Changing Trends in ODA
For many years, beginning from the 1950s, South Asia was the poster-child of foreign aid, known also as Official Development Assistance (ODA), offered by the Advanced Industrial Countries (AIC) as well as the Communist countries to the developing world. South Asia was seen as the first battleground of the cold war for the hearts and souls of the Third World. India and China were projected as the symbolic protagonists of this epic struggle between democracy and communism and foreign aid was seen as the currency which could influence the course of this struggle. In the halcyon days of foreign aid in the 1960s, posters of earnest young Peace Corp workers- the children of the Kennedy era of an idealistic United States- vaccinating impoverished Pakistanis or teaching school to earnest Indian villagers were part of the popular image of foreign aid. U.S. food surpluses distributed as aid under the PL480 programme were expected to remove hunger and provide jobs to the rural unemployed through various public works programmes. It was believed that aid, wisely invested in building power stations, roads, bridges, schools, hospitals, even industries, would banish poverty from South Asia, and help to sustain the democratic project. Five Year Plans were seen as the appropriate vehicle for absorbing and programming foreign aid because the planners could model the two resource gaps of savings/investment and foreign exchange which were seen as the principal constraints to economic growth in developing countries.

Much aid has since been invested in South Asia over the last four decades. Between 1980 and 2001, US$ 17 billion has flown into South Asia as ODA. However, in recent years, aid flows into this region have been exposed to steady decline in both absolute and relative terms. Table I shows that ODA inflows per capita as well as a % of GDP declined between 1990 and 2001. This decline was registered in all countries of the region, except Pakistan. The share of aid flowing to South Asia as a proportion of total ODA also declined from 18.5 per cent in 1980, to 12.6 per cent in 1990 and, eventually, 11.7 per cent in 2001. In real terms total aid received in 2001 was discernibly less than in 1980. The only country which enjoyed an increase in aid in recent years has been Pakistan which led to an increase in per capita inflows in 2001 compared to 1990 as well as in the ODA/GDP ratio. As we will discuss later, Pakistan's unique experience derived largely from its changed strategic circumstances in the wake of the war in Afghanistan.

As a consequence of this restructuring of global aid flows, South Asia has become increasingly less dependent on aid in relation to its development process. Some countries, such as India and Pakistan, have been conspicuous in their resort to private capital inflows to compensate for the decline in ODA. Table 1 also shows the changing composition of external capital inflows and the declining shares of aid in relation to underwriting total capital inflows. India is now for all practical purposes no longer an aid recipient, even though in 2001, it did receive US$1.7 billion in ODA. This amounts to 0.4 of its GDP and a small part of its total public expenditure. A great part of this aid inflow to India comes in the form of non-concessional loans from multilateral financial institutions.

South Asia's reduced dependence on aid owes in large measure to its robust export performance in the area of goods and services since the 1990s. However, this export boom, particularly in the area of goods, has been heavily concentrated in North America and the European Union, and for countries other than India, on a narrow range of items such as textiles and ready-made garments or tourism. This has opened up new sources of dependency and erosion over policy autonomy.

Changing Donor Perspectives
The decline in aid inflows into South Asia is reflected in the changing perspective of aid donors to the development discourse which underlies donor-recipient relations. The principal aid donors to South Asia, led by the World Bank, have begun to change both the composition of their aid and also the underlying policy advise associated with such aid. In South Asia, ODA flows had traditionally been heavily concentrated in the more capital intensive areas of physical infrastructure such as energy, transport and communications and even industry. Until the mid-1980s, multilateral institutions such as the World Bank and Asian Development Bank were the principal financers for infrastructure projects and were even financing investment in state owned enterprises. Bilateral donors such as U.K., Federal Republic of Germany (FRG) and Japan were particularly active in financing investments in the power, transport and communications sectors. In the 1960s the World Bank was the principal financer of the state owned Ghorasal Fertilizer Factory in Bangladesh and in the 1970s the ADB was the principal investor in the Ashugonj Fertilizer factory. In the 1960s and 70s, bilateral donors were particularly active in financing public sector investments in the industrial sector. The U.K. and FRG pioneered public investment in the steel sector in India which was matched by the USSR investments in the steel sector as well as a broad range of SOEs in India and Pakistan, including what is now Bangladesh. The USSR finance for the steel mill in Karachi in the late 1960s was one of the biggest projects of its kind in Pakistan. These investments in capital intensive, highly visible public projects, provided tangible evidence of donor realpolitik in the 'great game' in South Asia.

The changing composition of ODA coming into South Asia in the 1980s and 90s reflected the growing influence of ideology over realpolitik in the aid practice of both bilateral and multilateral donors and its intrusion into their aid priorities. While the U.S. was always reluctant to invest in public sector industry, it continued to support infrastructure projects and was one of the principal financiers of the Rural Electrification Project in Bangladesh beginning in the late 1970s. Although the principal aid donors to South Asia were never averse to using the leverage provided by their aid to influence the political complexion of regimes, their aid leverage in the last two decades was mostly directed to influencing policy choices, at least in South Asia.

The change in the policy regime and direction of aid to South Asia in the 1980s reflected the growing disillusion with the effectiveness of aid not just to South Asia but to most of the developing world. By the end of the 1970s it was increasingly believed in the AICs that the massive aid flows to South Asia in the last two decades had neither alleviated poverty nor generated sustained economic growth. This disillusion with aid originated from among the tax payers of these aid giving countries as much as from the particularist constituencies of the 'right' and 'left' who questioned the quality of aid effectiveness. In the AICs, the resistance to a rising tax burden was growing and tax payers were particularly incensed that their taxes may end up in developing countries to be dissipated in wasteful public expenditures permeated with corruption. The juxtaposition of persistent poverty with the growing affluence of a narrow elite in the developing world enabled tax payers to join hands with aid critics in questioning the efficacy of aid. Within the developing world the costs of aid dependence were being recognised and the hegemonic influence of aid donors on the policy discourse of aid dependent countries was being challenged. Academic work on the limitations of aid in stimulating development was very much in evidence.

As the cold war drew to its conclusion, it was no longer acceptable for once strategically favoured states to go on misusing aid with impunity. The main challenge to the sustainability of aid budgets came from the disillusion of tax payers in the North who questioned the complicity of the aid agencies in the donor countries in contributing to this misdirection as well as misuse of aid and their collusive role in building up a class of people who prospered from aid at the expense of the majority of the citizens in developing countries (DCs). The response of aid agencies in the AICs to this rising sense of outrage in the donor countries was thus driven both by the expectation that this disillusionment with aid could be reversed as well as by their compulsions for institutional survival. Aid agencies, seeking to protect their budgets focussed on two themes in seeking to redesign aid strategies:

(i) Getting policies right. (ii) Redirecting aid to the poor.

The second part was, however, largely subordinated to the first because it was believed by the dominant aid donors through the decade of the 1980s that the right policies would stimulate growth which in turn would alleviate poverty. In order to get policies right, aid was increasingly offered on conditional terms that policy reforms, on lines suggested by the donors, would be put in place in the respective developing countries (DCs). This agenda for policy reform was, in turn, heavily influenced by the ideological input emanating from the Reagan and Thatcher administrations which underwrote the so called Washington Consensus. In country after country, the World Bank and IMF, known collectively as the Bretton Woods institutions (BWI), put in place stabilisation programmes followed by a package of structural adjustment reforms (SAR) inspired by the Washington Consensus.
It was, with some distinguished exceptions- in the DCs and the transitional economies (TEs)- the apparent failure of the aid driven reforms of the 1980s to either promote sustained growth or alleviate poverty which has now inspired a further change of direction in donor aid strategies. The sense of frustration amongst the taxpayers of the North had by now extended from the `right' to the `left' led by the NGOs, radical academics and church groups. The critics projected the 1980s as an era of failed reforms, which not only did not improve growth but made a small fraction of these Third World countries very rich whilst the poor remained poor. The `right' continued to challenge the very assumptions of aid and remained unimpressed by the decade of reforms initiated in many developing countries under the leadership of the World Bank and IMF.

Putting Governance First
To cope with critics from both the 'left' and 'right' the new focus on aid strategy in South Asia appears to be directed to the establishment of good governance and targetting aid to the poor through what James Wolfensohn, the incumbent President of the World Bank, termed the challenge of inclusion. The literature of the World Bank in the 1990s indicated that the World Bank, at least, had recognised that a combination of getting policies and governance right was likely to alleviate poverty. The World Bank's widely discussed empirical work on Assessing Aid claimed that 'with' sound country management, 1 per cent of GDP in assistance translates into 1 per cent decline in poverty. Thus, it stated that a US$ 10 billion increase in aid would lift 25 million people a year out of poverty- but only if it favours countries with sound economic management. By contrast, the Bank paper argued that an across the board increase of US$ 10 billion would lift only 7 million out of their hand to mouth existence if economic management was weak.

This World Bank study further argued that 'improvements in economic institutions and policies in the developing world are the key to a quantum leap in poverty reduction'. Such effective use of aid is also seen to complement private investment. Promoting aid effectiveness thus demanded the use of aid in strengthening institutions as well as policies and bringing about an active engagement of civil society in the design and delivery of aid. These conclusions of the World Bank study are apparently derived from intensive empirical work on aid effectiveness based on reviewing a large sample of DCs and aid projects.

The original paper on Assessing Aid contained a number of serious flaws in the assumptions as well as design of the analytical model used in the study whilst their empirical evidence merited more careful scrutiny. The original definition of sound management incorporated a mix of three policies: reducing the budget surplus as well as the rate of inflation and realising increased trade openness. These reforms were packaged with institutional quality which was defined as an admixture of strength of the rule of law, quality of the public bureaucracy and pervasiveness of corruption. It would be necessary to examine the metric for such abstract concepts as rule of law and bureaucratic quality before assessing the weights assigned to these four variables and three sub-variables of institutional quality. Such an exercise would permit a fuller appreciation of the empirical work correlating GDP growth with economic policy and institutional quality. It will, however, be argued that the available evidence from the South Asian experience does not conclusively support the conclusions of the World Bank study on the role of governance.

Notwithstanding its technical limitations, the World Bank study on Assessing Aid was an important document. Its currency and extent of analysis on aid effectiveness and the attempt to use empirical evidence to question the efficacy of a decade of donor driven policy reforms underwritten by conditional offers of ODA, made it a landmark document. The study appears to reflect a willingness of the World Bank to encourage a more endogenous process of promoting policy reforms within not just South Asia but also in the Third World. This rethinking in the World Bank was further reaffirmed by a series of conferences organised by the World Bank around the world to address the issue of policy ownership as a critical ingredient in any move to promote better governance. The emphasis by the World Bank on prioritising poverty was highlighted in their World Development Report (WDR) of 2001 whilst the role of institutions was highlighted in the WDR of 2002.

This rethinking of aid policy was not limited to the World Bank. Other aid donors such as the OECD, the U.K., Canada, the Nordics countries, and the Netherlands also sought to link good governance with aid effectiveness and argued that policy ownership was crucial to the exercise of effective governance over development policy in the developing countries. All such agendas to promote governance reform focused on the need to prioritise the poor in the donor's allocative regimes. Such poverty alleviating agendas are now increasingly concerned with issues of empowerment of the poor and of women as integeral to the process of poverty alleviation.

Contradictions in the World Bank's New Aid Strategy
Bank programmes designed in an era when growth was prioritised over poverty have not quite worked out how poverty alleviation could be integrated into the earlier generation of structural adjustment reforms (SAR) programmes. The belief of the 1980s that high growth will reduce poverty may be something of a truism. However, the earlier reforms neither generated sustained growth nor alleviated poverty so that a new development model to reconcile growth with poverty alleviation is still awaited. The current practice of simply adding on poverty related projects to the old adjustment model appears to be a self-defeating exercise. If the original development design was itself perpetuating poverty, accentuating inequalities and empowering a small elite who use their wealth to monopolise state power, a few so- called poverty centred projects will not ensure a sustainable assault on poverty or the empowerment of the poor. Prioritisation of poverty in the aid agenda thus demands that the original design of the reform process has to incorporate institutional mechanisms for ensuring inclusion of the poor in the development process, giving them competitive access to the market and institutionalised claims on resources, and scope for participating in political power. Attempts to step up allocations for the poor through targetted aid is hardly likely to disturb the realities of power in most DCs and transitional economies. Serious contradictions also appear to arise between the prioritisation of governance in aid agendas and the BWI commitment to policy lending. The distorting impact of policy lending derives from its impact on policy ownership as well as the limited access of the poor to the benefits of such reforms. It has been recognised by all donors from the World Bank to the OECD studies, that reforms without ownership have proved to be unsustainable.

This failure to address the structural sources of poverty and the compulsion to adhere to the macro-economic policy model associated with the World Bank's structural adjustment reforms (SAR) has now been internalised in the Poverty Reduction Strategy Papers (PRSP) adapted by Bangladesh, Nepal, Pakistan and Sri Lanka, under pressure from the World Bank and IMF. This design failure in the PRSPs reflects the weak ownership of the South Asian governments over the current new policy fashion of the donors. It would thus appear that the newthink on aid and its manifestation in the PRSP process has not really resolved the tension between the flawed policy design of the original structural adjustment reforms model and the Bank's new commitment to putting poverty and governance first. The World Bank has in fact not succeeded in developing a coherent macro-model which links such reforms with the process of poverty eradication. Nor is there any indication that policy ownership in the DCs is being more actively promoted rather than talked about. This weakness in the PRSP process has now intruded into the report of the Independent South Asian Commission on Poverty Alleviation (ISACPA) which was approved by the SAARC Summit in Islamabad in January 2004. This report is a useful document but is essentially astructural in its conception and is thus likely to have a minimal impact on poverty in South Asia.

All such arguments about the counter-productive nature of donor driven policy reforms have been part of the critique of foreign aid and external dependence for at least two decades and particularly during the high tide of adjustment reforms in the 1980s. For the academics, NGOs and some political parties who had been challenging the donor driven reform process of the 1980s, it is welcome news that the World Bank has seen the light. Empirical research is now deployed by the Bank to demonstrate that lack of policy ownership contributes to the failure of reforms. They could have learnt as much by a careful reading of writings on the subject published in the 1980s.

A South Asian Perspective
Sound economic policies
The available evidence from South Asia indicates that by the standards set by the World Bank, the region's policy regimes remains reasonably sound. Between 1997-2003, the South Asian countries had, by DC standards, lower fiscal deficit/GDP ratios which on average remained below 10 per cent. The deficits, in South Asia, where they persisted, were designed to accomodate inflows of aid. These budget deficits were not the result of governmental extravagance but part of a structural problem originating in the process of aid dependency. To draw any conclusion, at least within South Asia, about the relative policy merits of the fiscal deficit/GDP ratio would thus appear to be misleading. All South Asian countries have, again by global DC standards, enjoyed relatively low rates of inflation, mostly in single digits. Donors could, thus, not fault the region's policymakers on monetary profligacy.

Trade openness as a measure of policy, has also shown considerable improvement in South Asia, though the region still has some way to go to match the East and South East Asian experience. It is argued that the extent of openness in East Asia is open to question. The covert protectionism practiced by Republic of Korea in the dynamic phase of its growth persisted until well into the 1980s and continues to be practiced today in Japan. It is argued that a policy of domestic protection appears to have co-existed with considerable policy support for exports throughout the decades of high export growth in several East Asian countries from 1965 to 1985. Many of these export promotion measures through the 1960s and 1970s were not very consistent with the tenets of economic liberalisation and would be deemed today as unacceptable by the WTO. Even the South East Asians protected some key parts of their economy and nurtured these for entry into the export market, as for example the case of the Proton car in Malaysia. China and Vietnam, who have been enjoying high rates of GDP and export growth over the last 15 years, for all their reforms, remain even today the most protected economies in South Asia.

Conversely, Bangladesh, Sri Lanka and Nepal's opening up of the economy has not yet yielded the benefits promised by economic reformers. These arguments could be applied even more strongly to Sub-Saharan Africa (SSA) where many countries have liberalised their import regimes at the cost of a deterioration in domestic industry and the ushering in of a process of de-industrialisation. It is thus arguable that open economic policies may be a necessary, but far from sufficient, condition to stimulate growth. The correlation between policy and outcomes needs to be made country specific if we are to draw any policy conclusions as has been the practise in the World Bank's report on Assessing Aid.

Measures of institutional quality
The link between institutional quality and growth in South Asia is far more problematic. Measuring the strength of the rule of law is as difficult as the comparative measure of corruption, introduced by Transparency International (TI). It is thus difficult to assess whether, for example, the rule of law in Thailand was better established than in Bangladesh or Pakistan, in order to explain their consistently higher growth rates. No doubt what passes for the rule of law was more in evidence in India than in Zaire. But for this proposition to hold good it must also explain differential performance within South Asia itself.

As far as corruption is concerned, there is no evidence at hand which would indicate that Indonesia was more or less corrupt than Bangladesh or Pakistan. Indeed countries such as Bangladesh, Pakistan and Nepal, which rank quite unfavourably in the lists of Transparency International, have on average performed better in the last 5 years than many developing countries (DCs). It is by no means conclusive from the evidence provided by Transparency International that South Asia is conspicuously more corrupt than Sub-Saharan Africa or even Indonesia.
As far as a comparative assessment of bureaucratic quality is concerned, it is not clear what measures are used for this by the Bank. Again, it would be difficult to argue that India or Sri Lanka's bureaucracy is less competent than that of Thailand, Indonesia, China or Vietnam as to qualifications, systems of recruitment and career advancement. Bureaucratic quality thus appears to be measured by economic performance and can hardly serve as an explanatory variable for this economic performance. It could thus be argued that the South Asian bureaucracy, compared to that in Francofone Africa, may, on anecdotal evidence, look more meritorious but within South Asia the application of these measures in assessing economic performance would need to be much more sensitively analysed to permit for any conclusions to be drawn.

The East Asian crisis of 1997 suggested that many of the features of weak governance once associated with South Asia such as corruption, crony-ism, political patronisation, lack of transparency, lack of rule of law, personalised regulatory practices, were in existence in East Asia and are only today being identified as explanations for their financial crisis. But these flaws in governance were also present in the East Asian system during its miracle phase when few donors sought to highlight these as constraints to their economic performance. It would thus appear that whilst the World Bank's position on the value of sound governance appears to be intuitively acceptable more robust evidence, within Asia at least, needs to be generated before we can use these measures as a yardstick for guiding aid policy.

The follow up arguments posed by the World Bank for reducing poverty thus also need to rest on more robust evidence establishing the causal link between sound economic management and policy success. Unfortunately, the conceptual link between governance and economic performance remains far from clear. In these circumstances, the recommendation that aid be targetted to low income countries with sound economic management appears to be sensible in principle but difficult to operationalise. Obviously China has fared much better than India whilst Vietnam has done better than Bangladesh in reducing poverty. But whether this owes to their policy and allocative priorities, their better economic management or stronger political commitment, remains again open to debate.

The World Bank study made the sensible point that experience shows that donor financing with strong conditionally but without strong domestic leadership and political support has generally failed to produce lasting change. This statement could certainly be written as an epitaph on the era of conditional aid offered to South Asia (excluding India). There is no evidence that any of these countries made strong political commitments to economic reforms or sought to build a political constituency behind their economic reforms. Even in India the strong commitment demonstrated by Dr. Manmohan Singh to economic reforms, when he took over as Finance Minister in 1991, was not fully endorsed by his Cabinet colleagues in the ruling Congress Party. Thus, the pace of his particular reforms visibly decelerated in the second part of the Congress regime as general elections approached in 1996. The approach of the successor BJP regime in India has faced its own ebbs and flows. The contestation between reform minded cabinet members such as their successive finance ministers or the minister in charge of privatisation and the more swadeshi school of thought remains unresolved.

Using aid to promote good governance
The Bank's recommendation to direct aid to countries with a strong track record of concrete performance behind domestically initiated reform would thus again favour China and Vietnam over Bangladesh or Nepal. The Bank's recipe for dealing with countries with poor policies and no credible reform movement suggests a patient role of disseminating ideas, transmitting experiences of other countries, training future policymakers and leaders. This again is a paradoxical position. Poor policies and a credible reform movements defined by the Bank's yardstick could exclude China and Vietnam but include Sri Lanka or even Nepal who were very faithful adherents to structural adjustment reforms. Their poor outcome may thus originate in weak implementation. This has enabled the Bank to now pass on the responsibility of poor performance in South Asia not to any design flaw in the structural adjustment reforms but to poor governance which remains the responsibility of the host government.

Endogenising policy reforms
Once the Bank and other donors embrace the proposition that reforms depend mainly on domestic political and social factors, the donors have to come to terms with the limited influence they can exercise over domestic policy agendas in South Asia. In the wake of this renovation in the Bank's approach to policy reforms, conditional lending would need to be phased out. The Bank again recognises that conditionality is unlikely to bring lasting reform if there is no strong domestic movement for change. Thus, only when domestic constituencies are committed to reform, adjustment loans and foreign aid can help consolidate policy gains. In such a context the donors can and indeed should do no more than suggest to the concerned governments that they need to get their act together, design reforms and commit themselves to the implementation of these reforms. Out of this reform process the need for aid can be articulated in a variety of areas from Technical Assistance (TA), to budget and balance of payments support offered for a finite period whilst revenue and export earnings capacities are built up.

Throughout South Asia, with perhaps the exception of Bhutan and Maldives, there is no country which lacks the domestic capacity to design its own reforms. This capacity must be mature enough to recognise where skill and knowledge gaps exist so that donor resources can be solicited to fund the necessary Technical Assistance. India has exercised ownership over their reforms and have articulated their own need for Technical Assistance which has, as a result, been much more effectively used than was the case of Technical Assistance imposed from without upon Bangladesh or Nepal. If donors are to recognise the need for policy ownership and the role of civil society in promoting this ownership there is not a great deal that they can do except react to such local initiatives.

Donors have, for too long, attempted to lead reforms. This often follows in the wake of slow progress by a country in designing its own policy reforms. The World Bank or UNDP tend to lose patience with such tardiness and prefer to call in expatriate consultants but with a facade of local participation added on. Donors thus also need patience and self-discipline. They should not make the mistake of promoting ownership which would itself be a contradiction in terms.

The circumstances governing the assumption of local ownership will vary from region to region. South Asia is a region with the strongest potential for assuming ownership over its policy agendas. It has a longer democratic tradition than many other regimes but its roots constantly need fertilisation since persistent malgovernance endangers democratic institutions in most countries of the region. In India, Pakistan, Bangladesh, Nepal and Sri Lanka free elections have ended in periodic regime changes. But the working of parliamentary institutions leaves much to be desired and the recent lapse into authoritarian rule in Pakistan, Nepal and even Sri Lanka points to the shallowness of these roots. The press is relatively free and lends itself to extracting transparency from the government of the day. However long exposure to autocracy and a tradition of bureaucratic concealment leaves much scope for making public affairs more transparent. Both accountability and transparency need, however, to be extended to the private sector which tends to conceal a variety of misdeeds which are not exposed to the public or penalised in the market place because of their collusive association with the state and the imperfections of the market.

The role of civil society
South Asia also has a highly pro-active civil society manifest not just in the profusion and quality of its NGOs, some of which are world famous, but in the growth of civic activism. Its professional resources are comparable to any in the Third World so that its capacity to design its own reform agendas waits on the will of the governments of South Asia to follow India's lead in reducing their dependence on donor advise and on the part of donors to practise what they preach over policy ownership. South Asia, outside India, has for two decades been innundated with expensive expatriate TA, usually of poor quality and with negligible use value due to lack of ownership.

The role of aid in moving South Asia towards better governance is thus likely to be minimal since in most of Asia, donors lack the leverage to do this. This has not prevented them from trying to influence not just economic policy but the promotion of transparency and even free elections in a variety of countries. In such a process, faced by recalcitrant governments, donors have sought to go over the head of governments to deal with civil society. Unfortunately civil society itself is an elusive concept. Donors, in search of civil society in South Asia, have often been tempted to use their aid to fabricate a civil society by using NGOs as a surrogate for civil society. This donor approach to building civil society through NGOs creates new channels of dependency manifest in the plethora of NGOs throughout South Asia whose institutional existence and the livelihood of hundreds of thousands of their employees now depends on foreign aid. In South Asia, when the dependence of the state on aid to underwrite its activities has been in visible decline over the last decade, the external dependence of the NGOs has expanded exponentially. This escalation in aid dependence of most of the NGOs raises serious problems for their sustainabili