centile	216,388	39.4	38.4	1.0	131,210
Average	13,088	35.2	28.0	7.2	8,479
Gini index	0.542				0.468

Average marginal income tax rate, weighted by number of taxpayers: 33.1 percent.
Average marginal income tax rate, weighted by taxpayers' taxable income (regular rates): 43.6 percent.

Source: State Revenue Administration tax model for 1995.

Notes to the table:

1. Including directors of companies that they control.
2. Taxable income is not adjusted for full-time positions or twelve months' salary. The adjustment is described in Part 1.