Killing aid

About the author
Chola Mukanga is a London-based economist specialised in justice and constitutional economics, transport and development, local government, and urban / rural issues. He is the founder of the Zambian Economist.

Dambisa Moyo is tired and frustrated by the aid apparatus that has not only come to “trap” poor and indebted African states but is, in her view, the root cause of poverty. The central argument of Dead Aid is that aid is the fundamental cause of poverty and therefore eliminating aid is critical to spur growth in ailing African states. Aid is the disease that we must treat to bring us back to full economic health. A bold and daring statement built around the central belief that aid distorts incentives among policymakers and society at large. It makes governments less accountable to their citizens and has led to civil wars, rampant corruption (electoral and otherwise) and has been central to an undercurrent of irresponsibility culminating in increased and self-reinforcing poverty since the end of colonialism. None of these arguments are new of course, but Dambisa is probably the first economist to boldly claim that aid causes poverty.

This article is a review of Dambisa Moyo "Dead Aid"

Also in openDemocracy:

Anna Lekvall, "Democracy and Aid: the missing link" (13 May 2009)
If aid is the disease that causes endless bleeding, to stop the bleeding you simply need to stop aid, the only challenge therefore is how to do it. The Dead Aid solution is a five year exit strategy built around the idea of incentivising poor countries to access finance on international markets, supported by the tripod of microfinance, trade/foreign direct investment (FDI) and remittances. In the Dead Aid world there’s a stash of money out there on the international financial markets that is just waiting to be tapped by any African country willing to invest in a credit rating. If African countries can enter these markets and borrow, it would provide the right incentives to spark good governance since the international markets would be more willing to “punish” bad behaviour compared to those that provide aid at infinitum. In other words, borrowing through international financial markets is a sort of "self commitment mechanism" to good governance, and with that comes better long term prosperity.

It is certainly likely to be slightly more expensive than “easy money” that concessional loans and grants bring, but by rejecting these overtures nation states will find themselves on a better path to prosperity. The trouble is that African governments have limited incentives to do this on their own, though some have made progress in this direction, so they need to be compelled through the Dead Aid proposal of terminating aid completely within a five year period.

Radical stuff indeed, but is it too radical? Depending on your view of aid, this is either the most ingenious idea you have ever come across or the most naive, if not downright reckless. At this present time when many western countries are tightening their belts and some are seeking aid themselves due to the fallout from the credit crunch and many people are growing weary of Darfur, Guinea Bissau, Mauritania and Zimbabwe, the Dead Aid message is likely find some appeal not just in your Daily Mail or Fox News . I am afraid to say, and with deep sorrow, that the Dead Aid proposal falls far short in many areas, with at least four worth highlighting.

First, there’s a general lack of clear analytical rigour evidenced by elementary confusion in key areas : correlation/causality issues; definitional problems; poor evidence on policy counterfactuals; incomplete and unbalanced citation of evidence; and, perhaps more worryingly lack of general familiarity with refined areas of existing literature. Too many problematic issues to cover within this short review, but some key examples are worth highlighting.

In a number of instances Dead Aid embarrassingly confuses correlation with causality. For instance the correlation between foreign aid and savings, which Dambisa takes as strong evidence that foreign aid reduces domestic savings. It does not take a genius to work out that one expects poor nations to correlate with reduced domestic savings, and in so far as foreign aid is prevalent in poor countries, the issue of correlation between higher aid and low domestic savings becomes meaningless. Perhaps more worryingly is that in a number of places Dead Aid seems to rely on evidence just from single sources that always reinforces its general argument that aid is bad. So when Dead Aid posits that remittances are more effective than international aid, it ignores other studies that have shown remittances can also be a “curse”.

Evidence of poor research abound, with one of the glaring examples being the lack of reference and consideration of new emerging literature led by Daron Acemoglu and others on the importance of drawing a distinction between proximate and ultimate causes for underdevelopment. In many respect if aid was going to be a factor it would be nothing more than a proximate cause because ineffective aid preys on inefficient states, which are strongly determined by the existing distribution of power in society (ultimate cause).

Secondly, the treatment of aid in a homogeneous and aggregate way is particularly problematic. Dead Aid defines aid as the “sum total of concessional loans and grants”, but excludes “emergency aid” e.g. help for Darfur or the Asian Tsunami. There’s no distinction within Dead Aid between budget support, infrastructure aid, person to person aid, heath related aid, grants or concessional loans for discretionary spending. It is all discussed under one umbrella and handed the same fate. This is a remarkable assumption, especially given that the same book acknowledges the effectiveness of the Marshall Plan which largely focused on infrastructure spend. Surely the Marshall Plan demonstrates that a more nuanced assessment of aid has the potential to reach different conclusions? We may for example find that some of the aid is bad, some good and some requires further study.

This distinction is also important because we are now seeing a plethora of literature that suggests that some mechanisms work better than others e.g. cash based incentives as recently argued by Göran Holmqvist . When Britain gave Zambia £40m in 2007, I remarked that "I hope the money was new but not given freely". It presented a new opportunity for Britain to think outside the box and consider the possibility of converting this "new cash" into long term Kwacha bond claims of Zambians on the Zambia Government. Such a move would have helped restore much needed accountability in our system as well as strengthening our debt management practices. Britain could have allocated a share of the bonds to civil servants as part of civil service pay increase and so forth. The underlying point here is that not all form of aid leads to perverse incentives and indeed not all forms of aid perpetuate dependency. To put all aid in one basket makes the book appealing to the uninformed but it does not make for convincing argument to policymakers.

Thirdly, Dead Aid is characterised by a plethora of inconsistent arguments. A key example that stands out is the emotive issue of Chinese investment. Dambisa dedicates a whole chapter explaining why the “Chinese are our friends”, largely arguing from their historical involvement in Africa and their renewed commitment to trade and FDI. However, against a backdrop of Dead Aid’s “anti-dependency” rhetoric , the chant for China appears odd. Let us be clear, China is not only bringing FDI to Africa but it has also brought concessional loans and long term dependency. Zambia’s external debt has now risen to about $2bn since the HIPC completion point, a significant part of that is through new agreements with the Chinese government and Chinese businesses.

A closer look at Angola reveals the same truth. Not only is China investing heavily in that country but in exchange it is tying Angola and other countries to China for a long time reducing their options to renegotiate in the future. That is not necessarily bad, but if the central worry is that dependency leads to ineffective governments with poor incentives we should be honest enough to consider the possibility that China’s closeness to many African governments (which are not all democratic) may have similar negative impacts as aid. In addition, a more refined assessment of the China – Africa relationship would reveal that the issues go far beyond simple FDI but also relates to military cooperation and sometimes creating instability in various parts of Africa (see Michael Sata’s paper). More recently we have witnessed General Nkunda during the recent upsurge of violence in DRC use the China-DRC deal as a pretext for his insurrection, part of the so-called Coltan wars.

Another glaring inconsistency relates to the preferred metrics of measuring the extent of Africa’s aid led failures relative to the assumed metrics for measuring the success of proposed solutions. In assessing the state we are in, Dead Aid relies on national indicators such as GDP, life expectancy, level of external debt and so forth. However when it comes to assessing the extent to which the proposed solutions might be useful the book does not always stick to a consistent set of measures. For example to support the argument for microfinance, we are told Grameen Bank has helped lift many poor people out of poverty through helping “bank the unbankable”.

I am a fan of microfinance and a strong believer that aid properly directed at providing the right sorts of incentives, like IFAD are pursuing in Zambia to boost rural finance through the NARBARD style model , can produce positive results. What is particularly puzzling about the Dead Aid position is that if the metric for judging the effectiveness of microfinance is “lifting people out of poverty” at the micro level why not use the same measure for aid? If we are going to argue that remittances help bypass bureaucracy and can be effective in tackling schooling, not necessarily increase national GDP, why can’t we accept that the metric of “school attendance” is just as good a measure for assessing the effectiveness of certain aid interventions? Conversely if we are to judge the failure of aid interventions on their inability to raise national GDP (all things being equal) why don’t we accept that no empirical study to date has demonstrated that large initiatives of providing microfinance (e.g. in Bangladesh) has led to increases in GDP? The underlying point is that Dead Aid too often moves around between inconsistent measures for the problem and suggested solutions. Incidentally the IFAD initiative is a good example of effective aid that is unfortunately ignored by Dead Aid.

Fourth and finally, the solutions proposed by Dead Aid are ineffective. This is not surprising because without a clear definition of the problem, it is inevitable that the solutions would not work. But even if one was to accept Dead Aid’s basic premise that aid is bad, its solutions come far short. In order to assess whether any proposals would present an overall improvement beyond the status quo, we need to define what happens in the counterfactual carefully and then judge that against proposed policy initiatives.

In our scenario the counterfactual is the situation where we continue with the current process. We know already that Dead Aid has not demonstrated that this situation would lead to more aid driven poverty . More importantly, evidence in recent years from Zambia, Uganda, Kenya Tanzania and other countries shows an improving picture in terms of economic performance. This doesn’t mean aid causes good performance, but it does suggest growth is possible in the presence of aid even for nations at the bottom. It is therefore possible that in the presence of aid we may witness an improving counterfactual over time.

Two important questions flow from the above discussion : (1) what would be the impact of turning off the aid tap on poor nations relative to the counterfactual?; and (2) would these developing nations be able to borrow on the international markets, as an alternative to aid?

On (1) there’s no doubt that the answer largely depends on the economic and political situation in relevant nation states. For those countries with 20 % – 50% of national budgets supported by donor partners the adjustment would be too difficult and politically infeasible within the suggested five year time frame. The failure to implement their budgets would significantly weaken the human and physical infrastructures rendering these states ungovernable. More importantly locally targeted aid that is spearheaded by many aid organisations divorced from budget support would dwindle, possibly leading to multiple failed states. Dead Aid misses the point that even without aid, the incentive for military coups and emergence of vampire states would be remain because of the lucrative mineral wealth that exists. So the incentives for seeking alternative funding through financial markets as a way of survival are not always going to be as strong. Simply put for some countries turning off the aid tap would lead to chaos and breakdown in the rule of law.

As for replacing aid with borrowing, dwindling international capacity following the credit crunch (likely to persist beyond 2011/12) means there is no immediate prospect of accessible markets with significant cash to spread around. Even if African governments had strong incentives to enter these sorts of arrangements and with good initial credit ratings (which is highly unlikely) the process may be too prolonged and the outcomes would be uncertain given prevailing global economic conditions.

In short on both theory and practice, Dead Aid falls far short of what is expected of a book advocating such a radical proposal of “turning off the aid tap”. If there’s any consolation in this assessment, it is that Dead Aid will hopefully not find any intellectual traction. The analytical consensus remains that aid is important and the challenge is how to make it smarter, better and ultimately beneficial to the poor. This question has never been more urgent given the limited aid resources around. Dambisa is certainly right that now is the time to examine these issues and we can certainly do better than the present.