Budgeting in the Real World - What Do We Know? What Should We Do?

This is the keynote speech given last week, November 13^th, by Antoinette Sayeh, Director of the IMF’s African Department at the UK’s Overseas Development Institute’s annual CAPE Conference in London on why PFM matters, why reforms are difficult, and what we know to make them successful…..

I am delighted to have the opportunity to deliver this keynote address and would like to thank Messrs. Ed Hedger, Kevin Watkins, and Philip Krause for inviting me to this important conference and for that generous introduction.

Let me start by saying that from the IMF’s perspective, good governance is important for countries at all stages of development. Transparent government accounts and effective public resource management are preconditions for sustained economic growth and prosperity. Indeed, budget formulation, implementation, and oversight lie at the core of good economic governance. Strong budget institutions are essential for countries to achieve sound fiscal policies and effective expenditure programs. Budgets can only be spent once. Getting the priorities right all the way from formulation to execution, and being efficient at it, is all the more important. Transparency and fairness are most important in ensuring that expenditures are aligned with broadly agreed priorities, and in securing society’s buy-in. While most can agree to the underlying principles, the hard part is to have systems and capacity in place that actually ensure that they are respected all along the process chain. As so often, the devil is in the detail. 

In my current role, I am confronted with the severe implications of weak Public Financial Management or PFM systems on a daily basis. During the recent IMF and WB Annual Meetings, PFM issues were front and center in the discussions with country delegations. Many sub-Saharan member countries want to pursue ambitious plans to increase infrastructure investment. These countries have low debt ratios and funding for investments would be available. However, without strong PFM, governments cannot ensure that investments bring value for money and that fiscal risk remains contained. So, I have had to advise to move slowly in scaling up investment, even though infrastructure needs are pressing bottlenecks to growth and poverty alleviation. I worry also that the focus on new infrastructure investments relegates to the back-burner the equally important challenge of ensuring adequate resources for maintaining existing infrastructure.

But problems with budget processes don’t only occur in low-income countries. The recent budget crisis in the U.S. brought poor budget practices in the international headlines. In Europe, a monetary union has proven very vulnerable without strong fiscal rules and procedures, including on reliable data provision.

For emerging economies, enhancing PFM is essential to preserving stability during the process of high investment and growth.  Here, strong PFM systems are especially important due to the increased fiscal risks we see appearing from state-owned sectors, as well as emerging social and pension obligations. With strong development ambitions, emerging economies also tend to aim for second-generation PFM-reforms such as performance budgeting, medium-term expenditure frameworks, and accrual accounting. These ambitions reinforce the need to have the ‘basics’ in place.

But let me return to sub-Saharan Africa. To better understand the weaknesses of PFM systems there, I will dig a bit deeper into their characteristics, into what works and what does not. This list of characteristics summarizes my own experience and also a comparative study of African countries, using PEFA scores, prepared by Professor Matt Andrews, who will speak at tomorrow’s morning session.

I want to flag three characteristics on this list:

1.         First, budgets are often better prepared than they are executed: Even though the credibility of the annual budget remains work in progress in many African countries, the budget preparation PEFA scores are on average better than scores for budget execution and oversight.
2.         Second, there is a severe implementation gap: Laws and processes may be in place, but are not necessarily implemented. I am very much looking forward to a more in depth discussion about how to close the implementation gap later in this conference.
3.         The third characteristic to flag is that consolidation of responsibilities often yields better results. The success of Treasury Single Account development and strengthened cash management in some countries is an impressive example of how consolidation can achieve excellent results. For macroeconomic stability, centralized decision-making is important to enforce fiscal rules and budget discipline. These topics, with reliable fiscal reporting, still comprise the “bread and butter” of technical assistance by the IMF in African countries.

The reasons for weaknesses in PFM systems are certainly numerous and differ from country to country. I look forward to hearing more of your insights on this issue in the discussion following my speech. From my experience, four sets of factors play an important role in countries with weak PFM performance:

• First, I have no doubt that the political economy in many low income countries is an obstacle to good PFM practices. Weak governance creates disincentives for transparent budget processes because they would enhance accountability. The fact that budget preparation PEFA scores are better than execution and oversight reflects these political economy dynamics. Preparing a reasonable budget does not interfere with political interests if execution and oversight offer loopholes to reallocate funds.
• Second, the capacity constraints in LICs’ public administration also play an important role. I am sure that many of the practitioners among us could tell frustrating stories about the struggle to train accountants, auditors, and IT experts. And often it is even more difficult to stop them from taking better paid jobs in the private sector once they are well trained. The extensive use of international consultants financed by donors in many countries also weakens the incentive to strengthen local capacity. Using the tremendous potential of modern IT is difficult because the operation of IT systems and the use of the internet remain costly in many countries. Even basic problems like unreliable power supply can turn a well intended IT program into a white elephant. Or they could frustrate the energy and determination of the few motivated staff there are in post-conflict ministries of finance.
• Let me now comment on the third set of factors. Having worked as Minister of Finance in Liberia, I am convinced that a country’s history and culture play an important role when it comes to PFM. After years of civil war, it takes time to re-establish respect for the rule of law and formal arrangements like contracts. One should have no illusions about the impact of introducing state of the art budget legislation in such an environment. Change will only happen once institutions have evolved that are capable of implementing these laws and have the authority to do so. And political will as well as strong leadership are essential elements of institution building
• And now the fourth set of factors. Related to the bias in favor of changing laws and procedures instead ofstrengthening institutions, we also have to critically re-think the role of donors and international institutions, including the IMF. Too often, our expert advisors have strong ideas about what a “good” budget should look like. They then advise on how to change laws and budget manuals to bring them in line with “best practices”. Such legal changes can be made rather quickly which allows us to report progress on PFM reform to our boards. Strengthening the institutions that are supposed to implement these laws is a much more tedious task. It takes much longer and is more prone to failure—something us bureaucrats in Washington and elsewhere don’t always appreciate.

Let me now try to respond to the key questions this conference will try to address: That is, “What should we do?” How should PFM reform be designed to achieve tangible results in enhancing service delivery in LIC’s? What do donors and international institutions have to change to be more helpful in supporting the reform process?

My main message is simple: That is,

• Reforms have to be well tailored to the country and genuinely owned by national governments rather than responding to the sometimes fleeting needs and objectives of the donor community and international financial institutions.
• Or, to put it differently: When defining reform priorities, country specific institutional constraints and political economy considerations have to be taken into account. Cutting edge PFM concepts (such as program budgeting, medium-term budget frameworks, accrual accounting, etc) are often premature in countries that don’t have the basics right. And these basics are critical for fiscal policy: reliable projections of key macroeconomic and fiscal indicators; a credible annual budget; fully-functional input accounting; complete and accurate financial reports; efficient cash management, including a Treasury Single Account; elimination of ghost workers; comprehensive and reliable control systems that avoid the accumulation of arrears; and a robust system of external audit and oversight. But even the audit and oversight basics can be hugely complicated and indeed circumvented by dysfunctional parliaments in post-conflict environments. So, basics are anything but easy!

Let me spell out a bit more what I consider the most important elements of a new reform approach, well aware that the list is not comprehensive and will benefit from our subsequent discussion.

• The first issue is that attention to political economy is essential for the success of PFM reforms. Let me elaborate on two points about how we can better integrate political economy considerations in the reform approach:
  • First point is that understanding the political economy environment requires presence on the ground. At the IMF, more than 70 percent of our TA is now being provided by experts from the regional AFRITAC centers, and through resident advisors. Thanks to generous funding by donors, some of them represented here, our total PFM TA to Africa has almost quadrupled in the last 5 years. The involvement of seasoned experts over a number of years can give them insights about the political economy that help in identifying realistic reform targets. Representation of TA recipients on the AFRITAC steering committees can reinforce ownership of TA.
  • Second point: I am also convinced that the IMF and the donor community should not only be observers and analysts of the political economy in a country, but be a reasoned voice in the process. For example, if TA providers identify a severe PFM problem that is difficult to address in the current political environment, I would encourage them to think strategically about how to influence the political process. In meetings with ministers, parliamentarians, and civil society they should explain the implications of the weak practices and present realistic reforms to address them.
  • I also believe in the power of transparency as a key element of a new reform approach. Especially in a multi-layer government setting or a difficult political environment, publishing information can be a powerful tool to enhance transparency and promote change. Let me tell you a story that I heard from the Nigerian Minister of Finance Ngozi Okonjo-Iweala which echoes earlier experiences in Uganda. The ministry was increasingly frustrated when being accused that the federal government did not make the constitutionally-required transfers to state and local governments on time. So they decided to publish all transfers to regional and local governments on a regular basis in a wide range of newspapers. Suddenly, local citizens were empowered to inquire with their leaders why local services remained mediocre despite all the money received from the central government.
  • In an environment where institutions are weak, prioritization and careful sequencing are essential for the success of any reform effort. Addressing the most severe weaknesses in the PFM system should have priority. Once basic standards are met in core PFM functions, more demanding reforms can be initiated.

There are some good country examples that have followed this approach. Kenya’s macro-fiscal consolidation was accompanied by reforms in basic budget functions. This included the introduction of a comprehensive IFMIS system and substantive improvements in internal audit. The strengthening of the role of the Treasury Department was part of the success story, in line with my previous observation that the consolidation of responsibilities often improves outcomes. With basic functions well established, the focus has now shifted to what I would call “second generation” reforms, including introducing a performance-orientation in the budget and improving the medium-term budget planning process. Moving forward, a major challenge in Kenya will be to implement basic PFM practices in the new counties that have been established following the decentralization initiative reflected in the 2010.

• More emphasis should be given to strengthening PFM systems and processes (that is de facto reform) rather than bringing the legal framework into line with best international practice (which is de jure reform). Unless there are important gaps in existing laws – such as the absence of appropriate rules on accountability and transparency – or the laws create obstacles to practicing good PFM, reforms should focus on enhancing actual practices, especially during the early stages of the reform process. Again, supporting the institutions that have to implement these reforms, including line ministries, local governments, and oversight agencies, is essential for the success of the overall reform.

A good example of a coherent focus on institution building and de facto reforms is Rwanda. Improvements in the budget preparation process there went hand in hand with equally impressive improvements in cash management (especially through the introduction of a Treasury Single Account), public procurement, audit, and payroll management. As a result, budget execution has become better aligned with the original budget, cash flow control and arrears management have improved, real time fiscal information has become more readily available, the possibility of "ghost workers" on the public payroll has been eliminated, and the number of complaints about the public procurement process has fallen sharply. The organization of the finance ministry was also strengthened and refocused on core budget and fiscal tasks.  

Mali is another example where building institutions has proved useful. Although the Ministry of Finance faced historical challenges in 2012, it succeeded in preserving the PFM system throughout the crisis, and previous progress has been used as the basis for reviving the reform agenda as soon as stability was achieved.

• A final comment to make on approach is that realism about the pace of reform is essential to focus scarce administrative capacities: Developing and implementing the practices that are considered PFM’s “gold standard” has taken decades in advanced economies like Singapore and New Zealand. These best practices of high-achievers in the world of PFM provide useful long-term objectives for ambitious LIC reformers, but most LIC’s should focus on more down-to earth objectives. There are quite a few examples where timelines were overly ambitious—even to the point that they slowed down progress in PFM reform because resources were invested in unrealistic tasks.

A positive example of a country that over time gradually established good basic PFM functions is Ethiopia. Continuous reforms since the mid-1990’s have led to significant improvements in core budget practices. Over more than a decade, the government has strengthened cash management (again, through a treasury single account) and established an IFMIS system that facilitates recording and control of budget data. With those basics well established, the government is now aiming for the gradual introduction of output orientation into the budget.

So, let me conclude: What are the challenges ahead?

The reform challenges that countries are facing today are already demanding, but I can see even more challenges ahead. Let me focus here on two:

• First, the accelerating discovery of large mineral and oil resources in sub-Saharan Africa, including in countrires that were not natural resource producers before, will put significant pressure on PFM systems. If large windfall profits hit a weak PFM system, it is likely that the “resource curse” will result in unsustainably high domestic spending and loss of budgetary control. I am afraid that reform incentives to reduce waste would then be substantially reduced. Therefore, it is urgent that we make progress in establishing good PFM practices now so that countries are prepared to manage natural resource revenues well.
• The second challenge is the rising popularity of decentralization. In some countries, like Kenya, decentralization is a response to political tensions between different regions or ethnic groups. In other cases, natural resource discoveries trigger demands for more fiscal independence in the regions where the resources were discovered. These more decentralized spending models have potential for better delivery of services, but in a weak institutional environment they create an additional challenge to establishing good PFM practices.

To sum up, good public financial management lies at the heart of development. Resources are scarce, which puts a premium on using them well. Accountability and good governance are critical, since public financial management is by definition about public money. The primary responsibility therefore lies with governments to ensure accountability to their citizens. But the international community also has a role to play by providing appropriate assistance that strengthens the weakest links first.

Thank you very much, and I look forward to your questions and comments. 

Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.