Israel Finance Ministry, International Division

Excerpts from the
Annual State Revenue Report for 1997

Tax Burden and Public-Sector Tax Revenue

THE TAX TREND IN 1997

The tax burden in 1997 was 40.8 percent of Gross Domestic Product as against 39.4 percent in 1996. Two factors explain most of the increase: legislative changes that magnified the burden by 0.8 percent of GDP and a decrease of 0.4 percent of GDP in income-tax refunds.

However, if we keep in mind the inclusion of the health tax in the definition of the tax burden (begun in 1995) and clearances to the Palestinian Authority, the tax burden in 1997 was 7 percent lower than its peak level in 1986.

Total tax revenues of the public sector in 1997 were NIS137.8 billion, 5.7 percent higher in real terms than in 1996. Direct taxes rose by 10 percent and indirect taxes by only 1 percent (see Table IV-1). Consequently, the share of direct taxes climbed from 50 percent of total taxes in 1996 to 52 percent in 1997, similar to the 1994 level. (See Table IV-4.)

1. THE TAX TREND IN 1997

The trend in taxes in 1997 was affected by three factors:

a.	Legislative changes that went into effect in 1997 (non-adjustment of personal income-tax brackets, an increase in employers' contributions to National Insurance, an increase in the rate of withholding at source, and increases in fuel excise, tobacco tax, and purchase tax on automobile air conditioners) and changes that went into effect in 1995-1996 but did not have their full effect until 1997 (for example, reduction of the "parallel tax" on employers in 1995, which raised taxable income and tax payments in 1997) boosted total tax receipts in 1997 by NIS2,650 million net: NIS2,000 million in direct taxes and NIS650 million in indirect taxation (for a more detailed presentation, see Chapter I). Legislative changes this year added 0.8 percent to the burden and thereby helped lower the government deficit. Notably, legislative changes reduced the tax burden by 0.9 percent of GDP in the previous two years and thus have had no substantive effect on tax revenues for the past three years.
b.	Deferral of NIS500 million in tax refunds from late 1995 to early 1996 NIS300 million in income-tax returns and NIS200 million in VAT statements. This deferral affected the rates of change in net tax refunds and collections in 1997 as against 1996.
c.	Clearances to the Palestinian Authority (PA)-In the economic appendix to the Interim Agreement between Israel and the PLO (the "Paris Accord"), Israel undertook to forward to the PA taxes that the Israeli tax divisions collect on account of imports to PA-controlled areas and on trade with Israel residents. In 1994, Israel forwarded NIS70 million to the PA on this account. As the agreement matured, these clearances increased to NIS750 million in 1995, NIS1.36 billion in 1996, and NIS1.66 billion in 1997.

Net of the effects of these three factors, collections in 1997 rose by 3.3 percent in real terms-all originating in a 6.5 percent increase in direct taxes; collections of indirect taxes showed no substantial change.

Table IV-1
Tax Collections (NISbillions)

	Total	Direct taxes	Indirect taxes
Actual collections
1996	119.618	59.644	59.974
1997	137.763	71.567	66.196
Real percent change	5.7	10.1	1.3

Collection adjustments
(a) Legislative changes in 1997	2.65	2.00	0.65
(b) Deferral of refunds from 1995 to 1996	0.50	0.30	0.20
(c) Clearances to Palestinian Authority
1996	1.36	0.15	1.345
1997	1.66	0.20	1.64

Collection at uniform rate
1996	121.478	59.959	61.519
1997	136.773	69.587	67.186
Real percent change	3.3	6.5	0.2

The disparity between the 3.3 percent increase in total taxation-the figure that takes account of the aforementioned collection adjustments-and the real GDP increase of 1.9 percent, on the one hand, and the different trends in direct and indirect taxes on the other hand, must be explained. By using the increase in GDP as a benchmark, we see that 1997 ended with a total collection surplus of NIS1.7 billion-an NIS2.9 billion surplus in indirect taxes and an NIS1.2 billion shortfall in indirect taxes. A similar development occurred in 1994, while the collection potential was not attained in 1995 and 1996.

In addition to collection efforts, the deviation from the potential evidently reflects a combination of several factors:

	The composition of Gross Domestic Product: Since the various components of GDP are taxed at different rates, tax collections change more (less) rapidly than the change in GDP if tax-intensive components increase more (less) rapidly than the average. GDP in 1997 rose by 11.2 percent in nominal terms, but imports of goods (excluding defense imports, fuel, and diamonds) increased by only 3.1 percent, and tax-intensive imports of motor vehicles decreased considerably, thus explaining the standstill in indirect taxes.
	Capital gains: Capital gains are not reflected in Gross Domestic Product. In a year when the stock market rises or when a larger number of businesses change owners, accrued capital gains are realized and tax revenues may increase more swiftly than GDP. Such was the case in 1997. We cannot estimate collections of capital-gain taxes with precision, but they seem to have grown by NIS500 million to NIS1 billion in 1997 relative to 1996.
	Real-estate taxes: Another explanation for the stagnation in indirect taxes is the decrease in Value Added Tax on the component of land in the sale of new dwellings and in property tax on land. The bases of these two taxes, like capital gains, are not reflected in the GDP data.
	Lags in collection: Except for import taxes, there is a time lag between the creation of a tax liability and remittance of the tax, especially with respect to business income. Companies and the self-employed pay advances during the year, but only several months after the end of the year does the exact tax liability become clear and the difference is paid in case of liability in excess of the advances or a refund is given in case of overpayment. As Table IV-2 shows, tax refunds to businesses decreased by 21 percent in real terms, so that real net collection from this sector rose by 12 percent even though real collections by means of advances and withholding grew by only 7 percent. NIS0.3 billion of the NIS1.1 billion decrease in tax refunds originated, as stated, in deferral of income-tax refunds from 1995 to 1996. Consequently, NIS0.8 billion of the NIS2.9 billion "surplus" in direct-tax collections in 1997 may be traced to a lag in collection.
	Progressivity of the tax system: In addition to the non-adjustment of personal income-tax brackets, the effect of which was measured in the section on legislative changes, the rate of personal income tax rose because the average wage per employee post increased by 2 percent in real terms, thereby augmenting direct-tax revenues by NIS0.3 billion. Inequality in income distribution may also have grown; if this occurred, it would also have prompted an increase in personal income-tax revenues.

Table IV-2

Income-Tax Collection from Businesses

(NISbillions)

	1993	1994	1995	1996	1997
In current prices
Net collection from companies and self-employed	10.318	13.915	14.971	15.455	18.940
a. Advances and withholding at source	11.404	13.544	14.983	16.369	19.121
b. Tax remainders less refunds	-1.086	0.371	-0.012	-0.914	-0.181
Tax remainders	2.253	2.964	3.406	3.892	3.976
Less refunds	3.339	2.594	3.417	4.806	4.158
Real percent change relative to previous year
Net collection from companies and self-employed		20.1	-2.2	-7.3	12.4
a. Advances and withholding at source		5.8	0.5	-1.9	7.2
b. Tax remainders less refunds
Tax remainders		17.2	4.4	2.6	-6.3
Less refunds		-30.8	19.7	26.3	-20.6