3. Long term perverse effects of aid policies

"One major issue is the appropriate sequencing of policy and institutional reforms. No poor country has the capacity to move forward with equal vigor on all these fronts at once, so it will be important for the country, with external support, to focus on identifying and grappling with the main obstacles to growth," noted the World Bank in a paper earlier this year (Stern 2002:xiii). In this part of the paper, I will try to demonstrate that the World Bank and other donors have never seen corruption and state capture as "main obstacles to growth"; even when senior Mozambican officials highlighted corruption issues, the Bank had other priorities. The perverse effect, I argue, has been to promote corruption and convince key figures in the Mozambican elite that not only is corruption acceptable, but that it is the normal route to capitalist development.

Unless noted, the material in the next four sections is drawn from Hanlon (2001, 2002a).

3.1 The transition to capitalism

In the mid-1980s Mozambique was deeply immersed in a war which was to cost more than one million lives and $20 billion. Mozambique's government knew the country had become a Cold War battlefield, but it took two donor strikes, in 1983 and 1986 when food aid was withheld, before Mozambique made its "turn toward the West". Mozambique agreed to structural adjustment, privatisation and a transition to the market economy (Hanlon 1996:15-17). Aid jumped from $359 mn in 1985 to $875 mn in 1988 (Hanlon 1991:269). Government spending was cut, including on health and education, and privatisation - which had begun in 1980 - was accelerated.

In this period, donor support was important in four areas of corruption.

1) The first was privatisation itself, which was seen as a high priority to be carried out as quickly as possible. Donors looked the other way when small firms were passed on to friends and family of the leadership; the view seemed to be that transparency would slow the process. There was also some support for the view that because of colonial restrictions, there was no national business class, and that the nomenklatura were the most experienced administrators.

2) In 1988 the Caixa de Crédito Agrario e de Desenvolvimento Rural (Agricultural and Rural Development Fund) was set up using donor counterpart funds to give "loans" to military men and party officials, with no intention that the loans would be repaid. Donors accepted that the money was being used to buy out military people and Frelimo party officials opposed to ending the war and abandoning socialism.

3) The World Bank's 1989 Small and Medium Enterprise Development Project was intended to help the new owners of privatised businesses. Nearly $33 mn was lent, and the World Bank's 1998 evaluation admitted that 90% of the loans would never be repaid. The Banks evaluation admitted that "the Bank is alleged to have put substantial pressure on the management of the banks to ensure the expedient disbursements of project funds; this undermined even further the credit quality of the subloans." A World Bank Industrial Enterprise Restructuring Project was similar and gave $30 mn in loans to larger privatised state companies, most of which will probably never be repaid (Landau 1998:62-63).

4) An organisation close to the President was appealing to donors for project funding, and by the late 1980s was already considered corrupt. At the time, I was writing a book on the aid agencies (Hanlon 1991) and I asked donors why they continued to support this organization. They replied, quite openly, that they were having trouble spending their increased aid budgets and they saw this as a way of buying influence to the President's office to approve their projects. Projects which violated government policies were approved by this route.

This period of the transition was important, because Mozambican officials and newly emergent businesspeople with little experience of the world of capitalism were, in effect, being given a crash course by the donor community. And the lesson was that capitalism is not about profit but about patronage - businesses are "privatised" and given "loans" that need not be repaid according to who you know and donor whim.

Despite Landau's report, the World Bank has never accepted any responsibility for putting "substantial pressure" on honest Mozambican bankers to bend the rules to give loans they knew could not be repaid. Indeed, World Bank Resident Representative James Coates stressed to me in an interview in 2001 that he still expected Mozambique's government to repay those loans to the World Bank (although part will be written off under HIPC debt cancellation).

3.2 Minimal government

The late 1980s were the period of the shift to market capitalism, under the tutelage of the World Bank. The first half of the 1990s were the era of inflation control and minimal government, imposed by the IMF (Hanlon 1996:24ff). It was the era in which the international financial institutions believed the less government the better, and that development must be left to the private sector. The IMF imposed savage cuts on government spending. It was not until the IMF began to bar donor countries from giving aid for post-war reconstruction, provoking an unprecedented public protest from donors in October 1995, that the IMF was forced to ease slightly its spending cap (Hanlon 1996:134).

Salaries were the biggest component of the government spending. A UN study showed that of 110,000 civil servants, more than half were in health and education, and the army had only 12,000 people. The study concluded that far from being too big, Mozambique's civil service was already too small to provide basic services (Adedeji et al. 1995). The only way to meet the savage IMF spending cuts was to cut wages; within five years, salaries of front line staff such as teachers and nurses were one-third of what they had been in 1991. Corruption was inevitable, as front line staff demanded extra payments or took time off to earn money or till fields, in a desperate attempt to feed their families. A woman going to a maternity hospital had to have $2 to pay the midwife. "Demanding money is illegal. But the midwives say 'we work so hard all day here that we don't have time to grow food as other women do'," a Nampula woman told me. A Sofala primary school teacher commented "we in education have one foot inside and other out, because we are parents and we don't like to see our children dying of hunger." (Hanlon 1996:2,4)

It was obvious at the time that paying poverty wages would create corruption.

In parallel with the low civil service salaries, donors and non-government organisations began to pay key technicians and civil servants high salaries to work for them instead of the government, directly decapacitating and weakening the government. More seriously, donors began to give key civil servants extra money - for attending donor-run seminars during the business day, and for doing consultancies instead of their government job. Donors encouraged civil servants to steal time and do outside work for others instead of what they were being paid to do by the government, creating a climate of donor-approved corruption.

3.3 Privatising the banks

The Mozambican bank scandal has been the subject of other articles (Hanlon 2001, 2002a) and only a few points will be repeated here. In the early 1990s banking was liberalised; the first new private bank, Banco Internacional de Moçambique (BIM, Mozambique International Bank), owned 50% by Banco Comercial Português (BCP) and 25% by the World Bank's International Finance Corporation (IFC), opened in 1994. There was growing discussion about the privatisation of Mozambique's two state-owned banks, Banco Popular de Desenvolvimento (BPD, People's Development Bank) and Banco Comercial de Moçambique (BCM, Commercial Bank of Mozambique).

Privatisation of BCM became a "necessary condition" for World Bank aid in 1995. The only candidate was a consortium put together by António Simões, a Portuguese businessman with interests in the Mozambican insurance and metal-working sectors. His group included Banco Mello of Portugal and a company believed to be fronting for the family of President Joaquim Chissano. The leadership of the central bank, Banco de Moçambique (BdM), was noted for its integrity and honesty, despite the growing corruption in public life, and made it known that Simões was not acceptable because he had a number of bad debts with local banks, was failing in his efforts to rehabilitate the metal-working sector, and was not accounting for concessional loans he received from donors for this purpose. BdM began a desperate search for an alternative bidder, but the World Bank backed Simões and said BCM had to be given to him - and it was, on 26 July 1996.

BCM already had corruption problems, and the new owners did not do the normal due diligence audit of the bank - meaning it would be impossible to find out which frauds occurred before privatisation and which after. Some new management was brought in, and one official said they found a wide range of frauds. "The bank needed a total clean-up. But it never happened. The shareholders told us not to." In 1998, Simões sold his shares to Banco Mello, which was subsequently sold to BCP, and in 2002 BCM was merged into BIM. The amount of total losses is still disputed, but is close to $200 mn.

Meanwhile the IMF demanded the BPD be privatised by the end of 1996, in early 1997 it said aid to Mozambique would be cut off if the bank was not privatised soon, and on 8 May 1997 it set a deadline of the end of June 1997. A Mozambican group close to President Chissano's family had been set up in 1996 but could not find a foreign partner, until Chissano made a personal request to the Malaysian Prime Minister Mahathir Mohammed, who instructed the Southern Bank Behard to become a partner. Privatisation went ahead on 3 September 1997. As with BCM, corruption was endemic from the first, and no due diligence audit was done. The bank made loans to members of the Mozambican elite who seemed to have no intention of repaying. BPD, by now renamed Banco Austral, collapsed and the private owners handed their shares back to the government on 3 April 2001. Losses will exceed $150 mn, and a substantial amount of money will be required from the government. In 2002, Banco Austral was taken over by ABSA of South Africa.

In another article (Hanlon 2002a) I argue that there is an ongoing struggle within the Mozambican elite, between a "predatory" faction which sees state capture as the only way to rapidly develop a national bourgeoisie and a "developmental" faction which promotes a longer term entrepreneurial perspective which requires a more interventionist, functioning and honest state. Bank privatisation was an important site of struggle between these two groups, with the IFIs backing the predatory faction. The banking scandal brought the issue to the forefront, and caused two very public assassinations of figures linked to the developmental faction.

Carlos Cardoso, editor of the faxed business daily Metical, became the defacto spokesperson of the developmental group. He had been investigating the bank scandals when he was machine-gunned in a drive-by shooting on 22 November 2000. With the collapse of Banco Austral, the developmental faction tried to regain control and impose some integrity by appointing António Siba-Siba Macuácua, the respected Banco de Moçambique head of banking supervision, as acting head of Banco Austral. But when he began to try to collect loans from the Frelimo elite and to repossess properties, he was killed and thrown down the stairwell of the bank's headquarters on 11 August 2001. At first neither murder was investigated, with the police either unwilling or not permitted to pursue what were assumed to be high level killers. Because he was an internationally known journalist, Cardoso's killing became a subject of an international campaign, and eventually an investigation began; those alleged to have done the actual killing have been arrested and charged. Siba-Siba was not well know internationally or in the donor community; there has been no international campaign about his assassination and there has been little investigation of his murder.

With one exception, there has been no investigation of the thefts of nearly $400 mn from the banking system.

3.4 Ignoring the crisis

The issue is not new. Back in 1996, in a book on Mozambique, I attacked the widespread corruption at all levels of government. But I also wrote: "Donors have fallen into corrupt practices as well, and may well have led the descent. They, too, must change. This means the obvious line of neither paying nor offering bribes and of denouncing thefts and misuse of donor funds." And I called for "an end to indirect bribery … where a donor or NGO threatens to withdraw assistance unless the recipient does something it doesn't want to do." (Hanlon 1996:144) This is not to claim any prescience, but in fact, to claim the opposite - that the issue was already well known and discussed then.

The banking scandals did not occur in isolation. From 1998, Carlos Cardoso began to raise the issue of donor funding, and point out that loans from Norway, Sweden, France, Germany and Switzerland seemed to have been used by António Simões to buy BCM instead of to rehabilitate the metal-working industry. Cardoso continued to raise the issue until he was murdered. Yet none of the donors would say if the loans had been repaid, nor would any admit to even asking the government what happened to their money. Privately they were embarrassed; some admitted they could not find the documentation on loans made six years earlier, but none asked Cardoso to see his copies of the loan agreements.

Despite the campaign by Cardoso and the pressure from honest government and civil society forces in Mozambique, donors still did not see the bank scandal and other corruption as a problem for them. But the assassination of Cardoso, who was known personally by many donor staff, the large injection of government money into BCM, and then the collapse of Banco Austral combined to cause disquiet in some parts of the donor community. It was pointed out that with donors funding a significant portion of the government budget, it was the donor money that was plugging the hole in the banking system, replacing the money stolen in part by senior government and Frelimo people.

The first test came in mid-2001, when the donor community was asked to approve the government's poverty reduction strategy paper (PRSP) and, with it, debt relief under the Enhanced HIPC (Heavily Indebted Poor Countries) Initiative. Mozambican civil society reminded sympathetic donors that the donor community carried far more weight than civil society, and appealed to them to put some pressure on the government. Some Nordic donor officials in Maputo called for approval of the PRSP to be delayed until the government at least provided more information on the banking scandals.

Again, other donor priorities took precedence. At that time, only two countries, Uganda and Bolivia, had had any debt cancelled under the Enhanced HIPC (World Bank 2002d:7), and the IMF and World Bank were under heavy pressure from campaigners. This was felt particularly in the United States, where the government had initially opposed debt cancellation and was seen as blocking HIPC. Instructions came from Washington to the US embassy in Maputo that Mozambique had to be approved for HIPC at all cost, and USAID officials intensely lobbied the Nordic donor representatives, and successfully convinced them that debt relief was more important than corruption. The matter was not raised, the PRSP was approved, and Mozambique gained debt relief.

The next opportunity came at the donor Consultative Group (CG) meeting in Maputo 25-26 October 2001. This was just two months after the murder of Siba-Siba Macuácua; no investigation was under way and his efforts to collect bad debts had been stopped. There was much high-flown rhetoric from donors about the assassinations, corruption, and the bank scandal. Chairman Darius Mans noted that "most delegates urged further actions including: aggressive efforts to recover non-performing loans [and] legal prosecution of perpetrators of crimes to the full extent of the law". He added that delegates "welcomed [Finance Minister Luisa] Diogo's commitment to ensuring that financial expenditures related to recapitalizing the banks do not crowd out poverty-related spending" (Mans 2001b). Mozambique asked for $600 mn in aid and was given $722 mn - the extra money was enough to plug the hole in the banking system. It was after this that former security minister Sérgio Vieira wrote that the donors recognise "the good performance of the government" and this "overrides the bank scandal and the assassinations of Siba-Siba Macuácua and Carlos Cardoso" (Domingo, 2 Dec 2001).

The issue came up again in May 2002, when the G-10 group of donors which provides direct budget support to the government met to discuss the mid-year position. Several noted that there had not been "aggressive efforts to recover non-performing loans" nor had there been any investigations of various major crimes. Two donors wanted to delay the second tranche of budget support to bring pressure to bear, but others did not support this.

3.5 Is there no liability?

The World Bank now has a special anti-corruption website which admits "it did not explicitly address corruption in its development strategies" until 1996 (World Bank 2002b). And it now admits that some of its policies which caused so many problems in Mozambique, were, in fact, wrong.

World Bank Senior Vice President Nicholas Stern (2002:xii) noted that it is now recognised that "the minimal-government free-market approach advocated by many people in the 1980s and early 1990s" will not achieve the millennium development goals - a slightly coy comment since the "many people" mainly worked for the IMF and World Bank, and a bit late, considering the damage that was done. By 1995, the IMF and World Bank had forced Mozambican public service wages down to one-third of their level four years earlier; nurses and teachers had fallen below the "abject poverty line" (Hanlon 1996:49). In 1995, UNICEF and UNDP published a remarkable booklet called Pay, Productivity and Public Service (Adedeji et al., 1995) which looked at five African countries, including Mozambique. In their introduction, UNICEF head Richard Jolly and UNDP Africa head Ellen Johnson Sirleaf wrote that "an efficient and effective public service … is a sine qua non for human development." The booklet pointed to the sharp decline in morale and quality of public service across Africa, and argued that "a key cause of the decline is the erosion of basic service providers' remuneration." The result, it said, was less time and less attention to work, as well as corruption and "privatised user fees". It concluded that "the elements associated with the decline in public service quality are all elements of coping strategies designed to ensure household survival. They cannot be eradicated so long as additional incomes beyond government pay are necessary for the survival of the majority of public servants who are front-line service deliverers."

But the World Bank is rarely open to criticism of its core policies. Instead of listening to what two major United Nations agencies were saying, it called on the UN to suppress the booklet because of the implicit criticism of structural adjustment; the UN did, and few copies were ever distributed.

Seven years later, the World Bank (2002c) anti-corruption website says that "adequate pay" in the public service is essential for preventing corruption.

Years of poverty wages have left the civil service deeply corrupted, as the Ética Moçambique study showed. Wages have again risen above the poverty line, but the "privatised user fees" and other coping strategies built up during the years of poverty have not gone away. It is true to argue that Mozambican civil servants are now being corrupt when it is no longer necessary, but it is also important to remember that it was the international financial institutions (IFIs) which forced them to be corrupt in the first place and which suppressed any criticism of policies it now admits were misguided. The Mozambique government carried out policies imposed by the IFIs which created the administrative corruption problem, but the IFIs have washed their hands of the problem and blame Mozambique.

Nines years after the event, the World Bank admits it pushed Mozambican banks into making corrupt loans, in order to speed privatisation - but it still expects the government to repay those loans.

Stern (2002:3) says proudly that "the Bank has also learned from its failures". But it is the Mozambican predatory elite which has profited from those "failures" and ordinary Mozambicans, not the Bank, who are paying for those lessons.

3.6 What are 'good policies'?

The World Bank sets the tone for the donor community in Mozambique. The Bank talks frequently about the need for "good policies", but these are rarely actually defined in any but the vaguest terms. A search of World Bank literature, however, shows that much of the work on "good policies" was done by David Dollar and the Bank's Macroeconomic and Growth Division, and that the seminal paper to which all others refer was produced in 1997 by Dollar and Craig Burnside. They say "the heart of structural adjustment are fiscal discipline, trade liberalization and other market friendly policies." They explicitly equate "good policies" with "good economic policies", and then define an index of good policy based on just three factors: government budget surplus, inflation (as a measure of monetary policy), and trade openness (Burnside & Dollar, 1997:1,2,16).

In practice, as distinct from the rhetoric, the World Bank measures "good policy" purely in monetary and trade terms. When the crunch comes and the World Bank is actually evaluating a country, the talk of poverty focus, democratisation, anti-corruption and a host of other issues count for nothing.

[click here to go to the last part]
[click here to go back to the first part]