‘Africa is bleeding money, as capital flows into the private accounts of African elites and their accomplices in Western financial centres,’ write Léonce Ndikumana and James K. Boyce, in an excerpt from their new book.
In March 2010, the African Union and the UN Economic Commission for Africa jointly convened the annual Conference of African Ministers of Finance in Lilongwe, Malawi, around the theme of 'Promoting high level sustainable growth to reduce unemployment in Africa'. The agenda included a high level session on 'the phenomenon of illicit financial flows from Africa and its devastating impact on development prospects'.  This reflected increasing recognition that capital flight poses a major development challenge for African countries. The issue is at the heart of discussions of development finance, transparency in public resource management, and the sustainability of external borrowing.
The magnitude of African capital flight is staggering both in absolute monetary values and relative to GDP. For the thirty-three sub-Saharan African countries for which we have data, we find that more than $700 billion fled the continent between 1970 and 2008. If this capital was invested abroad and earned interest at the going market rates, the accumulated capital loss for these countries over the thirty-nine-year period was $944 billion. By comparison, total GDP for all of sub-Saharan Africa in 2008 stood at $997 billion.  Comparisons to Asia and Latin America have found that capital flight from Africa is smaller in sheer dollar terms, but larger relative to the size of the African economy. 
READING THE HIDDEN BALANCE OF PAYMENTS
Measurement of capital flight poses daunting challenges, and requires some rather sophisticated statistical detective work. Funds that are acquired illegally, or funnelled abroad illegally, or both, are not entered into the official accounts of African countries. At the same time, the perpetrators of capital flight benefit from the complicity of bankers and other operators who assist in the placement of the funds in foreign havens. The identities of asset holders are often concealed through proxies and by taking advantage of legal screens available in bank secrecy jurisdictions. Nevertheless, researchers have made substantial progress in developing ways to estimate the magnitude of capital flight. This section reviews the methods used in this book. 
RESIDUAL MEASURES OF CAPITAL FLIGHT
Our starting point is the balance of payments (BoP), each country's official record of inflows and outflows of foreign exchange. These data are compiled annually by the IMF on the basis of reports from the central banks of its member governments. The 'current account' of the BoP records international flows arising from trade in goods and services, interest payments and transfers - transactions that do not lead to future claims on resources. The 'capital account' records flows of loans, investments and other financial transactions that entail future claims. Outflows of foreign exchange to the rest of the world, such as debt service or payments for imports, are recorded as debits (denoted by a negative sign). Inflows, such as loan disbursements or payments for exports, are recorded as credits (with a positive sign).  The net sum of the current account and the capital account gives the country's overall BoP position, which in principle corresponds to the net change in the country's official reserves of foreign exchange. A BoP surplus, when foreign exchange inflows exceed outflows, translates into a gain in international reserves. A BoP deficit, when outflows exceed inflows, translates into a loss of reserves. In practice, recorded inflows and outflows of foreign exchange seldom match exactly the changes in the country's official foreign exchange reserves. The missing money, or residual, is labelled 'net errors and omissions' in the BoP.
In the wake of the 1983 Third World debt crisis, it was discovered that the inflows of foreign borrowing recorded in the official BoP were often understated by substantial amounts. As a result, the total stock of external debt, built up over years of borrowing, often exceeded the cumulative borrowing as reported in the BoP. The World Bank independently assembles annual data on the stock of debt. This information, contained in a World Bank database called Global Development Finance (GDF), provides the basis for corrections to the BoP figures. By taking GDF data on changes in debt stocks, substituting this for the BoP data on foreign borrowing, and recalculating net errors and omissions, we can obtain a new residual estimate of missing money. The World Bank (1985) and others pioneered this technique to derive a measure of capital flight. 
TRADE MISINVOICING AS A SOURCE OF CAPITAL FLIGHT
In addition to incomplete recording of debt inflows, another well-known source of errors in the official BoP accounts is trade misinvoicing. This can take several forms. Both importers and exporters may manipulate the reported values of their transactions in order to conceal foreign exchange transactions from the country's monetary authorities. In the case of exporters, under-invoicing (by falsely underreporting the quantity of goods exported and/or the price received) evades tax liabilities and reduces the amount of foreign exchange that must be surrendered to the authorities from export receipts. In the case of importers, over-invoicing (by inflating the quantity and/or price of imports) increases the amount of foreign exchange they can obtain on favourable terms from the central bank to pay for imports. Both export under-invoicing and import over-invoicing are important mechanisms for capital flight. When exporters understate the value of their export revenues, they often retain abroad the difference between the true value and the declared value. Similarly, when importers send extra foreign exchange abroad, ostensibly to pay for imports, the excess (minus a commission for their partners) is often deposited in a designated foreign bank account.
The extent of trade misinvoicing can be estimated by comparing the export and import data provided by an African country to the corresponding import and export data of its trading partners. Both sets of figures are reported in another annual IMF publication, Direction of Trade Statistics. If we assume that the trade data provided to the IMF by the industrialized countries are relatively accurate, the discrepancy between these figures and the data from their African trading partners yields a measure of trade misinvoicing.
DISCREPANCIES IN WORKERS' REMITTANCES
One more item in the BoP statistics that can be an important source of error is workers' remittances. Over the past few decades, many African countries have recorded large and increasing inflows of remittances from their citizens who are working in other African countries, Europe and, to a lesser extent, the United States and other industrialized countries. In some African countries, remittances are now larger than conventional external financing from aid or foreign direct investment. However, a substantial fraction of remittance inflows is transferred through informal channels that escape recording in official BoP statistics. The World Bank estimates that unrecorded remittances in African countries account for more than half of total remittance inflows. 
Adjusting for remittance discrepancies is important for accurate measurement of capital flight, as the unrecorded inflows increase the amount of foreign exchange that is available to the country. The effect of unrecorded remittances thus is similar to that of unreported export earnings: the amount of foreign exchange actually entering the African country is greater than what is captured in the official BoP.
The International Fund for Agricultural Development (IFAD) has computed alternative measures of workers' remittance inflows by using survey data.  The IFAD estimates were derived by combining data on total numbers and locations of migrant workers in 2006 with survey data for various host origin country pairs on the percentage of migrants who send remittances and the average amount sent. The results indicate that the true magnitude of remittance inflows to Africa is substantially underestimated in the BoP data. For example, IFAD estimates that remittance inflows from industrialized countries to Nigeria in 2006 amounted to $5.4 billion, compared to $3.3 billion reported in the official BoP statistics. For Angola, the BoP reports no remittances whatsoever in that year, whereas the IFAD estimate shows an inflow of $969 million.
ADJUSTMENTS FOR INFLATION AND INTEREST EARNINGS
To obtain measures for the period from 1970 to 2008, the final step is to convert the annual flows into figures that are comparable across different years, since a dollar outflow in 1970 is not the same as a dollar outflow in 2008.
COUNTING THE MISSING MONEY
Using the method described above, we estimated the amount of capital flight from thirty-three sub-Saharan African countries for which adequate data are available for most years.  The numbers are eye opening. Total capital flight from these countries over the 1970-2008 period (in 2008 US dollars) amounted to $735 billion. This is equivalent to roughly 80 per cent of the combined GDP of these countries in 2008.
If the funds that left African countries during this period were invested in assets that earned the interest rate on short-term US treasury bills, the cumulative stock of flight capital with imputed interest earnings in 2008 would amount to $944 billion. In practice, of course, the fate of the missing money in most cases is unknown. Undoubtedly some of it was not invested, but instead was dissipated in Parisian shopping sprees and other consumption. On the other hand, some may have yielded returns above the fairly conservative US treasury bill rate. Whatever the rate of return that accrued on average to African flight capital, its cumulative stock with imputed interest earnings is a reasonable indicator of the opportunity cost of the failure to invest these funds productively in Africa. It also provides the most appropriate measure for comparison to Africa's external debts, since these accrue interest regardless of how the borrowed money was used.
Our $944 billion estimate of the cumulative stock of African flight capital closely matches the total wealth of Africa's high net worth individuals (HNWIs) as reported in World Wealth Report, an annual publication of the financial services firms Capgemini and Merrill Lynch Global Wealth Management which tracks the holdings of HNWIs around the globe. The report defines HNWIs as people with investable personal assets of $1 million or more. The total wealth of Africa's HNWIs peaked, according to this source, at $1 trillion in 2007 before slipping to $800 billion in 2008 as a result of the global financial crisis. 
Table 2.2 African capital flight: the top ten [table abridged to show only total real capital flight, in $ billion">
Source: Authors' computations
Focusing on the dollar amount of capital flight may give a misleading sense of relative burdens, since in smaller economies even a modest outflow could represent a substantial drain. For example, total capital flight over the period was equivalent to 614 per cent of the 2008 GDP for São Tomé and Príncipe, 493 per cent for Seychelles, 384 per cent for Burundi, and 312 per cent for Sierra Leone. By this measure the burden of capital flight was substantial for a number of large economies, too: 807 per cent of 2008 GDP for Zimbabwe, 265 per cent for the Democratic Republic of Congo, 223 per cent for the Republic of Congo, and 194 per cent for Côte d'Ivoire.
AFRICA AS A NET CREDITOR
It is now time to balance the books. How much net financial wealth does Africa have, given its external assets and liabilities? To answer this question, we compare external assets, as measured by the cumulative stock of capital flight, to external debt. The assets accumulated by means of capital flight are private, while the external debts are public liabilities owed to the creditors by the people of Africa through 'their' governments.
Not all of the capital that fled sub-Saharan Africa can be presumed to have been saved and invested so as to earn normal rates of return. As we have noted, some of the money was spent on consumption, and some savings may have earned sub-normal rates of return. Our measure of cumulative capital flight, including interest earnings, therefore does not exactly equal the external assets held by private Africans today. We nonetheless believe that a comparison between the stock of capital flight and the external debt can provide a reasonable indicator of Africa's net wealth. By this measure, sub-Saharan Africa is a net creditor to the rest of the world by a substantial margin. The cumulative stock of capital flight from the thirty-three countries covered in this book stood at $944 billion in 2008, compared to external debts of $177 billion. By this measure, these countries had positive net external assets to the tune of $767 billion (see Table 2.2). In other words, the rest of the world owes more to these African countries than they owe to the rest of the world. This suggests that Africa could expunge its entire stock of foreign debt if it could recover only a fraction of the wealth held by Africans in foreign financial centres around the world.
Many millions of Africans are desperately poor. But the continent is rich. According to the World Wealth Report, the continent had roughly 100,000 high net worth individuals in 2008, twice as many as a decade before. Of these, about 1,800 were 'ultra-high net worth individuals', with at least $30 million each in investable assets.  Together these rich Africans held about $800 billion in investable assets in 2008.
Compared to other regions, African private wealth holders exhibit a stronger preference for foreign assets as opposed to domestic assets. According to a study by researchers at the World Bank and IMF, an astonishing 40 per cent of Africa's total private wealth was held abroad as flight capital in 1990. The corresponding figure for South Asia was 5 per cent. For East Asia it was 6 per cent, and for Latin America 10 per cent. Sub-Saharan Africa and South Asia had similar levels of total private wealth per worker, but in sub-Saharan Africa capital flight amounted to $696 per worker whereas in South Asia it was only $90 per worker. As a result, private domestic capital per worker in Africa was less than 60 per cent of what it was in South Asia. 
High net worth individuals typically have more internationally diversified portfolios than their poorer countrymen. According to the World Wealth Report, high net worth individuals in the Asia-Pacific region hold 32 per cent of their assets abroad and those in Latin America hold 55 per cent abroad, percentages roughly five times higher than the overall averages for these regions reported by the World Bank and the IMF.  If the same pattern holds in Africa, this would suggest that the greater part of the wealth of high net worth Africans is invested abroad. In this respect, the ultra-rich of Africa today are unlike the robber barons of years gone by in the industrialized countries, who whatever their misdeeds at least did invest in their nations' economies.
The preference for foreign assets and aversion to domestic investment comes at a high opportunity cost to African economies. In the case of legally acquired assets, the continent is deprived of the gains that would accrue from investment at home, not only losing income and jobs, but also forgoing government revenue that could fund public services. In the case of illegally acquired assets, African countries lose twice: first, they are robbed through fraud and embezzlement; then they are further deprived of any benefits that would trickle down if the loot were invested at home.
BLEEDING A CONTINENT: THE COSTS OF CAPITAL FLIGHT
Africa is bleeding money, as capital flows into the private accounts of African elites and their accomplices in Western financial centres. At the same time, the continent is in dire need of financing. For Africa to overcome widespread and extreme poverty, it needs sustained and sustainable economic growth. This will require very large increases in the levels of domestic investment, especially in infrastructure. 
Researchers and development institutions have invested considerable time and energy to prove that African countries need more resources to meet their infrastructure financing needs. The 2009 Africa Infrastructure Country Diagnostic report concluded that Africa's middle-income countries need investment of about 10 per cent of GDP per year in infrastructure alone. [28"> Investment needs for low-income African countries are higher at about 15 per cent of GDP annually. To achieve these levels, the continent's investment would need to be scaled up by at least $100 billion per year to nearly double the current level.
To get a first-hand sense of the immensity of the problem, one need only experience any of the cities in sub-Saharan Africa. In July 2009, one of the authors of this book, Léonce Ndikumana, visited Freetown, the capital of Sierra Leone, to attend a meeting of the African Caucus of Finance Ministers and Central Bank Governors. Although this was his first trip to Sierra Leone, he did not expect any surprises; after all, he was going to an African country, and he had been in quite a number of them. But the visit to Freetown turned out to be an opportunity to experience something new: a city that is effectively cut off from its own airport. To get from the airport to the city, there are three options: a very long road trip around the bay that separates the two; a ferry that is subject to long waits and the risk of capsizing due to overloading; or a seven-minute ride in a helicopter, which is most convenient (for those who can afford it) but also risky, since these have been known to drop in bad weather with no survivors. On this occasion, the helicopter made the journey safely. The cost of building a bridge to link the city to the airport has been estimated at $400 million, a modest fraction of the $6 billion in capital flight that has left Sierra Leone since 1970. [29"> Despite the economic benefits that such a bridge would bring to the country and the region, the government has not been able to mobilize the necessary money.
Sierra Leone is by no means alone in its dire lack of basic infrastructure. In fact, apart from the helicopter, Ndikumana's experience in Sierra Leone was little different from what he encounters in his native country, Burundi, when he visits his commune of Vugizo in the south. The commune has the agricultural potential to feed the towns and cities in the province and beyond, but it is landlocked and has poor access to markets. During the rainy season, it can take two hours to drive the 40-kilometre stretch of dirt track linking it to the nearest paved road. [30">
The Millennium Development Goal (MDG) of halving extreme poverty by 2015 remains elusive for much of Africa. The MDG Africa Steering Group estimated in 2008 that for Africa to achieve this and related development goals, public external financing would have to increase by $72 billion per year in the medium term. [31"> Were Africa able to recoup only a fraction of what it has lost in capital flight, this would go a long way towards filling this gap. The United Nations Economic Commission on Africa estimated in 1999 that an investment/GDP ratio of 34 per cent would be required to achieve a 7 per cent GDP growth rate in Africa, which would cut poverty by half by 2015. [32"> This investment target is in reach for African countries if they can manage to stem capital flight and recoup some of the money stolen in the past. [33"> Otherwise, efforts to mobilize additional development financing for growth and poverty reduction will yield only limited results.
CAPITAL FLIGHT AND TAX REVENUE
Sub-Saharan African governments badly need tax revenue to bridge the large deficits in the provision of public goods, including not only infrastructure but also health and education. [34"> Some resource-rich countries have seen revenue gains thanks to natural resource booms, but these may prove to be transient. Meanwhile, very few non-resource-rich African countries have recorded sustained increases in revenue. [35">
Countries with higher capital flight tend to have lower tax revenue, as can be seen in Figure 2.3. There are two reasons for this negative relationship. First, capital flight directly erodes the tax base by subtracting from it private wealth and income earnings on that wealth. Second, high capital flight is symptomatic of an environment characterized by corruption and weak regulation, circumstances that both promote capital flight and undermine tax administration. (In contrast, if capital flight were motivated primarily by a desire to escape high taxes, one would expect the opposite correlation: countries with less tax revenue would tend to have less capital flight.)
If we look at the 'tax effort' - the ratio of the actual tax revenue to the potential revenue based on the country's economic structure and level of development - we find that actual tax performance in sub-Saharan Africa generally remains well below potential, and that resource-rich countries tend to perform even worse in this respect than resource-scarce countries. [36"> In the case of Nigeria, for example, when oil rents are excluded, the tax effort index is 0.44, meaning that the country is generating only 44 per cent of its potential tax revenue from non-oil sectors. In Angola, the corresponding index is 0.39. Natural resource revenues are often poorly mobilized, too. In the Democratic Republic of Congo, for example, it is reported that gold exports can reach up to one billion dollars a year, but these exports generate a negligible $37,000 in tax revenue. [37">
Rampant tax exemptions contribute to low revenues. Often exemptions are awarded not on the basis of the criteria set by the law - which typically aim to stimulate private economic activity, for example by means of tax incentives - but rather on the basis of the political influence of individuals and firms. As a result, tax revenue may not follow the expansion of private sector activity and private wealth. A case study on Ethiopia, where resource inflows to the private sector are increasing but the proceeds from corporate taxation are declining, estimates that the revenue forgone through exemptions doubled between 2005 and 2007. [38"> At the same time, Ethiopia has a relatively high nominal income tax rate, which may contribute to greater tax fraud. Taxes that are high in theory thus can be low in practice, owing to both legal exemptions and illegal evasion.
Capital flight has substantial adverse distributional effects, too, exacerbating gaps between rich and poor. The rich, by virtue of the fact that they hold a larger share of their assets abroad, are shielded from the wealth effects of devaluation of the national currency. Indeed, they may benefit from devaluation, as this allows them to reap windfall gains if they bring some of their capital back into the country. Since capital flight itself puts pressure on the exchange rate, it increases the likelihood of this exchange rate effect.
At the same time, by depressing government tax revenue, capital flight adversely affects the poorer segments of the population who depend most heavily on publicly funded services. For example, when the government is unable to provide adequate medical supplies and qualified health personnel for public hospitals, the poor who cannot afford the alternative of going to private clinics suffer the most. The same goes for education.
BROUGHT TO YOU BY PAMBAZUKA NEWS
* Excerpted with permission of the authors from: ‘Africa's Odious Debts: How Foreign Loans and Capital Flight Bled a Continent’, by Léonce Ndikumana and James K. Boyce. Zed Books, 2011.
* This excerpt was first published by AfricaFocus Bulletin, an independent electronic publication providing reposted commentary and analysis on African issues, with a particular focus on US and international policies. AfricaFocus Bulletin is edited by William Minter.
* Please send comments to editor[at]pambazuka[dot]org or comment online at Pambazuka News.