Review of African Political Economy No.91:53-72
? ROAPE Publications Ltd., 2002
Bank Corruption Becomes Site of
Struggle in Mozambique
Three people have been murdered for investigating corruption in the Mozambican banking system and the loss of more than $400 million. All countries use banks politically, and in Mozambique, the banks were first used to build socialism, then to keep the country running during the war, and finally in the new capitalist era to promote local entrepreneurs and keep the economy out of foreign hands. But the nature of socialist banking and the process of transition combined to create the conditions under which powerful individuals could use the banking system for accumulation. But this has been contested, and there is an ongoing struggle within the elite between those groups which back what Peter Evans calls the ‘predatory’ and ‘developmental’ states. The recent murders suggest this contest is becoming more acute. Finally, we note that a key role has been played by the international financial institutions, which in their doctrinaire opposition to any serious role for the state chose to back the predatory state faction.
From Socialist Banking to War Economy to Capitalism
At independence, the Bank of Mozambique (Banco de Moçambique, BdM) was created as the new central bank on 17 May 1975 by simply transforming the Banco Nacional Ultramarino. But the new Frelimo government did not give priority to the banking sector, and it was more than two years before any further action was taken. With growing evidence that the private banks were hostile to the new government and were facilitating capital flight, particularly by Portuguese who had left the country, the government finally nationalised all but one of the remaining banks on 31 December 1977 (Wuyts, 1986:172). Four banks were merged into BdM and two were merged to form the People’s Development Bank (Banco Popular de Desenvolvimento, BPD); only Banco Standard Totta de Moçambique (BSTM) remained private. The BdM became the only bank that could deal in foreign exchange on 1 September 1978. It was the treasury, central bank, the controller of the execution of the state plan, and a commercial bank. BPD was a credit institution taking small deposits and specialising in agricultural credit.
In part, the slow movement on the banks was because Frelimo followed the Soviet view that in a socialist centrally planned economy, material balances become the principal method of allocating resources and money loses its importance. The state budget financed capital expenditure, while the banking system financed circulating capital. In the first years of independence, the priority was to keep production going and not dismiss workers, despite the flight of Portuguese managers and technicians; state-controlled banks were instructed to finance the deficits of these enterprises to keep them running. Marc Wuyts (Wuyts, 1986:171) comments that despite being inflationary, it
was the most direct and effective instrument in the combat against economic sabotage. Far from being a destabilising factor, this policy was crucial in stabilising employment, preventing a further collapse in production, and preserving stable prices.
With the intensification of the cold war, South Africa in 1981 launched its war of destabilisation and
over the next decade, caused damage of more than $20 billion and 1 million deaths (Hanlon, 1996:15). The government printed money to finance the war and again banks were lending to cover enterprise deficits, in this case increasingly caused by the war. The amount of internal credit doubled between 1981 and 1984, mainly due to the need to reschedule debts of enterprises with persistent deficits (Wuyts, 1986:70, 96, 160, 172, 193).
However, the failure to transform the state sector of the economy and its interaction with the rest of the economy led to a separation between official and parallel economies, with the peasantry located in the parallel economy. Money supply grew, and surplus money migrated into the parallel economy, and quickly moved into the hands of speculators and of a growing private commercial capital (Wuyts, 1986:9, 215, 257). Prakash Ratilal, then governor of BdM, told parliament ‘each month the stock of money in the hands of the public increases without there being any outlet for it.’ This, in turn, created
social instability, he said (Notícias, 25 Mar 1983).
The year of dramatic change was 1986. Faced with a worsening war and intense pressure from the West and donors, the government moved to liberalise the economy and increase the role of private business, in particular trying to control the money supply by raising prices and by allowing more private investment. In April 1986 Abdul Magid Osman became finance minister and Eneas da Conceição Comiche became governor of BdM (he had been President of BPD since 1978 and vice-minister of Finance 1984-86). On 19 October 1986, Samora Machel was killed. On 14 January 1987 Mozambique introduced the Economic Rehabilitation Programme (Programa da Reabilitação Económica, PRE), which devalued the local currency (the metical), raised prices, and allowed workers to be laid off. Bank credit regulations were tightened and credit was to be directed more to importing raw materials, and less to covering on-going deficits. In 1987 the government recognised 34% of the BdM portfolio – then
about $160mn – as bad loans of public enterprises. Of this, $40mn was rescheduled and $35mn actually paid off by the government. Government recognised 78% of BPD’s portfolio – $80mn – as bad debts and
rescheduled or paid off most of that (World Bank, 1990:104).
Although there was an attempt in the PRE to move to more traditional banking, the ongoing war meant that bank credit remained important to keep cooperatives, state companies and even quite large private companies functioning. José Miguel Sequeira Braga, a BPD administrator until he retired in 1995, noted that the ‘poor economic and financial situation of nearly all companies during the war resulted in an accumulation of debts to the bank.’ Furthermore, ‘the government instructed that capital be injected into these companies to permit them to continue to function and prevent the dismissal of tens of thousands of workers.’ But Braga added that although perhaps half of BPD loans would not be repaid, ‘banks in Europe after the Second World War were in as bad shape, or even worse’ (MediaFax, 5
And in the new free market economy, money from the banking system was essential to promote Mozambican business and to prevent a complete foreign take-over and re-colonisation. Finance Minister Abdul Magid Osman said that ‘Mozambique needs an elite of entrepreneurs who feel that society values their importance. … In my opinion, the creation of a large, dynamic and enterprising Mozambican entrepreneurial class is essential’ (Hanlon, 1991:224). There was a widespread view of the old socialist elite that they should have help to become the new capitalist elite. Honour, Corruption, Development & Predation
Nothing we have discussed so far is in any way ‘corrupt’. In a socialist, centrally planned state, it was correct and proper to give money as directed by the plan. During the war, it was essential to use the banks’ money to keep the economy running. Any country – socialist, capitalist or mixed – supports its
preferred entrepreneurs, whether they are big state companies, cooperatives, private companies, or individuals. This normally involves regulatory preferences, priority access to state resources, and access to credit and the banks. Thus it was obvious that the state and the banking system would give easy credit to the new private businesses. And the existing banking system, which was accustomed to propping up state enterprises and co-ops, switched smoothly into propping up the new private
There has been extensive discussion of corruption in this journal (Szeftel, 2000a; 2000b) and we take Morris Szeftel’s definition that corruption is ‘the misuse of public office, public resources or public responsibility for private – personal or group – gain’ (Szeftel, 2000b:427). Much of the rest of this article discusses corruption in the banking system. But in this section we need to make a number of preliminary points.
First, we argue that the socialist and wartime banking system created the conditions which made later corruption easy. In the early years of independence, Mozambique showed an incredible degree of integrity, honesty and even puritanism (Hanlon, 1991:230-2). Because Frelimo did not give a high priority to the banking system, proper systems were never created and procedures remained irregular. So long as it was according to the plan, company deficits were covered by loans which were never expected to be repaid. The overlap between the treasury and the banking system was total – when
Samora Machel was travelling, someone from the presidency would simply phone BdM or BPD and ask for thousands of dollars, in cash, for the delegation. In a highly authoritarian country, any bank official will follow the instructions of the president. The system may have been irregular, but it was not corrupt, and it worked because of the honesty, integrity, and good will of the people in the banking system. But that left a system dependent on individuals, with few regulatory systems and controls in place. By the end of the Samora era, extravagance was creeping in, as presidential delegations going abroad took more money from the banks for shopping on foreign trips. With no rules, no one in the banking system could say no. That created fertile ground for the corruption that came later. Second, we argue that the growth in corruption after 1986 was intimately linked to the process and model of economic ‘reform’ imposed on Mozambique by the Bretton Woods institutions and the international donor community. Szeftel (2000b:429) notes that ‘far from arresting the upward spiral of
corruption, the economic liberalisation and attendant governance reforms imposed by the donors have sometimes intensified it beyond anything the government can manage or control’ and that ‘increasing
levels of corruption and criminality are often the consequence of crisis and restructuring rather than their
cause’. We will argue that exactly this occurred in Mozambique, in several interrelated ways. Szeftel argues that ‘the legacy of imperialism excludes the indigenous petty bourgeoisie from entry into the
most advanced or capital-intensive sectors of the local economy, confining it to peripheral activities.’ Accumulation is dependent on access to the state and its resources, and it is ‘unlikely that the market
forces which had produced underdevelopment and exclusion would mysteriously reverse themselves once independence was attained.’ (Szeftel, 2000a:301; 2000b:427, 431) The result of this, in Mozambique and other countries, was that the new capitalist state continued to direct resources to its preferred entrepreneurs, to try to keep some domestically controlled economy. In Mozambique, privatisation was particularly opaque, and the government tried to keep at least medium-size companies out of the hands of foreign owners.
The World Bank, which was pushing privatisation, could never allow the state to publicly say it was restricting bidding to Mozambicans, but the Bank was willing to turn a blind eye as long as privatisation accelerated. The effect then was that any efforts to build an indigenous entrepreneurial group had to be done in the shadows – which obviously created the conditions for corruption. Szeftel
(2000a:303) makes two other points which are relevant. First, although corruption in Africa has transferred substantial money to the ruling elite, it has not promoted the development of an autonomous capitalism, and this too, is the case in Mozambique. Second, as reforms and donor pressures reduce the available resources, this actually increases the pressure for corruption and people compete for diminishing resources; this also happened in the Mozambique banking sector. The other issue about World Bank and IMF-driven reform is that its core assumption is that if the state is made smaller and weaker, then there are fewer opportunities for corruption. But the opposite has occurred in Mozambique, both in the privatisation process itself proving a source of corruption, and because the state has reduced capacity for regulating and policing corruption. Riley (2000:139; 1998:135) notes that the World Bank and other donors until recently turned a ‘blind eye’ to corruption in developing countries. We argue that in Mozambique this continues with an implicit agreement that
corruption will be tolerated as long as Mozambique remains the Bank’s star pupil.
Finally, we turn to the ongoing struggle inside the Mozambican elite. Peter Evans (2000:44-8) writes of the ‘predatory state’ which is characterised by total marketisation, where everything is for sale, including judges. It is based on a patrimonial bureaucratic system, with state power controlled by personally connected individuals. And it tends to disorganise civil society. The ‘developmental state’, he says, ‘foster[s] long-term entrepreneurial perspectives among private elites by increasing incentives to engage in transformative investments and lowering the risks.’ This is not about capitalism versus socialism, but rather between capitalist roads – between visions of developmental social democracy
and of primitive capitalism. Evans also notes that neo-liberalism is not necessarily antithetical to the predatory state. He cites Brazil, where traditional families moved into positions of power and turned modernising projects into support for a traditional oligarchy.
In this paper, I want to argue that the struggle is between that proponents of the predatory state, placed closely to President Joaquim Chissano, and the proponents of the developmental state, who tend to dominate the Ministry of Finance and the central bank (BdM). The developmental state advocates also include a number of people who played important roles under Samora Machel, and tend to retain from the socialist era a stricter set of ethics.
Moving to the Market
The first overt corruption in the banking system started in the late 1980s, when there was significant and quite open illegal trading in foreign exchange. Dollars could be bought and sold at a number of shops in major cities, particularly those run by Asian-origin traders, and the parallel metical to dollar rate was quite widely quoted. A few senior BdM officials were corrupted and began assisting trading families in their illegal foreign exchange dealings. One particular scam took advantage of the dual exchange rate for the Rand maintained by the apartheid government in South Africa; Asian traders were seen openly going to the house of a BdM official with bundles of Rand.
With the ‘turn toward the West’, aid doubled from $359 million in 1985 to $710 million in 1987 (Hanlon, 1991:61), putting hard currency into the economy but also serving as a basis for corruption. By the early 1990s, at least two senior officials had more than $3 million each in London banks. People close to the President’s office were known in the aid community as corrupt – but donors queued to give them
money, hoping to curry favour with the presidency. Links to the banking system were direct, with officials in the Presidency telephoning senior officials of the two banks to demand that loans be given to certain companies or cash be given to individuals.
The transition to the market economy led increasingly to practices within the banking system which may have been politically necessary, but which were in banking terms questionable. In 1988, the government set up the Agricultural and Rural Development Fund (Caixa de Crédito Agrario e de Desenvolvimento Rural, CCADR), which used donor counterpart funds to give ‘loans’ to military men and party officials, with no intention that the loans should ever be repaid. This was an attempt to buy out people opposed to the move to the market economy and attempts to end of the war. As the World Bank (1990:105, 107, 115) commented, the CCADR recognised ‘the need to finance the resettlement of veterans in civilian life in the productive sector (rather than through pensions)’, but it noted ‘that
CCADR credit had been extended mostly to urban borrowers rather than rural ones, and that it had financed mostly transportation and marketing equipment rather than long-term investment.’ CCADR
was managed by BPD.
The World Bank’s 1989 Small and Medium Enterprise Development Project was similar. By 1997, $32.6mn had been lent though BdM (later BCM, see below) and BDP. The World Bank in its subsequent evaluation said that 90% of these loans would never be repaid (Landau, 1998:63). In a 12 October 2000 letter to Carlos Cardoso, editor of Metical, the World Bank Resident Representative James Coates
claimed that the ‘the World bank was not directly associated with the management of these funds’. But the World Bank’s own evaluation contradicts this. It says ‘The [World] Bank is alleged to have put substantial pressure on the management of the banks to ensure the expedient disbursement of projects
funds; this undermined even further the credit quality of the subloans.’ Evaluation team head Luis
Landau goes on to note that:
The implementation of the project was adversely affected by a high staff turnover of [World Bank Washington-based] task managers; this also had a negative impact on the quality of the project (Landau,
A World Bank Industrial Enterprise Restructuring Project was similar and gave $30 million in loans to privatised state companies, probably never to be repaid. This raises a fundamental question: What message was the World Bank sending to the Mozambican elite when it was putting heavy pressure on banks to make loans that it was obvious would never be repaid? At a crucial time when Mozambican entrepreneurs were just learning how to be capitalists, the World Bank told them loans did not need to be repaid.
In addition, privatisation had begun and banks lent money to Mozambican entrepreneurs to buy and rehabilitate state companies – as well as for cars and other non-productive expenditure – and seemed
not to expect borrowers to repay. BdM and BPD officials began approving loans to people in exchange for a commission, knowing that the loans would not be repaid. Finance Minister Magid Osman in 1990 noted that
the current tendency is towards the creation of a class based on dubious business deals, and that requires various ‘bonuses’ and protectionism from the state, to the detriment of the consumer (Hanlon, 1991:224).
Proper record-keeping would have picked up some of these frauds, but both BdM and BPD had inadequate systems. Attempts were made in the late 1980s and early 1990s to computerise and to modernise and upgrade record-keeping, but this was consistently blocked by people inside the banks who wanted the old systems as cover.
Back to Private Banking
The early 1990s saw two processes running in parallel – an increasing and eventually dominant role for
the private sector in banking, and the beginning of the struggle between the developmental and predatory state forces. The former tried to curb the worst excesses of the latter, and also tried to create islands of integrity.
In response to the explosion of illegal foreign exchange dealings, the international financial institutions put pressure on the government, and on 18 September 1990 a decree was issued allowing a free market in foreign currency and allowing individuals and institutions to open foreign currency exchange bureaus. The exchange bureaus were supposed to keep records and file reports with BdM, but in reality there was never any proper control. In 1991 the state monopoly on insurance was ended, and a private insurance company Impar (Companhia de Seguros de Moçambique) was established. Lead shareholders were a Portuguese businessman, António Simões; Inocêncio António Matavel, an important Frelimo-linked businessman; Madal, which Industry Minister António Branco joined after ending his term earlier that year; and BPD. Portuguese investors included Companhia de Seguros Império, which was controlled by Banco Mello and which provided technical support. BSTM had never been nationalised, but the first new private bank, Banco Internacional de Moçambique (BIM, Mozambique International Bank) opened in 1994. Initial capital was $24mn, and it was owned 50% by Banco Comercial Português (BCP), 25% by the World Bank’s International Finance Corporation, 22.5% by Mozambican government organisations, and 2.5% by Graça Machel’s Community Development Fund (Fundação para o Desenvolvimento da Comunidade, FDC). Its chair is former Prime Minister Mário Fernandes da Graça Machungo and its managing director was from BCP, José Alberto de Lima Félix. Banking sources say that although Machungo is in overall control, most key day-to-day decisions are taken by Portuguese staff named by BCP. Jorge Jardim Gonçalves is chairman and managing director of Banco Comerical Português, first set up in 1985 but which grew to be ones of the largest private banks in Portugal; Gonçalves is a prominent member of Opus Dei.
Meanwhile, attempts were underway to sort out the state-owned commercial banking sector. The commercial and central banking sectors of BdM began to be separated in 1987, but only moved forward in 1991 when Eneas Comiche was promoted to be Finance Minister and Adriano Afonso Maleiane was promoted to replace him as Governor of BdM. Within two months of his appointment, Maleiane had issued regulations and named directors for the commercial banking sector of BdM. On 25 February 1992 this was formally split into a new bank, Banco Comercial de Moçambique (BCM). Maleiane and senior BdM officials refused to talk to us. But in interviews, senior banking officials all made clear that Maleiane’s priority was transforming BdM into an effective and honest central bank, and he succeeded. BCM was given what was left. Maleiane knew about corruption in BdM and he moved the corrupt, incompetent and lazy staff over to BCM – mixed in with good staff to keep the bank going.
Thus the state owned two commercial banks, BCM and BPD, and a 1991 World Bank study suggested that both were so deeply in debt and in such a mess that they should simply be closed. This was never a serious option, but privatisation was on the cards. Opinion within the government was divided. Magid Osman as Minister of Finance wanted them privatised – independently of the World Bank he had
concluded that they were already too corrupted and could not be controlled as state banks. However bad the privatisation, it would be cheaper to hand them on to the private sector. Between when it was established as a separate bank and when it was privatised (1992-96), the government had put in more than $100mn, according to Finance Minister Tomaz Salomão (Mozambiquefile, September 1996:12).
Others opposed privatisation. Some wanted the banks cleaned up and to have an active state banking sector, while some senior Frelimo figures opposed privatisation because they wanted to continue to be able to allocate funds on both political and personal grounds. Awaiting firm political guidance, Comiche and Maleiane initially aimed at improving the efficiency of the banking sector and the regulatory capacity of BdM. The World Bank stepped up pressure, and privatisation of BCM was one of seven ‘necessary conditions’ of the World Bank’s 7 November 1995 Country Assistance Strategy, meaning
that if BCM was not privatised, the World Bank would end its programme, which would cut off all aid to Mozambique. (Ending cashew industry protection was a much more controversial necessary condition of that CAS; see Hanlon, 2000). The World Bank seems to have been convinced that at least middle level international banks would be interested in BCM and BPD. In both cases, there was initial interest, but prospective buyers dropped out as soon as they had a look at the books. There were too many bad loans and chaotic accounting systems. The 11 April 1996 joint IMF-World Bank Policy Framework Paper, which set conditions that the government must meet, required BCM be brought to point of sale by June 1996 and BPD by the end of 1996. Relations between BdM and the World Bank were poor on the privatisation issue. Eventually, BCM staff were told they could not talk to the World Bank resident economist directly, and that he would have to deal with them via BdM. BCM was the largest commercial bank in Mozambique with a reasonable system of local branch banks and more than $100mn in foreign deposits, much of which could be transferred to accounts in the parent bank if it was taken over. It also had chaotic accounts and administration, which could be an advantage or disadvantage. Banco Português de Investimento (BPI) which coordinated the privatisation process said in its sale memorandum that $22mn of its loans should be treated as bad debts, and that a further $13mn in bad debt provisions was required. The government said that the Treasury (effectively BdM) would assume this liability (Metical, 16 March 1998).
In the end, there were only two proposals to take over the 51% of BCM that was on offer. One was a consortium put together by António Simões which was 50% Impar, 35% National Merchant Banks of Zimbabwe(NMBZ), and the rest Banco Mello of Portugal. (Público, 16 April 1998). The involvement of
NMBZ was never clear, but a senior NMBZ official told us quite openly that he was ‘fronting for a group of people in Mozambique’ which was unable to invest directly because there needed to be a bank involved. He refused to say who the Mozambicans were, but a series of senior banking figures said that NMBZ represented the Chissano family. The other prospective bidder was Caixa Geral dos Depositos of Portugal.
But who was António Carlos de Almeida Simões? He came from an old Portuguese industrial family which had one small company in Maputo, Empresa Metalúrgica de Moçambique (EMM), which
continued to operate after independence. In the mid-1980s, Simões came to Maputo and developed plans to unify and modernise the metal-working sector in Maputo. Simões won the backing of the then Industry Minister António Branco and of Octávio Filiano Muthemba, the then vice-minister in charge of heavy industry who replaced Branco as minister in 1992. Simões expanded rapidly and in 1990-92 developed a network of companies which pulled together these people and their families. EMM and the state jointly set up two companies, CSM (Companhia Siderúrgica de Moçambique), which took over the government-run Cifel steel mill (Hanlon, 1984:191), and Trefil (Companhia Moçambicana de Trefilarias). Inocêncio António Matavel was administrator of CSM. António Branco became administrator for the state in Trefil. EMM also set up a transport company Transmap (Transporte Rodoviário de Maputo) with Levy Filiano Muthemba, brother of Octavio Muthemba.
With the support of Branco and Octávio Muthemba, Simões borrowed money to import equipment to modernise CSM and develop Trefil. Between 1992 and 1994 CSM and Trefil received $17mn in highly concessional long-term loans with aid money from Norway, France, Germany, Sweden and Switzerland. In addition, Simões companies owed at least $1mn to BPD (Metical, 16 March 1998). ‘As it
happened, CSM was never rehabilitated, leading informed sources to suggest that Simões used part of the $20mn to buy BCM’, wrote Carlos Cardoso, editor of Metical (9 October 2000) Government and
bank sources refuse to say if those loans were repaid. What is notable is the very incestuous nature of these loan contracts. Everyone on the government side moved into bank posts. For example, Ricardo David, then National Treasury Director who signed all the agreements, became an administrator at BCM. Branco, Matavel and BPD were also involved in Simões insurance company, Impar. But the metal industry was not revitalised. The equipment proved to be expensive and much of it was never installed, and despite the loans, CSM received no raw materials after 1990, prior to privatisation (Mozambiquefile,
November 2000). ‘To function, CSM and Trefil needed raw materials, but Simões had no money for raw
materials’, an ex-employee commented. ‘By the end of 1996, he ran out of money, even for wages. He was always at least four months behind in wages, and often more.’ In the end, some people were given equipment and unused material in lieu of back pay.
The BCM privatisation was similarly personalised and politicised. Banco Mello had no interest in Mozambique, but Simões was friends with Vasco de Mello, chairman of Banco Mello, who agreed to participate. Meanwhile Caixa Geral dos Depositos suddenly dropped out; several sources told us that Simões used his political connections to convince Almeida Santos, president of the Portuguese parliament, to in turn convince Caixa Geral dos Depositos that it was politically unwise to compete against a privatisation bid that included the President of Mozambique.
This left only one bidder – António Simões. BdM officials strongly opposed his bid on the grounds that he was already considered a bad debtor. Mozambicans in the banking and business community also opposed the bid on the grounds that Mozambicans in a similar precarious financial position would not be allowed to bid for the bank. Nevertheless, the IMF and World Bank were increasing the pressure on Mozambique to privatise BCM during 1996, threatening to cut off funding if this was not done. Simões complained to the World Bank in Maputo that senior BdM officials were trying to block him, and Bank staff backed Simões, assuming that BdM and Frelimo were simply trying to prevent privatisation. A World Bank official went to Maleiane and ‘read him the riot act’, saying BCM had to go to the only bidder left. At the last minute, BdM found an alternative Portuguese bidder, but this was blocked by the World Bank, which did not want to delay any longer. Faced with a potential cut-off in foreign aid, BdM capitulated. On 14 May 1996 BdM announced that the only offer was from the Simões consortium. This was enough for the IMF, which agreed an Enhanced Structural Adjustment Facility (loan) for Mozambique on 21 June. On 26 July, BCM was formally privatised to the consortium, which had offered $107mn for 51% of BCM – although it is not clear how much of this was paid.
Simões took over the top floor of the BCM building, including the old BCM president’s office, as the
office of his metalworking companies EMM, CSM and Trefi, but he kept a low profile within the bank. Simões hoped to use his new bank to make loans to his metalworking companies, but BdM issued orders to all banks that they should report weekly on the state of Simões accounts, and not grant him credit without permission. Curiously, the new owners of BCM never did the normal due diligence
audit of the bank, so there was never a clear picture of what bad debts had been carried forward. One official of the BCM at that time said they began looking at the books, and found a wide range of frauds. ‘The bank needed a total clean up. But it never happened. The shareholders told us not to.’
BCM was soon in trouble again, with further fraud and deficits. Simões sold his shares in Impar to Banco Mello for $20mn, giving it control of BCM, and in 1999 a new president nominated by Banco Mello finally took over.
BPD was next, but it had major problems. The accounting firm of Deloitte & Touche said BPD needed to make bad debt provisions of $23mn, 52% of its entire loan book. Banco Português de Investimento (BPI) noted in its sale memorandum that credit control was ‘weak’ (BPI, 1996). The IMF insisted that BPD be privatised by the end of 1996 (IMF, 1996), but no foreign bank wanted it. By early 1997 the IMF said that aid would be cut off if BPD was not privatised soon. On 8 May 1997, the IMF set a deadline of the end of June.
In 1996 a Mozambican group close to the Frelimo leadership had been put together under the name Invester, headed by former Industry Minister Octávio Muthemba and with shareholdings of one-third each for Muthemba, Jamú Hassan, and Alvaro Massingue (Metical, 19 September 1997). Invester tried to
find a South African partner, but failed. President Joaquim Chissano made a state visit to Malaysia from 19 to 21 March 1997 with Muthembe and Hassan in the delegation. Chissano is said to have made a personal request to Malaysian Prime Minister Mahathir Mohamed to provide a partner for Invester, and the Prime Minister requested Southern Bank Berhad (SBB) to participate. This was before the Asian financial crisis hit Malaysia in July 1997 and Malaysia was looking for involvement in southern Africa and pushing the concept of ‘smart partnership’. Chissano is said have offered Malaysia preferential
treatment in other areas, such as mining and Maputo property development, in exchange for solving the BPD problem.
Privatisation went ahead on 3 September 1997, with the state keeping 40% and a holding company Investil (Investimentos Associados) taking the other 60%. Investil, in turn, was 51% SBB and 49% Invester. These two new investors were to pay $21mn, but more than $2.5mn of this was never paid (KPMG, 2001). SBB was to provide know-how and new capital, and turn BPD into a modern bank within three years; the bank was renamed Banco Austral (which means ‘Southern Bank’ in Portuguese). SBB insisted on control of the bank, and named K Muganthan as managing director; Muthemba became chairman of the board. There are many party and family links. Muthemba is chair of SPI (Metical, 19
September 1997), the Frelimo party holding company, and he is clearly seen as one of the businesspeople who negotiates for Frelimo as well as himself. President Joaquim Chissano has always refused to publish a list of his property, and the press has always assumed, on circumstantial grounds, that the Chissano family has close links with Banco Austral. At the time of the signing of the agreement, a photograph of the President’s businessman son Nympine Chissano with the Malaysian buyers was
published (Metical, 4 & 19 September 1997). The Malaysian director of SBB, Dato’ Tan Teong Hean, and Muthemba both set up companies with Nympine. Muganthan set up local businesses with Muthemba and other prominent figures.
As with BCM, no due diligence audit was ever done of BPD when it was privatised, so it was impossible to see what was done by the new management and what was done by the old. This was SBB’s first foreign investment and at home it was staggering under the impact of the financial crisis,
and was never able to put in the required money and technical support. Both Mozambican and Malaysian officials began making questionable loans to friends and family, Frelimo party people, and Invester members and companies. Unnamed Banco Austral staff said that loans were being given to people without any guarantees, sometimes in exchange for a 10% commission to a senior official. The trade union complained that loans had been given to friends and family of senior administrative personnel of the bank (Savana, 6 April 2001) and MediaFax (18 April 2001) talked about the ‘generosity’
of K Munganthan. By 2000, BdM had intervened to restrict new lending and force an audit. What is clear in this period is that central bank authorities were fighting a rearguard action to maintain some integrity in the banking system, first by establishing a competent and honest central bank, then by
trying to halt a questionable privatisation led by Simões, then by curbing lending to Simões himself, and finally by belatedly stepping in to Banco Austral. But this came against intense pressure from the Bretton Woods institutions to privatise that banks, no matter how corruptly. In effect, the World Bank and IMF were backing those proponents of the predatory state, in Frelimo and close to the President, who wanted to take money from the banking system.
Bank Officials Plunder the Banking System
Throughout the 1990s, money was being siphoned off from the banks. There had been no due diligence audit of either bank when they were privatised, and computerisation and tighter controls and audits were delayed until well after privatisation. It was 1998 before Banco Mello staff in BCM began to impose tight enough controls, including daily reconciliations and a requirement for daily reports of all large movements of money. The continued use of paper files made fraud easier. Loans were granted and then the original file simply lost, so no one would know, for example, what guarantees had been promised. It was possible to create fake accounts which could be used as part of complex transactions, and then simply delete all record of the accounts. Inactive accounts were drained or used for illegal transactions. Money was taken from government and project accounts which were not closely monitored.
Central to most fraud has been the suspense, internal and balancing accounts which cover the business between branches, between branches and headquarters and between Maputo and foreign banks. For a bank, everything has to be accounted for somewhere, so when a transaction is not yet completed (such as an uncleared cheque), the item goes into a ‘suspense account’. Under normal backing practice, including the rules of BCM, no item should remain in a suspense account for more than 45 days – which
should be enough time even for post from remote branches. Annual audits should check to see that there are no old items hanging about in suspense accounts. Normal practice is to do a ‘reconciliation’ in which various accounts and sets of books are compared, to ensure that all agree. In Mozambique this was not done. In an article in Savana (7 April 2000), an unnamed ‘ex-director of BCM’ said that BCM
had billions of meticais, as well as substantial foreign exchange, in suspense accounts for years, and that it was a strategy of BCM and BdM officials, and of the auditors, to pretend that money was there when it was not. Several banking officials we talked with confirmed that no reconciliations were done. BCM officials were authorising transfers from accounts which had no money and were allowing overdrafts which could not be covered. Another fraud is to issue letters of credit without adequate cover, and when the letters are presented, simply take the money from the suspense account – which
becomes a never empty honey pot. The ‘ex-director of BCM’ claimed that between 1993 and 1996,
$40mn was stolen in this way, ‘on orders from above’. KPMG (2000) audits show the same thing was happening with BPD and Banco Austral.
The first reported fraud of this type occurred in 1993 in Maputo, involving more than $1 million (Mozambiquefile, February 1993). Pedro Pinto and Julio Tandane, who operated cafes and restaurants, were allowed to cash cheques without having money to cover them. BPD took over Pinto Group properties and Umberto Fusaroli Casadei, a well-known Italian-Mozambican businessman, was appointed to run them. He was shot and wounded twice, on 22 April and 12 May 1993, and blamed the Pinto Group for the attempted assassinations. He then left Mozambique (Mozambiquefile, May, June
There was a similar $4 million fraud in 2000. Cheques drawn on one branch of Banco Austral were deposited in other banks in a second city; they passed the cheques on to Banco Austral in that city, and the manager said they had been covered. But he did not forward the cheques for collection, and increasing amounts built up in the suspense and internal (inter-branch) accounts. In violation of normal procedures, no reconciliation of these accounts was done by Banco Austral head office. The biggest single theft is the ‘144 billion meticais’ fraud, then about $13 million, which occurred in the six months before BCM was formally privatised on 26 July 1996. A series of bank accounts was opened at the Sommerschield (Maputo) branch by Abdul Satar Carimo and his family and companies. Cheques
issued on a series of accounts outside Maputo, apparently without the knowledge of the account holders, were deposited in Satar accounts, peaking at $6.6mn in the single month of July 1996. The branch manager, Vicente Ramaya, allowed the money to be withdrawn from the Satar accounts before the cheques cleared, and he then destroyed the cheques so that they did not go through the system. The amounts remained in the suspense accounts. Demos (30 May 2001) claims that at least $7mn in cash was
taken out of Mozambique and some of it was deposited in a British bank. This would not be hard to organise, as the Satar family runs an exchange house, Unicambios. When the fraud was discovered, the Satar’s were alerted and fled the country. Abdul Satar Carimo and his wife now live in Dubai. The government’s failure to prosecute the case soon became the touchstone for endemic corruption in the prosecutors’ office and in the courts, and also convinced many people that the fraudsters had very high
level protection. Diamantino dos Santos was Maputo city prosecutor and he prevented the case from being investigated in 1996 and 1997; he was first transferred to Sofala and later was himself charged with fraud and fled the country. The press, notably Metical edited by Carlos Cardoso, continued to raise
the case. But it only began to move when it was raised in parliament on 14 March 2000 by Eneas Comiche. He had been Minister in the Presidency for Economic and Social Affairs, until he was elected an MP in December 1999; shortly thereafter he was named chairman of the BCM board. He used his parliamentary position to denounce the attorney-general’s office and name Ramaya and the Satar
family as culprits.
The official view put forward by Comiche and BCM is that Ramaya was sophisticated and experienced enough to have hidden a $13mn fraud, not simply by destroying cheques, but also by falsifying various levels of accounts and reports. This is widely rejected by banking experts who argue it was not possible for $6.6mn to suddenly appear in the accounts of a small branch, and not to be noticed at head office; to have not spotted the frauds, even with the weak controls, was either incompetence or corruption. Someone at higher level must have known and participated, or at least chosen to look the other way. And it was clear that the perpetrators were protected, because the investigation was blocked. The ‘ex-director of BCM’ wrote that it was ‘completely impossible’ to have a multi-million dollar fraud
involving suspense or internal accounts without the knowledge or a director or administrator, in particular the director responsible for accounting (Savana, 24 March & 7 April 2000). At that time, the
director responsible for accounting was Teotónio Comiche, younger brother of Eneas. Both Dos Santos and Asslam Abdul Satar have written letters published in the local press in which they claim that officials at the top of BCM were involved, and were also involved in illegal foreign exchange transactions (Domingo, 26 March 2000; Demos, 23 May 2001). There have been at least two shootings
related to the Mt 144bn fraud. On 29 November 1999, shots were fired into the car of Albano Silva, just missing him. Silva is a prominent lawyer who was representing BCM in the Mt 144bn fraud case and who was publicly criticising the attorney general’s office for blocking prosecution of the case. He is the husband of Lúisa Diogo, then Deputy Finance Minister, now Finance Minister. On 14 Feburary 2001, Albino Macamo, who had been appointed assistant attorney general on 28 December, was shot and seriously wounded. He had been investigating corruption in the attorney general’s office and had been part of a commission on inquiry to investigate Diamantino dos Santos.
Another important internal fraud has been bank officials taking money out of the accounts in ‘correspondent banks’ in other countries – the banks that hold foreign reserves, make payments and
collect cheques and other due items. Before electronic banking became common, there is a high error rate; typing mistakes could take months to trace when a transfer failed to appear. Thus there were always a large number of items in suspense accounts, and reconciliation was a genuine nightmare, even with good will. Thus bank officials could siphon money out of foreign accounts by issuing a payment order from the account – against a telex, letter of credit, cheque etc. – but ensuring that the payment did
not appear on the books in Maputo. A gap grew between the amount people in Maputo believed was in an account in, say, New York and the amount actually in the account. But if no proper audit or reconciliation was done, no one would ever know about the difference. The ‘ex director of BCM’ claimed that in May 1995, he detected a gap of $12 million between the accounts abroad and records in Maputo, which the KPMG audit had not noticed because there was a counter-entry in the suspense account. Other bank officials we have spoken to suggest that the gap was larger. There are strong suspicions that more money disappeared in this way from Banco Austral and BCM after privatisation.