Halving poverty in HIPC countries by 2015: How costly if achievable?

Available online at www.sciencedirect.com

Journal of Policy Modeling 33 (2011) 213–225

Halving poverty in HIPC countries by 2015: How costly if achievable? Jacinta Nwachukwu Salford Business School, University of Salford, Maxwell Building, Greater Manchester M5 4WT, UK Received 1 January 2010; received in revised form 1 June 2010; accepted 1 July 2010 Available online 20 August 2010

Abstract This article assesses the likelihood and costs of halving the poverty headcount ratio by 2015 from its 1990 levels in sixteen post-HIPC–MDRI countries. An optimistic pro-poor growth scenario indicates that, on average, they will attain this goal 2 years before the end date. An estimated annual cost of 16 percent of the recipients’ GDPs suggests that currently available funds will be sufficient to finance the MDG poverty target, provided that they achieve a 6 percent annual economic growth, improve their equality of incomes and implement policies to raise absorptive capacity to levels obtained by East Asian countries in the mid-1990s. © 2010 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved. JEL classification: F21; F34; F37; H62; O55 Keywords: HIPC; MDRI; Dual-gap; Domestic savings gap; Trade gap

In September 2000 the member countries of the United Nations set a goal of halving extreme poverty throughout the developing world from its 1990 level by 2015. The extent and depth of poverty in low- and middle-income countries (LMICs) was defined as the percentage of population living at under US$1.08 dollars a day measured in 1993 Purchasing Power Parity (PPP) terms. To achieve the UN Millennium Development Goals (MDGs), estimates made by earlier studies, including Hanmer, Jong, Kurian, and Mooij (1999), the World Bank (2000), Devarajan, Miller, and Swanson (2002) and the UN Millennium Project Task Force (UNDP, 2005), suggested that the least developed economies, particularly the highly indebted poor African countries, would need to grow on average by 8 percent a year over the period 2000–2015. However, sustaining a steady 8 percent growth path over time would likely require a much faster rate of investment accumulation than many of these countries would be able to finance from conventional sources. E-mail address: [email protected]. 0161-8938/$ – see front matter © 2010 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.jpolmod.2010.08.002

214

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

Thus, to ensure that the poverty MDG is successfully reached in severely indebted poor states, appropriate resources need to be made available by their donors. This paper therefore assesses the total cost of, and likelihood that the goal of reducing by half the 1990 income poverty headcount ratio (MDG1) in highly indebted countries will be achieved by the 2015 target date. The discussion contributes to the literature on the subject in three ways: First, it focuses on sixteen out of the seventeen post-Highly Indebted Poor Countries (HIPCs) that have taken steps to reform their economic polices and institutions and which were also provided with multilateral debt relief in January 2006.1 These post-HIPC-Multilateral Debt Relief Initiative (MDRI) economies comprised: Benin, Bolivia, Burkina Faso, Ethiopia, Ghana, Honduras, Madagascar, Mali, Mozambique, Nicaragua, Niger, Rwanda, Senegal, Tanzania, Uganda and Zambia. Guyana was excluded because its exports plus imports were outlying at 180 percent of GDP averaged from 1980 to 2004. Second, it simulates future income poverty headcount ratios in each of our sixteen postHIPC–MDRI states under different poverty reduction elasticities and economic growth scenarios. The estimated poverty elasticities cover the range from minus 0.3 to minus 1.5 commonly reported in the literature with respect to the responsiveness of poverty rates to changes in mean income and its distribution in developing countries over time. Our projections highlight the impact that alternative steady growth-cum-inequality prospects could have on poverty incidence. Third, it projects the aggregate external finance requirements adjusted for total interest charges on the stock of foreign debt arising from a planned growth rate. The analysis uses five different estimates of income growth rates suggested by earlier researchers as necessary for meeting the MDG1 over the 10 years 2005–2015. Then too, most existing studies have tended to assume that the potential financing-gap will be covered entirely by non-debt creating flows, including official grants-in-aid. But data shows that the provision of most of the external finance to lowincome states has been in the form of long-term official loans, both with and without concessions. This implies that the planned poverty-reducing growth targets envisaged by the organisers of the MDGs will almost inevitably be accompanied by further external indebtedness in many of the least developed countries. The article is organised as follows. Section 1 briefly reviews the literature on the different approaches for analyzing resources needed to meet the MDG1 by 2015. Section 2 describes past trends in headcount poverty rates. Section 3 estimates the likely future headcount poverty ratio arising from five different growth-inequality forecasts and compares the results with the MDG1 target. Section 4 projects the aggregate net external financing needs of our sixteen postHIPC–MDRI countries. Section 5 remarks on the policy implications of our results. 1. Costing the MDGs: a brief review of literature Since the signing of the UN Millennium Declaration, a number of researchers have evaluated the total costs and benefits of alternative policy strategies for MDG1 and the sharing of costs among national governments, private investors and official donors. One of the earliest of such studies on the cost of halving global poverty by 2015 was the Report of the High Level Panel on Financing for Development (more commonly known as the Zedillio Report) published by the United Nations Development Programme (UNDP) in 2001. The Zedillio report (named after the former Mexican president who chaired the panel) suggested

1

See the World Bank website: www.worldbank.org for the key features of the HIPC and MDRI Programs.

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

215

that an average of 22 percent investment as a share of GDP a year would be required to achieve a 6 percent annual average economic growth rate in the group of developing countries over a 10–12-year period. This corresponded to an additional US$50 billion per year. Several later studies were commissioned by the UNDP and World Bank to provide detailed regional cost estimates using the approaches and economic growth assumptions underlying the Zedillo report. They included a paper by Pettifor and Greenhill (2003) which calculated that the total cost of financing the investments required to generate the MDG1 target growth in output in the developing world at US$76 billion a year; one-and-half times the US$50 billion reported by the Zedillo Commission. This study estimated that US$46 billion per year would be required to meet the MGD1 in the 42 HIPCs. A World Bank study by Devarajan et al. (2002) estimated that the resources needed to achieve the MDG1 for Africa would range from US$54 to US$62 billion a year. A major criticism of these papers, including the Zedillo Report, derived from the ad hoc manner with which the calculations were produced. Consequently, the projections provided by these early UNDP and World Bank studies were judged to merely indicate “the order of magnitude” of the extra funds required to achieve the MDGs. By the latter half of the decade, an increasing number of individual country-level studies began to emerge in order to deal with the inadequacies of the World Bank and UNDP global estimates of the cost of meeting MDG1. Kakwani and Son (2006) predicted that an average per capita aid of about US$35.4 per person would be required to reach the MDG1 in 15 Sub-Saharan countries under the pro-poor growth strategy, rising to US$129 per capita under an anti-poor scenario. A UNDP study published in 2003 calculated that in order to achieve MDG1, Cameroon would need to grow at 7, Malawi at 6 with Tanzania and Uganda at 5 percent. Bussolo and Medvedev (2006) employed the Maquette for MDG Simulation (MAMS) to analyse the economy-wide consequences of the pursuit of accelerated growth and MDG policy targets in Honduras with its infrastructure bottlenecks in education, health, water, sanitation, roads and energy. MAMS drew heavily on the Computable General Equilibrium Model (CGE) popularized by Dorosh and Sahn (2000), Addy (2001) and Löfgren, Lee, and Robinson (2001). A key advantage of MAMS is that it explicitly builds into the simulation the growth implications of the interaction between the package of policy interventions that contribute to the achievement of the different MDGs (poverty, education, infant mortality, maternal mortality, water and sanitation) and the additional resources – physical and human capital, labour and intermediate inputs including public infrastructure – that become available to the rest of the economy. Under an assumption that MDG financing is entirely covered by foreign grants-in-aid, the authors predicted that the minimum aid per capita needed for achieving MDG1 in Honduras would be US$190 in their business-as-usual scenario. The MAMS model was also used by a host of authors to assess the effects of alternative MDG target policies on the various economic sectors in developing economies. They included the papers by Löfgren and Díaz-Bonilla (2006) and Bussolo and Medvedev (2007). Evidence from the foregoing literature suggests that the different methodologies utilized to date have produced varied estimates of the resources needed to attain the MDG targets. Nevertheless, their results indicate that the total cost of halving the poverty headcount ratio by 2015 would likely be higher than resources currently available to developing countries. Given the large magnitude of the needed external resource flows, our post-HIPCs in particular may need to consider alternative or complementary sources of financing, including improvements in the efficiency of service delivery, domestic taxes, or domestic borrowing. Besides, if their governments are able to address the existing infrastructure gaps in information technology, energy, water, sanitation and rural roads, the additional growth rate in income could accelerate further, enhancing poverty reduction and the attainment of the MDGs.

216

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

Fig. 1. Poverty trends in developing regions (percentage of people with income below US$1.08 a day, 1993 PPP).

2. Trends in poverty and development finance The decision to make poverty eradication the central objective of the development agenda in 2000 followed from the fact that at the turn of the 21st Century, more than a billion people (almost one in five of the world’s population) continued to live below the internationally agreed poverty threshold with an average income of 77 cents a day in 1993 PPP terms. Fig. 1 shows that in 1981 some 40 percent of the total population of low- and middle-income countries (LMICs) lived on incomes below the international poverty line of US$1.08 per day in 1993 PPP terms. By the end of that decade, the number of extreme poor had declined to 29 percent of the total population of the LMICs in 1989. The percentage of population in LMICs still mired in severe misery declined to 18 percent in 2004. This last figure suggested significant progress towards the MDG1 target. However, the evolution in average headcount poverty ratios for LMICs as a whole masks important disparities across the four major developing regions. In 1981, Sub-Saharan Africa, including South Africa (SSA or Africa hereafter), South Asia and East Asia all had more than 40 percent of their citizens living on less than the internationally agreed US$1.08 threshold. But while the proportion of poor people declined in South and East Asia over the subsequent years, it increased steadily from 46 percent of population in 1984 to a peak of 48 percent in 1996 in SSA. Although poverty rates declined to 41 percent in 2004, there were then still around 299 million people in SSA living in abject misery compared to 168 million in 1981. The rise in the absolute number of poor there was mainly because of continuing high population growth rate, but it was also due to civil unrest and a succession of natural disasters which pushed large numbers below the international poverty line Aryeetey (2004). Most importantly, the literature attributed Africa’s relatively slow reduction in poverty to a deterioration in the region’s economic performance by comparison with the rest of the developing world Sachs and Warner (1997). However, the commodity boom of the latter half of the past decade should have improved things. Africa’s growth has certainly increased markedly. Nevertheless, many authors,

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

217

including Dollar and Kraay (2002), expressed concern about the rising income inequality in SSA and emphasised the importance of access to credit, investment in basic infrastructure and social services for an improvement in the wellbeing of the poor. 3. Will extreme poverty be halved in post-HIPC–MDRI economies by 2015? Debates about the likely future rates of poverty in developing countries are common, although views differ on how poverty lines should be set and what measures of poverty should be used. In this section, we concentrate on the headcount ratio which is the mostly commonly used poverty measure. It captures the percentage of population living below the MDG specified income threshold of US$1.08 dollars a day in 1993 PPP terms. We present our simulations for the impact on the headcount poverty ratio of changes in mean per capita GDP growth rates and income inequality for each of our sixteen states. Unlike previous studies, our assessment will involve five alternative scenarios of per capita GDP growth with associated poverty-reduction elasticities in order to account for the influence of policy reforms on the poor. These are: (1) Pro-poor growth which reflects the trend in the headcount poverty ratio that will materialize if our post-HIPCs achieve a target GDP growth rate of 5.6 percent per year and an income poverty elasticity of minus 1.5 averaged from 2005 to 2015. (2) Improved equality which assumes that an average GDP growth rate of 6.6 percent a year will be combined with a modest reduction in income inequality. The anticipated average income poverty elasticity is minus 1. (3) Unchanged inequality with an annual average GDP growth rate of 7.6 percent and a poverty elasticity of minus 0.7. (4) Slightly widening inequality with an average GDP growth rate of 8.6 percent a year. It is also assumed that a marginal worsening in income distribution will cause a fall in the average poverty elasticity to minus 0.5. (5) Anti-poor growth which assumes that an average GDP growth rate of 9.6 percent a year will be accompanied by a considerable deterioration in the equality of income distribution leading to a low average poverty elasticity of minus 0.3. Fig. 2 depicts our alternative estimates of the unweighted prospective mean of the percentage of people living below the internationally agreed poverty line at a given time period t for our sixteen post-HIPC–MDRI economies as a whole.2 For all these projections, we have taken the initial headcount poverty ratio in 1990 as the baseline level in line with the original MDG poverty reduction targets. For brevity, the discussion here is confined to the pro-poor growth scenario. The reader may infer the changes in the poverty MDG observed under the other four alternative growth forecasts. Comparing the end points, 1990 and 2015, headcount poverty ratios declined significantly under each of our growth-inequality propositions. Such emphasises the importance for poverty reduction of high per capita income growth rates sustained overtime. But the MDG1 target is reached at different dates. For instance, when the additional income is more equally distributed as it is under our optimistic pro-poor growth strategy, the percentage of poor people is cut in half by 2011 from 46.66 percent 2

The corresponding year-on-year projections for each country of study are available from the author on request.

218

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

Fig. 2. Projected incidence of extreme poverty in 16 HIPCs under different growth scenarios (unweighted average from 2004 to 2050).

of population in 1990 and it later falls to 16.97 percent in 2015. Nonetheless, it is not until 2023 that the absolute poverty headcount is projected to decline to half of its initial figure in 1990. Such conflicting evidence on the outlook of relative and absolute poverty incidence underlines the sensitivity of the MDG target to the choice of the poverty line and the initial poverty rate, as well as mean income and its distribution (Chen, Datt, & Ravallion, 1994; Hanmer et al., 1999; Ravallion & Sen, 1996). 4. The total cost of halving poverty in the post-HIPCs The purpose of this section is to project the aggregate external financing requirements (adjusted for interest payments) of our group of sixteen post-HIPC–MDRI economies under the five different poverty-reducing growth scenarios outlined in Section 3 above. The discussion is developed under: (4.1) methodology, (4.2) the gross external resource gap and (4.3) aggregate net foreign capital requirements. As we noted earlier, the arguments here are confined to the pro-poor growth scenario. Once again, the reader may infer the financing-gap consequences of our other four alternative growth-income inequality choices. As in Section 3, the reported estimates are the unweighted annual averages for our sixteen post-HIPCs as a whole. 4.1. Methodology The literature reviewed in Section 1 showed that there is more than one method that can plausibly help to estimate the cost of achieving the MDGs. The choice of technique is largely influenced by the limited availability of data, as well as the need to simplify the calculations. Consequently, most studies tend to use some variant of the Chenery–Strout dual-gap financing model with its assumption of a hypothetical optimistic medium- to long-term growth outlook and contributions of labour, physical capital and total factor productivity Chenery and Strout (1966). These theoretical values, including the investment-to-GDP ratio, export growth rate, import propensity and efficiency gains in capital and labour were usually based on favourable assumptions and estimates drawn from an IMF-World Bank Debt Sustainability Analysis for the country of study.

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

219

An application of the Chenery–Strout dual-gap model involves four steps: First, an estimate of the target GDP growth rate needed to attain the Millennium Development Goal of halving poverty between 1990 and 2015 under our five alternative poverty elasticity and inequality scenarios. Achieving MDG1 would require a reduction in headcount poverty of around 2.74 per annum over the 25-year period. Therefore, the MDG1 target per capita income growth rate is calculated by dividing the estimated 2.74 percent by the poverty reduction elasticity underlying each of our five alternative income-inequality scenarios. In addition, utilizing the population data from 1990 to 2004, we fitted a time-trend regression of the natural logarithm of actual population against time. The coefficients were then used to project the total population with corresponding annual growth rates for each of our post-HIPC countries, starting from 2005. The target GDP growth rate was then obtained for every year by subtracting the estimated annual population growth rate from the MDG1 required per capita growth rate. Second, a calculation was made of the excess of the ex ante gross domestic investment over savings (gap 1) using estimates of the MDG1 target GDP growth rate, the incremental capitaloutput ratio (ICOR) and the marginal gross domestic savings rate. Third, the excess of the ex ante “minimum” required imports over exports (gap 2) was calculated using estimates of the marginal propensity to import and the expected real rate of growth of exports. Fourth, estimates of the ex ante foreign exchange gap (gap 2) were then compared with those for the anticipated gross domestic savings gap (gap 1). The larger of these two gaps at any given time period t determines the total financing-gap associated with the proposed target growth rate in the gross domestic product. This potential resource gap must be filled through foreign financing on an annual basis if the MDG1 target GDP growth rate is to be achieved.3 4.2. The gross external resource-gap Based on our optimistic performance parameters, we simulated the aggregate resource-gap which needs to be covered year-on-year by foreign donors if MDG1 is to be attained. Fig. 3 presents the results of the projected unweighted annual average gross external financing need for our sixteen HIPCs as a whole under the five alternative growth-cum-income inequality scenarios described earlier in Section 3. These aggregate external flows were frequently derived from the dominant gross domestic investment minus savings gap which was persistently the key constraint limiting accelerated growth in output in many of our post-HIPC economies. Such provides additional support for the decision of many authors, including those in the UN Millennium Task Force (UNDP, 2005), to estimate the financing needs of developing countries using the savings-gap approach rather than the two-gap (with foreign exchange) or three-gap (with fiscal gap) models. It is noteworthy that the estimated aggregate external financial flows remain persistently positive throughout the projection period. We may therefore infer that none of our post-HIPC-states is likely to graduate from the need for donor assistance in the foreseeable future despite our assumption of favourable adjustments in underlying performance parameters. On the whole, the results of our simulations indicate that the gross external financing needed to halve poverty in our sixteen post-HIPCs would be around US$897 million a year averaged from 2005 to 2010, increasing to US$1.2 billion (14.01 percent of their GDPs) from 2011 to 2015. Thus, given an average ratio 3 Description of model assumptions, parameter estimations and associated financing gaps for each of the sixteen countries in our study is available from the author on request. This data was drawn from the World Development Indicators for 2010 (World Bank, 2010) provide online by the Economic and Social Data Service.

220

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

Fig. 3. Potential gross external financing requirements under different growth scenarios (unweighted mean 16 HIPCs).

of foreign aid to GDP of around 11 percent in 2008, our post-HIPCs will need approximately 3 percent of their prospective GDPs a year in further financial assistance from their sponsors in the last 7 years of the MDG programme. Interestingly, our projections for the annual average resource-gap of 14 percent of the recipients’ GDPs, are within the MDG financing-gap of 10–20 percent of GDP reported by the UN Millennium Project (UNDP, 2005) for HIPCs such as Ghana, Uganda and Tanzania. Then too, despite their intricate production functions for MDG outcomes, our results are comparable to the 13.2 percent of GDP per annum projected by Bussolo and Medvedev (2007) as the level of foreign grants-inaid needed to reach the poverty MDG in an archetypal developing country like Ghana using the MAMS approach. Such similarity underscores the relatively small differences between the key parameter values underlying both the MAMS and the traditional financing-gap models for those developing countries where an econometric estimation of relevant growth elasticity of poverty is problematic due to data limitations. This outcome supports the argument by Sanchez and Vos (2007) that the calibration of parameters for use in any simulation exercise is essentially one of an educated guess. The use of such “guesstimates” does not necessarily invalidate the basic operation of alternative approaches provided that these are in line with the stage of development, income, market conditions and other key characteristics of the country which is being analysed. However, these results should be sensitivity tested as they are here under our five alternative growth-inequality scenarios in order to inform policy makers of the implications of different courses of action (Easterly, 1999; Gottschalk, 2004). Moreover, the external resource flows estimated by these conventional financing-gap models are not equivalent to the countries’ expected current account deficits as they exclude profit remittances, interest charges on the stock of loans and unilateral transfers. The manner in which they are adjusted for debt interest payments is discussed in Section 4.3. 4.3. The aggregate net external capital requirement Statistics from the World Bank database suggest that in addition to the need to finance the dominant resource-gap, most developing countries also require further transfers in order to

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

221

Fig. 4. Projected aggregate net external financing requirements under different growth scenarios (unweighted annual average for 16 HIPCs from 2005 to 2015).

cover interest charges on both their new disbursements and their outstanding stock of loans. The most problematic states are the group of severely indebted poor countries. This is largely because of their lower rate of domestic savings and greater requirement for investment in the basic infrastructure needed to support the higher desired GDP growth rate of more than 5.6 percent a year. Such implies a larger residual financing-gap than that reported in Section 4.2. Fig. 4 charts the results for the unweighted mean net external finance in constant 2000 US$ on the left vertical scale averaged from 2005 to 2015. This amount will be required to support the higher MDG1 target growth rate and cover the residual interest charges on external loan obligations under our five alternative growth-inequality prescriptions. To assess the financial implications of the poverty alleviation goal for donors, we depict on the right hand vertical axis of Fig. 4, the estimated total net external finance needs expressed as a percentage of the preceding 3-year average of the group of OECD-DAC countries’ projected GDPs that have been promised in official assistance to developing nations. These estimates assume that each donor country will endeavor to raise its ODA commitment to the target ratio of 0.7 percent of its potential gross domestic product a year. As expected, the potential total net external capital needed to achieve the MDG1 in our post-HIPCs varies considerably across the different growth-cum-inequality trajectories. It is projected to range from 0.55 percent of the prospective total OECD countries’ target aid commitment to 0.83 percent averaged from 2011 to 2015 depending on whether the growth pattern was pro- or anti-poor. To make sense of how our estimates compare with those of previous studies, we adopt the approach outlined by the UN Millennium Project Task Force and express our results in terms of total net capital requirements per person per year. To halve the share of the extreme poor in total population, a typical post-HIPC would require an aggregate net transfer of US$65 per capita per year from 2005 to 2010, rising to US$81 from 2011 to 2015. This last is slightly higher than the US$70 per person per year observed for our sample of post-HIPCs in 2008. This should encourage augurs the organisers of the MDGs to lobby donors to increase aid to the world’s poorest states to the full 0.7 percent target. However, any addition to aid flows must be supported by sound economic policies, good governance and the rule of law in the beneficiary nations.

222

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

5. Policies for mobilizing the additional resources needed to fund MDG1 Four major policy implications emerge from our estimates in Section 4. First, for many of our post-HIPCs with their underdeveloped infrastructures and limited supply of skilled and tertiary-educated workers, the anticipated surge in external resources may well be associated with an upward pressure on their relative price of non-traded goods and services, in particular health and education (UNDP, 2005; Vos, Sanchez, & Inoue, 2007). Such real rate overvaluation would erode the competitiveness of domestic import-competing and export firms with a consequent reduction in domestic savings therefrom. What is more, if any resulting deterioration in the trade deficit is covered by further external borrowing, then the anticipated spending on MDG-related programmes will exacerbate the debt problems of our post-HIPCs. Therefore, a key priority for achieving and sustaining the MDGs in our severely indebted poor countries is to implement policies aimed at relieving the constraint posed by the expected real exchange rate appreciation effect of foreign capital inflows. Vos et al. (2007) outlined the importance of enhancing the capital absorptive capacity of beneficiary countries through investment in supply side activities, including the provision of infrastructure, information and communication technology, agricultural support, extension workers, teachers, nurses and doctors. Such investments would help remove bottlenecks on domestic production and so should be made a precondition of additional assistance to our-post-HIPCs. Then too, evidence suggests that government spending in many LDCs is dominated by a demand for non-tradables largely arising from public sector wages in administration, the army and the police. Thus, another way of ameliorating the harmful effect of large capital inflows may be for public spending on such “non-productive” activities to be cut, although any decision to retrench civil servants may be politically difficult to implement and there is a case for raising the salaries of those who remain in the hope of reducing corruption and the emigration of skilled workers. Nwachukwu (2008) discusses further policy initiatives to contain the real exchange rate appreciation effects of external finance. They include the adoption of a more flexible exchange rate regime and the easing of import tariffs, quotas and licensing procedures. However, any tradeoff between public revenue and these reforms may undermine the attempt of our post-HIPCs to achieve the MDGs unless donors provide bridging finance to cushion them from the political and economic shocks inherent in such liberalization policies. Second, given the magnitude of the needed aid flows, which may or may not be available from donor countries, particularly in the wake of their austerity measures following the financial crisis of autumn 2007, our group of HIPCs may need to consider alternative or complementary sources of financing; in particular emigrant remittances which are more stable than portfolio investment and international bank loans, not to mention the fact that they represent a direct transfer of finance from migrant workers to their families in their home countries. The international market for remittances is inefficient, uncompetitive and dominated by a handful of key money transfer operators, most notably, Western Union, MoneyGram and Thomas Cook (Solimano, 2003). This paper outlined policies for reducing the costs of sending money to developing countries, as well as for raising competition in the global market. They included a reduction in the cost of licensing for new operators as well controls on the money transfer industry. Then too, measures to formalise the legal status of migrant workers in advanced would encourage their integration into their societies and help provide greater access to the variety of services offered by commercial banks, including the use of ATM cards, telephones and the internet to make transfers. Also, banks in the receiving countries should be supported by their governments to form alliances with money transfer operators in sending countries. What is more, the income distribution

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

223

effect of remittances would be enhanced if banks in the receiving nations were to develop new product lines and accounts that channel migrant foreign currency into poverty-reducing ventures such as housing, education and business enterprise. Third, the mobilisation of the domestic resources needed to achieve the MDGs will necessitate further reforms in tax administration to broaden its base, increase the efficiency of its collection and diversify its sources away from export duties, as well as customs and tariffs on imports. Moreover, simplification of income tax documents and schedules for customs and excise duties could save time, improve transparency and accountability and help increase investor confidence in the revenue collection authorities. Then too, many reforming African countries, such as Uganda, have increased recurrent revenues by replacing taxes on imports with a domestic value added tax (VAT) and by establishing an independent agency charged with the collection and administration of taxes and other public sector revenues. The Monterrey Consensus further identified corruption, inefficiency, embezzlement and smuggling as major barriers to effective resource mobilisation and its allocation to productive investments in many developing countries. Good governance is essential for financial intermediation and requires the rule of law, respect for human rights and civil liberties, market-orientated policies, transparent regulatory frameworks and prudential supervisory systems. Fourth, the diversification of micro-finance services to include savings schemes to fund small and medium scale enterprises (SMEs), particularly for those in households in rural areas and for women, are important for raising income with associated private savings rates. Savings facilities provided by commercial banks are seldom adapted to the needs of rural households and SMEs, partly because of the cost of making small loans. Moreover, given low-population density and poor infrastructure, rural services for the saving, payment and transfer of funds are often limited. As a result, the very few household and SME investors that save do so by stockpiling goods, by investing in commodities like gold and by hoarding cash. Such non-financial forms of savings are sub-optimal in the sense that they are subject to fluctuations in prices, as well as destruction by pests, fire and theft as compared to deposits in formal banks which may re-lend them for further investment. More important still, perhaps, is the raising of auto-finance opportunities through training and the provision of agricultural and other extension services. Such household savings needed to finance improved seeds, fertilizer and small-scale irrigation appeared as if by magic in Asia’s self-evidently profitable green revolution. 6. Conclusions The MDGs have become the yardstick upon which the performance of developing countries and their donors will ultimately be judged at the end of the target year in 2015. In this paper we have looked at whether the income poverty goals are likely to be met across a group of sixteen post-HIPC–MDRI states and how much this will cost under five alternative growth-cum-income inequality outlooks. The main findings may be summarized as follows: First, our estimates, like those of earlier researchers, show that the most likely route for our sixteen post-HIPCs to achieve the MDG1 target is for their real per capita incomes to grow year-on-year by a minimum of 3 percent for the foreseeable future in a more equitable income redistribution pattern. With roughly 46.66 percent of population living in extreme hardship in the base year 1990, our forecasts of a reduction to 23.33 percent under our equitable growth scenarios 1 and 2 suggest that their headcount poverty ratio will be halved by the end of 2013. But the absolute number of poor is projected to be cut in half of initial headcount in 1990 by the MDG1 target end date despite our assumption that the anticipated additions to their per capita incomes

224

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

will be more equitably distributed. Such ambiguity in the attainment of relative and absolute poverty reducing objectives underscores the importance of the choice of the poverty line and its measurement in promoting human development in the coming years when claims are made by the sponsors of the MDGs with respect to target fulfillment. Second, the challenge of halving the headcount poverty ratio by 2015 is generally confronted by the problem of income inequality rather than accelerated growth in income per se. Examples of redistribution policies that could make a more effective contribution to poverty reduction may include the development of family-orientated small-scale enterprises with, perhaps, the provision of micro-finance services including savings, credit, and insurance as an enabling factor. Then too, agricultural extension services, irrigation, farm-to-market roads, rural electricity and water supply, village paramedical services and universal primary education are all self-evident contributors to the well-being of the indigent. But where the provision of subsidized services is concerned, its welfare effects over and above those of allocating the same finance to directly productive investments should always be carefully considered. Third, based on a persistent gross domestic savings gap, the aggregate net external finance needed to halve poverty in a typical post-HIPC–MDRI economy is likely to range from 0.55 percent to 0.83 percent of prospective total OECD countries’ target aid commitment averaged from 2011 to 2015 depending on whether the growth trajectory is pro- or anti-poor. The last of the two is a mere 0.13 percent higher than the 0.7 percent of the potential OECD gross domestic product a year promised in additional aid to the group developing countries. This extra funding, if eventually provided, will almost cover both their potential MDG-financing gaps and the interest payments on new and outstanding debt stock. Our post-HIPCs, for their part, should take steps to implement policies that contain the Dutch-disease effect of external finance, as well as encourage alternative source of foreign and domestic resource flows and their effective use in poverty-reducing ventures such as health, education and business enterprises.

References Addy, S. (2001). Note on computable general equilibrium model for Ghana identifying growth areas. Journal of Policy Modeling, 23, 821–824. Aryeetey, E. (2004). Financing Africa’s future growth development: Some innovations. , available on the Internet: www.g24.org/aryeetey.pdf Bussolo, M., & Medvedev, D. (2006). Millennium Development Goals for Honduras: Current Achievements and Forthcoming Challenges. In Paper presented at the GTAP conference Addis Ababa, Ethiopia, June 15–17, available on the Internet at: http://www.un.org/esa/policy/cairo training mdgs/bussolo and medvedev 2006.pdf Bussolo, M., & Medvedev, D. (2007). Accelerated Growth and MDG Achievement in Ghana: A General Equilibrium Analysis of Fiscal Policy Alternatives. In Paper presented at PEGNET conference Berlin, Germany, September 6–7, available on the Internet at: http://www.pegnet.ifw-kiel.de/papers/conference-2007/bussolo medvedev.pdf Chen, S., Datt, G., & Ravallion, M. (1994). Is poverty increasing in the developing world? Review of Income and Wealth, 40(4), 359–376. Chenery, H., & Strout, W. (1966). Foreign assistance and economic development. American Economic Review, 66, 679–733. Devarajan, S., Miller, M. J., & Swanson, E.V. (2002). Goals for development: History, prospects and costs. World Bank Policy Research Working Paper No. 2819. Washington, DC. Dollar, D., & Kraay, A. (2002). Growth is good for the poor. Journal of Economic Growth, 7(3), 195–225. Dorosh, A., & Sahn, D. (2000). A general equilibrium analysis of the effect of macroeconomic adjustment on poverty in Africa. Journal of Policy Modeling, 22(6), 753–776. Easterly, W. (1999). The ghost of financing-gap: Testing the growth models used in the international institutions. Journal of Development Economics, 60, 423–438.

J. Nwachukwu / Journal of Policy Modeling 33 (2011) 213–225

225

Gottschalk, R. (2004). How much external finance will be needed to meet the poverty-reduction target by 2015? Mimeo Institute of Development Studies, University of Sussex. Hanmer, L., Jong, N. D., Kurian, R., & Mooij, J. (1999). Are the DAC targets achievable?: Poverty and human development in the year 2015. Journal of International Development, 11(4), 547–563. Kakwani, N., & Son, H. (2006). How costly is it to achieve the millennium development goal of halving poverty between 1990 and 2015? International Poverty Centre, UNDP. Löfgren, H., & Díaz-Bonilla, C. (2006). MAMS: An economy-wide model for analysis of MDG country strategies. Washington, DC: Development Prospects Group, DECPG, World Bank. Löfgren, H., Lee, R., & Robinson, S. (2001). A standard computable general equilibrium (CGE) model in GAMS. In TMD Discussion Paper N◦ 75. Washington, DC: International Food Policy Research Institute. Nwachukwu. (2008). Foreign capital inflows and the real exchange rate. Journal of Quantitative and Qualitative Analysis in Social Sciences, 2(3). Pettifor, & Greenhill. (2003). Debt relief and the millennium development goals. In Occasional paper, background paper for the human development report 2003. UNDP. Ravallion, M., & Sen, B. (1996). When Method Matters: Monitoring Poverty in Bangladesh. Journal of Economic Development and Cultural Change, 44(4), 761–792. Sachs, J., & Warner, A. (1997). Sources of slow growth in African economies. Journal of African Economies, 6, 335–376. Sanchez, M., & Vos, R. (2007). Informative note on elasticities and calibration of MAMS. , available on the Internet at: www.un.org/esa/policy/cairo training mdgs/sanchez and vos 2007 elas calib.pdf Solimano. (2003). Remittances by emigrants: Issues and evidence. , available on the Internet at: http://website1.wider.unu.edu/conference/conference-2003-3/papers/Solimano-2208.pdf United Nations Development Program. (2005). Investing in development: A practical plan to achieve the millennium development goal. New York: United Nations. Vos, R., Sanchez, M. V., & Inoue, K. (2007). Constraints to achieving the MDGs through domestic resource mobilization. United Nations Department of Economic and Social Affairs, DESA Working Paper No. 36. United Nations, New York. World Bank. (2000). Can Africa Claim the 21st Century. Washington, DC. World Bank. (2010). World bank development indicators online from the economic and social data service. ESDS.