Document of
The World Bank
FOR OFFICIAL USE ONLY
Report No: ICR00004866
IMPLEMENTATION COMPLETION AND RESULTS REPORT
IBRD — 82610
ON A
LOAN
IN THE AMOUNT OF EURO 70 MILLION
(US$ 91.8 MILLION EQUIVALENT)
TO
Romania
FOR THE
Revenue Administration Modernization Project
03 October 2019
Governance Global Practice
Europe And Central Asia Region
CURRENCY EQUIVALENTS
Exchange Rate Effective: September 16, 2019
Currency Unit = Romanian Leu (RON)
RON 4.29 = US$1
US$0.23 = 1 RON
FISCAL YEAR
January – December
Regional Vice President: Cyril E. Muller
Country Director: Arup Banerji
Regional Director: Lalita M. Moorty
Practice Manager: Roberto Adrian Senderowitsch
Task Team Leader(s): Moses Sabuni Wasike
ICR Main Contributor: Runyararo Gladys Senderayi
ABBREVIATIONS AND ACRONYMS
CIT Corporate Income Tax
CNIF National Center for Financial Information
COTS Customized-off-the-shelf
CPF Country Partnership Framework
CPS Country Partnership Strategies
EU European Union
FM Financial Management
GDP Gross domestic product
GoR Government of Romania
HNWI High Net Wealth Individuals
ICR Implementation Completion and Results Report
ICT Information and Communications Technology
IMF International Monetary Fund
IT Information Technology
M&E Monitoring and Evaluation
MoPF Ministry of Public Finance
NAFA National Agency for Fiscal Administration
PAD Project Appraisal Document
PDO Project Development Objective
PIT Personal Income Tax
PMU Project Management Unit
RAMP Revenue Administration Modernization Project
RARP Revenue Administration Reform Project
RMS Revenue Management System
SPV Virtual Private Space
TAXGIP Tax Administrators eXchange for Global Innovative Practice
TTL Task Team Leader
VAT Value added Tax
TABLE OF CONTENTS
DATA SHEET .......................................................................................................................... 1
I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 5
A. CONTEXT AT APPRAISAL .........................................................................................................5
B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) .......................................9
II. OUTCOME .................................................................................................................... 10
A. RELEVANCE OF PDOs ............................................................................................................ 10
B. ACHIEVEMENT OF PDOs (EFFICACY) ...................................................................................... 11
C. EFFICIENCY ........................................................................................................................... 17
D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 19
E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 19
III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 21
A. KEY FACTORS DURING PREPARATION ................................................................................... 21
B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 22
IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 26
A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 26
B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 28
C. BANK PERFORMANCE ........................................................................................................... 29
D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 31
V. LESSONS AND RECOMMENDATIONS ............................................................................. 32
ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 35
ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 50
ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 52
ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 53
ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 56
ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) ..................................................................... 58
The World Bank
Revenue Administration Modernization Project (P130202)
DATA SHEET
BASIC INFORMATION
Product Information
Project ID Project Name
P130202 Revenue Administration Modernization Project
Country Financing Instrument
Romania Investment Project Financing
Original EA Category Revised EA Category
Not Required (C) Not Required (C)
Organizations
Borrower Implementing Agency
Romania, Romania National Agency for Fiscal Administration (NAFA)
Project Development Objective (PDO)
Original PDO
The development objectives of the project are (i) to increase effectiveness and efficiency in collection of taxes and
social contributions; (ii) to increase tax compliance; and (iii) to reduce the burden on taxpayers to comply.
PDO as stated in the legal agreement
The objectives of the project are to: (a) increase effectiveness and efficiency in the collection of taxes and social
contributions; (b) increase tax compliance; and (c) reduce the burden on taxpayers to comply.
Page 1 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
FINANCING
Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$)
World Bank Financing
91,800,000 23,734,072 19,275,789
IBRD-82610
Total 91,800,000 23,734,072 19,275,789
Non-World Bank Financing
0 0 0
Borrower/Recipient 220,000 0 0
Total 220,000 0 0
Total Project Cost 92,020,000 23,734,072 19,275,789
KEY DATES
Approval Effectiveness MTR Review Original Closing Actual Closing
26-Apr-2013 09-Aug-2013 21-Mar-2016 31-Mar-2019 31-Mar-2019
RESTRUCTURING AND/OR ADDITIONAL FINANCING
Date(s) Amount Disbursed (US$M) Key Revisions
01-Nov-2016 11.04 Change in Loan Closing Date(s)
25-Jan-2019 19.26 Change in Loan Closing Date(s)
Cancellation of Financing
Reallocation between Disbursement Categories
KEY RATINGS
Outcome Bank Performance M&E Quality
Unsatisfactory Moderately Unsatisfactory Modest
Page 2 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
RATINGS OF PROJECT PERFORMANCE IN ISRs
Actual
No. Date ISR Archived DO Rating IP Rating Disbursements
(US$M)
01 01-Sep-2013 Satisfactory Satisfactory 0
02 19-Apr-2014 Satisfactory Satisfactory 0
03 25-Nov-2014 Satisfactory Moderately Satisfactory 0
04 07-Jun-2015 Moderately Satisfactory Moderately Unsatisfactory 1.30
05 07-Aug-2015 Moderately Satisfactory Moderately Unsatisfactory 4.23
Moderately
06 22-Feb-2016 Moderately Unsatisfactory 8.68
Unsatisfactory
07 23-May-2016 Moderately Satisfactory Moderately Satisfactory 9.68
08 16-Jan-2017 Moderately Satisfactory Moderately Satisfactory 15.57
09 18-Jun-2017 Unsatisfactory Unsatisfactory 16.70
10 06-Feb-2018 Unsatisfactory Unsatisfactory 19.18
11 03-Oct-2018 Unsatisfactory Unsatisfactory 19.18
SECTORS AND THEMES
Sectors
Major Sector/Sector (%)
Public Administration 100
Central Government (Central Agencies) 100
Themes
Major Theme/ Theme (Level 2)/ Theme (Level 3) (%)
Economic Policy 20
Fiscal Policy 20
Tax policy 20
Page 3 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Public Sector Management 80
Public Finance Management 20
Domestic Revenue Administration 20
Public Administration 60
Administrative and Civil Service Reform 5
E-Government, incl. e-services 55
ADM STAFF
Role At Approval At ICR
Regional Vice President: Philippe H. Le Houerou Cyril E Muller
Country Director: Mamta Murthi Arup Banerji
Director: Yvonne M. Tsikata Lalita M. Moorty
Practice Manager: Adrian Fozzard Roberto Adrian Senderowitsch
Task Team Leader(s): C. Bernard Myers Moses Sabuni Wasike
ICR Contributing Author: Runyararo Gladys Senderayi
Page 4 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES
A. CONTEXT AT APPRAISAL
Context
Country Context
Like many countries, in 2013, Romania was still reeling from the negative effects of the global economic and
financial crisis. It was among the poorest countries in the European Union (EU) with a Gross Domestic Project (GDP)
per capita of US$9,585. The crisis hit Romania hard, slowing convergence and necessitating changes to her growth and
reform strategy to protect the gains of almost a decade of rapid growth and poverty reduction. The focus was on
improved economic management and finalizing public sector and governance reforms. Romania was committed to
deepening structural reforms and strengthening public administration in order to reap the benefits of EU membership.
Sectoral and Institutional Context
Propelled by the EU accession process, when the project was appraised in 2013 Romania’s tax sys tem had
already undergone substantive changes. The system was fairly well-designed and supportive of growth. It was broad-
based with low rates and few exemptions.1 Much effort had been put into improving the administration of the system
internally and the efficiency of tax collection, especially from a technological perspective.
Still, tax revenue collection remained low and Romania stood out in the EU for having a very low tax-to-GDP
ratio. In 2013, Romania’s revenue envelope had shrunk, dropping to 27.3 percent of GDP from 28.1 percent of GDP in
2007 when Romania became an EU Member State. The tax administration system was built around an outdated
business model, supported by a heavy network of local offices spread countrywide.2 From an information technology
(IT) perspective, the software in use was obsolete and out of step with ongoing technological advances. Consequently,
the system was strained and not operating at capacity. Endeavors to strengthen compliance were floundering; despite
a robust policy framework, in 2013, Romania’s estimated compliance gap was the highest in the EU. This was
exacerbated by the suboptimal operational model and poor infrastructure.
Recognizing the need to improve the situation, modernization of Romania’s revenue administration was
embedded into the Government’s economic reform program. The Government placed importance on having an
effective tax administration that could encourage voluntary compliance and deal with tax evasion in a more
professional and technology aided environment. Political support for revenue administration modernization was
broad based and spanned different political parties and government stakeholders. This was important because
previous efforts to pursue comprehensive reforms were unsuccessful for lack of consensus. Under its 2011-2014
1 According to the findings of the IMF’s tax policy diagnostic from 2011, “… after the 2008 crisis, the tax structure became more supportive of
economic growth, with successive tax rate cuts and other reforms to growth-harmful taxes such as social security contributions and corporate
income tax”.
2 At the time of appraisal, the National Agency for Fiscal Administration had over 400 such offices and the majority of taxpayers paid their income
taxes in person and often in cash.
Page 5 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Strategic Plan, the National Agency for Fiscal Administration’s (NAFA) medium term objective was to transform itself
into a revenue administration with a streamlined network of field offices, efficient business processes based on self-
assessment, supported by risk-based compliance management, and skilled and professional staff, all of which would
contribute to a taxpayer-friendly and client-oriented environment.
Bank Rationale
The World Bank’s support for tax administration reform and improvements in tax efficiency in Romania was
well-established and spanned two Country Partnership Strategies (CPS). The cornerstone of the Bank’s program was
to identify reforms to build a stronger public administration. This included a functional review of the Ministry of Public
Finance (MoPF) and efforts to improve communication with tax payers in the framework of the 2009-2013 CPS (Report
No. 60255-RO) which was designed around three pillars: (i) reforming the public sector, (ii) resuming growth and
enhancing competitiveness, and (iii) promoting social and spatial inclusion. In the second half of the CPS cycle, the
focus shifted to supporting the Government to implement the identified reforms and this was continued under the
2014-2017 CPS (Report No. 84830-RO) which identified improvements in revenue and expenditure administration as
a key objective. Prior to that, the World Bank had conducted a significant amount of diagnostic work in preparation
for the Revenue Administration Reform Project (RARP) in 2005. The project was designed to support the
implementation of a sustainable revenue collection system in Romania, but was subsequently dropped, although some
of the planned activities were undertaken under the Public Sector Institution Building Loan.
Given the strategic importance of tax administration reform in Romania and the need to accelerate
convergence with other EU Member States, the rationale underpinning the Bank’s investment was strong. The
Revenue Administration Modernization Project (RAMP) was fully embedded in the Government’s own reform priorities
and leveraged the progress made through investments from other donors. The project was a natural follow up to the
extensive work undertaken over the years to support NAFA in its pursuit of efficiency and efficacy in order to facilitate
tax compliance. Furthermore, given the Bank’s role in supporting Romania to weather the financial crisis and its deep
sectoral knowledge and reform experience in Romania, but also in the Europe and Central Asia region and beyond, the
Bank emerged as a strategic and trusted advisor.
Theory of Change (Results Chain)
The Project Appraisal Document (PAD) did not delineate the theory of change for the RAMP as this was not
required by Bank procedures when the project was prepared. However, the logic behind the operation was clear and
can be summarized as follows:
Addressing weaknesses in tax administration would help to improve tax efficiency and narrow compliance gaps. This
would be achieved in two phases. First, by supporting NAFA’s existing revenue collection capacity and developing
detailed reforms for each operational and managerial area of NAFA, as well as a modern revenue management system
(RMS). Second, by implementing the designed reforms and the RMS to support it in a modern information and
communication technology (ICT) environment. The result would be a modernized revenue administration with business
processes and workflows befitting an EU Member State. In the long run, this would contribute to improving the business
environment and enhancing Romania’s attractiveness to private sector investment.
Page 6 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Page 7 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Project Development Objectives (PDOs)
The PDO in the Loan Agreement was to: (a) increase effectiveness and efficiency in the collection of taxes and
social contributions; (b) increase tax compliance; and (c) reduce the burden on taxpayers to comply.
Key Expected Outcomes and Outcome Indicators
The key expected outcomes and their indicators are reflected in Figure. 2.
Figure 2. Key Expected Outcomes and Outcome Indicators
Outcomes Outcome Indicators
Effectiveness in revenue collection – measured as the rate of tax arrears
Increased effectiveness in the
collection of taxes and social
Effectiveness in revenue collection – measured by productivity rate for Value
contributions
Added Tax (VAT) and Corporate Income Tax (CIT)
Increased efficiency in the collection Efficiency in revenue collection – measured as the revenue collected per NAFA staff
of taxes and social contributions
Increased tax compliance Compliance as measured by the tax gap for VAT, CIT and Personal Income Tax (PIT)
Reduced burden on taxpayers to Compliance cost – measured as the average time for preparation of tax returns of
comply large, medium and small corporate taxpayers
Components
The RAMP had three substantive components, supported by a project management component:
Component 1: Institutional Development (Cost US$ 9.00 million). Activities included supporting the
development of a modern organization and management structure, instilling strategic focus, emphasizing the
importance of integrity, rationalizing the organization structure to increase NAFA’s effectiveness, and strengthening
human resource management through targeted professional training and technical assistance. This component was
also intended to improve the analytical capacity required to support the strategic management of NAFA and support
an improved legal framework and appeal process and review all business procedures in these areas and reengineer
the operational business processes to institute a modern revenue administration at par with best practice in the EU.
Component 2: Increasing Operational Effectiveness and Efficiency (Cost US$ 75.69 million). This component
aimed to streamline business processes and work-flows and invest in the modernization of core IT systems to take full
advantage of the possibilities offered by automation in NAFA’s back office operations, as well as in the interaction with
taxpayers and contributors. The component would also strengthen the effectiveness of audit and enforcement
functions, including the collection of arrears. This would be achieved by: (i) modernizing tax audit functions,
intelligence and preventive functions; (ii) upgrading IT management function within NAFA to manage all NAFA IT
Page 8 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
resources countrywide, including staff, following office consolidation; (iii) implementing electronic document
management and archiving; (iv) developing and implementing an integrated RMS; and (v) upgrading NAFA’s
information and communications technology (ICT) infrastructure, including developing a centralized data processing
center, data warehouse and a business continuity center capable of sustaining all NAFA operations with no data loss
in case of a disaster.
Component 3: Taxpayer Services and Corporate Communication (Cost US$ 3.83 million). Activities included
modernizing taxpayer services; implementing mechanisms for transparency and accountability, such as consultative
technical panel and community surveys, to ensure fair community input; modernizing service delivery channels to
provide advanced services to taxpayers; promoting community understanding of revenue laws, procedures and rights
and obligations of taxpayers; promoting taxpayer efficiency through simplification and provision of electronic services;
and strengthening external and internal communication.
Component 4: Project Coordination and Management (Cost US$ 3.31 million). A Project Management Unit
(PMU) was put in place to manage the day-to-day implementation of the project, including ensuring integration across
components. The PMU also comprised a Change Management Team responsible for managing the change process and
coordinating institutional change.
B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE)
Revised PDOs and Outcome Targets
The PDO and the outcome targets remained unchanged for the duration of the project.
Revised PDO Indicators
The PDO indicators remained unchanged for the duration of the project.
Revised Components
The project components and sub-components remained the same for the duration of the project.
Other Changes
The RAMP was restructured twice. In 2016, following a request from the MoPF, on behalf of NAFA, the project
was extended by two years to ensure sufficient time to implement the revenue management system. The closing date
was changed from 31 March 2019 to 31 March 2021. In November 2018, the Bank received an official request from the
MoPF, on behalf of NAFA, to cancel Euro 51,902,124 of the loan amount, effectively cancelling the undrawn balance of
the loan that was not committed. On December 13, 2018, the Bank received a supplemental request to establish March
31, 2019 as the revised project closing date.
Page 9 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Rationale for Changes and Their Implication on the Original Theory of Change
The request to extend the project in August 2016 came on the back of implementation delays in the
procurement of the customized off-the-shelf (COTS) solution for the RMS. The request was granted to ensure sufficient
time to procure and implement the RMS; the linchpin in modernizing Romania’s tax administration system, and a key
aspect in achieving the project’s objectives. The extension was therefore instrumental to fulfilling the theory of change.
The subsequent request to cancel the project was based on changes in circumstances and approach in pursuing
implementation of the national revenue management system, a core part of the project. Specifically, the Government
cited the following reasons for the cancellation: (i) there had been numerous changes to tax legislation that could not
be accommodated in the revenue management system as designed3; (ii) procurement of the revenue management
system was significantly delayed and too slow to match NAFA’s urgent and priority information technology system
needs; (iii) the implementing entity, NAFA, did not have sufficient capacity to implement the project; (iv) NAFA
management was unwilling to sign off on procurement of the revenue management system owing to likely retribution
by state auditors – the Romanian Court of Accounts; and (v) Government was in the process of reorganizing its
information technology support structures, including replacing NAFA’s IT department with a new institution, the
National Center for Financial Information (CNIF) under the MoPF. As a result of the cancellation the project was unable
to fulfill its development objectives.
A detailed discussion about the cancellation can be found in Section III. B under Key Factors During Implementation.
II. OUTCOME
A. RELEVANCE OF PDOs
Assessment of Relevance of PDOs and Rating
Rating: Modest
At the time of this Implementation Completion and Results Report (ICR), the relevance of the PDO was modest.
All told, the purpose of the RAMP was to strengthen NAFA’s institutional capacity which is consistent with the overall
focus of Romania’s Country Partnership Framework (CPF) for the period FY19 – 23 (Report No. 126154-RO) to build
institutions fit for a prosperous and inclusive Romania. Specific priority areas include; (i) ensuring equal opportunities
for all; (ii) catalyzing private sector growth and competitiveness; and (iii) building resilience to shocks. The longer-term
vision of the RAMP was to improve the business environment and enhance Romania’s attractiveness to private sector
investment in line with the second priority area of the CPF. That said, support for revenue and expenditure
administration are not central to the current CPF demonstrating a clear shift from the previous CPF under which efforts
to procure an RMS had already started to fumble “due to the loss of the original reform champion early on in
3During ICR discussions, NAFA officials further explained that any future modifications to the (COTS) RMS would have significant cost
implications.
Page 10 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
implementation and the inability to rebuild the commitment”4 with several changes in Ministers of Public Finance and
Presidents of NAFA.
Enhancing business processes and implementing and operationalizing new IT infrastructure in Romania continue
to be key priorities. However, the Government’s approach to revenue administration modernization has changed.
Romania’s National Reform Program (NRP) for 2019 cites the need to improve tax collection through modernization, but
not necessarily through the implementation of an integrated RMS.
The PDO had three sub-objectives, which would be achieved through a combination of analytical work and the
implementation of an RMS. Specifically, substantive achievements of the sub-objectives to increase effectiveness and
efficiency in the collection of taxes and social contributions and increase tax compliance would only occur after the
operationalization of the COTS procured RMS. As the desire to implement the COTS system faded (more on this in
subsequent sections of the ICR), the relevance of the PDO, to those who were no longer keen on the implementation of
the COTS system, also diminished. The “all-or-nothing” way in which the achievement of the sub-objectives was
designed (in terms of their dependence on the implementation of the RMS) limited the relevance of, and scope to
achieve, the PDO in the face of changes in priorities and shifts in the political context. Failure to revise the PDO to keep
the objectives fully relevant when circumstances changed precludes a higher rating.
A stand-alone investment loan was the right instrument given the project’s focus on the procurement of
revenue management software and ICT infrastructure.
B. ACHIEVEMENT OF PDOs (EFFICACY)
The RAMP was designed to comprehensively transform the way NAFA conducted its business by modernizing,
standardizing and integrating its business processes and leveraging IT in a way that would increase effectiveness and
efficiency in the collection of taxes and social contributions, increase tax compliance and reduce the compliance burden.
Although there were certain achievements that enhanced NAFA’s performance, the project did not achieve all its
objectives. Project interventions included: business process re-engineering, reorganizing key functions and establishing
a performance framework for NAFA, purchasing and installing 80 new licenses and providing enhanced training, creating
an annual tax audit implementation plan, adopting new procedures to increase the transparency of the audit process,
improving the quality of audit and legal resources that deal with high net wealth individual taxpayers’ appeal cases and
training courses on strategic management, change management and project management. These resulted in, among
other things, the partial consolidation of local offices,5 improved audit capacities, improved high net wealth individual
taxpayers’ monitoring and legal frameworks, and numerous departmental skills and capacity improvements.
The procurement of the revenue management software was the fulcrum of NAFA’s transition to an EU level
revenue administration. However, the project was cancelled before the software was procured although the
procurement had already been started with the technical evaluation of submitted bids almost completed.6 It is also
worth noting that even before the project was cancelled, progress towards achieving the PDO was limited and the RAMP
4 Cited from the Romania Country Partnership Framework for the period FY19-23.
5 Which only partially reached the target agreed to.
6 Deviations in the technical specifications had not yet been clarified and agreed.
Page 11 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
struggled to graduate from the analytical phase to the implementation phase, which is where the fundamental reforms
would have taken place.
Table 1. PDO Performance Indicators
PDO indicators Baseline Target Actual Status
2013 2018 2018
Effectiveness in revenue collection
Tax arrears recovered/stock of tax arrears, % 40.0 50.0 33.2 not achieved
VAT productivity, % 35.8 46.0 34.9 not achieved
CIT productivity, % 11.9 21.0 10.3 not achieved
Efficiency
Revenue collected per NAFA staff, thousand RON 6,310 8,715 9,147 exceeded
Compliance rate
VAT gap 36.2 26.2 36.0* not achieved
CIT gap 45.6 40.6 29.9* exceeded
PIT gap 41.5 36.5 43.4* not achieved
Compliance burden
Average time for preparation of tax returns, hours 224 176 163 exceeded
*Data refer to 2017 as 2018 information was not available at the time of ICR preparation.
Page 12 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Assessment of Achievement of Each Objective/Outcome
Achievement of Objective 1: Increased Effectiveness and Efficiency in the Collection of Taxes and Social
Contributions
Rating: Modest
Increased effectiveness in the Collection of Taxes and Social Contributions
Project interventions that were supposed to increase effectiveness in revenue collection included supporting
the modernization and enhancement of businesses processes and workflows through: the development, improvement
and implementation of a core revenue management system, data warehouse, and backup, security and disaster
recovery capacity; strengthening the effectiveness of audit and enforcement functions, improving documentary control,
and establishing mechanisms for registration, declaration and payment processing.
Without the completion of one of the key activities under the project — the RMS implementation —the project
was not successful in increasing effectiveness in revenue collection. In fact, instead of improving, effectiveness,
measured as the rate of recovery of tax arrears and VAT and CIT productivity rates, deteriorated. In each case, the
project failed to meet the stipulated targets. Although there were other factors that influenced the non-achievement
of the targets (each discussed in turn below), the failure to implement the RMS contributed to the project’s inability to
increase effectiveness in revenue collection.
For the rate of recovery of tax arrears, the baseline was 40% and the target was 50%, yet in 2018 the rate was
33.2%, backsliding from 36.7% at the mid-point of the project in 2015. Recommendations for improvements to arrears
management were developed under the project but some of them required legislative changes, which did not occur
during project implementation. For example, in the absence of clear criteria for the write-off of irrecoverable tax
arrears, amounts are retained in accounting records and tax officers are held personally liable for write-off decisions
queried by the Romanian Court of Accounts. As a result, arrears continued to accumulate and negatively impact the
effectiveness of revenue collection. In the circumstances, it can be concluded that the likely positive impacts of project
interventions were, to an indeterminate extent, negated by weak legislation and management of tax arrears.
The target for VAT productivity was 46%, but in 2018, the productivity rate was 34.9% — lower than the baseline
(35.8%). As presented in figure 3, compared to similar economies, while Romania registered the lowest VAT weighted
average rate of 14.5%, it ranked lowest among EU Member Countries, collecting 6.4% of GDP.
Page 13 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Figure 3. VAT revenues in 2018 compared with the previous year (% of GDP)
In 2018, the CIT productivity rate was 10.3% — lower than the baseline (11.9%) and far from the envisaged
target of 21%. A 2019 “Impact of the change in taxation of company earnings in Romania between 2015-2018” study
by the Romania Fiscal Council7 observes that between 2015 and 2018, there were considerable legislative changes
related to corporate income tax. These changes substantially changed the scope of firms included in the
microenterprises’ category, the tax rate applied to them and, implicitly, the budget receipts coming from this type of
taxes. The threshold for the annual income up to which a company is included in the microenterprises’ category
increased by more than 15 times, equivalent to the tax base shifting towards revenues rather than profits, while
diminishing and differentiating the tax rates. In addition, for companies operating in hotels, restaurants, cafes, a specific
tax was introduced, replacing the profit tax. The study drew on information from over 70% of financial statements
submitted to the MoPF by non-financial sector firms operating in Romania. It concludes that the significant reduction
in the number of profit tax payers negatively affected aggregate budget receipts by an estimated 1.5-2 billion RON
(equivalent to 10 to 13% of 2018 gross corporate tax revenues) over the 4 years. It can therefore be concluded that if
the legislative changes had not occurred, the project’s interventions were likely to have contributed to an increase in
CIT productivity.
Increased efficiency in the Collection of Taxes and Social Contributions
By reorganizing NAFA, streamlining business processes and workflows (e.g., establishing a legal basis for the
electronic transmission of garnishment notices, upgrading the management of the HNWI Unit by splitting it into high
wealth individuals directorate and high-risk individuals directorate, adopting enhanced appeals procedure), and
modernizing IT systems (excluding the RMS) to establish high volume processing, the project supported improvements
7The Fiscal Council is an independent authority established by the Fiscal Responsibility Law No. 69/2010 (FRL), which aims to support the
Government and the Parliament in designing and implementing the fiscal policy and to promote the transparency and sustainability of
public finances.
Page 14 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
in increasing efficiency in revenue collection, measured as revenue collected per NAFA staff. This is one of the project’s
achievements that if sustained, will have a measurable benefit to Romanian taxpayers for years to come. By 2018,
targeted improvements in increasing efficiency in revenue collection had been exceeded by 5%. The target was RON
8,715, starting from a baseline of RON 6,310, and the project achieved RON 9,147. The institutional reorganization
included staff rationalization that reduced the number of positions from 28,904 to 27,100. Other reforms included
organization of the territorial tax administration in 8 regional directorates of public finances, abolition of the financial
guard, and a reallocation of positions to subordinated territorial units and understaffed functions (antifraud, tax
inspection, IT, legal).
Achievement of Objective 2: Increased Tax Compliance
Rating: Modest
Improvements in increasing efficiency in revenue collection did not necessarily strengthen tax compliance. The
project aspired to reduce the VAT gap from 36.2 to 26.2. However, in 2018, the VAT gap was 36.0 resulting in Romania
continuing to have the largest VAT gap in EU, a phenomenon that had been in existence for several years. Furthermore,
according to a 2018 report8, Romania was one of the few countries that saw a widening of the tax gap in 2016. It is also
worth noting that the Organization for Economic Cooperation and Development and International Monetary Fund (IMF)
calculations9 indicate that tax efficiency in Romania, especially for VAT was lagging. This could also explain the large
compliance gap. Considering factors outside the scope of the project, as of January 2016, Romania reduced its standard
VAT rate from 24% to 20%. This change had a substantial impact on the effective rate which declined from 17.2% to
13.5%. Despite the reduction of the rate, VAT non-compliance increased and the VAT gap widened.
Instead of decreasing, the PIT gap increased from 41.5 to 43.4 and remained out of range of the target which
was 36.5. There were also factors beyond the control of the project’s activities that contributed to the increase in the
PIT gap. According to the 2018 Romania Fiscal Council Annual report, legislative changes resulted in the standard PIT
rate decreasing from 16% to 10% resulting in reduced revenue.
However, there was a reduction in the compliance gap for CIT which declined from 45.6 to 29.9, exceeding the
set target (40.6) by 15%. Activities that were intended to increase tax compliance included modernization and
enhancement of business processes and workflows; including an RMS and IT infrastructure, upgrading systems for
internal alerts, installing and using risk compliance and VAT fraud software, establishing a taxpayer assistance
Monitoring Unit and call-centers, delivering educational campaigns on tax awareness, implementing electronic
garnishment of debtor’s bank accounts and electronic communication for all administrative acts of enforcement , and
moving from face-to-face to electronic payments. Of these planned activities, the implementation of an RMS did not
take place.
Achievement of Objective 3: Reduced Burden on Taxpayers to comply
Rating: Substantial
8 Study and Reports on the VAT Gap in the EU-28 Member States: 2018 Final Report, TAXUD/2015/CC/131
9 https://www.imf.org/en/Publications/CR/Issues/2018/06/06/Romania-Selected-Issues-45944
Page 15 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
The compliance burden was significantly reduced, exceeding the set target. The average time to prepare tax
returns was 224 hours at the beginning of the project. In 2018, when the project closed, the average time was 163
hours, considerably less than the target (176 hours). The broad and complex process of simplifying the tax system and
reducing bureaucracy has taken place gradually, and recognition of this progress has been highlighted in annual Paying
Taxes reports issued by PricewaterhouseCoopers (PwC)10 and the World Bank's Doing Business reports. Project
interventions included modernizing taxpayer services, establishing mechanisms for transparency and accountability,
promoting community understanding of tax revenue laws, developing NAFA’s taxpayer services program and reviewing
the charter of taxpayer rights. Specifically, NAFA made the electronic filing of declarations by taxpayers mandatory (by
the end of the project the electronic filing rate by large and medium size firms was 99%). Taxpayer services were also
improved through a multi-channel approach which included face-to-face interaction, post, emails, phone etc. providing
taxpayers with various options to engage and seek assistance. Eight call centers were established at the regional level
and a total of 400 taxpayer services staff were trained on improving taxpayer services. In 2018, over 24,000 taxpayers
were served using the various platforms for assistance developed under the RAMP. It is important to note that the
activities highlighted above related to the modernization of taxpayer services were implemented using existing systems
and did not include the electronic services essential for voluntary compliance that could only be implemented once the
RMS was in place. A high rating is precluded because the RMS was not procured and implemented which prevented a
comprehensive reform of NAFA’s systems to reduce the compliance burden on taxpayers.
Justification of Overall Efficacy Rating
Rating: Modest
The RAMP’s efficacy in achieving its objective was limited. Out of the eight PDO indicators, three were
achieved—mostly related to increasing efficiency on revenue collection and reducing the compliance burden. At the time
that the project was cancelled, Romania still raised the least amount of revenue as a percentage of GDP in the EU, and it
continued to have the largest VAT tax gap in the EU. Significantly, in 2018, NAFA’s IT systems were in critical need of an
upgrade with demand far exceeding capacity.11 There is still much room for NAFA to become a fully service-oriented
revenue body, which allows for maximum simplification in the fulfillment of tax obligations, through extended use of IT.
It is also important to note that the Quick Wins Program12 which was established at mid-term review, was never
fully implemented and was eventually canceled. This resulted in certain improvements being undone. For instance, the
gains in effectiveness of audits have been lost, with yield per auditor for firms declining to below the baseline, yet at mid-
project, the indicator exceeded the target. Effectiveness of dispute resolution deteriorated too, with the number of days
required to settle a dispute rising to 111 at the end of the project, compared to a baseline of 60 days. With growing
numbers of appealed cases and reduced effectiveness of audits, costs of revenue collection remained above the target
10 https://www.pwc.ro/en/brochure/2018Paywell.pdf
11 As of 2019, NAFA had invested in upgrading system hardware, reorganizing its IT support function under the National Center for
Financial Information (CNIF), and establishing virtual private spaces.
12 According to an aide memoire from 2018, the Quick Wins Program identified a number of policy recommendations from the various
consultancies to modernize NAFA’s activities. Many of these did not depend on the new RMS. The program consisted of about 125
activities covering most, if not all, of NAFA’s operational activities. Slightly less than half of the activities were comple ted. The aide
memoire states that important activities to improve NAFA operational management were not executed. These included; developing and
implementing a performance management methodology; implementing internal control procedures; management and coordination of
the internal control territorial units by the headquarter GDI; establishing the tax audit Integrated Quality Framework (IQF); and
elaborating the Compliance Strategy/ Program.
Page 16 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
of 0.9 percent of revenues. In addition, expected improvements in taxpayer services, critical for voluntary compliance,
did not materialize. During the ICR discussions, NAFA officials explained that this was because electronic services could
only be implemented once the RMS was in place. The expansion of the taxpayer services staff, while proceeding, did not
reach the expected level. NAFA discontinued surveys of taxpayers and staff when there was indecision about the
implementation of the RMS. According to NAFA officials, the surveys were postponed with the intention to resume them
once the RMS was in use in order to collect “before and after data” that would have been more comparable. In any case,
the suspension of the surveys was a missed opportunity for NAFA to continue receiving feedback on its performance.
However, the project executed almost all the analytical work intended to define the requirements for a modern-
highly automated revenue administration system. NAFA adopted a strategic and performance management
methodology to plan its activities. The first plan was for the period 2017-2020. This improvement was internalized by
NAFA becoming an integral part of the processes of strategic planning and implementation of the Fiscal Budget Strategy
elaborated at the MoPF level. NAFA also improved its internal control system by adopting a risk-based approach including
the modification of existing systems to generate 17 alerts of possible unauthorized access and data manipulation,
important safeguards against possible corrupt practices. The nominal improvements in effectiveness and efficiency did
not result in the project meeting its objectives, partially because of factors outside the scope of the project but also
because the RMS was not implemented.
C. EFFICIENCY
Assessment of Efficiency and Rating
Rating: Modest
The efficiency of the project is rated as modest. There were significant delays in implementation. For the most
part, funds lay idle between the time that the project was declared effective in August 2013 until 2015 when the pace
of disbursement increased. Lags between loan effectiveness and disbursement are systemic and adversely impact the
Bank’s investment lending portfolio in Romania. This is because these operations are pre-financed from the State’s
Budget and are affected by limitations in the country’s public finance systems, including excessive controls and often
“ad-hoc” suboptimal budgetary allocation processes. After modest spending in 2013 and 2014, disbursement improved
in January 2015 reaching 5.4% in June 2015. Still, the project remained behind its planned disbursement profile and
even as it progressed, contracted activities were consistently delayed and/ or were not fully executed. This was primarily
because certain deliverables related to the implementation of the RMS had to be postponed. Consequently,
disbursements were low throughout the implementation of the project given that the bulk of expenditures related to
the RMS. Ultimately, only 20.98% of the loan amount was disbursed by the end of the project.
Although a substantial portion of the loan amount was cancelled, the USD 17.34 million that was expended,
primarily on technical assistance, contributed to strengthening NAFA’s institutional capacity and laid the foundation for
modernization through the RMS. The amount spent under the RAMP is comparable to amounts disbursed over similar
timeframes for projects with similar designs, i.e. the implementation of an RMS, prefaced by technical assistance. The
important results achieved through the technical assistance demonstrate some efficiency in expenditure.
Page 17 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Economic and financial analysis
At appraisal, the economic analysis focused on the reduction of compliance costs for corporations due to: (i)
simplified filing and payment of taxes and social contributions; and (ii) decreased number of tax inspections and
reduction of time dealing with tax officials. It recognized that Romania was among the EU countries that had
cumbersome procedures for tax compliance as documented in the Doing Business 2013 report. Other benefits, such as
increased transparency and accountability, reduced avenues for corruption and improved trust in the tax administration
were deemed difficult to measure, though their economic effect could be significant.
The ICR team conducted an ex post economic analysis, focusing on the same areas highlighted in the PAD. The
Doing Business 2018 indicators for paying taxes showed improved performance for Romania from 36 to 14 payments
per year, better than the overall ECA average of 16.5, however, the direct contributions by the RAMP to this change
remain unclear. The Doing Business 2018 report also observed a reduction in the time spent by corporates filing taxes
(from 224 hours to 163 hours). Based on that data the compliance costs for corporate taxpayers are estimated to have
decreased by 29% . The full economic analysis, which estimates a net economic benefit, can be found at Annex 4.
The financial analysis at appraisal and at ICR was based on improvements in tax revenue contribution to GDP.
At ICR, the financial analysis estimates a net fiscal deficiency. The overall tax revenue contribution to GDP was negatively
impacted by numerous changes to the Tax Code. These changes were not anticipated in the projection model at
appraisal. The full financial analysis can be found at Annex 4.
Aspects of design and implementation
The design of the project was sound and the political environment was propitious. Conditions for engagement
and implementation seemed ideal; political consensus, at least on the need for reform, was at an all-time high, and there
was a push to climb through the window of opportunity before it closed. Between the decade long dialogue on revenue
administration, extensive diagnostic work done by the Bank and the IMF, and a comprehensive analysis of the revenue
administration reform needs and priorities, as well as risks and challenges conducted when preparing the Revenue
Administration Reform Project (RARP) in 2006, the stage for reform had been set. Nonetheless, the reality was that these
were difficult, ambitious radical reforms that had failed to take off in the past or that had resulted in limited traction.
Against that background, it would have been prudent to slow down in the preparation phase, if for nothing else,
then to at least monitor the political economy, which was, and still is, very transitory, and to the extent possible phase
the engagement based on political ownership and drive for the reforms. For instance, there was never complete buy-in
on the idea of implementing a COTS solution for the RMS. During preparation, NAFA’s IT department advocated for an
in-house system but was overruled by the Government. Nonetheless, the idea of an in-house system never really
disappeared and continued to feature in dialogue up until the project was cancelled. Although the RAMP was designed
to implement a COTS solution, for some stakeholders, the implementation of an in-house system remained an option.
This suggests that there was no collective understanding, from the very beginning, and indeed throughout the project,
of why a COTS solution was most ideal in the Romanian context. A more detailed discussion on this can be found under
section III.
Page 18 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
During ICR discussions, NAFA officials also observed that the five-year implementation schedule was
inconsistent with the ambitious program. They also noted that they would have appreciated more time in the preparatory
stages to consider the different modernization options and would have benefitted from more in-depth consultation.
Limitations in the implementation of the project were related to:
(a) Political and institutional transitions that led to shifts in priorities, approaches and political will. As the project
progressed, it lacked key decision makers that could keep the reform momentum, and in the end, the project did not
have a reform champion who was committed to following through with the implementation of a COTS solution for the
RMS.
(b) The lack of a “procurement-safe” environment perpetuated by audit fears— a systemic issue throughout the
country stemming from the Romanian Court of Accounts’ prosecutorial approach to any irregularities in the
procurement process.
(c) Suboptimal implementation of the change management strategy for a project that was so change-intensive, and
a lack of tangible short-term solutions or positive reform scenarios for NAFA in the run up to the longer-term structural
reforms. This is particularly important in light of the agreement during the project appraisal decision meeting that “the
project should be viewed primarily as a change management rather than an IT project.”
(d) Difficulties sustaining dialogue and engagement in a changing political environment that warranted a more risk-
based approach to project implementation support and supervision.
These limitations will be further discussed in subsequent sections of the ICR, including in the context of lessons learned.
D. JUSTIFICATION OF OVERALL OUTCOME RATING
The overall outcome rating for the project is Unsatisfactory. This rating is based on assessments of project
relevance, average efficacy of the three objectives, and efficiency. The final rating noted in Table 1 is based on the
methodology provided in Annex H of the updated World Bank Guidance Note for Implementation Completion and Results
Reports for Investment Financing Operations, issued September 27, 2018.
Table 1: Calculation of the Overall Project Outcome Rating
Outcome Component Rating
Relevance Modest
Overall Project Efficacy Modest
Objective 1 - Increased Effectiveness and Efficiency in the Collection of Taxes and Social Contributions Modest
Objective 2 - Increased Tax Compliance Modest
Objective 3 - Reduced Burden on Taxpayers to comply Substantial
Efficiency Modest
Overall Outcome Rating Unsatisfactory
E. OTHER OUTCOMES AND IMPACTS (IF ANY)
Page 19 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Gender
The project’s activities were gender neutral in nature, and no discernible gender differentiated impacts and/
risks were identified at appraisal. That said, as part of the monitoring and evaluation arrangements, surveys were
supposed to be conducted to assess NAFA’s taxpayer services and provide NAFA with strategic information to sustain
large scale improvement efforts. Where feasible, these surveys would have provided sex-disaggregated data, in the
case of individual taxpayers, and also would have monitored any gender differences in the feedback provided. However,
NAFA postponed the surveys in anticipation of the implementation of the RMS.
Institutional Strengthening
The RAMP supported the design of the building blocks to transform Romania’s revenue administration. The
analytical phase of the project resulted in detailed reform recommendations for each operational and managerial area
of NAFA, many of which were implemented. The project improved NAFA’s management systems through business
process re-engineering, reorganizing key functions and establishing a performance framework for NAFA along with a
management training manual. Training courses on strategic management, change management and project
management were provided to over 500 NAFA managers and staff. Training was also provided on the Fiscal Code, the
Civil Code, and the Penal Code.
Support to upgrade NAFA’s analytical capacity resulted in the establishment of a new unit and development of
a revenue forecasting model, tax gap analysis, development of a voluntary compliance strategy, and training on
modelling software and eViews. NAFA strengthened its integrity management through improvements in ex-ante audit
and ex-post control activities, thereby enhancing internal control at NAFA. NAFA’s human resources (HR) policy and
strategy development were upgraded and a comprehensive training policy and strategy developed.
NAFA also made progress on the legislative framework by adopting an enhanced appeals procedure based on
the provisions of the new Fiscal Code in force from January 2016. NAFA improved its audit techniques through e-audits:
purchasing and installing 80 new IDEA software access licenses and providing enhanced training on the system. IDEA is
a computer-assisted audit technique for data mining and interrogation. NAFA also created an annual tax audit
implementation plan and adopted new procedures to increase the transparency of the audit process, including ten
specialized audit guidelines covering key sectors of the economy.
NAFA upgraded the management of its High Net Wealth Individuals (HNWI) unit which was split into High
Wealth Individuals (HWI) Directorate and High-Risk Individuals (HRI) Directorate, and an Office for Compliance
Strategies and Risk Management. The scope of work in this area was extended reaching more than 300 individuals and
family members involved in their businesses. Additionally, about 4,700 individuals were identified as high risk. On the
procedural side, a NAFA order enabled the use of indirect audit methods for fiscal verification.
NAFA also moved to improve client interaction by allocating a specific manager for each HNWI, and improving
the quality of audit and legal resources that deal with HNWI appeal cases. In the anti-fraud arena, NAFA created a new
Risk Analysis Unit and training of trainers on risk analysis was delivered, as was training on investigative techniques for
the anti-fraud staff. In terms of enforcement, NAFA developed the legal basis and systems for the electronic
Page 20 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
transmission of garnishment notices, which has progressed well with commercial banks implementing the counterpart
procedures to execute the garnishment.
Mobilizing Private Sector Financing
As a public sector project targeting the modernization of revenue administration, the RAMP did not have a
specific focus on mobilizing private sector financing. However, an improved revenue administration that can support
economic efficiency of taxation and reduce the compliance burden, two of the RAMP’s modest achievements, can
facilitate economic growth which can in turn accelerate convergence of income and institutions with the EU. Therefore,
the outcomes of the project, however limited, could improve the Government of Romania’s (GoR) ability to engage the
private sector and in time mobilize finance.
Poverty Reduction and Shared Prosperity
The support provided to the GoR to strengthen its revenue administration and improve NAFA’s effectiveness
had the potential to foster sustainable income growth. However, according to the European Commission’s 2019 report
on Romania, “the power of the tax system to reduce poverty and correct social disparities is limited”.13 At ICR, income
inequality in Romania remained one of the highest in the EU and a third of Romanians were at risk of poverty, the
highest rates in the EU. Therefore, the extent to which the project contributed to the World Bank’s twin goals is limited.
Other Unintended Outcomes and Impacts
None identified.
III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME
A. KEY FACTORS DURING PREPARATION
Project preparation was too ambitious. There were consistent calls, both from the Bank and NAFA, for quick
project preparation and approval. The important steps that NAFA had already taken to advance its revenue
administration and the preparatory work done under the RARP in 2006 (subsequently dropped) were cited as justification
for the expedited approach. However, operating environments are not static, and it is difficult to see how the RAMP
could, seven years later, simply “pick up where the RARP left off” and implement an ambitious change to NAFA’s business
process and ICT architecture, alongside major changes to the physical and operational infrastructure. Notably, at quality
enhancement review (QER), the team was reminded of the risks of implementing a project of such ambitious scope, given
the “prevailing volatilities on the macroeconomic and political fronts”. The QER suggested a more simple, focused
operation, but deferred to the team’s technical expertise and country-specific experience. The importance of closely
monitoring and managing risks associated with such an ambitious reform program was also reiterated.
The point is taken that, as a rule, investments in information technology should be predicated on the
modernization of corresponding business processes, therefore the full range of reforms was required. Also, maintaining
13 https://ec.europa.eu/info/sites/info/files/file_import/2019-european-semester-country-report-romania_en.pdf
Page 21 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
the wide system of tax offices around the country would have run counter to the project’s development objective,
therefore the consolidation of these offices was necessary. However, on hindsight, it seems that the project tried to do
too much, too soon. A more focused operation, as suggested by the QER, or a phased approach might have been more
appropriate as history had demonstrated that the operating environment, in general, but especially in fiscal sector, was
susceptible to the winds of political change, thereby increasing the level of risk.
It is also worth noting that during preparation there were differing views regarding the benefits of a COTS
solution versus an in-house developed solution, with the NAFA IT department strongly in favor of an in-house one,
arguing that it would be cheaper, easier to implement and customized to Romania’s needs. The Government and the
Bank team believed a COTS solution was more appropriate because it would quickly bring modern, consistent and
integrated processes to the NAFA. It was eventually agreed that a COTS solution would be implemented, and the MoPF
and NAFA agreed to reduce the number of tax offices and amend the Fiscal Code to facilitate this. However, although
there was agreement on the COTS solution, there wasn’t complete buy in from the IT department and the option to
implement an in-house solution lingered in the background and was raised periodically during implementation and even
just before the project was cancelled.
B. KEY FACTORS DURING IMPLEMENTATION
Various factors impeded the implementation of the project. Some played a marginal role and were symptoms
of the underlying issues, others were fundamental. The most significant factor was that sustained consensus on the need
for reform, and specifically the direction that reforms should take, did not last. It changed in tandem with changes in
political and institutional leadership and this hurt the RAMP. For example, prior to the start of the RAMP, project
proponents in the Government reduced the number of local offices from 400 to 205. During project negotiations, the
Government at the time agreed to bring the number down even further to just 50 offices before the system was
implemented. This was important as it was intended to lay the foundation for changes in NAFA’s business processes and
indeed the RMS system. However, the offices were not reduced further to the agreed 50. Maintaining the network of
local offices not only flew in the face of the modernization agenda, it also continued a system that was predisposed to
corruption. Also, NAFA’s reorganization was done hastily without due consideration for the substantial reallocation of
resources (human, IT and physical infrastructure). This got the project off to a shaky start and issues created by the
reorganization were never fully addressed.
Disbursement was problematic. About 18 months after the project became effective, it had only disbursed 0.8%,
and remained behind the disbursement schedule throughout the life of the project, notwithstanding a revision to the
project disbursement profile in 2015. In part, this was because of special disbursement arrangements in Romania which
stipulate that procurement processes can only be formally launched after the budget is allocated. As a result, it was not
possible to launch certain procurement activities in advance. This was exacerbated by the fact that requests for
disbursements by the MoPF were intermittent and inconsistent with the project’s planned funds-flow requirements.
Consequently, there were delays in procurement and in signing contracts, which culminated in implementation delays.
It is worth noting that based on prior experience, the Bank team had given sufficient consideration to a carefully staged
series of procurements to enhance disbursement and build experience with Bank procurement. However, the extent to
which this was implemented was limited because of the operating environment.
Page 22 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Delays in signing contracts were also underpinned by fear of the Romanian Court of Accounts which is renowned
for its rigorous audits, including deploying a prosecutorial approach. There are severe consequences for any irregularities,
including personal liability. The Romanian Court of Accounts’ heavy-handed approach created an atmosphere of
trepidation which was not conducive from a procurement perspective.
Although a Project Management Unit (PMU was established, all members of the PMU continued to fill several
roles in NAFA, therefore the PMU was stretched. Nonetheless, it still had sufficient capacity to manage the project,
update procurement plans, publish tenders, contract consultants and pay invoices etc. However, procurement
management related to the RMS could have been improved. There were inefficiencies in some IT related procurement
procedures, resulting in re-bidding. For example, specifications were overengineered, complicating the bidding process
and potentially impacting cost estimates. Also, as the project progressed, the PMU’s authority to take and implement
decisions seemed to diminish.
The Bank perceived the PMU’s capacity to manage the second and most critical phase of project
implementation; customization and roll out of the RMS as insufficient. Despite the Bank’s recommendation, there was
no dedicated IT project manager. The international IT consultant who was hired was let go, reportedly for
underperforming, and was never replaced.
From the Bank task team’s perspective, there was tremendous suspicion of consultants at NAFA and general
antipathy towards their perceived high fees. As a result, consultants were stymied and sidelined by reducing the scope
of their activities. NAFA management and the PMU believed that this would result in cost savings. The Bank team advised
against such ‘cost savings’, but all the consulting contracts were shut down post 2016. This was unfortunate because the
consultants added value to key aspects of the reform process; IT, tax policy, and perhaps most importantly, change
management. All of these aspects remained half-baked when the project ended. During the ICR discussions, NAFA
officials contended that the consultants brought on board did not have the requisite skills and expertise to meet the
institution’s needs. This could have been because recruitment procedures forbid interviews and consultants are recruited
on the strength of their CVs. In any case, NAFA did not feel that they were getting value for money from the consultants.
According to NAFA, consultant contracts were frozen after 2017 while awaiting a final decision on the direction of the
project.
Another key aspect of the project that was never fully implemented was the Quick Wins Program. The Program
was established at mid-term review, partially implemented and eventually canceled. This limited the outcomes of the
project and to some extent contributed to the back-sliding of certain achievements, such as the gains in effectiveness of
audit. Failure to fully implement the Quick Wins Program was a lost opportunity to shore up political support and counter
reform-fatigue.
Relatedly, efforts to provide visible changes and tangible results within a reasonable time were unsuccessful.
This, coupled with insufficient change management efforts, contributed to waning ownership of the reforms. It is also
worth noting that from a very early stage, there were challenges in identifying premises and facilities for the data
processing center in order to prepare for the acquisition of IT infrastructure for the RMS. Efforts to address these
challenges during project implementation did not gain much traction and were perhaps the harbinger of dwindling
appetite for the RMS.
Page 23 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Significant changes to personnel, on both the Bank and client side, during the life of the project hindered efforts
to build strong relationships with Government at various levels and tackle implementation problems. The project had a
total of 4 Task Team Leaders (TTLs), and a move to ensure that a TTL was based in-country did not necessarily result in
improvements in dialogue or implementation. Despite changes in task team leadership, the implementation support
team remained the same since the project’s inception, which ensured continuity. According to the PAD, the
implementation support plan called for two formal supervision and implementation support missions per year. The
record shows that nine months passed without a supervision mission between June 2015 and March 2016. In 2016, the
team conducted a supervision mission together with a mid-term review in March, but no other supervision missions were
conducted in that year. The next mission was a year later in March 2017. Therefore, the team did not fully comply with
the implementation support plan outlined in the PAD. Given the dynamic operating environment, and constant changes
in leadership (see the discussion on this below), a business as usual approach to supervision – even if implemented – was
not sufficient. The team would have done well to adopt a risk-based approach to supervision and increased the number
of supervision missions and facilitated more opportunities for engagement. This would have been consistent with the
spirit of the implementation support plan which outlined that “the objective of the supervision team is to maintain
continuous dialogue with Government counterparts on all operational, technical, policy issues relevant to revenue
collection in Romania and, particularly, in ensuring the Project achieves its development objectives .” Furthermore, at
QER, the team had been reminded of the “importance of closely monitoring and managing risks associated with the
ambitious reform program.”
There were also frequent transitions in political and institutional leadership in Romania which affected the
project’s tempo and engagement opportunities. Altogether, there were 8 different Ministers of Public Finance, with
different views on the importance of the reforms, and 6 Presidents of NAFA between 2013 – 2019. Senior management
changes at NAFA also affected the commitment and level of understanding of key officials. In 2016, NAFA spent three
months without a President after the President and Vice President were suspended following allegations of financial
impropriety that were not prosecuted in the end. A new President was eventually appointed in April 2016. The frequent
changes undermined ownership of reforms and disrupted project implementation. Interchanges of leadership personnel
amongst the MoPF, the Director General of Tax and NAFA were also common. Although personnel in the PMU remained
the same throughout, the project suffered from a lack of key decision makers and ultimately, did not have a reform
champion.
Insufficient attention was paid to the importance of change management for an operation of this nature.
Despite advice to launch a change management strategy well in advance and to ensure that an international consultant
with deep expertise in change management was onboarded in the nascent stages of the project, change management
seemed to play a peripheral, rather than a central role in the RAMP. Following this shaky start and half-hearted
embracement, efforts on change management were never fully comprehensive, neither were they sustained. Based on
the information available to the ICR team, the lack of a shared vision and commitment to the reforms, even and especially
in light of subsequent implementation difficulties and delays, contributed to the project’s inability to achieve its
objectives.
Bureaucratic resistance to any significant reforms impacting the current operating norms or employment
structure are expected in modernization initiatives. The RAMP was no different; there were concerns from NAFA staff
Page 24 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
that modernization of the agency would lead to redundancies. This could also explain why the initiative to downsize the
network of offices was abandoned. Without an effective change management initiative to mitigate these concerns and
a clear, consistent message from the top in support of the RAMP, the commitment to implement the project, even from
within the agency was weak.
Main Events/ Factors Leading to Cancellation
By early 2017, implementation progress under the RAMP had become labored. NAFA’s IT systems had not
received the required maintenance and upgrades that had been expected since 2012, and could not sustain the high
volume of activity generated by increased online interaction with taxpayers and were not in line with technological
developments. From the Bank task team’s perspective, the drive for reform in NAFA, among the political leadership and
the PMU was abating. At this stage of the project, NAFA lacked a project champion who was fully invested in
implementing the project. The Coordination Committee chaired by the Minister of Public Finance had not met since 2015,
signaling a decrease in stakeholder interest and engagement. PMU staff had begun to divert their attention to other
responsibilities, outside the scope of the RAMP, therefore focusing less on implementing the project. With the bulk of
resources dedicated to the RMS, disbursement was lethargic, and a contract that was intended to help with integration
and change management was in limbo.
During and subsequent to October 2017 Annual Meetings, discussions between the Client and the Bank focused
on considering options for the RMS. In November 2017, NAFA and the MoPF informed the Bank that NAFA management
had conducted an evaluation of the RAMP and based on the assessment, wanted to cancel the project. Several reasons
were cited, including; numerous changes to legislation which could not be accommodated in the COTS, slow progress
under the RAMP which did not match NAFA’s urgent and immediate needs, the need to implement a more flexible RMS
aimed at broader digitization of NAFA's activities, and lack of capacity to implement the COTS solution. The Bank team
acknowledged the authorities’ wish to cancel the project and requested a formal letter from the MoPF requesting
cancellation of the project. Following receipt of the letter, the Bank would move forward with cancelling the project.
Several months passed and the Bank did not receive the formal letter requesting cancellation of the project but
dialogue on options for the RMS continued. In June 2018, the Bank conducted another supervision mission. During that
mission, counterparts (i) presented the idea of replacing the COTS solution with an in-house upgrade of the RMS to the
task team and (ii) requested additional financial and technical support to upgrade IT equipment at the CNIF and to
develop a new data center building. The Bank team advised the counterparts that attempting to modernize NAFA’s
current RMS by updating existing business practices with in-house upgrades was likely to be suboptimal given that
existing systems were running on obsolete technology and the lack of internal capacity to undertake such upgrades.
Moreover, such upgrades could take up to a decade to design and implement. The Bank team also noted that the RAMP
was not designed or intended to support IT equipment upgrades at the CNIF and to develop a new data center building,
but advised the client to submit a formal restructuring request and procurement plan if it wanted to pursue this option.
A restructuring request was not submitted.
Instead, in November, 2018, the Bank received an official request from the MoPF, on behalf of NAFA to cancel
Euro 51,902,124 of the loan amount, effectively cancelling the undrawn balance of the loan that was not committed.
Page 25 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
It is worth noting that based on the Bank’s previous experience with tax administration reforms that like the
RAMP also included large and complex IT system procurement and implementation, the reasons proffered by the client
to cancel the project are difficult to substantiate. For instance, all tax systems have to adapt to changes in tax legislation
— the Russian RMS manages to adapt to several changes each fiscal year. Despite assurances from the task team that
previous changes in legislation would not affect the system, especially as it had not yet been procured or designed, the
client was not convinced. During the ICR discussions, the client observed that it would be difficult to assimilate the cost
implication of modifications in light of a 10 percent allowance provided in suppliers’ contract terms. Also, slow progress
under the RAMP could have been easily remedied by speeding up procurement of the RMS. This would not have been
too much of a heavy lift given that the documentation to launch the procurement of the RMS was ready by mid-2016.
While it is true that NAFA did not have sufficient capacity to implement a COTS solution, it is also true that throughout
the project, the Bank team impressed upon NAFA management the need for a highly-skilled and dedicated project
implementation team with the requisite technical capacity, led by a full-time project manager. Yet, NAFA management
did not move to appoint the required personnel.
The Romanian Court of Accounts’ prosecutorial approach does not lend itself to the creation and maintenance
of a “procurement-safe” environment. In Romania, civil servants are held personally liable for any cost irregularities and
in the past, project managers have been deemed to be liable, despite the Bank producing no objections and standing
behind the robustness of the procurement process. This is a systemic issue in Romania, not unique to the RAMP. Audits
are intense and could have career-ending consequences, hence the reticence to sign off on procurement. This slows
down project implementation, and in the case of the RAMP, contributed to the premature close of the project.
IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME
A. QUALITY OF MONITORING AND EVALUATION (M&E)
M&E Design
Rating: Modest
The M&E system described in the PAD was not particularly robust. The PAD mentions the results framework
“with project specific indicators and actionable monitoring arrangements” that were agreed with key counterparts, but
does not elaborate on the said actionable monitoring arrangements, neither does it delineate where responsibility for
monitoring project results would lie. A pre-appraisal aide memoire suggests that project results would be monitored by
the PMU in coordination with the Analysis Department of the MoPF.
The PAD also mentioned that NAFA would develop a robust system for monitoring and evaluation of the target
outcomes and outputs. This system would be used to maintain records on implementation and to generate performance
reports, semi-annual progress reports and annual reports. Surveys would also be conducted to assess NAFA’s taxpayer
services. These surveys would also provide gender-disaggregated data where feasible and monitor any gender
differences in feedback provided.
Page 26 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
It seems that the core of the M&E design was the results framework which provided basic information and
feedback for monitoring the results of the project during implementation, but less information on the processes
involved in effectively monitoring and evaluating key aspects of the project apart from the results. Also, the M&E design
did very little to bridge the various elements; the results framework, NAFA’s M&E system and the surveys, in a way that
made the overall M&E system all encompassing, or at the very least, capable of drawing on and leveraging all the data
available to effectively assess progress under the project. As a result, the information available only provides a snapshot
of the results and neglects other ‘softer’ process aspects of the project which would have done well to be monitored
and evaluated. Indeed, such monitoring and evaluation could have provided valuable insights that could have steered
the project.
During the ICR discussions, NAFA officials noted that most of the PDO level indicators were also influenced by
factors outside the scope of the project. Therefore, their achievement, or lack thereof, cannot be solely attributed to
the RAMP. For example, effectiveness in revenue collection measured as the rate of tax arrears is based on the
assumption that the arrears are recoverable. However, Romanian legislation and practice14 does not permit periodic
reassessment and write-off of unrecoverable arrears. This would therefore impact the effectiveness of revenue
collection as arrears would continue to accumulate. Also, the M&E framework was not revised to reflect delays in the
implementation of the RMS. Furthermore, after some time, the projected targets under the PDO level indicators
became irrelevant because without the implementation of the RMS, their measurement was no longer reliable in the
overall scope of the project’s development objective. Although NAFA suggested that the targets be formally revised in
light of the delay in implementing the RMS, this did not happen.
The project did not have an explicit theory of change, as this was not required when the project was prepared.
The ICR team ‘retrofitted’ the theory of change in an effort to crystalize the project activities and indicator targets.
M&E Implementation
Rating: Modest
The Bank team conducted supervision missions and assessed implementation progress. However, the
frequency of missions was not in accordance with the implementation support plan and implementation status and
results reports were not prepared at the required frequency. Also, the Bank’s engagement with political leadership
could have been more consistent. In March 2016, the team conducted a comprehensive mid-term review which
proposed midcourse adjustments to the project and outlined key actions to be undertaken. These included: appointing
a reform champion for the implementation phase, launching the Quick Wins Program, reconvening the high-level
coordination council, implementing the change management program and expanding the PMU’s capacity for the
implementation phase. In addition to the usual supervision team, high level tax experts from the Bank participated in
the review.
14 Tax officials are held personally liable for decisions declaring tax arrears unrecoverable.
Page 27 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
NAFA conducted surveys initially, but these were discontinued after 2017. NAFA also filed periodic reports on
time and these were consistently of high quality. In particular, the PMU prepared an exemplary mid-term report which
clearly indicated the status of project activities and the risks ahead.
M&E Utilization
Rating: Modest
Aside from isolated aide memoires that included the results framework and updated indicators, there is scant
evidence in project documentation that describes M&E implementation in a way that shows that available data was
used, either strategically or operationally to inform project implementation, policymaking or to engage the client. M&E
implementation was limited to reporting on the progress, or lack thereof, against indicators.
Justification of Overall Rating of Quality of M&E
Overall Rating of Quality of M&E: Modest
Although the Bank team was quite diligent about updating the results framework, information in project
documents about the entire M&E system is patchy and there is little evidence to demonstrate that the full M&E system
was implemented and utilized. Limited implementation and use of the entire system could have been a result of
inadequate design at preparation and little effort to consolidate the system during implementation. Given the
significant shortcomings in the M&E system, the overall rating of the quality of M&E is Modest.
B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE
The RAMP was rated as Category C for environmental purposes at appraisal. Therefore, no environmental
safeguards policies were triggered. The project did not comprise activities that could cause social risk or negatively affect
the population, or that required land acquisition or caused any physical or economic displacement. Appropriate
measures were taken to ensure that the consolidation of NAFA offices and consequent reduction in staffing levels did
not negatively impact staff.
NAFA’s procurement capacity was assessed at appraisal and deemed to be insufficient, and the procurement
risk for the project was rated high. Due consideration was given to the complexity and scope of the project, and the fact
that NAFA was not familiar with the Bank’s procurement procedures. Against this background, procurement
arrangements were made, and measures were taken to mitigate the procurement risks. During implementation, the
project followed all the necessary guidelines and procedures and the PMU had a good track record of preparing quality
procurement documents and leading the implementation of the procurement process. However, the PMU was not
always able to achieve timely procurement on several key contracts. In part, this was because the bids received by NAFA
greatly exceeded the pre-bid estimates of the contract value, and instead of accepting the offers, as is allowed under
the Bank Procurement Policies and Procedures, NAFA chose to cancel the procurement and re-open the bidding. There
were also problems with specifications being overengineered which complicated the bidding process and had the
potential to impact cost estimates.
Page 28 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Financial management (FM) performance was consistently satisfactory during project implementation in terms
of staffing, budgeting, flow of funds, accounting, internal controls, reporting and external audit. The project used a
comprehensive set of accounting policies and internal control procedures as described in the FM section of the Project
Operational Manual. The substantial delays in the procurement process for certain high value complex contracts (the
COTS RMS, in particular) and also in the acceptance of certain deliverables under several consulting contracts impacted
the realism of forecasting and predictability of in-year spending. During the early stages of the project, annual budget
allocations were often generous, reflecting the best-case scenario for implementation plans, with subsequent
reductions in the second half of the year. The yearly budget execution rates were high, close to 100%. Quarterly Interim
un-audited Financial Reports (IFRs) were sent to the Bank on time, in the formats agreed and were of adequate quality.
The yearly project financial audits were also submitted on time and had an unmodified (clean) opinion throughout
implementation.
C. BANK PERFORMANCE
Quality at Entry
At appraisal, the project reflected the government’s tax administration reform agenda and consensus on
proceeding with reforms had been shored up among the various stakeholders. This was important given that the 2006
RARP had failed to take off for lack of agreement among government stakeholders. However, buy-in on the direction
that the said reforms should take, especially regarding the implementation of a COTS solution or an-in house solution,
was not universal. This should have given the team pause. Later, fleeting political will and continued debate about a
COTS solution versus an in-house solution hampered the implementation of the project and led to cancellation and non-
achievement of the PDO. While there is no doubt that the project was strategically relevant at entry, the approach could
have been fortified by conducting a rigorous analysis and stakeholder consultations on the pros and cons of a COTS
solution versus an in-house solution. Agreeing, upfront on the most ideal and strategically relevant RMS choice for the
Romanian context and committing to a specific modernization pathway would have left little room to second guess the
RMS solution agreed to, thus facilitating the project’s ability to fully achieve its planned development outcomes.
The Bank team acknowledged the complexities involved in projects that were ICT-intensive but also
demonstrated that it had extensive experience with tax administration reforms in the region, including successful
implementation in neighboring Bulgaria. The risks were clearly outlined, and lessons learned from similar projects
adequately documented. With strong government ownership of the broader reform agenda, the Bank’s strong track
record in this area of the reform, and an EU level revenue administration system as the goal, the project seemed to be
primed for success.
As already alluded to, extensive preparatory work had been undertaken to understand the policy and
institutional aspects of revenue administration in Romania. The work included comprehensive diagnostics and an
analysis of Romania’s revenue administration reform needs and priorities. At entry, NAFA demonstrated that it was
technically competent as an implementing agency and was tasked with ensuring cross-departmental coordination with
the agency and externally with the MoPF and other stakeholders. In addition to operational aspects related to specific
project activities, implementation arrangements were also designed to cover high-level policy and strategic
management among key stakeholders and external coordination and accountability. For example, a Coordination
Page 29 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Council would be established to make key strategic decisions and ensure interagency coordination and a Consultative
Technical Panel would be established for technical discussions with stakeholders heavily affected by the project to
reflect their views on institutional changes made by the project.
The operation was consistent with the World Bank’s fiduciary role. Project activities were gender neutral and
did not have environmental implications. It was understood that if the project achieved its development objective, in
the long term, it could contribute to sustainable income growth, but project activities did not have any overt social
development ramifications or effects on poverty. The overall M&E framework was somewhat fragmented which limited
its ability to generate information that could be used to influence the design of the project. Some baseline data only
materialized in the second and third years of project implementation.
Quality of Supervision
Bank performance varied between different functions, as can be expected for a complex project. Overall, Bank
performance started out strong, with proactive monitoring and engagement, but weakened as the project progressed.
Implementation support missions became infrequent, adversely affecting dialogue with the client and resulting in
missed opportunities for meaningful engagement to course correct or foster a shared understanding of the difficulties
encountered during implementation and how best to address them. The constant changes in task team leadership
exacerbated the situation; the project had a total of 4 TTLs in the five years that it existed. Also, a move to have a TTL
based in country did not result in significant improvements.
Procurement and fiduciary arrangements were adequate. The Bank team provided support and training as
required to strengthen implementation capacity.
A robust mid-term review revealed that the necessary building blocks to achieve the PDOs were in place, but
the project now needed to transition from the analytical to the implementation phase. The review also warned that
there was a significant risk that some of the indicators might not be achieved in the absence of action, and highlighted
that NAFA faced serious capacity constraints related to ICT infrastructure, ICT staff and space for ICT infrastructure and
consultants. The importance of strengthening the change management process was also reiterated. Against this
background, the review recommended that a Quick Wins program be launched to ensure the implementation of the
second phase of the RAMP. The task team heeded the advice and a program was put in place but was ultimately
cancelled by NAFA. While it lasted, the program resulted in marginal improvements to the project but many of these
achievements were undone when the program was cancelled. It is also worth noting that the mid-term review occurred
in March 2016 before major problems with the procurement and implementation of the RMS had crystallized, therefore
a restructuring of the project was not recommended at mid-term review.
To expose NAFA to international best practice and to facilitate peer-learning, the task team invited the President
of NAFA and the PMU coordinator to the Tax Administrators eXchange for Global Innovative Practice (TAXGIP)
Conference in St. Petersburg in the summer of 2016. In 2017, thanks to the task team’s efforts, the TAXGIP Conference
was held in Bucharest. Having the Conference in Bucharest while the RAMP was being implemented gave NAFA and
opportunity to showcase and discuss the work it was doing under the project.
Page 30 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
When presented with alternatives to implementing the RMS, the Bank team emphasized that the RAMP was
designed to support the GoR to implement a COTS solution, and that COTS solutions for RMS were being adopted by
countries around the world, and many EU Member States had implemented second and third generation COTS
solutions. In any event, other options, including an in-house solution, would have required the project to be redesigned
and would have undermined the overall objective to change existing business processes. Furthermore, there was no
internal capacity to develop a new system or to include custom add-ins.
Taking into account the irregular implementation missions which affected dialogue with the client, in general,
the Bank team was proactive in identifying threats to the achievement of the development objectives and reported
these threats candidly and in a timely manner. However, this did not translate to action in the context of resolving some
of the issues related to delays in implementation, and in the end, the cancellation of the project.
While the Bank cannot manufacture political will, rather than wait to see if it will materialize with various
changes in government, it could have done more to find ways to push political support, especially given that the volatile
political environment had been flagged as a significant risk. At QER and in the PAD, it was recommended to engage the
private sector to help push reforms and monitor the project, but there is no evidence that this was done.
Justification of Overall Rating of Bank Performance
Moderately Unsatisfactory
RAMP had several shortcomings, some of which were foreseen but were not adequately addressed. In part, this
was because the task team was not sufficiently and consistently engaged in a dynamic environment that was prone to
change, while working on a complex project. However, there were also factors in the operating environment that were
beyond the Bank’s control, such as the constant legislative changes and, in the end, the GoR’s decision to take the
reform of its revenue administration in a different direction. All things considered, the Bank’s overall performance is
rated moderately unsatisfactory.
D. RISK TO DEVELOPMENT OUTCOME
Rating: High
In the absence of an RMS, the modest outcomes achieved under the project in pursuit of the development
objective (which was not achieved) are at risk. Essentially, the project only laid the foundation for the achievement of
the development objective and if an RMS is not implemented forthwith, even the modest outcomes achieved will be
undone. This is especially so given that since the discontinuation of the Quick Wins Program, some of the project’s
achievements were lost. In fact, implementation of intermediate targets deteriorated significantly since the mid-term
review and by the end of the project only 5 of the 18 indicators had been met because most of the outstanding indicators
depended on the implementation of the RMS.
Structural reforms– such as streamlining of processes, reorganizing key functions and establishing a
performance framework for NAFA have made significant improvements. However, these are not one-shot reforms, and
more work is needed. It is hoped that the catalytic effects of these reforms are harnessed as NAFA moves forward with
the modernization of its tax administration. At ICR, Romania had taken steps to ensure that taxpayers only communicate
Page 31 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
with NAFA online. To this end, NAFA had self-service computers in 98 percent of its offices to support this change. The
training provided under the project strengthened the capacities of key personnel in the administration. That said, such
training needs to be continuously updated to keep pace with new technologies and changes.
With EU funds at the GoR’s disposal, and indeed, the State’s own budget, it is possible for NAFA to capitalize on
key reform elements initiated under the RAMP. In fact, NAFA has already indicated its intentions to proceed in this way,
and as the modernization of Romania’s revenue administration is still embedded in the Government’s broader reform
agenda, it is hoped that some of the gains achieved under the RAMP can be salvaged. For example, computerized
systems for payment and checking taxpayers’ situation in order to improve fiscal administration services and combat
tax evasion are being developed. The Virtual Private Space (SPV) is a complex IT system that facilitates communication
with taxpayers and helps to simplify procedures. Functionalities include submission of fiscal statements and supporting
documentation. PatrimVen is dedicated to cutting red tape and simplifying procedures for taxpayers by maintaining a
common database for tax-related information.
V. LESSONS AND RECOMMENDATIONS
Sustained political and institutional consensus and ownership are essential for change intensive projects.
Consensus on the need for reform, but more importantly, on the direction that reforms should take is critical for projects
that introduce substantial changes to an institution and how it functions. As such, consensus on, and ownership of, the
reform agenda among key internal and external stakeholders is required from the very nascent stages of project
preparation and ought to be followed through and sustained throughout project implementation. Political and
institutional will are not static and evolve with time, and must therefore be monitored and managed closely to prevent
divergent views on reform options and changes in reform priorities that can easily derail a project. In the case of the
RAMP, conflicting views on whether to adopt a COTS or inhouse developed system remained unresolved and
contributed to the project’s premature closing.
Adopting a modular approach to complex revenue administration projects in difficult contexts can help to
build reform scenarios. Typically, revenue administration projects are procurement and IT heavy and are therefore
suited to investment loans. Incorporating a modular approach to such projects by including short-term solutions to
address urgent issues, and focusing on delivering a complete solution in a specific reform area, as opposed to a
comprehensive reform approach, can help to demonstrate tangible results, galvanize reform efforts and stakeholder
support, and buy time for more structural and longer-term reform processes. This is especially true for projects that in
addition to being IT intensive, intend to undertake comprehensive institutional reforms.
Projects with complex IT and procurement investments require sufficient technical assistance to ensure
technical readiness. In situations where the client’s capacity is constrained, it is important to bring technical experts on
board early. Leveraging the Bank’s internal capacity alongside the expertise of cutting-edge external consultants can be
an effective way of providing timely and sustained technical assistance. Looking inward is also important. If there are
other government agencies in the country that have experience with complex IT and procurement investments, it could
be helpful to bring them on board as an additional source of expertise and pillar of support for the project. In Romania’s
case, the Serviciul de Telecommunicatii Speciale has vast experience with large IT management systems.
Page 32 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Institutional reform projects require robust change management. Robust change management procedures
have the potential to ensure that top management and other functional managers fully understand the impact of
changes, creating opportunities to influence their direction and reduce resistance to reform efforts. Underestimating
the extent of the changes needed and the importance of sustained change management can be detrimental to a project.
Also, change management efforts should permeate all levels of the organization and be sustained throughout the life of
the project. Specifically, change management strategies should include a combination of bringing in international
expertise/ experts as well as building and sustaining in-country capacity for change among stakeholders and
implementors who are on the ground.
Risky projects require heightened supervision. Clients rely on the Bank’s guidance, advice and support in
implementing projects. Regular implementation support missions that monitor the project’s progress and perhaps more
importantly the pulse and direction of broader reforms and the operating environment are vital. The operational aspects
of the project are just as important as broader policy dialogue on fundamental reforms. It could also be helpful to ensure
that for complex projects and/ or projects in a dynamic environment, at least one of the TTLs is based in the country.
High-level engagement with political and institutional leaders is also important to maintain buy-in and facilitate timely
adjustments.
Strong M&E can create opportunities for dialogue which can be an effective way to sustain engagement and
communicate progress and interim results. Beyond reviewing and documenting the project’s progress, deliberate
efforts are required to explain to key stakeholders how activities are leading to results. Open and consistent dialogue
helps to manage expectations and ensure that everyone is still on the same page regarding the necessary reforms.
Additionally, these communication efforts should not be confined to key personnel in government, the implementing
agency or the sector but also to external stakeholders including the private sector and citizens. This way, M&E is not just
a compliance function, but an information generation and sharing exercise. Moreover, M&E should be utilized as a tool
by management to steer the project and make course corrections.
Project teams should consistently take heed of the lessons learned as reflected in the design of the project
during implementation. Some of the pitfalls that RAMP encountered were highlighted as risks in the PAD, yet when
these risks materialized, they were not adequately addressed. For example, political will waned as the project
progressed and it proved difficult to resuscitate through various mechanisms such as the project Coordination Council
and the Quick Wins Program. Also, the implementation of the change management strategy was suboptimal, yet it was
an integral aspect of the implementation and indeed the success of the project, as highlighted in the PAD.
Comments on Issues Raised by Borrower/Implementing Agencies/Partners
(a) Borrower/implementing agencies
The Borrower’s comments were received on September 13, 2019. In accordance with NAFA’s request, the comments are
verbatim in Annex 5.
There are no issues in the Borrower’s comments that would warrant a response here. We note the tremendous support
and commitment to the ICR process that has been shown by NAFA and the Romanian MoPF counterparts. The information
Page 33 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
that they have provided has greatly assisted the team in the completion of this report.
(b) Co-financiers
There were no other co-financiers to the project.
.
Page 34 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS
A. RESULTS INDICATORS
A.1 PDO Indicators
Objective/Outcome: (i) To increase effectiveness and efficiency in collection of taxes and social contributions
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Effectiveness in revenue Percentage 40.00 50.00 33.20
collection measured as the
rate of tax arrears 31-Jan-2012 30-Mar-2018 31-Dec-2018
Comments (achievements against targets):
Not Achieved. For the rate of recovery of tax arrears, the baseline was 40% and the target was 50%, yet in 2018 the rate was 33.2%,
backsliding from 36.7% at the mid-point of the project in 2015. Recommendations for improvements to arrears management were developed
under the project but some of them required legislative changes, which did not occur during project implementation.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Effectiveness in revenue Text VAT 35.8%, CIT 11.9% VAT 46.0%, CIT 21.0% VAT 34.9%, CIT 10.3%
Page 35 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
collection as measured by the 31-Dec-2012 30-Mar-2018 31-Dec-2018
productivity rate of VAT and
CIT
Comments (achievements against targets):
Not achieved. The target for VAT productivity was 46%, but in 2018, the productivity rate was 34.9% — lower than the baseline (35.8%). In
2018, the CIT productivity rate was 10.3% — lower than the baseline (11.9%) and far from the envisaged target of 21%. Between 2015 and
2018, there were considerable legislative changes related to corporate income tax. These changes substantially changed the scope of firms
included in the microenterprises’ category, the tax rate applied to them and, implicitly, the budget receipts coming from this type of taxes. The
threshold for the annual income up to which a company is included in the microenterprises’ category increased by more than 15 times,
equivalent to the tax base shifting towards revenues rather than profits, while diminishing and differentiating the tax rates. The significant
reduction in the number of profit tax payers negatively affected aggregate budget receipts by an estimated 1.5-2 billion lei (equivalent to 10 to
13% of 2018 gross corporate tax revenues) over the 4 years. It can therefore be concluded that if the legislative changes had not occurred, the
project’s interventions were likely to have contributed to an increase in CIT productivity.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Efficiency in revenue Number 6310.00 8715.00 9147.00
collection measured as the
revenue collected per NAFA 31-Dec-2012 30-Mar-2018 31-Dec-2018
staff (RON Thousands)
Comments (achievements against targets):
Achieved. By 2018, targeted improvements in increasing efficiency in revenue collection had been exceeded by 5%. The target was RON 8,715,
starting from a baseline of RON 6,310, and the project achieved RON 9,147. The project interventions which contributed to this increase in
Page 36 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
efficiency include: reorganizing NAFA, streamlining business processes and workflows, modernizing IT systems, and establishing high volume
processing.
Objective/Outcome: (ii) To increase tax compliance
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Tax Gap for VAT, CIT, and PIT Text VAT 36.2%, CIT 45.6%, VAT 26.2%,CIT 40.6%, VAT 36.0%, cit 29.9%,
PIT 41.5% PIT 36.5% PIT 43.4%
31-Dec-2014 29-Dec-2017 29-Dec-2017
Comments (achievements against targets):
Partly Achieved. The project aspired to reduce the VAT gap from 36.2 to 26.2. However, in 2018, the VAT gap was 36.0 resulting in Romania
continuing to have the largest VAT gap in EU, a phenomenon that had been in existence for several years. Furthermore, according to a 2018
report[1], Romania was one of the few countries that saw a widening of the tax gap in 2016. Considering factors outside the scope of the
project, as of January 2016, Romania reduced its standard VAT rate from 24% to 20%. This change had a substantial impact on the effective
rate which declined from 17.2% to 13.5%. Despite the reduction of the rate, VAT non-compliance increased and the VAT gap widened.
Instead of decreasing, the PIT gap increased from 41.5 to 43.4 and remained out of range of the target which was 36.5. There were also factors
beyond the control of the project’s activities that contributed to the increase in the PIT gap. According to the 2018 Romania Fiscal Council
Annual report, legislative changes resulted in the standard PIT rate decreasing from 16% to 10% resulting in reduced revenue.
Page 37 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
There was a reduction in the compliance gap for CIT which declined from 45.6 to 29.9, exceeding the set target (40.6) by 15%.
[1] Study and Reports on the VAT Gap in the EU-28 Member States: 2018 Final Report, TAXUD/2015/CC/131
Objective/Outcome: (iii) To reduce the burden on taxpayers to comply
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Compliance cost as the Days 224.00 176.00 163.00
average time for preparation
of tax returns for corporate 31-Dec-2012 31-Dec-2018 31-Dec-2018
taxpayers
Comments (achievements against targets):
Achieved. The compliance burden was significantly reduced, exceeding the set target. The average time to prepare tax returns was 224 hours
at the beginning of the project. In 2018, when the project closed, the average time was 163 hours, considerably less than the target (176)
hours.
Page 38 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
A.2 Intermediate Results Indicators
Component: Institutional Development
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 1-Instl Devt: Text No Yes Yes
Estimated tax gap is
calculated annually 31-Dec-2015 31-Dec-2018 31-Dec-2018
Comments (achievements against targets):
Achieved. From 2015 onwards the estimated tax gap was calculated annually.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 1-Instl Devt: Case Yes/No N Y Y
management tools
implemented for managing 30-Dec-2016 31-Dec-2018 31-Dec-2018
legal and appeals processes
Comments (achievements against targets):
Partly achieved. Case management tools to manage legal and appeals processes were implemented in 2018. However, the full
implementation of the tools was contingent on the implementation of the RMS.
Page 39 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 1- Instl Devt: NAFA Percentage 68.00 88.00 66.00
staff’s level of satisfaction
with new skills and ability to 31-Dec-2014 31-Dec-2018 31-Dec-2018
apply them for better
performance
Comments (achievements against targets):
Dropped. NAFA's staff level of satisfaction with new skills and the ability to apply them for better performance was expected to increase by
10% annually, starting from a baseline of 68% in 2014. In 2015, levels of satisfaction dropped to 63% then increased slightly in 2016 to 66%,
but remained below the level of satisfaction recorded in 2014. Also, NAFA discontinued the annual surveys in 2017.
Component: Increasing Operational Effectiveness and Efficiency
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 2-Increasing Text RON 608,000 RON 608,000 RON 454,000
Effectiveness & Efficiency:
Yield per audit of firms and 31-Dec-2015 29-Dec-2017 31-Dec-2018
High-wealth individuals
Comments (achievements against targets):
Page 40 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Not achieved. The yield per audit of HNWIs was 608,000 RON in 2015, and was expected to increase by 25% annually. However, by the end of
the project the yield per audit was 454,000 RON, significantly lower than the baseline. In fact, in 2016, the yield per audit decreased
significantly and was 161,000 RON, then in 2017 it increased dramatically to 2,064,700 RON before falling to 454,000 RON in 2018.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 2-Increasing Text VAT: 78.5%PIT: 81.5%, VAT:83.5%, PIT: VAT:86.1%, PIT:
Effectiveness & Efficiency: SC: 76.5% 86.0%, SC:82.5% 86.2%, SC: 82.45%
Payment compliance rate for
VAT, PIT, Social 31-Dec-2012 31-Dec-2018 31-Dec-2018
Contributions
Comments (achievements against targets):
Exceeded. The payment compliance rate for VAT, CIT exceeded the set targets. While the payment compliance rate for social contributions
met the stipulated targets.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 2-Increasing Oprtnl Days 61.00 45.00 111.00
Effectiveness & Efficiency:
Average time required to 31-Dec-2012 31-Dec-2018 31-Dec-2018
settle dispute resolution
Page 41 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Comments (achievements against targets):
Not Achieved. The average time required to settle disputes was expected to decrease from 61 days to 45 days. However, by the end of the
project, the number of days required to settle disputes had increased to 111 days.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 2-Increasing Oprtnl Text 1.1% 0.9% 1.03%
Effectiveness & Efficiency:
Cost of revenue collection 31-Dec-2012 31-Dec-2018 31-Dec-2018
Comments (achievements against targets):
Not achieved. Although the cost of revenue collection decreased slightly from 1.1% to 1.03%, it did not meet the stipulated target of 0.9%.
Component: Taxpayer Services and Corporate Communications
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 3-Taxpayer Srvcs and Percentage 2.50 8.30 3.28
Corp Comm: Ratio of NAFA
staff in Taxpayer Services 31-Dec-2012 31-Dec-2018 31-Dec-2018
Page 42 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Comments (achievements against targets):
Not achieved. The ratio of NAFA staff in taxpayer services was expected to increase from 2.5% to 8.3%. However, by the end of the project
only 3.28% of NAFA staff were in taxpayer services. Although this was a slight improvement from the baseline, it fell short of the set target.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 3-Taxpayer Srvcs and Percentage 95.00 98.00 99.00
Corp Comm: Electronic filing
rate by large and medium 31-Dec-2012 31-Dec-2018 31-Dec-2018
size firms (legal persons)
Comments (achievements against targets):
Exceeded. The electronic filing rate by large and medium size firms exceeded the set target of 98%. By the end of the project, the electronic
filing rate by large and medium size firms was 99%.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 3-Taxpayer Srvcs and Percentage 54.00 56.00 52.00
Corp Comm: Taxpayers’
satisfaction with NAFA’s 31-Dec-2014 31-Dec-2018 31-Dec-2018
quality of taxpayers services
Page 43 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Comments (achievements against targets):
Dropped. NAFA discontinued the annual surveys in 2017. However, when the last survey was conducted in 2016, taxpayers' (self-employed)
satisfaction with NAFA's integrity was 52%, marking a slight decrease in satisfaction from when the first survey was conducted in 2014 (545).
The levels of satisfaction were expected to increase by 15% annually.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 3-Taxpayer Srvcs and Percentage 48.00 52.00 50.00
Corp Comm: Taxpayers’
satisfaction with NAFA’s 31-Dec-2013 31-Dec-2018 31-Dec-2018
integrity
Comments (achievements against targets):
Dropped. NAFA discontinued the annual surveys in 2017. However, when the last survey was conducted in 2016, taxpayers' (corporate)
satisfaction with NAFA's integrity was 50%, marking a slight increase satisfaction from when the first survey was conducted in 2014 (48%). The
levels of satisfaction were expected to increase by 15% annually.
Formally Revised Actual Achieved at
Indicator Name Unit of Measure Baseline Original Target
Target Completion
Comp 3-Taxpayer Srvcs and Hours 180.00 163.00 168.00
Corp Comm: Taxpayer
perception of the time 31-Dec-2014 31-Dec-2018 31-Dec-2018
Page 44 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
required to comply with tax
procedures
Comments (achievements against targets):
Dropped. NAFA discontinued the annual surveys in 2017. However, when the last survey was conducted in 2016, taxpayers' (corporate)
perception of the time required to comply with tax procedures had decreased to 168 hours per year from 180 hours per year in 2014. The
target was to decrease the number of hours taxpayers perceived were required to comply with tax procedures by 20% annually.
Page 45 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
B. KEY OUTPUTS BY COMPONENT
Objective/Outcome 1a: Increased effectiveness in the collection of taxes and social contributions
1. Effectiveness in revenue collection— measured as the rate of tax
arrears.
Outcome Indicators 2. Effectiveness in revenue collection —measured by productivity rate
for VAT and CIT.
1. Annual increase in the payment compliance rate for: VAT, PIT, Social
Intermediate Results Indicators
Contributions.
1. Supported the modernization of business processes and workflows
by implementing a data warehouse system, and back up, security and
disaster recovery capacity.
2. Training conducted to improve NAFA's operational functions with
regard to tax audit, documentary control, anti-fraud, business fiscal
intelligence, enforcement, registration, declaration and payment
processing, and high-volume revenue processing in line with
international best practices, particularly those followed in the
Key Outputs by Component European Union member states.
(linked to the achievement of the Objective/Outcome 1a) 3. Reorganized key functions and established a performance
framework for NAFA along with a management training manual
4. Training courses on strategic management and change management
provided to over 500 NAFA managers and staff.
5. Training provided on the Fiscal Code, the Civil Code and the Penal
Code.
6. Revenue forecasting model developed.
7. Tax gap analysis conducted.
8. Training provided on modelling software and eViews.
Page 46 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
9. NAFA’s human resources policy and strategy development were
upgraded.
10. Training policy and strategy developed.
11. Annual tax audit implementation plan developed
12. New procedures to increase the transparency of the audit process
adopted.
Objective/Outcome 1b: Increased efficiency in the collection of taxes and social contributions
Outcome Indicators 1.Measured as the revenue collected per NAFA staff
1. 50% annual increase in the yield per audit of firms
2. 25% annual increase in the yield per audit of HNWI
Intermediate Results Indicators
3. Annual decrease in the cost of revenue collection
4. Annual decrease in the average time required to settle disputes
1. NAFA reorganized; staff rationalization that reduced the number of
positions from 28,904 to 27,100, abolition of the financial guard,
reallocation of positions to subordinated territorial units and
understaffed functions (antifraud, tax inspection, IT, legal.
2. Streamlined business processes and workflows and modernized IT
systems (excluding the RMS)
3. Established high volume processing
Key Outputs by Component
4.Enhanced appeals procedure adopted.
(linked to the achievement of the Objective/ Outcome 1b)
5. Management of the HNWI Unit upgraded by splitting it into high
wealth individuals directorate and high-risk individuals directorate.
6. Specific managers allocated to each HNWI.
7. Quality of audit and legal processes that deal with HNWI appeal
cases improved.
8. Legal basis for the electronic transmission of garnishment notices
developed.
Objective/Outcome 2: Increased tax compliance
Page 47 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Outcome Indicators 1. Compliance as measured by the tax gap for VAT, CIT and PIT
Intermediate Results Indicators 1. Electronic filing rate by large and medium size firms (legal persons).
1. Modernized and enhanced businesses processes and workflows
(except the RMS). For example, the risk compliance and VAT fraud
software system.
2. Developed a voluntary tax compliance strategy.
3.Risk Analysis Unit created and training of trainers on risk analysis
delivered.
4. Training on investigative techniques for anti-fraud delivered to staff.
Key Outputs by Component 5. Upgraded systems for internal control alerts.
(linked to the achievement of the Objective/Outcome 2) 6. Established a taxpayer assistance monitoring unit and call centers
7. Delivered educational campaigns on taxpayer awareness
8. Executed information campaigns.
9. Moved from face-to-face to electronic payments.
10. Implemented electronic garnishment of debtor’s bank accounts and
electronic communication for all administrative acts of enforcement.
11. Established an Office for Compliance Strategy and Risk
Management
Objective/Outcome 3: Reduced burden on taxpayers to comply
1. Compliance cost — measured as the average time for preparation of
Outcome Indicators
tax returns of large, medium and small corporate taxpayers.
1. Taxpayers' satisfaction with NAFA's integrity (corporate taxpayers
and those who are self-employed)
2. Taxpayers' satisfaction with NAFA's quality of taxpayers services
Intermediate Results Indicators (corporate taxpayers and those who are self-employed)
3. Taxpayer perception of the time required to comply with tax
procedures (corporate taxpayers and those who are self-employed)
4. Ratio of NAFA staff in taxpayer services.
Page 48 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
1. Implemented mechanisms for transparency and accountability
2. Modernized service delivery channels to provide advanced services
to taxpayers.
3. Promoted community understanding of revenue laws, procedures
Key Outputs by Component
and rights and obligations of taxpayers.
(linked to the achievement of the Objective/Outcome 2)
4. Promoted taxpayer efficiency through simplification and provision of
electronic services.
6. Strengthened external and internal communication.
7. Established mandatory electronic filing of declarations.
Page 49 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION
A. TASK TEAM MEMBERS
Name Role
Preparation
Alberto Leyton Task Team Leader(s)
Valeria Irmanova Nikolaeva Procurement Specialist(s)
Anneliese Viorela Voinea Financial Management Specialist
Supervision/ICR
Moses Sabuni Wasike Task Team Leader(s)
Shawkat M.Q. Hasan Procurement Specialist(s)
Anneliese Viorela Voinea Financial Management Specialist
Ismail Radwan Team Member
Rajul Awasthi Team Member
Stella Ilieva Team Member
Valerie Morrica Social Specialist
Cesar Niculescu Environmental Specialist
Marielle Emma Wessin Severino Team Member
Runyararo Gladys Senderayi Team Member
B. STAFF TIME AND COST
Staff Time and Cost
Stage of Project Cycle
No. of staff weeks US$ (including travel and consultant costs)
Preparation
Page 50 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
FY12 15.897 229,260.45
FY13 18.620 254,861.67
Total 34.52 484,122.12
Supervision/ICR
FY13 1.125 46,101.82
FY14 8.962 86,642.03
FY15 12.450 152,168.72
FY16 15.913 105,799.57
FY17 6.850 88,911.55
FY18 2.200 68,485.88
FY19 4.850 75,461.61
FY20 9.750 55,770.95
Total 62.10 679,342.13
Page 51 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 3. PROJECT COST BY COMPONENT
Amount at Approval Actual at Project Percentage of Approval
Components
(US$M) Closing (US$M) (%)
Institutional Development 9.00 2.77 31
Increasing Operational
75.69 11.52 15
Effectiveness and Efficiency
Taxpayer Services and
3.83 1.89 49
Corporate Communications
Project Coordination and
3.31 1.16 35
Management
Total 91.83 17.34 19
Page 52 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 4. EFFICIENCY ANALYSIS
Table A4.1. Calculation of Compliance Costs for Corporate Taxpayers Before and After RAMP
Unit Taxpayer Categories Totals
Large Medium Small Micro
Compliance cost before RAMP - 2013
Number of taxpayers - 2013 2,000 22,692 571,899 68,107 664,698
Average time for preparation of tax returns days 31 27 24 12
Total time for preparation of tax returns days 62,000 612,684 13,725,576 817,284
Average remuneration of accountant Euro/day 50 38 31 29
Total cost of compliance Euro 3,100,000 23,281,992 425,492,856 23,701,236 475,576,084
Compliance cost after RAMP – Estimated in 2013
Number of taxpayers - 2013 2,000 22,692 571,899 68,107 664,698
Average time for preparation of tax returns days 24 21 20 10
Total time for preparation of tax returns days 48,000 476,532 11,437,980 681,070
Average remuneration of accountant Euro/day 50 38 31 29
Total cost of compliance Euro 2,400,000 18,108,216 354,577,380 19,751,030 394,836,626
Compliance cost after RAMP – Actual in 2018
Number of taxpayers - 2013 2,000 22,692 571,899 68,107 664,698
Average time for preparation of tax returns days 23 20 17 9
Total time for preparation of tax returns days 46,000 453,840 9,722,283 612,963
Average remuneration of accountant Euro/day 50 38 31 29
Total cost of compliance Euro 2,300,000 17,245,920 301,390,773 17,775,927 338,712,620
Estimated annual economic effect (Euro) 136,863,464
Economic effect in percentage 29%
Page 53 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Key assumptions
For purposes of calculating the effect of project interventions:
a) the same number and categories of taxpayers used at appraisal are used at ICR. The total number
of taxpayers at the close of the project is estimated at 692,968.
b) The average time to prepare tax returns at the close of the project is prorated based on the results
of the Doing Business 2013 and Doing Business 2018 reports, respectively estimated as 224 and
163 hours.
c) The average remuneration of accountants used at appraisal is used at ICR. Based on
PricewaterhouseCoopers’ Paywell Romania Salary and Benefits surveys, the average annual
increase in remuneration during the implementation of the project is estimated at 4.8%.
Financial Analysis
The financial analysis shows whether the project accumulated tax revenues in volumes exceeding the
capitalized cost of the project. It considers actual revenues (growth of revenues) and expenditures (project
costs) and is based on the increase of tax revenues as a result of improvements in tax administration and
reductions in tax arrears under the project. Key assumptions at the time of appraisal, based on similar
reforms in other countries, were that: (i) improved tax administration would facilitate the growth of tax
revenues measured as a percentage of GDP, from 28.7% in 2012 to 30.8% in 2017; and (ii) annual
improvements in tax compliance rates would occur as follows - VAT by 10%; profit tax by 5%; and payroll
taxes by 5%. Other assumptions were that there would be a discount rate of 10% and an annual inflation
rate averaging 3%.
In 2018, the ratio of tax revenues to GDP (taxes and social contributions) reached 26.7%15, ranking
Romania penultimate among EU Member States. There was a gap of 13.2 percentage points between
Romania and the EU average (39.9%). Major changes introduced by the new Tax Code resulted in shortfalls
in annual revenues contribution to GDP.
15 Source: Romania Fiscal Council 2018 Annual Report
Page 54 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Table A4.2: Calculation of Additional Fiscal Benefits from Project Implementation
(in million Euros)
2013 2014 2015 2016 2017
Tax revenues before RAMP
Tax and social contributions – baseline 43,247 44,929 46,512 48,043 49,624
Tax and social contributions – projected 43,247 44,929 46,829 48,854 51,289
Memo: GDP 149,126 158,483 167,839 177,356 187,424
Tax and social contributions to GDP percentage - 29.0 29.2 29.4 29.6 29.8
baseline
Tax and social contributions to GDP percentage - 29.0 29.2 29.6 30.1 30.8
projected
Projected benefits compared to baseline 0 0 316 812 1,665
Tax revenues after RAMP
Tax and social contributions to GDP percentage - actual 29.0 27.6 28.0 25.9 25.7
Actual benefits compared to baseline 0 (2,462) (2,215) (6,005) (6,827)
Table A.3: RAMP Additional Fiscal Benefits
(in million Euros)
2013 2014 2015 2016 2017 Total
World Bank financing 0 0 4 6 7 17
Total Cost 0 0 4 10 17
Benefit/(deficit) from improved collection 0 (2,462) (2,215) (6,005) (6,827) (17,510)
RAMP Net Fiscal Benefit/(Deficit) 0 (2,462) (2,219) (6,011) (6,834) (17,526)
Page 55 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS
The activities implemented through the Revenue Administration Modernization Project (RAMP) offered the
National Agency for Fiscal Administration the means necessary to increase the skills of its staff at different levels,
as well as the necessary material resources.
The decision regarding the cancellation of the project was taken after a long analysis, carried out with our
partners. But, because the evolution of the Romanian public administration recorded a number of key moments
that narrowed the actual possibilities of the project to contribute to the reform of NAFA, the steady decision
was to cancel the loan agreement and to continue the development of the fiscal system based on other strategic
options taking into consideration the fact that the procurement of a COTS RMS system is not the only one option
for modernization.
The modernization of the agency continues to be one of the priorities on the agenda of the Romanian
Government, and a number of initiatives are being developed to accelerate its informatization.
Already in 2018 and 2019, a number of investments have been made in the modernization of infrastructure of
the agency and the increase of the data storage capacity (procurement of servers, workstations, licenses, etc.).
On another hand, the new guideline continues, meaning the transfer of interaction with taxpayers exclusively in
electronic format: expansion of the Virtual Private Space, creation of new functions indispensable for taxpayers,
and opening of this facility to other categories of payers.
At the same time, a series of extensive operations are underway for strengthening existing IT applications,
simplifying interaction with taxpayers and developing new functionalities. The MoPF and the National Agency for
Tax Administration have created a framework plan for development and modernization, this solution being more
flexible, faster and more cost-effective.
The main elements of the NAFA IT modernization program will be able to produce results by the end of 2020.
Listed below are a number of issues for consideration with priority, in the immediately following period,
supposed to generate short, medium- and long-term effects:
- Full computerization of the enforcement activity (on central databases, together with all the necessary
support modules) and implementation of bank account garnishment by electronic means;
- Simplifying the analytical sheet on the taxpayers and introducing the obligation to enroll the territorial
administrative units in the national system of records of the goods subject to local taxation;
- Development and implementation of the centralized risk analysis system;
- Implementation of the standard audit file (SAF-T);
- Development of IT applications necessary for collaboration with partners from EU Member States (AEOI,
MOSS, VIES, etc.)
- Modernizing the NAFA portal and facilitating taxpayers' access to tools dedicated to taxpayer assistance.
Page 56 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
Beyond doubt, the results obtained during the RAMP implementation are significant. The benefits of the project
are very visible both inside and outside the Agency. Among the notable results of the project are:
- redefining the strategic framework of the agency and its operationalization at the level of all managers,
- significant development of the analysis and development system,
- implementation of new facilities dedicated to taxpayers (one stop shops in every tax administration unit,
- unified system for the management of the waiting lines,
- expansion of the single contact number at national level for the call center),
- redefining the risk analysis system, and many others.
The creation of the internal analytical capacity for analyzing the tax gaps was also an important moment in the
development of NAFA, considering that the activity of revenue collection is currently in reorientation of the main
approaches in the field. Prioritization of the action fields is closely tied to the results of such analysis.
We also mention the Agency's present efforts towards opening up to the demands of the private business
environment, increasing transparency of decision-making and, above all, involving all the stakeholders in
establishing NAFA's immediate priorities.
According to this guideline, representatives of the business environment, the academic environment, other
institutions directly involved in simplifying the administrative procedures in Romania have been invited to
support NAFA in initiating and developing the dedicated and timely activities necessary to modernize the
institution.
Page 57 of 58
The World Bank
Revenue Administration Modernization Project (P130202)
ANNEX 6. SUPPORTING DOCUMENTS (IF ANY)
Page 58 of 58