Cameroon - Financial sector report

Report No. 6028-CM Cameroon Financial Sector Report lune 2,1986 Western Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorizationm  ABBREVIATIONS AND ACRONYMS Commercial Banks BCC - Bank of Credit and Commerce BIAO - Banque Internationale de l'Afrique Occidentale BICIC - Banque Internationale pour le Commerce et l'Industrie du Cameroun BPPC - Banque de Paris et des Pays-Bas du Cameroun SCB - Societe Camerounaise de Banque SGBC - Societe Ggngrale de Banques au Cameroun Financial Establishments SCE - SocitE Camerounaise d'Equipement SOCCA - Societe Camernunaise de Credit Automobile SOCABAIL - SociWtE Camerounaise de Credit Bail SOGELEASE - Societe Generale de Leasing au Cameroun Other Institutions of Financial Nature or Function BCD - Banque Camerounaise de Developpement FGEN - Fonds de Gestion de l'Epargne Nationale FOGAPE - Fonds d'Aide et de Garantie aux Petites et Moyennes Entreprises FONADER - Fonds National de D6veloppement Rural SNI - Sociftg Nationale d'Investissement CFC - Crgdit Foncier du Cameroun CAA - Caisse Autonome d'Amortissement CNPS - Caisse Nationale de Pr6voyance Sociale ONCPB - Office Nationale de Commercialisation de Produits de Base SNH - Socift6 Nationale des Hydrocarbures Central Banks BCEAO - Banque Centrale des Etats de l'Afrique Occidentale BEAC - Banque des Etats de l'Afrique Centrale AVERAGE EXCHANGE RATES Fiscal Years CFAF/US$ 1980 - 209.2 1982 - 296.7 1984 - 409.5 1981 - 235.3 1983 - 354.7 1985 - 471.1  FOR OWFIVIL USE ONLY CAMEROON FINANCIAL SECTOR REPORT Table of Contents Psae No. Executive Sumary *i A. Overview: Performance and Issues 1 B. Sources, Uses and Deployment of Savings ifi C. Financial Intermediation v D. The BEAC, Monetary Policy and Interest Rates viii E. Money and Capital Markets x F. Summary Recommendations xv I. Public Resource Mobilization and Asset Management A. Introduction i B. The Mechanics of Public Resource Mobilization 1 C. Planning and the Extra-budgetary Accounts 2 D. Government Savings Performance 3 E. Caisse Autonome d'Amortissement 3 F. Public Liquidity: Quasi-governmental Bodies 4 C. Conclusions and Recommendations 7 II Private Resource Mobilization and Financial Intermediation A. Introduction 8 B. Financial Intermediation 9 C. The Adequacy of the Cameroonian Financial System 10 D. The Institutional Perspective on Limits to Finaucial Deepening 14 E. The Efficiency of the Financial Intermediation System 19 F. Conclusions and Recommendations 20 Annex 23 This Report is based on the findings of a Bank financial sector mission to Cameroon in June-July, 1985 led by Philip Berlin. The green cover version was discussed during a subsequent mission of February - March 1986. Mission participants included Pedro Alba, James Houston, Cristian de BoIssieu (consultant:) and Hafes Ghanem (economists) as well as Jacques Toureille (financial specialist) and Jorge Calderon-Rossell (UFC, capital narkets specialist). mhe Report benefited from the important work of Nancy Benjamin on financial social accounting matrices, and with respect to the latter enjoyed cons.-%rable support from Iradj Alikhani. Michel Drouin, Cameroon-based consultant, protided s ibstantial background material for the mission and the ultimate Report. Roger Tchoung.i of -he Ministry of Finance, whose services were generously provided by the Government, ply'- - entral role in the overall effort. This document hbas estrtd distibution an may be ud by reipInts on in te pefornune of thei offcl dutes. I contents my not othrwbe be dicked without Wod Bank autnation.  - ii - III. -Flows of Funds and the Cameroonian Financial aystem A. Introduction 27 B. The Savings-Investment Process in Cameroon 28 C. Summary and Conclusions: The Impact of Oil on the Financial System 41 IV. Central BankinA in Cameroon A. Money and Credit 43 1. Introduction 43 2. Institutional Elements 45 3. Monetary Policy and its Conduct 46 4. The Rediscount System: An Evaluation 52 5. A Reform Program 53 B. Interest Rates 57 1. Introduction 57 2. Interest Rates and Monetary Policy 58 3. Interest Rate Selectivity 59 4. A Revision of the Interest Rate Structure 59 C. A Possible BE.AC-Zone Money Market 63 V. The Cameroonian Mor ' and Capital Market A. Introduction 65 B. The Commercial Banking Sector 66 C. Development Banking Institutions: BCD, FOGAPE and FONADER 74 D. Capital Market Operations: Non-bank Financial Institutions 81 E. A Fonds de Gestion de l'Epargne Nationale (FGEN) 87 Appendix I : Financial and Capital Market Statistics Appendix II : The Financial Social Accounting Matrix  EXECUTIVE SUMMARY Prefatory Remarks Cameroon is a basically well-managed economy with a population of nearly 10 million people enjoying a per capita income of some $800 per year. The country also enjoys a certain financial ease stemming from its oil resources which generate somewhat more than 15% of its GDP, at the same time contributiag about 45% of total budgetary resources. Proven oil reserves are,expected to be exhausted within the next ten years, If not sooner, and with the recent precipitous drop in prices, prospects for-this sector are growing increasingly dim. Conscious of this, the Government has been concerned to establish more firmly the basis for on-going growth in the post-oil era. In both the imediate sense--the commercial banks are in poor straits, too poor to carry out much of a development mandate--and in the longer-run perspective as well, Cameroon's financial system is increasingly ill-equipped to meet the needs of a sophisticated economy, tnd it is in this context that the Government requested the World Bank to carry out a thorough-going survey of its financial system. For the most part the Report is addressed to the Government itself. Description is provided more as a backdrop for the analysis and recommendatione and less for general background. The reader seeking general information or data (such as monetary surveys, balance of payments or debt statistics) is unlikely to find them, except as they advance the argument. Recommendations with respect to sector reform (including under the franc zone system) touch on all aspects of the system in a way which focusses on its inter-relatedness; thus they are fairly all-encompassing. Proposed reforms, however, are not developed to the full extent of the detail which such reforms would merit. Here, It was felt most appropriate for these proposals to be discussed with the Government at the general level, following which more specific recommendations, particularly with respect to timing, could be worked out. This said, the Report is rather ambitious in its present scope. A. Overview: Performance and Issues 1. The performance of the Cameroonian financial system, dominated by the franc zone apparatus within which it must operate, has been mixed. With the important exception of the near-insolvency of the banking system, there are no gross or obvious problems which threaten Cameroon's financial or economic stability, its creditworthiness, or the ability of its financial system to perform sdequately on a day to day basis. Unlike the situation in many other developing countries, Government finances do not pre-empt the banking sector, but rather work to support it via budget surpluses or the savings of the quasi-public institutions. The system has certain structural weaknesses due to what remains an excessive dependence upon the Central Bank, especially for medium-term loans, as a source of funds. However, It is essentially the substantial level of public deposits which has permitted the system to function relatively smoothly. The process does, however, mask the factors which ultimately compromise the  - ii - ability of the financial system to function as effectively as it should. A longer-run problem is that the system has not developed sufficiently effective intermediary structures to mobilize financial savlngs and eventually replace Government deposits which the depletion of oil revenues will ultimately require. 2. The Csmeroontan financial system essentially functions in sub- optimal fashion with respect to its ability to provide a firm underpinning for growth. Not only does the system fail adequately to mobilize private financial savings, but it tends to constrain the development of financial intermediation as well. This, in turn, impedes future economic growth because savers are discouraged from participating in the domestic financial system and because such financial savings as are available are diverted from sectors of maximum productivity. With the rapid exhaustion of oil resources and revenues, this also implies that the non-oil private sector must be sufficiently strong, not only to replace the oil sector as a source of financial savings necessary for groi-th, but also to provide the wherewithal, through domestic taxation, Lo replace the oil sector as a source of Government revenues. If the financial system remains as weak as it presently is, there is some doubt as to whether these goals may be achievable. Cameroon and the Franc Zone 3. The BEAC/franc zone system has been an important element in the development of the francophone African economies, 1I particularly in maintaining the links of these economies with the rest of the world through the full convertibility of their currencies. Beyond maintaining the conditions for full convertibility, the system was originally developed to give support to the financially underdeveloped economies of the region, particularly with respect to the nearly non-existent commercial banking sector, for which the common Central Bank represents a major complement. Despite the growth of large commercial banks in the main urban centers, there are, however, strong indications that this aspect of the system as it currently operates is increasingly ill-adapted to Cameroon's economy, to the point where it is more and more working to impede the development of Cameroon's domestic financial intermediation system. 4. The system is, moreover, not as transparent as it should be for effective management. Instr ents used under the regulatory framework of the franc zone system for the control of the money supply are basically overdetermined for that particular task, and mix quantity and quality considerations in such a way that neither is adequately dealt with. As a consequence, it is practically impossible to determine the impact of 1/ Besides Cameroon, its members include Equatorial Guinea, Chad, the Central African Republic, the Congo and Gabon.  - iii - different policy changes either on the domestic money supply or on other target variables sought. Commercial Banking 5. The commercial banking system is highly concentrated and fragile and exemplifies many of the weaknesses of the overall financial system. Four of a totad. of ten banks cortrol roughly 85X of all lending, thus making free-market solutions, particularly with respect to interest-rate determination, impracticable, at least until concentration is diminished. 6. The commercial bauks are for the most part technically insolvent, the sector being informally (but not in fact) recapitalized by Government deposits of undefined term. The insolvency of the sector is the result of a number of ill-considered loans which have led to non-performing assets which are some six times greater than bad debt reserves as presently constituted, and which represent as much as four times banks' current capitalization. This in turn has made the banks considerably more risk- averse with respect to their domestic lending. This has had the result of reducing the already limited access of smaller-scale Cameroonian investors to capital and has probably made them even more dependent upon the informal sector (the tontines), whose lending costs are extremely high and whose maturities are short. The existence of these institutions, whose activities -epresent a substantial share of total credit to the Cameroonian economy, points up the sharp segmentation of the financial sector: there is practically no contact, either between tontines and commercial banks or among tontines themselves. Thus in many ways the tontines, whose participation is usually limited to relatively few eambers, function much like individual savers in a barter economy financing their own investments. Because the level of intermediation, i.e., the ability of the financial system to transfer investable resources between savers and investors remains generally embryonic in Cameroon, high interest costs and substantial impediments to term-transformation will continue to represent a structural constraint to private investment until a broad-front attack is made on the interlocking weaknesses of the system. B. Sources, Uses and Deployment of Savings Public Savings 7. Cameroon's public sector generates a surplus of savings over those required to finance domestic investments at current levels, a phenomenon which is rare among developing countries-even the oil-exporting ones. However, should the recent sharp decline in oil prices and revenues  -iv- be permanent, this situation will not be maintained, and considerable adjustments with respect to fiscal policy will be required. / 8. While the Government's fiscal policies have been prudent, resulting in a favorable evolution of public savings, the management of the Government's portfolio is generally acknowledged to have been weak, and a portfolio cum debt-management strategy should be established to manage current surpluses and ultimately to enunciate government borrowing strategies. While the Treasury currently manages such resources, there is, by common consensus, little short or long-term rtrategy. A decree establishing a new Caisse Autonome d'Amortissement (CAA), among others to manage Government debt was recently promulgated. This new organization will also, according to its new mandate, manage other surplus resources of the Treasury, as well as carry out new borrowings on its account. While the precise mandate of the CAA has yet to be defined, it should begin to define a medium-tenm borrowing strategy consistent with the expected evolution of Government surpluses/deficits, taking into account existing interest-rate structures as well as expected oil revenue flows. For a number of reasons essentially bearing upon the quality of overall portfolio management, oil resources should be managed concurrently with those of the Treasury proper, whether they have been formally repatriated and "budgetized" or not. Quasi-public Resources 9. In addition to Central Government savings, public resources available from the important quasi-public bodies such as the marketing board (ONCPB), the national provident fund (CNPS), and the national savings fund, are substantial. These are structurally surplus organizations presiding over a large volume of investable resources. These are separately managed, often employed for unprogramed public investments, and--even thoug). being important potential resources for term-lending--are mostly deposited in short-term time deposits with the commercial banks, in effect contributing heavily to negative term transformation. It will be argued that these resources should be used, in part to capitalize, and in part as a deposit base for, a new non-bank financial intermediary which would help reinforce the now very weak financial intermediation apparatus existing in Cameroon. They should also ultimately be used to support a secondary market for Government (or other) securities, as well as in the support for the eventual privatization of selected public enterprises. Unlike the case with the CAA resoutces, strict attention would be paid to ensuring that the new intermediary's investments were financially sound. 2/ Required Government adjustments to the sharper than expected oil price decline, as well as projections for public savings and consumption under different policy scenarios, are examined in a forthcoming Country Economic Memorandum, and are not addressed here.  v - Private Savings 10. Private savings must play an lAcreasingly limportant role in financing domestic investments as oll resources dwindle although public savings will for the next five- to ten-year period play the major role in overall Camnroonian resource mobilization. In this context, however, It is important to understand that the concern will be largely with private financial savings, and not savings ln the national accounts sense of consumption foregone. It La accepted in this report that savings in the latter sense are not easily amenable to policy instruments, and particularly interest rate levels, at- least in the relevant range appropriate for an economy as open as Cameroon's. Thus it is important to attract savers to ,iolding the proceeds of their savings in deposit accounts -or In other financial Instruments. This is so for a number of reasons, the most important being that the financial system itself should be in a position to allocate investable resources among the most remunerative investments. C. Financial Intermediation 11. Observers agree that the level of financial sophistication of Cameroon's economy is far less than it should be for its per capita income. It is very likely that Came-oon's economic growth is being increasingly constrained by the inability of its financial institutions to mobilize and channel resources as required. 12. As a general rule, the capability of a financial syst , to mobilize (financial) savings is closely associated with its efficiency in distributing or Intermediating these resources. The better a system is set up to supply Investable resources with characteristics of specific attraction to borrowers with a range of different financing requirements, the more likely It will be able to offer deposit or other Instruments to lenders which combine, for example, different "packages" of return, risk and maturity. These in turn should engender a greater supply response from lenders making the choice between physical assets, currency and financial instruments. 13. It is generally believed that as financial intermediation increases, the Institutional distance between ultimate lender and borrower becomes greater, thereby requiring a greater volume of money or near-money relative to GDP to support the process. However, in many developing countries the opposite phenomenon, termed "financial repression," is frequently observed. This is essentially a fllght from the financial intermediation system which is generally associated with an interest rate structure vhf4' is Incommensurate with .nflation, risks or availability of  - vi banking services. 3/ Thus savers, fhequently faced with negative real interest rates on their deposits. may turn away from the banking system, using their savings to acqulre real assets (such as gold or livestock, or--in the open franc zone system--foreign assets), which may diminish the efficiency of the financial system In chamaeling resour_es to their most productive uses. 14. A commonly used measure of the degree of financial intermediation ("financial deepening") is M2/GDP, i.e.* money plus quasi-money (time and savings deposits). As suggested above, the level of financial Intermedic.tion I:n an economy is Important, not only because it suggests something about the ability of the financial system to offer Instruments which may be of interest to savers in the constitution of their pcrtfolios, but also because it provides some indication as to the efficiency of the system in channeling resources to the productive sector. The greater the range of financial instruments and institutions available, the more capable the system is likely to be in fulfilling these requirements, and the higher the relationship between the money stock and GDP. While there is no firmly established causality between the financial deepening ratio and the level of per capita GDP itself, research has shown that, up to a certain point, a consistent and positive correlation has tended to exist. 15. Cameroon's financial intermediation level is much below that suggested by its per capita income. This may, of course, indicate that the relationship is weak at best. In the case of Cameroon, however, the figure is equal to 19% of GDP, or less than two-thirds of the average level for developing countries of similar per capita income levels. 41 16. Quantitative work suggests that domestic as we}l as international interest rates have a very important bearing upon financial intermediation levels. The results pertaining to interest rates are particularly Important, for they indicate that the negative real interest rates prevailing in Cameroon for much of the period under study have negatively affected financial deepening and thus the efficiency of the system in mobilizing and allocating domestic savings. They also suggest that the differential between these rates and international rates--and more 3/ cf. Shaw & McKinnon (1973), who argue that negative real interest rates, along with other governmet-induced distortions in the financial system inhibit the development of financial intermediation and reduce real growth rates. 4/ Most striking is the fact that Cameroon's financial intermediation level is lower than all countries in the West African franc zone save Niger; it is even somewhat lower than the ratio for the Central African Republic.  - vii - specifically Paris rates--have had the same impact. 5/ Thus to the extent that domestic rates are kept below Paris rates, savings will flow to the latter market and the domestic financial system will remain under- developed. Given the fact that under the franc zone system the economy of Cameroon Is, and will remain hlghly open, this effect-which bears upon domestic prices,-real exchange rates and domestic monetary policy generally-is unavoidable. It ts probably exacerbated by the fact that large or expatriate firms will continue to have access to the more sophisticated financial services of the Paris market, thereby tending to stultify the development of such services, and financial intermediation generally, on the Cameroonian market. -The Informal Sector 17. Car.eroon's informal sector plays a very important role in the overall f'-gancial system and appears to be impeding the development of a more formal sector. A country's financial intermediation system may be underdeveloped for a nuKber of institutional reasons having relatively little to do with outward indicators (e.g., negative real interest rates) of financial repression. This appears to be the situation in Cameroon, where the informal system developed early in Cameroon'e history as an important source of financial developmet outside of the colonial sector. However useful the informal system may be for individuals and affinity groups, a more fo-nal system would ultlmately provide greater benefits to the typical clients of the existing system. 18. For the period studied (1979-1984), flow of funds analysis showed that the informal sector played a substantial role in the provision of credit to the domestic economy. Unfortunately, it does this under circumstances which are not particularly conducive to financial development, or, ultimately, to economic growth. 19. While the tontines are very effective in marshalling resources and generating credit within social or clan affinity groups, so that failure to repay is practically unknown (with implicit guarantees being provided by the extended family), they do so at interest rates which are several multiples of those of the formal sector, and for maturities which are seldom, if ever, in excess of one year. Segmentation between formal and informal markets, as well as among informal markets is, however, practically complete, so that there is practically no chance for term transformation, lending rates which are consistent with risk, or for financial packaging which responds to specific requirements of borrowers. 5/ Specifically, for each percentage point change in domestic real interest rates thare is a nearly identical change in the M2/GDP ratio. Similarly, a one point increase in Paris rates (domestic rates remaining unchanged) would bring about an almost identical fall in the M2/GDP ratio.  viii - (There is also no chance for borrowers from the tontines to make effectively risky investments which may be in the public interest if not that of the individual private saver.) D. The BEAC, Monetary Policy and Interest Rates 20. There is good reason to believe that the franc zone system has played a salutary role in the development of the francophone economies associated with it. Rowever, there is also growing reason to believe that certain aspects of its regulatory framevork may, particularly for the more advanced countries like Cameroon, increasingly inhibit not only private resource mobilization but also the financial development of the economy. 21. The BEAC (Banque des Ptats de l'Afrique Centrale),-since W13, and its predecessor, the Banque Centrale des Etats de l1Afrique Equatoriale et du Cameroun (1959) wern established under a philosophy reflecting the assumption that domestic savings were likely to be so limited in the extremely poor member countries that interest rates were relatively unimportant for resource mobilization. Since it was also accepted that low interest rates would stimulate investments, cheap money was seen as desirable, and the Central Bank, rather than being lender of last resort, became a co-financer of domestic investments. Moreover, because local banking systems were so rudimentary, the Central Bank, through its rediscount policy, played a role which in many ways substituted for the local banking system and the market which it was meant to service. 22. Under the regulations of the common system, the BEAC is responsible for settlng a common monetary policy for its member countries. With a common currencf, the CFA franc, being fully convertible into the French franc, the rules of operation for the BEAC zone countries are basically established to ensure that domestic demand is managed in such a way as to avoid a continuing drain on the operations account. This account, managed by the French treasury, is In effect an overdraft facility used to purchase all excess supplies of CPA francs on the foreign exchange market, and thus to ensure full convertibility. 23. An important aspect of the BEAC framework is that the purely quantitative notion of rediscount cellings is severely complicated by a number of diverse quality elements. 6/ Rediscounts in the BEAC zone are 6/ Although following the practice of the last few decades the terms "rediscount" or rediscountable are used throughout this Study, current practice is to talk of "mobilization." The difference is a technical one having to do with whether the loan documents being rediscounted are physically delivered or not. At present the loan documents are kept by the banks themselves, and credits are spoken of as being mobilized, rather than rediscounted. The practical effect, however, is identical.  - ix - articulated as to whether they are for privileged or non-privileged uses; whether loans are expected to be made to enterprises which may have exceeded individual borrowing limits; whether working capital levels are appropriate, whether the expected loan is short, medium or long-term in maturity. and whether firms have received prior authorizations (autorisations prealables) for medium-term borrowings. Accordingly, different rediscount rates and bank margins may be applied, different amounts rediscounted, or rediscounts not provided, all depending upon a complex interplay of elements. Thus the relationship between the more important quantitative targets and instruments Is partly obscured. 24. The system is, moreover, substantially complicated by the multiplicity of administratively set interest rates, of which there are 21 for borrowing, and 49 for deposits. Interest rate structure and levels are jointly determined by the BEAC--which sets base (discount) rates--and the individual Mtnisters of Finance, who set individual margins (covering banks' costs) which are added to the BEAC-determined base rates, so that ultimate borrowing and lending rates may differ between countries of the zone. With permissible margins being lower for privileged sector loans than for ordinary re-discountable loans, commercial banks are as a practical matter discouraged from making loans to sectors (such as small- and medium-scale enterprises) for whom lending costs and risks are inherently higher. 25. Fundamental changes in the overal' system are required. First, it has been noted that nominal interest rates much below international (Paris) rates have constrained not only resource mobilization but also the financial deepening of the system overall. 7/ Thus to the extent possible within the BEAC system--i.e., via the margins which are under the purview of the Mlnisters of Finance-Cameroon's borrowing and lending rates should be aligned with international rates. Second, a sharp compression in the number of borrowing and lending rates is also required with a simultaneous upward adjustment in certain banking margins to remove the disincentive effect of low nargins. 26. Simultaneously, the overall system should be sioplified, among others, in order to increase its transparency. In particular, rediscount determinations by the Central Bank should bear uniquely upon determining the overall money supply. To this end, the category of non-rediscountable loans would be abolished so that in principle all loans would be rediscountable. Thus quality-based considerations would be removed from 7/ It must be noted that nominal rates are important in the present context. While negative real rates have probably contributed to financial repression, if nominal rates are much out of line with Paris neates (whatever the domestic rate of inflation), this will induce capital flows which may either inhibit financial intermediation or be destabilizing.  -x- the rediscount process itself. 8/ Moreover, the alignment of domestic interest rates on International rates, apart from being indicated from a resource mobilization/financial deepening viewpoint, would also have the virtue of insulating the system from the imnact of interest-rate induced capital flows, thereby simplifying demand- magement through rediscount controls and permitting a-better assessment of the impact of the new system on the domestic money supply. In this context, it must be understood that real control over the domestic money supply is difficult in an open system, and that policy instruments should be strictly reduced to a level commensurate with achievable policy targets in order to permit a ideally transparent relationship between policy measures and domestic money and credit. 27. Ideally, interest rates should be left to find their own level consistent with Central Bank monetary policy; these would In practice probably closely approximate international levels. However, with the highly oligopolized commercial banking sector, some regulation of domestic interest rates based upon international rates seems advisable until a greater degree of competition is introduced into the sector. E. *oney and Capital Markets Commercial Banking 28. A fundamental fact about the Cameroonian commercial banking sector is its state of near-insolvency. Partly in recognition of this fact the Treasury has maintained large deposits with the banking system; together with the deposits of the quasi-public organizations these have averaged roughly 30% of total deposits. 9/ While the basic state of the commercial banking sector'& balance sheets with respect to its non- performing assets remains murky, there is little doubt that if bad debts were written off as normally required by international standards the sector's capital base would be fully eradicated: total capitalization of these banks (excluding the already bankrupt Cameroon Bank) according to their own balance sheets was about CFAF 30 billion; the volume of reputed bad debts is CFAF 120 billion. 8/ Quality control would be assured through other measures, namely those based on permissible ratios for particular types of lending. However, banks would have greater responsibility (with increased Ministry of Finance supervision) of determining the quality of their portfolios themselves. 9/ This does not, however, include the deposits of the public enterprise sector, which would bring the total closer to 50% of aggregate bank deposits.  - xi - 29. In any event, public sector deposits have not necessarily rectified matters with respect to banks' overall balance sheets, and the banks have clearly become more risk-averse with their weakened capital base. The deposits are recognized as having contributed to banks' liquidity, however and the banks in fact perceive themselves as being excessively liquid. Irdeed, loan to deposit ratios have fallen sharply and banks' foreign asset holdings have shot up since 1980, probably as virtually risk-free loans to parent banks have become increasingly preferred to riskier domestic loans. Thus while such assets were CFAF 1.4 billion ln January, 1980, by May, 1985 they amounted to CPAF 190.1 billion despite a relative narrowing of interest-rate differentials vis-a-vis the Paris interbank rate. 10/ 30. Another source of weakness of the commercial banks is the exposure they face through making term-loans sometimes considerably in excess of term resources, thus among others making them reticent to make term loans of increasing risk to them. An Important part of the overall problem is their obligation to use term funds to purchase five-year bonds in support of the activities of SNI, the national investment corporation. Because the latter has stopped its new lending activities itself, these resources are re-deposited with the commercial banks at considerably higher rates and at much shorter maturities, thereby not only taxing the banks through a form of seigniorage at a time of weakness but also potentially contributing to their disinclination to make term loans. 31. Few of the commercial banks have the expertise to appraise, service or monitor loans to SMEs. They are legally obliged to devote 10% of their portfolio to SME's but this obligation is honored in the breach, largely because it is virtually impossible for commercial banks to lend to this risky sector at margins fixed at 2.5-3.5%. Guarantees for SME loans have not helped much for legal and institutional reasons. 32. Also, the structure of taxation of oank operations is highly dysfunctional. Not only are banks taxed to support the activities of SNI and FOGAPE (the loan guarantee agency recently converted into a development bank largely because of the failure of the BCD), but they are also obliged to add taxes to margins which themselves are often inadequate to cover the costs of servicing their loans. This tends to contribute significantly to financial repression. If these taxes were removed while permitting banks to increase the relevant margins roughly equivalently-particularly those for the SMHs-lending which is now unprofitable could be made attractive, 10/ Government and quasi-Government deposits ordinarily earn an interest rate of 10%. It seems to be tacitly understood that these deposits may be placed in virtually risk-free foreign deposits (with a spread of perhaps 2-3%), essentially as a measure toward reconstituting bank profits. In any event, although having the regulatory apparatus to do so, the BEAC has not put an end to this practice.  - xii - and resources would be channeled more nearly In line with their ultimate productivity. Sitilarly, taxes on most deposit instruments unnecessarily reduce the rateb of return to such deposits and discourage financial resource-mobilization. While Cameroonian ittereat rates must In any circumstance be roughly in line with itinal international (Paris) rates. the removal of such taxes would permit rates more nearly positive In real terms. l1/ Estimates are that the removal of taxes on financial instruments or loans would result In less than a one percent reduction of total tax revenues. 33. Neither the Cameroonian Development Bank (BCD) nor the loan guarantee organization, FOGAPE, is able to reach the SME market effectively. BCD's portfolio is more heavily weighted to term-lending than that of the commercial banks; a greater portion of their lending is also slanted toward the SHE's. However, BCD's lending has been almost completely stagnant for the last two years and It is increasingly orienting its portfolio toward the virtually risk-free agriculture export sector, lending which has few benefits to the producers and which probably substitutes for foreign financlng. About 602 of FOGAPE's guarantees benefited the wholly Government-owned BCD and FOGAPE has since been transformed into a development banking institution itself, leaving BCD a poorly-managed virtual shell. But FOGAPE, too, suffers from many of the deficiencies of BCD, most notably a lack of qualified manpower. 34. Grass-roots Institutions do exist in Cameroon, but linkages between them and apex institutions which could strengthen their effectiveness are lacking. While the informal financial market, largely consisting of tontines, has its weaknesses, particularly in that it contributes little to the financial unity of the country, the tontines play an important role in providing the social and institutional framework for guai mteeing the lending of the resources which they collect. While it will be difficult-although not necessarily impossible-for the formal financial system to construct direct links to the tontines themselves, there are a number of features which could well be emulated. Thus in many developing countries (including 19th century Europe) one finds chains of popular credit institutions which have some of the attributes of the tontines in their closeness to the ultimate borrowers, whether these are professional associations (whose members are most likely to be able to evaluate loan proposals for related enterprises) or other types of credit 11/ Under the current BEAC system, real rates for the most part winl have to be adjusted via changes in the inflation rate rather than the nomiual rate (assuming the absence of taxation of financial instruments) since nominal interest rates above international rates could well result in real appreciation of the exchange rate by inducing net capital inflows which in turn would raise the internal price level. This effect has been documented for Cameroon (see Annex, Ch. 2)  - xiii - unions based on affinity groups, much as the tontines, Popular credit institutions in Cameroon are usually independeut and rarely have financial or tutelary associations with apex institutions. Not only are they thus dependent upon their own deposits, but frequently they have little access to the kind of-assistance or information required if they are to participate In the modern economy. In view of this, BCD/FOGAPE could probably most usefully function principally as suppliers of funds and technical assistance to the grass-roots institutions; their resources would, in turn come from a new non-bank financial intermediary (and not necessarily BEAC) to be discussed below. The Capital Market 35. The Cameroonian capital market can scarcely be said to exist, but this is neither for lack of term funds, nor of demand, but purely for the lack of the necessary institutional framewoik. To some extent this is due to the ready possibility of medium-term rediscounts from the BEAC, since this has to some extent made the development of term-lending facilities unnecessary. Ultimately, however, a country's financial system must develop as a separate entity to provide a range of Intermediation services far beyond those provided through Central Bank rediscounting. 36. ONCPB, CNPS, the insurance companies, the savings fund and other capital market institutions dispose of large volumes of assets which can be considered as term resources in the sense that a substantial proportion of these resources could ultimately be lent at medium and long-term, particularly if aided in this by an overarching non-bank Intermediary. However, for the most part these institutions now practice neative term transformation, in that these term resources are deposited in short-term time deposits with the commercial banks, where they must ordinarily be on- lent as short-term credits if the banks are to avoid excessively endangering their overly weak (or non-existent?) capital base. It appears desirable that the resources available from these institutions should be used, in part to capitalize, and in part as deposit base for, a new non- bank financial intermediary which would help reinforce the now very weak financial intermediation apparatus existing in Cameroon. Unlike the case with the CAA resources, strict attention would be paid to ensuring that the new intermediary's investments were financially viable: much of the weakness of the existing system is due to the commingling of subsidies with the financial system. 37. Thus in accepting deposits (together with some capital contribution) from the structurally surplus quasi-public institutions, the new intermediary institution could ensure the appropriate term transformation of resources. 12/ The new institution's operations would be 12/ If producer prices are raised substantially, as the World Bank has (Footnote Continued)  - xiv - totally free of financially unprofitable ventures or subsidies in order to avoid undermining the re-established financial system. 38. A major way in which the new institution would ensure term transformation would be through the purchase of long-term debt instruments from the commercial banks (once solvent), which, acting as-universal banking institutions could use these resources, among others, to make long- term loans, which are not rediscounted by the BEAC. The new Institution could play an Important role as well as a secondary market. either (as a separate operation from that described above) in supporting the newly- established perticipatory loans (uprgts participatifs") designed to, recapitalize the banks, or as an instrumentality of state divestiture of assets in public enterprises which could be privatized, whereby such assets could be purchased, for ultimate resale, from SNI, the state holding company. Ultimately, as the Cameroonian capital market developed, the new institution 13/ would play a major role in the gradual establishment of a capital market-"bourse des valeurs". (Footnote Continued) argued, ONCPB (the export crop marketing board) assets may face slower growth, if not decline. For this reason a substantial part of these assts should be held with the new Intermediary Institution at short- term. Should the ONCPB be phased out, the new lnstitution might borrow at term from the CAA/Treasury to the extent required. 13/ The new institution might be baptized "Fonds de Gestion de l'Epargne Nationale."  - xv - SUMMARY RECOMMENDATIONS (a) Bank Recapitalization. The most Important first step for the Government is to nove toward the recapitalization of the country's commercial banks, in vhich the Government owns a minimum share of 30%. It should proceed on the basis of independent audits by internationally recognized auditing firms. While the Government has made some important first steps in re- capitalization, overall financial reform is not possible without a settlement of this issue. In negotiating the re-capitalization of the banks with their parent organizations the Government should make clear its intentions with respect to future reforms, e.g., with respect to taxation of financial instruments, as an inducement to the parent institutions. (b) Public Finances. The Government should act to consolidate SNH resources with other oil and non-oil resources; if it wished to limit public clamor for the expenditure of oil resources it might consider doing this through the establishment of a "petroleum resources stabilization fund," according to which financial resources would be annually released from the fund/SNH to the budget according to a legally established formula based upon present and projected petroleum prices, international interest rates and changes in proven reserves. Moreover, resources from this source, as well as from foreign borrowing, should be consolidated with normal budget resources with respect to the plan. (c) An Analytical Budget. In order to make the best use of public savings, a rolling plan consolidating budgetary revenue and, ultimately, domestic and foreign savings should be drawn up. This could be done on a purely analytical basis, thereby avoiding the necessity for a formel reconstruction of the current budgeting system. The consolidated rolling plan would represent the envelope within which recurrent and capital expenditures would be cast, with capital expenditures essentially representing the residual once current expenditures-themselves linked with previous years' capital expenditures-were budgeted for. Part of budget resources would be the payments from the petroleum resources stabilization fund. (d) Taxation, Still with respect to its public finances, the Goverament should consider tax reductions as a meane of (a) transferring public savings to private investors, particularly with respect to agricultural prices, and (b) improving the operation of the financial markets in mobilizing and allocation resources, specifically through the reductions of the taxes on borrowing (TDC and ICAY' or on lending, (TPCRH). The abolition of these taxes would cobt the budget less than one percent of total revenues.  -xvi- (e) Establishment of the CAM. In setting up the CAA the Government should immediately proceed to establish the extent of its mandate, and determine its strategies and priorities, particularly with respect to establishing a borrowing strategy which would manage the exploitation of the dwindling petroleum assets (either in the ground or as converted into financial assets) along with other foreign and domestic assets and liabilities. Because the new institution would be oriented towards activities such as assisting in the recapitalization of the banks its aspets would largely be those of the eentral government, and - of the quasi-public bodies such as the ONCL'B. As is the case in the Ivory Coast, this aspect of the CAA's activities should be kept separate from those involving Government debt management. (f) Changes in the BEAC System. 14/ In order to mitigate the constraints of the BEAC system on its financial resource mobilization and deepening, Cameroon should (i) align its interest rate structure on world rates; (ii) make major changes in the rules under which it works within the system, to wit: abolish the distinction between rediscountable and non- rediscountable assets and make all loans rediscountable in principle, and move to global rentraints on credit through an "encadrement de credit" system; and (iii) implement the current de lure reserve requirement, enacted in 1977 but not yet applied, in a way which is consistent with the system of encadrement de credit. (g) The Establishment of a "Fonds de Gestion de l'Epargne Nationale. The new Fonds would be a multi-purpose non-bank financial iutermediary which might be run by an official board of directors including Ministers of Finance and Plan and National Governor of the BEAC which would set its priorities and strategies. 15/ 14/ The proposed changes have been designed to be consistent with the overall operating rules of the BEAC and within the authority of the individual members (e.g., Cameroon) to carry out. However, the proposed changes would probably have zone-wide implications. In this context, proposals put forward internally for an interbank money market now before the BEAC Board of Directors can only be considered for the zone as a whole. 15/ In any event, the Fonds should enjoy a considerable degree of autonomy from the Government. Thus there may be some argument for the exercise of its control through a supervisory body (Commission de Surveillance, rather than a Board of Directors) which might be broadly representative of the economic interests of the country, thus possibly (Footnote Continued)  - xvii - Increasingly, the Fonds could substitute for the BEAC in its operations. It would use the financial resources of quasi-public surplus agencies, such as the ONCPB and the CNPS, as well as those from some private institutions (e.g., insurance companies) and transform these into term resources through long-term lending to the commercial banks. It would also function as a secondary market, particularly for SNI-centered operations (such as providing the resources for purchasing financial assets associated with the privatization of a part of the SNI portfolio). It might also support,-through its resources, the development of a network of grass-roots lending operations. In any event, the full range of Fonds interventions in the Cameroonian financial sector needs to be articulated with considerable care. Should the new institution come under the aegis of the CAA, its capital market functions should be kept organically separate from other functions - such as debt management and Treasury portfolio management, which may be assigned to it. (h) The SNI. Consistent with the public enterprise rehabilitation program, the SNI would most appropriately become a manager of enterprises selected to remain in the Government's portfolio; others would gradually be privatized (or liquidated) through the support of the Fonds. The bons d'equipement supporting SNI current operations would be retired at maturity; any new financial resources would be borrowed at market rates, possibly from the Fonds. (i) BCD/FOGAPE/FONADER and the SME-Rural Credit Network. BCD and FOGAPE should ultimately be merged, with their manpower resources being used to develop and support a new and/or expanded network of grass-roots institutions providing SME credits; FONADER might play the same role with respect to rural credit (although some of the institutions in question would doubtlessly provide credits to both SME's and agricultural producers). There would be two levels of intermediation: (i) the Fonds. which would rediscount the loans of BCD/FOGAPE/PONADER 16/; and (ii) the latter. which, along with self-generated savlngs, would be a source of funds to the grass-roots institutions. (Footnote Continued) including elected representatives, higher level civil servants and representatives of the private sector. 16/ Here an as yet unresolved question arises as to whether the Fonds loans would be rediscouncable with the BEAC. Ultimately they probably should not be if Cameroon's financial system is to develop as it ought.  CHAPTER I PUBLIC RESOURCE MOBILIZATION AND ASSET MANAGEMENT A. Introduction 1.1 The ability of Cameroon's government to mobilize the ample resources evailable for public--and indirectly, private--investments through its budget has been closely linked to its-status as an oil exporter. However, unlike many developing countries, whether oil- exporting or not, Cameroon was able to finance virtually all of its public investments with relatively little recourse to foreign borrowing even before the oil revenue boom at the beginning of the eighties. As a result, its current debt-service ratio is below 10Z. While the growth in oil revenues was accompanied by a substantial upsurge in the public investment program in 1981, the Government has nonetheless managed to run an overall budget surplus for most years since the onset of oil exports in 1979. This basically sound position, coupled with conservative budget management policies, would have brought Cameroon relatively comfortably through the next ten years had the sudden drop in oil prices not intervened. A corollary of the new situation is th-t the Government, and its new Caisse Autonome d'Amortissement, will have to manage carefully its dwindling public savings, introducing new fiscal policies to offset the expected sharp declines in oil revenues, and developing a new borrowing strategy consonant with the new revenue situation. This matter will be addressed at length in a forthcoming CEM and will not be further treated here. B. The Mechanics of Public Resource Mobilization L.2 Cameroon's public finances are dominated by the ex*rabudgetary account ("Compte hors budget": CHB) through which roughly 15-20% of total budget revenues have been recently channeled, and which is directly managed by the Presidency. While the existence of the CHB preceded the onset of revenue flows from the oil sector, it is nonetheless the channel by which the bulk of oil revenues enter Into the expenditure stream. The CHB itself is, in effect, a transit account outside the regular budget into which oil revenues are paid as they are believed to be required to finance certain types of expenditure, mainly Investments, which account for nearly 802 of total uses of these revenues. Particularly since these expenditures bear no necessary relationship to the Plan, they have a certain ad hoc quality despite the fact that they finance about 502 of public capital expenditures. The major source of CHB revenues--there are some minor exceptions--is the earnings of the National Hydrocarbon Corporation (Sociftg Nationale des Hydrocarbures: SNH), the Government's shareholding arm in the three petroleum companies now active in Cameroon, and with which it has production-sharing agreements. It is these earnings which are paid into the CHB accounts as required, i.e., in conformity with the desires of the Presidency. 1.3 Accumulated (unrepatriated) earnings of SNH are strictly confi- dential and for the most part appear to be held as foreign assets; as sucL,  -2'- they do not enter into Cameroon's banking system or its formally reported foreign exchange reserves. The amounts of the resources in question have been kept secret in order to dampen the ardor of domestic spending constituencies; since the revenues Involved are assumed to be temporary, the Government feared that making public the amounts in question would give rise to expectations which would raise expenditures to levels unsustainable in tne post-oil era. Assuming no transfers from SNH to the CHB in 1980 and 1981 kfor which no data are avsilable) total accumulations of unrepatriated oil revenues were probably somewhat less than CFAF 150 billion, or roughly $350 million. C. Planning and the Extra-budgetary Accounts 1.4 TO a considerable extent, the ad hoc quality of CHB management clearly impedes the effective management of aggregate surpluses, whether directly budgetized, transferred to the budget via the CHB, or maintained by the SNH as unrepatriated overseas balances. Not only is it nearly impossible under the CHB regime to make a reasonable estimate of total resources available to distribute between recurrent and investment expenditures (since the total volume of resources expected to be available can only be guessed at by the relevant decision makers), but the planning process itself is essentially weakened. 1.5 The non-programmability of CHB resources is, if anything, aggravated by the fact that externally-financed investments are themselves not programmed together wich the (non-CHB) domestically-financed public investment budget. Thus there are in effect three investment budgets: (i) domestically-financed (36% of the total in 1984); (ii) CHB-financed (42% of the total, and managed by the Presidency); arA (iii) externally-financed (22%) which are separately programmed and developed, essentially by the Ministry of Finance. This means that carefully drawn up plans based upon available factors of production and integrating the intersectoral relationships embodied in models now applied by the Ministry of Planning can be undermined by ad hoc decisions taken outside the nexus of the Plan itself. 1.6 Moreover, as is the case in many other developing countries, Cameroon's institutional structure is poorly adapted to strike an appropriate balance between capital expenditures financed out of public (or foreign) savings and the recurrent expenditures required, among others, to maintain the capital infrastructures financed by these resources. Thus--and again it is a commonplace among developing countries (and not only those)-that the use of public resources may not be as efficient as might be the case because there is no built-in incentive, such as the need to maximize profits on borrowed resources, which will ensure the appropriate maintenance of productive capital stock. This suggests that the country striving to mobilize domestic resources should make a special effort (a) to ensure that public expenditures are well-planned; and (b) to ascertain that the mix between current and capital expenditures Is sufficient to maintain the country's capital stock. Current procedures based upon the use of the COB, particularly given the lack of any rolling  -3- plan incorporating the recurrent expenditure implications of public investments, militate against the effective execution of both points. This is compounded by separate programing procedures for externally-financed investments. D. Government Savings Performance 1,7 In sum, while the CHB process appears to introduce dyafunctional elements onto the overall budget process, the GoverDment's general perfor- mance in the area of public resource mobilization has been relatively good. Moreover, while there was a considerable upsurge In public investments early in the period to absorb these resources, the Government still managed to run overall surpluses, and at the same time applied considerable modera- tion, following the initial upsurge, in its public investments policy. Although recurrent expenditures have shown a tendency to outstrip GDP growth, central government expenditures-were 21% of GDP in 1984, which compares very favorably with the 30.4% average for oil-exporting developing countries as well as with the 25.5% average for middle-income developing countries. While there is room for individual economies--the area of public enterprises comes to mind--it is in fact difficult to isolate obvious areas where the Government should be reducing expenditures, particularly in the area of public investments. E. Caisse Autonome d'Amortissement 1.8 The Caisse Autonome d'Amortissement (CAA) is a newly created institution in Cameroon which has its counterparts in many other franco- phone African countries. It was established by decree on August 28, 1985, with the objective, among others, of expanding the functions of the Public Debt Directorate in the Ministry of Finance. This latter has been respon- sible for maintaining public debt accounts pertaining to disbursements and debt service payments, for public domestic and foreign debt. Record- keeping has been poor, although some progress has been made recently. The conversion of the Directorate into a CAA will give the new organization greater independence in overall debt management, particularly as the CAA will, as an independent public establishment, have its own budget. The latter may be financed by its own public borrowings, direct budgetary allocations from the Treasury, and by other levies and taxes, as yet undefined. From these resources it is to ensure debt service payments as required. It will also be respor3dible for articulating a borrowing strategy by the Government, as well as carrying out other pertinent studies (particularly with respect to borrowing guarantees by the Government) and may even itself be involved in the financing of priority development projects. Perhaps most importantly in this coatext--the relevant decree is not entirely explicit-it may be involved in the management of Treasury assets. In effect, this would involve a duplication of effort with respect to what is essentially a Treasury function (although at present not very well carried out)--the management of government finances. However, it is not entirtly inappropriate that the CAA be called upon to carry out certain asset management functions. In particular, this would permit the fusion of the functions of managing both debt and financial assets, and so permit the  -4- articulation of an overall borrowing strategy in light of the probable Government surpluses. Moreover, its status as an autonomous public entity (rather than a Department of the Ministry of Finance) will permit it greater flexibility in using greatly needed outside assistance in carrying out this particular task. F. Public Liquidity: Quasi-governmental Bodies 1,9 With the Central government having been in a continuing state of (excess) liquidity, borrowings from the domestic market (as opposed to foreign borrowings financing domestic Investments) 1/ have been limited. Thus the Government's Impact on the domestic financial market has been via its surpluses, rather than via deficit financing found in most other modern states. While this has to a certain extent impeded the development of a financial system based upon an array of public debt instruments, it has also meant that many of the proolems witnessed in-other countries of the franc zone have been avoided in Cameroon. In these countries, more fre- quently member countries of the West African franc zone (UMOA), restric- tions on domestic borrowing by governments have brought unsustainable external debt burdens and a basic corruption of financial systems. This has occurred as governments incur large payments arrears (which ultimately must be financed by the domestic banking system), are forced to plunder postal savings and checking systems, and to shift essentially governmental functions--in particular, subsidy payments-to quasi-governmental bodies and public enterprises. The latter, not facing the same borrowing restrictions as the central government, then proceed to overwhelm the domestic banking sector under circumstances where the interest-rate rationing function is irrelevant, since interest charges are simply compounded without payment. The result is a virtual preemption by the state of private financial circuits. 1.10 No such problems have been witnessed in Cameroon. Not only have public surpluses supported the liquidity of the domestic financial system, but the major quasi-governmental bodies are themselves in considerable surplus, the proceeds of which may be deposited with the Treasury or with domestic financial institutions. Although there is a number of such bodies 1/ While disbursements from net new foreign financing have been larger than amortization payments (interest-payments are already included in the budget projections), these have, in the aggregate, been rather small. Moreover, our current estimates suggest that the increases in CFAF-denominated debt service payments arising from these new borrowings will be offset by the decline in the dollar, the fact that some loans are being prepaid, and the apparent associated stretching out of grace periods. Thus the Government's surplus figures would not change much if the foreign sector were explicitly taken into account. Given the relatively minor magnitude of the amounts in question this was not felt to be cost-effective.  -5- which in one way or another collect taxes, earmarked or otherwise, the most important among them are the commodities marketing board (Office ilational de la Commercialisation de Froduits de Base: ONCPB) and the social insurance fund (Caisse Nationals de Pr4voyance Sociale: CNPS). 2/ SNH might be considered to be In the same category, but we have considered it as an essential part of the Central government whose resources should be managed by the Treasury, or eventually the CAA. The municipalities, themselves generally in surplus in the aggregate, are In the same category. 1.11 The ONCPB is perhaps the most important of the above. Established in 1978, it has two principal functions; price stabilization and marketing of export crops. Although under the control of the Ministry of Commerce and Industry, ONCPB has legal and financilu autonomy. Its operating revenues come from two sources, levies on sales made by authorized private traders and the proceeds of sales made directly by ONCPB's agents. 31 These are, in essence, export taxes, the burden of which is borne by the producers. Net revenues after operating expenses are split evenly between two reseTve funds (which are distinguished for accounting reasons only; the actual funds are merged for purposes of asset management). 1.12 The first of these ONCPB funds Is the price stabilization fund; the second is the "free" reserve fund, which is used to finance investments and subsidies in the agricultural/rural developwent sector. This mandate is conceived broadly; apart from equity investments in tha ailing public enterprise sector not very directly associated with agriculture totalling CFAF 14.7 billion, the fund is also used to finance certain public invest- ments not inscribed in the Plan. Even with such expenditures, considerable liquidity has been built up, so that ONCPB had, as of June 30, 1985, a total of CFAF 98.5 billion of accumulated liquid assets. Despite its legal obligation to place the total of its liquid resources with the Treasury, as of that date, ONCPB maintained CFAF 60 billion of its liquid assets as term deposits with commercial banks, apparently because it earns no interest on its Treasury accounts. 2/ To some extent this selection of institutions is quite arbitrary. Although the CNPS is more of a non-bank financial institution it is included here because its accounts are normally consolidated with the Central government accounts. It could as easily have been included in the discussion on capital markets, where it plays (or should play) an important role. In this It does not much differ from the national housing scheme (Credit Foncier du Cameroon), the main source of whose funds is also earmarked taxes. However, its accounts would not normally be consolidated with the Central government accounts. 3/ The different systems result from a merger of anglophone (using direct agents) and francophone (private traders) systems.  6 1.13 Theie are a number of slements to be noted here. lhe first is that ONCPB is rzaking a substantial contribution to public sector liquidity. Thle. of course, essentially flows to the domestic banking system which itself is ex-cessively liquid, thereby giving rise, among other things, to capital oUtflkwt. Second, the price stabilization fund does not exist as such; apparently since the establishment of ONCPB there have been no draw downs required on reserves since domestic prices are left far below world parity levels. The Government Is on record as wishing to transfer public savings as invesetible resources to the deserviag private sector, i.e., SMEs and the rural agricultural sector. One way in which this transfer might most easily be achieved is through some reduction in current agricultural taxation; i.e., through the raising of producer prices. Third, any teaningful price stabilization fund should be "sterilized" via deposits with the Central Bank, and not with the commercial banking system. 1.14 Once ONCPB resources have financed outlays associated with agricultural development. the residuals should be used to contribute to the overall development of the financial sector, including the term- transformation of resources (which the current commercial bank deposit structure does not permit) and, ultimately financial institutions which serve priority sectors. This might best be carried out through a new non- bank financial intermediary institution (discussed further below) which would also receive resources from other collectors of savings, most notably the CNPS, but including other institutions, such as the National Savings Bank, the postal checking system, and the Housing Bank as well. (Details of this institution are provided in Chapter V). Thus ONCPB's residual assets would no longer flow to the Treasury, but would be assigned to the new institution. 1.15 Unlike the ONCPB, the CNPS is an institution which collects long- term resources whose Inflows and outflows are easily predictable, making it a valuable resource in the transfer of investible surpluses to the domestic tinancial system. particularly with respect to longer-term funds which might eventually support the development of a capital market, now virtually nonexistent in Cameroon. 41 At present these resources are "wasted" via a CFAF 38.5 billion term loan to the Government for on-lending to the public enterprise sector (there is no obligation for CNPS to place its resources with the Treasury) and CFAF 70.6 billion in sight and time-deposits with the commercial banking sector. The amounts in question represent the virtual totality of CNPS's investible resources, and in effect contribute virtually nothing to the development of a real capital market. 4/ While it is true that CNPS provided equity funding of roughly CFAF 4.5 billion to the public enterprise sector (mainly CELLUCAM, CAMSUCO), CFAF 3.8 billion has had to be written off.  -7- G. Conclusions and Recommendations 1.16 The Cameroonian government has performed well in generating public savings and In keeping public investments sufficiently in bounds to generate an increasing overall surplus. Should some satisfactory resolu- tion to the public enterprise problem be brought about, these surpluses should eventually be such as to permit the absorption of the decline in oil revenues at the end of the period without major difficulties. 1.17 In line with its stated interest, the Government should begin to develop mechanisms for the orderly and efficient transfer of investible public surpluses to the private sector. As suggested, some of this might best be carried out through well-targeted tax reductions, particularly with respect to financial instruments (see Chapter II for further discussion of this question) and agriculture producer prices. Apart from any such fiscal measures, however, the Treasury (or the new CAA) needs to define a consis- tent financial asset management strategy which includes SN! resources (as well as those from external financing) particularly since a problem of the Government's budget management system involves the non-programmed usage of SNH-financed extra-budgetary accounts. Indeed, all three separate invest- ment budgets should be amalgamated, at least on an analytical basis (as opposed to changing the legal accounting framework), to permit the establishment of a forward rolling plan including recurrent as well as capital expenditures. Simultaneously, the Government needs to consider centralizing the management of non-Treasury resources from the ONCPB, the CNPS and other collectors of savings into a new non-bank financial intermediary, which would utilize these funds for the development of the financial sector, and particularly for the term-transformation of resources, as well as other functions which will be described at greater length in Chapter V.  -8- CHAPTER II PRIVATE RESOURCE MOBILIZATION AND FINANCIAL INTERMEDIATION A. Introduction 2.1 We have seen that the Cameroonian system is reasonably effective in the mobilization of public resources through the budget; where it is less effective is in the management of such resources. As indicated in Chapter I, a large part of public savings is devoted to public investments, and with the programming of these investments being poorly Integrated as between various sub-budgets, there is considerable likelihood that t-he profile of these investments will be far from ideal. 1/ Furthermore, the management of that share of public savings which is not devoted to public capital expenditures leaves much to be desired. First, beyond their being utilized to shore up a nearly bankrupt banking system there appears to be little consistent portfolio strategy for the utilization of public finan- cial surpluses. Second, with these surpluses being divided between Trea- sury and SNH, there is little possibility of having an Integrated portfolio strategy; Treasury officials charged with portfolio management are left in the dark about SNH resources and their deployment. Finally, the surpluses of the quasi-government agencies, principally ONCPB and CNPS, play virtually no role in the development of the Cameroonian financial system: they exercise little or no function in encouraging financial deepening in the unusually "shallow" system. Thus they contribute little or nothing to financial intermediation. 2.2 This chapter provides an overview of the concept of financial intermediation and its empirical manifestations in Cameroon both as com- pared with other developing countries and In terms of Cameroon's own financial institutions. All indicators suggest an unusually low level of sophistication in Cameroon's financial intermediation svstem, particularly compared with its relatively high per capita income. A technical annex sets forth the results of quantitative analysis indicating that Cameroon's interest rate structure, both absolutely (in real terms) and in comparison with international rates, has a statistically significant bearing upon the level of financial intermediation in that country. International interest rate levels (i.e., those of the Paris financial market) are especially Important given the high degree of openness which characterizes the Cameroonian economy. 1/ In all fairness, however, any such conclusion should be supported by a careful public investment review: as a practical matter, the Cameroonian public investment budget does not appear to be encumbered by obvious "white elephants."  2.3 Cameroon, as a member of the southern tier franc zone (the BEAC, discussed at greater length in Chapter IV) to subject to a number of the same influences as its fellow member countries of the zone, and it is possible that a number of the same conclusions found below would also apply to the other members of BEAC. This Report, however, necessarily confines itself to Cameroon alone; in any event, there is no way of knowing whether the externally-imposed institutions of the BEAC will react in identical ways with the internal institutionis and economic configurations of each individual member country. Indeed, the likelihood is that they will not. B. Financial Intermediation 2.4 The relationship between private financial resource mobilization and financial intermediation is a critical one. If the level of financial intermediation (or degree of "financial deepening") is low, this will tend to inhibit the mobilization of private resources. Moreover, for a given level of private financial resource mobilization, if the intermediation system is inadequate, this will imply that the transformation of the financial resources in question will be only imperfectly translated into economic growth. Other things being equal, the lower the level of finan- cial intermediation, the greater the Impediment to economic growth: limited financial savings will rarely be channeled to their most productive uses. 2.5 It is important at this point to define some of the above con- cepts. By financial resource mobilization we mean basically financial savings as distinguished from "real" savings, or the abstention from consumption in the national accounts sense. The distinction is made: (a) because real savings are generally inelastic to the relevant policy variables (such as interest rate variations); and (b) because if private savings cannot be captured by the financial system they cannot easily be mobilized for investments. The mobilization of financial resources, on the other hand, is much more susceptible to changes in policy instruments. As suggested, financial intermediation is a notion complementary to financial resource mobilization which sheds light on the efficiency of the financial system. Often encapsulated by the summary measure M2/GDP, 2/ the level of financial intermediation reflects: (a) the range of financial instruments available to savers wishing to diversify their portfolios with respect to rates of return, risk and maturity structure; and (b) the related set of 2/ M2 is defined as currency in banks, demand deposits and time and savings deposits. This ratio is designed throughout the discussion to exclude Government (Treasury) deposits, mainly since such deposits will bear no relationship to interest rates or other relevant variables (such as income or number of banks) and their inclusion would tend to distort the relationships being studied. While there are some unresolved theoretical uncertainties concerning Treasury balances in oil-rich countries, Cameroon's M2/GDP ratio was at low levels even before the oil boom in the late 1970s.  - 10 - financial packages available to investors who seek investible resources with different mixes of equity and debt, different maturities in the debt structure and different rates of interest associated with varying degrees of risk. 2.6 All too often in developing countries, including in Francophone West Africa, the financial system offers the saver little more than bank deposits for the most part paying negative real interest rates; simultaneously, few borrowers have more than limited access to equity financing or term loans. Moreover, access to the system is frequently limited by credit rationing associated with too low interest rates. In such a system the level of financial intermediation ratio is likely to be. low, and the system "financially repressed." Savers will often opt to hold "treal" assets, e.g., real estate, livestock, gold, etc., in-preference to financial assets, 3/ and prospective investors will frequently find investible resources from the formal financial system scarce at any interest rate. On the other hand, higher intermediation levels will be found where the system comprises a number of institutions which would ultimately work to pool risks which would not be acceptable to the average depositor, transform maturities and offer different portfolio character- istics which would be attractive both to potential borrowers and potential lenders. The more "layering" of financial institutions this implies, the greater the amount of financial resources (K2) required to sustain this layering, relative to GDP. C. The Adequacy of the Cameroonian Financial System 2.7 Available evidence suggests that In the case of Cameroon finan- cial deepening in the economy is less than what it should be, particularly for the level of its per capita GDP. Thus the M2/GDP ratio is, as Table 2.1 shows, substantially below ratios for African countries of similar or lower per capita incomes, and even more so for other LDC's. Thus the Cameroonian level of 0.19 was a full one-third below the average for selected African countries of roughly similar per capita levels (or, like Kenya, economic makeup), and even lower as compared with the other LDCs shown. The comparison with other members of the franc zone (UMOA) is particularly instructive: M2/GDP for that zone as a whole was 25.1% in 1983; in that year, Cameroon's ratio was lower than that of all UMOA countries save one, which suggests that the UMOA system may be relatively 3/ Real assets in this sense of the word will include holdings of net foreign assets (which are not technically available to domestic investors through the intermediation system) and, to a certain extent, asset holdilgs In the tontines. This is discussed at greater length below.  - il - less conducive to weaknesses in financial intermediation than the BEAC system. 4/ Table 2.1 C(HPARXTIVE WBASUtES OF PINAWCIAL DEVELOPHEIT. 1983 Countxy GNP/capita 2P LB/GDP QMPGDP Cameroon 820 0.19 0.15 0.09 Other Africa Kenya 340 0.27 0.22 0.10 Nigeria 770 0.37 /a 0.26 /a 0. 15 /a C8te d'Ivoire 710 0.26 0.17 0.08 Senegal 440 0.31 lb 0.21 /b 0.08 lb Average African Sample 565 0.30 0.22 0.10 Other LDCs Morocco 760 0.44 /a 0.29 /a 0.10 /a Thailand 820 0.48 0.42 0.39 Indonesia 560 0.20 0.16 0.10 Philippines 760 0.25 0.20 0.17 Average Others 725 0.34 0.27 0.19 Overall Average 645 0.32 0.24 0.14 /a 1982 data. 7 1981 data. Source: World Development Report, IFS, and Bank staff estimates. LB and QM refer to bank liabilities and quasi-money, respectively. 2.8 A similar picture is seen with respect to bank liabilities relative to GDP. This figure is instructive because, being exclusive of currency balances, it is more clearly associated with the use of the banking system, through which intermediation must ultimately be carried out. Here, with the Cameroon ratio being .15 (as compared with other Africa at .22) this indicator of financial deepening is again more than one-third below that of its African neighbors. 2.9 Since Cameroon's economic performance has basically been of very high quality, particularly with respect to its growth performance, which 4/ This suggests that a comparative study of the BEAC and UMOA systems may well be desirable with respect to the impact of the two different franc zone systems on the financial characteristics of their member countries, particularly since the UMOA system is in many ways more advanced than the BEAC system. Such a study, however, is beyond the scope of this Report.  - 1? - was good even in the pre-oil period, it Is not particularly easy to suggest that it should be doing better in this respect, especially since some of the poorer performers (i.e., Senegal or Nigeria) have financial intermedia- tion levels considerably in excess of that of Cameroon. Or since a better performer, Indonesia, shows indicators closer to those of Cameroon. However this may be, there is considerable empirical evidence which sug- gests that relatively low level of sophistication of the Cameroonian system may well work as a constraint to growth in the years to come. Sf Before we consider the institutional failures of the Cameroonian financial system which underlie or point up the low level of financial deepening it is useful to examine some of those elements which are thought to bring about the institutional weaknesses in question. Table 2.2 : Cameroon: Real Interest Rates, 1975-82 Ex-post Real Ex-ante Ex-post Deposit Real Deposit Real Lending Rate /a Rate /b Rate Ic 1975 -7.2 -2.4 -2.4 1976 -3.5 -2.8 0.8 1977 -7.2 -3.3 -3.4 1978 -5.2 -4.3 -1.4 1979 0.1 -5.0 3.9 1980 -2.1 -5.2 2.9 1981 -4.5 -5.5 -0.4 1982 -7.9 -6.3 -3.4 /a The nominal deposit rate used here is the average on different types of deposits, and calculations of the real rate are based upon current inf"ation rates. /b Ex-ante rates are expected inflation rates estimated from a model where agents form their expectations on the 4oasis of the current CPI and inflation in the two preceding periods. /c The nominal lending rate underlying these calculations is defined as the rate on medium-term ordinary non-rediscountable loans. Current inflation rates are used for the calculations. Source: IFS, MOF, and BEAC. 2.10 Analysis of financial intermediation puts considerable emphasis on the behavior of real interest rates. These have (see Table 2.2) been for the most part negative in Cameroon. Concern has been expressed in many quarters that such interest rates discourage real savings in the national accounts sense of consumption foregone; obviously this would simultaneously tend to discourage the use of the financial system, and thereby negatively affect financial deepening, as well. However, the empirical evidence of such a relationship is weak, and for Cameroon no statistical support for it was found; analysis showed real savings to be primarily a function of income, as suggested by Keynes. This being said, a more real concern is 5/ Cf. Shaw and McUnnon (1973).  - 13 - that low or negative real rates, while not necessarily discouraglng real savings for the relevant range of rates, may tend to encourage a shift in savings from financial instruments to real assets. In the latter case, the associated "flight from the financial system" will inhibit the access of deficit units (net investors) to surplus units (net savers) and so under- mine the efficiency of the resource allocation system. 2.11 A critical element in the equation, for an open economy such as Cameroon, is foreign interest rates. With a fixed exchange rate system ruling out exchange risks, and with capital movements relatively free-both characteristics of the franc zone system--capital flows are likely to be responsive to even relatively minor disparaties in interest rates. To the degree that this Is the case (and assuming that domestic rates are lower than foreign rates, almost inevitably the case in Cameroon) foreign assets will be much like real assets in that they will be basically inaccessible to domestic borrowers& This will be shown by a low level of financial intermediation. 2.12 Financial deepening-or lack thereof-is also generally thought to be associated with institutional factors, foremost among which is the accessibility of branch banks or other types of savings outlets; obviously the presence or absence of these will have much to do with the holding of bank or savings deposits. 2.13 Empirical evidence (see annex to this Chapter) shows that there Is indeed a strong correlation between financial deepening (as measured both by M2 and bank liabilities relative to GDP) and the above phenomena. Particularly important were real interest rates, both domestic and interna- tional (Paris), where both nominal rates were deflated by the Cameroonian Inflation rate; for a Cameroonian saver operating under a fixed exchange rate system the French inflation rate is basically irrelevant. Summarizing the results, we find that a one percentage point Increase in the real interest rate in Cameroon would lead to a 0.9 point increase In the K2/GDP ratio. Conversely, and other things being equal, a one percentage point increase in real foreign rates would lead to an identical decrease in the M2/GDP ratio in Cameroon. These findings are of considerable importance since they suggest that not only do real rates in general have much to do with financial savings and the strength of the financial system, but the differential between these rates may be of equal importance. Assuming that the differential is in Paris's favor, this will tend to inhibit the deepening of the financial system and its ability to intermediate between domestic savers and investors. Should, however, the differential between Paris and Cameroon be in Cameroon's favor, quantitative evidence (see Chapter annex) shows that, by encouraging capital inflows, this would tend to lead to real exchange rate appreciation. 2.14 The latter finding is instructive since it has some unexpected policy implications. That is, to the extent that the establishment of positive real rates in Cameroon brought nominal rates in Cameroon above nominal Paris rates, this would tend to encourage capital inflows and domestic inflation (and real exchange rate appreciation). This suggests  - 14 - that, short of reducing taxes on financial instruments (discussed below), a policy adjusting interest rates (rather than, say, inflation itself) to keep them at real positive levels, may be both costly and self-defeating. 2.15 The importance of the rate differential between Paris and Cameroon with respect to financial deepening--given the openness of the Cameroonian economy toward that of France-strongly suggests the need to align Cameroonian rates on Paris rates. Apart from the fact that this would tend to Improve financial deepening, it also has important implica- tions for monetary policy, and will be discussed further in that context. 2.16 Although having little to do with interest rates, our quanti- tative work shows that the availability of bank branches has a real impact on financial deepening (see chapter annex). Table 2.3 indicates that Cameroon is relatively underdeveloped, as compared with other African countries of or near its income class, with respect to the number of bank branches per 10,000 persons. Although the sample is limited because of a lack of data, what was available shows that with a per capita income averaging 43% more than the comparator countries, the number of bank branches per 10,000 was 17% lower than the average. Table 2.3 : Number of Bank Branches per 10,000 of the Population (1977 data) Permanent Bank Country GNP/capita Branches/10.000 persons Cameroon 340 0.149 Ghana 380 O. 25 Kenya 270 0.209 Tanzania 190 0.188 Somalia 110 0.087 Average 238 0.172 Source: Ministry of Finance, World Development Report and [warteng (1982). D. The Institutional Perspective on Limits to Financial Deepening 2.17 It is useful to consider the somewhat theoretical relationship between financial deepening and "real interest rates" from a more institutional perspective in order to understand better how financial deepening has been impeded in Cameroon. Following that, we will examine how efficient the system itself is; that is, the degree to which the system--deep or shallow-delivers financial services effectively.  - 15 - 1. Bank Margins 2.18 The theoretical perspective on financial intermediation in Cameroon is pointed up by specifIc institutional elements. While low or negative real interest rates may discourage the accumulation of financial instruments, margin limitations on bauk lending may bring about the same problems. Thus if a bank using deposits to finance its non-rediscountable banking operations 6/ calculates the cost of funds at the maximum rate on six-month deposits of 12 percent, it earns margins of between -2.5 percent (for medium term privileged loans) and +1.75 percent (medium-term non-privileged loans). Apart from the fact that this will tend to discourage loans to the so-called privileged sectors (e.g., small- and medium-scale enterprise) it will also discourage commercial banks from seeking deposits except at rates very likely to be negative In real terms, so that there is a direct disincentive for banks to mobilize savings. This will be reinforced by the ability of banks-under the logic of the franc zone system to use rediscount facilities, in which case they will earn margins of 2.25 and 3.25 percent on the two types of loans, respectively. While this may work to orient lending to rediscountable uses or borrowers, at the same time it will restrict the capability of the system as a whole to mobilize resources. In any event, a margin of 2.25 percent will rarely be sufficient to cover all but the most risk-free of privileged lending (e.g., for agricultural export credits, which may well have the effect of displacing foreign lending for the same purpose), and it seems clear that this margin policy, in the guise of assisting socially-favored sectors, has quite the opposite effect. 2. The Paris/International Market 2.19 It has been suggested that one reason for the relatively low degree of financial deepening in Cameroon is the existence of the easily accessible Paris money and capital market. 71 Accordingly, domestic institutions have not developed because given easy access to Paris, this has been unnecessary. There is some empirical evidence ror this from the 6/ The concept of rediscountability will be discussed more extensively in Chapter IV, but essentially rediscountable credit includes most short- and medium-term loans except consumer loans and long-term loans. However, qualitative criteria relating to financial soundness of the operation as well as potential profitability, and quantitative criteria (via determinations as to the rediscountability of the asset) are also applied, and their outcome cannot easily be known in advance. 7/ It must, however, be noted that the Cote d'Ivoire and Senegal benefit from the same conditions of access to this market, yet their financial deepening position is more advanced than that of Cameroon, a country of higher per capita GDP.  - 16 - 1979 census on Investments in manufacturing. This shows that for firms in the modern sector having total annual sales of more than CFAF 5 million' (see appendix Table 16), somewhat over 60X of total equity capital had its origins with foreign investors, almost entirely private. With government capital financing 17% of the remainder, this left only 23% to be supplied by private domestic investors. This does not include debt, but since there is practically no formal long-term debt in the Cameroonian economy at present, (although doubtlessly considerable short-term debt is in fact rolled over), any such debt must come from abroad, most probably via parent companies. 2.20 Thus in considerable measure capital financing must come from overseas or the Government (whose own policies supporting domestic enter- prise have been little short of disastrous and which in any event has since 1983 supplied little new investment capital). Those other institutions capable of jointly playing the role of capital market (e.g., CNPS) have, for the most part practiced negative term transformation and deposited assets essentially associated with long-term liabilities In commercial banks, where interest rate differentials have tended to draw these resources back to overseas markets. Indeed, the open financial system of the BEAC zone works in such a way as to encourage banks themselves to use foreign financial end capital markets for intermediation and term- transformation, thereby reducing or discouraging local lntermediation. Thus over the six-year period between 1979 and 1985 average gross foreign assets of the commercial banks, at CFAP 45.0 billion, were roughly equal to the banks' medium and long-term borrowings (i.e., debt) of CFAF 42.3 billion over the period. (However, with increasing bank liquidity this relationship has shifted to roughly three-to-one in favor of gross foreign assets.) 3. The Tontines 2.21 The tontines, essentially social or clan affinity groups (averaging perhaps 15-20 members) which play the role of informal credit cooperatives, are responsible for a substantial degree of credit creation in the Cameroonian economy. 8/ Interest rates tend to be quite high and 8/ Tontines are the standard form of credit creation in the West and Central African economies and occur also in parts of Southeast Asia. The simplest form of operation is that by which the members make monthly payments into a revolving fund, the totality of which is withdrawn by one of the members by prior agreement each month, each according to his turn. No formal interest payments are involved. Other arrangements involve bidding for the (generally monthly) "pot," where the winner leaves in the pot the amount of the winning bid, which becomes a part of the new pot. The borrower may then no longer bid on future pots and in effect pays back the loan by continuing to (Footnote Continued)  - 17 - maturities practically always under one year. While these institutions play an extremely Important role in the Cameroonian economy, they are not directly connected with either the formal banking sector or with one another; thus they are highly segmented. Not operating through the formal banking sector they do not directly influence money supply (so that their impact is o4 velocity rather than morey itself). Even though their role is in many ways a useful one because the formal sector does not ordinarily supply the financial requirements of tontine members, It has tended to represent an impediment to the financial deepening of the Cameroonian economy. However, the importance of these institutions may also be regarded as symptomatic of the relative underdevelopment of the formal system, and will most likely dimdnish in importance as the sophistication of the financial system grows.. 4. Lack of a Non-Bank Financial Intermediary 2.22 Particularly given the lack of project evaluation capabilities, general banking, accounting and auditing skills, etc., in Cameroon (includ- ing the effective retreat of the Cameroon Llevelopment Bank from the capital market scene), there is no effectively functioning apparatus for term- transformation or risk diversification in Cameroon. As seen above, most of the important potential capital market institutions in Cameroon practice effective negative term-transformation, particularly in the absence of a non-bank intermediary which could accept deposit liabilities of all terms from institutions such as ONCPB and transform them, in tandem with banking institutions, into long-term debt, diversifying risk through this opera- tion. Similarly, there is no secondary market in Cameroon to guarantee the liquidity of primary s.zcurities and this also tends to limit the development of the market for such financial instruments. The lack of an institution performing these functions--common in other countries of Cameroon's per capita income level-probably contributes considerably to the measured shallowness of Cameroon's financial system. The introduction of an institution of this nature (discussed in detail in Chapter V) would very likely have a significant effect on strengthening Cameroon's financial system. (Footnote Continued) contribute every month until each member has t ,rrowed at least once. Since the amounts left in as discounts are combined to form new pots, so that the number of months of a cycle will be fewer than the number of participants; the last borrower may receive the pot in considerably fewer than the months in the cycle. Those who make their contributions In the first half of the cycle will be net borrowers, those contributing in the last half will be savers, whose interest payements for their savings are reflected in the shorter number of months for which they are required to make payments.  - 18 - 5. Socifta Nationale d'Investissement (SNI) 2.23 The SNI was established In 1962 as a state holding company to take equity shAres in Cameroonian enterprises thought to have difficulties in access to capital despite their expected profitability (or because of their social interest to the state). With the bulk of these enterprises in SNI's portfolio being technically bankrupt, SNI is technically bankrupt as well, having a negative net worth of CFAF 13,7 billion. However, because it continues to receive the proceeds of "bons d'Xquipement" (issued at 4.5% with five years maturity) 9I to which commercial banks a e~ obliged to sub- scribe 10% of their total assets, it Is fitnacially quite liquid: because the SNI has made no new investments since 1983, these assets are "recycled" back to the commercial banks at a 10% deposit rate. Obviously reducing overall rates of return to Investments, and given narrow margins,-this means that banks can pay all the less on average deposits, thereby inhibiting financial savings mobilization. In addition to this, the negative term transformation Involved with the re-deposit of these resources and the weakening of the commercial banks' maturity structure which this engenders, overall fitancial intermediation suffers. We will have considerably more to say on this subject below. 6. Taxes on Financial Instruments 2.24 Apart from the disguised tax which is represented by the bons d'equipement, financial intermediation in Cameroon is subjected to--and ultimately impeded by-two types of taxes: taxes on lending rates and taxes on interest income. These taxes represent income to the Treasury, but amount to less than 1% of its total tax revenues. Given the potential damage to the intermediation system, these taxes can scarcely be justified. 2.25 Two taxes are imposed on borrowers: (a) the tax on the distribution of credit (TDC); and (b) the turnover tax (imp6t sur le chiffre d'affaires: ICAI). The former is set at one percentage point, and is added to the lending rate, and the latter, also added to the lending rate, is equal to 10.998% of that rate. These taxes increase the cost of funds to investors, and represent an equivalent reduction in margins earned by the banks (discouraging lending), or tending to reduce average deposit rates paid (discouraging financial savings). Either way this works out badly for the financial system. Thus for a med4um-term, ordinary, non- rediscountable loan for which the bank receives 13.5% the borrower will pay 15.25%; yet the bank's margin will only be 1.75%, which is surely insufficient to cover the cost of processing, particularly: (a) since excessive term-transformation may put the bank at risk; (b) non- rediscountable loans are probably somewhat riskier than rediscountable loans; and (c) because 1.75% is considerably lower than the average margin required by Cameroonian commercial banks. A removal of this tax (or its 9/ This was recently raised to 7.0%, however.  - 19- replacement by an increased tax on bank profits) would permit banks to earn acceptable margins (or increase average deposit rates) and thereby expand lending where unfilled demand existed. 2.26 Savers are also heavily taxed, and all income from savings and time deposits at commercial banks is subject to two taxes: (a) the "propor- tional tax" on revenues earned on finarcial capital (la taxe proportionelle sur le revenu des capitaux mobiliers: TPRCM) of 16.5%, which is retained at source; and (b) the normal progressive income tax. Thus if a saver earning 12% on a six-month deposit is in the 33% bracket, he only receives a 6.7% after-tax rate of return; this is currently negative in real terms and an induceaent to return to the tontine, reversing financial deepening. The elimination of the TPCRM would permit a rise of about 2 percentage points in the after-tax return to the saver in question, which could obviously stimulate financial savings and some strengthening of the financial system, all at minimal cost to the Treasury. E. The Efficiency of the Financial Intermediation System 2.27 We have seen that there are a number of institutional elements- which tend tc limit financial deepening even given a low interest-rste structure which both on a priori and empirical grounds are found to Inhibit the development of the financial system. In addition, there are also important weaknesses with respect to the efficiency of the system in delivering financial services as it currently exists. 2.28 Foremost among these weaknesses is the insolvency, near- insolvency, or outright bankruptcy of the Cameroonian commercial banking system, caused, among others factors, by a substantial number of ill-ad- vised loans. 10/ Total amounts of non-performing assets are said to amount to about CFAF 120 billion, cr about seven times loss-reserves. There are doubtlessly considerably more loans which are of dubious value. In any event, it is clear that Cameroonian banks are badly undercapitalized and that this has a strongly negative Impact on the efficiency and behavior of the sector. 2.29 First, this undercapitalization has meant that only two of the ten commercial banks in Cameroon are profitable, and these only marginally so. This, in turn, has led to the Government's reducing all sight deposit rates to zero in order to increase bank profitability (thereby, of course, increasing financial repression). Second, it has meant that banks have 10/ It is often suggested that these loans, in large part made to merchants in North Cameroon, were politically motivated. It is, however, unlikely that there was any outright political coercion forcing bankers to make these loans; at most, many may have hoped to gain political favor by making them. It is universally agreed that these loans are unrecoverable.  - 20 - become much more risk-averse as potential losses have become far more costly, compared with wealth (banks' remaining net capital resources), than before. This appears to have encouraged banks to make loans only to the most risk-free borrowers--e.g., expatriate firms-or to transfer funds overseas, where banks' deposits have grown from CFAF 4.6 billion In March 1980 to CFAF 190 billion in March, 1985, probably with the tacit agreement of the BEAC and even the Government. Thus despite government efforts, for example, to encourage lending to the SME sectors, bank undercapitalization works strongly against such loans (as do clearly inadequate margins permitted the commercial banks for lending to this sector). 2.30 The Government has sought to compensate for this problem by maintaling substantial deposits with the comercial banks. Together with parastatal deposits, these now amount to about 50% of total deposits and have thus supplied a considerable amount of liquidlty to the sector. So much so, in fact, that some of this is recycled to overseas balances. Moreover, since government deposits, essentially time deposits, are remunerated at '0%, this means that there are relatively few loans where margins are sufficient to make lending out of these deposits interesting, especially in view of the fact that overseas placements are virtually risk free, particularly as compared with domestic loans. However stable these balances may be, they do not substitute for capital, so that banks remain, if anything, excessively conservative in their domestic lending policies. 2 .31 The problems In the banking sector are compounde by the fact that the Cameroonian banking structure is highly concentrated, with four of the ten banks controlling 85% of total assets. With an oligopolistic market structure, this means that the commercial banking sector must remain subject to government regulation, at least until (a) the other banks gain a greater degree of market power; and (b) the sector as a whole is adequately recapitalized. Perhaps most Importantly, this suggests that changes in the management of interest rates, while requiring some liberalization and allgnment on international rates, cannot immediately lead to full depen- dence upon the market for rate determination. F. Conclusions and Recommendations 2.32 Evidence is conclusive that Cameroon's interest-rate structure leads to financial repression: not only is this effect noted with respect to real interest rate levels taken in Isolation (and thus encouraging the accumulation of real, as opposed to financial assets), but also in comparison with the Paris market. With the average divergence in rates being so great (nearly five percentage points over the 1975-84 period), this has led to financial savings flowing to Paris. While from a balance of payments viewpoint the flows have been manageable, the differentials in  -21- question have also constrained the development of the Cameroonian financial system. 11/ 2.33 For this reason alone, Cameroonian rates should be aligned with rates on the Paris market, probably on a quarterly basis. 12/ Thus interest-rates must be a good deal more flexible than before, and while the flexibility here being urged would be required to limit speculation, it is also desirable as a means of avoiding the macro-economic destabilization which wide swings in the difference between domestic and Paris rates will very likely lead to. Although rate alignment will be an important element in improving domestic resource mobilization and financial deepening, as we will see in Chapter IV, it will also be an important adjunct to policy changes leading to an improvement in the management of monetary policy. Thus it will be a pivotal feature of any financial sector program. 2.34 A number of institutional features Inhibiting the development of the financial intermediation system were noted. One of these, the tontine system, is for the most part outside the scope of policy changes, and this will probably remain a permanent feature of the financial landscape. However (and this will be developed in further detail in Chapter V), some attempt to integrate aspects of this system with the formal financial sector appear to be justified. This would be a step toward overcoming the market segmentation noted and, to the extent feasible, would represent a definite step in strengthening the Cameroonian financial system, particu- larly with respect to its ability in delivering investible resources to sectors which the Government wishes to favor. Moreover, an overriding concern would seem to be the development of a new, non-bank financial intermediary, now conspicuously lacking in Cameroon, which might perform term-transformation (or reverse the tendency toward negative term- transformation) and perform a greater degree of risk diversification and secondary market support for new or existing financial instruments. 2.35 One of the most promising possibilities for Improving financial deepening is the adjustment of taxes on financial instruments: these tend to discourage both financial savings and limit the efficacy of the system in delivering available resources to where the return is greatest. They reduce the rates of return to financial savings, encourage shifts to the Paris market, reduce the incentives to banks to solicit deposits, represent 11/ To the extent that all Cameroonian investors had access to the Paris market for loans (or "packages" of loans involving a range of portfolio instruments) this would involve no welfare loss. Evidence suggests, however, that access by Cameroonian investors is not equal to that of expatriate-run enterprises. 12/ It should be noted that this can be achieved within the framework of the BEAC system by adjusting margins over BEAC-set base rates, a prerogrative of the Ministers of Finance.  - 22 - a wedge between borrowing and lending rates (and thus distort the "true" costs of capital). Moreover, to the extent that the difficulties inherent in the Cameroonian financial system result from inadequate margins (per- versely, mainly for socially desirable lending), these also inhibit bank lending, at least domestically. Particularly in view of the fact that their return to the Treasury is minimal, these taxes should be sharply reduced or better, eliminated altogether.  - 23 - ANNEX TO CHAPTER II An Empirical Analysis of Financial De.epenipg In Cameroon 1. This annex describes and explains the empirical analysis of private financial savings in Cameroon as summarized In Chapter II. It to shown through an econometric analysis of the determinants of private financial savings that increases in permanent income, in the domestic real interest rate or in the accessibility of commercial banks (and improvements in the quality of services they offer) have a positive effect of financial deepening. On the other hand, an increase in the differential between domestic and foreign interest rates leads to a decrease in financial deepenirg. 2. The analytical model, based upon work of Tobin (1982), McKinnon (1973) and Shaw (1973), from which the estimated equations are derived is as follows: (1) J.-L(Y,d,f,r,x), (2) C=C(y,d,f,r,x), (3) F-F(Y,d,f,r,x,), (4) V=V(Y,d,f,r,x), (5) T-L+C+F+V, (6) x-x(g,n), where L, C, F, V, and T are the demands for the liabilities of domestic banks, currency, foreign assets, real assets and total assets respectively. All of these variables are expressed as ratios to GDP. The variables Y, d, f, r, x, g, and n are permanent Income, the real Interest rate on domestic deposits, the real return on foreign assets, the return on real assets, a variable measuring the quality of services offered by domestic commercial banks, the ratio of government to private deposits at commercial banks and the number of commercial banks respectively. The system as defined implies that the change in total assets (T) is equal to private savings in the national accounts sense, i.e., non-consumption. 3. Following Tobin, the demand for any asset is a function of income, the real rate of return on that asset and the rates of return on other alterna- tive assets. An increase in income or in the real rate of return of an asset is expected to increase the demand for that asset, while an increase in the rates of return of its substitutes decreases the demand. It is also assumed here that improvements in the quality of services offered to bank customers will lead to an increase in the demand for domestic financial assets. Equation (6) postulates that the quality of services offered by commercial banks is a function of the number of banks operating In Cameroon and of the relative importance of government deposits in their portfolio. The rise in competition which can be associated with an increase in the number of banks can be expected to lead to better services. The effect of government deposits on the quality of services is ambiguous. By increasing  - 24 - bank's liquidity and hence their ability to make loans, government deposits exert a positive effect. On the other hand, if at the fixed margins banks find further lending unprofitable, this increase in liquidity would reduce their incentive to mobilize private deposits and thus have a negative impact on financial development. 4. The impact of changes in these variables on total private savings, however, is not as clear. For example, a rise in the return on domestic financial assets will lead to an increase in L but also to a fall in C, V, and F. Thus, the effect of such changes on T will be ambiguous. The same is true for increases in f and r which will lead to a rise in F and V respectively but also to a fall in the demand for other assets. On the other hand, an increase in income will lead to a rise in the demand for all types of assets; hence total private saving will also increase. 5. The estimated equations explaining financial deepening in Cameroon, obtained by substituting (6) into (1) and (2), are, after correcting for serial correlation: (7) 1 - -7.8 + l.l*y + .03*d - .04*f + .08*r - .05*g +.04*n (-9.7) (6.4) (4.7) (-4.5) (1.5) (-.03) (3.1) R-squared - .98 D.W. - 2.5 rho - -.49 (8) m - -4.1 + .47*y + .02*d - .03*f + .07*r + .17*g + .03*n (-4.9) (2.5) (3.2) (-2.7) (1.3) (.94) (2.3) R-squared - .96 D.W. = 2.4 rho = -.31 where 1, m, and y are the natural logarithms of domestic bank liabilities, the ratio of M2 to GDP and a measure of permanent income (a distributed lag of real per capita GDP) respectively. The real return on domestic finan- cial assets is defined as the interest rate on six-month deposits minus ex post inflation; similarly, the real return on foreign assets is defined as the yield on French government bonds deflated by ex post inflation in Cameroon. Measures of the return on real assets, l.e., time series of the rate of return on marginal private sector projects or interest rates charged in the curb markets, are not available. Therefore, as a proxy for this variable the inverse of the incremental capital output ratio is used. The variables g and n are defined as the ratio of government to private deposits in commercial banks and the number of these banks operating In Cameroon respectively. The equations were estimated using data from 1965 to 1982 (18 observations). 6. Some conclusions can be drawn from these equations. (a) Income variations affect financial development because the coefficient on permanent income is positive and statistically significant in both equations. It is also interesting to note that the income elasticity of the ratio of bank deposits to GDP is higher than that of the ratio of M2 to GDP. This is to be  - 25 - expected because M2 includes currency which should decline relative to demand and time deposits as income grows. 13/ (b) Both measures of financial development are sensitive to changes in the real deposit rate--the coefficient on d is positive and statistically significant in both equations. The elasticity of the two measures of financial deepening with respect to real deposit rates are such that a one point rise in the real interest rate would increase the ratio of bank deposits to GDP by one point and the ratio of N2 to GDP by 0.9 points all other variables remaining constant. It should be understood that these estimates are subject to the errors typical of econometric analysis and should be taken with caution. (c) Moreover, the elasticity of the two measures of financial deepening with respect to foreign interest rates are such that a one point increase in foreign interest rates would decrease the ratio of bank deposits to GDP by one point and the ratio of M2 to GDP by 0.9 points, all other variables remaining constant. Again, as explained above, these figures should be taken with caution. (d) Government deposits with the banking system do not seem to have a measurable and consistent impact on financial deepening, at least as measured by these two equations. On the other hand, the coefficient on the number of banks is positive and statistically significant in both equations. This seems to indicate that the quality of services provided by banks is an important determinant of financial development. However, since the number of banks in Cameroon is positively correlated with the creation of BEAC in 1973, It is not clear whether the apparent improvement in bank services is a result of increased competition as originally postulated or simply a result of the new institutional structure that was more conducive to financial development. In either case, the evidence indicates that the structure of the banking system is an important determinant of financial development. 7. The equations above indicate that the demand for domestic financial assets is negatively correlated with the return on French assets. This indicates that despite some exchange controls, private capital movements between Cameroon and France do exist and seem to respond to profit incen- tives. Public authorities, therefore, should be cautious in implementing changes in their financial (interest rate) policies as such changes would 13/ Obviously deposits will in any event continue to grow, and especially as branch banking develops (an area where Cameroon is relatively weak, cf. para. 2.16) in a situation where real interest rates are oitly modestly negative.  - 26 - affect the level of net foreign Inflows which may be destabilizing in the short-run. 8. In order to further clarify the effects of interest rate policy on short-run stability, the relationship between those policies and the real exchange rate was analyzed. The real exchange rate In this context Is defined as the purchasing parity real exchange rate vis-&-vis France, which, since the nominal rate is fixed, simply becomes the ratio of the French to the Cameroonian CPI. The real exchange rate is postulated to be a function of flows of net foreign assets, the difference between the domestic and foreign rates of growth of real GDP and the terms of trade (Edwards, 1985). Net foreign assets are in turn assumed to be a function of the interest rate differential between France and Cameroon. 9. This reduced form equation, after correcting for serial correlation was found to be: e - 0.08 + 0.08 i - 0.00 gd - 0.04 t (0.19) (2.5) (-1.1) (-0.46) r-squared - 0.73 D.W. = 1.4 rho = 0.83 where e, i and t are respectively, the natural logarithms of the real exchange rate, the interest rate differential (foreign minus domestic rate) and the barter terms of trade, and gd is the difference between real growth rates (domestic minus foreign). Since the coefficient on i is positive and statistically significant, a fall in domestic relative to international Interest rates leads to a decrease in net capital outflows and hence to a real depreciation. This result provides further evidence to support our previous conclusion that the demand for domestic assets is sensitive to changes in international interest rates.  - 27 - CHAPTER III FLOWS OF FUNDS AND THE CAMEROONIAN FINANCIAL SYSTEM A. Introduction 3.1 This chapter provides an overview of the process by which Investable resources are allocated through the Cameroonian financial system to ultimate users. The technique used, flow of funds analysis, permits an identification of the ultimate sources and uses of funds, and allows an overview of the major Intermediation channels through which the financial system allocates funds from surplus sectors and institutions to deficit ones. Thus the flow of funds analysis provides an accounting of all major financial transactions In the Cameroonian economy showing the sources-and uses of funds among firms, households, the banking system, the public sector and external flows. This analysis can serve as a quantitative framework for the description and analysis of the diverse financial markets and for the interrelationships between monetary and fiscal policies described in greater detail elsewhere in this Report. 3.2 The flow of funds is portrayed by means of a series of f-.nancial social accounting matrices (FSAMs) (described in Appendix 1) which detail the distribution of real and financial resources among the different production activities, agents and institutions in the Cameroonian economy during a given year. This approach not only provides an accounting frame- work for financial transactions, as does the more traditional flow of funds accounts, but it also extends this framework to the real side of the economy. The approach therefore allows us to capture more exactly the interrelationships between the real and monetary sides of the economy. This is particularly important for Cameroon where public sector funds are not only a very large share of total real resources, but are also an important source of liquidity to the banking system. Another advantage of the FSAM is that it necessarily assures consistency between real and monetary data as with, for example, monetary surveys and national accounts. This allows a departure from dependence upon published data and permits the estimation of unknown flows as a residual. 3.3 It should be understood that this flow of funds is based on data which in many cases are incomplete and/or Inconsistent. Although the FSAM methodology to some extent allows us to correct these deficiencies, a number of judgmental factors necessarily enter into the analytical process (see Appendix 1). In particular, the matrices which are central to the analysis can show us how funds are channelled from surplus to deficit sectors of the Cameroonian economy and provide a quantitative framework of the diverse financial markets as they manifest themselves in that economy. The FSAM methodology in question Is based upon a series of Interrelated accounts for each of the productive sectors, economic agents and institu- tions which follow the principle of double entry bookkeeping ensuring that for each account expenditures will equal revenues. The advantages of this approach are that consistency between national accounts data and financial  - 28 - data is assured. Moreover, from an analytical point of view the FSAM approach allows the calculation of unknown flows as a residual, showing the implications across accounts of different assumptions with respect to the various magnitudes in question. In addition, the FSAMs permit a clear overview of the interrelationships between the real and monetary sides of economy. B. The Savings-Investment Process in Cameroon. 1. Savings, Investment and Institut-ional Financing Needs 3.4 The advent of major oil revenues by 1981 11 changed the character of the Cameroonlan economy both in terms of investment and financial flows. Investment has grown from approximately CFAF 85 billion to CFAF 760 billion -between 1973 and 1984, or from 20X to 25Z of GDP. Prior to 1979, Cameroon had moderate deficits on current account; gross national savings (GNS) grew apace with gross domestic investment. Between 1979 to 1981, GNS did not grow as rapidly as investment, and as a result the current account deficit increased substantially. The advent of oil revenues in 1981 reversed this trend; GNS Increased at a faster rate so that by 1984 Cameroon was a net provider of savings to the rest of the world. Table 3.1 : Total Savinas and Its Distribution (billion CFAF and percent) Fiscal House- Govern- Foreign Year GUS holds Firms ment Savings 1980 240.8 46,8 113.1 80.9 81.3 100.0% 19.4% 47.0% 33.6% 1981 390.7 72.7 193.5 124.5 108.6 100.0% 18.6% 49.5% 31.9% 1982 461.7 90.1 181.3 190.3 71.7 100.0% 19.5% 39.3% 41.2% 1983 612.0 119.6 215.4 277.0 39.6 100.0% 19.5% 35.2% 45.3% 1984 859.5 190.0 334.0 335.5 -96.2 100.0% 22.1% 38.9% 39.0% Note: The firm sector includes public enterprises. Source: The FSAMs for Cameroon. 11 All years refer to fiscal years, i.e., from July to June.  - 29 - 3.5 There has been a marked.change in the sources of savings during these years; Table 3.1 shows the decline In the importance of net foreign savings (i.e., from abroad) and private savings (households + firms) and the relative increase in government savings which took place over the period. As explained above, the savings from the rest of the world (RoW) have not only declined In relative terms over this time period but Its absolute value has decreased as well, so that by 1984 Cameroon became a net provider of savings to the RoW. Particularly notable is the fact that the advent of oil had the effect of increasing the relative contribution of the Government to GNS by 10 percentage points, from 32% to 42X. The proportion of GNS provided by the private sector dropped commensurately from 68% to 58%, with all of thls relative decline originating in retained earnings of firms. 3.6 There are some marked differences in saving patterns between Cameroon and C6te d'Ivoire, a Sub-Saharan African country with a comparable level of GDP per capita. First, Cameroon's overall savings effort has been stronger than that of Cote d'lvoire as measured by GNS as a proportion of GDP. This was already the case in the pre-oil era because,the enterprise sector seems to have been able to generate a higher proportion of savings in Cameroon as compared with C8te d'Ivoire; this difference has, however, diminished during the oil era (Table 3.2). After the onset of oil revenues, all three institutional sectors in Cameroon were contributing more to savings as a percent of GDP than in C8te d'Ivoire. Finally, because of the apparent strength of Its domestic enterprise sector, government in Cameroon contributes relatively less to GNS than is the case in Cote d'Ivoire, although differences have narrowed since oil production began. The Government's share in C8te d'Ivoire is probably artificially high because it reflects the commodity boom taking place during the years in which the comparisons were made.  - 30 - Table 3.2 5 GNS and Its,Distribution in Cameroon and Cfte dILvoire (percent) Cameroon Cote d'Ivoire 1980-81 1982-84 1974-76 GNS (% of GDP) 19.8 .24.6 15.0 of which: Households (6 of GVP> 3.8 5,0 3.4 Firms (% of GDP) 9.6 9.3 3.6 Government (% of GDP) 6.5 10.3 8.1 DISTRIBUTION OF GNS Households (% of GDP) 19.0 20.4 22.4 Firms (% of CDP) 48.3 37.8 23.7 Government (% of GDP) 32.7 41.8 53.9 Notes: GNS as a percent of GDP for CSte d'Ivoire refers to 1980. The contribution of each Institutional sector has been calculated using the distribution in 1974-76. Sources: The FSAMs for Cameroon. World Bank: Finance in the Development of C8te d'Ivoire (1981). 3.7 An alternative way of identifying the growing savings effort in Cameroon is to track the evolution of the average propensities to save of the institutional sectors. While there has been some growth in the average propensity to save of households, it Is that of the Government which has most increased most sharply, from 28.2% in 1980 to 39.7% in 1984. This reflects the fact that current expenditures of the Government have risen at a slower rate than total revenues, perhaps unsurprising in view of the very rapid growth In oil revenues over the time period, particularly given the conservative policies typically followed by Cameroonian authorities with respect to expenditure growth. 3.8 There has been no significant change in the basic pattern of the sectoral distribution of gross investment; government has remained the origin of 1/4 of total gross Investment, firms approximately of the other 3/4, with households being responsible for a negligible amount. 2/ There has been some Instability in this pattern; thus government's share between 1980-84 has varied from 20% to 28%, with compensating changes taking place in the share of the en-erprise sector. 21 Investments performed by households when acting in their capacity as entrepreneurs, e.g., a farmer building a tool shed, have been assigned to the firm sector. By assumption, only those investments performed by individuals not directly related to a productive sector, e.g., (Footnote Continued)  - 31 - 3.9 The distribution of investment and savings across institutional sectors determines the net financing needs (savings - investment) of each of these sectors, that is which sectors are in surplus and which in deficit. Table 3.3 derives these surpluses and deficits and shows households growing in importance as a net provider of funds, government net surpluses increasing very rapidly as oil revenues themselves expanded rapidly over the period compensating for the growth in public investment (gross fixed capital formation) and net financing from foreign sources declining quickly. 3/ Some of these surplus funds are being transferred overseas but the rest are being channeled through the domestic financial system to firms whose net financing needs have Increased substantially. Table 3.3 : Savings, Investment and Net Financing bY Institutional Sector (billion CFAF) Households Firms Government RoW Ne-t Net het Net Fiscal Invest- Finan- Invest- Finan- Invest- Finan- Finan- Year Savings ment cing Savings ment cing Savings ment cing cing 1980 46.8 2.5 44.3 113.1 237.8 -124.7 80.9 81.8 -0.9 81.3 1981 72.7 5.0 67.7 193.5 392.2 -198.7 124.5 101.9 22.6 108.6 1982 90.1 10.0 80.1 181.3 392.2 -210.0 190.3 131.3 59.0 71.7 1983 119.6 13.0 106.6 215.4 502.6 -287.2 277.0 136.0 141.0 39.6 1984 190.0 20.0 170.0 334.0 533.5 -199.5 335.5 209.7 125.8 -96.2 Note: The firm sector includes public enterprises. RoW = Rest of World Source: The FSAMs for Cameroon. (Footnote Continued) housing improvement, are assigned to the household sector. These are considered to be very small. 3/ Public investment in this flow of funds analysis will differ from that reported in the tables and analysis of Chapter I on public resource mobilization. This is because investment here includes only fixed gross capital formation and does not include expenditures related to new capital expenditures normally reported in public budgets under investment, particularly with respect to start-up costs of investment projects.  - 32- 2. Sources and Uses of Funds: Households 3.10 The pattern of household finance In Cameroon is suggestive (a) because of its large net surplus position which implies that households are an important source of financing for the (deficit) enterprise sector; and most importantly, (b) because of the existence of the informal sector, which is a major component of the household sector's portfolio, and which represents a substitute to the formal financial system. Table 3.4 shows the main sources and uses of total investable funds of households from 1980 to 1984. 3.11 The most Important source of funds for households is their savings, which have been the origin of approximately 96? of their total available funds on average. Loans from the commercial banking system are only a minor component, averaging only about 4% of their total funds. 3.12 In terms of uses of household funds the most important are bank deposits and the informal sector. The other two uses which account for about 15% of the total are real investment (e.g., subsistence housing) and currency holdings. What is particularly noticeable during this time period is the success which the informal sector had in capturing funds from the household sector; it accounted for 66% of household savings in 1982/83. The figures show that during the 1980s the banking system was not a very attractive proposition to households in terms of portfolio choice in comparison to the informal sector, although in 1984 the banking system recovered the relative position it had in 1980. It seeus likely that the negative real deposit interest rates, the lack of formal financial institu- tions in the rural areas, and the higher rates paid by the informal sector were major contributing factors. This suggests that tontines collect a non-negligible proportion of household savings. In any event, perhaps as much a one-fifth of informal sector savings are expended on essentially subsistence housing. 3.13 The informal sector has both negative and positive aspects. On the positive side, the informal sector has provided Cameroon with a form of financial intermediation, albeit rudtimentary, that has been an important complement to the limited funds which the commercial banking sector provides to the agricultural and SME sectors; it has also been important in providing an outlet for savings which are not attrav.ted by the banking sector. On the other hand, the tontine system is not integrated either among individual tontines or with the formal financial sector, limiting the degree of financial deepening in the economy; 4/ In fact, the institutional and social strength of this sector may have impeded the 4/ Basically, then, savings which are "deposited' with the tontines are much like real (non-financial) savings in that they are not available to the economy as a whole for intermediation to the most productive (Footnote Continued)  - 33 - development of the formal sector. All this suggest the need for the banking system to be reformed so as to be able to attract those funds. e.g., raising interest rates on deposits and more importantly offering services to small savers similar to those provided by the tontines including easier access to savings Instrumeuts and investable funds. Table 3.4 Uses jnd Sources of Funds for Households (in billions of CFAF and percent) (fiscal years) USES OF FUNDS 1980 1981 1982 1983 1984 Informal Sector 22.0 29.9 51.6 82.3 70.7 44.6% 41.1% 53.0% 65.8% -36.8% Commercial Bank Deposits 19.9 26.1 27.8 24.9 93.1 40.4% 35.9% 28.5% 19.9% 48.4% Direct Investments 2.5 5.0 10.0 13.0 20.0 5.1% 6.9% 10.3% 10.4% 10.4% Currency 4.9 11.7 8.0 4.9 8.5 9.9% 16.1% 8.3% 3.9% 4.4% Total Funds 49.3 72.7 97.4 125.2 192.3 100.0% 100.0% 100.0% 100.0% 100.0% SOURCES OF FUNDS 1980 1981 1982 1983 1984 Real Savings 46.8 72.7 90.1 119.6 190.0 94.9% 100.0% 92.5% 95.6% 98.8% Bank Loans 2.5 0.0 7.3 5.6 2.3 5.1% 0.0% 7.5% 4.5% 1.2% Total Funds 49.3 72.7 97.4 125.2 192,3 100.0% 100.0% 100.0% 100.0% 100.0% Notes: (1) Direct investments by households are investments not directly related to a productive sector, housing imporvements, for example. They are equal to cell N2 of the FSAHs (see Appendix 1). (2) The informal sector is defined to be total real savings of households minus their direct investments, that is (413-N2) (see Appendix 1). Source: The FSAMs for Cameroon. (Footnote Continued) sector. But they are nonetheless more "productive" than purely real savings.  - 34 3. Sources and Uses of Funds: Government 3.14 The pattern of government finance in Cameroon has been determined by several factors. The flrst is that of public savings, which have been growing very rapidly both in absolute and relative terms because of oil revenues. The second has been the commensurate decline of foreign loans both relatively and absolutely as a source of funds. In this respect a major element in the pattern of public finlnces has been the changes in public deposits in the banking system which despite the relative stability of Treasury deposits have on the whole been quite unstable. 3.15 The expansion of oil revenues has had a major impact on govern- ment finances and lndirectly on the whole financial system. Oil has permitted government to expand its savings by 42.7% p.a. in nominal terms from 1980 to 1984, and the share of public savings in total investable funds doubled over this time period from 44Z to 88% (Table 3.5). This has allowed government to reduce its dependence on foreign borrowing as a source of finance for capital expenditures. As a result, the contribution of foreign savings to total investable funds has fallen from 53.32 to 7.9% over the period. 3.16 This increase in resources available to the Government has been used not only to finance the expanding public investment program but also to increase official holdings of foreign assets. Thus public investment has increased by 26.5% p.a. over the period (Annex Table 5) while government's purchases of foreign assets have increased by 69.3% p.a. Taking into account the flows throughout the entire period, government has also used the increases in its resources to expand its holdings of deposits at both the Central Bank and especially in commercial banks. 3.17 These rapid changes in the composition of the Government's portfolio reflect the major difficulties encountered by the public sector in channelling an investable resource without overly complicating the management of liquid assets. The issue faced by Government is how to channel the resources collected from the successful domestic resource mobilization effort into the rost productive uses. 3.18 The most straightforward solution-that of increasing public investments-can only be a partial one. Indeed, given the limited capacity of the Cameroonian economy to absorb the large amounts of funds available, the present levels of public Investment are probably close to the absorp- tive capacity of the economy. The Government, in any case, should elaborate a post-oil strategy and identify what sectors should be considered priority sectors for public Investment. Only after such an exercise can levels and sectoral distribution of public investment be determined as well as areas in which the absorptive capacity needs to be increased.  - 35 - Table 3.5 : Sources and Uses of Funds of the Central Government. CNPS and SUH (billions CFAF and percent) USES OF FUNDS Chane8 in Deoosits at Changes in Fiscal Direct Loans C4amercial Buying of Overseas TOTAL Year Investment to Firms B