What hinders development budget implementation?

The budget of the Government of Bangladesh is about what the government promises to give the citizens in a fiscal year in the form of macro-economic and structural policies, economic services and investments; in turn what the citizens are required to give the government in the form of taxes, now and in future (borrowing is nothing but a tax on future generations); and what the consequences of these will be on issues we care about, such as prices, production and employment.
The economy weathered a political tsunami in fiscal year (FY) 14. The major challenge for the government was to consolidate the gains in political stability since the beginning of 2014; restore and strengthen the climate enabling production, trade and investment; and sustain progress on human development. The FY15 budget needed to set the level and composition of expenditures and financing embedded in structural reforms in such a way as to meet these challenges. It is a promising budget, with mixed reform priorities, that will need continued attention to challenges in some key reform areas such as the public banks, public-private participation, land for infrastructure and industry, and decentralization (to broaden and deepen the role of local governments in budget formulation and implementation).
This analysis focusses on the Annual Development Programme (ADP) - the most critical instrument from the point of advancing the country's infrastructure and human development agenda.  The FY15 ADP accounts for 32.1 per cent of the FY15 budget.
Once again, the fiscal year will start with a very large ADP, of about Tk 800 billion, comprising of 1,034 investment projects in the main programme and another 959 projects awaiting approval.  Tk 30.07 billion is also set aside for unapproved projects. Around 308 "expired" projects have also been reportedly included; not completed within the scheduled deadline; the government has extended the time to utilize the funds allocated for these projects.

The original ADP size always becomes an issue at the time of budget preparation.  Most often, analysts and observers describe it as too large to implement.  To date, they have been proven right in varying degrees, as shown in the table here.  Note that only 67 per cent of the FY14 original ADP has been spent in the first eleven months. To achieve the revised budget target, ADP spending in June alone had to be 24 per cent of the original ADP, implying as usual hurried release of funds to bolster spending record without necessarily actually spending the funds. Surely, this breaches expenditure discipline.
When we say that the ADP is too large, we do not mean it is too large relative to Bangladesh's needs.  Take the Tk 800 billion (equivalent to over $10 billion) ADP for FY15.  Bangladesh needs to invest $7.4 to $10 billion per annum in infrastructure alone. From this perspective, the ADP is in fact not large enough. However, the question is implementation capacity.
The ADP for FY15 is too big to be fully implemented, given the lack of project readiness, procedural constraints, capacity deficiency in the line ministries, and the usual delays in procurement of goods and services. The number of projects in the ADP has been swelling year after year because of low project completion rate in relation to the inclusion rate of new projects.  New projects for setting up large power plants, constructing expressways and so on are certainly needed, but they ought to be implementation ready in addition to being economically viable. There are several good ongoing projects in the ADP which need to be speedily implemented to address the country's infrastructure, energy and human development needs. It is time to take a proper account of the trade-off between the populist mileage obtained from appeasing constituencies favouring more and more new projects versus the longer term economic and political gains from completing high-priority ongoing projects.   
Implementation deficiencies are not necessarily in the areas of policies and availability of project opportunities.  Take the case of the road sector.  A Road Master Plan covering the next 20 years has been prepared to coordinate development and expansion of the road network. There is a Strategic Transport Plan covering the next 20 years to reduce road accidents and eliminate traffic jams. For maintenance of the road network, a Road Fund Act 2012 has been approved by the Cabinet. A Strategy Master Plan for Chittagong port covering the next 50 years will be done by June 2015. About 174 investment projects on rehabilitation, repair, maintenance, and extension of road networks are now under implementation.   
Yet progress on building and maintaining roads remains in a sorry state. Take the case of the Dhaka-Chittagong four-lane highways. The government in January 2010 signed deals with three firms to expand the highway in three years. But the work began in 2011, as problems relating to the appointment of project's consultant took a year. The work was further held up due to scarcity of dirt to fill the lanes, inadequate allocation and rent-seeking by local politicians and cronies from the contractors. The delay led the ministry to revise the Development Project Proposal (DPP) and re-fix the cost. Later, the cost was raised again when the DPP was revised a second time, bringing some crucial changes to the project, including its design. Cutting a long story short, the project is still less than 40 per cent done.
According to a recent review, on average no less than half the project time elapses by the time 20 per cent fund is released, primarily because of poor project readiness. Apart from procedural constraints, there is a capacity problem in the planning cell of executive agencies and line ministries.  Delays in procurement and recruitment of consultants remain endemic in infrastructure projects.  Low levels of delegation of power in the government approval system also contribute to delays. Resettlement and land acquisition delay project implementation by 2.0 to 2.5 years, on average.  Project budgets for land acquisition are made in line with a 1982 ordinance which stipulates payments for acquired land to be made at government fixed rates, not market rates. The ordinance does not allow giving compensations to people affected by a project, resulting in legal disputes that delay project implementation.
Surprising as it may be for a poor country like Bangladesh, financing is not really a binding constraint on the implementation of the development budget.  The government has undertaken measures to increase domestic resource mobilization through better compliance, collecting arrears, extending value added tax (VAT), rationalizing existing tax policy and improving tax administration. Even during a bad year such as FY14 from a revenue collection point of view, the revenue surplus (revenue receipt minus non-development revenue expenditure) in the revised budget amounted to over 547.2 billion. External assistance is the other significant source in financing development projects. The opening pipeline as of June 30, 2013 stood at about US$ 16.62 billion.  Disbursement as a percentage of pipeline opening balance was 18.2 per cent in FY13.  This is estimated to have declined to 16 per cent in FY14. With rising aid commitments, the opening pipeline rose to about US$ 19.4 billion by end-June 2014. Slow implementation of projects leads to slow disbursement of aid, not vice-versa.
Incentive structures play a crucial role when it comes to strengthening capacity in development administration. But, while recent literature draws the conclusion that incentives matter, incentive schemes do not always contribute to improved implementation performance and may even represent an obstacle to capacity building. The limited impact of many incentive schemes is linked to competing disincentives leading to capacity erosion, such as sustained low wage levels, lack of transparency in recruitment or promotion based on political affiliation. When it comes to public sector management, existing capacity level and political will are crucial elements.
'Political will' is commonly used as a catch-all concept, the meaning of which is vague unless we assign it a specific and narrow meaning such as 'the determination of an individual political actor to do things that will produce a desired outcome'. Defined this way, 'political will' is a neutral concept. It can produce both benign and invidious outcomes. Lincoln had 'political will' in abundance, but so did Hitler. We therefore need to ask what purpose it is serving. The over-riding purpose to which many policy makers apply their 'political will' is the enhancement of their influence and reputation. If such policy makers believe that constructive reform serves those purposes, they are likely to pursue it. If reform appears not to serve those ends, they will do little. Thus efforts need be made to persuade politicians that prioritized implementation of the development budget do indeed serve those ends. This is easier said than done because reform champions often lack influence over key players. But there is much they can do to shape public debates and elite perceptions. This is one sphere in which attempts to raise the bar on the pace and quality of implementation progress are often worth making.
The point that opportunities, policies, and finance are not the binding constraints on the implementation of the development budget in Bangladesh can hardly be over-emphasized. The binding constraint is implementation which in turn is constrained by inadequate capacity, an incentive system that fails to reward risk-taking to achieve development results, and the lack of appropriate political will.
(The authors are respectively Lead Economist and Country Director (Bangladesh, Bhutan and Nepal), the World Bank. Email: mmahbub@worldbank.org)