Pete’s blog

Abstract

Africa is the poorest and most indebted continent in the world. Many explanations and solutions have been given for this persistent poverty, but one factor that has received very little attention is the nature and structure of Africa’s debt. The majority of Africa’s debt is in foreign currency and is not tradeable, and very few countries in Africa have a viable bond market. Having a liquid local bond market and the majority of debt denominated in local currency provides significant benefits, whilst its absence can create a spiral of poverty and indebtedness. The potential benefits include better fiscal planning and discipline; a reduced drain on foreign exchange reserves; easier implementation of export orientated policies; increased local participation in debt markets; increased investment in small to medium enterprises; more effective money supply control; improved financial information; more effective market corrective forces on over-indebtedness; less leakage from the savings-investment cycle; and the development of other financial markets.

Given all the benefits, creating these debt markets should be made a priority. Since Africa is already at the limit of its borrowing capacity, the change will require the involvement of the lenders as well as the African countries. The World Bank, as one of the largest single lenders to Africa, should play a central role in this change and should convert a large portion of its debt holding to tradeable bonds in order to provide capital for these new markets. The upfront challenges and costs of implementing bond markets may be significant, but are far outweighed by the long-term benefits. Now that the major debt relief programmes have been finalised, efforts should be concentrated in this direction to make Africa’s debt sustainable and manageable.

Bond Markets: The Key to Africa’s Financial Independence?

A bond market is one of the most fundamental financial markets – textbooks normally take it for granted that governments borrow money by issuing government or sovereign bonds and that these bonds are freely traded; basic economic theory on monetary supply assumes that a bond market exists; risk pricing and investment valuation models rely on data from bond markets. Yet for most African countries the bond market is insignificant or non-existent, even though Africa has some of the most heavily indebted countries in the world. The lack of a functioning bond market could be a fundamental contributor to these African countries’ persistent poverty and indebtedness – preventing proper debt management systems from forming, blocking the functioning of corrective market forces, and slowing investment flows and the formation of more complex financial markets. After several major debt relief initiatives, many countries are left with large but ‘affordable’ debt stocks at highly concessional rates. It would be difficult, if not impossible, for these countries to be able to issue further debt at market related prices in order to create a functioning bond market. As most of the debt of these countries is concentrated in multilateral debt with the World Bank/ International Development Association (IDA), the World Bank would be an integral part in any initiative to create a bond market. The World Bank should pursue a programme to convert a large portion of its debt with these African countries into tradeable bonds and assist the countries in creating a market in which these bonds can be traded.

The size and nature of Africa’s debt

Africa’s ratio of external debt to GDP in 1998 was the highest in the world at 65.5%, with approximately 6% of GDP being used annually to service this debt. This however, was the picture for Africa as a whole – the picture was more alarming when only countries with a severe debt problem were examined. Of the 38 countries currently identified in the Heavily Indebted Poor Countries (HIPC) debt relief initiative, 32 are in Africa. For these HIPC countries, their ratio of external debt to GDP in 1998 was 103%. The debt relief initiatives have had a major impact on these ratios, reducing the external debt to GDP ratio of HIPC countries to 69.7% in 2005 with further reductions in 2006 owing to the implementation of the Multilateral Debt Relief Initiative (MDRI). However, this is still far above the 30% average for developing countries[1].

In terms of debt structure, South Africa is the one major exception in Sub-Saharan Africa. South Africa has a very well-developed local bond market and almost its entire debt is held by private creditors, mostly in the form of bonds. For the rest of Sub-Saharan Africa, excluding South Africa, the debt is split largely between multilateral (45%) and bilateral (43%) debt with the minority as private (12%). In 2005, the IDA (the arm of the World Bank dealing with poor countries) held 30% of the total long-term outstanding debt of these countries[2]. In theory only the privately held debt is tradeable, i.e. less than 12% of Sub-Saharan Africa’s debt. In the context of the global market, Africa is insignificant. In 2001, Africa as a whole only accounted for 7% of the total tradeable debt of emerging markets or less than 0.4% of the total world bond market. Of Africa’s total tradeable debt, Sub-Saharan Africa, excluding South Africa, only accounted for 15%[3].

Loans to Africa have furthermore become highly concessional, with low rates and long maturities. The average interest rates payable on new loans to Sub-Saharan Africa in 2004 was only 2%, compared to the average of 4.4% for all developing countries. The average time to maturity was 22.1 years compared to 14.8 years for all developing countries[4].

Characteristic and benefits of a bond market

Firstly, as bonds can be issued in various forms in various markets, it is important to define the characteristics of the bonds and market that are required. The bonds themselves need to have a variety of term lengths and be fully transferable. The majority of the bonds should be denominated in the local currency of the country. The market should be very accessible locally and have a large percentage of local trade. New bond issues should be made via auction or similar mechanism, allowing the bonds to trade at a market related price. With these characteristics, the following benefits can be obtained:

1. Foreign versus local currency

Most multilateral debt is denominated in a foreign currency; in particular IDA debt is denominated in Special Drawing Rights (SDR), a currency used by the International Financial Institutions that is derived from a basket of international currencies. Having debt in a foreign currency adds an extra dimension of volatility to debt management and can create a host of problems. Exchange rate movements were cited as one of the three main contributing factors towards increasing debt ratios in an assessment of the debt sustainability of post-HIPC countries[5]. The biggest advantage of having debt in local currency is that it allows the government to prepare a long term fiscal budget with greater confidence. The government would know exactly how much debt it has and how much the debt repayments will be in local currency, without having to estimate exchange rate movements. A carefully planned budget cannot be rendered useless by a sudden exchange rate fluctuation. Under these conditions, governments would be able to learn and implement effective fiscal planning and discipline far easier, thereby improving long-term debt sustainability and economic growth. Having a stable, reliable, long-term fiscal plan will increase investor confidence in the government and the country.

Another problem with debt being denominated in foreign currency is that the drain on the country’s exchange reserves is unaffected by exchange rate movements – a $1 debt service remains $1 even if the currency severely weakens. This creates a conflict of interest within the government’s economic planning – a weak exchange rate can spur future export growth and therefore economic growth; however, it will mean that debt service payments will become immediately more expensive in local currency terms. Faced with the choice between potentially having greater economic growth in the future or having cheaper debt payments now, the safer choice would be to keep the currency strong and have the cheaper payments. If debt is dominated in local currency, the government is freed from this dilemma, allowing them to pursue export orientated policies.

Finally, having debt in a foreign currency effectively excludes local participation in the debt market. The local investment would be a complicated two-step investment, with both a foreign currency and a debt component.

2. Risk management and investment

Government bonds are considered the lowest risk instruments in capital markets. In the absence of tradeable government bonds, investors are forced to invest in higher risk instruments. Investors have a certain ‘risk appetite’ when it come to how they structure their investment portfolio. If an investor is looking for a medium risk exposure, they can either invest solely in medium risk instruments or alternatively invest in a spread of investments from low to high risk that on average gives them a medium exposure to risk. If an investor is unable to invest in low risk government bonds and instead has to invest in medium risk instruments, they will not be able to invest in any high risk investments. By providing a market for low risk government bonds, capital is effectively freed up for investment in high risk investments. What are high risk investments? They are small to medium enterprises, entrepreneurs and start up companies – the drivers of economic growth. In effect, the absence of government bonds has been stifling growth in this critical economic sector.

Similarly, capital is being tied up in banks. Banks have what is known as ‘capital adequacy requirements’, whereby they have to maintain a certain amount of capital to provide security against deposits that they take. Without low risk investments, banks are forced to maintain a greater amount of capital in reserve. Providing government bonds as an investment option for banks would also free up capital, increasing the bank’s overall loans and investments.

In addition, having a bond market provides space for the operation of enterprises with specifically lower risk appetites – in particular pension funds.

3. Monetary policy

In the current situation, many African governments have access to loans at highly concessional rates – if certain conditions are met, they can even access IDA loans interest free. Although this is arguably highly beneficial, the negative side effect is that it becomes very difficult to control money supply. The method favoured by most developed countries, that of using interest rates to control inflation, is limited in its effectiveness. Any interest rate increases in the country would have no impact on the biggest spender – the government. No matter how high local interest rates rise, government will continue to have access to debt at these concessional rates. If interest rates are used to control inflation, far greater adjustments would be required to have an impact on money supply. Over time, persistently high inflation and interest rates will result in an emaciated private sector and an over-inflated government sector. Conversely, if new government debt had to be issued at market related prices, local interest rate increases would also increase the cost of government debt, thereby reducing government spending. The burden of interest rate targeting of inflation would be equitably shared between the public and private sector and therefore the average cost of borrowing for the private sector would be lower.

4. Information

An established bond market can provide key investment information, including interest rate expectations and inflation rate expectations. The yield on government bonds is often used to estimate the risk free rate of return of a country. This risk free rate is in turn used to value other investments and to calculate country and sovereign risk rates. Having this information available allows investors to better value potential investments and to better predict future market movements, making the markets more accessible and attractive to investors.

This information can also be used in fiscal and monetary planning. Knowing what the market expects interest rates to be in the future allows the government to better estimate the future cost of borrowing. Also knowing market expectations, allows the monetary supply regulator to have a better idea of how the market will react to interest rate adjustments or how concerned the market is about inflationary pressures.

5. Market forces

The fact that African countries were able to become indebted to the extent that the possibility of repayment was remote indicates a severe market failure. No market mechanism kicked in to prevent or correct this severe over-borrowing. The first corrective mechanism that was bypassed was that of inflation and exchange rate adjustments. In theory a government that overspends and therefore over-borrows should cause their currency to devalue as money supply outstrips demand. As the currency devalues, the exchange rate against other currencies weakens. If the debt is denominated in local currency, the devaluing of the currency will decrease the relative cost of the debt until it is once again affordable. A government attempting to over-borrow would therefore be faced with high inflation and economic decline and would therefore in theory be restrained. However, if the majority of debt is denominated in foreign currency, this mechanism cannot function. A government that over-borrows a foreign currency would be faced with an increasing relative cost of debt (as the currency devalues) and a shortage of foreign currency – forcing them to borrow more foreign currency.

This currency devaluation mechanism also functions as a restraint on foreign lenders. If a foreign lender lends too much to the government, they run the risk of causing the country’s currency to weaken and thereby reduce the value of the loan in the lender’s currency. The risks and costs of over-borrowing are shared between the government and the lenders. On the other hand, if the loan is denominated in the lender’s currency, the value of the loan remains constant for the lender even if they over-lend. Since the lender carries none of the costs of excessive lending, they are willing to lend far beyond sustainable levels.

The second market mechanism that should restrain over-borrowing is that of increasing costs of borrowing. In a normal situation, as the government begins over-borrowing, the cost of the debt should increase as lenders demand a greater return for the greater risk. The supply of debt will also diminish as lenders are more reluctant to lend money. Eventually either the debt will become unaffordable to the government or the lenders will refuse to lend any further money.

Two factors prevented this from happening in many African countries. Firstly, the governments were able to source debt in foreign currency from international markets and were able to negotiate consistently lower rates. By denominating the debt in foreign currency, the risk and costs of currency depreciation was effectively removed from the negotiated rate. If the debt was in local currency, the foreign lender would demand a higher rate in order to compensate for the risk of currency devaluation. Once the multilateral institutions became involved, the governments were able to negotiate even lower concessional rates. Secondly, as Arslanalp and Henry note in a review of the Brady Plan, Africa never suffered from a shortage of capital flows or negative net resource transfers[6] – the supply of debt never ran out.

6. Debt/investment cycle

The last benefit is gained if there is a large percentage of local investment and trade in government bonds. A loan in foreign currency results in a large inflow of capital at initiation, followed by a steady outflow of repayments until the loan is repaid. The outflows will generally exceed the inflows for a particular loan, therefore if looked at independently a foreign loan will ultimately result in a net capital outflow from the country. The money borrowed must therefore be very wisely spent so as to create more capital than this outflow. Even if the entire amount of the loan is used to buy fixed capital assets, the assets would have to be able to generate more revenue than the interest costs on the loan, otherwise the overall result would still be a reduction of capital. This makes a standard Keynesian fiscal strategy unusable – normally any general increase in government spending will have a beneficial impact on economic growth, especially if the economy is not at full employment. But in this situation, only government spending that results in economic growth will result in economic growth – somewhat of a tautology.

If, instead, money is borrowed from local investors, the capital remains within the country and there is no loss of capital. Interest on debt increases the revenues of the local investors, which in turn can increase the tax revenues of government. Increasing government spending can get money flowing more freely in the economy, promoting trade, investment and economic growth.

Changing the current system

If a bond market does convey such important benefits, why have they not been created? There are many possible explanations why African countries initially chose foreign loans over local bonds – a lack of local capital, an abundance of apparently cheap foreign capital, the lower administration costs of foreign loans over issuing bonds – but once this decision was made, it became a self-reinforcing cycle. As the foreign loans drain local capital and foreign exchange reserves, more foreign capital and currency is required. As the economies stagnate for a variety of reasons, local capital becomes more scarce and more expensive, whilst foreign capital is readily available at cheaper rates. Any weakening of the currency reduces the relative size of local capital and increases the need for foreign exchange. A government eventually has no interest in issuing local bonds at very expensive rates, which will not provide vital foreign exchange. From the foreign lenders’ point of view, they would have no interest in buying local bonds carrying additional currency risk, when they are able to issue debt in the currency of their choosing. As the economic situation declines in the country, the lenders will insist on having their loans denominated in foreign currency so as to limit their risk. Eventually the multilateral institutions are brought in and, in an attempt to help, offer the countries even more foreign debt at even lower rates. A bond issued in local currency at market prices stands little chance of competing against these cheap and relatively freely available loans.

To break this cycle a concerted effort by both the borrowers and the lenders is required. All parties need to recognise that the current position is unsustainable and that a bond market will provide long-term benefits. In fact this shift in thinking is currently underway in the world debt markets. The HIPC and associated debt-relief programmes were implemented in recognition of the fact that the current debt burden was unsustainable and in very recent years, bonds have been creeping up the agenda internationally. The G8 is currently working on an action plan for developing local bond markets in emerging market economies and developing countries, a topic that was discussed at a meeting of the G8 Finance Ministers in Potsdam in May 2007[7]. The World Bank is currently assisting Ghana with its first issue of sovereign bonds on international markets[8], with Nigeria, Kenya and Zambia expected to follow[9]. Current thinking around bonds appears to be that once countries have stabilised their debt, implemented sound fiscal policies and shown some degree of financial stability, then any new debt requirements can be met through the issue of bonds on international markets. This is a big step forward, as new debt will now be in the form of bonds in the country’s local currency and there is strong interest in these bonds from both sides, even with their additional costs. The bonds will also be tradeable, thereby creating the seed of a market.

However, this falls well short of what is required for the full benefits to be realised. As these bonds will be issued on top of existing debt, the risk and therefore the cost of the debt will be higher. The market might be reluctant to participate fully in these bonds, knowing the size of the existing debt burden. The amount of bonds that can be issued will be small and it will take a long time for a significant amount of debt to be issued in the form of bonds. Furthermore the bonds are targeted at international investors and not local investors. The local market must not be ignored; it is critically important that local investors are able to and are encouraged to participate in bond issues. The full benefits of a bond market will not be realised without an actively traded local market. Finally, the belief that countries need to show fiscal discipline and economic stability before they are allowed to issue bonds is a “chicken-and-egg” problem. Given the benefits listed above, the creation of a bond market first will make fiscal discipline and economic stability a far easier task to achieve; however until the country can show fiscal discipline and economic stability, the cost of the bonds may be unaffordable as creditors demand higher rates for the greater risk. Once again the local market should not be ignored – bonds can be issued on the local market only. The local market will not demand a premium to compensate for foreign exchange risk and an affordable rate may be achieved.

What Africa needs is an ambitious plan involving the systematic conversion of existing debt stocks into tradeable bonds, the implementation of local bond markets and an education and promotional campaign encouraging locals to invest in these bonds. This may sound overly ambitious, yet the desire, the skills and the components are already in place. Firstly bonds are a very common financial instrument. There is a very wide body of knowledge on bonds and bond markets – everything from establishment to management of markets. Secondly, the debt relief programmes have resulted in a concentration of debt within the multilateral institutions, in particular the World Bank (IDA debt for Sub-Saharan Africa increased from 15% in 1995 to 27% in 2005[10]). Having one major creditor makes the conversion of debt a far easier task. It would take a strategy shift within the World Bank, but since the World Bank already deals with many countries that issue debt primarily in bonds (mostly through its International Bank for Reconstruction and Development (IBRD) arm), it will not require any new systems or processes. The World Bank will, however, have to be willing to accept upfront losses if the bonds are to be issued at viable rates. With the necessary political will, these losses could be accepted, especially considering the following factors: the World is already taking losses on the loans issued below market rates; the conversion to bonds will result in upfront cash flows into the Bank; and the losses should approximate a write-down to the actual net realisable present value of the loans.

In terms of the creation of bond markets, the G8 as part of its action plan has indicated an interest in providing technical assistance for the creation of local bond markets[11]. If a continent-wide programme is followed, a central trading system could be developed and implemented – thereby providing economies of scale and reducing costs. Such a system may even be sourced from the African continent itself – South Africa is interested in positioning itself as a “financial centre for Africa”, opening its markets to African listings and providing the infrastructure and market for capital investments[12]. South Africa already has a liquid bond market with a world-class trading system. The Bond Exchange of South Africa has also stated an interest in being involved in the development of bond exchanges in Africa[13].

Even with all this interest, there are still some significant initial challenges – getting the market size and trading volumes high enough to create a sufficiently liquid market; implementing measures to protect against speculative attacks; finding the balance between a market driven rate and an affordable rate; promoting local interest and investment in the bonds; implementing mechanisms to prevent countries reverting back to ‘cheaper’ foreign debt; and educating government, central banks and industry on the benefits, functioning and management of bond markets.

Conclusion

The establishment of local bond markets in African countries should be prioritised by the international community. A successful change in debt structure will have a profound impact on investment, economic growth and industry development. A viable bond market will free up capital for investment in small businesses; allow the government to pursue export orientated policies and to fuel economic growth through government spending; increase investment and money flows within the country; and support the development of other financial markets like stock exchanges and corporate bond markets. In short, it will increase the sources of investment for the local private sector. As many countries are still trapped by heavy debt burdens, a programme to implement bond markets would not be effective without the involvement of the international community. The question has often been asked – “what do we do after debt relief?” The momentum of the debt relief programmes should be channelled into this direction, changing the nature of the debt into one with greater long-term sustainability and benefits.

Bibliography

Arslanalp S, Henry PB, Debt Relief: What Do The Markets Think?, National Bureau Of Economic Research, http://www.nber.org/papers/w9369, December 2002

Debt Relief For The Poorest: An Evaluation Update of the HIPC Initiative, World Bank Independent Evaluation Group, Washington DC, 2006

Developing a Capital Market in Africa, Bond Exchange of South Africa, http://www.bondexchange.co.za/besa/action/media/downloadFile?media_fileid=4440, 17 October 2005

G8 Action Plan for Developing Local Bond Markets in Emerging Market Economies and Developing Countries, www.g7.utoronto.ca/finance/g8finance-bond.pdf, May 2007

Global Development Finance 2006: Summary and Country Tables, World Bank, Washington DC, 2006

Global Development Finance 2007: The Globalization of Corporate Finance in Developing Countries, World Bank, Washington DC, 2007

Kganyago L, Speech at “Reuters Economist Of The Year” Award Ceremony: “South Africa As A Financial Centre For Africa”, http://www.treasury.gov.za/speech/2004081101.pdf, 1 August 2004

Size & Structure of the World Bond Market: 2002, Merrill Lynch, http://info.worldbank.org/etools/docs/library/154716/domestic2003/pdf/MerrillLynch_Bondmkts.pdf, April 2002

World Economic Outlook 2006: Globalization and Inflation, International Monetary Fund, Washington DC, April 2006

Yeboah LA, World Bank Passes Vote of Confidence In Ghana Cedi, The Daily Graphic no 150146, Accra, 7 September 2007

---

 

[1] World Economic Outlook 2006: Globalization and Inflation, International Monetary Fund, Washington DC, April 2006, pg. 173 – 174 & 238 – 245

 

[2] Global Development Finance 2006: Summary and Country Tables, World Bank, Washington DC, 2006, pg. 21 & 358

 

[3] Size & Structure of the World Bond Market: 2002, Merrill Lynch, http://info.worldbank.org/etools/docs/library/154716/domestic2003/pdf/MerrillLynch_Bondmkts.pdf, April 2002, pg. 1 & 19

 

[4] Global Development Finance 2006: Summary and Country Tables, World Bank, Washington DC, 2006, pg. 4 & 22

 

[5] Debt Relief For The Poorest: An Evaluation Update of the HIPC Initiative, World Bank Independent Evaluation Group, Washington DC, 2006, pg. 19

 

[6] Arslanalp S, Henry PB, Debt Relief: What Do The Markets Think?, National Bureau Of Economic Research, http://www.nber.org/papers/w9369, December 2002, pg 24

 

[7] G8 Action Plan for Developing Local Bond Markets in Emerging Market Economies and Developing Countries, www.g7.utoronto.ca/finance/g8finance-bond.pdf, May 2007

 

[8] Yeboah LA, World Bank Passes Vote of Confidence In Ghana Cedi, The Daily Graphic no 150146, Accra, 7 September 2007, pg. 1, 3

 

[9] Global Development Finance 2007: The Globalization of Corporate Finance in Developing Countries, World Bank, Washington DC, 2007, pg. 59

 

[10] Global Development Finance 2006: Summary and Country Tables, World Bank, Washington DC, 2006, pg. 21

 

[11] G8 Action Plan for Developing Local Bond Markets in Emerging Market Economies and Developing Countries, www.g7.utoronto.ca/finance/g8finance-bond.pdf, May 2007

 

[12] Kganyago L, Speech at “Reuters Economist Of The Year” Award Ceremony: “South Africa As A Financial Centre For Africa”, http://www.treasury.gov.za/speech/2004081101.pdf, 11 August 2004

 

[13] Developing a Capital Market in Africa, Bond Exchange of South Africa, http://www.bondexchange.co.za/besa/action/media/downloadFile?media_fileid=4440, 17 October 2005, pg. 17