STUDIES IN DOMESTIC FINCE NO. 29 CONTRACTUAL SAVNGS AND THE MOBILISATION OF IRESOURCES THE ROIE OF CONTRACTUAL SAVIGS INSTITUTIONS: A ?RELIINRY SURVEY By Parthasarathi Shome Public and Private Finance Division Development Economics Department * I wish to acknowledge the illuminating conversations I have had with Messrs. V. V. Bhatt, J. K. S. Ghandhi and W. 5. Kee and Mrs. K. W. Saito during the course of writing this paper. The responsibility o2 errors however, is mine alone. November 1976 PREFACE To facilitate the process of economic development is a distinct function performed by the financial structure; the nature and structure of this process are conditioned by the characteristics of the evolving financial structure. One of the significant aspects of this structure has been the role and function of the social security institutions. From a cursory comparative study of the pattern of household sector's financial saving, it appears that one can venture the following ccnjectures about the impact of soc-ial security institutions on saving and its allocation, and on wealth distribution. 1. They seem to have promoted the growth as well as the stability of the household sector saving - particularly in the form of financial assets. And this for several reasons. They have tailored financial instruments to some important saving motives. Households seem to treat this saving initially as a tax; hence these instruments do not adversely affect the relationship between saving and disposable incomes. By reducing uncertainty, they seem to have induced the households to take greater risks with regard to their saving-investment decisions. The contractual nature of this saving imparts stability to household saving behaviour. 2. Since these liabilities are of a longer-term nature, they have been able to provide medium and long-term finance for durable projects with long gestations lags in the public as well as the private sector. By promoting the growth and stability of the longer-term end of the capital market, they have improved the resource allocation process. Since their investment decisions are guided more by the investment motive than by the speculative motive, they induce greater efficiency in the use of resources. 3. Since they have been one of the principal sources for medium and long-term finance, and hence one of the major investors in long-term financial assets, they seem to have reduced the degree of concentration in the distribution of wealth and incomes. The term 'Pension Fund Socialism' in a way describes this phenomenon. If these hypotheses be true, it should be one of the major objectives of financial policies in the LDCs to promote the growth of social security institutions. The Division has planned some comparative studies to discuss and test some of these hypotheses on the basis of the experience of countries like India, Brazil, Malaysia, Turkey, Sri Lanka, Philippines and Korea. This paper provides merely a preliminary survey of this field of inquiry. V. V. Bhatt CONTENT S Pa.ge Number INTRODUCTION .......................................... 1 I. THE PLACE OF CONTRACTUAL SAVINGS AMCNG FINANCIAL INSTIRUMNTS AND INSTITUTICNST.................... 3 Financial Instruments and Institutions. Contractual Savings Institutions. 7 II. THE EXTEKT AND SCOPE OF CONTRACTUAL SAVINGS.. 13 Relation to Gross National Savings. 14 Relation to Gross National Product. 18 Implications for Aggregate Savings. 20 III. SAVINGS MOBILISATION AND INVESTIENT OF SURPLUSES 24 Factors Influencing Savings Mobilisation. 24 Investment Criteria. 28 State Influence on the Funds. 30 Different Country Experiences. 34 LIS3T OF TABLES Page Number Table 1-1 Different Forms of Contracts ................... 8 Table II-1 Social Security Savings: Relation to GNP and GNS in Developing Countries (percentage)....... 1 Table 11-2 Social Security Savings: Relation to G'P and GNS in Developed Countries- (percentage) ........ 19 Table 11-3 Household Contractual Savings: Relation to Total Financial Assets (percentage)............ 22 Table III-1 Contractual Savings and Savings Mobilisation (percentage indicators)........................ 27 Table 111-2 Contractual Savings: Contributions to Central I Government (percentage of Current Revenue and Public Financg) .............................. 31 Table 111-3 Different Investment Patterns of Contractual Savings (local currencies/percentages)......... 3 Table 111-4 Investments (national currencies) of: a) National Insurance in Trinidad- Tobago.. . . 38 b) National Insurance in Jamaica ..a.i.c..... 39 c) Social Security System in Panama ......... 40 d) Local Life Insurance Companies in Guyana. 41 INTRODUCTION In this preliminary survey, we analyse the possible role of contractual savings in mobilising savings and affecting the development process by studying a wide sample of countries. In Chapter I, we define contractual savings, focus on their role among other financial instruments and categorize different forms of contractual savings according to their potential capacity in mobilising savings. We observe that since the household sector is the only surplus sector, the government and the corporate sector often being deficit sectors, we should concentrate on the savings potential of the household sector if we are interested in the mobilisation of savings from the surplus to the deficit sectors. Chapter II studies the relationships of contractual savings with the gross national product, gross national savings and total financial assets of the household sector, in an attempt to study the extent and scope of contractual savings. We note that while all these relationships move in line with one another, they may not have any direct implication on the effect of contractual savings on aggregate national savings. Indeed, it may not be pertinent to analyse the effect of contractual savings on national savings; rather, it is concluded, the effect of contractual savings on the total financial assets of the household sector may be more indicative of the answer to our query, since we have to be ultimately interested in the total savings of this surplus sector. Finally, in Chapter III, we study the factors which may have some special influences on contractual savings institutions, especially pension schemes (since provident fund schemes are more widely accepted -2- as mobilisers), in mobilising savings. We also study the factors that must be borne in mind when these funds are invested, for it is this investment that affects development. In this connection we study, in some detail, the actual experiences of some countries in the investment process. The most striking phenomenon emerges to be the influence of the State over the use of funds. Channelisation of funds through the State, it is noted, may or may not mean contribution towards development. -3, I. THE PLACE OF CONTRACTUAL SAVINGS AMONG FINANCIAL INSTRUMENTS AND INSTITUTIONS The domestic savings performance of developing countries has been impressive during the development decade (1961-70), the savings rate rising from 12.8% in 1951-6o to about 17.0% in 1961-70, as a percentage of GDP, a performance, indeed better than that of today's developed countries during their developing stage (1861-90). The level and the rate of growth of income, the size of the country, inflation and inflationary expectations, the existing and changing financial structures affecting the savings of the different sectors in the economy, were among the important factors 2/ that conditioned the performance. Inspite of the exemplary savings performance during the devel- opment decade, however, the savings rate dropped in the 70's, due to structural, behavioral changes such as the violent recession in devel- oped countries in the early 70's,the breakdown of the international monetary system, the world-wide crop failures and the structural shift in the relative price of oil. The need to generate new savings in the domestic economy thus seems to have raised its head anew. The household 1/ These developing countries refer to a sample of 33 low and middle income countries according to the Bank's classification in the "Simlink Model of Trade and Growth for the Developing World", World Bank Staff Working Paper, No. 220, 1975. 2/ For a detailed description of savings performance and constraints on savings for developing countries during the development decade, refer, "Resource Mobilisation in Developing Countries: Financial Institu- tions and Policies". Studies in Domestic Finance, No. 23, World Bank 1976. One of the objectives of the paper is to focus on possible institutional reforms for the removal of some of the constraints that may exist on savings. The current study emanates from there, namely the role of contractual savings as a financial institution for savings mobilisation. sector being the only surplus sector, one needs to study greater savings mobilisation possibilities from this sector to the deficit (corporate and sometimes, government) sectors. Several constraints may operate to curb household savings perfor- mance, an important constraint being the lack of proper financial insti- tutions. Even among different financial institutions, some may be more appropriate and may need greater encouragement at a particular time and place than others. Our purpose here is to focus on the role of contractual savings institutions. FINANCIAL INSTRUMENTS AND INSTITUTIONS: Financial instr-uments such as deposits of different duration and maturity, evidences of participation in the equity (ie. profit minus loss) of a business enterprise, claims arising out of life insurance and pension contracts, are issued by a corresponding variety of institutions. Each financial instrument may have special advantages or disadvantages regarding its development in a particular environment. Countries may thus differ substantially with respect to the character and operational aspects of financial instruments and institutions. They may also differ regarding the way their financial structures have changed over time. These differences are evident in the sequential development of the various instruments and institutions, through the speed of their adaptation as the economies have grown, as well as in the differences in their penetration into the different sectors of the economies. 1/ Other causes may be unsuitable budgetary-monetary policies, predominance of the farm economy with an outmoded land tenure system, a lack of financiers, managers and entrepreneurs, social and cultural constraints, inflation etc. The effectiveness of bank deposits in mobilising savings cannot be denied. Since thrift deposits seem to be a preferred asset among households, the commercial banking system needs expansion. However, the banking system may not always be able to reach small savers and enterprises, especially in I/ rural areas, at controlled administrative costs and risks,7 and it may be time-consuming to lay the foundation for such an infrastructure. This may lead to a levelling off of savings through bank deposits, beyond which it may be necessary to encourage alternative forms of savings within the structure of financial intermediaries. The role of credit co-operatives has been mentioned in this regard as complementary since they are good lending agencies, able to reach small enterprises at low administrative costs. However, both rural devel- opment banks as well as credit co-operatives are "investor-oriented" insti- tutions with probably a limitation withiespect to savings beyond a certain ceiling. Instead, it may be better to foster "saver-oriented" financial 2/ institutions for generating higher savings Such a source of generating additional financial savings may be a predominantly institutional and large- ly mandatory flow of personal savings through contractual savings institutions. For this purpose, particular policies by governments may need to be altered significantly, for example, they may need to lay only tertiary importance on the creation of a market for government and corporate securities among savers. The limitations of effectively enunciating capital markets in 1/ For example, time deposit certificates have to be printed in advance, their transportation and storage may involve risk, their destruction may curb the possibilities of recovery; safe deposit boxes introduced to counter such risks may increase costs for the saver. 2/ For an exposition of this distinction between saver-oriented and investor- oriented financial institutions, refer M.S. Joshi, "The Role of Contractual Savings", Finance and Development, Volume 1, 1972. developing countries seem to be many and it may not be essential *for them to pass through the same historical process of the emergence of a financial structure as was experienced by some of today's developed countries. For one thing, prerequisites for the successful operation of a securities market are diffusion of ownership and proper dissemination of information regarding the asset-liability position of the enterprises concerned. in developing countries both these criteria may not be well satisfied due to the household-limited nature of many enterprises, relegating ownership within the extended family and keeping the asset-liability position, very often, a well-guarded secret. secondly, apart from the difficulties of developing a sound securities market, the ability of a securities market to generate and mobilise savings itself may be questioned. Even from the experience of some developed countries, it may be pointed out that it has played only limit'ed roles for example, in Japan, France and Germany, when compared to others like the U.S.A. and U.K. If this is so, contractual savings such as life insurance or provident and pension contracts may need to be given more attention and prove very attractive in today's developing countries towards their drive for greater savings mobili- sation. This is especially true since these are 'saver-oriented' institutions, geared for the express purpose of servicing the buyers of the issues of these institutions. In many countries provident and pension fund schemes have been successful, such as Turkey, Malaysia and India. Social security institutions have also figured significantly in several countries such as Sri Lanka. Jamaica and Brazil. Life insurance policies are very popular in many countries and need to be expanded in others. For the farm sector, crop insurance may be more attractive than life insurance and this possibility needs to be looked -7- into. These saver-oriented institutions may be used not only to provide the necessary cushion a:gainst risk-ridden situations, but also as a specific tool to generate financial savings in the economy,and a careful method of allocating such savings for development may be worked out. CONTRACTUAL SAVINGS INSTITUTIONS: There is a reasonable measure of doubt regarding the definition of and any disaggregated taxonomi, on contractual savings due to the wide nature and forms of these institutions. Some schemes may be voluntary (individual endowment insurance), while others mandatory (group insurance, pension funds), sometimes leading to a definition of contractual savings excluding the latter. However, voluntary schemes may also be considered 'compulsory' by the contributors once the contract is entered into, for fear of Josing the principle paid.. Thus, for our purpose, a contract is defined as any definite commitment (voluitary or mandatory, individual or social) by an economic agent to enter into a periodic payment scheme towards covering future risks. These contracts may, in addition to covering risks, account for savings of the contributors if there is an endowment element in them. Such contracts may then generate contractual savings, defined as the difference between the revenues collected and the benefits paid out in a particular year. The accumulated surplus or fund as of any year is the sum of all annual savings plus any income on the investment of these savings. Table I-1 exposits a simple taxonomy on such contracts and those that may generate savings. -8- Table I-1: DIFFERENT FORMS OF CONTRACTS Provident Funds (funded) Social Security Pension Funds Contracts (pay-as-you or funded) Individual endowment Those which generate (funded) savings Life annuity Insurance (funded) Group insurance Those that may not generate savings Property insurance Term insurance All forms of contracts shown in Table I-1 have the potential of generating savings except, as depicted, group, property and term insurance. This is because these types of insurance cover only the risk of a future contingency and have no annuity element for the insured over and above the risk coverage. All other categories of contracts may thus be called contractual savings institutions. 1/ The profits from such contracts accrue to the companies. These are therefore not earnings d the household sector. Refer J. Ghandhi, "The Role of Commercial Banks in Sri Lanka n, Division of Private and Public Finance, World Bank, 1976. -9- Another categorization that emerges from Table I-1 is that of contracts based on the funded vis-a-vis the pay-as-you-go principle of benefit payments. In the full funding (sometimes referred to as the general average premium) system, demarcated funds are set aside for contributors, creating a reserve and earning investment inc6me (interest and appreciation of value). These funds are then used at retirement etc. to pay the claimed benefits. In the pay-as-you-go (or annual assessment) system, benefits are paid from the annual revenue usually with no prior reserve formation except may be a precautionary fund. Usually short-term programs are covered by the pay-as-you-go system; but long-term programs may be covered by either, as is clear from Table I-1. The fully-funded method has a greater capacity to generate surpluses due to the creation of reserves (eg. India and Malaysia). However, a program based on the pay-as-you-go principle may also generate savings through contingency reserves (eg. U.S.A., Sri Lanka and Turkey). The accumulated funds may be mobilised for necessary consumption for long-run investment if collection of revenues occurs much ahead of payments. indeed these funds may constitute a stable source of long-run investment given the predictable nature of the premiums paid. It may therefore be worthwhile to study briefly the different forms of contractual savings institutions and their probable effects on savings mobilisation. Social Security Schemes: Provident Funds: In a provident fund scheme, each individual contributor usually has an account of his contributions, to be paid back in a lump-sum at the end (however defined) of the contract. Since they are based on the - 10 - funded principle they are expected to generate considerable surpluses. Pension Funds: Pension fund schemes appear in a variety of forms and are, therefore,mor.e difficult to handle as a sub-group. Some may be contributory, some not; some funded, others pay-as-you-go. Some schemes may be insured, some self-administered (for example, bank trusteed), etc. Further, the method of payment may vary widely. There are several methods: (i) In the accumulation method, payments are based on accumulated savings; (ii) in the 'pure savings, method, they are based on what the accumulated savings will purchase, rather than the savings per se; (iii) in the 'formula' method, payment may depend on accumulated savings, number of years of employment, final year salary etc. All these methods of calculating benefits may be found in the same country, for example in Trinidad and Tobago. The ability of a particular pension fund to generate savings will depend on all these factors. Insurance Policies: Individual Life Insurance: Individual life insurance may vary from contract to contract. If it is an endowment insurance, providing a lump-sum at a fixed maturity date (to the insured or a survivor) or life time insurance protection, then the contributor saves through the scheme. On the other hand, if the 1/ Refer M.A. Odle, Pension Funds in Labor Surplus Economies, Institute of Social and Economic Research, University of West Indies, 1974, for a detailed analysis of pension funds vis-a-vis life insurance policies. benefit is paid only if the insured dies within a specified number of years, as in a term insurance, there is no savings element involved for the contributor, but, only a coverage for risk. Life Annuity Contracts: These contracts, without any cash value, enable the beneficiary to receive an annuity for the rest of his life and stop when he dies. They can be issued to an individual, group of individuals or to the employees of an enterprise. Over and above the risk element, they constitute savings for the households. Property, Term or Group Insurance: These are just risk coverage devices with no annuity or lump-sum element at maturity. Group insurance, for example, provides short-term insurance and is usually based on wide participation in order to average out the risk from unhealthy or over-age participants. Therefore, they do not generally comprise savings for the participating households. In conclusibn, it may be noted that the prevalence of contractual savings institutions is becoming inereasingly perceptible throughout developing nations. The contingencies as well as the persons covered have increased. Among the contingencies may be named employment injury, maternity, sickness, invalidity, survivors and family benefits apart from old age. With respect to the persons protected, it is common to find protection for civil servants, army personnel and wage and salary earners, especially those unionised. In some countries, all employees are covered though in very few countries has social security effectively penetrated -12 - 12 - the rural areas. Here crop insurance may play an important role and its potential needs to be studied further. As the degree of organisation of the employees as a group is increasing, and as the governments of the developing countries are looking more into the issues of rural sector development, more people are being covered by social security schemes. This results in an increasing compulsory inflow of funds and if those newly covered belong mostly to the younger generation, the initial benefit payments may remain low. This then may enable the social security sector to generate a surplus, to be used in other deficit sectors. 1/ Differences in coverage may also occur due to ethnic or racial differ- ences, especially in Africa. For example, in. Southern Rhodesia, the indigenous population is covered only for employment injury, and in South Africa, African and Asian workers are excluded from family benefits and pension are paid according to ethnic origin. Refer P. Mouton, Social Security in Africa, International Labor Office, Geneva, 197. - 13 - II. THE EXTENT AND SCOPE OF CONTRACTUAL SAVINGS Whether or not contractual savings of households increase aggre- gate household savings has been debated in the literature without a definite conclusion. In the case of the U.S., for example, Cagan finds, in a survey of 11,000 households, that pension plans do not lead the households to curb other fcrms of savings and may even motivate them to save more. In other words, contractual savings by housholds are madeat the cost of consumption and therefore add to household sector savings. On the other 2/ hand, Feldstein concludes, using a life-cycle model of household consump- tion, that households may indeed decrease other forms of saving- over- whelmingly if they participate in social security savings. However, it is generally agreed that the institutionalisation of the saving investment process improves the financial market mechanism and the role of contractual savings institutions in this market, namely the security market as well as the markets for long-term industry and housing loans becomes increasingly imnportant. To this extent, the scope of contractual savings institutions has been expanding. 1/ P. Cagan, "The Effect of Pension Plans on Aggregate Savingit, National Bureau of Economic Research, New York, 1955. 2/ M. Feldstein, "Social Security and Saving: The Extended Life Cycle Theory", American Economic Review, Papers and Proceedings, May 1976. Data on household sector savings for a reasonably large sample of countries and an acceptable time series are extremely difficult to obtain. In their absence, other indices for the measurement of the extent and scope of contractual savings have been employed in the literature. One is the relation of contractual savings to the gross national product (GNP); another is the relationship to gross national savings (GNS). We therefore, use these indices for a sample of 33 developing and 18 developed countries. 2/ presenting the relationships among social security savings (SSS), GNP and GNS in Tables II-1 and II-2. The data are mostly for the years 1964 to 966, these years being the latest years for which comprehensive data for a sample of 51 countries allowing broad comparisons. are available. How- ever, we also propose to study the relationships between household contrac- tual savings and total household financial assets for a selected group of countries and for more recent years in Part B of this study; here, as a preliminary measure, we present the relationship of contractual savings to total financial assets for the household sector, for a sample of 6 developing f countries, (1963-1971). in Table 11-3. RELATION TO GROSS NATIONAL SAVINGS: In Table II-1, countries have been divided into (a) high ratio and (b) low ratio countries on a rough-and-ready basis of including countries in the high ratio group if SSS forms approximate2y 3% (or more) of GNS. On this 1/ Refer Franco Reviglio, "Social Security: A Means of Savings Mobilisation for Economic Development", IMF Staff Papers, 19673 Public Finance and Social Security: Papers and Proceedings of the Congress at Turin, Inter- national Institute of Public Financ e, 1967; and Mouton, op. cit. 2,' 'Social Security Savings" (SSS) is being interpreted as the difference between the total receipts and expenditures by the social security sector in any particular year. Refer Reviglio, op.cit.) for elaboration of this definition. Also, refer Chapter I of this study. ..... .... -15- Table II-1: SOCIAL SECURITY SAVINGS (SSS): RELATION TO GROSS NATIONAL PRODUCT (ONP) AND GROSS NATIONAL SAVINGS (GNS) IN DEVELOPING COUNTRIES 2/ (3 Year Averages 1964-1966-) 3/ High Ratio Developing Countries Low Ratio Developing Countries 4/ Country S5 SSS GNS Country SSS GNS GNP CM'NS GNP GNP GNS GNP Ghana .49 4.63 10.56 ..Zambia .30 .82 36.28 Togo .86 6.87 12.55 Cameroon .08 .74 10.58 Malaysia 2.17 12.60 17.19 Upper Volta .04 .89 4.75 India .57 4.03 14.12 Pakistan .00 .00 11.66 Sri Lanka .78 6.04 12.83 Iraq .14 .68 20.89 Cyprus .81 5.09 15.80 Colombia .04 .25 16.25 Greece .96 5.84 16.48 El Salvador .17 1.40 12.24 Israel 1.06 8.40 12.64 Guatemala .06 .58 9.54 Malta .47 3.14 14.89 Honduras .16 1.36 11.99 Portugal 1.69 9.59 17.64 Jamaica .05 .32 15.19 Spain .66 2.91 22.58 Mexico .14 .77 18.42 Syria 1.04 10.66 9.84 Trinidad- Tunisia .37 2.92 12.59 Tobago .22 1.06 20.85 Turkey 1.42 8.64 16.42 Brazil .70 3.96 17.62 Costa Rica 1.59 10.91 14.61 Ecuador 1.43 13.77 10.38 Nicaragua .44 2.98 14.87 Panama 1.38 9.19 15.03 Paraguay .74 6.47 11.48 Uruguay .43 2.98 14.59 1/ Percentage Figures. 2/ 63-65 figures for the following countries: India, Sri Lanka, Cameroon, Iraq, Malta, Jamaica. 3/ Ad-hoc definition of SSS comprising approximately 3% or more of GNS. 4/ SSS was calculated as total receipts minus benefit payments in a particular fiscal year. Source: (1) The Cost of Social Security, Seventh International Inquiry, 1964-66, International Labor Office (Geneva 1972). (2) Basic Economic Data Sheet - World Table I, World Bank. - 16 - basis, 21 countries belong to the high ratio group while only 12 countries belong to the low ratio group (with a ratio of 1.h% or lower). On this ad-hoc basis, therefore,one may observe that SSS represents a non-negligible percentage of GNS in developing countries with social security systems. In some countries, notably Malaysia, Portugal, Syria, Costa Rica and Ecuador, SSS comprises approximately 10% of GNS, while in Togo, Sri Lanka, Cytrus, Greece, Israel, Turkey, Panama and Paraguay it is more thah $%. These countries have all had expanding social security schemes between 1964 and 1966. While most of them were in the initial stages of their working, the surplus they generated were high. Pension programs in Turkey and Israel, for example, were fo-unded in 1957 and 1953, respectively. In Costa Rica, the pension programs for employees in industry and commerce generated a substantial surplus. In Fanama, even though the pension program had been started in 1941, an expanding coverage together with a strict set of regulations for benefit payments permitted surpluses to develop continuously. The provident funds of India (1952), Malaysia (1951) and Sri Lanka (1959) were relatively new and were very successful in mobilisationi of resources. In India, employees of firms with 20 workers or more were covered after a 12-month employment, as well as public employees (defense, railways, post and telegraph, central and state governments). Between 1954 and 1962, the annual rate of mobilisation' doubled, with a net increase of reserves of Rs. 10.6 billion. The Malaysian Enployees' Provident Fund, covering all non- public employees, was found to be an important source of surpluses needed for domestic developental financing. The Sri Lanka Emplcyees' Provident Fund for 1/ However, this surplus was used to balance much of the deficit in the Government employees, pension program. - 17 - both private and public employees also generated remarkable surpluses. On the other hand, some other countries in the high ratio group, especially those in South America, were beset with several problems. The social security system in Brazil (included among high ratio countries) for example, worsened since 1960 due to a growing proportion of inactive members, unpaid contributions by government and employers and a high rate of infla- tion. In Uruguay and Nicaragua, the benefit payments absorbed much of the social security receipts, leaving negligible surpluses. The most harmful effect on the real value of the assets held by social security institutions has been through inflation. which also accounted for the glaring shortfalls 2/ in the contributions by employers and government. Among the low ratio developing countries, there was no pension scheme for non-government employees in Columbia, Guatemala, El Salvador and 3/ Honduras. In Iraq, Jamaica, and Trinidad-Tobago, benefit payments left negligible surpluses. In Upper Volta and Cameroon, International Labor Organisation (ILO) projects (reflecting the financial prospects of social security schemes) were established only in 1967. However, for Zambia, even though the a964-66 average was low, the rate of growth of accumulated funds has been very high in recent years, being 2541 during the 1972-73 financial year alone. 1/ Refer Reviglio, op.cit., for figures on resource mobilisation in these countries. Also refer K. Saito, "Household Savings: An Estimation for Sri Lanka", Studies in Domestic Finance No. 27, World Bank, 1976. 2/ During this time period, the Uruguayan social security sector received as much as 13" of GNP (one of the highest ratios in the world) as a result of ccntributions of 38% to 4% of wages and salaries. Social security tax revenues almost equalled other Central Government tax revenues, but they werj almost wholly absorbed by benefits. 3/ The social security sector in many Latin American countries would have shown a higher surplus if the revenues were all received in cash. Instead large amounts of obligations by the employer or government remained unpaid or were paid in government securities. - 18 - Table 11-2 provides the same figures, SSS as a percentage of GN? and GNS, for a group of 18 developed countries. Of these, two countries, Netherlands and Sweden, had the latter ratio of over 10% si. countries, Belgium, Canada, Finland, Iceland, Japan and Switzerland, had a ratio of more than 5%, while six countries, Austria, Denmark, France, West Germany, New Zealand and U.K., had a ratio below 3%. Given these numbers, the patterns of savings generated by social security in developing, as opposedto developed countries, are not distinct. RELATION TO GROSS NATIONAL PRODUCT: So far we have been concerned with the relation between SSS and GNS. However, the relative importance of SSS can be measured also in terms of its relation to the GNP. It can be seen from Table II-1 that most devel- oping countries with high SSS ratios in terms of GNS, also had relatively high SSS ratios in terms of G4P compared to the low ratio countries. For example, all countries with a more than 1% GNP ratio also had a more than 8% GMS ratio. Similarly, countries with low CNS ratios have low MP ratios as well. This hypothesis is confirmed by an observation of Table II-2 for developed countries. Countries with the highest GVS ratios (Netherlands and Sweden) also had the highest GNP ratios. Further countries. with more than 5% GNS ratios all had approximately 1.3% GNP ratios or more. Countries with the lowest GIS ratios were also beset with the lowest ONP ratios. Therefore, for countries for which aggregate savings (or investment) data are not easily available, it may be possible to study the importance of the social security sector in terms of its relationship with the GNP. - 19 - 1/ Table 11-2: SOCIAL SECURITY SAVINGS (SSS)- REIATION TO GROSS NATIONAL PRODUCT (GNP) AND GROSS NATIONAL SAVINGS (GNS) IN DEVELOPED COUNTRIES 2/ (3 Year Averages: 1964-1966 ) 3/ Country SSS GNS GNP GNS GNP Australia .81 3.16 25.57 Austria .43 1.$5 27.67 Belgium 1.28 $.51 23.28 Canada 1.49 6.55 22.79 Denmark 0.39 1.81 21.39 Pinland 1.73 6.35 27.19 France .06 .22 26.81 W. Germany .6 2.10 28.56 Iceland 1.51 5.50 27.40 Italy .83 3.70 22.52 Japan 2.09 6.46 32.34 Netherlands 2.73 10.26 f 26.64 New Zealand .24 .97 24.74 Norway .90 3.17 28.36 Sweden 3.01 12.21 24.67 Switzerland 1.38 5.34 29.63 Uhited Kingdom .41 2.07 19.62 U.S.A. .91 b.77 18.96 1/ lercentage Figures. 2/ 63-65 figures for the following countries: Australia, Canada, Denmark, New Zealand, U.S.A. 3/ SSS was calculated as total receipts minus benefit payments in a particular fiscal year. Source: Refer Table II-1, op.cit. - 20 - IPLICATICNS FOR AGGREGATE SAVIGS: While it is clear that the extent and scope of contractual savings may be studied by their relationship with GNP and GNS, it is not clear whether these ratios have any implications for aggregate national savings. For example, from the percentage figures of GNS to GP presented in Tables II-1 and I1-2 for both developing and developed countries, one may note that high (SSS:IS) ratio countries need not necessarily have relatively high savings rates ((IS:GNP). Nor is there an ostensible relationship between the (SSS: GNP) and(GNS:GNP) ratios. This brings to focus the extreme importance of not over-extending the possible implications of contractual savings institutions. Instead of concentrating on the effects on GN1S, it may be more pertinent to study the effects on the aggregate savings of the household sector alone. Indeed because one is interested in the potential of contractual savings institu- tions as a savings mobiliser in terms of the flow of funds from surplus to deficit sectors, one could focus on how contractual savings affect the 1/ savings of the surplus household sector, rather than their implications on 2/ GNS. 1/ Since the size of the corporate sector is small in developing countries, it is to be expected that the savings of this sector will not be high. However, savings rates of the corporate sector in developed exonomies are also not high: 4.71 in Japan, 2.3% in USA, 3.5% in Germany, and 3.9% in France. The result of this is that the corporate sector is a deficit sector usually receiving from the household sector. 2/ It may be worthwhile to recall at this point that the potential of different contractual savings institutions in generating savings was analysed, in Chapter I, in terms of savings of the household sector. -21- Table 11-3 (1963-1971) presents contractual savings (CS) as a percentage of total household financial assets (TFA) for 6 developing countries. It may be interesting to note that India and Sri Lanka with high (CS:TFA) ratios in Table 11-3 (1963-1971) also had high (SSS:GNS) ratios in Table 11-1 (1963-1965); Colombia and Jamaica, with relatively low (CS:TFA) 2/ 3/ ratios (upto 1968), had low (SSS:GNS) ratios (1963-1966). From these figures it may be possible to conclude that if contractual savings are high as a percertage of TFA of the household then it is also a high (low) percentage of GNS. This is a reflection of the fact that TFA of the household sector is an extermely -1mportant component of GNS and, therefore, the two may be expected to move together. However again, it may be of great interest to note from Table II-1 that both India and Sri Lanka had lower (GNS:GNP) ratios than Jamaica and Colombia during the 1963-1966 period. Thus it may be difficult to make conclusions on aggregate savings performance of an economy from contractual 1/ Contractual savings basically comprise all forms of social security savings such as pension funds and provident funds, as well as those from insurance. Refer Chapter I for clarification. 2/ Between 1969 and 1971, the (CS:TFA) figures improved substantially for Colombia and Jamaica. 3/ The same holds true for GNP. This is because, as we viewed in Table II-1 and 1-2, the (SSS:GNS) and (SSS:GNP) figures move together. )/ One may note that the Korea figures in Table 11-3 are relatively low. This may be due to the fact that prior to 1976 when the National Welfare Pension Law became effective, contractual savings were not dominant in Korea. For a detailed analysis, refer Part B of this study. - 22 - Table II-3: HOUSEHOLD CONTRACTUAL SAVINGS: 1/ RELATION TO TOTAL FIANCIAL ASSETS ( 3-Year Average Percentages) 1963-65 1966-68 1969-71 2/ Jamaica -10.7 7.3 26.0 India 27.5 33.6 33.5 Korea 10.4 13.8 14.1 3/ Si Lanka 7h.3 64.9 52.3 Colombia 7.0 13.1 20.0 Peru 9.5 12. 4 1/ Basically, contractual savings are inclusive of provident and pension funds, and life insurance in each country. 2/ 1965 figure. 3/ Average of '64 and 165. 4/ Average of '69 and '70. Sources: 1. Reserve Bank of India Bulletin, February 1972, May 1973 and August 1975. 2. Flow-of-Funds, Bank of Korea, 1972. 3. K. Saito, op. cit. (for Sri Lanka). 4.. Clark W. Reynolds, "The Use of Flow-of-Funds Analysis in the Study of Latin American Capital- Market Development", unpublished paper, Organi.zation of American States, 1974. - 23 - savings alone. Instead, the role of contractual savings in generating household sector savings, to be transferred to deficit sectors thereby improving savings mobilisation, needs to be analysed. With this preview of the extent and scope of contractual savings, we go on to study different aspects of savings mobilisation in the next chapter. III. SAVINGS MOBILISATION AND INVESTMENT OF SURPLUSES It has been argued that contractual savings schemes based on the funded principle have a greater savings mobilisation capacity than those based on the pay-as-you-go principle. The usual argument for this is that the former is based on the accumulation of resources unlike the latter. Yet it is an empirical fact that pension systems in several countries generate substantial savings even though based on the pay-as-you-go principle. Hence an elaboration of how pay-as-you-go plans may mobilise savings, in addition to funded plans, is attempted below. FACTORS INFLUENQING SAVINGS MOBILISATION: While plans are newly established, deferred benefits and broader or expanded coverage may bring in added revenues. As the economy grows, if the benefits are controlled and the retirement age is not early, then the period of savings generation can be extended. If pension claims are illiquid in the sense that it is difficult for employees to borrow on the basis of their contributions, or that em- ployers do not get back their contributions if an enterprise is liquidated, then again the savings potential is high. Household sector savings may also increase through pensions if the poor employees will not save except 1/ Refer Reviglio and Mouton, op. cit. through compulsion. Workers may increase other forms of saving after becoming participants in the pension plans since now they are more aware of the need to save and believe that a comfortable retired life is not impossible to achieve. Also if pension contributions are illiquid, people may feel the need to save in more liquid forms. Cash contributions and benefit payments of pension plans are, by and large, 1/ regulated and, therefore, predictable. Thus there is a greater freedom in mobilising these savings on a long-teim basis. Taxation for pension and other social security schemoes faces less political opposition than other taxes. Employers like it since it is not progressive and can pass on its incidence; employees like it since it is not progressive and cannot often perceive the tax element, regarding it as an insurance premium - an illusion often encouraged by govemments! Also social security taxation has a high income elasticity of yield as the tax base increases during the early stages of development. 3/ This tax is easily collected if withheld at the scurce. 1/ Cash contributions go on throughout the working life; benefit payments are actuarially determined. 2/ Refer J. Gandhi, op.cit., for a detailed analysis of how this may happen. 3/ It is not as though social security schemes do not possess sparks of curbing savings, especially in other Ibrms and other sectors. For poor workmen whose consumption cannot reduce, savings in other forms will. Business savings will fall unless prices or productivity increases. Pension deductions allowable in income tax rates may lower government savings. A detailed analysis of the overall effects of such substitutions will be attempted in Part C of this study. - 26 - In developing countries, with higher rates of growth, expanding financial sectors and improving structures, more savings may be generated through contractual savings. As unemployment is controlled and the share of wages and salaries improves (with a curb on the bias towards capital-intensity), as the small scale sector is increasingly absorbed under social security schemes, and as the inci- dence of crop-insurance becomes wider in the rural areas, more savings mobilisation may be expected from these sectors of the economy. Table III-1 presents figures attempting to indicate savings mobilisation by contractual savings instituticns for a sample of countries for which recent data were available. The contribution of total public sector saving may be one such index: it is seen that in both Egypt and India, they have been substantial. In Brazil it is increasing. Annual social security savings as a percentage of the total accumulated surplus is high in Panama, but in Agentina, savings have indeed been negative. In several African countries, newly instituted social security schemes have led to accumulation of reserves due to a low average age of workevs as well as the recency of the schemes, with pro- visions for a few benefits only, for example, in Gabon, Ivory Coast, Mauritania and Togo. Indeed, the accumulated reserves were multiples of the annual resources of the social security institutions in some coutries (3½ times in Zaire, 3 in Ghana, 21 in Niger, 2 in Ivorty Coast and Togo) . 1/ Refer P. Moutpn, op.cit., for detailed figures on social security reserves of African countries. Table III-1: CNTRACTUAL. SAVINGS AND SAVINGS MOBILISATION (PERCENTAGE IJDICATORS) Type of COUNTRY Contract 1967 1968 1969 1970 1971 1972 1973 3974 1975 1976 1977 1978 (a) Social Security. Savings (SS) as % of Total Fund: Argentina Social .40 -11.98 -10.40 -5.89 -8.i -4.9h -6.51 (1973 Estimated) Security Panama/ Social 22.69 18.01 17.15 20.90 18.91 17.63 21.50 22.18 28.33 (1975 Estimated) Security (b) SSS as % of Public Sector Saving: EYpt Social 66.92 83.80 87.58 77.1. 67.06 61.86 55.75 (Projection Eccept Security 1972) Brazil Social (1977 Estiiated) Security 1.83 3.52 8.54 8.62 Integration Fund Lndia2/ Provident 19.57 21.74 19.21 17.28 Fund 1/ Note that these figures can also be calculated for Jamaica, T"inidad-Tobago, and Guyana from Table III-4. 2/ Percentage of gross household savings. Source: World Bank Country Reports (1974-]976). - 28 - Given the transitory nature of the savings mobilised through contractual savings institutions, it is of the utmost importance as to how these funds are invested. INVESTMENT CRITERIA: The use of the reserves of contractual savings institutions should be based on three principles, namely, genuine security,optimum yield and contribution to economic development. These-criteria, however, may not always be mutually compatible. For example, it is sometimes assumed that if savings are lent to the State, then they will automatically be channelled for development together with p oviding security and optimum yield. However, even though loans to the state are usually secure, they may not always be used for developmental purposes, going instead towards current expenditure of non-developmental types. On the other hand, it may seem at first glance that investments in the hotel industry may give an optimal yield but may contribute little towards development. This may not be true, however, inasmuch as funds invested in the hotel industry may contribute towards a viable national tourist industry with a compara- tive advantage for earning hard currencies, and hence to development. This has been the case in Zaire, Senegal, Zambia and Gabon. 1/ One alternative may be to make it illegal for the government to use these funds in other than capital expenditure. However, even this does not ensure proper allocation since the capital budget may include non-productive investments in government buildings and 'sick' nation- alized industries. On the other hand, some current expenditures in education (acquisition of skills) and health care increase future productivity and possess connotations of development. -29- A developmental use of the accumulated surpluses may be investing in the housing industry. This channel of investment generally does not lack security nor give sub-optimal yields. Such investments have been made in Costa Rica- On the other hand, purely speculative investment in land or real estate for the purpose of beating inflation may not contri- bute towards development. Such investments have been common in Zaire. A clearly non-developmental channel is investing in foreign securities, favorable on the other counts of security and yield. This has happened in Ivory Coast, Zaire and Madagascar but is now declining. An investment in the State is clearly a better alternative for development. Indeed even though channelling the funds through the State may not be a guarantee towards development, the funds may increase the pool available to the government, which can then be utilised for different purposes in the private or public sector according to national priorities in a way that will increase productivity and taxable capacity, as has been the practice in Brazil. The state may achieve this by channelling the funds through deve- lopment banks, state housing agencies etc. which are relatively safe channels of institutionalised investment. This has been achieved in the Philippines, Turkey, Chile, Ivory Coast, Togo and Zaire. The State may also use such funds to stabilise the economy and check inflation by investing in fixed deposits (Jamaica, Argentina). 1/ One extreme case of this phenomenon was the West African Retirement Fund which invested all its surplus in Europe till 1970, after which it has invested substantially in Senegal's hotel industry. - 30 - If this checks the overall slide in national productivity, then such investments may be justified. On the other hand, such investments may erode in value as occurred throughout the 160's in Argentina, Colombia and Uruguay. If the investments are in government bonds, the erosion signifies a redistribution from the participants. Direct investments by contractual savings institutions in public and private enterprises may turn out to be a viable method of resource exploitation. But direct investment methods calling for consumption loans to the contributors neither add to national assets nor have welcome redistribution effects, subsidising the borrowers effectively, in an inflation-ridden world. STATE INFLUENCE ON THE FUNDS: Contributions to the central (and provincial) governments by the social security sectorhave been remarkable in some developing coun- tries. Table 111-2 presents percentage data on how much contractual savings have contributed towards government current revenue as well as other forms of public financing. Noteworthy are the figures for Argentina where social security savings represent a substantial part of all taxes 1/ combined. Even in Ghana and Togo, insurance and pension funds comprise a large percentage of government current revenue and operate to fill the deficits. The other extreme is Cyprus where the social security sector 1/ Note that these figures for Argentina represent only the size of the social security sector. The surpluses generated by this sector for Argentina on the other hand, were indeed minimal (or negative) due to inflation, lack of government or employer contributions etc. Refer to Table IDI-1 for figures shoiing social security savings in relation to total fund. Table 111-2: CONTIACTUAL SAVINGS: CONTRIBUTIONS TO CMTRAL GOMENMET (Percentages) Type of COUNTRY Contraet 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 (a) Contractual Sivings as % Current Revenue: Trinidad-Tobago Pension & h.57 2.59 2.25 2.20 1.99 1.85 (74-78 Projecbed) Insurancel/ Jænaica National 6.08 6.09 5.79 (76-80 Projected) -Insurance Cyprus Social Security & .08 .65 -.31 .60 .73 -.75 Insurance Argentina2 Social 28.86 27.87 26.98 30.03 28.30 28.78 25.36 25.52 25.61 29.69 (74-77 Projected) Security Tax (b) ConLractual Savings as % of Public Financing: Israel National 6.1 3.23 11.15 14.41 15.53 h.58 4.88 3.61 Insurance Togo-/ Pension 25.00 22.24 19.66 52.40 42.05 32.74 Funds Ghana Insurance & Social 36.63 25.98 no.h5 26.85534.95 Security Colombia Social 2.32 1.63 .58 .23 Security 1/ Insurance Funds invested in government securities. -32- has even absorbed current revenue funds during 1972 and 197$. Figures for Israel, Jamaica, Trinidad-Tobago and Colombia are also high. Sometimes the pattern of investment by contractual savings institutions may be strictly regulated by the State. Colonbia's 'forced investment regime', for example, requires insurance companies and 1/ mutual funds to invest in securities of the national government, auto- nomous government-owned entities and several development organisations (Agrarian bonds, Economic development bonds, Public Debt bonds and twenty-five other such securities). These securities generally carry a lower rate of interest than the banking system can obtain from ordinary loans and are less profitable than corporate securities. Indeed, at the current rate of inflation, the real rate of return has been negative. But the forced investment regime cannot be reduced significantly, inspite of such a recommendation by the President's Commission (1971), because of the outstanding volume of forced investment on which development and official entities are dependent. Another case of State influence in the use of contractual savings is that of Brazil. Here, the government has attempted to influence production and investment decision by channelling savings through inter- mediaries and controlling their allocation of credit. The government's influence is strong inasmuch as a part of the savings is generated by its own fiscal operations or may result from programs such as the Tenure Guarantee Fund, Social Security Social Integration Program (OTS), or the 1/ Commercial and private development banks are subject to the same statute. -33- Progra for the Formation of Patrimony of Public Employees (PASEP). The PIS scheme, for example, required employers to contribute a percentage of their gross sales (startin-g at .15% in 1971 and rising to .51 in 1974) and of their income tax liabilities (2% in 1971 to 5% in 1974) into a fund for employees, drawing an annual interest of 3% and being lent for industrial development through development banks. Similarly, PASEP is financed by the federal and local governments, public and mixed enterprises, its fund-being lent to both the public and private sectors. Life insurance funds are coming under government regulatory control also in the Philippines. Recently, due to the anomalous effects of the lack of control on the non-bank financial intermediaries with respect to reserve requirement and interest ceilings in the Philippines, a joint.IMF-Central Bank of Philippines Banking Survey Commission was established. The resulting reforms give the Central Bank regulatory powers over the non-bank financial intermediaries regarding interest rate ceilings, reserve requirements and use of funds. As a result, life insurance companies, both private and government owned, have the potential of being used for long-term finance to agriculture and industry either directly (purchase of corporate securities) or indirectly (purchase of government and Development Bank of Philippines bonds). Insurance companies may act as catalysts in developing the mortgage market by acting as financial intermediaries (purchase of consolidated mortgage bond issues of savings and mortgage banks), thereby reducing the importance of direct mortgage lending. All this may lead to a better 'qualitative' structure of the savings investment process in the economy. If the State does, however, exert this much influence over the investment of surpluses, then it should be obliged to a positive rate of return, comparable to those from alternative investment possi- bilities. The State should bear the capital loss, if any, of social investments. This can be done through State guaranteed bonds rather than compulsory direct investment in development projects. Thus, State responsibility must accompany State control and regulation of the investment of the accumulated surpluses. DIFFERENT COUNTRY EXPERIENCES: Table 111-3 makes an attempt to show the variety of investment patterns for surpluses of contractual savings institutions for 5 develop- ing countries. In Tbgo, the pension fund surpluses comprise a high per- centage of total Treasury deposits whereas in Costa Rica substantial 2/ investments are made in the housing industry. In Brazil, the social 1/ The issue of the role of contractual savings in improving the 'quality, of savings even when their effect on aggregate savings is minimal has been raised by Joshi, op. cit., and is indeed very interesting. It is our purpose to make an attempt to study this issue in some depth in Parts B and C. 2/ The Social Security Fund and the National Insurance Institute have invested heavily in housing, constructing from 480 housing units between 1955-60, to 3750 in 1971 alone, and investing $6,100 per housing unit (in housing costs) between 1955-60, going up to $18,500 in 1971. These investments have yielded optimum yields as well as contributed towards development Table 111-3 DIFFERERT INVEST4ENT PATTERNS BY CONTRACTUAL SAVINGS Type of , COUNTRY Investment 1966-68 1969 1970 1.971 1.972 1973 1974 Jamaica National Insurance (million J$) Fund deposits in .9 .3 .1 .1 .9 the banking system Sri lanka7 Pension & Insurance (Rupees million) Corporation owner- 81.6 310.88 133.60 123.20 204.62 ship of non-bank (24) (24) (20) (22) (26) securities Brazil 2/ Social Integration (percentage)- Fund: 0 1.9 Credit to comimiodity producing sector Costa Rica Investment in (million colone.) - Housing by Social 18.6 23.7 19.1 25.5 Security & National (32.74) (53.49) (31.46) :(29.01) Insurance Togo 4/ Pension Fund: (million CFA Francs) Deposits with 2172 768 654 666 730 629- Treasury (25.38) (22.24) (19.65) (52.40) (42.05) (32.74) l/ 1973 is a provisional figure; the numbers in the brackets are percentages of total non-bank securities. 2/ Percentages of total credit to the comrodity producing sector. 3/ Figures in brackets show percentage of total investment in housing. 11./ Figures in brackets show percentage of total Treasury deposits. 5/ Up to October 1973. SOURCE: World Bank Country Reports (1973-76). - 36 - Integration Fund has been extending credit to the commodity producing sector in recent years. In Jamaica and Sri Lanka, surpluses are deposited in the banking system and as non-bank securities respectively, which is a common feature in several other countries. Such investments of a purely economic character (treasury and other bonds, loans to national development banks and to national investment companies, loans to public undertakings, semi-public companies etc.) have also occurred in several African countries such as Gabon, Ivory Coast, Madagascar and Togo. Indeed, it has been claimed that the purpose of social security taxation in some countries has been fcr the purpose of mobilising resources for government pr6grams. Thus, Ethiopia's Third Development Plan extends the pension scheme to the private sector explicitly for the purpose of mobilising resources for the Plan. Similarly, Kenya' s Social Security Fund is an instrument for domestic savings mobi- lisation for its Second Plan. Finally, in Tables III-, we provide some detailed figures on the composition of assets, i.e. the nature of investment with contractual savings and the receipts therefrom for four countries for which recent data (or projections) are available: (a) Trinidad-Tobago, (b) Jamaica, (c) Panama; and (d) Guyana. In all countries (except Guyana, for which total revenue figures are unavailable) the accumulated fund surplus is a significant proportion of the total revenue. In Trinidad-Tobago, the single most important area of investment has been government securities, and it is expected to grow. On the other extreme is Guyana, where government securities do not occupy a significant (though growing) allocation, while the private sector is the most important and rising. Investments in government in Jamaica also rank high among the different forms of investment. One interesting fact is that much of the - 37 - investments have been of financial variety, even though buildings and equip- ment (Panama) and Mortgage and Trust funds (Trinidad-Tobago) are also important in some countries. Clearly non-productive overseas invest- ment has been increasing in Guyana where most of the investments are made in the private sector. Among the financial investments, government (Treasury bills, Debentures) and private securities, fixed deposits and cash have figured significantly, and the returns from these investments have also contributed significantly to the total fund. To conclude one may note the variety of investment possibilities by contractual savings institutions. Keeping the main purpose of contractual savings institutions in mind namely, providing security to the contributors in defined contingencies, the best use needs to be made of the funds if they are to affect the development process. 一 Table III-4(b): INVESTMhUTS OF NATIONAL INSURANCE IN JAMAICA (1970-71 to 1975-76) (J $ millions) Upto 1970-71 1971-72 19'72-73 1973-74 3-974-75 1975-7-6 Accumulated Surplus 12.7 15.0 19.0 19.8 27.5 32.6 Invested In: JUL& 2/ 8.2 12.1 27.3 19.5 31 Jamaica Treasury Bills 0.8 -1.4 - - 25. 30.0 Overseas Stocks - - - 01 Deposits (Dank of Jamaica) 0.7 5.7 -6.4 - 2.5 2.6 Other (including cash) 3.0 1 - 5 -1.8 0.3 1 Investment Th-come: 2.8 3.9 5.0 6.8 1-0.0 13.2 Interest on J1,RS 2.7 3.8 5.0 6.8 On Jamdica Treasury Bills .1 0.03 0.02 (neg) 10.0 13.2A \ o Overseas Thvestvent, - 0.03 - Other Financial Assets (neg) 0.10 0.03 (neg) - Total :Income 14.0 16.6 21.0 23.0 32.0 38.5 al TZ (or Annu evenue, 1/ 1974-75 and 1975-76 are estimates. 21 Jamaica Jocal, Registered Stock (at cost). 3/ The flower brackets denote the subtotals of items for which further break-up is not available. SOURCE: World Bank Co-Lmtry Report (1976). Table III-4 (c): IVISTMENTS OF SOCIAL SECUIRITY SYSTEM IN P1ANAIA (1965-75) (Thousands of Balboas) upto 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 Accumulated surplus 6765 7405 7714 6828 7243 11017 12426 13180 18881 21959 (of which cash) 2219 756 342 -2023 -772 -976 -4772 622 2591 (n.a.) Invested In: Loan (net) 1710 2795 1814 -1123 -1067 710 1850 1576 3025 2507 Bonds (net) 423 1815 11039 -746 27392 5778 -2147 19764 -940 -3473 Reinvestment of Repayrents 2987 3441 3338 3413 3964 5041 5614 6194 6570 6710 Buildings and Equipment 888 1187 1529 1994 1258 1377 3759 4973 4542 5152 Investmnent Incone Interest on Joan to private sector 2_/ 1736 1843 2071 2057 2111 2381 2541 2908 3626 3712 Interest on loan to public sector 2 1663 1939 1885 2419 3514 3963 4414 4596 5358 5485 Total Incoe 25994 30230 33990 37915 42233 52712 65699 74751 87810 98984 (or Ainual Revenue) 1/ Not available. 2/ Actual leceipts. SOURCE: World Bank Country Report (1976). Table III-h (d): INVESTI'ITS OF LOCAL LIFE INSURANCE COMPANIES IN GUYANA (Million Guyana dollars) Upto 1970 1971 1972 1973 1974 Accumulated Surplus 24.3 25.4 29.9 33.7 38.1 Invested In: Central Government: (i) Treasury Bills 0.6 2.6 (ii) Debentures 2.8 2.8 3.1 3.9 4.8 Rest of Public Sector 0.7 0.8 0.6 0.6 0.6 Private Sector 24.3 25.9 27.5 30.8 30.4 Inclassified (net) -2.7 -4.6 -3.1 -.0 -3.8 Claims on Financial Institutions: (i) Cash 0.2 0.2 0.5 0.1 0.2 (ii) Bank Deposits 0.3 0.4 - 0.2 0.3 Foreign Assets (net of Liabilities) -1.3 -0.2 1.3 1.5 3.2 SOURCE: World Bank Country Report (1966). BIBLIOGRAPHY Andrews, V. L. "Non-Insured Corporate and State and Local Government Funds in the Financial Structure", in Private Capital Markets, Prentice Hall: 1969. Cagan, P. "The Effect of Pension Plans on Aggregate Savings", National Bureau of Economic Research, (New York:195). The Cost of Social Security. Seventh International Inquiry, 1964-66, International Labor Office, (Geneva: 1972). Feldstein, M. "Social Security and Saving: The Extended Life-Cycle Theory" American Economic Review, Papers and Proceedings, May 1976. Ghandhi, J. K. S. "The Role of Comercial Banks in Sri Lanka", Studies in Domestic Finance (forthcoming), Division of Public and Private Finance, Development Economics Department, World Bank, 1976. Goldsmith, R. W. Financial Structure and Development, Yale University Press, (New Haven:719Z9). Joshi, M. S. "The Role of Contractual Savings", Finance and Development, Vol. I, 1972. Mouton, P., Social Security In Africa, International Labor Office, (Geneva: 1974). Odle, M. A., Pension Funds in Labor Surplus Economies, Institute of Social and Economic Research, Uhiversity of the West Indies, (West Indies: 1974) Reserve Bank of India Bulletins, February 1972, May 1973, Aug. 1975. "Resource Mobilisation in Developing Countries: Financial Institutions and Policies", Studies in Domestic Finance No. 23, Division of Public and Private Finance, Development Economics Department, World Bank, 1976. Reviglio, F. "Social Security: A Means of Savings Mobilisation for Economic Development", IMF Staff Papers, May 1967. Reynolds, C. W., "The Use of Flow-of-Funds Analysis in the Study of Latin American Capital Market Development", unpublished paper, Organisation of American States, 1974. Saito, K. W., "Household Savings: An Estimation for Sri Lanka", Studies in Domestic Finance No. 27, Division of Public and Private Finance, Development Economics Department, World Bank, 1976. Schoeplein, R. N., "The Effect of Pension Plans on Other Retirement Saving", Journal of Finance, June 1970. "Simlink Model of Trade and Growth for the Developing World"., World Bank Staff Working Paper No. 220, 1975. World Bank Country Reports, 1974-1976. World Table I, Basic Economic Data Sheets, World Bank, 1976. -