Benefits of Boosting Quality Public Infrastructure Spending in Romania1
Core Questions and Findings
Why is the quality of infrastructure in Romania relatively poor? Several factors likely contribute to the poor quality of infrastructure. Despite some recent reforms, Romania still lacks a robust framework for developing, vetting, prioritizing, and executing public investment projects. A coherent and predictable strategic plan is lacking to guide the project selection process and preserve priorities over the medium term. The prospect of electoral gains may encourage politicians to announce and begin new infrastructure projects, but provide little incentive to see the projects through to completion. In addition, the transportation network is dominated by large, inefficient, and financially vulnerable state-owned enterprises.
What is the status of the investment project portfolio? The central government project pipeline is clogged with old projects. There are over 500 central-government investment projects with a value equivalent to 31 percent of GDP. Many projects have long implementation periods with 78 projects expected to take more than 10 years to complete, including one with a duration of 42 years. In addition, there is little transparency about the thousands of local government investment projects. Moreover, the integration of project prioritization process into the budget and medium-term spending plans is limited.
Why is Romania’s absorption of EU structural and cohesion funds (SCF) low and what could be done to improve absorption? For the 2007–13 programming period, Romania has absorbed about 52 percent of the EUR 19.1 billion (15 percent of 2007 GDP) of SCF at its disposal. It is eligible for another EUR 22.5 billion (14 percent of GDP) in SCF during 2014–20. Romania’s low absorption rate reflects weak project management capacity in the government and in beneficiaries and a preference for initiating projects funded through the state budget.
What is the size of the growth dividend Romania could expect from increasing quality public investment that is primarily financed through increased EU-funds absorption? Using three different techniques, staff estimate that a 0.7 percent of GDP annual increase in investment over the period 2016–20, funded by SCF grants and cuts in national capital spending, could increase output in the medium term up to 1½–3 percent over the baseline forecast. Based on a production function model, the potential GDP growth rate would be 0.5 percentage point higher in 2020.
1. Since the financial crisis, fiscal and current account deficits have been tackled, but the infrastructure deficit has widened. Romania’s infrastructure is suboptimal due to a backlog of incomplete projects, as well as administrative deficiencies in the maintenance and operation of the infrastructure (Romania, 2014). Romania needs to make quality investments, particularly in infrastructure, for today and tomorrow. Based on the EU’s production function methodology and demand side projections using conservative multiplier estimates, this paper shows that increased quality public investment in infrastructure can boost growth in Romania, consistent with recent research. Greater absorption of available EU grant funding could play an instrumental role in increasing the quality and quantity of infrastructure spending in a tight budget environment. In addition, acceleration of project management reforms and stricter implementation of medium-term budgets are needed to realize the full growth benefits of infrastructure investment.
B. Infrastructure—Low Quality
2. Romania’s public-sector capital spending has significantly outpaced that of other EU members, but its infrastructure is perceived as being low quality. Several factors likely contribute to the poor return on public investments in infrastructure. Despite some recent reforms, Romania still lacks a robust framework for developing, vetting, prioritizing, and executing public-investment projects. The prospect of electoral gains may encourage politicians to announce and begin new infrastructure projects, but provide little incentive to see the projects through to completion. Large state-owned enterprises (SOEs) that are subject to political interference dominate key network industries. This is particularly true in the transportation sector, where inefficient SOEs are responsible for building and maintaining the rail and road infrastructure and delivering rail services (see Selected Issues paper “Romanian State-Owned Enterprises: Challenges and Reform Priorities”).
Efficiency of Capital Spending in the EU-28
Sources: Eurostat; and World Competitiveness Report.
3. In addition, infrastructure density, particularly the road network, is relatively low. The low density may be attributable to broader issues. A 2012 World Bank study notes that “there is a serious problem of leadership, management, and governance in the transport sector, including in roads.” A separate World Bank report, using a specialized dataset of road sector contracts under Bank-funded projects in 14 countries in East Europe and Central Asia, found that Romania had the second highest average number of red flags per contract. It is not surprising that this environment would yield a low road density.
Total Length of Motorways
1/ Unweighted average of Bulgaria, Czech Republic, Hungary, Poland and Slovakia
4. The legacy of poor project planning coupled with political expediency is a large, unwieldy central government project pipeline clogged with old projects. The central government’s project database at end-2013 included 543 projects with an investment value equivalent to 31 percent of GDP. Based on World Bank analysis, only 85 of the projects receive non-reimbursable funding from the European Union, while another 97 projects received no funding at all in 2013 (World Bank, 2014). Cost overruns amount to 2 percent of GDP. Many projects have long implementation periods with 78 projects expected to take more than 10 years to complete, including one with a duration of 42 years. Moreover, the budget does not provide funding for all projects with a duration of less than a year. Funding for some of these projects will inevitably have to be postponed to future years. More broadly, given the long duration of the portfolio, it is quite possible that many projects are no longer consistent with current objectives.
Total Cost of Public Investment Projects by Duration (Years), 2013
1/ Reflects 322 central-government projects with known dates.
Sources: World Bank, Romanian authorities and IMF staff calculations.
C. Absorption of EU Structural and Cohesion Funds (SCF)
5. European Union (EU) funds available to Romania represent a significant source of financing for economic growth.2 During the 2007–13 programming period, Romania was eligible for EUR 19.1 billion (15.3 percent of 2007 GDP) of external financing from the EU’s structural and cohesion funds (SCF).3 These funds were allocated across seven operational programs (OP) designed to foster Romania’s competitiveness and cooperation and convergence with other EU members. The OPs for transportation and environment, primary infrastructure development, were the largest, accounting for 46 percent of the available resources. Romania is eligible for another EUR 22.5 billion (13.7 percent of estimated 2014 GDP) under the 2014–20 programming period. Six OPs are being considered for this period, with the large infrastructure OP allocated EUR 9.4 billion, part of which is contingent upon EC approval of a Master Transportation strategy.
|Absorption Rate||Estimated budget|
|Operational Program||Budget 2007–13 1/||SCF payments 2/||(Percent)||2014–20 3/|
6. Romania’s absorption of SCF is the lowest in the EU, despite a recent pick up. As of early November 2014, Romania’s absorption rate, including pre-financing, was 50 percent, compared to an average of 71 percent in other new member states. The human resources OP has the lowest absorption rate of 41 percent among the operation programs. Romania has until the end of 2015 to absorb the remaining resources for each OP (a cumulative EUR 8.6 billion, 5.3 percent of GDP) from the 2007–13 programming period.4 Nonetheless, more important than the degree of absorption is the effectiveness with which SCF transfers have been used.
EU Funds Absorption, 2014
Source: European Commission.
7. Absorption of SCF for the 2007–13 programming period has suffered largely due to domestic governance and capacity issues. Romania’s difficulties in absorbing SCF are well documented (Lungu, 2012; Romania, 2011; KPMG, 2014). These deficiencies have resulted in the suspension or interruption of some operational programs and reimbursement of some amounts received. While some of the obstacles to absorption have been addressed, others remain. For example, beneficiary capacity and skill to identify, elaborate, and manage projects continues to be weak. Public administration of the operational programs also faces deficiencies in evaluating and monitoring projects. Finally, it is possible that public officials may prefer to initiate and implement investment projects that are financed only by Romania’s central and local governments because these projects do not carry the level of monitoring and evaluation that accompanies in EU-funded projects.
D. Recent Reforms to Improve the Quality of Public-Sector Investments
8. Recent reforms under the IMF/EU supported-program have focused on improving project prioritization and EU funds absorption. The authorities adopted a legally-binding project prioritization methodology,5 with World Bank assistance, applicable to all significant projects (value greater than RON 100 million) that are funded by the state budget. The framework was applied during the 2014 budget cycle and a list of priority projects was published (Annex). However, the Public Investment Plan (PIP) lacks a multi-annual time dimension that should accommodate the full life cycle of projects, based on Fund technical assistance. As with the entire central-government project portfolio, the highest priority significant projects have long durations (average duration is 10.7 years).
9. In 2013, the Public Investment and Valuation Unit (PIEU) was created in the Ministry of Public Finance (MoPF), but more can be done to enhance its effectiveness. The unit’s mandate is to help strengthen quality control in the preparation, prioritization, appraisal and management of new significant public investment projects financed by the central government (IMF, 2014a). However, in practice, many of the project life cycle steps remain the responsibility of line ministries. To date, PIEU activities have been limited to reviewing pre-feasibility studies for projects with a value greater than RON 100 million that are prepared by line ministries and compiling a list of priority projects for an inter-ministerial committee, which approves final project selection. In addition, the prioritization process needs to be better integrated into the budget preparation process (the priority list prepared for the 2015 budget cycle was not approved), including medium-term budget projections. Moreover, project design does not follow good practices in terms of definition of objectives and performance indicators, specification of results, and risk analysis. With respect to EU-funded projects, the principle of first come, first-served guides the prioritization process. Responsibility for managing the operating programs under the 2014–20 programming period has been centralized under the Ministry of European Affairs (MEF), with the exception of the regional development OP. This could foster more efficient absorption. Nonetheless, coordination between the MEF and MoPF in key stages of the project and program cycles is limited. Better cooperation would help ensure that overall financial resources are consistent with national policy priorities.
10. Critically, most projects, including local projects, fall outside the ambit of the prioritization framework. There are more than 350 central governments and over 3000 local government projects that are not subject to the government-approved prioritization framework. Efforts to improve the prioritization of these projects and the level of funding transparency have been ad hoc. For example, the Ministry of Regional Development (MoRD) maintains a database of local government projects but prioritization of the projects is rudimentary (e.g., in 2014 those exceeding a certain percentage-of-completion threshold receive financing). The MoRD published a list of local government projects that received budget financing in 2013. This practice should continue. In addition, recurrent costs arising from investment projects do not appear to be fully recognized and provided for in budgets, according to the findings of recent Fund technical assistance.
E. Quality Infrastructure Investment Can Boost Growth
11. Quality public investment in infrastructure can boost domestic demand and potential GDP growth, particularly in low growth environments such as Romania’s. Since the global financial crisis, Romania’s economy has underperformed and the output gap is unlikely to close within the next few years (IMF, 2014a). The labor participation rate also remains one of the lowest in the European Union (see Selected Issues paper on “Cutting Labor Taxes in a Constrained Budget Environment”). At the same time, government borrowing costs and the policy rate are at or near all time lows. In short, Romania’s current situation—economic slack, accommodative monetary policy, cheap money, and an infrastructure deficit—presents a strong case for increasing quality public infrastructure investment, provided the effort to improve the public investment process is accelerated.
12. Supply and demand side estimates yield similar growth benefits from increased quality public investment. EU-funded investment is subject to a higher level of monitoring and evaluation than other investments. Therefore, higher EU-funds absorption brings not only additional investment financing but also higher quality investment. Using three different techniques, staff estimate that raising the EU funds absorption rate from 75 percent in the baseline to 100 percent over the period 2016–20, equivalent to an average of RON 6.2 billion annually (¾ percent of GDP), could increase output by 1½–3 percent over the medium term, based on the most plausible assumptions. The production function approach also shows that the potential GDP growth rate could be 0.5 percentage points higher by 2020.
Production Function Approach
13. Staff simulations employing the EU’s production function methodology show significant growth benefits from higher quality infrastructure spending. Using the Cobb-Douglas production function together with projections for capital stock, labor supply and total factor productivity (TFP), staff calculated potential growth and output gap under baseline and upside scenarios. Compared with the baseline projection, the upside scenario assumes additional investment of ¾ percent of GDP per annum, which boosts investment growth by about 1.4 percentage points on average from 2015 to 2020. Simulations suggest that the growth benefits from higher and more efficient investment could be up to 0.5–0.7 percentage point on real growth and potential growth by 2020. The cumulative impact on the level of GDP and potential GDP could be up to 1.3 to 1.5 percentage point by 2020.
Growth Benefits From Higher Investment
Sources: IMF staff estimates.
Level Impact From Higher Investment
Sources: IMF staff estimates.
The Growth Impact of Public Investment—a Model-Based Approach
14. We also use a model-based approach to assess the potential impact of the increased investment financed by EU funds. Specifically, we use the IMF’s Global Integrated Monetary and Fiscal model (GIMF) calibrated for the Romanian economy to simulate the impact of such an increase in investment on economic growth.6 The additional investment is modeled as a temporary increase for five years in government capital spending equal to 0.8 percent of GDP per annum and financed by EU grants.
15. The output elasticity of public capital is an important factor in our simulation. Various studies have attempted to gauge this elasticity by estimating a production function that includes the public capital stock as an input. Lighthart and Suarez (2011) conduct a meta-analysis and report the simple average of output elasticity of public capital of their meta sample at 0.2, compared to their meta-regression analysis estimate of 0.1. In Romania, the public infrastructure has a relatively low quality and, according to some estimates, is substantially below the productivity frontier range where incremental public investment contributes less and less to productivity. Hence, one can argue that in the case of the Romanian economy the output elasticity of public capital would be higher than 0.1. We use in our benchmark simulation an output elasticity of public capital of 0.2 and also simulate a scenario with a higher elasticity.
16. As a result of higher investment, real GDP would increase by around one percent initially with the impact peaking in 2025. After the shock in government spending, real GDP increases by approximately 1 percent on impact because of the increase in domestic demand, particularly investment. In the next periods, the real GDP is higher by around 1.5 percent. A simulation with higher output elasticity of public capital of 0.3 shows a bigger impact in case the government chooses the “right” investment projects. These estimates are broadly in line with estimates of the medium-term effect of SCF absorption on Bulgaria’s output of 1½–3 percent, using the GIMF model (Paliova and Lybek, 2014).
Source: IMF staff estimates.
17. Demand side projections of the effect of increased quality public infrastructure investment on output hinge on the robustness of projections of fiscal multipliers. However, there is little consensus on the size of fiscal multipliers in the economic literature, particularly for emerging and developing countries. Multipliers can vary over time and depend on a range of factors, including the state of the economy, the fiscal instrument used, and other discretionary policies, particularly monetary policy (Batini et al., 2014). Recent research, after reviewing large public investment booms in emerging and developing countries, suggests a public investment multiplier of about 1–1.3. Alternative approaches to estimating the public investment multiplier reviewed in the same study, indicate that a 1 percentage point of GDP increase in public investment has a multiplier of 0.5–0.9 over the medium term (IMF, 2014b).
18. The method of financing public infrastructure spending also influences the size of the multiplier and the potential impact on fiscal sustainability. Using a microfounded dynamic general equilibrium model, Varga and Veld (2010) calculated the impact multiplier of SCF funding for infrastructure investment in EU members to be around 0.5, after accounting for unproductive spending, import leakage, and possible crowding out of private spending to the extent that the increased investment leads to higher wages and real interest rates. Another study found the multiplier for projects funded by SCF in Romania to be in the range of 0.5–0.9 across different operational programs (Lungu, 2013).7 Based on an effective national cofinancing ratio of 25 in 2013,8 the fiscal multiplier effect of budgetary spending on EU co-financing would be significantly higher.
19. Moreover, the persistence of the multiplier plays a significant role in the impact of public investment on growth. Studies find that the output effect of an exogenous fiscal shock, in general, vanishes within 5 years even if fiscal measures are permanent. However, the effect of a permanent change in public investment may be permanent, with multipliers steadily rising toward their long-run values. Varga and Veld’s (2010) model produced a multiplier for all SCF financed investment that rises gradually to one after seven years and 2.6 after nearly two decades. Persistence is also likely to be stronger when economic activity is weak and/or monetary policy is accommodative (Batini et al., 2014).
|Increase in GDP growth rate||Cummulative increase in GDP|
|(Percentage points relative to baseline)|
20. Given the multiplier uncertainty, a range of multipliers were used to project the effect of increased quality infrastructure spending on growth. The increased quality of investment is modeled through an annual “shock” to expenditure on EU-funded projects by RON 1.7 billion per year relative to the baseline over the period 2016–20. The increase is assumed to be funded by SCF grants and cuts in national capital spending;9 thereby maintaining budget neutrality. The output effect of the fiscal shock (increased spending on EU-funded shocks) is projected to vanish after five years. This is modeled through a persistence factor. The increase in the GDP growth rate in 2020 relative to the baseline ranges from 0.3 to 1.2 percentage points. The cumulative change in GDP in 2025 relative to the baseline ranges from 1.9–6.2 percent. The analysis is conservative in that it does not account for topping up EU-reimbursements, which are roughly equivalent to 10 percent of the additional expenditure, nor the dynamic affect on productivity of more and higher quality investment and infrastructure.
F. Considerations and Recommendations
21. Resources for quality public investment must contend with other spending priorities and an overall budget constraint. Shifting state- and local-government capital spending toward projects financed by the European Union SCF is an obvious way to increase quality investment while maintaining fiscal discipline. The 2015 budget moves in this direction but more funds could be diverted. At the same time, the maintenance costs need to be adequately reflected in medium-term budget projections. More broadly, Romania’s strategic framework should be improved and made more predictable, which would help facilitate alignment of EU and national strategies and serve to preserve project priorities over the medium term.
22. The quality of project selection, management, and implementation plays a crucial role in determining the return on investment. While some progress has been made in prioritizing spending on significant investment projects, responsibility for implementation of all phases of a public investment’s project life cycle remains with sponsoring line ministries. The MoPF should consider undertaking an assessment of the effectiveness of the project selection methods and take corrective measures as necessary. In addition, the MoPF should mainstream the good management practices of EU-funded programs (project design and monitoring and evaluation) across all projects. Over the medium term, the scope of projects overseen by the technical team should grow to encompass most central and local-government projects. Specific measures that could be taken during 2015 to improve project management include:
Amend Government Emergency Ordinance (GEO) 88/2013 to align the project prioritization process with the budget cycle timeline.
Amend the Public Finance Law to include the substantive elements of GEO 88/2013 and to ensure the prioritization system covers all projects over time.
Upgrade the database of central-government projects to allow for a cost/benefit approach to project selection and better monitoring and encourage more consistent reporting by line ministries of project values and costs.
Update, approve, and publish a project prioritization list prior to commencement of the 2016 budget process, and, within 15 days after approval of the 2016 budget by parliament, update the published list with actual amounts allocated to each project in the 2016 budget and estimated expenditures over the medium term.
23. Romania’s political and legal environments also influence the allocation of scarce budget resources for public investment. The perception exists, borne out to some extent by the average length of time it takes to complete projects and the number of projects that remain incomplete, that project selection in many cases is driven by political considerations rather than economic and financial ones. At the same time, the significant shortfall in capital spending in 2014, amid evidence of increased vigilance in tackling governance issues by Romania’s anti-corruption agency, suggests a possible reluctance of officials to sign-off on projects. To partly address the former concern, the PIEU could be empowered to reject projects based on the existing prioritization methodology.
24. Finally, SOEs in Romania play a critical role in infrastructure development, maintenance, and service provision. Further reforms of these enterprises will be important to realize the full benefit of increased quality public investment in Romania.
|Name||Value||Remaining amount||Starting year||Estimated completion year|
|Non-reimbursable funds (EU grants)||49.7||30.0|
|o/w top 3 projects of 36 total|
|Railway rehabilitation Brasov-Simeria||5.0||3.5||2007||2018|
|Reimbursable funds (loans)||7.9||5.7|
|o/w top project of 10 total|
|Metro line to Otopeni Airport||5.3||5.3||2009||2019|
|State-budget and pre-accession funds||55.5||34.6|
|o/w top 6 projects of 54 total|
|Railway rehabilitation Bucuresti-Constanta||2.1||0.1||2006||2014|
|Bucharest by pass roadway||3.2||3.2||2008||2020|
|Ministry of Economy building||0.1||0.1||2004||2016|
Alexeeva, V., C.Queiroz and S.Ishihara, 2011, Monitoring Road Works Contracts and Unit Costs for Enhanced Governance in Europe and Central Asia World Bank, Washington, DC.
Anderson, D., Hunt, B., Kortelainen, M., Kumhof, M., Laxton, D., Muir, D., Mursula, S., and S.Snudden, 2013, “Getting to Know GIMF: The Simulation Properties of the Global Integrated Monetary and Fiscal Model,” IMF Working Paper 13/55, Washington, DC.
Batini, N., Eyraud, L., Fomi, L., and A.Weber, 2014, “Fiscal Multipliers: Size, Determinants, and Use in Macroeconomic Projections:IMF Technical Notes and Manuals, Washington, DC.
Dabla-Norris, E., Brumby, J., Kyobe, A., Mills, Z. and C.Papageorgiou, 2011, “Investing in Public Investment: An Index of Public Investment Efficiency,” IMF Working Paper 11/37, Washington, DC.
European Commission, 2014, Summary of the Partnership Agreement for Romania, 2014–2020, Brussels.
European Commission-Romania Partnership Agreement, 2014, http://www.fonduriue.ro/res/filepickerusers/cd25a597fd-62/2014-2020/acord-parteneriat/Partnership_Agreement_2014RO16M8PA001_1_2_ro.pdf.
International Monetary Fund, 2013, “Republic of Poland: 2013 Article IV Consultation,” IMF Country Report No. 13/219, Washington DC.
International Monetary Fund, 2014a, “Romania: First and Second Reviews Under the Stand-By Arrangement and Requests for Waiver of Nonobservance of a Performance Criterion, Modification of Program Conditionality, and Rephasing of the Availability Date of Purchases,” IMF Country Report No. 14/87, Washington, DC.
International Monetary Fund, 2014b, Is it Time for an Infrastructure Push? The Macroeconomic Effects of Public Investment, World Economic Outlook, Washington, DC.
KPMG, 2014, EU Funds in Central and Eastern Europe Progress Report 2007–2013.
Kumhof, M., Laxton, D., Muir, D., and S.Mursula, 2010, “The Global Integrated Monetary and Fiscal Model (GIMF) – Theoretical Structure,” IMF Working Paper 10/34, Washington, DC.
Lighthart, E. and M.Suarez, 2011, “The Productivity of Public Capital: A Meta-Analysis” in Infrastructure Productivity Evaluation, Manshanden, W., and Jonkhoff, W., (eds.), SpringerBriefs in Economics.
Lungu, L., 2013, The Impact of EU Funds on Romanian Finances, Romanian Journal of European Affairs, Vol. 13, No. 2, June2013.
Paliova, I. and T.Lybek, 2014, “Bulgaria’s EU Funds Absorption: Maximizing the Potential!,” IMF Working Paper 14/21, Washington, DC.
Romania, 2011, Priority Action Plan for Increasing the Capacity to Absorb Structural and Cohesion Funds, Bucharest.
Romania, 2014, Romanian Partnership Agreement for the 2014–2020 Programming Period, Bucharest.
Stoian, A., 2012, “The Macroeconomic Effects of Fiscal Policy in Romania,” Presentation.
Varga, J. and J.Veld, 2010, “The Potential Impact of EU Cohesion Policy Spending in the 2007–13 Programming Period: A Model-Based Analysis,” Economic Papers (Brussels: European Commission).
World Bank, 2012, “Assistance to the Ministry of Transport and Infrastructure to Strengthen Strategic Planning in the Transport Sector Component 3: Road Construction Costs,” Washington, DC.
Prepared by John Ralyea, Vahram Stepanyan, and TengTeng Xu (all European Department).
EU funds typically refer to grants received from the European Union for development and income support purposes. Funds are allocated in seven-year programming periods. The last one covered the period 2007–13. The current programming period covers 2014–20. Total EU funds available to Romania during 2007–13 were EUR 34.6 billion, which included support from Structural and Cohesion Funds (SCF), European Agricultural Fund and Rural Development, European Fishery Fund, and European Guarantee Fund. This section and the analysis in this paper focus only on financing from the SCF.
SCF include the European Fund for Regional Development (EFRD), the European Social Fund (ESF), and the Cohesion Fund (CF).
The projects have to be approved by 2013, but the last payments can be made in 2015 (the N+2 rule).
Emergency Ordinance 88/2013 and Government Decision 225/2014.
The GIMF is a multi-region, forward-looking, DSGE model developed by the IMF for policy analysis and international economic research. For details on GIMF, see Kumhof et al. (2010), and Anderson et al. (2013).
This is consistent with Stoian’s (2012) estimate of a short-term multiplier of 0.5 for Romanian government investment.
National co-financing ranges from 15 percent for ERDF and ESF projects to 20 percent for CF projects. National authorities also have to finance the VAT and 90 percent of the cost of land acquisition for infrastructure projects. Romania also receives top-up financing from the European Union of up to 10 percent for SCF projects. Taking into consideration all these factors, as well as corrections, the average effective national co-financing rate of SCF projects was 25 percent in 2013.
EU-reimbursements for projects sponsored by private beneficiaries (e.g., NGOs) are passed onto the private beneficiary while the associated national co-financing is recorded on the budget. As a result, only about 60–70 percent of total EU-funded project expenditures incurred on the budget are covered by EU-reimbursements.