Richard Allen

Fiscal Affairs Department (IMF) & METAC

Workshop on MTEFs

December 16th–19th, 2014, Beirut, Lebanon

Institutional, Sequencing and Capacity Building Challenges

I. Outline of Presentation

II. PFM reform on the rise

III. Evidence of success and failure

IV. Strategies for reform

V. Why things go wrong

VI. Lessons for the future

VII.Capacity building issues

VIII.Conclusions

2

II. Interest in PFM Reform has increased enormously in the past 20 years

• PFM received its most recent impetus from the “New Public Management” reforms in New Zealand, Australia, and the UK in the 1980s.

• Reforms were largely successful in these countries, it seemed. Aim was to transform input-oriented, control and compliance focused bureaucracies into flexible, responsive, efficient service delivery units focused on outputs and outcomes.

• Reforms were copied in Europe, the US and advanced Asian countries. Some countries did better than others.

• From the mid-1990s, many emerging market and developing countries tried to copy advanced country PFM reforms

• There was a surge in technical assistance (TA), often supported by the World Bank and IMF

3

II. Some Evidence of the Growth of TA……

• Between 1990 and 2005 the number of WB projects with a significant Public Sector Reform component quadrupled

• OECD data show that donor funding for PFM reform rose from US$ 85 million in 1995 to US$ 931 million in 2007

• IMF PFM TA almost tripled from 2006 to 2014

4

II. What types of PFM reform are we talking about?

• Medium-term budget/expenditure frameworks

• Program/performance budgeting

• Deconcentration / Decentralization

• Development of agencies/outsourcing of services

• Government-wide IFMIS

• Accrual accounting

• Fiscal risk management

• Fiscal rules and fiscal responsibility laws

• Independent fiscal agencies

5

III. What evidence of success……?

• MTBF: Countries with (some form of) MTBF have increased from fewer than 20 in 1990 to more than 130 in 2008.

• Performance Information: since 2007, 80% percent of OECD countries produce performance information in their budget, and in 2011 about 2/3rds indicate that they have a performance budgeting framework.

• Fiscal Reporting: countries reporting at least a financial balance sheet to the IMF has increased from 21 in 2004 to 41 in 2011.

• Accrual Accounting: 26 countries have moved to full accruals and this will reach – reportedly – 63 within the next 5 years

• Fiscal Rules: Countries with fiscal rules have risen from 5 in 1990 to 82 in 2013.

• Fiscal Councils: the number of countries with Fiscal Councils grew from about 6 in 1990 to around 27 in 2013.

6

III. Improvements have been less in Developing Countries …..

1.5%

-9.8%

5.4% 5.1%

-1.9%

25.5%

2.9%

-15.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

Latin America and Caribbean

(LCR)

South Asia (SAR) East Asia and Pacific (EAP)

Africa (AFR) Middle East and Northern Africa

(MNA)

Europe amd Central Asia

(ECA)

Total

% In

cre

ase

/De

crea

se

Growth Change in the Average CPIA Rating for PFM for the period 2001-2011

7

IV. Advanced Reforms not immediately suited for Developing Countries

• Need for supporting “basic reforms”

– Credible and comprehensive budgets

– Smooth and controlled budget execution

– Accurate and timely fiscal reporting

– Effective external audit and oversight

– Concentration of PFM functionality (Treasury, Debt, Budget, internal audit)

• To be done first? The purists say yes!

• Not everything can be done at once: sequencing the reform strategy is necessary in any case

8

• No coherent framework of macroeconomic analysis, or credible fiscal projections

• An annual approach to preparing the budget• An “incremental” approach to preparing the budget• Undisciplined approach to supplementary appropriations• Limited capacity in the MoF to analyze spending requirements

on a sector-by-sector basis, or to exercise a “challenge function” with spending ministries

• An approach based on analysis of inputs, with little attention given to the outputs or outcomes of spending programs

• An absence of aggregate expenditure ceilings on which spending ministries are requested to prepare their bids.

• Under-developed ICT systems

IV. Example: Two models of budget preparation (1) Traditional

9

• Budgets are based on an analysis of medium-term economic projections and fiscal indicators or targets – supported by a fiscal responsibility law

• A medium-term budget framework (MTBF) is in place, together with estimates of key fiscal risks

• Spending programs are presented in terms of program classification and analysis of outputs and outcomes

• Disciplined approach to supplementary appropriations• MoF exercises a strong challenge role and uses advanced analytical

techniques • Budget process is supplemented by a comprehensive expenditure

review• Focus on accountability and transparency (e.g., citizen’s budget)• Modern ICT systems (e.g., FMIS) support these advanced systems

IV. Example: Two models of budget preparation (1) Evolving

10

V. Different Reform Strategies were developed during the 2000s

• “Basics first” (or even one thing at a time) – provides capacity

• “Buy the Ferrari” – provides motivation

• Law driven reforms – provides authority

• IT driven reforms – provides a framework

• Everything at once – because issues are related

• Supply/donor-driven – because its free money

One of the more sophisticated strategies:

• The platform approach – focuses on synergies and real outcomes, and is realistic about capacity

• Tailor-made

>>No strategy has been fully successful;

>> Even limited successes often superficial11

V. “Basics First” usually not enough

• Not always focused on issues at hand

• Provides little motivation/buy in from political level, donors or counterparts

• Does not always address related reforms

• Does not provide a roadmap of where to go

• Underestimates the pace at which reform can take place

• Challenges of reform are often more institutional and political than technical

12

V. Supply/donor-driven agenda

• Overestimates capacity; burden for ongoing budget process

• Focused on formal rather than de facto reforms

• Takes little account of local circumstances and institutional constraints (blueprint driven)

• Long list of to do’s = overloaded agenda

• Sequencing = everyone takes a slice of the pie

• Political will/ownership lacking

• Pace usually overambitious

13

V. Issues lie deeper than sequencing

• Reforms implemented without clear reflection on what problems they would solve

• Blueprints do not work: do not take into account context, culture and political economy

• Incentives overlooked: power and money interest block real change

• Leadership and change agents are critically important, but often underestimated

• Finance ministries cannot push reforms by themselves, nor can line ministries

• Diagnostic tools are useful in identifying systemic weaknesses but do not provide reform strategy

• Skepticism on achievements of “New Public Management” in OECD

14

V. Who are the Bad Guys?

• Consultants: Want quick “results” without accepting context; do not provide honest feedback to either countries or IFIs

• IFI’s: Stuck to their blueprints? Do not want to get hands dirty. Need to spend money.

• Countries themselves: Are too ambitious, or just want to go through the motions; do not want to invest enough of their own people

• Civil servants: Do not really want change. See reform as a danger for their interests

>>> All of the above?

>>> Or misunderstanding of change processes in institutional development?

15

VI. What do we know about successful change from country case studies?

Positive:– Sweden: fiscal/budget management reforms in the early nineties

– China: treasury single account/cash management reforms

– Brazil: fiscal management/decentralization

– Egypt: budget classification, TSA

– Kuwait: macro-fiscal forecasting

Negative: many examples– Performance Budgeting in OECD

– MTEF in many countries worldwide

– IFMIS in many countries worldwide

– Reform “walls” often obstruct institutionally challenging reforms

>>>Successful reforms often preceded by powerful economic or institutional crisis!

16

VI. What are the elements of a successful reform?

– Real problems are addressed

– Strategy is aligned with context, culture and political economy

– Leadership, authority and ownership are essential

– Incentives and motivation of participants are managed well

– Capacity and capability are aligned with agenda (and part of the reform)

– Pace and magnitude of reform is realistic

– Process change has to go hand in hand with institutional change (this applies especially to Ministries of Finance)

– Reform agenda allows learning and change

– Bureaucrats need to feel stress not to Fail (not to Succeed)

– Planning, sequencing and prioritization are important (but different paths can be taken)

– Last but not least: use the power of transparency and accountability

17

VI. What have we learned about change since the 1960s: the “Hiding” Hand Principle and PDIA

In the 1960s economist Albert Hirschman found:

• People only undertake reforms if they underestimate the difficulty initially

• Successful change is usually: 10% imitation, 90% invention

• Only if forced to struggle to succeed the force of “creativity” is unleashed

• Conclusion: Reform is always a hard slog

Fifty years later Harvard Economist Matt Andrews calls this:

>>>Problem Driven Iterative Adaptation: Reforms require identifying problems, and taking first experimental steps to get quick wins. Then learning, building support and capacity are a long journey. Take steps in a problem driven iterative process that promotes adaptation.

>>>Not a tidy process; can be perceived negatively. 18

VI. What has been learned from advanced countries?

• Reform does not usually proceed on a recognizably predictable path

• Advanced countries have tended to approach reform in a logical order, namely more basic reforms are carried out first, more advanced reforms later

• Don’t attempt too many reforms at once• Reforms may take many years to complete,

requiring patience and perseverance• Reform is driven by a range of factors, internal

and external, political and technical

19

VI. Follow MTFF ＞ MTBF ＞ MTPF sequence (“platform approach”)

• Many basic elements of fiscal reporting in place, but countries need to strengthen forecasting fiscal developments and related fiscal risks

• Put in place basics – solid macro forecasting, credible budget, top-down budget process and medium-term fiscal objectives , then add more binding direction

• Develop stronger interaction between central government and sub-national governments

• Change parliamentary procedures to follow a top-down sequence

• Put in place restrictions on overspending during budget execution and gradually introduce limits on multi-annual spending commitments

20

VII. Capacity building issues

• Strengthen the organization of the MoF• Strengthen center of government’s role in policy

coordination• Change (radically) role, leadership and skills of

central budget office• Build budget and finance departments in LMs• Strengthen linkages between MoF and other

ministries• Avoid the “fragmentation” problem• Learn from experience of advanced countries

21

VIII. Conclusion

1. Developing countries struggle with both technical and institutional aspects of reform, despite the huge growth of TA on PFM topics

2. Some positive examples of PFM reform – including MTBFs - can be found but also many examples of partial success, or failure

3. Advanced country models and experience offer useful guidance but are not templates for reform

4. “Basics first” and the platform approach also offer many important lessons but other elements need to be woven into a successful reform strategy

5. For medium-term frameworks, the MTFF ＞ MTBF ＞ MTPF sequence provides a useful guideline for reform

22