10 - The Changing Dependence on Donor Countries

Summary

After independence, Mozambique relied on international solidarity and managed donor relations well. Donor dependency entailed loss of agency, allowing donors to challenge government capacity but never its authority. During the 2010s, donors expressed disappointment with reforms and challenged government legitimacy — not only due to developments in Mozambique. Donors are less enthusiastic about harmonised development cooperation and less concerned with aid effectiveness. Aid budgets are under pressure and development finance links more to other foreign policy concerns. Mozambique should expect increasing instrumentalisation of aid budgets. Institutions Mozambique developed to deal with donors are not well suited to present challenges. They focus on less relevant areas of the relationship with foreign countries, which often serve other agendas. Reforms could start with strengthening Mozambique’s Foreign Service as a genuine coordinator of foreign relations and the establishment of greater discipline around national plans and strategies. Institutionalising strong links between the foreign ministry and key economic ministries under the leadership of the prime minister could help.

I Introduction

Mozambique has been receiving foreign aid for more than forty-five years since its independence from Portugal. It would be unrealistic to imagine that Mozambique could have no relations with donors and therefore no dependency. In fact, independence from colonial rule depended on international cooperation and solidarity, including support for its armed struggle. The interactions of the government and the ruling party with donors can be traced back to the early 1960s when the exiled FRELIMO (Frente da Libertação de Moçambique) embryo of diplomacy argued the case for independence with the world’s leaders and nations. There is no doubt that donors have an influence on political and institutional development in poor countries including in Mozambique. This is often the purpose of their interventions, and, indeed, many donor-funded projects and programmes in Mozambique aim at strengthening or reforming the country’s institutions.

How has the massive presence of donors in Mozambique since its independence in 1975 helped to shape key sovereign institutions in the country and how will this affect inclusive growth in the future?

This chapter studies Mozambique’s donor relations and sovereignty. It is organised as follows: Section II provides an overview of the recent performance of donor relations, aid effectiveness, and sovereignty, drawing on available evidence and insights. Section III looks at institutional factors: What are the underlying drivers that influence how donor relations affect economic institutions and sovereignty? Finally, Section IV concludes and looks ahead: given the challenges foreseen for the future, which policy recommendations can be made about what to do to promote future inclusive development?

II How Aid Influences Institutions

There is no generalised evidence that donor aid harms institution building. Although there may be situations where this happens, it is not an inherent feature of donor aid, as shown by Jones and Tarp (2016). They conclude that there is no evidence that aid has a systematic negative effect on political institutions.

Mozambique is, however, a rare case with extremely high donor dependency. It is therefore pertinent to enquire whether there are serious impacts arising from aid dependency and donor relations that significantly hamper sound and sustainable economic growth. Based on the data up to 2004, Arndt et al. (2007) concluded that aid to Mozambique works and has a reasonable rate of return. At the same time, they cautioned that some features of the aid that Mozambique receives might deepen rather than reduce dependency in the medium term. This evidence is solid and therefore important in the debate about the aggregate effects of aid. While the overall impact of aid seems positive, the debate about exactly how aid affects institutions is much less solid. Bourguignon and Gunning (2016) identified six possible channels of influence, grouped here into three clusters:

The first cluster relates to systemic efficiency, ‘Dutch disease’, and the cost of taxation. In a very donor-dependent country, Dutch disease is difficult to mitigate. However, we do not debate this aspect of donor relations further in this chapter as it is essentially one of appropriate macroeconomic management. The same is true in relation to the cost of taxation, and the majority of poor countries are comfortably on the safe side of the Laffer curve. Furthermore, distortive (and regressive) redistribution can be carried out in a variety of ways, not just through taxation.

The second cluster, as argued by Moss et al. (2006) and Birdsall (2007) among others, relates to the direction of accountability and government survival. It points to a view that donors may pervert institutional development and, in some cases, keep regimes afloat that would otherwise have given way to democratic change and more endogenous and sustainable institutional renewal. This belief is often built on the argument that the regime would have performed worse if it had not had access to donor funds, and this situation has created a more uneven playing field for local political opposition and other interest groups. The countries that have been able to raise large sums of donor support compared to their total government expenditure and income may be less concerned with accountability vis-à-vis their own populations and may be under less pressure to build and maintain domestic legitimacy. Such governments, it is argued, have less interest in building and investing in effective public institutions, and aid sometimes even helps to finance the tools of repression.

Similarly, rent seeking and corruption are sometimes claimed to be associated with, or even the result of, development aid. Not surprisingly, there is only anecdotal evidence of this, and the few solid empirical studies that exist point to different results, as highlighted by Bourguignon and Gunning (2016). The issue here is that all economies experience rent seeking and corruption, particularly if institutions are weak. If aid is a significant source of funding for an economy, it will somehow help to feed rent seeking and sometimes corruption. This is typically the case in poor aid-dependent countries with few other large revenue flows. If they do have alternative large revenues to tap, such as from natural resources, such funds are probably more fungible due to donor conditionality and public expenditure reviews.

One frequently discussed feature of this cluster of arguments is that of taxation and nation-building. If governments can raise large sums of money from external donors, they do not need to tax their population, which can have an impact on their popularity and sometimes their legitimacy. If this is true, it is likely to apply primarily to direct taxes on income and holdings that people really feel and understand as their contribution to society. Also, the taxation argument assumes that in any country, public revenues, or significant amounts of them, will have to be raised eventually by taxing the broad population directly. To raise sufficient revenue, the tax base must go beyond indirect taxes and a narrow base of foreign companies, the local elite, and middle class in the formal sector.

This argument reflects European history rather than evidence from Africa.¹ Purely from a revenue point of view, the gas bonanza expected in the years to come may postpone the financial need for Mozambique to broaden the tax base to the population at large through direct taxes on income and holdings. What is at stake here is the basic contract of nation-building, and that is a much more complicated issue than finding ways to finance the budget.

There is ambiguity, to say the least, in the literature about the role of taxes in nation-building and democracy. Causality is thoroughly debated. Here only two arguments need to be remembered: The first relates to how European regimes extracted taxes in the eighteenth and nineteenth centuries, when their income levels were similar to those of poor African nations today. The second feature of taxation relevant to our study is the content of the nation-building contract in Europe and the USA at that time, which did not include the provision of social services to the broad population. The population paid taxes (and contributed their adult sons as conscripts to the army) in exchange for infrastructure and protection (security). The tax financing of education and health started much later, when countries were much richer (late nineteenth century even in the Nordic countries), and some rich countries are still fiercely debating whether it is a good idea (USA). In any case, it is almost impossible to tax finance the general provision of education and health² in countries with income levels below US$2,000 per capita. Intriguingly, many donors promote this model when they give aid to very poor countries.

This brings us to the third cluster of arguments, which centres on the concept of conditionality. Bourguignon and Gunning (2016) introduce the distinction between aid for finance and aid for reform, and show that, theoretically, conditional aid is more efficient than unconditional aid when it comes to ‘stopping the leakage’ of funds aimed at the poor. They also argue, however, that true and enforced conditionality is very hard to come by.

When analysing donor relations and sovereignty, this channel of influence on institutions is highly relevant. If there is complete harmony and agreement about objectives and priorities between the donor and the recipient, conditionality can serve as a reinforcement to the receiving government vis-à-vis other interests (rent-seeking lobby or political opposition). More frequently though, conditionality is a means to align the interest and priorities of the receiving government to those of the donor, and this tampers with the nature of sovereignty.

There is no claim that there is inherent or embedded dysfunctionality in institutional development due to conditionality from donors. The claim is that reforms are much more likely to succeed if the recipient government truly believes in the reforms promoted by donors, internalises them, and takes responsibility for implementing them. Conditionality can be dysfunctional either because it cannot be implemented or because it is built around poor (unsuitable) policy advice and impatience. Failure to implement reforms without reprisals occurs when donors are unable to act on the non-fulfilment of the contract by the receiving government. This can be for various political reasons³ and because of the so-called Samaritan problem. Even when conditionality is observed, it can be dysfunctional, simply because the reforms it supports are badly designed. This can have severe consequences for a poor country.

Donors are keen to promote development models that reflect international best practice and that often mirror their own efforts in sectors where they have been successful. Arguably, Denmark’s health system, Finland’s education sector, and the Netherlands’ water management and ports administration appear to be very successful undertakings that a developing country could benefit from copying. The export of such systems may, however, be harmful and lead to overly optimistic policy choices in poor countries. When we want to understand the impact of donor relations on institutions, it is not a matter of considering how well or bad is the reform or institution suggested for copying. What matters is how recipient countries deal with donor-driven reforms that they are unable or unwilling to implement. One approach is to make believe, or to pretend, all is going well – a feature that is characteristic of Mozambique!

In many other cases, the proposed systems may be necessary and appropriate for the recipient country. Yet, they seldom receive enough time to mature or to be fully assimilated and implemented due to budgetary considerations, or changes of government, relevant officials, and policies in the donor or recipient country. Whatever the reasons may be, the empirical evidence shows that donor-prompted reforms of institutions often happen too frequently, without sufficient time for them to consolidate and stabilise.

The literature uses a metaphor from biology called ‘isomorphic mimicry’ (Andrews 2009), which describes a situation where an animal looks like another animal that has capabilities it does not have (a non-venomous snake looking like a very poisonous relative). Such an animal can benefit from this similarity in some cases (to scare off enemies and predators), but it cannot develop the capability of the animal it looks like (it cannot hunt like a venomous snake if it is not one).

Isomorphic mimicry is attractive when an institution replicates the processes, systems, and outputs of other (often foreign) successful institutions without developing the capabilities of the institutions it seeks to replicate. When donors push for best-practice reforms, it can be difficult for institutions to develop the new skills needed in the local context. Arguably, donor requirements for agenda conformity close the path to novelty (Andrews et al. 2017, Chapter 2). The phenomenon relates to a rather blurred concept introduced by donors, referred to as ‘ownership’. Ownership often equates the ability to control the formulation and outcome of policy and to act as a proxy for sovereignty. Recipient government ownership of reforms has become increasingly important to donors, at least formally. Consequently, if poor governments want donors to finance institutional development, they must ask for funding of programmes and reforms that donors believe in. This can still have positive outcomes, as mimicry can be an early part of learning.

Nevertheless, isomorphic mimicry becomes problematic when poor governments start asking donors what they would like them to pretend to do (to keep funds flowing). Then, it approaches institutional ventriloquism. This happens when best-practice reforms are articulated, planned, and implemented following external prompting with help from externally funded advisers and consultants. In its extreme form, reforms become fragmented and governments lose agency (Krause 2013). However, loss of agency is not equal to loss of sovereignty, though it may defeat its purpose.

Sovereignty is more than ownership (the capacity to control formulation and implementation) of policy. It is the right to rule and defines the ultimate location of authority. Sovereignty, as the right to rule, constitutes the very basis of the aid relationship and endows African states with the agency with which to contest the terms of aid deals. In Table 10.1, the first column represents the unavoidable loss of national control in the globalised world. The second and third columns represent policy choices that are influenced by the conditionality of aid, the latter being the politically most controversial. The fourth column refers to the abrogation of sovereignty itself – the denial in one way or another of the politico–legal independence of the state.

Table 10.1 Degrees of dependency

External influences
	National control issues			Sovereignty issues
Type	General external social, cultural and economic influence of societal development	External influence on state policy, particularly macroeconomic policy	External influence over constitutional structure	Sovereignty annulment
Examples	Foreign investment and trade, world markets, connectivity, information technology and communications	Aid conditionality, bilateral and multilateral agreements, lobbying, bribery	Political conditionality, state-building, post-conflict reconstruction	Occupation, colonization, military intervention, non-recognition
Impact	Implications for developmental, cultural, or economic purity		Shape and foundation of internal authority	Forfeiture of political and legal independence
Is aid a factor?	Indirectly	Directly, various extents	Can be	No

Source: Data from Brown (2013).

When Mozambique won its independence, the location of ultimate authority over the Mozambican territory shifted from Portugal to Mozambique. The purpose of sovereignty is largely to be able to define and implement policy in one’s own territory. As such, sovereignty becomes the legal base from which the Government of Mozambique (GoM) negotiates the terms and conditions of its international engagements, including its relations with donors.

A Donor Characteristics

The rhetoric around donors and aid is strong. Donors like to be portrayed as do-gooders who administer funds in an attempt to alleviate poverty. They are concerned with the effectiveness of the aid they provide to help poor people. However, this is only a partial understanding of the true nature of donors.

To figure out what drives donors, we find good guidance from the Danish philosopher Søren Kierkegaard and former French President Charles de Gaulle. According to Kierkegaard (1855, cited in Hong and Hong 2009), ‘all true helping begins with a humbling’:

This is the secret in the entire art of helping. Anyone who cannot do this is himself under a delusion if he thinks he is able to help someone else. In order truly to help someone else, I must understand more than he—but certainly first and foremost understand what he understands. If I do not do that, then my greatest understanding does not help him at all. If I nevertheless want to assert my greater understanding, then it is because I am vain or proud, then basically instead of benefiting him I really want to be admired by him. But all true helping begins with a humbling. The helper must first humble himself under the person he wants to help and thereby understand that to help is not to dominate but to serve, that to help is not to be the most dominating but the most patient, that to help is a willingness for the time being to put up with being in the wrong and not understand what the other understands.

(Hong and Hong 2009: 45)

The first thing to remember when studying donors and their behaviour is that they are foreign governments, or international institutions, which foreign governments influence and control. Some bilateral donor agencies are relatively independent institutions, separate from their governmenťs foreign service, either as separate ministries or agencies (e.g., until recently the UK’s Department for International Development and the United States Agency for International Development (USAID)). Others are integral parts of their governmenťs foreign ministries (e.g., the Netherlands and Denmark). However, whatever the institutional set-up is, or appears to be, development aid is an integral part of any donor country’s foreign policy, though rarely the most important one. Security and commercial concerns normally far outweigh the aid considerations, and the question relates rather to the extent to which aid is instrumentalised in donor countries’ pursuance of the other two.

The differences between countries’ institutional set-ups of foreign policy formulation and management lie only in the level at which, and in how much detail, the co-ordination between aid and other foreign policy areas is pursued. To maintain the cash flow, donors must satisfy the interests, values, and incentives of their governments while at the same time providing them with the expected results. Therefore, one cannot understand donor behaviour and the institutional development around aid (instruments, architecture, conditionality, and volume) without analysing the evolution of donor countries’ foreign policy concerns. For any country, foreign policy is a derivative of domestic policy. To understand donors, a good point of departure is to remember the words of President Charles de Gaulle: ‘Les etats n’ont pas d’amis, ils n’ont que des intérets’ (translation: ‘States do not have friends; they only have interests’) (Haskew 2011: 187). Such interests may change from time to time, leading the same donor countries to adopt different policies over time. Their behaviour, but not always their narrative, will reflect this.

In the following, we discuss trends and defining moments in Mozambique’s relations with its donors. What has driven the donor agenda, and how has Mozambican society absorbed this? How has it mitigated and adapted to donor demands, and how has it influenced key institutions for economic development?

III Donor Relations and Trends in Underlying Drivers

The nature of relations between donor and recipient countries has evolved over time, but there is still fundamentally a lot of truth in the old saying: ‘He who pays the piper calls the tune’. However, it sometimes happens in perfect disguise:

The right to self-determination for the Mozambican people was the objective of the struggle for independence (from Portugal). The Lusaka Agreement is the internationally recognized legal instrument that grants this right to the Mozambican people.

(Chissano 2019)⁴

Mozambique’s independence from Portugal, and indeed the country’s sovereignty, is the result of a persistent and successful diplomatic campaign supported by armed struggle. The creation of a sovereign Mozambique came about because FRELIMO’s young diplomacy won the hearts and minds of countries of the world, which, with few exceptions, extended their diplomatic and sometimes economic and military assistance to the Mozambicans’ struggle for the right to self-determination. According to Panguene (2019),⁵ ‘Mozambique is the result of international solidarity’. Being able to rally international solidarity and support around its struggle for independence was the first step and, in substance, what granted Mozambique sovereignty. The country’s founders were united in FRELIMO, an institution, which by nature was dependent on international support and donations.

A The First Decade after Independence – A True Alliance

After independence, Mozambique sought help from the socialist bloc, because its leaders believed in and wanted Mozambique to aspire to their development model. Mozambique served as a frontline player in the cold war, in exchange for which the Soviet bloc provided economic and military assistance. This was a true alliance founded on shared visions and foreign policy concerns. The donors exercised little coercion in political, economic, and social policy, and when some donors occasionally tried to impose military conditionality, Mozambique was able to resist it⁶ (see Veloso 2007).

Mozambique’s donors were few and, although they sometimes included co-founding programmes (especially the UN and the Nordic countries), they can be grouped into four clear groups: (i) the Soviet/Eastern bloc including Cuba, (ii) the UN agencies, (iii) non-governmental organisations (NGOs) from Europe and the USA (mostly left wing and faith based), and, increasingly, (iv) the Nordic countries, the Netherlands, and Switzerland (the like-minded group).

Socialist development models were dominant at the time. Although some of the policy choices in economic development and agricultural strategies followed in this period failed to produce positive results and prosperity, they were not seriously contested. The Soviet Union, GDR, Bulgaria, and the UN agreed with the GoM on the way forward for agriculture. Price controls (inherited from colonial rule), accelerated mechanisation, and state farms, including conscripted labour in some cases, were the recipe everyone agreed on in the policy circles in Maputo.

Extensive consultation and other forms of interaction with the peasants were common practice in the framework of a ‘people’s democracy’. Likewise, central planning was the norm at that time, but the State Central Plan was a harmonised set of sector policies and provincial plans created through exercises that started at the district level. These exercises were undertaken and predominantly led by Mozambicans. The plans, particularly at district level, included the contributions of communities and workers.

Up to 20,000 foreign technical advisers and volunteers, known as cooperantes, helped to implement the policies and strategies. Unlike donor-funded consultants and advisers decades later, they did not usually coordinate and prioritise policy. Policy was formulated and controlled by FRELIMO and the government, whereas implementation was heavily assisted by donors, as testified by Mutemba and Salomao.⁷

The policies and institutions of price controls (even on traded stables), mechanisation, and state ownership of the production apparatus developed under conditions of extremely high state legitimacy. There were no issues around voice, participation, and political accountability. President Samora Machel enjoyed extraordinary popular support (Newitt 1995). The high degree of political instability and violence, successfully attributed to foreign countries funding insurgencies, did not affect state legitimacy. The local media labelled the Mozambican National Resistance (MNR) insurgency as ‘armed bandits’, and donor countries knew that the MNR had been created and funded by elements from ex-colonial rule, Rhodesia, and South Africa, with a view to destabilising and eventually provoking regime change. Mozambique’s very sovereignty and independence as a state were at risk, and the image of the enemy was clear. The war effort was justified, and low state capacity and lack of economic progress were explained by external destabilisation and colonial heritage.

Even in the early 1980s, when the Nordic countries established embassies in Mozambique, there was little debate about legitimacy, accountability, and participation. Mozambique was a front-line state and important in the struggle against apartheid, which is probably the strongest and most resilient political solidarity project between Western Europe and Africa ever known. The anti-apartheid solidarity project at large commanded broad-based political backing in most European parliaments and populations as well as in the UN, and criticism of Mozambique was rare. The few concerns that did arise among donors typically related to human rights. The concept of judicial independence was not recognised by the socialist regimes, and when dissidents to the FRELIMO government were sentenced to death or long prison terms by executive order (or courts under direct instruction from the executive), Nordic donors were uncomfortable and faced questions. However, the brutality and cruelty of the MNR insurgency and the apartheid regime’s brutality at home overshadowed this and were debated much more in European capitals. Gersony’s account expressed the common understanding of donor countries about the insurgency in Mozambique:

Most significantly, RENAMO’s⁸ inward expansion into Mozambique required rapid recruitment, achieved overwhelmingly by the forced conscription of unwilling civilians.[iv] Direct killing of civilians, along with a myriad of human rights violations, manifested in murders, routine brutality, and large-scale massacres. In addition to using indiscriminate violence during military operations, RENAMO leveraged terror to enforce control over new recruits and the local population. New recruits were coerced to murder their family members, while other common acts ranged from facial and bodily mutilation to the use of land mines and burning people alive.

(Morgan 1990: 49)

To understand where the Mozambican elite and FRELIMO come from as regards their relations with donors, it is important to remember the nature of the relationship. There was no explicit conditionality. The former Eastern bloc donors introduced no policy-based development contracts, no performance indicators, or no structural or quantitative benchmarks. It was an alliance built on shared ideology and shared foreign policy concerns that defined a common enemy as well as mutual security concerns. The relationship was much more than one of aid. Economic and technical assistance agreements were technical, not policy conditioned, and embedded in agreement on policy at a much higher level.

To understand the Mozambican elite’s practices today, it is equally important to remember how this relationship changed and when. The spirit of the ‘Decade of the Victory over Underdevelopmenť was in trouble before Samora Machel died in 1986.

Although the peace agreement with RENAMO was still ten years away, and the structural adjustment programme only started in earnest in 1987, the FRELIMO government was already beginning to lose its grit in the early 1980s. The combined effects of war, the second world economic crisis (a sharp increase in interest rates and oil prices), and a persistent drought lasting more than three years caused sharp declines in the economy, and Mozambique’s sovereignty was increasingly being threatened.

Mozambique’s joining the International Monetary Fund (IMF) and signing the Nkomati Accord with South Africa show that the FRELIMO government was already on its knees in 1984, despite a very different public rhetoric from Samora Machel. The Nkomati Accord, in particular, was regarded as a defeat in many circles and by some regional leaders as an outright treason. Mozambique had to make enormous diplomatic efforts to try to explain to its peers why a no-aggression pact with the apartheid regime was necessary.

The disillusionment and loss of grit already in the early 1980s influenced the strength of many government institutions. Civil servants experienced great difficulty in implementing the plans and projects they were supposed to manage due to an increasingly obvious discrepancy between means and ends (human and capital resources and policy objectives). The difficulties were explained by the armed insurgency (por causa da guerra), but it became increasingly clear that the war was not the only problem.

We know from mainstream management literature that staff motivation is a main driver of institutional efficiency. Staff need motivation to believe in what they do, feel that they have the confidence of their managers, and relate to clear performance criteria, that is understand their institution’s mission and their own role in it. Remuneration is important insofar as it has to reach a certain level but, above that, salary can largely be taken off the table in terms of motivation.

When the World Bank and the IMF came to Mozambique in 1986 to formulate and help implement the structural adjustment programme (SAP), civil servants’ self-esteem and confidence in the GoM were already low. Mozambique had become a demoralised country in many ways.

The traditions and approach of the Bretton Woods institutions did not make things better. The imposition, especially by the World Bank, of new working methods and the support of hordes of foreign consultants often led to weakening national institutions, turning them into rubber-stamping entities. The SAP itself caused public sector salaries to shrink significantly, often below the subsistence budget level for a family.

B Structural Adjustment and Privatisations before Peace and Elections

The timing of the SAP from 1987 to 1995, the Rome Peace Agreement in 1992, and the multiparty elections in 1994 created an institutional framework where formal political power, but not economic power, was shared. Donor conditionality was introduced, and it changed the rules of the game, but the main players stayed the same.

By the mid-1980s, Mozambique was economically broke and faced increasing trouble with its creditors. The government, especially Joaquim Chissano, saw that Mozambique’s sovereignty was threatened by the combination of domestic economic factors and geopolitical developments in the world, not least reduced support from the socialist countries, in particular the GDR whose intelligence had concluded that the war could not be won by military means (Wolf and McElvoy 1997).

Referring to Table 10.1, the constitutional reviews in 1986 and 1989 bear witness to externally prompted policy changes of the most fundamental calibre (Table 10.1, column 3). The first review was necessary to accommodate structural adjustment and privatisations, and the second in 1989 paved the way for multiparty democracy.

The sequence of policy changes shows that the FRELIMO government took the difficult decision to lead the institutional changes, although some powerful leaders in the party were not convinced that the capitalist model was better. However, Chissano’s giving up control of policy formulation was necessary to defend sovereignty and FRELIMO’s leadership in the long term. ‘In the words of Luis Bernardo Honwana: “Chissano assured the preservation of the state and the continuation of the state without interruption”.’⁹

The IMF and the World Bank introduced a much more direct contractual relationship with Mozambique as donors. The relationship with the IMF and the World Bank was the start of direct and explicit conditionality in development assistance. There was little ex ante belief that the former socialist GoM would wholeheartedly implement the SAP with fiscal austerity and privatisations as the main ingredients. It was evident from the start that the policy and institutional reforms embedded in the SAP would have negative effects on influential parts of Mozambican society and political costs for the government and ruling party. In exchange, Mozambique received loans, technical assistance, and, not least, rescheduling with the Paris Club of creditors.

Along with the introduction of capitalism, the most important institutional change in independent Mozambique was the adoption of multiparty democracy. This was not directly included in the agreements with the World Bank and IMF, but it was supported by a number of countries that were also Mozambique’s main bilateral donors and creditors in the Paris Club, as well as by influential board members in the IMF and World Bank. Some of these countries were even in ‘constructive engagemenť with the apartheid regime underpinning the RENAMO insurgency. (The Reagan administration introduced only very limited sanctions in 1985).

Although somewhat controversial, it can be debated how conducive multiparty democracy is to economic growth (e.g., in the case of the ‘Asian tigers’ in the 1980s), especially in Mozambique’s income level in the early 1990s. Particularly in a poor country like Mozambique, democracy itself constrains the ability of institutions to formulate and implement reforms. In a governmenťs lifetime of five years, there will be two years with elections (municipal and parliamentary/presidential), which will divert efforts and drain resources from managing the country. This is caused to some extent by the incumbent governmenťs abuse of public resources for campaign purposes, but it is by no means the only reason. Even in mature democracies where governments do not practise outright abuse of public resources, the waste of bureaucratic resources for short-lived campaign-motivated policy formulation and dissemination will drain resources (e.g., the preparation of two sets of policies and fiscal acts and Medium Term Expenditure Frameworks (MTEFs) in election years).

Short-lived governments, even from the same political party, will experience frequent changes of policy and of personnel in key positions. All politically headed institutions rely on personal networks, particularly when institutions are weak. Loyalty becomes more important than technical/professional ability, and high attrition, a sense of a lack of fairness, and insecurity of tenure undermine morale. The contribution of multiparty democracy has a negative effect on institutional efficiency.

In addition to the above, there is little doubt that the sequencing of the introduction of capitalism and multiparty elections in Mozambique has affected the country’s current economic performance. It allowed the GoM (FRELIMO and the Mozambican elite) to create new rules of the game and to continue to occupy the economic sphere before political power sharing took place. Economic adjustment was carried out and state assets were sold to leading Mozambicans (allies or members of the government) while RENAMO was still a guerrilla force, rather than a political party, fighting or waiting in the bush to be demobilised – far away from the decisions about economic reforms being taken in Maputo. When successful elections finally took place, and the new parliament began work in earnest in 1995, the SAP was all but over and most state assets had been sold, often to people with sector knowledge from the defunct state companies. This is not remarkable as there was no private sector to recruit from or sell to, but it supported an understanding or underlying notion of entitlement in the elite, who claimed and possibly believed that they could make the companies viable in a liberal economy.

Until the SAP, the political leaders and the emerging middle class in general had been complying with the code of values and practices of probity and respect for public property. The privatisations that occurred started to change this with the emergence of an incipient business class, more affluent and dominated by former political leaders and civil servants who had benefited from the privatisations. Those who did not benefit felt excluded. Furthermore, the first and subsequent multiparty elections created, for the first time, the awareness that FRELIMO might eventually lose the elections with subsequent widespread purges to follow, as RENAMO would need to accommodate its members and sympathisers in the state apparatus. All of this brought growing levels of insecurity and fear and a sense of urgency to ‘get something before it is too late’. It entailed corruption, compounded by a pervasive feeling and belief that FRELIMO members and sympathisers were more entitled to grab public assets than RENAMO members, who were portrayed as having colluded with external forces to destroy the country: ‘Mozambique privatized more than 4000 companies, about 80% benefitting Mozambicans and on highly advantageous conditions’ (Diogo 2013: 158).

There is speculation that the sequencing of the SAP was a deliberate tactic by FRELIMO to retain control of economic assets and that the Bretton Woods institutions accepted this – knowing that the economy would suffer in the medium term – to put the GoM in the driver’s seat of reforms. While the deep motivation of either party is difficult to ascertain, there was at least one additional element in FRELIMO’s motivation: The prospect for a lasting peace would be considerably better if demobilisation and reintegration happened before the elections. (See Appendix with separate note on peace process).

Contractual policy-based lending was not the only new feature in Mozambique’s relations with donors. The IMF required negotiations to take place directly with Mozambique’s Central Bank and Finance Ministry, whereas assistance from other donors was negotiated through the Ministry of Foreign Affairs and Cooperation. This separated the groups and privileged policy dialogue between the IMF and the Ministry of Finance as something all other donors depended on.

It had been intended that the SAP would address the structural imbalances in the de facto bankrupt Mozambican economy, introduce the fundamental institutional changes that capitalism entails, and succeed in implementing them. However, the whole operation was poorly timed and sequenced, and this had significant consequences for the way the Mozambican elite chose to mitigate and adapt to those changes. Even on its own isolated terms, sequencing of the SAP was unfortunate. While ‘re-installing’ market-based prices (which had never been relied on in colonial times) was meant to reinvigorate Mozambique’s comparative advantage, the economy was unable to produce a supply response to changes in relative prices.

There were a host of reasons for this, the most fundamental being that the programme went ahead in the middle of a devastating war, while Mozambique’s main comparative advantage from factor endowments lay with agriculture and natural resources. The war kept the farmers away from their land, as most were waiting in refugee camps or in the suburbs of garrison towns. The war also made it impossible to transport and market coal and other natural resources, to get inputs and spares, and to market the little output farmers managed to produce in less war-affected areas. Even the distribution and sale of cheap power from Mozambique’s dams crippled, as RENAMO sabotaged power lines. Few tourists dared visit the beautiful beach resorts and nature parks in the war-ravaged country.

Almost all the sectors where Mozambique could conceivably have a comparative advantage in a market-priced economy – except fisheries, which was already in Japanese and Spanish hands (through in-flagging and Mozambican silent partners) – were unable to respond to the changes in relative prices.

Therefore, while the expected positive effects of the SAP were late to arrive, the negative effects for previously established industries that had been protected by administered prices and exchange rates, on the other hand, took almost immediate effect. State enterprises across the manufacturing sector, which were overwhelmingly involved in import-substituting production, became unprofitable. The companies were subsidised by the government to avoid massive lay-offs and limit payroll cuts. However, once such subsidies became visible in the budget, they formed part of the fiscal equation in the IMF programme and were therefore short-lived. As the GoM still owned and controlled large sections of the banking sector, many subsidies were eventually moved to the banks, which were obliged to extend loans (e.g., under the heading of restructuring for viable commercial use) to unviable companies. This mining of the banking sector still affects the financial sector today.

At first, these companies could not make money, so they had to be subsidised in some way by the GoM until peace and foreign investors returned to favour their businesses. Eventually most companies’ assets were used for other purposes and/or partly sold to foreign investors:

The case of cashew showed that the World Bank/IMF imposed reforms with a view to securing the best possible farm gate prices to farmers (privatise and remove export bans and tax). When the government resisted in an attempt to save the cashew processing industry, the World Bank blocked a US$50 million loan tranche. When the GoM finally agreed, it became clear that the people who had bought the industries from the GoM – claiming they would try to safeguard ten thousand jobs – had colluded with the bank and Indian traders and used the installations for storage and trade only, exporting only raw nuts. The industry died and no new trees were planted (see, e.g., Diogo 2013).

When contract aid in the form of policy-based lending was introduced into macroeconomic management and reforms in key industries, the GoM managed the legitimacy aspects of the situation in two ways. To the Mozambican population, the negative effects of the economic policy of the period were successfully attributed to the conditionality of the IMF. The GoM played the ‘victim card’. It still controlled the press at home and took great care to protect other donors from bad press. In addition, in much of the international media, SAPs were reported to leave a trail of social havoc in Africa. This created the momentum for a group of countries and the UN to say that structural adjustment could be done with a human face. The same group of countries were the main bilateral donors to Mozambique. They found the IMF’s policy recommendations and conditionality to be misguided and were optimistic about prospects for Mozambique.

In the mid-1990s, state legitimacy was still relatively intact in Mozambique. FRELIMO’s election victory in 1994 was credible and accepted by RENAMO, although only after pressure from peers in the region (the ANC). Donors helped to convert Maputo’s finest cinema to serve as the parliament, and opposition politicians started to appear in local media on a daily basis and in the streets and cafés of Maputo. Things were also opening up in terms of accountability. The budget, which had previously been a state secret, appeared on the front page of a new and popular daily publication, Media Fax. Most people witnessed a remarkable improvement in living standards compared to the times of war. Double-digit growth rates, including in family agriculture, made the GoM popular at home and abroad.

At the Consultative Group meeting in Paris in 1994, donors pledged almost US$1.4 billion in new investment projects (including technical assistance), against a gross domestic product (GDP) of around US$2.5 billion and a government budget of around US$1 billion. Clearly, the GoM was not able to absorb this money in any sustainable way, as it was unable to finance the ensuing recurrent costs.

These conditions led to a handful of donors (Swiss Development Corporation, USAID, EU, and Denmark) working successfully with the Central Bank and Ministry of Finance and Planning to make the Bretton Woods institutions accept a more expansionary fiscal policy, enabling faster growth of social sectors after the war. This limited group of donors financed the servicing of some of Mozambique’s debt prior to the highly indebted poor countries’ debt cancellation initiative and import support schemes, which generated counterpart revenue in Metical for the government, which the IMF agreed not to sterilise.

This all took place in a ‘Marshall Plan’ atmosphere, and it was believed that subsequent economic growth and the cancellation of HIPC¹⁰ debt would enable medium-term sustainability. Mozambique was internationally recognised as a post-war rehabilitation success (OECD 2002). The real test for this to become a more sustained operation was whether donors in general would be prepared to finance recurrent costs, including salaries.

C Scaling Up for a New Darling

This preparedness came from a strong international agenda at the time, which drove donors to provide higher volumes of aid and new instruments. It helped the GoM to expand services to the population but also maintained the country’s dependency on donors even as GDP and domestic revenue grew. It created an aid architecture in the name of host country ownership that allowed and increasingly encouraged isomorphic mimicry, bordering on ventriloquism.

The World Summit for Social Development in Copenhagen in 1995, the largest gathering of world leaders at the time, reached a new consensus to put people at the centre of development. Its declaration and agreed programme of action made donors in all countries increase their volumes of aid especially towards social sectors. The summit cemented the discourse by UNICEF and the Nordic countries of structural adjustment with a human face and obliged governments to formulate poverty eradication policies and plans (PRSPs), which became a condition for debt relief under HIPC.

Universal access to education and health was a tall order at the time, and pursuance of the ‘20/20 objective’ faced particular obstacles. This objective required governments to allocate 20 per cent of their budget and donors to allocate 20 per cent of their support to social sectors. However, the Mozambican government established a budgetary regime of 20 per cent for education and 14 per cent for health (Diogo 2013: 157).

If most donors had one strong notion at the time, it was a narrow concept of sustainability, requiring Mozambique to meet future recurrent costs on the investments made by donors. Bilateral agreements between donors and the GoM usually included clauses about counterpart funds and promises by the GoM to staff and maintain the infrastructure created by donor-funded projects. At that time, few donors would get board approval for projects if this were not addressed.

Especially in the social sectors, the cost structure was heavily tilted towards recurrent costs. Unlike roads or power lines, where recurrent costs are closer to physical maintenance, the social sectors needed staffing and consumables on a completely different scale. The costs of the salaries of teachers, nurses, and doctors, and of drugs, which the GoM and most donors agreed were necessary for rapid social expansion, were impossible to finance given traditional conventions in development aid. Thus, the need to find a way to help the GoM finance the recurrent costs of rapid social expansion was a main driver of the establishment by bilateral donors of a new institution known as general budget support (GBS).

This drive by donors to increase absorption capacity by financing recurrent costs, including salaries, made Mozambique exceptionally dependent on donors. The GBS instrument also relocated the centre of dialogue between Mozambique and the donors from the Ministry of Foreign Affairs and Cooperation to the Ministry of Finance and Planning. In this way, one of the most important elements of Mozambique’s foreign policy – its relations with donors – separated from other foreign policy elements in a different institutional frame. This compartmentalisation of the formulation and management of foreign policy was appropriate and worked at the time.

The GoM never formulated a full and coherent aid strategy, although donors encouraged it to do so. It was not required to and probably did not want to. Indeed, in 1994, when the Ministry of Cooperation and the Ministry of Foreign Affairs merged into the current Ministry of Foreign Affairs and Cooperation (MINEC), there were discussions about the need to formulate an international cooperation strategy. However, this was abandoned to make room for the flexibility that is always needed when dealing with different donors. Experience had shown that donors had different policies and procedures, and, quite often, it was difficult to predict the amount of funding the donors would pledge and actually disburse. Against this backdrop of uncertainty and sizeable challenges, it was felt that the broad objectives stated in the governmenťs five-year plan were good enough to guide all stakeholders.

Given the overall context of the development challenges, the GoM had to be pragmatic and agree to meet the donors’ essential objectives. At the same time, a government that is under pressure from donors to accept reforms, and from local interest groups to resist reforms, can manage the situation by avoiding taking (explicit) strategic decisions (Castel-Branco 2008).

The GoM embedded its priorities in the international aid effectiveness agenda, which was gaining strong momentum at the beginning of the century. The donor countries themselves were driving this agenda primarily through the UN, OECD, and the EU, which showed committed and competent leadership on the issue (e.g., by Poul Nielson, Richard Manning).

The Millennium Development Goals, New Economic Partnership for African Development, and the Paris Declaration, through which donor countries agreed to harmonise and increase aid/GDP ratios to get closer to the UN target of 0.7 per cent, were the main signposts for donors. Volume, ownership, and harmonisation through alignment and division of labour became the key elements in almost all donor countries’ aid policies and their country assistance strategies in Mozambique.

The GBS and sector-wide approach (SWAP) became significant instruments for aid delivery, and the dialogue structure was built around them. Consequently, it became increasingly difficult and awkward for Western donors to be outside the group. The group all but monopolised policy dialogue with the GoM, and produced and shared important information only internally. A number of donors joined the GBS instrument for the wrong reasons, some with very small budgets both in absolute terms and relative to their total country budget. Aid effectiveness at the aggregate level may not have been their main concern, but a seat at the table conducting policy dialogue was. Free riding became a problem (Bourguignon and Platteau 2015). More importantly, though, there was – among most donors, including the pioneers of GBS – an underlying assumption that the instrument and broad co-ordination around it could be used for coercive purposes.

Since 2000, the government had been conducting half-yearly policy dialogue meetings with the EU, as agreed under Article 8 of the Cotonou Agreement. At these meetings, cooperation was evaluated and political issues were discussed. The GBS donors who were not members of the EU wanted to have a similar type of dialogue with the GoM. As the EU’s own code of conduct obliged its Commission and member states to subscribe to the broadest-possible host country-led dialogue mechanisms, the solution adopted was to establish a forum for all GBS donors.

Donors organised themselves into sector working groups with common funds, each with a lead donor. The working groups reported to the deputy heads of missions (heads of agencies where these were outside the diplomatic missions), who reported to ambassadors and high commissioners. By 2009, donors had a dozen common funds and working groups reporting to the top two levels, where the 19 donors selected a rotating chair in a troika format that was entrusted to undertake policy dialogue with the governmenťs key ministers. The donor troika was given a secretariat, paid for by donors with overseas development assistance (ODA). The GoM’s troika was headed by the Minister for Finance and Planning for fifteen years and by the Minister for Planning and Development during President Guebuza’s ten-year tenure, when the Ministry for Finance and Planning was split in two. The other members were the Governor of the Central Bank and the Minister of Foreign Affairs and Cooperation.

The whole machinery was based on a memorandum of understanding (MoU) between donors and the GoM, which set out the underlying principles, that is of democracy, good governance, and sound macroeconomic management (column 3 in Brown’s model – see Table 10.1) for budget support. The MoU also described the main tasks for implementing sector reforms, plans and targets (column 2 in Table 10.1), joint reviews and assessment of progress (PAP), and financial pledges each year. The matrix contained more than forty indicators, often tied to different tranches of donor disbursements, but they were mostly output and process indicators rather than real outcome assessments. The indicator framework measured how many children were enrolled in primary school and how many legal cases were awaiting settlement in the courts. It said little about learning outcomes and access to justice. It allowed for isomorphic mimicry, sometimes even encouraged it, because meeting output indicators became the objective. The system was successful in increasing aid, which became a reference point at many international gatherings on aid efficiency and harmonisation.

Although many reforms remained shallow because of the high ambitions of donors and frequent changes of direction, and because GoM would not allow reforms to deepen (i.e. in agriculture and the justice sector), much of the institutional capacity building undertaken with donor assistance in the 2000s was solid.¹¹ This was the case with the strengthening of public financial management, including a strong tax authority and public budget and accounting system. According to the IMF (Dabàn and Pesoa 2007), ‘Mozambique is a promising example of successful budget reform in Africa’.

Even where the reforms were shallow, there was little long-term systemic damage to the institutions due to isomorphic mimicry. Donor money and Mozambican human resources were wasted at times, but because the reforms remained shallow, they did little real damage. This occurs primarily when ill-conceived reforms and policies are implemented.

The GoM had few concerns about harmonisation, which was considered a win–win situation. The increased aid flow allowed the government to expand the social sectors and infrastructure, etc., which underpinned its legitimacy. The alignment with Mozambican procedures made the donors more anonymous and the GoM more visible. It also made the government a larger economic agent in the market (as a buyer of goods and services with donor money) helping to feed the bourgeoisie.

Aid dependency did not increase in nominal terms during the 2000s, as GDP and the governmenťs domestic revenue picked up. However, more donor funds enabled the expansion of government spending in the social sectors and infrastructure, etc., above what would have been possible without increasing and more aligned assistance. More importantly for ownership, donors expanded policy-based granting and lending (which had previously been restricted to macroeconomic and strategic sectors imposed by the World Bank/IMF) to almost all sectors and main donors. Common funds and donor-financed technical advisers and consultants, embedded in Mozambican institutions to assist primarily with policy formulation, became an implicit part of conditionality. Policy formulation and dialogue, along with target setting and measurement of results, took place at sector level. As the whole aid architecture was organised around the group of donors providing GBS (and the USA), donors were able to influence policy formulation at the sector level at very early stages and to follow through with implicit conditionality in policy dialogue with the government all the way to the macro level, that is in discussion related to fiscal policy such as MTEFs, yearly budget allocations etc.

This had two profound effects: (1) it weakened the GoM’s policy co-ordination efforts by strengthening the sectors vis-à-vis the Ministry of Planning and Development, enabling isomorphic mimicry and approaching ventriloquism and (2) it side-lined the Ministry of Foreign Affairs and Cooperation (to ceremonial duties) and compartmentalised foreign policy.

The governmenťs policy-coordinating ministries, which are vital for ensuring true government agency (the ability to formulate and manage a central policy and plan), were not strong and were further weakened in terms of resources vis-à-vis sector ministries. At the same time, they were subject to unclear mandates that the donors could influence.

The international winds were favourable to Mozambique through most of the 2000s. The Paris Declaration came shortly after FRELIMO’s and Guebuza’s disputed election victory in 2004, and the aid effectiveness agenda remained alive, albeit weaker, with the Accra Agenda for Action in 2008. Although donors expressed increasing concerns about issues of democracy, human rights, and corruption, the aid kept flowing and even increased. The GoM got used to the donors ‘barking but never really biting’.

Economic growth and aid flows continued strongly correlated throughout the 2000s. However, the Mozambican economy did not achieve structural transformation and became a consumption-driven economy where agriculture fed the rural population while the formal sector – urban elite and middle class – depended on the export of a few commodities, a handful of large foreign direct investments, and development aid. As much of the development aid was delivered as programme aid, it financed public salaries and procurement. This meant it was strongly focused on public servants and – via public procurement – the private sector, which supplied goods and services to the public sector, often through shallow trading companies owned by the Mozambican elite and dependent on foreign supplies.

The notion of entitlement in elite quarters thrived in this environment, and Mozambique’s ranking in many international indexes of governance and corruption dropped. However, legal convictions in a few high-profile corruption cases (i.e. the Director of Airports) calmed things down. Donors agreed that Mozambique was heading in the right direction despite occasional setbacks.

D The Thrill Is Gone

All this changed slowly but surely as the decade of the 2000s approached its end and donor countries faced a series of new challenges in the world. Mozambique lost its status as a donor darling as aid money was needed elsewhere for less altruistic purposes, while the GoM did not respond to Western governments’ concerns about security and commerce. Donor assessments of Mozambique were carried out in a new light, and the attraction of isomorphic mimicry became more difficult. Contested elections, which were the norm, were no longer accepted. Donors evaluated reforms related to legitimacy and accountability as being too shallow, and the GBS ran into increasing trouble as an instrument of aid. At the same time, the aid architecture built around this institution remained intact.

The Western governments’ war on terror became wider and costlier. Afghanistan and Pakistan, in particular, consumed enormous resources, but events in West Africa and the Horn of Africa also caused Western countries to reconsider their priorities. A worldwide economic crisis also strengthened critical voices in Western parliaments on aid spending and conditionality, including on migration. Donor budgets reduced and moved increasingly north in Africa.

The developments at the Horn of Africa spilled over and had repercussions in Mozambique. Somali piracy and human traffickers reached the shores of Inhambane, and Mozambique was unable to patrol its territorial waters – 200 nautical miles along 2,000 kilometres of coastline. This was also one of the reasons, why the smuggling of drugs from south Asia through northern Mozambique intensified, and it was almost risk free. At sea, there was no law enforcement and on shore, there was ‘loosely governed territory’ as the CIA put it.

Western shipping was threatened in Mozambican waters and the drug trade that ran almost unhindered through Mozambique helped to finance the Taliban and Al Qaeda killing of Western soldiers in Afghanistan. This was bad enough on its own, but Western governments were also thinking ahead about the safety of future gas operations in the Mozambican seas. Some of the major donor countries at the heart of the G19 structure now also had security and commercial concerns about Mozambique. (As one senior diplomat in Maputo put it at the time: ‘There needs to be a roll-back of sovereignty’.)

President Guebuza and his government were insensitive to this. Guebuza forged ever-closer ties with China and Vietnam (awarded the third Global System for Mobile licence in Mozambique in 2010 along with enormous infrastructure contracts for a stadium, the Maputo circular road, and the bridge to Catembe). Furthermore, parts of the GoM, and Guebuza himself, openly supported a local business person, known to Western governments as a drug dealer.¹²

The Western governments’ security and commercial concerns had little to do with the underlying principles of GBS. Nevertheless, they influenced the donor–GoM dialogue because this was where Western governments had leverage. Donors raised their commercial and security concerns in discussions under headings such as corruption, the Extractive Industries Transparency Initiative, state capacity, and independence from private interests. However, above all, the donors’ new priorities and concerns influenced their assessments, and they became more determined to act together.

The GoM had plenty of opportunity during the 2000s to observe the donors’ rising concerns about democracy, governance, and human rights elsewhere in Africa without cutting aid, but unlike Nigeria, Uganda, Ethiopia, and Tanzania, for example, the GoM did not address Western security concerns. Therefore, Mozambique was measured by another yardstick, and as the donors’ list of critical issues grew, an increasing number of the G19 donors wanted to test the underlying assumption that GBS could be used for coercive purposes.

The opportunity to do so came with the troubled elections in 2009. A new political party, Mozambique’s Democratic Movement, popular with intellectuals, NGOs, and donors, was prevented from running in seven of the country’s eleven provinces. Donors delivered a démarche to President Guebuza on the issue prior to elections but to no avail. FRELIMO declared an enormous victory (with a qualified majority in parliament) and inaugurated President Guebuza’s second term in office. All GBS was frozen shortly after. The decision followed many weeks of discussion between donors and was conveyed to the government in a short letter signed by all nineteen heads of missions, which claimed there had been a breach of the underlying principles in relation to democracy and governance. (See Table 10.1, column 3.)

This was a wake-up call for the GoM. A number of Mozambique’s ambassadors in donor countries were called home to explain how this had surprisingly happened. The GoM concluded that donor agencies were highly decentralised and that ambassadors of donor countries in Maputo had been calling the shots. While this was partly true, many heads of missions had just been delivering what their governments wanted. The institutional factors that contributed to the GoM’s poor assessment of the general donor drift was a more fundamental reason why the GoM was taken by surprise.

Donors discussed policy co-ordination and priorities primarily with the Ministry of Planning and Development (MPD), and the presidency took over an increasing number of the activities that the Ministry of Foreign Affairs and Cooperation (MINEC) had traditionally been tasked with. State protocol moved from MINEC to the presidenťs office, and ad hoc consultations with resident ambassadors by the president without the participation of the Foreign Ministry began to occur along with a very busy calendar of international visits by heads of state. MINEC was sidelined from the aid dialogue and Mozambique made little use of its embassies abroad to understand the policy shift under way in most Western donor countries. Foreign policy co-ordination had been weakened.

The freezing of GBS hardened the relationship between the GoM and donors from then on. However, at the time, the GoM decided to please the donors, at least superficially, and initiated a series of reforms on transparency, participation, and state legitimacy (election law, EITI, etc.).

GBS resumed later the same year, but the policy dialogue became increasingly rigid and unproductive. If the GoM told donors about its fundamental challenges and troubles, these issues would soon become the subject of donor-prompted reforms and be measured in the joint indicator frameworks that guided GBS disbursements. Neither donors nor the GoM tried (in earnest) to change the dialogue architecture and ways of discussion to open the debate, separate policy debate from conditionality, and analyse issues in fora where they naturally belonged by people who knew what they were doing. Everything entered the policy dialogue in the G19. Donor-prompted reforms were formulated with the assistance of donor-funded consultancies and advisers and were presented to the MPD together with an already agreed financial pledge from donors.

In identifying institutional drivers, it is important to remember that development cooperation is by nature a very long-term endeavour, whereas politicians in Western democracies are elected for and often optimise their political capital over the medium term. There are of course exceptions to this but, generally speaking, bilateral donors are impatient by nature because ministers want to attract domestic attention. Thus, personally ambitious ambassadors and heads of aid agencies frequently mirror this and lack the courage to propose and design aid finance that reflects the true cost, in terms of time and money, of reforms and capacity building.

Very often, development ministers get attention by announcing new initiatives and funding to assist poor countries that have specific issues. This is often through vertical programmes (like support to combat HIV) rather than horizontal ones, as are often more called for (e.g., integrated support to primary healthcare). Such support programmes are much more complicated and difficult to explain, and GBS is only a front-page story when a development minister decides to stop it (showing resolve to taxpayers). This means that results achieved are not at the centre of the political debate about aid in Western democracies. Many aid agencies struggle to document their results and disseminate them to the broader public, but there is little uptake of this by journalists and politicians.

Almost all the successful cooperation programmes that have delivered good results are likely to have been initiated by the predecessors of current development ministers who are often from a different political party, as alternation of political parties holding power is common in Western democracies. No development minister dreams of extending the timescale and providing new funds for a support programme that was optimistically designed and approved by a predecessor from a different political party. As a result, there was constant pressure on the GoM to adopt and implement new reforms designed by donor-financed consultants in the image of donor countries themselves. This drive often led to isomorphic mimicry, either by default or by design, to keep the funds flowing.

Arrogance and urgency on the donor side (we need to discipline the GoM before it is too late) and on the GoM side (revenues from natural gas will soon outstrip donor assistance by far, so we do not need donors any more) made the atmosphere worse. Donors began to pay more attention to the household surveys and poverty assessments showing a very slow decline in poverty rates and no decline in the numbers of poor people (headcount), as well as marked regional differences that favoured the ruling party and discriminated against the opposition’s heartlands.

Another contested election in 2014 proved to donors that the reforms initiated after the freezing of GBS in 2010 were shallow. Laws had been passed and procedures adopted. Nevertheless, they did not really change the state of democracy. The reforms were isomorphic mimicry – Mozambique was still not a multiparty democracy as defined and understood by Western governments. The main opposition party, RENAMO, gave up on democracy and went back to the bush to reinitiate insurgency to claim control over at least a part of the country. The GoM’s legitimacy was seriously questioned, although irregularities were no more serious than at earlier elections and no more problematic than in other countries receiving GBS at the time, that is Burkina Faso, Uganda, and Ethiopia.

The drivers that had made the institutions of SWAPs and GBS a success and had pushed aid volumes to unprecedented heights had all but vanished. The international winds that had carried the agenda forward weakened even more during the 2010s as the Arab Spring, the success of ISIS, and the war in Syria created enormous security problems on the EU’s doorstep and produced huge flows of refugees into Europe. This strengthened right-wing nationalist forces in the EU and the USA and put further pressure on aid budgets (cuts and re-direction to emergency aid in North Africa and the Levant).

Western donor countries were already starting to turn their backs on GBS at the beginning of the decade and would have abandoned the instrument no later than the early 2020s anyway (leaving such instruments to a few such as the EU and World Bank). The aid strategies, such as harmonisation, including GBS, that were formulated from around 2012/13 by countries, which had previously been stalwarts of aid effectiveness, clearly show that they were moving on and had new concerns and priorities. As international solidarity weakened, Western governments were clearly looking for more direct and endogenous reasons in recipient countries to justify reduced budgets and the move away from GBS and the Paris agenda at large towards results-based contracts and projects that could underpin their commercial and security interests.

The Guebuza government provided plenty of ammunition for this move – even a crash landing of donor relations – when it guaranteed three large commercial loans for partly fraudulent purposes to three state enterprises without informing parliament and donors. More than US$2 billion of new commercial debt was contracted in an attempt to create a maritime security infrastructure in Mozambican hands before the gas bonanza, when almost all security contracts would traditionally go to international companies with close links to foreign governments. A combination of incompetence and greed overwhelmed whatever good intentions there had been in the project, and Mozambique as a state was left with massive (defined as unsustainable by the IMF) levels of new debt that would inevitably be partly repaid by donor funds.

To Western donor countries, this was proof of further deception and mimicry. In partnership with the GoM and the IMF, they had promoted and financed what had been acknowledged as successful public financial management reform, and it was now clear that it had not solved the problem of accountability in any way. Donors acted swiftly and united around a single condition for resuming GBS: an international forensic audit showing what the money had been used for and by whom, along with appropriate legal follow-up of the findings. This condition was unmanageable for the GoM, as it would create decay in the ruling party. When the audit was finally carried out and revealed massive fraud and embezzlement, little or no action was taken by the Attorney General. After months of waiting and pressure from civil society and donors, lower-level officials were indicted and the legal processes crept forward very slowly.

Unlike the situation in the late 1980s, when Western donor countries had insisted on fundamental reforms (Brown’s column 3 – see Table 1), Mozambique’s sovereignty was not under threat in 2016. RENAMO’s new armed insurgency was containable and the GoM knew that oil and gas revenues could replace donor aid within a decade. As it was not threatened, sovereignty served as FRELIMO’s bulwark.

It was only when the British and South African courts complied with requests from the US courts and arrested former Finance Minister Manuel Chang and a number of foreign bankers involved in the scandal that the Mozambican Attorney General became busy. Although the Mozambican elite would not be threatened by withheld donor funds, high-ranking and centrally placed bankers and former members of the GoM who would cooperate with the US courts in a plea bargain could cause serious trouble.

While the freezing of funds by donors did not lead to change in the Mozambican leadership or its attitude, events in the justice sector in the region may well have influenced the mind-set in the ruling party, not least the indictment and order for preventive prisoning of former President Jacob Zuma. There is little doubt that these events inspired FRELIMO and the government to demonstrate stronger efforts to pursue corruption, although there still seemed to be some selectivity in deciding which cases to pursue.

Moreover, an often-ignored part of this phenomenon is the delays and the destruction of projects that infighting among the elite caused to programmes and projects financed by donors and foreign direct investment (FDI). When elements from different factions of the elite (FRELIMO) were destroying and stealing each other’s projects (rent seeking or not), this helped to create a corrupt or ‘do nothing’ approach in key institutions that prioritised, coordinated, and approved vital investments for economic growth. It seems that the GoM was trying to use the greater caution of economic actors and the damage caused to FRELIMO’s image by the debt scandal to reduce and control such infighting.

The political consequences of the debt scandal are far from over, but three things stand out:

1. 1. FRELIMO is still in government and has probably strengthened its position as compared to 2016. This has happened partly as a result of the manipulation of election results – intimidation, rigging voter lists, and direct ballot stuffing – but it is also the result of a completely disorganised political opposition and a new resolve within FRELIMO itself, to not lose power.

2. 2. Donors are coming back on stream, despite the lack of consequences for top party officials of the debt scandal and despite the election cheating that is incomparable to any previous elections.

3. 3. Donors have undermined their own previous institutional strengthening and capacity-building efforts by starving the government institutions of finance and choosing to finance NGOs, the UN, and standalone, vertical projects instead. The organisations pay good salaries (well above those paid by GoM) and have now hired a significant number of the best higher-level government officials and specialists in various sectors.

The donor’s financial body language did not lead to any change of attitude in the FRELIMO leadership. Rather, it may have helped to strengthen the resolve of the party leadership, as the Mozambican economy proved remarkably resilient to the financial crisis that followed the loans and was made worse by freezing of donor funds. Events have proved that FRELIMO’s sense of entitlement is still very strong, and it continues to be possible to equate Mozambique’s sovereignty with the party’s grip on power. Concessions were certainly given to some donor countries with large commercial investments, but not to the donor group at large.

Now, a few years later and after one much contested election since the non-declared state guarantees were revealed, many donors, including the largest and most intense defenders of the disbursement freeze, seem to have reviewed their standards for governance and democracy. The US Millennium Challenge Corporation has approved Mozambique for massive new financing and the IMF is finalising a new programme with the GoM, which will pave the way for the EU to re-engage. Most bilateral donors will follow, although their budgets have been reduced compared to the last decade (for other reasons) and their aid instruments have changed.

While the SWAPs in education and health may survive for some years in significantly reduced versions and with more strings attached, GBS is de facto dead for bilateral donors in Mozambique. No other known aid instrument can provide the volume of funds that GBS did, and Mozambique will have to raise revenue in other ways, including closing deals with foreign companies in the gas and natural resource sector. Most of these companies are from Mozambique’s donor countries.

There will be new instruments and new players in combinations that cannot be assessed and dealt with in a narrow donor-recipient perspective. Investments from private foreign companies, pension funds, government guarantees, loans, and grants will be bundled and advanced under strategic international agendas like climate change and job creation to suit donor countries’ interests.

When Mozambique begins to develop serious international commercial cooperation, it will with large and powerful foreign companies. They will use their governments to pave the way for general conditions for commercial exchange, sometimes even conditions for specific transactions. These foreign governments are Mozambique’s main donors and there is an increased risk that Mozambique will be totally dependent on foreign governments and companies in the coming decades. As revenue from gas and coal depends on the operations of a handful of foreign companies from almost the same countries as the main donors, the risk to Mozambique of external shocks – political or market based – may be harder to mitigate than ever before.

It is interesting that many of the large contracts for oil and gas exploration were signed at the time of the crisis between the GoM and its donors. The same is the case for some of the security contracts for exploration, and foreign security companies are now openly operating in northern Mozambique.

IV Conclusions and Looking Ahead

For more than three decades Mozambique’s government nurtured donor relations in ways that allowed its partners to pursue the objectives, targets, and modalities set by the international development agenda and even make Mozambique a reference case in international aid. It allowed donors to challenge its ability, but never its authority. Consequently, the basic political fabric stayed the same, while the social and economic sectors were subject to ambitious donor-prompted reforms in exchange for high levels of external finance. Some reforms were shallow because donors were too ambitious or impatient, or because Mozambique’s ruling party would not allow them to deepen. Nevertheless, much capacity has been built, which makes Mozambique comparable with many of its African peers, which had a thirty-year head start on Mozambique in independent development.

This changed over the last decade. Long-term, harmonised development aid fell out of fashion as Western countries became much more concerned with instrumentalising aid in pursuit of security and commercial interests. In this situation, it is detrimental to Mozambique’s sovereignty to approach foreign countries primarily as donors. The loss of agency that Mozambique’s government has experienced – and deliberately allowed in significant areas of the social and economic sectors in exchange for finance – must be rolled back. Development partners have changed their ways because their objectives have changed. The GoM is no longer dealing with a group of relatively well-coordinated donors, which are primarily pursuing effective development assistance and overwhelmingly playing transparently and by internationally agreed objectives and standards. Mozambique must prepare to no longer deal with donors pursuing poverty alleviation, but with foreign countries with a much more intertwined agenda of foreign policy concerns – security, commercial, and aid – that are bundled and labelled in tactical ways to serve each country’s interests.

Mozambique’s government would be well advised to revamp institutions (organisations and ways of working) used to dealing only with donors and to build institutions and structures suited to dealing with foreign countries. A holistic view of relations with foreign nations is now more relevant than a primary and overwhelming focus on the donor aspect of that relationship. A ministry of finance and planning should probably not be at the centre of dealing with foreign nations. It has other important functions (the plan, budget, and treasury) that deserve strong and persistent performance, including building coherence and discipline around national plans.

Mozambique would benefit from having a strengthened foreign service, which ensures that all aspects of foreign policy are (un)covered in relations with foreign countries. It would be advisable to create clear rules of the game for all foreign investments (public and private) through a solid public investment plan, with priorities based on clear objective criteria (internal rate of return). To do so, and indeed to police such plans and strategies, the government could establish strong interfaces between MINEC, MEF, and key sector ministries like the Ministry of Industry and Trade, the Ministry of Energy and Mineral Resources, and the Ministry of Agriculture and Rural Development under the leadership of the Prime Minister.

The proposed institutional arrangements could ensure better co-ordination and mutual understanding among the agencies and, most importantly, more coherent interaction with foreign countries and organisations to ensure that they support Mozambique’s priorities rather than a rapidly changing kaleidoscope of their own ideas. The government should sometimes say ‘no thank you’ to aid money. If the government can align all foreign investments (ODA and FDI) to a coherent development strategy and investment plan, there is a chance that Mozambique can choose and receive the investments that will help finance inclusive growth, not just those that will ensure the greatest profit to foreign companies and sections of the Mozambican elite.