Abstract
Algeria’s local corporate bond market provides a good basis for further diversifying the economy’s financing channels. This study analyzes the nexus of bank financing of public enterprises in the ongoing national investment program (NIP) and also sets Algeria’s local corporate bond market in a cross-country perspective. It develops options for developing local capital markets during the timeframe of the NIP, and considers improvements in some aspects of market functioning. The authorities have taken measures to help the banking system increase lending to public enterprises over the next few years.
I. Domestic Capital Markets and the Financing of the National Investment Program1
A. Introduction and Summary
1. The massive national investment program may have a significant impact on financial intermediation in Algeria. Public enterprises will implement some industrial projects under the national investment program (NIP). Though minor in relation to the public investments financed by the budget, these projects will nevertheless require substantial new credit from Algeria’s small banking system. Whether this new credit is provided by banks or by local capital markets will influence the capacity of Algeria’s banking system to finance private sector activity.
2. The fact that public enterprises need significant new financing gives Algeria an opportunity to develop local capital markets. This means a key role for banks as catalysts for expanding corporate debt markets. Large, transparent borrowers such as public entities typically put capital markets and banks in competition for their debt finance. This process pushes banks toward earning fee income from helping large clients issue bonds, while focusing pure lending on smaller borrowers. Modern banks retain a comparative advantage over capital markets in processing opaque financial information typical of small and medium enterprises (SME) borrowers.
3. Algeria’s local corporate bond market provides a good basis for further diversifying the economy’s financing channels. The government has pursued deliberate policies aimed at developing this market. A modern government debt market was launched in 2002, with maturities now extending to 15 years. The authorities then encouraged large public enterprises to issue corporate bonds. The Algerian corporate market is significantly larger than in other countries at the periphery of the EU-15. Meanwhile, the authorities have launched several initiatives to improve the governance of banks, especially as regards their large credit exposure to public enterprises.
4. Benefits from further developing Algeria’s local capital markets include:
giving banks more time to upgrade their credit practices and risk management;
reducing the banking system’s exposure to public enterprises; and
spurring banks to develop new sources of income through competition with capital markets. These sources include bringing large issuers to the market, managing mutual funds, and developing SME loans.
5. The further development of Algeria’s capital markets involves:
encouraging large public entities to finance NIP-related projects by issuing bonds and further extending bond maturities;
welcoming issuance by firms incorporated in Algeria and owned by foreign interests;
launching the commercial paper market;
maintaining a public debt issuance program that ensures a reliable government yield curve as a benchmark for corporate debt; and
evaluating ways of boosting secondary trading for government debt, including through foreign participation in the market.
6. The paper is organized as follows. Section II analyzes the nexus of bank financing of public enterprises in the ongoing NIP. Section III sets Algeria’s local corporate bond market in a cross-country perspective. Section IV develops options for developing local capital markets during the timeframe of the NIP. Section V considers improvements in some aspects of market functioning.
B. Bank Financing of Public Enterprises in the NIP
7. Algeria has embarked on a massive national investment program. The NIP has two components: (a) the large public investment program financed directly from budgetary resources (115 percent of 2007 GDP); and (b) a small one (5 percent of GDP) undertaken by public enterprises in power, water desalination, and transport systems. Budgetary lending to public enterprises will remain minimal; the government has also ruled out foreign financing. Thus, the banking system is expected to increase credit by more than three quarter over the next few years to public enterprises for NIP-related projects.
8. Lack of capital and stricter enforcement of prudential rules constrain banks’ ability to increase credit at the pace required by the NIP. Foreign banks established in Algeria had no lending to public enterprises as of end-2006 and are too small to extend significant new credit. The public banks, which represent 90 percent of system assets, also have a small capital base relative to the need for new credit because of the prudential rule on risk concentration to public enterprises. The banking system also lacks long-term funding to match the maturities of credit that certain projects require.
9. The authorities have taken measures to help the banking system increase lending to public enterprises over the next few years. The measures help banks to (a) comply with prudential rules on risk concentration; (b) mitigate maturity mismatches; (c) boost lending capacity through higher capital; (d) marshal enough liquidity for the upcoming large disbursements; and (e) sustain credit to SME (Box I.1).
Measures to Facilitate Bank Credit during the Time of the NIP1
To comply with prudential rules on risk concentration, the treasury provides a guarantee for exposure to a single public enterprise in excess of 15 percent of bank capital. The treasury estimated that the guarantee would cover 90 percent of new NIP-related credit, the rest being the banks’ own risk. The banks used 20 percent of the guarantees in 2006. The risk concentration regulation has also been amended to exclude special purpose corporations from group consolidation by using project financing techniques.
To comply with rules on maturity mismatch, the treasury assumes maturity risk through the following facility: banks deposit funds with the treasury overnight and it extends them 17-year loans, with a two-year grace period. Banks using the facility can charge up to 3.75 percent interest on long-term credits to public enterprises. The facility amounts to one third of the new credit required by public enterprises under the NIP.
To increase the banks’ lending capacity, the treasury plans to provide additional tier 2 capital for public banks, which would allow them to increase credit by 25 percent. In addition, (a) remaining loans to public enterprises that were closed in the 1990s were swapped out; (b) the minimum capital of banks was raised to DA 2.5 billion; (c) two foreign banks have obtained licenses; and (d) two foreign banks already established received approval for major expansions.
To ensure liquidity, the treasury has encouraged syndicates, and also plans to reimburse ahead of time some portion of the stock of government securities held by public banks as a result of NPL buybacks.
To sustain the banks’ lending to SME, a large guarantee fund (CGCI-PME) has been created. The new fund can guarantee DA 360 billion of SME investment credit. This represents 35 percent of total credit to the private sector, but 100 percent when credit unrelated to SME investments is subtracted, e.g., consumer loans, mortgages, crop financing, or working capital credit. CGCI-PME is owned 60 percent by the treasury, 40 percent by banks. The precursor of CGCI-PME is FGAR, which was created in 2000 and has since guaranteed less than 2 percent of credit to the private sector.
1 Information from Algerian Treasury as of November 2007.10. The authorities have also started to restructure public enterprises to boost their ability to qualify for more bank credit (Box I.2). As was the case in the earlier round of 1997–98, the new round aims to restore key financial ratios of viable public enterprises to levels that conform to standard lending criteria. The authorities expect banks to give more credit on the basis of these new ratios.
11. These measures have already translated into strong bank credit to public enterprises. Between July 2005 and June 2006, only 14 percent of bank credit flows went to public enterprises. This rose to 64 percent in the period between July and December 2006.2
Financial Restructuring of Public Enterprises1
Algeria undertook a major financial restructuring of public enterprises in 1997–98.2 However, by 2004, many public enterprises had again key financial ratios (e.g., fixed assets/equity, working capital/turnover, etc.) in excess of standard values that banks use for lending. Accordingly, the 2005 Budget Law stipulated that overleveraged, but otherwise viable public enterprises will be restructured financially with budgetary support, including through the buyback of bank debt. The Algerian Treasury concluded the financial analysis of public enterprises in 2006 and announced a program concerning 378 public enterprises (workforce: 108,000) out of 1,002 reviewed (workforce: 320,000).
The total cost of the financial restructuring of the 378 enterprises is 4 percent of GDP, of which 3 percent for buying back bank debt and 1 percent as cash injection. Small amounts of tax liabilities and suppliers’ credit were also written down. These amounts exclude public real estate developers at CNEP, which are estimated to cost another 1.8 percent of GDP.
For 209 enterprises deemed viable (mainly larger entities), the restructuring involved: (a) new equity, in the form of cash, to cover 30 percent of net fixed assets; (b) a buyback of bank debt so that working capital credit does not exceed four months of turnover; and (c) a “supplementary” buyback to bring interest charges below four percent of turnover.
For 169 nonviable enterprises, the restructuring is limited to buying back debt and covering operating deficits through budgetary allocations until they are closed.
Finally, for 29 other public enterprises, the government will buy back enough bank debt to bring their interest charges below four percent of turnover.
In exchange for their financial restructuring, public enterprises agree to multi-year performance contracts, with quantitative targets for securing their economic and financial viability.
1 Information from Algerian Treasury as of November 2007.2 See (Nashashibi et al., 1998).12. The public banks have been so far the main lenders to public enterprises. Public enterprises represent about 3 percent of employment and one fifth of nonhydrocarbon GDP. Public banks had limited success in preventing NPLs on lending to public enterprises. Periodic swaps of NPLs for government securities—“rachats de créances”—helped the banks remain solvent and liquid (Table I.1).3 In 2004, the authorities launched reforms aimed at upgrading risk management, efficiency and services of public banks. The main planks of the reform were new managers bound by performance contracts, new boards with specialized committees, and the privatization process of a first bank.
Indicators of Financial Intermediation, 2003–07
NHM2 = M2 - Sonatrach deposits in the banking system; NHGDP = GDP - hydrocarbon GDP.
Bilateral swaps of government securities for bank claims on public enterprises, mostly NPLs.
Mainly insurance companies and postal accounts (CCP).
Unemployment insurance fund, medical benefits fund, and pension reserve.
Indicators of Financial Intermediation, 2003–07
| 2003 | 2004 | 2005 | 2006 | latest | |||
|---|---|---|---|---|---|---|---|
| (In percent) | |||||||
| NHM2/NHGDP 1/ | 91.4 | 89.6 | 88.3 | 90.0 | 93.1 | ||
| Credit to economy/NHGDP | 40.6 | 36.2 | 33.3 | 30.1 | 27.5 | ||
| (In billions of Algerian dinars) | |||||||
| Credit to private sector | 589 | 675 | 897 | 1,057 | 1,143 | ||
| Public enterprises, net of swaps | 792 | 860 | 883 | 850 | 898 | ||
| Idem, including “rachats de créances” | 1,349 | 1,414 | 1,517 | 1,563 | … | ||
| Government securities--“Rachats de créances” 2/ | 557 | 554 | 634 | 713 | 616 | ||
| Government securities--auctioned | 221 | 253 | 240 | 282 | 315 | ||
| Corporate bonds outstanding | 7 | 56 | 109 | 160 | 170 | ||
| Of which: Private placements | 7 | 42 | 79 | 107 | … | ||
| Assets of financial system | |||||||
| Banks | 3,534 | 3,893 | 4,205 | 5,382 | 6,120 | ||
| Of which: Public banks | 93% | 92% | 91% | 92% | … | ||
| Nonbank financial institutions 3/ | 197 | 212 | 246 | 316 | … | ||
| State insurance and pension reserve funds 4/ | 100 | … | … | 144 | … | ||
| GDP | 5,264 | 6,128 | 7,499 | 8,461 | 9,110 | ||
| Nonhydrocarbon GDP | 3,395 | 3,808 | 4,146 | 4,578 | 5,021 | ||
NHM2 = M2 - Sonatrach deposits in the banking system; NHGDP = GDP - hydrocarbon GDP.
Bilateral swaps of government securities for bank claims on public enterprises, mostly NPLs.
Mainly insurance companies and postal accounts (CCP).
Unemployment insurance fund, medical benefits fund, and pension reserve.
Indicators of Financial Intermediation, 2003–07
| 2003 | 2004 | 2005 | 2006 | latest | |||
|---|---|---|---|---|---|---|---|
| (In percent) | |||||||
| NHM2/NHGDP 1/ | 91.4 | 89.6 | 88.3 | 90.0 | 93.1 | ||
| Credit to economy/NHGDP | 40.6 | 36.2 | 33.3 | 30.1 | 27.5 | ||
| (In billions of Algerian dinars) | |||||||
| Credit to private sector | 589 | 675 | 897 | 1,057 | 1,143 | ||
| Public enterprises, net of swaps | 792 | 860 | 883 | 850 | 898 | ||
| Idem, including “rachats de créances” | 1,349 | 1,414 | 1,517 | 1,563 | … | ||
| Government securities--“Rachats de créances” 2/ | 557 | 554 | 634 | 713 | 616 | ||
| Government securities--auctioned | 221 | 253 | 240 | 282 | 315 | ||
| Corporate bonds outstanding | 7 | 56 | 109 | 160 | 170 | ||
| Of which: Private placements | 7 | 42 | 79 | 107 | … | ||
| Assets of financial system | |||||||
| Banks | 3,534 | 3,893 | 4,205 | 5,382 | 6,120 | ||
| Of which: Public banks | 93% | 92% | 91% | 92% | … | ||
| Nonbank financial institutions 3/ | 197 | 212 | 246 | 316 | … | ||
| State insurance and pension reserve funds 4/ | 100 | … | … | 144 | … | ||
| GDP | 5,264 | 6,128 | 7,499 | 8,461 | 9,110 | ||
| Nonhydrocarbon GDP | 3,395 | 3,808 | 4,146 | 4,578 | 5,021 | ||
NHM2 = M2 - Sonatrach deposits in the banking system; NHGDP = GDP - hydrocarbon GDP.
Bilateral swaps of government securities for bank claims on public enterprises, mostly NPLs.
Mainly insurance companies and postal accounts (CCP).
Unemployment insurance fund, medical benefits fund, and pension reserve.
13. Credit to the private sector remains small by international standards, despite its recent rapid growth. Lack of capital has constrained the banks in developing credit to the private sector.4 The public banks’ capital is only 4 percent of nonhydrocarbon GDP. This small capital allows low overall credit because of capital adequacy rules. Since a significant share of credit still goes to public enterprises, the scope for private sector credit is small.
Credit to the Private Sector in Selected Countries, 2005
(In percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: ECB, and Fund staff estimates.1/ Including public enterprises.Credit to the Private Sector in Selected Countries, 2005
(In percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: ECB, and Fund staff estimates.1/ Including public enterprises.Credit to the Private Sector in Selected Countries, 2005
(In percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: ECB, and Fund staff estimates.1/ Including public enterprises.14. More credit to public enterprises may distract banks from developing the practices and products to finance private sector activity. In particular:
a. instilling a modern risk culture in the banking system becomes more challenging;
b. large new lending to public enterprises will ensure several years of good earnings for the banking system, thus weakening incentives to develop other sources of earnings from financing private sector activity; and
c. the playing field in banking would remain distorted. The government has accepted a low return on equity for the banks it owns. This allows them to compete with the foreign banks, which require returns that are 4–5 times higher.
15. By contrast, the new credit needs of public enterprises are an opportunity to further develop local capital markets. Banks would facilitate access by large corporate clients to the debt markets, while focusing pure lending increasingly on SME.
C. Algeria’s Corporate Bond Market
16. The Algerian market is already sizeable and maturities are long. The government has encouraged large nonfinancial public enterprises to issue bonds since 2003. Two private entities have followed suit in 2006. Maturities extend to 11 years, with a median of six, partly reflecting the tax advantage given to maturities exceeding five years. The stock of bonds is equivalent to half of public enterprises’ medium- and long-term bank credit.
17. The Algerian market is larger than in other countries on the periphery of the EU-15. The Algerian market is about four times larger than the average for these countries. Market size is measured by the outstanding stock of bonds issued by nonfinancial corporations. Only Ukraine has a market of comparable size, reflecting the prevalence of large enterprises in Ukrainian industry.5
Nonfinancial Corporate Bonds in Selected Countries, 2005–06 Average
(Outstanding in percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: BIS; ECB; and Fund staff estimates.Nonfinancial Corporate Bonds in Selected Countries, 2005–06 Average
(Outstanding in percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: BIS; ECB; and Fund staff estimates.Nonfinancial Corporate Bonds in Selected Countries, 2005–06 Average
(Outstanding in percent of GDP; except Algeria, in percent of NHGDP)
Citation: IMF Staff Country Reports 2008, 104; 10.5089/9781451811575.002.A001
Sources: BIS; ECB; and Fund staff estimates.18. Some bonds were significantly oversubscribed in 2006, suggesting scope for growth. The largest issuer is Sonelgaz, the electric utility, which plans to invest about 8 percent of NHGDP until 2009. Half of Sonelgaz’s bonds were issued in the seven-eleven year segment. Other large issuers include Air Algérie and Algérie Télécom. The size of the Algérie Télécom 5-year bond of October 2006 was doubled during the public offering period, due to demand.
19. There has also seen a substantial increase in public offerings compared to private placements in 2006. Private placements, which are limited to institutional investors, represented three quarter of the value of bonds sold in 2003–05, mostly bought by banks. By contrast, public offerings, which are not limited to institutional investors, accounted for more than half of bonds sold in 2006. This suggests strong demand of the general public for nonbank saving vehicles.
20. Launching fixed-income mutual funds would be a natural evolution for financial institutions looking for new sources of income. Mutual funds would be attractive savings vehicle as they could pay at least 150 bp more than term deposits.6 If 10 percent of term deposits switched to mutual funds, corporate bond issuance could potentially double.
21. The corporate market took off after the government started issuing its debt more predictably. Debt issuance through auctions on a regular calendar was introduced in 2002. The aim was to grow the market gradually and ensure reasonable profitability for primary dealers.7 The yield curve reflects mainly the outcome of auctions because secondary market activity is limited. Thus, a new corporate bond offering may sometimes have to wait until the issuance of a government bond of comparable maturity to provide the market with a reliable pricing point.
22. Uncertain timing of budgetary outlays under the public investment program has made government debt issuance somewhat difficult to predict in 2007. The authorities have made provisions for accessing Algeria’s oil stabilization fund to finance these outlays. However, each drawing on the fund requires a decision of the minister of finance. The onset of bulky outlays in early 2007 generated large bridge financing requirements. As a result, the stock of auctioned government securities increased by 45 percent in the first two months of 2007, then dropped 20 percent until September 2007. This pattern has contributed to low yields. The yield curve became inverted at short maturities compared to 2006, when its shape was similar to curves of countries with persistent fiscal surpluses and to the euro and dollar curves (Box I.3). Going forward, the authorities plan to increase predictability by stabilizing the debt stock.
23. New issuance of corporate bonds has been weak in 2007. The stock of corporate bonds has increased marginally in 2007, from DA 150 billions to DA 170 billions, all in private placements.
D. Further Development of Algeria’s Capital Markets
24. A market for corporate debt securities spurs banks to increase efficiency and innovate. Capital markets provide debt finance at lower overhead cost than banks, once credit risks become predictable and bond features sufficiently standardized (Box I.4). As a result, banks must innovate by customizing their financing and services to clients with potential access to capital markets. To make up for straight loan income that is lost to the market, banks also direct their lending efforts to other borrowers, such as SME, also because SME loans are too small to justify the fixed issuance costs of capital markets.
25. A robust financial system requires complementary channels of financing, where banks and markets compete for borrowers. Through diversification of financing channels, capital markets help cushion the real economy from shocks to bank intermediation, and vice-versa.8 (Greenspan, 1999) compared the corporate debt markets to an economy’s “spare tire” as Asian banks became reluctant to extend credit. Local currency market finance also helps avoid currency mismatch in the real economy and mitigates maturity mismatch on banks’ balance sheets.
Yield Curves for Algerian Dinar and Selected Currencies, 2006–071
Key Features of Corporate Bond Markets
The corporate market is for borrowers of fairly well-established reputations. Financing extends from the short term—this is the commercial paper (CP) market of maturities up to one year, to the long term—the bond market per se. Issuers are financial institutions (the largest segment in the EU because of covered bonds) and nonfinancial corporations, either residents or nonresidents. Issuers in the market are fairly well-known “names” with acceptable risk characteristics for inclusion in fixed-income portfolios.
Buyers are typically prudent institutional investors, such as pension and life insurance companies, and investment funds (money market funds for CP; fixed-income funds for bonds). Mainly for regulatory reasons, institutional investors prefer bonds from reputable “names” with fairly low credit risk. In more advanced stages of development, certain buyers emerge that are willing to take on higher credit risk. At that point, entities with relatively less known track records can also issue in the market.
Bonds are cheaper than bank loans, for issuers that qualify and for bond offerings of sufficient size. In advanced financial systems, banks earn very little (a few basis points) on the financing of medium-large corporations, as these have achieved the needed reliability of financial information for accessing capital markets. The size of an individual bond must however be large enough (higher than $30–50 million equivalent) in order to dilute the overhead costs (legal, regulatory, marketing, custody, etc.). In addition, the terms of bond offerings (maturity, covenants, frequency of debt service, option features, etc.) are typically highly standardized, so that price differences across bonds reflect pure differences in credit risk.
The banking system benefits from a vibrant bond market in many ways. Certain banks of long standing reputation can specialize in “bringing issuers to market,” that is, ensuring strong demand for new corporate bond offerings through valuable marketing, distribution, and market-making services. These banks earn fees for arranging, underwriting, marketing, trading, and safekeeping. In addition, a vibrant market ensures a robust stream of information on the price of credit risk for various categories of borrowers. Banks use this information to price individual loans and manage the risk of their loan books. Finally, banks are spurred to innovate in order to retain the business of clients with potential access to market finance.
26. Competition between the corporate bond market and banks is important for Algeria’s financial sector development. The earnings of banks in 2006 show the effects of this competition. The banks would have earned DA 4–5 billion more, had the large public enterprises not replaced bank loans with cheaper bonds.9 The banks’ earnings in 2006 were boosted by income from liquidity absorption operations. Earnings from pure banking, which exclude this exceptional income, were about DA 9 billion. Therefore, bond market competition has lowered bank earnings by about a third, encouraging banks to develop other sources of income.
27. The government could encourage public entities to continue issuing bonds and extending maturities, especially on NIP-related projects. Public entities comprise public enterprises and other entities that have their own income stream to service bonded debt, for example the new public corporations being set up through project financing (sociétés publiques de projet). And also the Algerian banks: they are sufficiently known by the public to issue their own bonds. This would help them diversify funding away from term deposits. Issuance by Algerian public banks would also signal a willingness to be judged by investors, and provide the government with information on their value. Among other countries, Croatia, Czech Republic, and Hungary have seen issuance of significant size by financial institutions.
28. Issuance by foreign companies incorporated in Algeria would enhance market credibility. Some such companies have a need for term loans in dinar. These companies are accustomed in other countries to high standards of financial services and would progressively impress these standards on the local financial institutions.
29. There is no market yet for commercial paper, although the potential is large. The commercial paper market competes with banks in providing short-term credit to reputable borrowers. Short-term credit is more than half of total credit in Algeria. Some substitution of short-term bank loans by commercial paper is thus possible, as happened with corporate bonds and longer term bank loans. The authorities have largely completed the legal and tax groundwork for the commercial paper market. Some public enterprises could show the way, as they have for bonds, with private entities following, including foreign ones that need working capital finance in dinar.
30. Corporate debt markets rely on government yield curves for the pricing of new corporate debt securities. The yield on a new corporate debt security is the sum of the yield on the most recent government security of the same maturity plus a premium for the credit risk of the corporate.10 The reliability of government yields as a basis for pricing new corporate debt is open to question for governments that have no need to borrow. In these cases, the supply of government securities naturally falls below the demand of investors that have to hold such securities for regulatory purposes and/or cannot invest abroad. In Algeria, government securities of short maturities have recently become insufficient in relation to the regulatory needs of the few and small local insurance firms.
31. A well-structured public debt issuance program underpins the reliability of the yield curve. Countries with fiscal surpluses have realized that continued issuance of government debt supports competitive financial intermediation.11 Continued issuance anchors a reference term structure of local interest rates, maintains a liquid and efficient government debt market, and facilitates the transmission of central bank decisions on policy rates to the cost of credit in the economy. After careful consultations with market practitioners, some countries have increased gross debt and concentrated issuance on a limited number of benchmark lines to capitalize on these benefits. The Algerian Government may also have to structure its issuance program so as to foster the development of local capital markets.
E. Improving the Government Benchmark for Corporate Bonds
32. Looking ahead, more secondary trading of government securities would improve the price benchmark for corporate debt securities. Secondary trading generates more continuous interest rate information than auctions. Meaningful secondary trading develops when a critical mass of investors has frequent needs to restructure their portfolio of securities (increasing or reducing the portfolio; changing its duration). Trading needs arise in greater quantities and at higher frequency, the more heterogeneous the investors’ needs and views on economic fundamentals.
33. More secondary trading will be challenging, because many domestic investors will hold securities to maturity for the foreseeable future. Some more trading may be possible if public banks, as the largest holders of government securities by virtue of the swaps, were compelled to trade more often. This may involve creating a structural liquidity deficit for more banks.
34. To address this issue, the authorities may have to consider a combination of:
increasing the required reserve ratio to remove liquidity from most banks on a permanent basis. The central bank would continue to pay a market rate on required reserves, to offset any cost to banks of meeting the reserve requirement;12
providing the central bank with government securities that it can sell into the market. Outright sales of government securities remove liquidity via a more price-based mechanism than required reserves;
developing central bank repos of government securities. Banks will feel comfortable buying the above securities if central bank repos ensure their liquidity;
requiring that interbank loans take the form of repos. Some marginal banks would thus need to increase their holdings of government securities; and
reducing the frequency of central bank operations from once a week to once every two weeks. Given the slow pace of financial intermediation in Algeria, a bank needing liquidity can afford to wait until the next weekly operation of the central bank, rather than seek an earlier deal in the interbank market. Two weeks may push more banks more often to the market.
35. Robust secondary trading further ahead will require significantly greater heterogeneity of investors, mainly by allowing foreign buyers. There is not much scope for diversifying the local investor base. The development of the local pension fund or life insurance industry is constrained by the government’s growing financial wealth. This weakens incentives for old-age saving and helps explain that life insurance has failed to develop despite a fairly modern legal framework. Meaningful investor heterogeneity may involve foreign investment into the government securities market.
Appendix I.1. Key Financial Stocks and Flows, 2006–07
(In billions of Algerian dinars; unless otherwise indicated)
Excluding Sonatrach.
Authorized coverage when scheme is fully deployed.
See box 2 for details.
As of end-June.
| 2006 | Sep. 2007 | ||
|---|---|---|---|
| Bank credit to public enterprises | 850 | … | |
| Of which: medium and long-term investment credit | 440 | … | |
| Public enterprise bonds outstanding | … | 150 | |
| Overall cost of NIP | … | 12,044 | |
| Public investment program (government budget) | … | 10,530 | |
| Public enterprises 1/ | … | 1,514 | |
| New credits required for NIP | … | 443 | |
| Of which: government guaranteed | … | 390 | |
| Credit to the private sector | 1,057 | … | |
| Of which: SME investment credits | 300 | … | |
| New scheme to guarantee SME investment credits 2/ | 360 | … | |
| Government securities outstanding | 1,118 | … | |
| Auctioned securities | 282 | … | |
| Swaps “rachats de créances” | 713 | … | |
| Other | 123 | … | |
| New swaps planned 3/ | 318 | … | |
| Oil stabilization fund 4/ | … | 3,195 | |
| Banking system deposits | 3,516 | … | |
| Of which: term deposits | 1,766 | … | |
| Memorandum items | |||
| GDP | 9,110 | … | |
| Nonhydrocarbon GDP | 5,021 | … | |
| International reserves (In billions of US dollars) | … | 100 | |
| Government external debt | … | 0.8 | |
| Average DA/$ exchange rate | 73 | … | |
Excluding Sonatrach.
Authorized coverage when scheme is fully deployed.
See box 2 for details.
As of end-June.
| 2006 | Sep. 2007 | ||
|---|---|---|---|
| Bank credit to public enterprises | 850 | … | |
| Of which: medium and long-term investment credit | 440 | … | |
| Public enterprise bonds outstanding | … | 150 | |
| Overall cost of NIP | … | 12,044 | |
| Public investment program (government budget) | … | 10,530 | |
| Public enterprises 1/ | … | 1,514 | |
| New credits required for NIP | … | 443 | |
| Of which: government guaranteed | … | 390 | |
| Credit to the private sector | 1,057 | … | |
| Of which: SME investment credits | 300 | … | |
| New scheme to guarantee SME investment credits 2/ | 360 | … | |
| Government securities outstanding | 1,118 | … | |
| Auctioned securities | 282 | … | |
| Swaps “rachats de créances” | 713 | … | |
| Other | 123 | … | |
| New swaps planned 3/ | 318 | … | |
| Oil stabilization fund 4/ | … | 3,195 | |
| Banking system deposits | 3,516 | … | |
| Of which: term deposits | 1,766 | … | |
| Memorandum items | |||
| GDP | 9,110 | … | |
| Nonhydrocarbon GDP | 5,021 | … | |
| International reserves (In billions of US dollars) | … | 100 | |
| Government external debt | … | 0.8 | |
| Average DA/$ exchange rate | 73 | … | |
Excluding Sonatrach.
Authorized coverage when scheme is fully deployed.
See box 2 for details.
As of end-June.
References
Bank for International Settlements, 2002, The Development of Bond Markets in Emerging Economies, BIS Papers No. 11 (Basel: BIS).
Bank of Algeria, 2002, Annual Report (Algiers: Bank of Algeria).
Bank of Algeria, 2007, Tendances monétaires et financières au second semestre de 2006, available on the internet: http://www.bank-of-algeria.dz.
Benachenhou, Abdellatif (ed.), 2004, Financer le développement autrement: du budget au marché (Algiers: Alpha Design).
Comley, Blair, and David Turney, 2005, Debt Management in a Low Debt Environment: The Australian Case, Australian Treasury Working Paper No. 2005–02.
Committee on the Global Financial System, 2007, Financial Stability and Local Currency Bond Markets, CGFS Papers No. 28 (Basel: BIS).
European Central Bank, 2006, Bond Markets and Long-Term Interest Rates in the European Union and Acceding Countries (Frankfurt-am-Main: ECB).
Greenspan, Alan, 1999, Do Efficient Financial Markets Mitigate Financial Crises?, speech at the Financial Markets Conference of the Federal Reserve Bank of Atlanta, Sea Island, Georgia, 19 October.
International Monetary Fund, 2004, Algeria: Financial System Stability Assessment, Country Report No. 04/138 (Washington: International Monetary Fund (IMF)).
Nashashibi, Karim, et al., 1998, Algeria: Stabilization and Transition to a Market Economy, Occasional Paper (Washington: International Monetary Fund).
Norges Bank, 2003, Why Government Debt, available on the internet: http://www.norges-bank.no.
Rajan, Raghuram, and Luigi Zingales, 2003, Banks and Markets: The Changing Character of European Finance, in Second ECB Central Banking Conference: The Transformation of the European Financial System (Frankfurt-am-Main: ECB).
Roldos, Jorge E., 2004, Emerging Local Bond Markets, in Emerging Local Securities and Derivatives Markets, World Economic and Financial Surveys (Washington: IMF).
World Bank and IMF, 2001, Developing Government Bond Markets: A Handbook, (Washington: World Bank).
Prepared by Gabriel Sensenbrenner. Figures in this paper generally reflect data as of end-2006, unless otherwise indicated. Appendix I.1 lists key financial stocks and flows for the Algerian economy.
The carrying cost of swaps has been immaterial to fiscal sustainability in an era of high hydrocarbon budget revenues. See (Bank of Algeria, 2002) for details on swaps between 1991 and 2001. IMF Country Report No. 04/138 estimated that swaps averaged 4 percent of GDP per year between 1991 and 2002.
Algeria’s ratio is expressed in percent of nonhydrocarbon GDP to facilitate comparison with countries with no hydrocarbon sector and because Algeria’s hydrocarbon sector does not borrow.
Algeria and Ukraine have by far the lowest density of SMEs among countries on the periphery of the EU-15 (see IMF Country Report No. 06/101).
A mutual fund composed of government bonds (yielding 3–5 percent per year) and corporate bonds (5–7 percent) should yield about 4 percent after expenses. Term deposits earned between 1.25 percent and 2.50 percent in 2006 (except at CNEP).
Primary dealers (PD) are most of the public banks, one foreign bank, and several nonbank financial institutions. The treasury informs PDs of the issuance calendar on a quarterly basis. Auction frequency is weekly. Three days before, the central bank as agent of the treasury informs PDs of intended amounts by type of security. Bids are submitted electronically. Settlement is DVP on the large value system.
Capital markets stepped in when U.S. banks retrenched in the late 1990s (real estate bubble), and again in late 1998 (Long Term Capital Management). Banks stepped in when capital markets retrenched in 2001 (downgrade of major traditional issuers of commercial paper).
The calculation is based on the fact that: (a) bonds have been about 200–300 bp cheaper than bank loans; (b) public banks held some DA 100 billion worth of corporate bonds in 2006, including clients’ securities accounts; (c) various government funds and insurance companies held the balance of some DA 60 billion.
Assuming the same liquidity for both securities.
See (Comley and Turvey, 2005) for Australia (gross debt: 7 percent of GDP) and (Norges Bank, 2003) for Norway (gross debt: 17 percent of NHGDP). The Singapore Government increased its gross debt from 20 percent of GDP in 1998 (at the time, mostly held by the state retirement fund) to 40 percent in 2004 with a view to building large benchmark to develop the corporate bond market. During this time, the stock of corporate bonds increased from 20 percent of GDP to 70 percent of GDP.
The interest rate on required reserves should vary automatically to be always equal to the central bank’s rate on its main refinancing operations. In this way, the reserve requirement becomes a pure monetary policy tool for creating or increasing a structural liquidity deficit of the banking system, not a tax on banks. The ECB uses this approach to increase significantly the structural deficit in order to ensure the clarity of its policy signal.
References
Areski, Rabah, 2007, “Determinants of Oil Exporting Countries Current Account Positions”, Mimeo, Washington D.C.: International Monetary Fund (IMF).
Cashin, Paul, Felipe Céspedes, and Ratna Sahay, 2004, “Commodity Currencies and the Real Exchange Rate”, Journal of Development Economics, Vol. 75, No 1, pp. 239 –68.
Davis, Jeffrey M., Rolando Ossowski, and Annalisa Fedelino, 2003, Fiscal Policy Formulation and Implementation on Oil-Producing Countries, Washington D.C.: IMF.
Davoodi, Hamid, 2002, “Assessing Fiscal Vulnerability, Fiscal Sustainability, and Fiscal Stance in a Natural Resource Rich Country”, Republic of Kazakhstan-Selected Issues, IMF Country Report 02/64, Washington D.C.: IMF.
Floerkemeier, Holger, 2004, “Fiscal Management of Hydrocarbon Revenues”, Algeria-Selected Issues, IMF Country Report 05/52, Washington D.C.: IMF.
International Monetary Fund, 2006, Methodologies for CGER Exchange Rate Assessments, Available via: http://www.imf.org/external/pp/longres.aspx?id=3957
Isard, Peter, and Hamid Faruqee, 1998, Exchange Rate Assessment. Extensions of the Macroeconomic Balance Approach, IMF Occasional Paper No. 167, Washington D.C.: IMF.
Isard, Peter, and Hamid Faruqee, G. Russel Kincaid, and Martin Fetherson, 2001, Methodology for Current Account and Exchange Rate Assessments, IMF Occasional Paper No. 209, Washington D.C.: IMF.
Koranchelian, Taline, 2005, “The Equilibrium Real Exchange Rate in a Commodity Exporting Country: Algeria’s Experience,” IMF Working Paper 05/135, Washington D.C.: IMF.
Loko, Boileau, and Anita Tulhadar, 2005, “Labor Productivity and Real Exchange Rate: The Balassa-Samuelson Disconnect in the Former Yugoslav Republic of Macedonia”, IMF Working Papers 05/113, Washington D.C.: IMF.
Spatafora, Nikola, and Emil Stavrev, 2003, “The Equilibrium Real Exchange Rate in a Commodity Exporting Country: The Case of Russia”, IMF Working Papers 03/93, Washington D.C.: IMF.
Zalduendo, Juan, 2006, “Determinants of Venezuela’s Equilibrium Real Exchange Rate”, IMF Working Papers 06/74, Washington D.C.: IMF.
Appendix II.1. Econometric Results
Unit-Root Tests 1/
REER, OIL, PROD, and G stand for the log of real effective exchange rate, real price of oil, and productivity differential between Algeria and its main partners. The lag length is determined mechanically based on Schwarz information criterion (SIC) *Significant at 1 percent.
| ADF | ADF (with trend) | PP | PP (with trend) | |
|---|---|---|---|---|
| Level | ||||
| REER | −0.175 | −2.325 | −0.369 | −1.765 |
| OIL | −2.272 | −2.248 | −2.274 | −2.248 |
| PROD | −0.262 | −2.158 | −0.179 | −2.080 |
| G | −3.005 | −3.012 | −2.429 | −2.333 |
| First Difference | ||||
| REER | −4.439* | −4.428* | −4.424* | −4.464* |
| OIL | −6.106* | −6.030* | −6.106* | −6.030* |
| PROD | −8.428* | −8.608* | −7.750* | −7.854* |
| G | −4.954* | −4.876* | −7.847* | −7.565* |
REER, OIL, PROD, and G stand for the log of real effective exchange rate, real price of oil, and productivity differential between Algeria and its main partners. The lag length is determined mechanically based on Schwarz information criterion (SIC) *Significant at 1 percent.
| ADF | ADF (with trend) | PP | PP (with trend) | |
|---|---|---|---|---|
| Level | ||||
| REER | −0.175 | −2.325 | −0.369 | −1.765 |
| OIL | −2.272 | −2.248 | −2.274 | −2.248 |
| PROD | −0.262 | −2.158 | −0.179 | −2.080 |
| G | −3.005 | −3.012 | −2.429 | −2.333 |
| First Difference | ||||
| REER | −4.439* | −4.428* | −4.424* | −4.464* |
| OIL | −6.106* | −6.030* | −6.106* | −6.030* |
| PROD | −8.428* | −8.608* | −7.750* | −7.854* |
| G | −4.954* | −4.876* | −7.847* | −7.565* |
REER, OIL, PROD, and G stand for the log of real effective exchange rate, real price of oil, and productivity differential between Algeria and its main partners. The lag length is determined mechanically based on Schwarz information criterion (SIC) *Significant at 1 percent.
Cointegration tests
To correct for a possible small sample bias, we use Reinsel and Ahn’s approach, which suggests that the critical value be adjusted upward by a multiplicative scaling factor, T/(T-nj), with T the sample size, n the number of variables, and j the number of lags. After correcting the critical values, both Trace and Max-Eigen statistics reject the null hypothesis of zero cointegrating equation. Both tests indicate the presence of one cointegrating vector.
Trace Statistics for Cointegrating Rank
Significant at 5 percent; r= cointegrating rank
Trace Statistics for Cointegrating Rank
| Eingevalue | Trace Statistics | Critical Value | Critical value adjusted | |
|---|---|---|---|---|
| r=0 | 0.86 | 117.04* | 47.85 | 95.70 |
| R≤1 | 0.73 | 54.62 | 29.79 | 59.58 |
| R≤2 | 0.33 | 13.06 | 15.49 | 30.98 |
| R≤3 | 0.00 | 0.02 | 3.84 | 7.68 |
Significant at 5 percent; r= cointegrating rank
Trace Statistics for Cointegrating Rank
| Eingevalue | Trace Statistics | Critical Value | Critical value adjusted | |
|---|---|---|---|---|
| r=0 | 0.86 | 117.04* | 47.85 | 95.70 |
| R≤1 | 0.73 | 54.62 | 29.79 | 59.58 |
| R≤2 | 0.33 | 13.06 | 15.49 | 30.98 |
| R≤3 | 0.00 | 0.02 | 3.84 | 7.68 |
Significant at 5 percent; r= cointegrating rank
Max-Eigen Statistics for Cointegrating Rank
Significant at 5 percent; r= cointegrating rank
Prepared by Boileau Loko.
Further details on these methodologies can be found in (Isard and Faruqee, 1998) and (Isard et al., 2001).
See (Spatafora and Stavrev, 2003); (Koranchelian, 2005); and (Zalduendo, 2006).
The hydrocarbon sector represented about half of GDP, nearly all exports of goods, and three quarter of fiscal revenues.
The national hydrocarbon company (Sonatrach) fully surrenders foreign currency receipts generated by hydrocarbon exports to Bank of Algeria. In this context, oil prices are strongly related to other usual determinants of exchange rate such as net foreign assets.
Nominal oil price is projected to increase to $80.2 per barrel in 2012 from $64.3 in 2006 (WEO, November 2007). During the same period, high productivity in trading partners such as China, Russia, and Ukraine will increase the differential productivity, while government spending to GDP will increased to 33 percent from 28.5 percent in 2006.
The forecast standard error of REER found in CGER is about 12 percent, which is reduced to 7–8 percent if one accounts for factors driving the REER in the short run through an error-correction specification.
(CGER, 2006) investigated the major determinants of the current account using a panel of 54 advanced and emerging countries, including Algeria. However, the sample does not include other major oil/gas exporting countries.
In 2004, the authorities launched the Growth Consolidation Plan, a multi-year public investment program (2005–09) that amounted to $50 billion (about 40 percent of 2007 GDP). The program has been revised upward several times to reach about $155 billion (120 percent of 2007 GDP).
This is in line with CGER results for Algeria.
Sonatrach, state-owned hydrocarbon company, is a major component of the nongovernment sector. On average one-third of hydrocarbon export revenues accrue to Sonatrach and two-thirds accrue to the government.
For more details, see (Jeffrey, Ossowski, and Fedelino, 2003).
See (Floerkemeier, 2004).
For more details, see (Davoodi, 2002) and (Floerkemeier, 2004).
The PIF requires assumptions about future hydrocarbon production and exports, and interest/discount rates.
