Israel: Selected Issues
Author: International Monetary Fund. European Dept.
Publication Date:
01 May 2018
eISBN:
9781484353387
Language:
English
Keywords:
ISCR; CR; Haredi; employment; productivity gap; investment process; rate of return; dual economy; car ownership; Infrastructure; Labor productivity; investment scale
Download PDF (2.6 MB)

Selected Issues

Abstract

Selected Issues

Unleashing Israel’s Potential: Is Boosting Public Investment the Answer? 1

Israel can be characterized as a dual economy, that is well known for its dynamic high-tech sector, but the remainder of the economy has relatively low productivity. One of the factors weighing on Israel’s labor productivity are its sizable infrastructure needs, which may also hinder the labor participation of some groups. To address this issue, the authorities are preparing a long-term strategy to support higher infrastructure investment. This paper analyzes the macro-fiscal implications of an increase in infrastructure spending, taking into account Israel’s dual economy character.

A. Israel’s Infrastructure Needs

1. Israel’s labor productivity is low and the gap with peers persists. Israel’s high-tech sector is famous for startups, but it also includes R&D centers of many major technology companies. The high-tech sector currently comprises around 21 percent of gross value added, up from 12½ percent in 1995. Despite the strong performance of this sector, Israel’s overall labor productivity rose by only ¾ percent annually in recent decades, leaving its sizable productivity shortfalls virtually unchanged.

uA01fig01

Share of High-Tech Sector

(Percent of total gross value added)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: OECD
uA01fig02

Hourly Labor Productivity 1/

(Ratios of Israel to Countries or Regions, Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: OECD1/ Data in US dollars at constant prices, 2010 PPPs.

2. Israel’s capital stock ratio has declined steadily to be among the lowest in advanced economies. Israel’s total capital stock is estimated to be around 170 percent of GDP,2 which is lower than a peer group of small European countries (left figure below). Following a surge in construction to absorb half a million immigrants from the former Soviet Union, total investment—both public and private—has been declining since the early 1990s. The reduction of public investment, from around three percent of GDP in the second half of the 1990s to less than two in recent years, was part of the fiscal consolidation process. But the amount and quality of infrastructure development was also affected by issues with public investment management, the lack of land registries in the Arab towns, and the regulation of network industries.3

uA01fig03

Total Capital Stock, 2015

(Percent of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: “IMF Investment and Capital Stock Dataset, 2017”Note: Data on PPP are not available for countries with “*.”
uA01fig04

Capital Stock and Government Investment

(Percent of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: “IMF Investment and Capital Stock Data base, 2017”

3. The infrastructure need is especially large in the transportation sector. Cross-country benchmarks suggest a public infrastructure gap of around 35–40 percent of GDP. A more granular sectoral analysis of infrastructure needs recently conducted by McKinsey & Company finds a gap of around 20 percent of GDP. In particular, the public transport infrastructure in major cities is clearly inadequate for Israel’s income level, failing to meet the needs of a growing population concentrated in the urban areas of central Israel. For example, the share of train travel in total motorized travel was only two percent in Israel, compared to eight percent in the OECD or OECD small states. Moreover, public transport does not operate during weekends and religious holidays, creating a strong incentive to own private vehicles. As a result, Israel’s road traffic intensity per network length is by far the highest in OECD. Traffic congestion was causing an average loss of 60 minutes per road-user per day in 2012, weighing on productivity, with an estimated cost to the economy amounting to 1.5 percent of GDP in 2012.

uA01fig05

Share of Train Travel in Total Motorized Travel

(Pessenger-km travelled, percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: OECD.
uA01fig06

Road Traffic Intensity per Network Length, 2014

(1000 vehicle-km/km)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: OECD

4. Prospects are for congestion to worsen. Since the mid-2000s, the expansion of Israel’s car ownership has substantially outpaced population growth and the increase in road volumes. Yet, Israel’s car ownership rate remains low relative to its per-capita income, suggesting that the number of vehicles in Israel may well continue to rise rapidly in the coming years. As a result, congestion will likely continue to increase even with further road expansion, partly due to physical limits on major roads.

uA01fig07

Population, Vehicles, and Road Volume

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: Israeli Ministry of Finance
uA01fig08

Projected Increase in Vehicles

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

5. To address these large infrastructure needs, the government is developing a multi-year strategy for a lasting boost in infrastructure investment. A government committee is developing an integrated long-term national infrastructure strategy through 2030 (“Infrastructure 2030”), while also preparing a list of additional projects for implementation for the next five years. Although the size and modalities of scaling up public investment is yet to be decided, the government has a strong preference to undertake infrastructure projects in a form of Private Finance Initiative (PFI) or Public Private Partnerships (PPPs). This plan is expected to succeed the ongoing five-year infrastructure program for 2017–21, “Investment for Growth,” which supports the development of transport, education, social services, and housing. The total cost of “Investment for Growth” totals NIS 107 billion (8.5 percent of the 2017 GDP), of which NIS 41.6 billion is expected to be financed by the budget and the rest by tariffs, public private partnerships (PPPs), and the private sector. Infrastructure 2030 is expected to target a denser network of urban rail transport, including underground lines in Tel Aviv, a high-speed link between Tel Aviv and Eilat, and new airports in the north and south of the country. Moreover, Infrastructure 2030 is expected to be accompanied by “Employment 2030,” which aims to boost the employment of population subgroups and their productivity, such as by enhancing technological and vocational education.

6. This paper analyzes the macro-fiscal implications of raising public investment. A dynamic stochastic general equilibrium model called DIGNAR (Debt, Investment, Growth, and Natural Resources) is adapted for this analysis. Section B introduces this model and discusses key modifications made for its application to the Israeli economy. Section C uses the model to analyze the macro-fiscal implications of boosting public investment under various scenarios of investment efficiency, financing options, and labor participation and productivity. Section D summarizes key findings and policy recommendations.

B. Adapting the DIGNAR Model to Israel

7. The paper uses a dynamic stochastic general equilibrium model for a small open economy with real variables. Our analysis largely relies on the Debt, Investment, and Growth for Natural Resource (DIGNAR) model, with the steady-state values and structural parameters calibrated for Israel (Annex I).4 This model has the following key features:

  • Comprises three economic entities: (i) households (the spending of a portion of households is subject to a liquidity constraint while the remainder are only subject to their intertemporal budget constraint); (ii) firms (a nontradable goods sector, a non-high-tech tradable sector, and a high-tech tradable sector); and (iii) the government.

  • Covers a relatively long horizon and abstracts from money and nominal rigidities.

  • Provides various financing options for the government: e.g., tax increases, tax benefit reductions, spending cuts, and debt financing.5 Implementing public investment in a PFI or PPP framework may not immediately raise government debt, yet it increases the government’s liabilities (or contingent liabilities). We, therefore, assume that all the additional public investment that is not financed by revenue measures or spending cuts is financed by higher deficits and debt.

  • Explicitly allows for the crowding-in and crowding-out effects of public investment:

    • Public investment can crowd in private investment as public capital enters the production function and increases the returns on private investment. In the non-high-tech tradable sector, higher public capital and hence higher output can also raise total factor productivity in the subsequent period via a “learning-by-doing” channel.

    • Public investment can also crowd out private investment and private consumption by raising real wages, the real exchange rate, real interest rates, and the sovereign risk premium, which also affects firms’ external borrowing costs.6

  • Incorporates the implications for the efficiency of public investment spending from the pace of scaling up public investment.

8. We adapt features of the original DIGNAR model to better reflect the Israeli economy:

  • The natural-resource sector is replaced with a high-tech sector. In DIGNAR, the resource sector facilitates the analysis of issues such as using revenues from exhaustible resources to develop domestic infrastructure and diversify the economy to facilitate continued growth after the resource is depleted. Despite significant gas discoveries, Israel’s resource sector is not sufficiently large to warrant such analysis. In contrast, the high-tech sector is sizable and its rapid growth does not appear to have been significantly impeded by domestic infrastructure needs. In another parallel to a resource sector, strong high-tech exports have supported Israel’s current account and an appreciation of the shekel over the past decade, implying that growth of the high-tech sector could contribute to “Dutch-disease;” indeed, the share of Israeli goods exports in global goods trade has declined notably. In the model, we assume the high-tech sector to grow independently by 4.7 percent annually, faster than the 2.5 percent rate in the rest of the economy, which leads to overall growth of three percent annually with some real exchange rate appreciation. The high-tech sector also has a lower effective tax rate (we assume four percent of high-tech GDP), therefore an increase in the share of high-tech sector in the economy reduces the effective tax rate for the total economy.7

  • Dynamic depreciation rates for roads. The allocation of investment between roads and mass transit is of interest in the case of Israel, requiring some differentiation of their contribution to economic performance in the model. A study finds that the economic rate of return of public infrastructure is on average higher for roads and highways compared to railways and underground transit. But, other literature finds that road-building disproportionately benefits vehicle-intensive industries, and while interstate highway networks offer a one-time productivity increase, gains from other roads are limited. This suggests that the rate of return of roads may initially be high, but it eventually declines below that of mass transit. In the Israeli context, assuming vehicle ownership continues to rise in line with per-capita income while there are physical limits on expanding major roads, congestion will likely continue to rise, eventually triggering a rise in road depreciation rates.8 We therefore assume that mass transit offers lower initial rate of return with a constant depreciation, while roads offer a higher initial rate of return but with faster depreciation rates once per-capita GDP exceeds a pre-determined threshold: 9

uA01fig09

Depreciation rates

(Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates
the elasticity of output to investment(αg)={0.15for mass transit0.21for roadsthe depreciation rate for roads δt={δ¯ifytyss−1≤C¯δ¯+ϕ(ytyss−1)ifytyss−1>C¯

Where C¯ is assumed to be 0.01, which means that depreciation starts to accelerate when output increases by more than one percent from its steady state; ϕ is the elasticity of congestion to output growth, which is assumed to be 0.8.

C. Macro-Fiscal Implications of Boosting Public Investment

9. This section describes macroeconomic dynamics in response to increases in public investment under DIGNAR, and discusses key factors shaping the impact on output and debt. The scale and pace of increasing public investment under Infrastructure 2030 is yet to be decided. For illustrative purposes, the simulations assume an increase in public investment by 1.5 percent of GDP per year—from two percent of GDP to 3.5 percent of GDP per year—over the next 15 years. Such a scenario is roughly the minimum consistent with addressing a public infrastructure need of 20 percent of GDP, and it is not out of line with historical levels of public investment. Key factors that shape the impact of public investment on output and debt include the efficiency of investment, the type of financing used, the pace of capital depreciation, the pace of increasing public investment, and the magnitude of spillovers to labor participation and productivity. The simulation results are presented as deviations from the steady state in which public investment remains at two percent of GDP per year, growth in the non-high-tech sector at 2.5 percent, and overall growth at three percent. The government’s current spending is assumed to grow at the same rate as output.

Baseline Results

10. In the DIGNAR model calibrated for Israel, an increase in public investment could initially crowd out private spending, but then gradually stimulates higher output and private investment. Assuming all the additional public investment is financed by higher deficits, the immediate impact of higher public investment is a rise in interest rates which leads to some initial small crowding out of consumption and private investment, resulting in a transitory decline in the private capital stock. There is also an immediate appreciation of the real exchange rate which unwinds gradually over time. Aggregate output benefits from the accumulation of public capital, together increasing labor supply (in the form of hours worked) owing to real wage increases.10 These developments increase the return on private capital over time, so there is a crowding-in of private capital with some time lags, which also supports growth. By the end of the projection period, public capital is above that without investment scale-up by 35 percent, private capital by four percent, output by five percent, and labor supply (hours worked) along the intensive margin by one percent.

uA01fig10

Dynamics of Key Macroeconomic Variables with Investment Scale-up

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Efficiency of Public Investment

11. The efficiency of public investment is the proportion of investment spending that is actually turned into an increase in public capital. For example, an efficiency rate of 85 percent indicates that 85 cents out of one dollar invested turns into public capital. Based on the literature, the baseline scenario assumes an efficiency level of 85 percent (medium). We also consider high and low scenarios with efficiency of 95 and 75 percent, respectively.

12. Growth benefits from public investment depend importantly on investment efficiency. At the end of the 15-year period of higher public investment with medium efficiency, the output gain relative to baseline is five percent, with the stock of public capital raised by 35 percent from baseline, private capital above by 10 percent, and labor supply up by one percent. Meanwhile, the end-period output with the high- and low-efficiency rate is 6.5 percent and four percent above baseline respectively, because (i) efficiency affects the stock of public capital at the end of the period, which increase by 44 and 25 percent respectively, and (ii) the crowding in effect is stronger with high efficiency, with private capital rising five and three percent respectively from baseline.

uA01fig11

Output Increase from Investment Scale-up 1/

(Percent deviation from output without investment scale-up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates.1/ The additional real output when investment increases by 1.5 percent of GDP per year over 15 years under different efficiency levels. The efficiency in public investment is defined as the proportion of investment that is translated into capital stock. High, medium, and low efficiency levels correspond to 85, 76, and 67 percent of an increase in investment turns into capital accumulation over 15 years, respectively.

Financing Options

13. Allowing all the additional investment to be reflected in higher deficits financed by debt leads to significantly higher debt ratios. Productive public investment can raise potential output and future revenue, which in turn could offset debt increases from financing the investment. However, our analysis suggests that GDP gains are not sufficiently large to self-finance higher investment, as—from an initial level of 60 percent—the debt ratio rises to 66 percent at the end of the period even in the case of high efficiency. Lower efficiency levels reduce growth impacts, which imply higher debt ratios, at 72 percent in the case of low efficiency. This suggests a need to explore options for nondebt financing. The government can either increase revenues or cut spending, but Israel’s very small civilian budget suggests focusing on different options to raise revenues.

uA01fig12

Government debt

(Percent of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates

14. Different financing options lead to different growth impacts from higher investment. For example, the simulations below show the impact of each financing instrument if it is used to finance the entire scaling up of investment:

  • VAT hike: initially weighs on consumption and output, reducing labor demand. But higher savings leads to higher private investment, such that output gains are second only to tax benefits in the long run.

  • Labor tax hike: reduces labor supplied by households, which in turn lowers household income and private consumption, resulting in the smallest output gain in the long run.

  • Cut in tax expenditure: limited impact on labor supply or consumption delivers the highest output increase.11

uA01fig13

Non-high-tech Output Increase

(Percent deviation from output without investment scale-up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates1/ Cut in tax benefits by 1.5 percent of GDP throughout the projection period
uA01fig14

Labor Supply

(Percent deviation from labor supply without investment scale up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates
uA01fig15

Private Consumption

(Percent deviation from private consumption without investment scale up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates

15. Fully financing higher investment with revenues would reduce debt ratios modestly. Reflecting the increase in output relative to baseline, across all scenarios with revenue financing, the debt ratio declines to between 57 to 58 percent of GDP. Financing with tax benefits results in a slightly faster pace of debt reduction—aided by more favorable output response—followed by VAT financing and labor tax financing.

uA01fig16

Government Debt

(Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates1/ Cut in tax benefits by 1.5 percent of GDP throughout the projection period.

16. But financing higher investment fully with taxes would require a rather sharp increase in tax rates. For example, assuming all the additional investment to be financed by additional VAT collections, the VAT rate would have to increase to near 21 percent in two years, up from 17 percent. Similarly, when financing higher investment with increased labor taxes, the labor tax rate would have to rise from 25 to 27 percent in two years.

uA01fig17

Tax Rates with Full Tax Financing of Investment Scale-Up

(Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates

17. Combining tax and debt financing can strike a balance, where the advantages of reducing tax benefits are clear. For example, assuming that a third of the higher investment is tax financed, the size and pace of tax rate increases are smoothed significantly while debt accumulation also remains contained.12 It is notable that financing high investment by cutting tax benefits by one percent of GDP per year throughout the projection period has the most positive impact on output, especially in the medium term, helping to prevent debt accumulation from the initial level.

uA01fig18

Tax Rate with A Third Tax Financed

(Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates
uA01fig19

Non-high-tech Output Increase

(Percent deviation from non-HT output without investment scale-up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF Staff Estimates1/ Cut in tax benefits by 1 percent of GDP throughout projection period.
uA01fig20

Government Debt

(Percents of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF Staff Estimates1/ Cut in tax benefits by 1 percent of GDP throughout the projection period.

Roads Versus Mass Transit

18. As a result of the dynamic depreciation described previously in paragraph 8, investing only in roads would initially raise output more than investing only in mass transit, but the impact would eventually fade. This will also lead to a higher debt ratio increase in the 15-year period if investing only in roads. The government could maximize the rate of return by combining roads and mass transit as output gains from well-integrated road and mass transit system can be higher than the weighted average if synergies between road and mass transit augment their economic returns.

uA01fig21

Non-high-tech Output Increase

(Percent deviation from the output without investment scale up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates
uA01fig22

Total Public Debt to GDP

(In percents)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates

Pace of Scaling Up Investment

19. The efficiency of investment can depend on the pace of scaling up investment. In addition to the efficiency rate, the model considers absorptive capacity constraints, which tend to arise from technical capacity limits—which impact project selection, management, and implementation—leading to waste and leakage of resources in the investment process. Such absorptive capacity constraints can have long-lasting negative effects on growth, Esfahani and Ramirez (2003). The government’s absorptive capacitive constraints can cause efficiency losses if the pace of scaling up investment is too fast. In the model, such constraints are governed by (i) the threshold of absorptive capacity, which is the point at which further raising the growth rate in public investment leads to efficiency declines; and (ii) the severity of absorptive capacity constraints, which determines the degree of such efficiency losses. Below the absorptive threshold, the effective investment rate equals the efficiency rate, but it declines above that threshold reflecting severity.

uA01fig23

Public Investment

(Percent of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff Calculations
uA01fig24

Net Efficiency of Public Investment 1/

(Percent)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff Calculations1/ Net efficiencies equal the initial efficiency level until public investment growth exceeds the 0.75 threshold, which then start decreasing once public investment growth exceeds the threshold, following the formula, ϵ(γ¯tGI)=exp⁡[−ζϵ(γ¯tGI−γ¯GI)]ϵ¯, where ζϵ ∈ [0, ∞).

20. Front-loading capital accumulation can have a lingering positive impact on output even with larger waste, but excessive front-loading can be counter-productive. Front-loading of public capital accumulation can crowd in private investment at earlier stages, which in turn can have a sustained positive impact on output. However, a fast pace of boosting public investment can lead to larger waste if the pace exceeds the government’s absorptive capacity, which in turn slows the increase in output and results in larger debt accumulation. At the same time, back-loading public investment also appears counter-productive as it delays the crowding-in of private investment. Debt accumulation would slow initially, but the limited positive impact on output would eventually weigh on debt dynamics.

uA01fig25

Non-high-tech Output Increase

(Percent deviation from the output without investment scale up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates
uA01fig26

Total Public Debt to GDP

(In percents)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates

Labor Force Participation and Productivity Gains

21. Growth benefits of public investment increase if labor participation and productivity are boosted. Good infrastructure not only facilitates trade, bolsters market integration and competition, and enhances access to resources and public services, it can also help some people enter the labor market. Prior to 2009, a significant number of minority communities had no bus connection to the rest of the country (Knesset, 2014). Although the situation has improved, Arab towns remain relatively poorly served by public transport, which affects their residents and creates barriers to employment, especially for women, who are less likely to be licensed to drive and/or own cars (OECD 2018). If part of Infrastructure 2030 is focused on addressing the remaining issues of access to public transportation, it could raise the labor participation of disadvantaged groups and their better access to public services, including higher education, could enhance labor productivity.

uA01fig27

Scenario: Employment Rate

(Percent of population)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Sources: Central Bureau of Statistics, IMF staff calculations
uA01fig28

Scenario: Hourly Wage

(Index, Non-Haredi Jewish Men=100)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Sources: Central Bureau of Statistics, IMF staff calculations

22. If the scaling-up public investment is coupled with deep structural reforms that together reduce the gaps in labor participation and productivity, the combined output impacts are much greater. In some illustrative scenarios, it is assumed that reforms are adopted such that during the 15-year period of higher public investment there are declines in productivity and participation gaps in relation to non-Haredi Jewish men.13 The output impact of scaling-up public investment together with halving the labor participation gap is double that from scaling-up public investment alone. Higher output would also help stabilize debt at around 66 percent of GDP even if all the additional public investment is debt-financed, assuming that all the structural reforms are funded without raising the deficit. If both the labor participation and productivity gaps are narrowed by half, the output gain would be triple that from just increasing public investment, and 0debt would be put in a declining path after peaking at around 63½ percent of GDP.

uA01fig29

Output Increase

(Percent deviation from the output without investment scale up)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates1/ Closing the labor force participation gap between non-Haredi Jewish men (the frontier) and the rest by half.2/ Closing the labor force participation and productivity gap between non-Haredi Jewish men (the frontier) and the rest by half.
uA01fig30

General Government Debt 1/

(Percent of GDP)

Citation: IMF Staff Country Reports 2018, 112; 10.5089/9781484353387.002.A001

Source: IMF staff estimates1/ All the scenarios assume full debt financing of investment scale -up.2/ Closing the labor force participation gap between non-Haredi Jewish men (the frontier) and the rest by half.3/ Closing the labor force participation and productivity gap between non-Haredi Jewish men (the frontier) and the rest by half.

D. Key Findings and Policy Discussion

23. The efficiency of investment is key to ensuring growth benefits are achieved and to containing increases in the public debt ratio. Selecting projects with low rates of return, managing public investment inefficiently, or raising investment faster than absorptive capacity, can lead to weaker growth benefits and higher debt ratios that reduce the room to sustain increased public investment. To enhance the efficiency of public investment, Israel should:

  • Establish a body with clear accountability and sufficient powers for upgrading Israel’s infrastructure, supported by staff with the necessary technical expertise.

  • Make project evaluation and selection more rigorous and transparent, including by ensuring consistency with a long-term infrastructure strategy.

  • Streamline zoning and permitting processes and address other bureaucratic impediments to timely project implementation.

  • Improve coordination between ministries and between the central and local governments. Broadening the coverage of the medium term fiscal framework to the general government could contribute to improved coordination and planning as local governments implement around three-quarters of public investment.

  • Phase any scaling up of public investment judiciously to avoid waste.

  • Use a public-private partnerships (PPP) only in cases where bringing in private sector know-how improves efficiency. Design and monitor PPPs carefully to protect the public interest.

  • Maintain a high level of transparency around the level and composition of investment, including to help protect public investment against short-sighted cuts.

24. Yet, growth benefits will likely be insufficient to prevent a significant increase in debt ratios, indicating a need for revenue measures, where reductions in tax benefits are preferable. Allowing the public debt ratio to rise as much as 10 percentage points appears too high as Israel faces wider uncertainties than most advanced economies and it should also preserve fiscal space to facilitate structural reforms for long-term growth. Given Israel’s very low civilian spending, the government should consider financing most of the additional investment with additional revenues. Israel’s sizable foregone revenue from various tax benefits—around five percent of GDP per year—suggests significant scope for revenue gains. Our analysis also suggests that reducing tax benefits is least detrimental to growth, which in turn would be most positive for debt dynamics.

25. PFI or PPP financing should be used prudently with the associated risks monitored carefully. PFIs/PPPs can defer spending on infrastructure without deferring its benefits.14 However, fiscal risks (e.g., direct costs, contingent liabilities) associated with PFIs/PPPs can be large, therefore such approaches should not be used as a means simply to avoid reporting higher public debt.15 The government should restrict the use of a PFI/PPP framework to those projects that fit in a clear overall investment strategy and provide value-for-money, and where bringing the know-how of the private sector has clear benefits. Liabilities, including contingent liabilities, from PFIs/PPPs should be managed carefully and reported in line with international best practices.

26. Growth benefits can be substantially augmented if higher public investment is coupled with structural reforms to enhance labor participation and productivity. Expanding public transportation can improve access to better jobs and education, which in turn can raise labor participation and productivity. In particular, providing public infrastructure that can enable workplaces to locate within or close to minority communities would enhance labor participation and allow increased work hours, especially for Arab and Haredi women. But structural reforms are needed to make more substantial progress in reducing participation and productivity gaps. In this context, it is welcome that the government is preparing strategies to enhance labor participation and productivity along with Infrastructure 2030.

Annex I. Baseline Calibration of Key Parameters

The model is calibrated for the economy of Israel using annual data. The initial values are based on the macroeconomic data for Israel in 2017. Most of the structural parameters are drawn from the Bank of Israel’s DSGE model for the Israeli economy (MOISE), while some other parameters follow the original DIGNAR model by Melina et al. (2016). The below lists the parameters that do not follow either source.

  • Returns on public capital depend on the combination of the elasticity of output to public capital, the depreciation rates, and the efficiency level.

    • Elasticity of output to public capital: the meta-analysis by Melo et al. (2013) lists a wide range of estimates for the elasticity of output to transport infrastructure from −0.148 to 0.315. Another meta-analysis by Bom and Ligthart (2014) narrows this range by controlling for the definition of public infrastructure capital and output, and whether capital is installed at the national level or by state and local government. They suggest an elasticity of core infrastructure installed by a national government to be 0.17 and by local government to be 0.193. Our mid-point calibration of 0.18 results in an initial return of public capital at 30 percent. As Israel is an advanced economy, this is slightly higher to other estimates such as 22 percent for World Bank transportation projects and 25 percent for scaling up public investment in developing economies (Box 3.4 WEO October 2014). Our cumulative 15-year fiscal multipliers are also in line with estimates for advanced economies.

    • Depreciation rates: we use the estimates for public and private capital depreciation rates based on country income accompanying the Investment and Capital Stock Database 2017. In the case of Israel public capital depreciation, since we are focusing on the non-high-tech sector, we apply an annual depreciation rate of four, which is closer to the estimate for middle-income countries.

    • The initial efficiency level: we assume 85 percent for Israel. This means that for every dollar spent on investment, 85 cents would translate into public capital stock. This parameter is often calibrated to be 50 percent for low-income countries (Prichett 2000, Melina et al. 2016, Box 3.4 WEO October 2014). The Regional Economic Outlook 2016 uses the Data Envelopment Approach to estimate the frontier of public investment efficiency across 115 countries. If the frontier economies have a score of one, Israel shows an average score of 0.95. In practice, there is some unproductive public investment even in the most efficient economy. Therefore, we consider three scenarios of a high efficiency level at 95 percent, a medium efficiency at 85 percent, and a low efficiency level at 75 percent.

  • Frisch elasticities for the response of labor supply to changes in income are sensitive to estimation methods and sample selection. For example, studies based on micro data often find lower elasticities than in macro studies and estimates are often lower for men than for women. Reichling and Whalen (2012) provides examples of Frisch elasticity estimates in the literature ranging from zero to 0.8 for men, and from 0.5 to one for women. Macro models have used larger elasticities such as 1.9 (Smets and Wouters 2007), 2.6 to 4.0 (Cho and Cooley 1994, King and Rebelo 1999). Chetty et al. (2011) suggests that macro-models should match a Frisch elasticity of aggregate hours of 0.75. To differentiate between two types of households, in our model, we choose a lower elasticity of 0.5 for the credit-constrained households and an elasticity of 0.8 for the optimizers.

Initial values for Israel (in percent, unless indicated otherwise)

article image

References

  • Barro, Robert J. 1990. “Government Spending in a Simple Model of Endogenous Growth.” Journal of Political Economy 98 (5), Part 2: S103–25.

    • Search Google Scholar
    • Export Citation

  • Barro, Robert J., and Xavier Sala-i-Martin. 1995. Economic Growth. New York: McGraw-Hill.

  • Bom, Pedro R. D., and Jenny E. Ligthart. 2014. “What Have We Learned from Three Decades of Research on the Productivity of Public Capital?” Journal of Economic Surveys 28: 889–916.

    • Search Google Scholar
    • Export Citation

  • Buffie, Edward E., Andrew Berg, Catherine Pattillo, Rafael Portillo, and Luis-Felipe Zanna. 2012. “Public Investment, Growth, and Debt Sustainability: Putting Together the Pieces.” IMF Working Paper 12/144, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation

  • Chetty, Raj, Adam Guren, Day Manoli, and Andrea Weber. 2011. “Are Micro and Macro Labor Supply Elasticities Consistent? A Review of Evidence on the Intensive and Extensive Margins.” American Economic Review 101 (3): 471–5.

    • Search Google Scholar
    • Export Citation

  • Cho, Jang-Ok, and Thomas F. Cooley. 1994. “Employment and Hours over the Business Cycle.” Journal of Economic Dynamics and Control 18 (2): 411–32.

    • Search Google Scholar
    • Export Citation

  • Currie, Janet, and Reed Walker. 2011. “Traffic Congestion and Infant Health: Evidence from E-ZPass.” American Economic Journal: Applied Economic Journal: Applied Economics 3 (1): 65–90.

    • Search Google Scholar
    • Export Citation

  • Fernald, John G. 1999. “Roads to Prosperity? Assessing the Link between Public Capital and Productivity.” The American Economic Review 89 (3): 619–638.

    • Search Google Scholar
    • Export Citation

  • Futagami, K., Y. Morita, and A. Shibata. 1993. “Dynamic Analysis of an Endogenous Growth Model with Public Capital.” Scandinavian Journal of Economics 95: 607–25.

    • Search Google Scholar
    • Export Citation

  • Glomm, Gerhard, and B. Ravikumar. 1994. “Public Investment in Infrastructure in a Simple Growth Model.” Journal of Economic Dynamics and Control 18 (6): 1173–87.

    • Search Google Scholar
    • Export Citation

  • Greiner, Alfred, W. Semmler, and G. Gong, 2005, The Forces of Economic Growth: A Time Series Perspective, 1st edition (Princeton, N.J.; Princeton University Press).

    • Search Google Scholar
    • Export Citation

  • Greiner, Alfred. 2007. “An Endogenous Growth Model with Public Capital and Sustainable Government Debt.” Japanese Economic Review 58: 345–361.

    • Search Google Scholar
    • Export Citation

  • Hodge, Andrew. 2016. “The Design of Fiscal Reform Packages: Insights from a Theoretical Endogenous Growth Model.” IMF Working Paper 16/146, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation

  • King, Robert G., and Sergio T. Rebelo. 1999. “Resuscitating Real Business Cycles.” In Handbook of Macroeconomics, Vol. 1, edited by John B. Taylor and Michael Woodford. Elsevier.

    • Search Google Scholar
    • Export Citation

  • Melina, Giovanni, Shu-Chun S. Yang, and Luis-Felipe Zanna. 2016. “Debt Sustainability, Public Investment, and Natural Resources in Developing Countries: The DIGNAR Model.” Economic Modelling 52 (B): 630–49.

    • Search Google Scholar
    • Export Citation

  • Melo, Patricia, Daniel Graham, and Ruben Brage-Ardao. 2013. “The Productivity of Transport Infrastructure Investment: A Meta-Analysis of Empirical Evidence.” Regional Science and Urban Economics 43 (5): 695–706.

    • Search Google Scholar
    • Export Citation

  • Munnell, Alicia H. 1992. “Policy Watch: Infrastructure Investment and Economic Growth.” Journal of Economic Perspectives 6 (4): 189–198.

    • Search Google Scholar
    • Export Citation

  • Organisation for Economic Co-operation and Development (OECD), 2018, “OECD Economic Surveys: Israel 2018,” OECD Publishing, Paris.

  • Pritchett, Lant. 2000. “The Tyranny of Concepts: CUDIE (Cumulated, Depreciated, Investment Effort) Is Not Capital.” Journal of Economic Growth 5 (4): 361–384.

    • Search Google Scholar
    • Export Citation

  • Reichling, Felix, and Charles Whalen. 2012. “Review of Estimates of the Frisch Elasticity of Labor Supply.” Working Paper Series, Congressional Budget Office, Washington, DC.

    • Search Google Scholar
    • Export Citation

  • Smets, Frank, and Rafael Wouters. 2007. “Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach.” American Economic Review 97 (3): 586–606.

    • Search Google Scholar
    • Export Citation

  • Turnovsky, Stephen J., and Partha Sen. 1991. “Fiscal Policy, Capital Accumulation, and Debt in an Open Economy.” Oxford Economic Papers 43 (1): 1–24.

    • Search Google Scholar
    • Export Citation
1

Prepared by Aiko Mineshima and Vina Nguyen (all EUR). The chapter benefited from comments and suggestions from Giovanni Melina (RES) and colleagues from the Bank of Israel and Israeli Ministry of Finance.

2

IMF estimate. The capital stock is estimated with the perpetual inventory method. The public capital stock does not include capital transfers to state-owned enterprises for investment purposes.

4

See Melina, Yang, and Zanna (2016) for technical details of the model.

5

Several studies (e.g., Barro 1990, Sala-i-Martin 1995, Futagami et al. 1993, Glomm and Ravikumar 1994, and Hodge 2016) analyze the growth impact of public investment in the context of endogenous growth models, but they assume government balanced budget rules (or for the case of Hodge (2016), the primary balance is imposed to maintain public debt at sustainable levels). Others (e.g., Turnovsky 1990, Greiner et al. 2000, and Greiner 2007) incorporate government debt in their endogenous growth framework, but do not allow for different financing schemes and ignore the role played by the structural and policy conditions for debt sustainability.

6

We assume a small elasticity of sovereign risk premium (0.001) given Israel’s access to international capital markets and moderate level of debt, which is slightly smaller than the Bank of Israel (BoI)’s finding (0.005 before the global financial crisis and 0.007 after the crisis). Simulations results, however, virtually do not change with the BoI elasticity.

7

Lower tax rates on the high-tech sector are motivated by the high mobility of capital in this sector in the long run.

8

Fernald (1999) models average congestion as the ratio of road stock divided by some measure of road use. Barro and Sala-i-Martin (1995, p.158) suggest that aggregate output or private capital are good proxies for congestion in a model with long-run growth.

9

In principle, roads and mass transit could be modelled as complementary capital inputs in a production function, but this was not feasible within DIGNAR.

10

The degree of response is governed by the elasticity of labor supply to income (so-called “Frisch elasticities”), which is assumed to be different between households with and without liquidity constraints.

11

We proxy cuts in tax benefits by an improvement in revenue efficiency—revenue increases without creating any distortions. In reality, however, the elimination of certain tax benefits—e.g., VAT exemptions on fresh fruit and vegetables, income tax benefits for families—is akin to raising VAT or labor tax, although primarily impacting the population that face fewer liquidity constraints.

12

19 percent is one percentage point higher than Israel’s historical high in the VAT rate of 18 percent.

13

Such reforms include education and training, product markets including regulation, and measures to support labor participation of women as discussed in the staff report for the 2018 Article IV consultation with Israel.

14

PPPs are typically long-term contracts where the private sector executes and finances public investment.

15

Queyranne (2014)

References

  • Azary-Viesel, Sagit, and Ben-David, Dan, 2016, “On the Gender Gap in Israel’s Labor Market,” Policy Brief (Tel Aviv: Shoresh Institution for Socioeconomic Research).

    • Search Google Scholar
    • Export Citation

  • Bank of Israel, 2015, “The Earned Income Tax Credit: A Preliminary Report on a Designated Survey among Eligible Individuals,” Recent Economic Developments 140 (Jerusalem: Bank of Israel).

    • Search Google Scholar
    • Export Citation

  • Bank of Israel, 2017, “Is there a Gender Difference in the Basic Skills of Workers and in the Return on those Skills in Israel?” (Jerusalem: Bank of Israel).

    • Search Google Scholar
    • Export Citation

  • Berman, Eli, 2000, “Sect, Subsidy, and Sacrifice: An Economist’s View of Ultra-Orthodox Jews”, The Quarterly Journal of Economics, Vol. 115, Issue 3, pp. 905–953.

    • Search Google Scholar
    • Export Citation

  • Ben-David, Dan, and Ayal Kimhi, 2017, “An Overview of Israel’s Education System and its Impact,” Shoresh Research Paper (Tel Aviv: Shoresh Institution for Socioeconomic Research).

    • Search Google Scholar
    • Export Citation

  • Bleikh, Haim, 2016, “Poverty and Inequality in Israel: Trends and Decompositions,” State of the Nation Report 2016 (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Fuchs, Hadas, 2016, “Gender Gaps in the Labor Market: Wage and Occupational Segregation,” State of the Nation Report 2016 (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Fuchs, Hadas, 2017, “Education and Employment Among Young Arab Israelis,” State of the Nation Report 2017 (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Christiansen, Lone Engbo, Huidan Lin, Joana Pereira, Petia Topalova, and Rima Turk, under the guidance of Petya Koeva Brooks, 2016, “Unlocking Female Employment Potential in Europe: Drivers and Benefits,” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Christiansen, Lone, and Robert Sierhej, 2016, “Female Labor Force Participation in Poland,” Republic of Poland: Selected Issues, IMF Country Report No. 16/211 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Cornfeld, Ofer, and Oren Danieli, 2015, “The Origins of Income Inequality in Israel—Trends and Policy,” Israel Economic Review Vol. 12, No. 2 (2015), pp. 51–95 (Jerusalem: Bank of Israel).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, 2014, “Fiscal Policy and Income Inequality,” IMF Policy Paper (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Israel Ministry of Finance, 2016, “The Next Locomotive of Growth? The Economic Benefit of Reducing Gender Disparities in the Labor Market,” Weekly Focus (Jerusalem: Ministry of Finance).

    • Search Google Scholar
    • Export Citation

  • Israel Ministry of Finance, 2017a, “The Effect of the Labor Grant (Negative Income Tax) on Poverty and Inequality,” Weekly Focus (Jerusalem: Ministry of Finance).

    • Search Google Scholar
    • Export Citation

  • Israel Ministry of Finance, 2017b, “Is there a glass ceiling for women,” Weekly Focus (Jerusalem: Ministry of Finance).

  • Israel Ministry of Finance, 2017c “Employment and Wages among Haredi Men”, Weekly Focus (Jerusalem: Ministry of Finance).

  • Israel Ministry of Finance, 2018, “Contribution of allowances and compulsory payments to reducing poverty in Israel”, Weekly Focus (Jerusalem: Ministry of Finance).

    • Search Google Scholar
    • Export Citation

  • OECD, 2018, “OECD Economic Surveys: Israel 2018.” (Paris: OECD).

  • Regev, Eitan, 2016, “Challenges of Haredi Integration in Academic Studies,” State of the Nation Report 2016 (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Regev, Eitan, 2017, “Patterns of Haredi Integration into the Labor Market,” State of the Nation Report 2016 (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Shachar, Ella, “The Effect of Childcare Cost on the Labor Supply of Mothers with Young Children,” Discussion Paper No. 2012.12 (Jerusalem: Bank of Israel).

    • Search Google Scholar
    • Export Citation

  • Taub Center, 2017, “The Arab Education System in Israel: Are the Gaps Closing?” Bulletin Article (Jerusalem: Taub Center for Social Policy Studies in Israel).

    • Search Google Scholar
    • Export Citation

  • Thegeya, Aaron, 2015, “Income Inequality in Israel,” Israel: Selected Issues, IMF Country Report No. 15/262 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Yashiv, Eran and Nitsa Kasir, 2015, “The Labour Market of Israeli Arabs: Key Features and Policy Solutions,” Policy Insight No. 78 (Washington: CEPR).

    • Search Google Scholar
    • Export Citation
1

Prepared by Aiko Mineshima (EUR) and Henrique Chociay (FIN). The chapter benefited from comments and suggestions from the Israeli authorities.

2

The relative poverty rate is the share of the population with income below 50 percent of the median income.

3

Market income includes income from work and capital, receipts from employment-related social insurance, and transfers from non-profit institutions and other households.

4

Disposable income is defined as the sum of the market income and social security payments minus employment-related contributions, taxes on wealth and income, and social security contributions.

5

The share of Arab-Israeli and Haredi population is around 25 percent, yet 44 percent of population under 14 years old are from Arab-Israeli or Haredi families.

6

In principle, Haredi students are part of the Hebrew system. However, in practice, Haredi boys do not participate in PISA, therefore the results for the Hebrew system can be interpreted as the performance of non-Haredi Jewish students.

7

The long-term trend of Haredi men employment, declining from 80 percent in 1979 to less than 40 percent in the mid-2000s, was similar to that of men with up to four years of formal education (Ben-David and Kimhi, 2017).

8

Azary-Viesel and Ben-David (2016) report that the gap in has been diminishing, but in the “wrong” direction, as the result of lower enrollment of boys in advanced math and science courses.

9

According to OECD (2018), adjusting Israel’s skills dispersion to the PIAAC average would reduce wage inequality by almost four percent.

10

These illustrative calculations, based on 2015 data, assume that the relative change in output is equivalent to the relative change in total wage payments.

11

MoF (2016) estimates that closing only the monthly wage gender gap between non-Haredi Jews by 40 percent (with no change in gender gaps of other populations) could yield a long-term output gain of about 7 percent, considering changes in the demographic composition. A simpler simulation by staff, without taking into account the demographic changes, finds a similar impact of around 7¾ percent.

12

Dan Ben-David and Ayal Kimhi, An Overview of Israel’s Education System and its Impact, Shoresh Research Paper, December 2017.

13

Taub Center, 2017, The Arab education system in Israel: Are the gaps closing?

14

Christiansen and Sierhej (2016) point that provision of childcare tends to have a positive association with labor force participation, while family cash transfers are negatively correlated.

15

Prior to May 2003, the monthly allowance increased substantially from the fourth child onwards, possibly creating incentives to higher fertility, with a negative impact on labor participation. The 2003 reform set a flat-rate benefit per child. However, as the children who were born under the old schedule were grandfathered, the impact was seen only gradually. In 2009, another reform was introduced to provide an additional allowance for the second to fourth children. Average annual allowances per child dropped from more than NIS 4,000 before the 2003 reform to around NIS 2,000 in 2013 (Thegeya, 2015).

16

According to the MoF’s estimates using the CBS Household Expenditure Survey 2015, the EITC reduces the poverty gap by 2.9 ppt to 35 percent, and the Gini coefficient by 0.2 ppt to 0.364.

17

In other countries, conditional cash transfers have been attached to attendance of children at health clinic and school (as in Brazil), or to education performance (as in the U.K.). These schemes have been successful in ensuring access to basic education and health services, and in increasing investments in human and physical capital, reducing the inter-generational transmission of poverty (IMF, 2014).

18

MoF (2018) finds that those allowances that are intended for individuals who cannot participate in the labor market, such as old-age and disability benefits, have the largest contribution to poverty reduction.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
Israel: Selected Issues
Author: International Monetary Fund. European Dept.
Volume/Issue:
Volume 2018: Issue 112
Publisher:
International Monetary Fund
ISBN:
9781484353387
ISSN:
1934-7685
Pages:
40
DOI:
https://doi.org/10.5089/9781484353387.002

Share of High-Tech Sector

(Percent of total gross value added)

Hourly Labor Productivity 1/

(Ratios of Israel to Countries or Regions, Percent)

Total Capital Stock, 2015

(Percent of GDP)

Capital Stock and Government Investment

(Percent of GDP)

Share of Train Travel in Total Motorized Travel

(Pessenger-km travelled, percent)

Road Traffic Intensity per Network Length, 2014

(1000 vehicle-km/km)

Population, Vehicles, and Road Volume

Projected Increase in Vehicles

Depreciation rates

(Percent)

Dynamics of Key Macroeconomic Variables with Investment Scale-up

Output Increase from Investment Scale-up 1/

(Percent deviation from output without investment scale-up)

Government debt

(Percent of GDP)

Non-high-tech Output Increase

(Percent deviation from output without investment scale-up)

Labor Supply

(Percent deviation from labor supply without investment scale up)

Private Consumption

(Percent deviation from private consumption without investment scale up)

Government Debt

(Percent)

Tax Rates with Full Tax Financing of Investment Scale-Up

(Percent)

Tax Rate with A Third Tax Financed

(Percent)

Non-high-tech Output Increase

(Percent deviation from non-HT output without investment scale-up)

Government Debt

(Percents of GDP)

Non-high-tech Output Increase

(Percent deviation from the output without investment scale up)

Total Public Debt to GDP

(In percents)

Public Investment

(Percent of GDP)

Net Efficiency of Public Investment 1/

(Percent)

Non-high-tech Output Increase

(Percent deviation from the output without investment scale up)

Total Public Debt to GDP

(In percents)

Scenario: Employment Rate

(Percent of population)

Scenario: Hourly Wage

(Index, Non-Haredi Jewish Men=100)

Output Increase

(Percent deviation from the output without investment scale up)

General Government Debt 1/

(Percent of GDP)